Sara Freeman

Older age, higher subjective disease activity, and exercising for more than 120 minutes per week were three factors linked to patients with ankylosing spondylitis (AS) who switched from their original tumor necrosis factor inhibitor (TNFi) treatment within 2 years in a U.S.-based study.

Data from the Prospective Study of Outcomes in Ankylosing Spondylitis (PSOAS) also found that higher levels of inflammation, but not radiographic disease, were linked to patients changing from one TNFi to another, or to an interleukin (IL)-17 inhibitor or Janus kinase (JAK) inhibitor.

“Different factors were associated in AS patients who switch from their initial TNF inhibitor to another TNF inhibitor, IL-17 inhibitor, or JAK inhibitor within 2 years versus after 2 years of treatment,” John D. Reveille, MD, professor and vice chair of rheumatology and clinical immunogenetics with McGovern Medical School at UTHealth Houston, said at the 12th International Congress on Spondyloarthritides.

“We’re currently looking at different approaches to analyzing these data. And, certainly, this needs need to be looked at in other cohorts,” Dr. Reveille said.

PSOAS is a prospective observational cohort study that has been looking at predictors of AS severity for almost 20 years. Started in 2002, the study has routinely collected a whole host of data on various demographic and disease-related factors.

The reasoning behind the current analysis of PSOAS data was that a relatively recent study based on a commercial claims database had found that many patients with AS do not remain on their initial TNFi 2 years after initiation. So, Dr. Reveille and associates decided to look at the factors that could be influencing whether patients who started a TNFi would remain on their original drug in the PSOAS cohort.

In all, 533 patients from the PSOAS cohort who had at least 2 years of follow-up were included in the analysis. The majority (n = 496) were treated with a TNFi, 34 had received an IL-17 inhibitor, and 3 had received a JAK inhibitor.

Of the 496 patients treated with a TNFi, almost 70% (n = 344) persisted with this treatment for the duration of the study. Of those that switched to another TNFi or IL-17 or JAK inhibitor treatment, 20% (n = 101) did so within 2 years and the remaining 10% (n = 51) after 2 years.

Multinominal logistic regression modeling revealed a number of different factors that were associated with switching within 2 years versus switching after 2 years.

Compared to patients who persisted with treatment throughout the study period, patients who switched from their original TNFi within 2 years were more likely to be older (odds ratio [OR], 2.0 for ≥ 40 vs. < 40 years; P = .002), have a higher Bath Ankylosing Spondyloarthritis Disease Activity Index (BASDAI) score at baseline (OR, 1.73 for ≥ 4 vs. < 4; P = .03), higher C-reactive protein levels (OR, 1.94 for ≥ 0.8 mg/dL vs. < 0.8 mg/dL, P = .004), and greater weekly duration of exercise (OR, 1.95 for ≥ 120 minutes per week vs. < 120; P <.001).

Switchers also were less likely to have severe radiographic disease at baseline, as determined by the modified Stoke Ankylosing Spondylitis Spinal Score (mSASSS, OR, 0.63; P = .03), and less likely to be current smokers (OR, 0.69; P < .001).

Factors associated with switching after 2 years versus persisting with treatment were higher baseline BASDAI (OR, 2.31; P = .01), exercising more than 120 minutes per week (OR, 1.66; P = .03), and having more comorbidities (OR, 1.63 for ≥ 2 vs. < 2, P = .04).

However, patients who switched after 2 years were less likely to be depressed (OR, 0.35; P = .002) or to have a longer baseline disease duration (OR, 0.27 for ≥ 20 years vs. < 20 years P < .001).

The association observed between switching within 2 years and lower likelihood of currently smoking was a “little bit puzzling,” one delegate said after Dr. Reveille’s presentation. “The opposite has been shown in the literature, and current smokers seem to be refractory to TNF inhibitor therapy,” the delegate observed.

“I was confounded when I saw the data,” Dr. Reveille acknowledged. Because this was an observational study, this finding needs more investigation, he agreed. “Interestingly, we have seen this negative association with some other parameters, too,” he added.

The HLA-B27 carrier and radiographic status were carefully checked, so there should not be a problem with the diagnosis, Dr. Reveille reassured. Further analyses of the findings are now warranted.

Funding for the study was provided by the U.S. Department of Health and Human Services, the National Institutes of Health, and the National Institute of Arthritis and Musculoskeletal and Skin Diseases. Additional funding was received from the Spondyloarthritis Association of America and Eli Lilly.

Dr. Reveille made no personal disclosures; a coauthor of the abstract was an employee of Eli Lilly.

Older age, higher subjective disease activity, and exercising for more than 120 minutes per week were three factors linked to patients with ankylosing spondylitis (AS) who switched from their original tumor necrosis factor inhibitor (TNFi) treatment within 2 years in a U.S.-based study.

Data from the Prospective Study of Outcomes in Ankylosing Spondylitis (PSOAS) also found that higher levels of inflammation, but not radiographic disease, were linked to patients changing from one TNFi to another, or to an interleukin (IL)-17 inhibitor or Janus kinase (JAK) inhibitor.

“Different factors were associated in AS patients who switch from their initial TNF inhibitor to another TNF inhibitor, IL-17 inhibitor, or JAK inhibitor within 2 years versus after 2 years of treatment,” John D. Reveille, MD, professor and vice chair of rheumatology and clinical immunogenetics with McGovern Medical School at UTHealth Houston, said at the 12th International Congress on Spondyloarthritides.

“We’re currently looking at different approaches to analyzing these data. And, certainly, this needs need to be looked at in other cohorts,” Dr. Reveille said.

PSOAS is a prospective observational cohort study that has been looking at predictors of AS severity for almost 20 years. Started in 2002, the study has routinely collected a whole host of data on various demographic and disease-related factors.

The reasoning behind the current analysis of PSOAS data was that a relatively recent study based on a commercial claims database had found that many patients with AS do not remain on their initial TNFi 2 years after initiation. So, Dr. Reveille and associates decided to look at the factors that could be influencing whether patients who started a TNFi would remain on their original drug in the PSOAS cohort.

In all, 533 patients from the PSOAS cohort who had at least 2 years of follow-up were included in the analysis. The majority (n = 496) were treated with a TNFi, 34 had received an IL-17 inhibitor, and 3 had received a JAK inhibitor.

Of the 496 patients treated with a TNFi, almost 70% (n = 344) persisted with this treatment for the duration of the study. Of those that switched to another TNFi or IL-17 or JAK inhibitor treatment, 20% (n = 101) did so within 2 years and the remaining 10% (n = 51) after 2 years.

Multinominal logistic regression modeling revealed a number of different factors that were associated with switching within 2 years versus switching after 2 years.

Compared to patients who persisted with treatment throughout the study period, patients who switched from their original TNFi within 2 years were more likely to be older (odds ratio [OR], 2.0 for ≥ 40 vs. < 40 years; P = .002), have a higher Bath Ankylosing Spondyloarthritis Disease Activity Index (BASDAI) score at baseline (OR, 1.73 for ≥ 4 vs. < 4; P = .03), higher C-reactive protein levels (OR, 1.94 for ≥ 0.8 mg/dL vs. < 0.8 mg/dL, P = .004), and greater weekly duration of exercise (OR, 1.95 for ≥ 120 minutes per week vs. < 120; P <.001).

Switchers also were less likely to have severe radiographic disease at baseline, as determined by the modified Stoke Ankylosing Spondylitis Spinal Score (mSASSS, OR, 0.63; P = .03), and less likely to be current smokers (OR, 0.69; P < .001).

Factors associated with switching after 2 years versus persisting with treatment were higher baseline BASDAI (OR, 2.31; P = .01), exercising more than 120 minutes per week (OR, 1.66; P = .03), and having more comorbidities (OR, 1.63 for ≥ 2 vs. < 2, P = .04).

However, patients who switched after 2 years were less likely to be depressed (OR, 0.35; P = .002) or to have a longer baseline disease duration (OR, 0.27 for ≥ 20 years vs. < 20 years P < .001).

The association observed between switching within 2 years and lower likelihood of currently smoking was a “little bit puzzling,” one delegate said after Dr. Reveille’s presentation. “The opposite has been shown in the literature, and current smokers seem to be refractory to TNF inhibitor therapy,” the delegate observed.

“I was confounded when I saw the data,” Dr. Reveille acknowledged. Because this was an observational study, this finding needs more investigation, he agreed. “Interestingly, we have seen this negative association with some other parameters, too,” he added.

The HLA-B27 carrier and radiographic status were carefully checked, so there should not be a problem with the diagnosis, Dr. Reveille reassured. Further analyses of the findings are now warranted.

Funding for the study was provided by the U.S. Department of Health and Human Services, the National Institutes of Health, and the National Institute of Arthritis and Musculoskeletal and Skin Diseases. Additional funding was received from the Spondyloarthritis Association of America and Eli Lilly.

Dr. Reveille made no personal disclosures; a coauthor of the abstract was an employee of Eli Lilly.

Older age, higher subjective disease activity, and exercising for more than 120 minutes per week were three factors linked to patients with ankylosing spondylitis (AS) who switched from their original tumor necrosis factor inhibitor (TNFi) treatment within 2 years in a U.S.-based study.

Data from the Prospective Study of Outcomes in Ankylosing Spondylitis (PSOAS) also found that higher levels of inflammation, but not radiographic disease, were linked to patients changing from one TNFi to another, or to an interleukin (IL)-17 inhibitor or Janus kinase (JAK) inhibitor.

“Different factors were associated in AS patients who switch from their initial TNF inhibitor to another TNF inhibitor, IL-17 inhibitor, or JAK inhibitor within 2 years versus after 2 years of treatment,” John D. Reveille, MD, professor and vice chair of rheumatology and clinical immunogenetics with McGovern Medical School at UTHealth Houston, said at the 12th International Congress on Spondyloarthritides.

“We’re currently looking at different approaches to analyzing these data. And, certainly, this needs need to be looked at in other cohorts,” Dr. Reveille said.

PSOAS is a prospective observational cohort study that has been looking at predictors of AS severity for almost 20 years. Started in 2002, the study has routinely collected a whole host of data on various demographic and disease-related factors.

The reasoning behind the current analysis of PSOAS data was that a relatively recent study based on a commercial claims database had found that many patients with AS do not remain on their initial TNFi 2 years after initiation. So, Dr. Reveille and associates decided to look at the factors that could be influencing whether patients who started a TNFi would remain on their original drug in the PSOAS cohort.

In all, 533 patients from the PSOAS cohort who had at least 2 years of follow-up were included in the analysis. The majority (n = 496) were treated with a TNFi, 34 had received an IL-17 inhibitor, and 3 had received a JAK inhibitor.

Of the 496 patients treated with a TNFi, almost 70% (n = 344) persisted with this treatment for the duration of the study. Of those that switched to another TNFi or IL-17 or JAK inhibitor treatment, 20% (n = 101) did so within 2 years and the remaining 10% (n = 51) after 2 years.

Multinominal logistic regression modeling revealed a number of different factors that were associated with switching within 2 years versus switching after 2 years.

Compared to patients who persisted with treatment throughout the study period, patients who switched from their original TNFi within 2 years were more likely to be older (odds ratio [OR], 2.0 for ≥ 40 vs. < 40 years; P = .002), have a higher Bath Ankylosing Spondyloarthritis Disease Activity Index (BASDAI) score at baseline (OR, 1.73 for ≥ 4 vs. < 4; P = .03), higher C-reactive protein levels (OR, 1.94 for ≥ 0.8 mg/dL vs. < 0.8 mg/dL, P = .004), and greater weekly duration of exercise (OR, 1.95 for ≥ 120 minutes per week vs. < 120; P <.001).

Switchers also were less likely to have severe radiographic disease at baseline, as determined by the modified Stoke Ankylosing Spondylitis Spinal Score (mSASSS, OR, 0.63; P = .03), and less likely to be current smokers (OR, 0.69; P < .001).

Factors associated with switching after 2 years versus persisting with treatment were higher baseline BASDAI (OR, 2.31; P = .01), exercising more than 120 minutes per week (OR, 1.66; P = .03), and having more comorbidities (OR, 1.63 for ≥ 2 vs. < 2, P = .04).

However, patients who switched after 2 years were less likely to be depressed (OR, 0.35; P = .002) or to have a longer baseline disease duration (OR, 0.27 for ≥ 20 years vs. < 20 years P < .001).

The association observed between switching within 2 years and lower likelihood of currently smoking was a “little bit puzzling,” one delegate said after Dr. Reveille’s presentation. “The opposite has been shown in the literature, and current smokers seem to be refractory to TNF inhibitor therapy,” the delegate observed.

“I was confounded when I saw the data,” Dr. Reveille acknowledged. Because this was an observational study, this finding needs more investigation, he agreed. “Interestingly, we have seen this negative association with some other parameters, too,” he added.

The HLA-B27 carrier and radiographic status were carefully checked, so there should not be a problem with the diagnosis, Dr. Reveille reassured. Further analyses of the findings are now warranted.

Funding for the study was provided by the U.S. Department of Health and Human Services, the National Institutes of Health, and the National Institute of Arthritis and Musculoskeletal and Skin Diseases. Additional funding was received from the Spondyloarthritis Association of America and Eli Lilly.

Dr. Reveille made no personal disclosures; a coauthor of the abstract was an employee of Eli Lilly.

How best to define axial spondyloarthritis (axSpA) flares in practice remains the subject of some debate as evidenced by the discussion that followed an abstract presentation at the 12th International Congress on Spondyloarthritides.

It’s an important topic, said Maxime Breban, MD, PhD, of Ambroise Paré Hospital in Paris, as flares can adversely affect patient outcomes. The absence of flares may also a useful measure of how well a patient is responding to treatment in clinical trials and whether a treatment can be tapered.

“There have been many ways to define flares in the past and there is no consensus,” he observed.

Although the Assessment of Spondyloarthritis International Society (ASAS) devised 12 preliminary definitions of flare in 2016, “these were not that good when we moved to patients,” Dr. Breban suggested.

The ASAS definitions were based on patient vignettes, he explained, and used a combination of variables from the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), a visual analog scale (VAS) of pain, and the Ankylosing Spondylitis Disease Activity Score based on C-reactive protein (ASDAS-CRP).

The study that Dr. Breban presented looked at the performance of the ASAS preliminary definitions of axSpA flares in a real-life patient population, as well as prospectively determining how variations in BASDAI and VAS pain were associated with patient-perceived flares of disease.

A total of 99 patients took part in the study, recruited through a secure e-health platform called SPONDY+. Once a week, patients completed the BASDAI questionnaire and the pain VAS, and stated whether their disease had flared in the past week.

Receiver operating characteristic (ROC) curves were calculated to see how well the BASDAI and pain VAS identified patients who were experiencing a flare or had a recently resolved flare of axSpA.

Dr. Breban reported that variation in the BASDAI “appears a suitable variable to monitor the occurrence and resolution of patient-reported flare in axial spondylarthritis.”

In predicting a flare, the area under the curve (AUC) was significantly higher for the change in BASDAI than for the change in pain VAS, at a respective 0.81 and 0.77 (P = .01). However, both variables were similarly accurate in predicting the resolution of a flare, with respective AUCs of 0.78 and 0.80 (P = .3).

A 0.22-point increase in BASDAI was reported to be the best balance between sensitivity (70%) and specificity (79%) for a flare. However, this is “outside of what is possible within a test–retest situation,” Désirée van der Heijde, MD, PhD, of Leiden (the Netherlands) University Medical Center, said during discussion.

Dr. van der Heijde told Dr. Breban: “I understand that that comes out of your data, that that’s the best combination for sensitivity and specificity, but the next step is to decide if that makes sense.”

The ROC curves that Dr. Breban presented showed the range of sensitivities and specificities that could be achieved. If the specificity was increased to be 90% or higher, the specificity fell to 55%, with the change in BASDAI being an increase of 0.8 points. Conversely, bringing the sensitivity above 90% meant the specificity dropped to 39% and the change in BASDAI was a decrease of 0.1 point.

“So that means you can choose whatever you want as a cutoff,” Dr. Breban said. It depends on what you are aiming to do. “If you want to identify a flare, you can increase sensitivity, or specificity, according to what your purpose is,” he suggested.

“The next step, of course, is what to choose as a flare. Then it depends on how you want to use a flare if you want to use a flare to change the treatment,” agreed Dr. van der Heijde. “That was why, in the ASAS group, it was decided to have a high specificity so that you are not changing treatment all the time.”

In the data that Dr. Breban presented, the ASAS preliminary definitions were highly specific but lacked sensitivity. None of the ASAS definitions yielded sensitivity values higher than 37%, whereas specificity was higher than 95% for all of them.

The study’s design did not allow researchers to test the ASDAS-CRP as a definition of flare in its real-world patient sample. Thus, it is looking only at the patient’s perspective on flare, and there is a “huge discrepancy” between patient and physician-reported disease activity, Dr. van der Heijde noted. “So, I think before using your data to really choose the flare definition, I think we need to take it all into account.”

Maxime Dougados, MD, PhD, of Cochin Hospital in Paris, who has been “deeply involved in the elaboration of the definition of flare” added his thoughts: “Flare means for me, not a status, but a change,” he observed.

But if the aim of treating people with axSpA is to achieve a good or acceptable state of health, he questioned whether work should be continued to define the concept of a flare.

The definition of a flare was conceived for use in clinical trials mainly, Dr. van der Heijde noted. It helped to assess how changes in treatment might affect the outcomes of patients. In clinical practice, especially now with treat-to-target gaining more and more traction in axSpA, she agreed that perhaps the goal should be to focus more on the health status of patients.

Dr. Breban acknowledged that the SPONDY+ platform has been developed by bepatient with support from Merck Sharp & Dohme. No other disclosures were made.

How best to define axial spondyloarthritis (axSpA) flares in practice remains the subject of some debate as evidenced by the discussion that followed an abstract presentation at the 12th International Congress on Spondyloarthritides.

It’s an important topic, said Maxime Breban, MD, PhD, of Ambroise Paré Hospital in Paris, as flares can adversely affect patient outcomes. The absence of flares may also a useful measure of how well a patient is responding to treatment in clinical trials and whether a treatment can be tapered.

“There have been many ways to define flares in the past and there is no consensus,” he observed.

Although the Assessment of Spondyloarthritis International Society (ASAS) devised 12 preliminary definitions of flare in 2016, “these were not that good when we moved to patients,” Dr. Breban suggested.

The ASAS definitions were based on patient vignettes, he explained, and used a combination of variables from the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), a visual analog scale (VAS) of pain, and the Ankylosing Spondylitis Disease Activity Score based on C-reactive protein (ASDAS-CRP).

The study that Dr. Breban presented looked at the performance of the ASAS preliminary definitions of axSpA flares in a real-life patient population, as well as prospectively determining how variations in BASDAI and VAS pain were associated with patient-perceived flares of disease.

A total of 99 patients took part in the study, recruited through a secure e-health platform called SPONDY+. Once a week, patients completed the BASDAI questionnaire and the pain VAS, and stated whether their disease had flared in the past week.

Receiver operating characteristic (ROC) curves were calculated to see how well the BASDAI and pain VAS identified patients who were experiencing a flare or had a recently resolved flare of axSpA.

Dr. Breban reported that variation in the BASDAI “appears a suitable variable to monitor the occurrence and resolution of patient-reported flare in axial spondylarthritis.”

In predicting a flare, the area under the curve (AUC) was significantly higher for the change in BASDAI than for the change in pain VAS, at a respective 0.81 and 0.77 (P = .01). However, both variables were similarly accurate in predicting the resolution of a flare, with respective AUCs of 0.78 and 0.80 (P = .3).

A 0.22-point increase in BASDAI was reported to be the best balance between sensitivity (70%) and specificity (79%) for a flare. However, this is “outside of what is possible within a test–retest situation,” Désirée van der Heijde, MD, PhD, of Leiden (the Netherlands) University Medical Center, said during discussion.

Dr. van der Heijde told Dr. Breban: “I understand that that comes out of your data, that that’s the best combination for sensitivity and specificity, but the next step is to decide if that makes sense.”

The ROC curves that Dr. Breban presented showed the range of sensitivities and specificities that could be achieved. If the specificity was increased to be 90% or higher, the specificity fell to 55%, with the change in BASDAI being an increase of 0.8 points. Conversely, bringing the sensitivity above 90% meant the specificity dropped to 39% and the change in BASDAI was a decrease of 0.1 point.

“So that means you can choose whatever you want as a cutoff,” Dr. Breban said. It depends on what you are aiming to do. “If you want to identify a flare, you can increase sensitivity, or specificity, according to what your purpose is,” he suggested.

“The next step, of course, is what to choose as a flare. Then it depends on how you want to use a flare if you want to use a flare to change the treatment,” agreed Dr. van der Heijde. “That was why, in the ASAS group, it was decided to have a high specificity so that you are not changing treatment all the time.”

In the data that Dr. Breban presented, the ASAS preliminary definitions were highly specific but lacked sensitivity. None of the ASAS definitions yielded sensitivity values higher than 37%, whereas specificity was higher than 95% for all of them.

The study’s design did not allow researchers to test the ASDAS-CRP as a definition of flare in its real-world patient sample. Thus, it is looking only at the patient’s perspective on flare, and there is a “huge discrepancy” between patient and physician-reported disease activity, Dr. van der Heijde noted. “So, I think before using your data to really choose the flare definition, I think we need to take it all into account.”

Maxime Dougados, MD, PhD, of Cochin Hospital in Paris, who has been “deeply involved in the elaboration of the definition of flare” added his thoughts: “Flare means for me, not a status, but a change,” he observed.

But if the aim of treating people with axSpA is to achieve a good or acceptable state of health, he questioned whether work should be continued to define the concept of a flare.

The definition of a flare was conceived for use in clinical trials mainly, Dr. van der Heijde noted. It helped to assess how changes in treatment might affect the outcomes of patients. In clinical practice, especially now with treat-to-target gaining more and more traction in axSpA, she agreed that perhaps the goal should be to focus more on the health status of patients.

Dr. Breban acknowledged that the SPONDY+ platform has been developed by bepatient with support from Merck Sharp & Dohme. No other disclosures were made.

How best to define axial spondyloarthritis (axSpA) flares in practice remains the subject of some debate as evidenced by the discussion that followed an abstract presentation at the 12th International Congress on Spondyloarthritides.

It’s an important topic, said Maxime Breban, MD, PhD, of Ambroise Paré Hospital in Paris, as flares can adversely affect patient outcomes. The absence of flares may also a useful measure of how well a patient is responding to treatment in clinical trials and whether a treatment can be tapered.

“There have been many ways to define flares in the past and there is no consensus,” he observed.

Although the Assessment of Spondyloarthritis International Society (ASAS) devised 12 preliminary definitions of flare in 2016, “these were not that good when we moved to patients,” Dr. Breban suggested.

The ASAS definitions were based on patient vignettes, he explained, and used a combination of variables from the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), a visual analog scale (VAS) of pain, and the Ankylosing Spondylitis Disease Activity Score based on C-reactive protein (ASDAS-CRP).

The study that Dr. Breban presented looked at the performance of the ASAS preliminary definitions of axSpA flares in a real-life patient population, as well as prospectively determining how variations in BASDAI and VAS pain were associated with patient-perceived flares of disease.

A total of 99 patients took part in the study, recruited through a secure e-health platform called SPONDY+. Once a week, patients completed the BASDAI questionnaire and the pain VAS, and stated whether their disease had flared in the past week.

Receiver operating characteristic (ROC) curves were calculated to see how well the BASDAI and pain VAS identified patients who were experiencing a flare or had a recently resolved flare of axSpA.

Dr. Breban reported that variation in the BASDAI “appears a suitable variable to monitor the occurrence and resolution of patient-reported flare in axial spondylarthritis.”

In predicting a flare, the area under the curve (AUC) was significantly higher for the change in BASDAI than for the change in pain VAS, at a respective 0.81 and 0.77 (P = .01). However, both variables were similarly accurate in predicting the resolution of a flare, with respective AUCs of 0.78 and 0.80 (P = .3).

A 0.22-point increase in BASDAI was reported to be the best balance between sensitivity (70%) and specificity (79%) for a flare. However, this is “outside of what is possible within a test–retest situation,” Désirée van der Heijde, MD, PhD, of Leiden (the Netherlands) University Medical Center, said during discussion.

Dr. van der Heijde told Dr. Breban: “I understand that that comes out of your data, that that’s the best combination for sensitivity and specificity, but the next step is to decide if that makes sense.”

The ROC curves that Dr. Breban presented showed the range of sensitivities and specificities that could be achieved. If the specificity was increased to be 90% or higher, the specificity fell to 55%, with the change in BASDAI being an increase of 0.8 points. Conversely, bringing the sensitivity above 90% meant the specificity dropped to 39% and the change in BASDAI was a decrease of 0.1 point.

“So that means you can choose whatever you want as a cutoff,” Dr. Breban said. It depends on what you are aiming to do. “If you want to identify a flare, you can increase sensitivity, or specificity, according to what your purpose is,” he suggested.

“The next step, of course, is what to choose as a flare. Then it depends on how you want to use a flare if you want to use a flare to change the treatment,” agreed Dr. van der Heijde. “That was why, in the ASAS group, it was decided to have a high specificity so that you are not changing treatment all the time.”

In the data that Dr. Breban presented, the ASAS preliminary definitions were highly specific but lacked sensitivity. None of the ASAS definitions yielded sensitivity values higher than 37%, whereas specificity was higher than 95% for all of them.

The study’s design did not allow researchers to test the ASDAS-CRP as a definition of flare in its real-world patient sample. Thus, it is looking only at the patient’s perspective on flare, and there is a “huge discrepancy” between patient and physician-reported disease activity, Dr. van der Heijde noted. “So, I think before using your data to really choose the flare definition, I think we need to take it all into account.”

Maxime Dougados, MD, PhD, of Cochin Hospital in Paris, who has been “deeply involved in the elaboration of the definition of flare” added his thoughts: “Flare means for me, not a status, but a change,” he observed.

But if the aim of treating people with axSpA is to achieve a good or acceptable state of health, he questioned whether work should be continued to define the concept of a flare.

The definition of a flare was conceived for use in clinical trials mainly, Dr. van der Heijde noted. It helped to assess how changes in treatment might affect the outcomes of patients. In clinical practice, especially now with treat-to-target gaining more and more traction in axSpA, she agreed that perhaps the goal should be to focus more on the health status of patients.

Dr. Breban acknowledged that the SPONDY+ platform has been developed by bepatient with support from Merck Sharp & Dohme. No other disclosures were made.

Interleukin (IL)–23 inhibition may still have a role to play in the treatment of patients with axial spondyloarthritis (SpA), suggests research presented at the 12th International Congress on Spondyloarthritides.

There is a strong rationale for using IL-23 inhibitors in patients with axial SpA, and the IL-23/IL-17 axis has been proposed as a critical player in the pathophysiology of the disease. But around 2018 it became clear from randomized, controlled trials that IL-23 inhibition was ineffective at improving key clinical outcomes, at least in patients with axial disease.

Although the overall results of a systematic review and meta-analysis that was presented at the meeting corroborated the negative results seen with IL-23–inhibiting agents in clinical trials, there were some data showing benefits of the IL-23 inhibitor risankizumab on secondary outcomes in one trial.

To look at the available evidence, Louise Vanhoutte, a 2nd-year internal medicine student at University Hospitals Leuven (Belgium) worked under the guidance of Rik Lories, MD, PhD, head of the division of rheumatology at University Hospitals Leuven. Together they searched known databases for randomized, controlled trials investigating the use of IL-23 and IL-17 inhibitors for the treatment of adults with axial SpA or psoriatic arthritis. Studies could be either phase 2 or phase 3, but had to have included a placebo and used the ASAS40 (40% Improvement in Assessment of SpondyloArthritis International Society Response criteria), ASAS20, Bath Ankylosing Spondylitis Disease Activity Index, or SPARCC (Spondyloarthritis Research Consortium of Canada) index score to assess outcomes.

The systematic review whittled the number of clinical trials in the meta-analysis to 12, which concerned the use of ustekinumab, an IL-12/23 inhibitor, and risankizumab, along with two IL-17 inhibitors, ixekizumab and secukinumab. Data for the IL-23 inhibitors guselkumab and tildrakizumab were not available.

“To no surprise, Forest plots showed that there was a lack of efficacy for IL-23 agents in the treatment of axial spondyloarthritis and a superior efficacy for IL-17 inhibitors in the treatment of axial spondyloarthritis,” Ms. Vanhoutte reported.

The respective odds ratios for IL-23 and IL-17 inhibitors in getting patients to meet ASAS40 response criteria in comparison to baseline were 1.51 (95% confidence interval, 0.98-2.31) and 2.54 (95% CI, 2.02-3.19).

“Does this mean it is a dead-end street for all IL-23 inhibition?” she asked. Not necessarily. In the meta-analysis, not only did risankizumab lower the Ankylosing Spondylitis Disease Activity Score based on C-reactive protein (ASDAS-CRP) by a mean difference (MD) of –0.30 (95% CI, –0.41 to –0.19) from baseline values, but it also led to statistically significant reductions in SPARCC index score for the spine (MD, –3.10; 95% CI, –4.50 to –1.70) and high-sensitivity CRP (MD, –2.10; 95% CI, –2.56 to –1.64). The risankizumab findings might suggest there are potential disease-modifying properties for specifically targeting IL-23p19. There could also be a window of opportunity to use IL-23 inhibitors earlier.

“These are only results from one randomized, controlled trial in a small sample size where outcomes were reported as medians and interquartile ranges, so they had to be converted to means and standard deviations to have an odds ratio in the end,” she explained.

“Also, these were results from a radiographic axial spondyloarthritis population and not a nonradiographic axial spondyloarthritis population,” she added.

While that might limit the interpretation of the findings, “what we see here is both reduction in inflammation and reduction in structural disease progression as [measured] by SPARCC,” Ms. Vanhoutte said.

“Since IL-23 is an upstream molecule from IL-17 it’s probable that IL-23 is present in the prephase of the disease, in a prephase inflammation state,” she hypothesized. “This is especially interesting because there are very few randomized, controlled trials that examine therapeutic agents in nonradiographic axial spondyloarthritis,” she observed. Looking at IL-23 in radiographic, or established, disease therefore may not be as useful.

“I’m thinking you’re making actually a very important point for us,” commented Robert Landewé, MD, PhD, of Amsterdam University Medical Center.

“We are discussing whether or not IL-23 is important in inhibiting the disease activity of patients with axial spondyloarthritis, and we are surprised that it is not shown in RCTs.

“Why is it completely ineffective in axial spondyloarthritis? You show us that that is probably not the case,” Dr. Landewé suggested.

“What you make very clear here is that indeed there is some efficacy, and from a pathophysiological way of thinking it might be slightly different as compared with what most clinicians nowadays think.”

The study had no specific funding, and no disclosures were reported.

Interleukin (IL)–23 inhibition may still have a role to play in the treatment of patients with axial spondyloarthritis (SpA), suggests research presented at the 12th International Congress on Spondyloarthritides.

There is a strong rationale for using IL-23 inhibitors in patients with axial SpA, and the IL-23/IL-17 axis has been proposed as a critical player in the pathophysiology of the disease. But around 2018 it became clear from randomized, controlled trials that IL-23 inhibition was ineffective at improving key clinical outcomes, at least in patients with axial disease.

Although the overall results of a systematic review and meta-analysis that was presented at the meeting corroborated the negative results seen with IL-23–inhibiting agents in clinical trials, there were some data showing benefits of the IL-23 inhibitor risankizumab on secondary outcomes in one trial.

To look at the available evidence, Louise Vanhoutte, a 2nd-year internal medicine student at University Hospitals Leuven (Belgium) worked under the guidance of Rik Lories, MD, PhD, head of the division of rheumatology at University Hospitals Leuven. Together they searched known databases for randomized, controlled trials investigating the use of IL-23 and IL-17 inhibitors for the treatment of adults with axial SpA or psoriatic arthritis. Studies could be either phase 2 or phase 3, but had to have included a placebo and used the ASAS40 (40% Improvement in Assessment of SpondyloArthritis International Society Response criteria), ASAS20, Bath Ankylosing Spondylitis Disease Activity Index, or SPARCC (Spondyloarthritis Research Consortium of Canada) index score to assess outcomes.

The systematic review whittled the number of clinical trials in the meta-analysis to 12, which concerned the use of ustekinumab, an IL-12/23 inhibitor, and risankizumab, along with two IL-17 inhibitors, ixekizumab and secukinumab. Data for the IL-23 inhibitors guselkumab and tildrakizumab were not available.

“To no surprise, Forest plots showed that there was a lack of efficacy for IL-23 agents in the treatment of axial spondyloarthritis and a superior efficacy for IL-17 inhibitors in the treatment of axial spondyloarthritis,” Ms. Vanhoutte reported.

The respective odds ratios for IL-23 and IL-17 inhibitors in getting patients to meet ASAS40 response criteria in comparison to baseline were 1.51 (95% confidence interval, 0.98-2.31) and 2.54 (95% CI, 2.02-3.19).

“Does this mean it is a dead-end street for all IL-23 inhibition?” she asked. Not necessarily. In the meta-analysis, not only did risankizumab lower the Ankylosing Spondylitis Disease Activity Score based on C-reactive protein (ASDAS-CRP) by a mean difference (MD) of –0.30 (95% CI, –0.41 to –0.19) from baseline values, but it also led to statistically significant reductions in SPARCC index score for the spine (MD, –3.10; 95% CI, –4.50 to –1.70) and high-sensitivity CRP (MD, –2.10; 95% CI, –2.56 to –1.64). The risankizumab findings might suggest there are potential disease-modifying properties for specifically targeting IL-23p19. There could also be a window of opportunity to use IL-23 inhibitors earlier.

“These are only results from one randomized, controlled trial in a small sample size where outcomes were reported as medians and interquartile ranges, so they had to be converted to means and standard deviations to have an odds ratio in the end,” she explained.

“Also, these were results from a radiographic axial spondyloarthritis population and not a nonradiographic axial spondyloarthritis population,” she added.

While that might limit the interpretation of the findings, “what we see here is both reduction in inflammation and reduction in structural disease progression as [measured] by SPARCC,” Ms. Vanhoutte said.

“Since IL-23 is an upstream molecule from IL-17 it’s probable that IL-23 is present in the prephase of the disease, in a prephase inflammation state,” she hypothesized. “This is especially interesting because there are very few randomized, controlled trials that examine therapeutic agents in nonradiographic axial spondyloarthritis,” she observed. Looking at IL-23 in radiographic, or established, disease therefore may not be as useful.

“I’m thinking you’re making actually a very important point for us,” commented Robert Landewé, MD, PhD, of Amsterdam University Medical Center.

“We are discussing whether or not IL-23 is important in inhibiting the disease activity of patients with axial spondyloarthritis, and we are surprised that it is not shown in RCTs.

“Why is it completely ineffective in axial spondyloarthritis? You show us that that is probably not the case,” Dr. Landewé suggested.

“What you make very clear here is that indeed there is some efficacy, and from a pathophysiological way of thinking it might be slightly different as compared with what most clinicians nowadays think.”

The study had no specific funding, and no disclosures were reported.

Interleukin (IL)–23 inhibition may still have a role to play in the treatment of patients with axial spondyloarthritis (SpA), suggests research presented at the 12th International Congress on Spondyloarthritides.

There is a strong rationale for using IL-23 inhibitors in patients with axial SpA, and the IL-23/IL-17 axis has been proposed as a critical player in the pathophysiology of the disease. But around 2018 it became clear from randomized, controlled trials that IL-23 inhibition was ineffective at improving key clinical outcomes, at least in patients with axial disease.

Although the overall results of a systematic review and meta-analysis that was presented at the meeting corroborated the negative results seen with IL-23–inhibiting agents in clinical trials, there were some data showing benefits of the IL-23 inhibitor risankizumab on secondary outcomes in one trial.

To look at the available evidence, Louise Vanhoutte, a 2nd-year internal medicine student at University Hospitals Leuven (Belgium) worked under the guidance of Rik Lories, MD, PhD, head of the division of rheumatology at University Hospitals Leuven. Together they searched known databases for randomized, controlled trials investigating the use of IL-23 and IL-17 inhibitors for the treatment of adults with axial SpA or psoriatic arthritis. Studies could be either phase 2 or phase 3, but had to have included a placebo and used the ASAS40 (40% Improvement in Assessment of SpondyloArthritis International Society Response criteria), ASAS20, Bath Ankylosing Spondylitis Disease Activity Index, or SPARCC (Spondyloarthritis Research Consortium of Canada) index score to assess outcomes.

The systematic review whittled the number of clinical trials in the meta-analysis to 12, which concerned the use of ustekinumab, an IL-12/23 inhibitor, and risankizumab, along with two IL-17 inhibitors, ixekizumab and secukinumab. Data for the IL-23 inhibitors guselkumab and tildrakizumab were not available.

“To no surprise, Forest plots showed that there was a lack of efficacy for IL-23 agents in the treatment of axial spondyloarthritis and a superior efficacy for IL-17 inhibitors in the treatment of axial spondyloarthritis,” Ms. Vanhoutte reported.

The respective odds ratios for IL-23 and IL-17 inhibitors in getting patients to meet ASAS40 response criteria in comparison to baseline were 1.51 (95% confidence interval, 0.98-2.31) and 2.54 (95% CI, 2.02-3.19).

“Does this mean it is a dead-end street for all IL-23 inhibition?” she asked. Not necessarily. In the meta-analysis, not only did risankizumab lower the Ankylosing Spondylitis Disease Activity Score based on C-reactive protein (ASDAS-CRP) by a mean difference (MD) of –0.30 (95% CI, –0.41 to –0.19) from baseline values, but it also led to statistically significant reductions in SPARCC index score for the spine (MD, –3.10; 95% CI, –4.50 to –1.70) and high-sensitivity CRP (MD, –2.10; 95% CI, –2.56 to –1.64). The risankizumab findings might suggest there are potential disease-modifying properties for specifically targeting IL-23p19. There could also be a window of opportunity to use IL-23 inhibitors earlier.

“These are only results from one randomized, controlled trial in a small sample size where outcomes were reported as medians and interquartile ranges, so they had to be converted to means and standard deviations to have an odds ratio in the end,” she explained.

“Also, these were results from a radiographic axial spondyloarthritis population and not a nonradiographic axial spondyloarthritis population,” she added.

While that might limit the interpretation of the findings, “what we see here is both reduction in inflammation and reduction in structural disease progression as [measured] by SPARCC,” Ms. Vanhoutte said.

“Since IL-23 is an upstream molecule from IL-17 it’s probable that IL-23 is present in the prephase of the disease, in a prephase inflammation state,” she hypothesized. “This is especially interesting because there are very few randomized, controlled trials that examine therapeutic agents in nonradiographic axial spondyloarthritis,” she observed. Looking at IL-23 in radiographic, or established, disease therefore may not be as useful.

“I’m thinking you’re making actually a very important point for us,” commented Robert Landewé, MD, PhD, of Amsterdam University Medical Center.

“We are discussing whether or not IL-23 is important in inhibiting the disease activity of patients with axial spondyloarthritis, and we are surprised that it is not shown in RCTs.

“Why is it completely ineffective in axial spondyloarthritis? You show us that that is probably not the case,” Dr. Landewé suggested.

“What you make very clear here is that indeed there is some efficacy, and from a pathophysiological way of thinking it might be slightly different as compared with what most clinicians nowadays think.”

The study had no specific funding, and no disclosures were reported.

Better retention and remission rates with tumor necrosis factor inhibitors (TNFi) have been observed in patients with spondyloarthritis (SpA) who were also treated with a conventional synthetic disease-modifying antirheumatic drug (csDMARD) rather than monotherapy.

Data from the EuroSpA Research Collaboration have shown that 82% of patients who received TNFi and csDMARD cotherapy were still taking their TNFi at 1 year versus 79% of those who were taking the biologic alone (P < .001).

Combination therapy led to remission in 21.8%, compared with TNFi monotherapy at 19.5% (P < .0001), with rates of remission according to Ankylosing Spondylitis Disease Activity Score <1.3 (23.7% vs. 21.8%, P = .011) and Bath Ankylosing Spondylitis Disease Activity Index <2 (5.9% vs. 7.7%, P < .0001) also favoring the cotherapy group.

“We did see a considerable heterogeneity across the countries in this study, both in the use of csDMARD therapy and in the TNF [inhibitor] retention,” Michael J. Nissen, MBBS, FRACP, MD, a rheumatologist at the University of Geneva, said in an oral abstract presentation at the 12th International Congress on Spondyloarthritides.

“Overall, the 1-year TNFi retention was better with cotherapy, particularly in countries in the Scandinavian region and Switzerland,” Dr. Nissen observed.

“The csDMARD cotherapy significantly improved remission; we could perhaps argue whether these small differences are clinically meaningful, but they were highly significant in the study.”
 

Why look at csDMARD and TNFi cotherapy?

European guidelines suggest that csDMARDs are not that useful for treating people with SpA, stating that they should not be used in axial disease and used only as monotherapy in those with peripheral disease.

“Nevertheless, these medications are often used, which begs us to ask the question why is that the case?” Dr. Nissen said.

As previous work by Dr. Nissen and others had suggested that there might be a benefit of combining a csDMARD and a TNFi, the aim was to build upon this and see if looking at a very large patient population might be able to provide some clear answers.

The analysis included data from 13 European registries and more than 24,000 patients who had started their first TNFi between 2006 and 2017. The mean age of the study population was 42.5 years, and 58% were male, with an average disease duration of 5.9 years. Nearly one-third of patients were receiving comedication with a csDMARD, although this ranged from 13.5% to 71.2% depending on the country.

The most frequently used TNFi were adalimumab (31% of patients overall), etanercept (24%), and infliximab (25%), and the most-used csDMARDs were methotrexate (56%) and sulfasalazine (45%).

The effectiveness of treatment in achieving clinical remission was examined according to the TNFi used and for the use of csDMARDs.

“We found that sulfasalazine and methotrexate were similarly effective in terms of improving outcomes in addition to a TNF inhibitor,” Dr. Nissen said. For example, the adjusted odds ratios for using sulfasalazine or methotrexate with infliximab were a respective 1.32 and 1.37, and the aORs for uses these specific csDMARDs with etanercept were 1.38 and 1.35.

The researchers reported finding significantly higher aORs if both sulfasalazine and methotrexate were used with a TNFi (1.67 for any TNFi, 1.95 for infliximab, 1.45 foretanercept, and 1.85 for adalimumab).

This perhaps suggests “there’s some role for combining csDMARDs to have an even greater effect,” Dr. Nissen put forward.
 

A ‘very provocative conclusion’

That’s “a very provocative conclusion” commented Robert Landewé, MD, PhD, of Amsterdam University Medical Center who chaired the sessions. “You can see it by the [number] of people that want to ask questions.”

Désirée van der Heijde, MD, PhD, of Leiden (the Netherlands) University Medical Center, commented these data were “timely as we are updating the ASAS/EULAR [Assessment of SpondyloArthritis international Society/European Alliance of Associations for Rheumatology] recommendations at the moment.” She queried, however, if the differences were clinically significant.

“I think if you have a database of 24,000 patients, highly, statistically significant doesn’t tell me a lot. So, I want to look at the clinical meaning of the differences. So, if you have a 2% difference in retention rate, is that worth prescribing comedication?” Dr. van der Heijde said.

“I think you’re absolutely right,” Dr. Nissen responded. “We’re obviously going to find highly significant differences with such a big patient group.” While the small statistical differences seen may not look very clinically relevant at face value, they could provide some guidance for clinical decision making.

“It may alter our approach if that patient is already on a csDMARD and has peripheral disease; perhaps there’s tendency to then to keep that csDMARD rather than then stopping it and switching to a biologic agent,” Dr. Nissen argued. It’s a topic that has been debated for some time, he added, and further study is needed, but “I think it helps give a little bit more clarity to the idea.”

Another point of discussion was the proportion of patients who had “pure axial disease.” More patients in the comedication group had peripheral disease, Dr. van der Heijde pointed out, “that’s also where we expect to see the differences.”

Further research is needed, but there doesn’t seem to be any impact in terms of the effect on retention, Dr. Nissen said, “but it’s a little bit harder to interpret for efficacy.”

Confounding by indication was another issued raised, which “is very difficult to control for in this type of study,” Dr. Nissen said.

“The hope is that you have a big enough population” so this doesn’t matter, but as is the nature there were lots of missing data that would have been useful to have and adjust for, such as psoriasis and smoking status. There was also no adjustment for the number of patient visits, as it’s very different from registry to registry.

Dr. Nissen disclosed grant/research support from AbbVie and Novartis. He also acknowledged acting as a consultant or speaker for AbbVie, Celgene, Janssen, Eli Lilly, Novartis, and Pfizer.

Better retention and remission rates with tumor necrosis factor inhibitors (TNFi) have been observed in patients with spondyloarthritis (SpA) who were also treated with a conventional synthetic disease-modifying antirheumatic drug (csDMARD) rather than monotherapy.

Data from the EuroSpA Research Collaboration have shown that 82% of patients who received TNFi and csDMARD cotherapy were still taking their TNFi at 1 year versus 79% of those who were taking the biologic alone (P < .001).

Combination therapy led to remission in 21.8%, compared with TNFi monotherapy at 19.5% (P < .0001), with rates of remission according to Ankylosing Spondylitis Disease Activity Score <1.3 (23.7% vs. 21.8%, P = .011) and Bath Ankylosing Spondylitis Disease Activity Index <2 (5.9% vs. 7.7%, P < .0001) also favoring the cotherapy group.

“We did see a considerable heterogeneity across the countries in this study, both in the use of csDMARD therapy and in the TNF [inhibitor] retention,” Michael J. Nissen, MBBS, FRACP, MD, a rheumatologist at the University of Geneva, said in an oral abstract presentation at the 12th International Congress on Spondyloarthritides.

“Overall, the 1-year TNFi retention was better with cotherapy, particularly in countries in the Scandinavian region and Switzerland,” Dr. Nissen observed.

“The csDMARD cotherapy significantly improved remission; we could perhaps argue whether these small differences are clinically meaningful, but they were highly significant in the study.”
 

Why look at csDMARD and TNFi cotherapy?

European guidelines suggest that csDMARDs are not that useful for treating people with SpA, stating that they should not be used in axial disease and used only as monotherapy in those with peripheral disease.

“Nevertheless, these medications are often used, which begs us to ask the question why is that the case?” Dr. Nissen said.

As previous work by Dr. Nissen and others had suggested that there might be a benefit of combining a csDMARD and a TNFi, the aim was to build upon this and see if looking at a very large patient population might be able to provide some clear answers.

The analysis included data from 13 European registries and more than 24,000 patients who had started their first TNFi between 2006 and 2017. The mean age of the study population was 42.5 years, and 58% were male, with an average disease duration of 5.9 years. Nearly one-third of patients were receiving comedication with a csDMARD, although this ranged from 13.5% to 71.2% depending on the country.

The most frequently used TNFi were adalimumab (31% of patients overall), etanercept (24%), and infliximab (25%), and the most-used csDMARDs were methotrexate (56%) and sulfasalazine (45%).

The effectiveness of treatment in achieving clinical remission was examined according to the TNFi used and for the use of csDMARDs.

“We found that sulfasalazine and methotrexate were similarly effective in terms of improving outcomes in addition to a TNF inhibitor,” Dr. Nissen said. For example, the adjusted odds ratios for using sulfasalazine or methotrexate with infliximab were a respective 1.32 and 1.37, and the aORs for uses these specific csDMARDs with etanercept were 1.38 and 1.35.

The researchers reported finding significantly higher aORs if both sulfasalazine and methotrexate were used with a TNFi (1.67 for any TNFi, 1.95 for infliximab, 1.45 foretanercept, and 1.85 for adalimumab).

This perhaps suggests “there’s some role for combining csDMARDs to have an even greater effect,” Dr. Nissen put forward.
 

A ‘very provocative conclusion’

That’s “a very provocative conclusion” commented Robert Landewé, MD, PhD, of Amsterdam University Medical Center who chaired the sessions. “You can see it by the [number] of people that want to ask questions.”

Désirée van der Heijde, MD, PhD, of Leiden (the Netherlands) University Medical Center, commented these data were “timely as we are updating the ASAS/EULAR [Assessment of SpondyloArthritis international Society/European Alliance of Associations for Rheumatology] recommendations at the moment.” She queried, however, if the differences were clinically significant.

“I think if you have a database of 24,000 patients, highly, statistically significant doesn’t tell me a lot. So, I want to look at the clinical meaning of the differences. So, if you have a 2% difference in retention rate, is that worth prescribing comedication?” Dr. van der Heijde said.

“I think you’re absolutely right,” Dr. Nissen responded. “We’re obviously going to find highly significant differences with such a big patient group.” While the small statistical differences seen may not look very clinically relevant at face value, they could provide some guidance for clinical decision making.

“It may alter our approach if that patient is already on a csDMARD and has peripheral disease; perhaps there’s tendency to then to keep that csDMARD rather than then stopping it and switching to a biologic agent,” Dr. Nissen argued. It’s a topic that has been debated for some time, he added, and further study is needed, but “I think it helps give a little bit more clarity to the idea.”

Another point of discussion was the proportion of patients who had “pure axial disease.” More patients in the comedication group had peripheral disease, Dr. van der Heijde pointed out, “that’s also where we expect to see the differences.”

Further research is needed, but there doesn’t seem to be any impact in terms of the effect on retention, Dr. Nissen said, “but it’s a little bit harder to interpret for efficacy.”

Confounding by indication was another issued raised, which “is very difficult to control for in this type of study,” Dr. Nissen said.

“The hope is that you have a big enough population” so this doesn’t matter, but as is the nature there were lots of missing data that would have been useful to have and adjust for, such as psoriasis and smoking status. There was also no adjustment for the number of patient visits, as it’s very different from registry to registry.

Dr. Nissen disclosed grant/research support from AbbVie and Novartis. He also acknowledged acting as a consultant or speaker for AbbVie, Celgene, Janssen, Eli Lilly, Novartis, and Pfizer.

Better retention and remission rates with tumor necrosis factor inhibitors (TNFi) have been observed in patients with spondyloarthritis (SpA) who were also treated with a conventional synthetic disease-modifying antirheumatic drug (csDMARD) rather than monotherapy.

Data from the EuroSpA Research Collaboration have shown that 82% of patients who received TNFi and csDMARD cotherapy were still taking their TNFi at 1 year versus 79% of those who were taking the biologic alone (P < .001).

Combination therapy led to remission in 21.8%, compared with TNFi monotherapy at 19.5% (P < .0001), with rates of remission according to Ankylosing Spondylitis Disease Activity Score <1.3 (23.7% vs. 21.8%, P = .011) and Bath Ankylosing Spondylitis Disease Activity Index <2 (5.9% vs. 7.7%, P < .0001) also favoring the cotherapy group.

“We did see a considerable heterogeneity across the countries in this study, both in the use of csDMARD therapy and in the TNF [inhibitor] retention,” Michael J. Nissen, MBBS, FRACP, MD, a rheumatologist at the University of Geneva, said in an oral abstract presentation at the 12th International Congress on Spondyloarthritides.

“Overall, the 1-year TNFi retention was better with cotherapy, particularly in countries in the Scandinavian region and Switzerland,” Dr. Nissen observed.

“The csDMARD cotherapy significantly improved remission; we could perhaps argue whether these small differences are clinically meaningful, but they were highly significant in the study.”
 

Why look at csDMARD and TNFi cotherapy?

European guidelines suggest that csDMARDs are not that useful for treating people with SpA, stating that they should not be used in axial disease and used only as monotherapy in those with peripheral disease.

“Nevertheless, these medications are often used, which begs us to ask the question why is that the case?” Dr. Nissen said.

As previous work by Dr. Nissen and others had suggested that there might be a benefit of combining a csDMARD and a TNFi, the aim was to build upon this and see if looking at a very large patient population might be able to provide some clear answers.

The analysis included data from 13 European registries and more than 24,000 patients who had started their first TNFi between 2006 and 2017. The mean age of the study population was 42.5 years, and 58% were male, with an average disease duration of 5.9 years. Nearly one-third of patients were receiving comedication with a csDMARD, although this ranged from 13.5% to 71.2% depending on the country.

The most frequently used TNFi were adalimumab (31% of patients overall), etanercept (24%), and infliximab (25%), and the most-used csDMARDs were methotrexate (56%) and sulfasalazine (45%).

The effectiveness of treatment in achieving clinical remission was examined according to the TNFi used and for the use of csDMARDs.

“We found that sulfasalazine and methotrexate were similarly effective in terms of improving outcomes in addition to a TNF inhibitor,” Dr. Nissen said. For example, the adjusted odds ratios for using sulfasalazine or methotrexate with infliximab were a respective 1.32 and 1.37, and the aORs for uses these specific csDMARDs with etanercept were 1.38 and 1.35.

The researchers reported finding significantly higher aORs if both sulfasalazine and methotrexate were used with a TNFi (1.67 for any TNFi, 1.95 for infliximab, 1.45 foretanercept, and 1.85 for adalimumab).

This perhaps suggests “there’s some role for combining csDMARDs to have an even greater effect,” Dr. Nissen put forward.
 

A ‘very provocative conclusion’

That’s “a very provocative conclusion” commented Robert Landewé, MD, PhD, of Amsterdam University Medical Center who chaired the sessions. “You can see it by the [number] of people that want to ask questions.”

Désirée van der Heijde, MD, PhD, of Leiden (the Netherlands) University Medical Center, commented these data were “timely as we are updating the ASAS/EULAR [Assessment of SpondyloArthritis international Society/European Alliance of Associations for Rheumatology] recommendations at the moment.” She queried, however, if the differences were clinically significant.

“I think if you have a database of 24,000 patients, highly, statistically significant doesn’t tell me a lot. So, I want to look at the clinical meaning of the differences. So, if you have a 2% difference in retention rate, is that worth prescribing comedication?” Dr. van der Heijde said.

“I think you’re absolutely right,” Dr. Nissen responded. “We’re obviously going to find highly significant differences with such a big patient group.” While the small statistical differences seen may not look very clinically relevant at face value, they could provide some guidance for clinical decision making.

“It may alter our approach if that patient is already on a csDMARD and has peripheral disease; perhaps there’s tendency to then to keep that csDMARD rather than then stopping it and switching to a biologic agent,” Dr. Nissen argued. It’s a topic that has been debated for some time, he added, and further study is needed, but “I think it helps give a little bit more clarity to the idea.”

Another point of discussion was the proportion of patients who had “pure axial disease.” More patients in the comedication group had peripheral disease, Dr. van der Heijde pointed out, “that’s also where we expect to see the differences.”

Further research is needed, but there doesn’t seem to be any impact in terms of the effect on retention, Dr. Nissen said, “but it’s a little bit harder to interpret for efficacy.”

Confounding by indication was another issued raised, which “is very difficult to control for in this type of study,” Dr. Nissen said.

“The hope is that you have a big enough population” so this doesn’t matter, but as is the nature there were lots of missing data that would have been useful to have and adjust for, such as psoriasis and smoking status. There was also no adjustment for the number of patient visits, as it’s very different from registry to registry.

Dr. Nissen disclosed grant/research support from AbbVie and Novartis. He also acknowledged acting as a consultant or speaker for AbbVie, Celgene, Janssen, Eli Lilly, Novartis, and Pfizer.

Receiving treatment with a biologic medication, compared with no biologic treatment, appeared to be associated with a lower risk for developing psoriatic arthritis (PsA) in patients with psoriasis.

That’s according to the results of a nested case-control study involving electronic medical record data from an Israeli health maintenance organization in Arthritis & Rheumatology. Compared with no biologic treatment, the risk for developing PsA among PsO patients was reduced by 39%.

This study shows “a statistically and clinically significant lower risk for developing PsA among patients receiving biologic medications for psoriasis treatment,” wrote Yael Shalev Rosenthal, MPH, of the Sackler Faculty of Medicine, Tel Aviv University and colleagues. “The results suggest considering treatment with biologic medications in patients [who] present with significant risk factors for PsA at an earlier stage of treatment.”

“It would be nice to believe this story, but I don’t think we can based on the evidence we’ve got so far,” commented Philip Helliwell, PhD, DM, in an interview.

Dr. Helliwell, who is professor of clinical rheumatology at the University of Leeds (England) and an Honorary Consultant Rheumatologist for the Leeds Teaching Hospitals and Bradford Teaching Hospitals NHS Trust, noted that there were several issues with the current evidence.

Aside from their often retrospective or nonrandomized nature, prior analyses, including the current one, were based on EMR data.

“There’s actually no face-to-face patient contact going on here. It’s all done on coding, and coding can be unreliable,” Dr. Helliwell said.

While the study’s findings are “in line with other studies that have looked at this, and suggest that if you get a biologic, you’re less likely to get PsA with your psoriasis, there could be lots of reasons why.”

The big problem here is confounding by indication. “You don’t get on a biologic unless you’ve got bad psoriasis,” Dr. Helliwell explained. The Israeli criteria for starting a biologic are much higher than in the United Kingdom, he added, requiring more than 50% of patients’ body surface area to be affected, or a Psoriasis Area and Severity Index score of more than 50. Moreover, people with bad psoriasis are more likely to get PsA. This, however, makes the results more impressive.

Confounding by indication is an issue with this study, agreed consultant rheumatologist Adewale Adebajo, PhD, in a separate interview. He acknowledged, however, that the study’s authors did try to account for this by limiting the timescale of their analysis to the first 10 years of biologic therapy. They also used the usual methods of propensity score matching and multivariate Cox regression analysis to hopefully iron out any differences between the two groups of patients.

Study details and results

Ms. Rosenthal and coauthors analyzed EMR data on patients with psoriasis but not PsA that were logged in the Maccabi Healthcare Services (MHS) database. The MHS is the second-largest health maintenance organization in Israel, insuring over 2 million members, the researchers said.

In all, 663 patients with psoriasis but not PsA before or at initiation of biologic treatment were included in their analysis and matched to a control group of 663 patients with psoriasis who had not received biologic treatment. Propensity score matching was used to iron out some differences in baseline characteristics that had been seen between the groups, such as older age at diagnosis, higher body mass index, and a longer time between diagnosis and treatment seen in patients treated without biologics.

After adjusting for multiple risk factors and confounders, “the control group still had a significantly higher risk for PsA, compared to the biological treatment group,” the researchers wrote. Indeed, the adjusted hazard ratio was 1.39, with a 95% confidence interval between 1.03 and 1.87.

An ‘intriguing study’

“This is a retrospective study, and it has all the faults of a retrospective study,” said Dr. Adebajo, associate medical director for research and development at Barnsley (England) NHS Foundation Trust. But “these were patients who hopefully hadn’t yet developed psoriatic arthritis, although it is difficult to exclude subclinical psoriatic arthritis.”

The ideal would of course be to look at patients prospectively, but a randomized clinical trial would be unlikely to ever be conducted, Dr. Helliwell noted. “It would be unfair to randomize people who have got bad psoriasis and need a biologic to placebo just to prove the point really,” he said. “Getting control groups in this arena is very difficult.”

That doesn’t mean that prospective evaluation is not possible. Dr. Adebajo noted that there were already cohorts of newly diagnosed patients who were being prospectively followed up and those could perhaps be used to look at the question again in the future.

“You’re then looking at the natural history, the natural outcome, and you don’t need to worry about confounding because you’re just collecting all of the information as you go along.”

The idea that biologics might slow or even prevent the onset of PsA is “an interesting and enchanting hypothesis,” Dr. Adebajo said. “The study doesn’t prove the hypothesis, but it’s an intriguing study because it doesn’t disprove the hypothesis either.

“It gives us food for thought and a basis for further studies,” as well as some “encouragement to perhaps use biologics earlier because there may be additional benefits of doing so.”

That’s still to be proven of course, as it has been reported that patients with psoriasis can develop PsA while taking biologics.

“Clinically, that’s what we see in the combined clinic. We get people referred with psoriasis [who are] already on a biologic who developed musculoskeletal problems,” Dr. Helliwell said.

“It would be nice to believe” that biologics prevent or slow PsA in patients with psoriasis, Dr. Helliwell added, but I’m not sure these data are conclusive. From this study we know nothing about the phenotype of psoriasis, which is important in the development of PsA. In addition, we know that of the 30% of people with psoriasis who develop PsA, about half of these are undiagnosed at the time of such studies. In that case, what the biologic is doing is just treating preexisting PsA. If you count those numbers up, some of the differences between the two groups seen in this study are accounted for. From registry data there is no way of checking this.”

No external funding was used for the study. One author acknowledged acting as an investigator, adviser, or consultant to several pharmaceutical companies including AbbVie, Boehringer Ingelheim, Bristol-Myers Squibb, Coherus, Dexcel Pharma, Eli Lilly, Janssen, Novartis, and Pfizer. All other authors had nothing to disclose.

Dr. Helliwell and Dr. Adebajo had no conflicts of interest.

Receiving treatment with a biologic medication, compared with no biologic treatment, appeared to be associated with a lower risk for developing psoriatic arthritis (PsA) in patients with psoriasis.

That’s according to the results of a nested case-control study involving electronic medical record data from an Israeli health maintenance organization in Arthritis & Rheumatology. Compared with no biologic treatment, the risk for developing PsA among PsO patients was reduced by 39%.

This study shows “a statistically and clinically significant lower risk for developing PsA among patients receiving biologic medications for psoriasis treatment,” wrote Yael Shalev Rosenthal, MPH, of the Sackler Faculty of Medicine, Tel Aviv University and colleagues. “The results suggest considering treatment with biologic medications in patients [who] present with significant risk factors for PsA at an earlier stage of treatment.”

“It would be nice to believe this story, but I don’t think we can based on the evidence we’ve got so far,” commented Philip Helliwell, PhD, DM, in an interview.

Dr. Helliwell, who is professor of clinical rheumatology at the University of Leeds (England) and an Honorary Consultant Rheumatologist for the Leeds Teaching Hospitals and Bradford Teaching Hospitals NHS Trust, noted that there were several issues with the current evidence.

Aside from their often retrospective or nonrandomized nature, prior analyses, including the current one, were based on EMR data.

“There’s actually no face-to-face patient contact going on here. It’s all done on coding, and coding can be unreliable,” Dr. Helliwell said.

While the study’s findings are “in line with other studies that have looked at this, and suggest that if you get a biologic, you’re less likely to get PsA with your psoriasis, there could be lots of reasons why.”

The big problem here is confounding by indication. “You don’t get on a biologic unless you’ve got bad psoriasis,” Dr. Helliwell explained. The Israeli criteria for starting a biologic are much higher than in the United Kingdom, he added, requiring more than 50% of patients’ body surface area to be affected, or a Psoriasis Area and Severity Index score of more than 50. Moreover, people with bad psoriasis are more likely to get PsA. This, however, makes the results more impressive.

Confounding by indication is an issue with this study, agreed consultant rheumatologist Adewale Adebajo, PhD, in a separate interview. He acknowledged, however, that the study’s authors did try to account for this by limiting the timescale of their analysis to the first 10 years of biologic therapy. They also used the usual methods of propensity score matching and multivariate Cox regression analysis to hopefully iron out any differences between the two groups of patients.

Study details and results

Ms. Rosenthal and coauthors analyzed EMR data on patients with psoriasis but not PsA that were logged in the Maccabi Healthcare Services (MHS) database. The MHS is the second-largest health maintenance organization in Israel, insuring over 2 million members, the researchers said.

In all, 663 patients with psoriasis but not PsA before or at initiation of biologic treatment were included in their analysis and matched to a control group of 663 patients with psoriasis who had not received biologic treatment. Propensity score matching was used to iron out some differences in baseline characteristics that had been seen between the groups, such as older age at diagnosis, higher body mass index, and a longer time between diagnosis and treatment seen in patients treated without biologics.

After adjusting for multiple risk factors and confounders, “the control group still had a significantly higher risk for PsA, compared to the biological treatment group,” the researchers wrote. Indeed, the adjusted hazard ratio was 1.39, with a 95% confidence interval between 1.03 and 1.87.

An ‘intriguing study’

“This is a retrospective study, and it has all the faults of a retrospective study,” said Dr. Adebajo, associate medical director for research and development at Barnsley (England) NHS Foundation Trust. But “these were patients who hopefully hadn’t yet developed psoriatic arthritis, although it is difficult to exclude subclinical psoriatic arthritis.”

The ideal would of course be to look at patients prospectively, but a randomized clinical trial would be unlikely to ever be conducted, Dr. Helliwell noted. “It would be unfair to randomize people who have got bad psoriasis and need a biologic to placebo just to prove the point really,” he said. “Getting control groups in this arena is very difficult.”

That doesn’t mean that prospective evaluation is not possible. Dr. Adebajo noted that there were already cohorts of newly diagnosed patients who were being prospectively followed up and those could perhaps be used to look at the question again in the future.

“You’re then looking at the natural history, the natural outcome, and you don’t need to worry about confounding because you’re just collecting all of the information as you go along.”

The idea that biologics might slow or even prevent the onset of PsA is “an interesting and enchanting hypothesis,” Dr. Adebajo said. “The study doesn’t prove the hypothesis, but it’s an intriguing study because it doesn’t disprove the hypothesis either.

“It gives us food for thought and a basis for further studies,” as well as some “encouragement to perhaps use biologics earlier because there may be additional benefits of doing so.”

That’s still to be proven of course, as it has been reported that patients with psoriasis can develop PsA while taking biologics.

“Clinically, that’s what we see in the combined clinic. We get people referred with psoriasis [who are] already on a biologic who developed musculoskeletal problems,” Dr. Helliwell said.

“It would be nice to believe” that biologics prevent or slow PsA in patients with psoriasis, Dr. Helliwell added, but I’m not sure these data are conclusive. From this study we know nothing about the phenotype of psoriasis, which is important in the development of PsA. In addition, we know that of the 30% of people with psoriasis who develop PsA, about half of these are undiagnosed at the time of such studies. In that case, what the biologic is doing is just treating preexisting PsA. If you count those numbers up, some of the differences between the two groups seen in this study are accounted for. From registry data there is no way of checking this.”

No external funding was used for the study. One author acknowledged acting as an investigator, adviser, or consultant to several pharmaceutical companies including AbbVie, Boehringer Ingelheim, Bristol-Myers Squibb, Coherus, Dexcel Pharma, Eli Lilly, Janssen, Novartis, and Pfizer. All other authors had nothing to disclose.

Dr. Helliwell and Dr. Adebajo had no conflicts of interest.

Receiving treatment with a biologic medication, compared with no biologic treatment, appeared to be associated with a lower risk for developing psoriatic arthritis (PsA) in patients with psoriasis.

That’s according to the results of a nested case-control study involving electronic medical record data from an Israeli health maintenance organization in Arthritis & Rheumatology. Compared with no biologic treatment, the risk for developing PsA among PsO patients was reduced by 39%.

This study shows “a statistically and clinically significant lower risk for developing PsA among patients receiving biologic medications for psoriasis treatment,” wrote Yael Shalev Rosenthal, MPH, of the Sackler Faculty of Medicine, Tel Aviv University and colleagues. “The results suggest considering treatment with biologic medications in patients [who] present with significant risk factors for PsA at an earlier stage of treatment.”

“It would be nice to believe this story, but I don’t think we can based on the evidence we’ve got so far,” commented Philip Helliwell, PhD, DM, in an interview.

Dr. Helliwell, who is professor of clinical rheumatology at the University of Leeds (England) and an Honorary Consultant Rheumatologist for the Leeds Teaching Hospitals and Bradford Teaching Hospitals NHS Trust, noted that there were several issues with the current evidence.

Aside from their often retrospective or nonrandomized nature, prior analyses, including the current one, were based on EMR data.

“There’s actually no face-to-face patient contact going on here. It’s all done on coding, and coding can be unreliable,” Dr. Helliwell said.

While the study’s findings are “in line with other studies that have looked at this, and suggest that if you get a biologic, you’re less likely to get PsA with your psoriasis, there could be lots of reasons why.”

The big problem here is confounding by indication. “You don’t get on a biologic unless you’ve got bad psoriasis,” Dr. Helliwell explained. The Israeli criteria for starting a biologic are much higher than in the United Kingdom, he added, requiring more than 50% of patients’ body surface area to be affected, or a Psoriasis Area and Severity Index score of more than 50. Moreover, people with bad psoriasis are more likely to get PsA. This, however, makes the results more impressive.

Confounding by indication is an issue with this study, agreed consultant rheumatologist Adewale Adebajo, PhD, in a separate interview. He acknowledged, however, that the study’s authors did try to account for this by limiting the timescale of their analysis to the first 10 years of biologic therapy. They also used the usual methods of propensity score matching and multivariate Cox regression analysis to hopefully iron out any differences between the two groups of patients.

Study details and results

Ms. Rosenthal and coauthors analyzed EMR data on patients with psoriasis but not PsA that were logged in the Maccabi Healthcare Services (MHS) database. The MHS is the second-largest health maintenance organization in Israel, insuring over 2 million members, the researchers said.

In all, 663 patients with psoriasis but not PsA before or at initiation of biologic treatment were included in their analysis and matched to a control group of 663 patients with psoriasis who had not received biologic treatment. Propensity score matching was used to iron out some differences in baseline characteristics that had been seen between the groups, such as older age at diagnosis, higher body mass index, and a longer time between diagnosis and treatment seen in patients treated without biologics.

After adjusting for multiple risk factors and confounders, “the control group still had a significantly higher risk for PsA, compared to the biological treatment group,” the researchers wrote. Indeed, the adjusted hazard ratio was 1.39, with a 95% confidence interval between 1.03 and 1.87.

An ‘intriguing study’

“This is a retrospective study, and it has all the faults of a retrospective study,” said Dr. Adebajo, associate medical director for research and development at Barnsley (England) NHS Foundation Trust. But “these were patients who hopefully hadn’t yet developed psoriatic arthritis, although it is difficult to exclude subclinical psoriatic arthritis.”

The ideal would of course be to look at patients prospectively, but a randomized clinical trial would be unlikely to ever be conducted, Dr. Helliwell noted. “It would be unfair to randomize people who have got bad psoriasis and need a biologic to placebo just to prove the point really,” he said. “Getting control groups in this arena is very difficult.”

That doesn’t mean that prospective evaluation is not possible. Dr. Adebajo noted that there were already cohorts of newly diagnosed patients who were being prospectively followed up and those could perhaps be used to look at the question again in the future.

“You’re then looking at the natural history, the natural outcome, and you don’t need to worry about confounding because you’re just collecting all of the information as you go along.”

The idea that biologics might slow or even prevent the onset of PsA is “an interesting and enchanting hypothesis,” Dr. Adebajo said. “The study doesn’t prove the hypothesis, but it’s an intriguing study because it doesn’t disprove the hypothesis either.

“It gives us food for thought and a basis for further studies,” as well as some “encouragement to perhaps use biologics earlier because there may be additional benefits of doing so.”

That’s still to be proven of course, as it has been reported that patients with psoriasis can develop PsA while taking biologics.

“Clinically, that’s what we see in the combined clinic. We get people referred with psoriasis [who are] already on a biologic who developed musculoskeletal problems,” Dr. Helliwell said.

“It would be nice to believe” that biologics prevent or slow PsA in patients with psoriasis, Dr. Helliwell added, but I’m not sure these data are conclusive. From this study we know nothing about the phenotype of psoriasis, which is important in the development of PsA. In addition, we know that of the 30% of people with psoriasis who develop PsA, about half of these are undiagnosed at the time of such studies. In that case, what the biologic is doing is just treating preexisting PsA. If you count those numbers up, some of the differences between the two groups seen in this study are accounted for. From registry data there is no way of checking this.”

No external funding was used for the study. One author acknowledged acting as an investigator, adviser, or consultant to several pharmaceutical companies including AbbVie, Boehringer Ingelheim, Bristol-Myers Squibb, Coherus, Dexcel Pharma, Eli Lilly, Janssen, Novartis, and Pfizer. All other authors had nothing to disclose.

Dr. Helliwell and Dr. Adebajo had no conflicts of interest.

Patients with rheumatoid arthritis currently being treated with low doses of oral glucocorticoids (GCs) had a 59% increased risk of sustaining a vertebral fracture when compared with past users, results of a retrospective cohort study have shown.

izusek/Getty Images

Although the overall risk of an osteoporotic fracture was not increased when comparing current and past GC users, with a hazard ratio of 1.14 (95% confidence interval, 0.98-1.33), the HR for sustaining a spinal fracture was 1.59 (95% CI, 1.11-2.29).

“Clinicians should be aware that, even in RA patients who receive low daily glucocorticoid doses, the risk of clinical vertebral fracture is increased,” Shahab Abtahi, MD, of Maastricht (the Netherlands) University and coauthors reported in Rheumatology.

This is important considering around a quarter of RA patients are treated with GCs in the United Kingdom in accordance with European recommendations, they observed.

Conflicting randomized and observational findings on whether or not osteoporotic fractures might be linked to the use of low-dose GCs prompted Dr. Abtahi and associates to see if there were any signals in real-world data. To do so, they used data one of the world’s largest primary care databases – the Clinical Practice Research Datalink (CPRD), which consists of anonymized patient data from a network of primary care practices across the United Kingdom.

Altogether, the records of more than 15,000 patients with RA aged 50 years and older who were held in the CRPD between 1997 and 2017 were pulled for analysis, and just half (n = 7,039) were receiving or had received GC therapy. Low-dose GC therapy was defined as a prednisolone equivalent dose (PED) of 7.5 mg or less per day.

The use of low-dose GCs use during three key time periods was considered: within the past 6 months (current users), within the past 7-12 months (recent users), and within the past year (past users).

The analyses involved time-dependent Cox proportional-hazards models to look for associations between GC use and all types of osteoporotic fracture, including the risk for incident hip, vertebral, humeral, forearm, pelvis, and rib fractures. They were adjusted for various lifestyle parameters, comorbidities, and the use of other medications.

“Current GC use was further broken down into subcategories based on average daily and cumulative dose,” Dr. Abtahi observed. As might be expected, doses even lower than 7.5 mg or less PED did not increase the chance of any osteoporotic fracture but there was an increased risk for some types with higher average daily doses, notably at the hip and pelvis, as well as the spine.

“Low-dose oral GC therapy was associated with an increased risk of clinical vertebral fracture, while the risk of other individual OP fracture sites was not increased,” said the team, adding that the main results remained unchanged regardless of short- or long-term use.

“We know that vertebral fracture risk is markedly increased in RA, and it is well known that GC therapy in particular affects trabecular bone, which is abundantly present in lumbar vertebrae,” Dr. Abtahi wrote.

“Therefore, we can hypothesize that the beneficial effect of low-dose GC therapy on suppressing the background inflammation of RA could probably be enough to offset its negative effect on bone synthesis in most fracture sites but not in vertebrae,” they suggested.

One of the limitations of the study is that the researchers lacked data on the disease activity of the patients or if they were being treated with biologic therapy. This means that confounding by disease severity might be an issue with only those with higher disease activity being treated with GCs and thus were at higher risk for fractures.

“Another limitation was a potential misclassification of exposure with oral GCs, as we had only prescribing information from CPRD, which is roughly two steps behind actual drug use by patients,” the researchers conceded. The average duration of GC use was estimated at 3.7 years, which is an indication of actual use.

A detection bias may also be involved with regard to vertebral fractures, with complaints of back pain maybe being discussed more often when prescribing GCs, leading to more referrals for possible fracture assessment.

Dr. Abtahi and a fellow coauthor disclosed receiving research and other funding from several pharmaceutical companies unrelated to this study. All other coauthors had no conflicts of interest.

Patients with rheumatoid arthritis currently being treated with low doses of oral glucocorticoids (GCs) had a 59% increased risk of sustaining a vertebral fracture when compared with past users, results of a retrospective cohort study have shown.

izusek/Getty Images

Although the overall risk of an osteoporotic fracture was not increased when comparing current and past GC users, with a hazard ratio of 1.14 (95% confidence interval, 0.98-1.33), the HR for sustaining a spinal fracture was 1.59 (95% CI, 1.11-2.29).

“Clinicians should be aware that, even in RA patients who receive low daily glucocorticoid doses, the risk of clinical vertebral fracture is increased,” Shahab Abtahi, MD, of Maastricht (the Netherlands) University and coauthors reported in Rheumatology.

This is important considering around a quarter of RA patients are treated with GCs in the United Kingdom in accordance with European recommendations, they observed.

Conflicting randomized and observational findings on whether or not osteoporotic fractures might be linked to the use of low-dose GCs prompted Dr. Abtahi and associates to see if there were any signals in real-world data. To do so, they used data one of the world’s largest primary care databases – the Clinical Practice Research Datalink (CPRD), which consists of anonymized patient data from a network of primary care practices across the United Kingdom.

Altogether, the records of more than 15,000 patients with RA aged 50 years and older who were held in the CRPD between 1997 and 2017 were pulled for analysis, and just half (n = 7,039) were receiving or had received GC therapy. Low-dose GC therapy was defined as a prednisolone equivalent dose (PED) of 7.5 mg or less per day.

The use of low-dose GCs use during three key time periods was considered: within the past 6 months (current users), within the past 7-12 months (recent users), and within the past year (past users).

The analyses involved time-dependent Cox proportional-hazards models to look for associations between GC use and all types of osteoporotic fracture, including the risk for incident hip, vertebral, humeral, forearm, pelvis, and rib fractures. They were adjusted for various lifestyle parameters, comorbidities, and the use of other medications.

“Current GC use was further broken down into subcategories based on average daily and cumulative dose,” Dr. Abtahi observed. As might be expected, doses even lower than 7.5 mg or less PED did not increase the chance of any osteoporotic fracture but there was an increased risk for some types with higher average daily doses, notably at the hip and pelvis, as well as the spine.

“Low-dose oral GC therapy was associated with an increased risk of clinical vertebral fracture, while the risk of other individual OP fracture sites was not increased,” said the team, adding that the main results remained unchanged regardless of short- or long-term use.

“We know that vertebral fracture risk is markedly increased in RA, and it is well known that GC therapy in particular affects trabecular bone, which is abundantly present in lumbar vertebrae,” Dr. Abtahi wrote.

“Therefore, we can hypothesize that the beneficial effect of low-dose GC therapy on suppressing the background inflammation of RA could probably be enough to offset its negative effect on bone synthesis in most fracture sites but not in vertebrae,” they suggested.

One of the limitations of the study is that the researchers lacked data on the disease activity of the patients or if they were being treated with biologic therapy. This means that confounding by disease severity might be an issue with only those with higher disease activity being treated with GCs and thus were at higher risk for fractures.

“Another limitation was a potential misclassification of exposure with oral GCs, as we had only prescribing information from CPRD, which is roughly two steps behind actual drug use by patients,” the researchers conceded. The average duration of GC use was estimated at 3.7 years, which is an indication of actual use.

A detection bias may also be involved with regard to vertebral fractures, with complaints of back pain maybe being discussed more often when prescribing GCs, leading to more referrals for possible fracture assessment.

Dr. Abtahi and a fellow coauthor disclosed receiving research and other funding from several pharmaceutical companies unrelated to this study. All other coauthors had no conflicts of interest.

Patients with rheumatoid arthritis currently being treated with low doses of oral glucocorticoids (GCs) had a 59% increased risk of sustaining a vertebral fracture when compared with past users, results of a retrospective cohort study have shown.

izusek/Getty Images

Although the overall risk of an osteoporotic fracture was not increased when comparing current and past GC users, with a hazard ratio of 1.14 (95% confidence interval, 0.98-1.33), the HR for sustaining a spinal fracture was 1.59 (95% CI, 1.11-2.29).

“Clinicians should be aware that, even in RA patients who receive low daily glucocorticoid doses, the risk of clinical vertebral fracture is increased,” Shahab Abtahi, MD, of Maastricht (the Netherlands) University and coauthors reported in Rheumatology.

This is important considering around a quarter of RA patients are treated with GCs in the United Kingdom in accordance with European recommendations, they observed.

Conflicting randomized and observational findings on whether or not osteoporotic fractures might be linked to the use of low-dose GCs prompted Dr. Abtahi and associates to see if there were any signals in real-world data. To do so, they used data one of the world’s largest primary care databases – the Clinical Practice Research Datalink (CPRD), which consists of anonymized patient data from a network of primary care practices across the United Kingdom.

Altogether, the records of more than 15,000 patients with RA aged 50 years and older who were held in the CRPD between 1997 and 2017 were pulled for analysis, and just half (n = 7,039) were receiving or had received GC therapy. Low-dose GC therapy was defined as a prednisolone equivalent dose (PED) of 7.5 mg or less per day.

The use of low-dose GCs use during three key time periods was considered: within the past 6 months (current users), within the past 7-12 months (recent users), and within the past year (past users).

The analyses involved time-dependent Cox proportional-hazards models to look for associations between GC use and all types of osteoporotic fracture, including the risk for incident hip, vertebral, humeral, forearm, pelvis, and rib fractures. They were adjusted for various lifestyle parameters, comorbidities, and the use of other medications.

“Current GC use was further broken down into subcategories based on average daily and cumulative dose,” Dr. Abtahi observed. As might be expected, doses even lower than 7.5 mg or less PED did not increase the chance of any osteoporotic fracture but there was an increased risk for some types with higher average daily doses, notably at the hip and pelvis, as well as the spine.

“Low-dose oral GC therapy was associated with an increased risk of clinical vertebral fracture, while the risk of other individual OP fracture sites was not increased,” said the team, adding that the main results remained unchanged regardless of short- or long-term use.

“We know that vertebral fracture risk is markedly increased in RA, and it is well known that GC therapy in particular affects trabecular bone, which is abundantly present in lumbar vertebrae,” Dr. Abtahi wrote.

“Therefore, we can hypothesize that the beneficial effect of low-dose GC therapy on suppressing the background inflammation of RA could probably be enough to offset its negative effect on bone synthesis in most fracture sites but not in vertebrae,” they suggested.

One of the limitations of the study is that the researchers lacked data on the disease activity of the patients or if they were being treated with biologic therapy. This means that confounding by disease severity might be an issue with only those with higher disease activity being treated with GCs and thus were at higher risk for fractures.

“Another limitation was a potential misclassification of exposure with oral GCs, as we had only prescribing information from CPRD, which is roughly two steps behind actual drug use by patients,” the researchers conceded. The average duration of GC use was estimated at 3.7 years, which is an indication of actual use.

A detection bias may also be involved with regard to vertebral fractures, with complaints of back pain maybe being discussed more often when prescribing GCs, leading to more referrals for possible fracture assessment.

Dr. Abtahi and a fellow coauthor disclosed receiving research and other funding from several pharmaceutical companies unrelated to this study. All other coauthors had no conflicts of interest.

The overactive bladder treatment mirabegron (Myrbetriq) is one of three new drugs to be linked to cases of drug-associated antineutrophil cytoplasmic antibody–associated vasculitis (DA-AAV) according to pharmacovigilance data.

Artfoliophoto/Thinkstock

The other two drugs are the antiviral agent sofosbuvir (Sovaldi) and the tyrosine kinase inhibitor nintedanib (Ofev), which is indicated for chronic fibrosing interstitial lung diseases (ILDs) with a progressive phenotype.

The study findings also strengthen previously suspected associations, Samuel Deshayes, MD, of Caen (France) University Hospital and coauthors report in Arthritis & Rheumatology. This includes associations between hydralazine, propylthiouracil, thiamazole, minocycline, and carbimazole to name a few more of the 15 drugs where associations were found.

“These data imply specific prescriber attention to these drugs and argue for experimental studies on AAV pathophysiology with [the] recent drugs,” Dr. Deshayes and colleagues wrote.

Identifying whether AAV could be due to drug treatment is important as there have been suggestions that this may confer a better prognosis than for primary AAV. In addition, “withdrawal of the culprit drug” may be enough to treat the patient in milder cases, the researchers suggested.

That said, the researchers acknowledged that it can be difficult to attribute AAV to a specific drug, particularly as data are often from case reports or small series of patients and, until now, research had considered older medications.

To update current knowledge and include newer treatments, Dr. Deshayes and associates took data from the World Health Organization pharmacovigilance database (VigiBase) and performed a retrospective analysis.

They collected data on all adverse drug reactions (ADRs) reported using the Medical Dictionary for Drug Regulatory Activities preferred term ‘anti-neutrophil’ cytoplasmic antibody positive vasculitis’ from its introduction in 2006 up until November 2020.

Drugs used in the management of AAV, such as azathioprine, cyclophosphamide, leflunomide, and methotrexate, among others, were not included in the analysis “due to potential indication bias,” they explained.

Among 483 individual “deduplicated” case safety reports, most cases of DA-AAV were reported in women (71.2% of cases) and the median age at onset was 62 years. The median time to develop AAV after drug initiation was 9 months.

DA-AAV was considered serious in almost all (98%) of the reported cases, meaning that it was life-threatening, had resulted in or had prolonged hospital treatment, caused significant or persistent disability or other anomalies such as birth defects, or resulted in the death of the individual. With regard to the latter, DA-AAV was reported as fatal in 8.9% of cases.

The highest rates of reported DA-AAV were seen in people treated with the antihypertensive drug hydralazine, the antithyroid drugs propylthiouracil, thiamazole, and carbimazole, the antibiotic minocycline, mirabegron, sofosbuvir, and nintedanib.

A further seven drugs were also linked to cases of AAV, bringing the total to list to 15 drugs associated with overreporting of AAV. This included the heavy metal antagonist penicillamine, the influenza vaccine, the gout medication allopurinol, the anti-tuberculosis drug rifampin, the antiviral agent peginterferon alfa-2b, the anti-asthma drug montelukast, and the anti-cholesterol agent rosuvastatin.

Dr. Deshayes and coauthors pointed out that the associations do not imply causality. The data mining approach used allowed for the “automated detection of drug safety signals associated with rare events such as AAV,” they said.

There are limitations, they acknowledged, which are common to studies that use pharmacovigilance data. For one, “notoriety bias” could be at play. This is where there is an “increased number of reports after safety alerts in the media.” Missing data, “including delay between the introduction of the drug and AAV,” could be a factor, as could “bias related to its retrospective and declarative design.”

The study received no commercial funding. All study authors declared no competing interests.

The overactive bladder treatment mirabegron (Myrbetriq) is one of three new drugs to be linked to cases of drug-associated antineutrophil cytoplasmic antibody–associated vasculitis (DA-AAV) according to pharmacovigilance data.

Artfoliophoto/Thinkstock

The other two drugs are the antiviral agent sofosbuvir (Sovaldi) and the tyrosine kinase inhibitor nintedanib (Ofev), which is indicated for chronic fibrosing interstitial lung diseases (ILDs) with a progressive phenotype.

The study findings also strengthen previously suspected associations, Samuel Deshayes, MD, of Caen (France) University Hospital and coauthors report in Arthritis & Rheumatology. This includes associations between hydralazine, propylthiouracil, thiamazole, minocycline, and carbimazole to name a few more of the 15 drugs where associations were found.

“These data imply specific prescriber attention to these drugs and argue for experimental studies on AAV pathophysiology with [the] recent drugs,” Dr. Deshayes and colleagues wrote.

Identifying whether AAV could be due to drug treatment is important as there have been suggestions that this may confer a better prognosis than for primary AAV. In addition, “withdrawal of the culprit drug” may be enough to treat the patient in milder cases, the researchers suggested.

That said, the researchers acknowledged that it can be difficult to attribute AAV to a specific drug, particularly as data are often from case reports or small series of patients and, until now, research had considered older medications.

To update current knowledge and include newer treatments, Dr. Deshayes and associates took data from the World Health Organization pharmacovigilance database (VigiBase) and performed a retrospective analysis.

They collected data on all adverse drug reactions (ADRs) reported using the Medical Dictionary for Drug Regulatory Activities preferred term ‘anti-neutrophil’ cytoplasmic antibody positive vasculitis’ from its introduction in 2006 up until November 2020.

Drugs used in the management of AAV, such as azathioprine, cyclophosphamide, leflunomide, and methotrexate, among others, were not included in the analysis “due to potential indication bias,” they explained.

Among 483 individual “deduplicated” case safety reports, most cases of DA-AAV were reported in women (71.2% of cases) and the median age at onset was 62 years. The median time to develop AAV after drug initiation was 9 months.

DA-AAV was considered serious in almost all (98%) of the reported cases, meaning that it was life-threatening, had resulted in or had prolonged hospital treatment, caused significant or persistent disability or other anomalies such as birth defects, or resulted in the death of the individual. With regard to the latter, DA-AAV was reported as fatal in 8.9% of cases.

The highest rates of reported DA-AAV were seen in people treated with the antihypertensive drug hydralazine, the antithyroid drugs propylthiouracil, thiamazole, and carbimazole, the antibiotic minocycline, mirabegron, sofosbuvir, and nintedanib.

A further seven drugs were also linked to cases of AAV, bringing the total to list to 15 drugs associated with overreporting of AAV. This included the heavy metal antagonist penicillamine, the influenza vaccine, the gout medication allopurinol, the anti-tuberculosis drug rifampin, the antiviral agent peginterferon alfa-2b, the anti-asthma drug montelukast, and the anti-cholesterol agent rosuvastatin.

Dr. Deshayes and coauthors pointed out that the associations do not imply causality. The data mining approach used allowed for the “automated detection of drug safety signals associated with rare events such as AAV,” they said.

There are limitations, they acknowledged, which are common to studies that use pharmacovigilance data. For one, “notoriety bias” could be at play. This is where there is an “increased number of reports after safety alerts in the media.” Missing data, “including delay between the introduction of the drug and AAV,” could be a factor, as could “bias related to its retrospective and declarative design.”

The study received no commercial funding. All study authors declared no competing interests.

The overactive bladder treatment mirabegron (Myrbetriq) is one of three new drugs to be linked to cases of drug-associated antineutrophil cytoplasmic antibody–associated vasculitis (DA-AAV) according to pharmacovigilance data.

Artfoliophoto/Thinkstock

The other two drugs are the antiviral agent sofosbuvir (Sovaldi) and the tyrosine kinase inhibitor nintedanib (Ofev), which is indicated for chronic fibrosing interstitial lung diseases (ILDs) with a progressive phenotype.

The study findings also strengthen previously suspected associations, Samuel Deshayes, MD, of Caen (France) University Hospital and coauthors report in Arthritis & Rheumatology. This includes associations between hydralazine, propylthiouracil, thiamazole, minocycline, and carbimazole to name a few more of the 15 drugs where associations were found.

“These data imply specific prescriber attention to these drugs and argue for experimental studies on AAV pathophysiology with [the] recent drugs,” Dr. Deshayes and colleagues wrote.

Identifying whether AAV could be due to drug treatment is important as there have been suggestions that this may confer a better prognosis than for primary AAV. In addition, “withdrawal of the culprit drug” may be enough to treat the patient in milder cases, the researchers suggested.

That said, the researchers acknowledged that it can be difficult to attribute AAV to a specific drug, particularly as data are often from case reports or small series of patients and, until now, research had considered older medications.

To update current knowledge and include newer treatments, Dr. Deshayes and associates took data from the World Health Organization pharmacovigilance database (VigiBase) and performed a retrospective analysis.

They collected data on all adverse drug reactions (ADRs) reported using the Medical Dictionary for Drug Regulatory Activities preferred term ‘anti-neutrophil’ cytoplasmic antibody positive vasculitis’ from its introduction in 2006 up until November 2020.

Drugs used in the management of AAV, such as azathioprine, cyclophosphamide, leflunomide, and methotrexate, among others, were not included in the analysis “due to potential indication bias,” they explained.

Among 483 individual “deduplicated” case safety reports, most cases of DA-AAV were reported in women (71.2% of cases) and the median age at onset was 62 years. The median time to develop AAV after drug initiation was 9 months.

DA-AAV was considered serious in almost all (98%) of the reported cases, meaning that it was life-threatening, had resulted in or had prolonged hospital treatment, caused significant or persistent disability or other anomalies such as birth defects, or resulted in the death of the individual. With regard to the latter, DA-AAV was reported as fatal in 8.9% of cases.

The highest rates of reported DA-AAV were seen in people treated with the antihypertensive drug hydralazine, the antithyroid drugs propylthiouracil, thiamazole, and carbimazole, the antibiotic minocycline, mirabegron, sofosbuvir, and nintedanib.

A further seven drugs were also linked to cases of AAV, bringing the total to list to 15 drugs associated with overreporting of AAV. This included the heavy metal antagonist penicillamine, the influenza vaccine, the gout medication allopurinol, the anti-tuberculosis drug rifampin, the antiviral agent peginterferon alfa-2b, the anti-asthma drug montelukast, and the anti-cholesterol agent rosuvastatin.

Dr. Deshayes and coauthors pointed out that the associations do not imply causality. The data mining approach used allowed for the “automated detection of drug safety signals associated with rare events such as AAV,” they said.

There are limitations, they acknowledged, which are common to studies that use pharmacovigilance data. For one, “notoriety bias” could be at play. This is where there is an “increased number of reports after safety alerts in the media.” Missing data, “including delay between the introduction of the drug and AAV,” could be a factor, as could “bias related to its retrospective and declarative design.”

The study received no commercial funding. All study authors declared no competing interests.

Tumor necrosis factor inhibitors reduce both spinal and sacroiliac radiographic progression in early axial spondyloarthritis (axSpA), but the effects take time.

According to data from the German Spondyloarthritis Inception Cohort (GESPIC), it could take 2-4 years after the initiation of TNF inhibitor treatment before x-rays will show any sign of a benefit in either the spine or sacroiliac joints (SIJ).

“The association between anti-TNF treatment and the retardation of radiographic spinal progression in spondyloarthritis appears to be time-shifted,” GESPIC study investigator Denis Poddubnyy, MD, said at the annual European Congress of Rheumatology.

Reporting the spinal findings, Dr. Poddubnyy, who is head of the rheumatology department at Campus Benjamin Franklin of the Charité University Hospital in Berlin noted that “at least 4 years of observation is required” to see an effect of TNF inhibitors on new bone formation. “This might be related to the intermediate and long-lasting process of bone repair that follows the resolution of inflammation and precedes the development of new bone,” he suggested.

The effect on the SIJ may precede the spinal damage, so it’s important to look at how anti-inflammatory treatment might affect the SIJ, Dr. Poddubnyy said in an interview. Results from the SIJ analysis, which were presented by Murat Torgutalp, MD, suggested that there might be a 2-year time lag between TNFi treatment and being able to see something on x-rays.

“Everybody in the past focused on the spine, considering that spinal changes are functionally more relevant than changes in the SIJ,” Dr. Poddubnyy observed.

Structural damage in the spine and SIJ in radiographic axSpA probably occur by different processes, he explained. While it’s more to do with new bone formation in the spine, it’s more to do with erosive damage followed by repair in the SIJ. “It is very likely that in the majority of patients we see that those two processes are somehow disconnected in time,” he suggested.

The GESPIC cohort provides a good base from which to examine the effects of TNF inhibitors on radiographic progression in patients with axSpA for several reasons, observed Dr. Poddubnyy. For one it was established in 2000, so “well before the introduction of TNF-alpha inhibitors.” That means that almost all patients would have been put on TNF inhibitors while enrolled in the cohort, he said, “so we could really follow up them prospectively for a few years.” In fact, just nine patients had been treated with TNF inhibitors at baseline.

Spinal findings

Although TNF inhibitors are widely accepted to be effective anti-inflammatory drugs that can relieve patients’ spinal symptoms, there were inconclusive data on whether they can also have an impact on radiographic progression.

To look at the possible immediate or later effect of TNF inhibitors on spinal radiographic progression, the GESPIC investigators looked at sets of radiographs, taken at least 2 years apart over a 10-year period, from 266 patients, 77 of whom who had been treated with a TNF inhibitor for at least 12 months or more. Overall, there were 103 2-year intervals covered by TNF inhibitor treatment of any duration, and 78 intervals in which treatment had lasted for at least 12 months.

“This cohort included patients with quite early disease, including both nonradiographic and radiographic forms,” Dr. Poddubnyy said, adding that some patients had been treated with a TNF inhibitor for up to 10 years.

While there were no significant changes in radiographic progression as measured using the modified Stoke Ankylosing Spondylitis Spine Score (mSASSS) between patients who had and who had not received at least 12 months’ TNF inhibitor therapy in the first 2 years, there was a substantial difference after 4 years.

The mSASSS change in scores over 2 years in patients who were and were not treated with a TNF inhibitor for 12 months or more were a respective 0.35 and 0.81 (P = .047) considering treatment in the previous 2-year treatment period, and 0.33 and 0.77 (also P = .047) considering treatment in the prior 2-year treatment period and the current 2-year period.

“Our interpretation is that within the first 2 years we see a resolution of inflammation, a process of repair, maybe even a process of bone formation, but then if you continue this inflammation control for longer than 2 years, then we see the effect on structural damage,” Dr. Poddubnyy said.

SIJ findings

The SIJ analysis included 301 patients who had at least two sets of radiographs taken every 2 years over a 10-year follow-up period, 87 (28.9%) of whom had been treated with at least one TNF inhibitor during follow-up.

Radiographic SIJ progression was defined as the change in the sacroiliitis sum score over 2 years. Calculations considered both the current 2-year period and the previous 2-year period, as well as any use of TNF inhibitors or more than 12 months’ use.

There were no significant differences in sacroiliitis sum scores comparing people who were and were not treated with TNF inhibitors for at least 12 months in the current 2-year period. However, when the previous 2-year period was looked at there was a significant difference in the score between those who had and had not been treated for 12 or more months (0.15 vs. 0.27, respectively; P = .024).

“These data indicate that TNF inhibitors have a late effect,” Dr. Torgutalp said. “You cannot observe the effect of the TNF inhibitor on radiographic progression in the same interval, you have to wait to see this effect in the next interval,” he explained. That means that when a patient uses a TNF inhibitor, you must wait to see the effect on SIJ progression in the subsequent 2 years, he qualified.

Lessons for practice

“If we control inflammation, we might control structural damage progression,” Dr. Poddubnyy said.

If this is the case, then it is likely that other drugs that control inflammation in axSpA will have a similar effect, such as those that target interleukin-17 or Janus kinases. This is something in the future to look at in the GESPIC cohort once enough patient-years of data are available.

Early treatment is probably important, Dr. Poddubnyy suggested, “because in patients with early disease we can expect not that much structural damage at baseline.” Starting treatment early, at the inflammatory stage rather than at the stage where there was a lot of damage already there, could yield better results, he proposed.

“As a clinician, as a researcher, I would be really interested in demonstrating in a prospective, clinical trial that treating patients to target, achieving a remission, would be associated with retardation of structural damage progression. This has only been shown in observational studies so far, so this is an open question.”

Questions on adjustments made for confounding factors

Questions on the spinal presentation included whether there were any differences in NSAID use, to which Dr. Poddubnyy responded, “Patients on TNF inhibitors normally reduce NSAIDs. We adjusted, therefore, for NSAID use.”

The SIJ presentation garnered more questioning. Dr. Torgutalp qualified that both radiographic and nonradiographic patients had been included in the analysis, and “we observe the same effect.”

Asked how clinically relevant the findings were, he noted that, “to the best of our knowledge, this is the first data for sacroiliac joint progression.” The data for spinal progression seem to “reconcile the results from previous investigators,” where there had been some conflicting data on whether 2 or 4 years’ TNF-inhibitor therapy had any effect on radiographic progression.

“I don’t think you’ve adjusted for the baseline level of damage of the SIJ?” queried Pedro Machado, MD, of University College London Hospital. “As you know, in the spine, but also at the SIJ level, the baseline level of damage is usually one of the strongest predictors of further damage.”

This is not something that was adjusted for, Dr. Torgutalp conceded. “We will do it for further analysis,” he said.

Since 2010, GESPIC has been supported by grants from AbbVie. Dr. Poddubnyy reported financial relationships with AbbVie and numerous other pharmaceutical companies. Dr. Torgutalp had no relevant disclosures.

Tumor necrosis factor inhibitors reduce both spinal and sacroiliac radiographic progression in early axial spondyloarthritis (axSpA), but the effects take time.

According to data from the German Spondyloarthritis Inception Cohort (GESPIC), it could take 2-4 years after the initiation of TNF inhibitor treatment before x-rays will show any sign of a benefit in either the spine or sacroiliac joints (SIJ).

“The association between anti-TNF treatment and the retardation of radiographic spinal progression in spondyloarthritis appears to be time-shifted,” GESPIC study investigator Denis Poddubnyy, MD, said at the annual European Congress of Rheumatology.

Reporting the spinal findings, Dr. Poddubnyy, who is head of the rheumatology department at Campus Benjamin Franklin of the Charité University Hospital in Berlin noted that “at least 4 years of observation is required” to see an effect of TNF inhibitors on new bone formation. “This might be related to the intermediate and long-lasting process of bone repair that follows the resolution of inflammation and precedes the development of new bone,” he suggested.

The effect on the SIJ may precede the spinal damage, so it’s important to look at how anti-inflammatory treatment might affect the SIJ, Dr. Poddubnyy said in an interview. Results from the SIJ analysis, which were presented by Murat Torgutalp, MD, suggested that there might be a 2-year time lag between TNFi treatment and being able to see something on x-rays.

“Everybody in the past focused on the spine, considering that spinal changes are functionally more relevant than changes in the SIJ,” Dr. Poddubnyy observed.

Structural damage in the spine and SIJ in radiographic axSpA probably occur by different processes, he explained. While it’s more to do with new bone formation in the spine, it’s more to do with erosive damage followed by repair in the SIJ. “It is very likely that in the majority of patients we see that those two processes are somehow disconnected in time,” he suggested.

The GESPIC cohort provides a good base from which to examine the effects of TNF inhibitors on radiographic progression in patients with axSpA for several reasons, observed Dr. Poddubnyy. For one it was established in 2000, so “well before the introduction of TNF-alpha inhibitors.” That means that almost all patients would have been put on TNF inhibitors while enrolled in the cohort, he said, “so we could really follow up them prospectively for a few years.” In fact, just nine patients had been treated with TNF inhibitors at baseline.

Spinal findings

Although TNF inhibitors are widely accepted to be effective anti-inflammatory drugs that can relieve patients’ spinal symptoms, there were inconclusive data on whether they can also have an impact on radiographic progression.

To look at the possible immediate or later effect of TNF inhibitors on spinal radiographic progression, the GESPIC investigators looked at sets of radiographs, taken at least 2 years apart over a 10-year period, from 266 patients, 77 of whom who had been treated with a TNF inhibitor for at least 12 months or more. Overall, there were 103 2-year intervals covered by TNF inhibitor treatment of any duration, and 78 intervals in which treatment had lasted for at least 12 months.

“This cohort included patients with quite early disease, including both nonradiographic and radiographic forms,” Dr. Poddubnyy said, adding that some patients had been treated with a TNF inhibitor for up to 10 years.

While there were no significant changes in radiographic progression as measured using the modified Stoke Ankylosing Spondylitis Spine Score (mSASSS) between patients who had and who had not received at least 12 months’ TNF inhibitor therapy in the first 2 years, there was a substantial difference after 4 years.

The mSASSS change in scores over 2 years in patients who were and were not treated with a TNF inhibitor for 12 months or more were a respective 0.35 and 0.81 (P = .047) considering treatment in the previous 2-year treatment period, and 0.33 and 0.77 (also P = .047) considering treatment in the prior 2-year treatment period and the current 2-year period.

“Our interpretation is that within the first 2 years we see a resolution of inflammation, a process of repair, maybe even a process of bone formation, but then if you continue this inflammation control for longer than 2 years, then we see the effect on structural damage,” Dr. Poddubnyy said.

SIJ findings

The SIJ analysis included 301 patients who had at least two sets of radiographs taken every 2 years over a 10-year follow-up period, 87 (28.9%) of whom had been treated with at least one TNF inhibitor during follow-up.

Radiographic SIJ progression was defined as the change in the sacroiliitis sum score over 2 years. Calculations considered both the current 2-year period and the previous 2-year period, as well as any use of TNF inhibitors or more than 12 months’ use.

There were no significant differences in sacroiliitis sum scores comparing people who were and were not treated with TNF inhibitors for at least 12 months in the current 2-year period. However, when the previous 2-year period was looked at there was a significant difference in the score between those who had and had not been treated for 12 or more months (0.15 vs. 0.27, respectively; P = .024).

“These data indicate that TNF inhibitors have a late effect,” Dr. Torgutalp said. “You cannot observe the effect of the TNF inhibitor on radiographic progression in the same interval, you have to wait to see this effect in the next interval,” he explained. That means that when a patient uses a TNF inhibitor, you must wait to see the effect on SIJ progression in the subsequent 2 years, he qualified.

Lessons for practice

“If we control inflammation, we might control structural damage progression,” Dr. Poddubnyy said.

If this is the case, then it is likely that other drugs that control inflammation in axSpA will have a similar effect, such as those that target interleukin-17 or Janus kinases. This is something in the future to look at in the GESPIC cohort once enough patient-years of data are available.

Early treatment is probably important, Dr. Poddubnyy suggested, “because in patients with early disease we can expect not that much structural damage at baseline.” Starting treatment early, at the inflammatory stage rather than at the stage where there was a lot of damage already there, could yield better results, he proposed.

“As a clinician, as a researcher, I would be really interested in demonstrating in a prospective, clinical trial that treating patients to target, achieving a remission, would be associated with retardation of structural damage progression. This has only been shown in observational studies so far, so this is an open question.”

Questions on adjustments made for confounding factors

Questions on the spinal presentation included whether there were any differences in NSAID use, to which Dr. Poddubnyy responded, “Patients on TNF inhibitors normally reduce NSAIDs. We adjusted, therefore, for NSAID use.”

The SIJ presentation garnered more questioning. Dr. Torgutalp qualified that both radiographic and nonradiographic patients had been included in the analysis, and “we observe the same effect.”

Asked how clinically relevant the findings were, he noted that, “to the best of our knowledge, this is the first data for sacroiliac joint progression.” The data for spinal progression seem to “reconcile the results from previous investigators,” where there had been some conflicting data on whether 2 or 4 years’ TNF-inhibitor therapy had any effect on radiographic progression.

“I don’t think you’ve adjusted for the baseline level of damage of the SIJ?” queried Pedro Machado, MD, of University College London Hospital. “As you know, in the spine, but also at the SIJ level, the baseline level of damage is usually one of the strongest predictors of further damage.”

This is not something that was adjusted for, Dr. Torgutalp conceded. “We will do it for further analysis,” he said.

Since 2010, GESPIC has been supported by grants from AbbVie. Dr. Poddubnyy reported financial relationships with AbbVie and numerous other pharmaceutical companies. Dr. Torgutalp had no relevant disclosures.

Tumor necrosis factor inhibitors reduce both spinal and sacroiliac radiographic progression in early axial spondyloarthritis (axSpA), but the effects take time.

According to data from the German Spondyloarthritis Inception Cohort (GESPIC), it could take 2-4 years after the initiation of TNF inhibitor treatment before x-rays will show any sign of a benefit in either the spine or sacroiliac joints (SIJ).

“The association between anti-TNF treatment and the retardation of radiographic spinal progression in spondyloarthritis appears to be time-shifted,” GESPIC study investigator Denis Poddubnyy, MD, said at the annual European Congress of Rheumatology.

Reporting the spinal findings, Dr. Poddubnyy, who is head of the rheumatology department at Campus Benjamin Franklin of the Charité University Hospital in Berlin noted that “at least 4 years of observation is required” to see an effect of TNF inhibitors on new bone formation. “This might be related to the intermediate and long-lasting process of bone repair that follows the resolution of inflammation and precedes the development of new bone,” he suggested.

The effect on the SIJ may precede the spinal damage, so it’s important to look at how anti-inflammatory treatment might affect the SIJ, Dr. Poddubnyy said in an interview. Results from the SIJ analysis, which were presented by Murat Torgutalp, MD, suggested that there might be a 2-year time lag between TNFi treatment and being able to see something on x-rays.

“Everybody in the past focused on the spine, considering that spinal changes are functionally more relevant than changes in the SIJ,” Dr. Poddubnyy observed.

Structural damage in the spine and SIJ in radiographic axSpA probably occur by different processes, he explained. While it’s more to do with new bone formation in the spine, it’s more to do with erosive damage followed by repair in the SIJ. “It is very likely that in the majority of patients we see that those two processes are somehow disconnected in time,” he suggested.

The GESPIC cohort provides a good base from which to examine the effects of TNF inhibitors on radiographic progression in patients with axSpA for several reasons, observed Dr. Poddubnyy. For one it was established in 2000, so “well before the introduction of TNF-alpha inhibitors.” That means that almost all patients would have been put on TNF inhibitors while enrolled in the cohort, he said, “so we could really follow up them prospectively for a few years.” In fact, just nine patients had been treated with TNF inhibitors at baseline.

Spinal findings

Although TNF inhibitors are widely accepted to be effective anti-inflammatory drugs that can relieve patients’ spinal symptoms, there were inconclusive data on whether they can also have an impact on radiographic progression.

To look at the possible immediate or later effect of TNF inhibitors on spinal radiographic progression, the GESPIC investigators looked at sets of radiographs, taken at least 2 years apart over a 10-year period, from 266 patients, 77 of whom who had been treated with a TNF inhibitor for at least 12 months or more. Overall, there were 103 2-year intervals covered by TNF inhibitor treatment of any duration, and 78 intervals in which treatment had lasted for at least 12 months.

“This cohort included patients with quite early disease, including both nonradiographic and radiographic forms,” Dr. Poddubnyy said, adding that some patients had been treated with a TNF inhibitor for up to 10 years.

While there were no significant changes in radiographic progression as measured using the modified Stoke Ankylosing Spondylitis Spine Score (mSASSS) between patients who had and who had not received at least 12 months’ TNF inhibitor therapy in the first 2 years, there was a substantial difference after 4 years.

The mSASSS change in scores over 2 years in patients who were and were not treated with a TNF inhibitor for 12 months or more were a respective 0.35 and 0.81 (P = .047) considering treatment in the previous 2-year treatment period, and 0.33 and 0.77 (also P = .047) considering treatment in the prior 2-year treatment period and the current 2-year period.

“Our interpretation is that within the first 2 years we see a resolution of inflammation, a process of repair, maybe even a process of bone formation, but then if you continue this inflammation control for longer than 2 years, then we see the effect on structural damage,” Dr. Poddubnyy said.

SIJ findings

The SIJ analysis included 301 patients who had at least two sets of radiographs taken every 2 years over a 10-year follow-up period, 87 (28.9%) of whom had been treated with at least one TNF inhibitor during follow-up.

Radiographic SIJ progression was defined as the change in the sacroiliitis sum score over 2 years. Calculations considered both the current 2-year period and the previous 2-year period, as well as any use of TNF inhibitors or more than 12 months’ use.

There were no significant differences in sacroiliitis sum scores comparing people who were and were not treated with TNF inhibitors for at least 12 months in the current 2-year period. However, when the previous 2-year period was looked at there was a significant difference in the score between those who had and had not been treated for 12 or more months (0.15 vs. 0.27, respectively; P = .024).

“These data indicate that TNF inhibitors have a late effect,” Dr. Torgutalp said. “You cannot observe the effect of the TNF inhibitor on radiographic progression in the same interval, you have to wait to see this effect in the next interval,” he explained. That means that when a patient uses a TNF inhibitor, you must wait to see the effect on SIJ progression in the subsequent 2 years, he qualified.

Lessons for practice

“If we control inflammation, we might control structural damage progression,” Dr. Poddubnyy said.

If this is the case, then it is likely that other drugs that control inflammation in axSpA will have a similar effect, such as those that target interleukin-17 or Janus kinases. This is something in the future to look at in the GESPIC cohort once enough patient-years of data are available.

Early treatment is probably important, Dr. Poddubnyy suggested, “because in patients with early disease we can expect not that much structural damage at baseline.” Starting treatment early, at the inflammatory stage rather than at the stage where there was a lot of damage already there, could yield better results, he proposed.

“As a clinician, as a researcher, I would be really interested in demonstrating in a prospective, clinical trial that treating patients to target, achieving a remission, would be associated with retardation of structural damage progression. This has only been shown in observational studies so far, so this is an open question.”

Questions on adjustments made for confounding factors

Questions on the spinal presentation included whether there were any differences in NSAID use, to which Dr. Poddubnyy responded, “Patients on TNF inhibitors normally reduce NSAIDs. We adjusted, therefore, for NSAID use.”

The SIJ presentation garnered more questioning. Dr. Torgutalp qualified that both radiographic and nonradiographic patients had been included in the analysis, and “we observe the same effect.”

Asked how clinically relevant the findings were, he noted that, “to the best of our knowledge, this is the first data for sacroiliac joint progression.” The data for spinal progression seem to “reconcile the results from previous investigators,” where there had been some conflicting data on whether 2 or 4 years’ TNF-inhibitor therapy had any effect on radiographic progression.

“I don’t think you’ve adjusted for the baseline level of damage of the SIJ?” queried Pedro Machado, MD, of University College London Hospital. “As you know, in the spine, but also at the SIJ level, the baseline level of damage is usually one of the strongest predictors of further damage.”

This is not something that was adjusted for, Dr. Torgutalp conceded. “We will do it for further analysis,” he said.

Since 2010, GESPIC has been supported by grants from AbbVie. Dr. Poddubnyy reported financial relationships with AbbVie and numerous other pharmaceutical companies. Dr. Torgutalp had no relevant disclosures.

Inhibiting granulocyte/macrophage–colony stimulating factor (GM-CSF) with mavrilimumab prevented some patients with severe COVID-19 pneumonia and hyperinflammation from needing mechanical ventilation and reduced their risk of dying versus placebo in a phase 2 study.

There was no difference in outcomes between the two doses of mavrilimumab used in the trial (6 mg/kg or 10 mg/kg) and combined data showed a higher percentage of patients achieving the primary endpoint of being alive and free of mechanical ventilation at 29 days, at 87%, versus placebo, at 74%.

The P value was 0.12, “which achieved the prespecified evidentiary standard of 0.2,” according to Lara Pupim, MD, vice president of clinical research and development at Kiniksa Pharmaceuticals in Lexington, Mass.

Importantly, there was a 61% reduction in the risk of dying if patients had received mavrilimumab rather than placebo, she reported at the annual European Congress of Rheumatology. Mortality at day 29 was 21% in the placebo arm but just 8% in the combined mavrilimumab arms (P = .07).

Hendrik Schulze-Koops, MD, called it a “surprising study” and that “the outcome is very spectacular” in his short appraisal of the study during the Clinical Highlights session on the final day of the congress.

Mavrilimumab was “a compound that we would not have thought that would have such an impact on the outcome of COVID-19 infected patients,” Dr. Schulze-Koops of Ludwig Maximilian University of Munich added.

In this small study, “there was a consistent suggestion of a biological effect across key endpoints,” Richard Conway, MBChB, PhD, a consultant rheumatologist at St. James’s Hospital in Dublin, pointed out in an interview.

“Similar to tocilizumab, the benefits with mavrilimumab appear to be in addition to those seen with glucocorticoids, as 96% of patients received dexamethasone,” Dr. Conway observed. Furthermore, nearly one-third received antiviral or remdesivir treatment.

“This study was likely underpowered to assess a clinically meaningful benefit,” he said, adding that “there is insufficient evidence at present to begin using mavrilimumab as an alternative to currently available agents.” That said, “these results are promising for future studies.”

Rationale for GM-CSF inhibition with mavrilimumab in COVID-19 pneumonia

“The cytokine GM-CSF is vital to both lung homeostasis and regulation of inflammation in autoimmunity,” Dr. Pupim explained.

She added that “GM-CSF is implicated in the mechanism of aberrant immune cell infiltration and activation in the lungs, and it may contribute to respiratory failure and death in patients with severe COVID-19 pneumonia and systemic hyperinflammation.”

The efficacy and safety of blocking GM-CSF with mavrilimumab have been shown previously in phase 2 studies in other diseases, Dr. Pupim noted. This includes patients with rheumatoid arthritis and those with giant cell arteritis.

“It was hypothesized that GM-CSF receptor–alpha blockade may reduce infiltration of pathogenic cells into the lung and may suppress inflammation in COVID-19 pneumonia in hyperinflammation,” she explained.

Study details and other outcome results

The study presented by Dr. Pupim was a phase 2/3 double-blind, placebo-controlled trial predominantly conducted in Brazil, the United States, and South Africa, with some participation in Peru and Chile.

Patients were eligible for inclusion if they had had a positive COVID-19 test within 14 days of randomization and had been hospitalized but not ventilated. Evidence of bilateral pneumonia on chest x-ray or CT scan and clinical laboratory evidence indicative of hyperinflammation were also prerequisites for study enrollment.

The ongoing study comprised two cohorts, Dr. Pupim explained: patients who have not been ventilated and those who have recently been ventilated. Dr. Pupim presented the data on the nonventilated cohort, noting that there was a total of 116 patients aged a mean of 57 years.

Patients were randomized to one of three treatment arms: two groups received a single intravenous infusion of mavrilimumab, either 6 mg/kg or 10 mg/kg, and the third group got a placebo.

“Using a time-to-event approach, looking at mechanical ventilation-free survival, mavrilimumab recipients experienced a 65% reduction in the risk of mechanical ventilation or death,” Dr. Pupim said (P = .0175).

“Separation in the Kaplan-Meier curves was evident very early after study drug administration,” she added.

There were trends toward a faster benefit with mavrilimumab than placebo in two other key secondary endpoints: the median time to achieving a two-point clinical improvement (7 vs. 11 days) and the median time to room air (7 vs. 9 days).

Timing of mavrilimumab administration and safety

Study coauthor and chief clinical development officer at Kiniksa, Arian Pano, MD, answered questions on the presentation. When asked about the timing of giving mavrilimumab, he said: “Based on these data it is before they go to ventilation, as soon as you have symptoms of hyperinflammation and a need for oxygen.”

Mavrilimumab is given as a single infusion “and has been well tolerated; virtually no interruptions occurred in this study.”

No serious adverse events related to mavrilimumab were seen, and adverse events, including secondary infections, which are known complications of COVID-19, occurred less frequently in mavrilimumab recipients, compared with placebo.

Dr. Pupim reported that there was a case of tuberculosis in one patient treated with mavrilimumab (10 mg/kg). That case had occurred in an “endemic area for tuberculosis,” and the patient had been screened before entry but only via a sputum sample.

“Prior to these events, the patient received high-dose corticosteroids, a known risk factor for reactivation of TB, and thus the potential additive contribution of mavrilimumab, if any, is uncertain.” Dr. Pupim said.

“Thrombotic events, another known complication of COVID-19, occurred in the placebo arm only,” she added.

Dr. Pano commented that the study has now “seamlessly continued to phase 3. So, basically, we did not stop the study. At the end of phase 2, we just locked the database and collected the data.” Both the 6 mg/kg and 10 mg/kg are being studied, but it’s “very likely [that] 6 mg/kg could be the dose that we may bring forward to the clinic in terms of registration, but that’s at this point in time. We will need to wait for the phase 3 data,” he observed. Those findings will hopefully be available later this year.

Kiniksa funded the study. Dr. Pupim, Dr. Pano, and multiple study coinvestigators are employees of the company.

Dr. Schulze-Koops was not involved in the study and had no specific disclosures. Dr. Conway had no financial disclosures to make in relation to his comments.

Inhibiting granulocyte/macrophage–colony stimulating factor (GM-CSF) with mavrilimumab prevented some patients with severe COVID-19 pneumonia and hyperinflammation from needing mechanical ventilation and reduced their risk of dying versus placebo in a phase 2 study.

There was no difference in outcomes between the two doses of mavrilimumab used in the trial (6 mg/kg or 10 mg/kg) and combined data showed a higher percentage of patients achieving the primary endpoint of being alive and free of mechanical ventilation at 29 days, at 87%, versus placebo, at 74%.

The P value was 0.12, “which achieved the prespecified evidentiary standard of 0.2,” according to Lara Pupim, MD, vice president of clinical research and development at Kiniksa Pharmaceuticals in Lexington, Mass.

Importantly, there was a 61% reduction in the risk of dying if patients had received mavrilimumab rather than placebo, she reported at the annual European Congress of Rheumatology. Mortality at day 29 was 21% in the placebo arm but just 8% in the combined mavrilimumab arms (P = .07).

Hendrik Schulze-Koops, MD, called it a “surprising study” and that “the outcome is very spectacular” in his short appraisal of the study during the Clinical Highlights session on the final day of the congress.

Mavrilimumab was “a compound that we would not have thought that would have such an impact on the outcome of COVID-19 infected patients,” Dr. Schulze-Koops of Ludwig Maximilian University of Munich added.

In this small study, “there was a consistent suggestion of a biological effect across key endpoints,” Richard Conway, MBChB, PhD, a consultant rheumatologist at St. James’s Hospital in Dublin, pointed out in an interview.

“Similar to tocilizumab, the benefits with mavrilimumab appear to be in addition to those seen with glucocorticoids, as 96% of patients received dexamethasone,” Dr. Conway observed. Furthermore, nearly one-third received antiviral or remdesivir treatment.

“This study was likely underpowered to assess a clinically meaningful benefit,” he said, adding that “there is insufficient evidence at present to begin using mavrilimumab as an alternative to currently available agents.” That said, “these results are promising for future studies.”

Rationale for GM-CSF inhibition with mavrilimumab in COVID-19 pneumonia

“The cytokine GM-CSF is vital to both lung homeostasis and regulation of inflammation in autoimmunity,” Dr. Pupim explained.

She added that “GM-CSF is implicated in the mechanism of aberrant immune cell infiltration and activation in the lungs, and it may contribute to respiratory failure and death in patients with severe COVID-19 pneumonia and systemic hyperinflammation.”

The efficacy and safety of blocking GM-CSF with mavrilimumab have been shown previously in phase 2 studies in other diseases, Dr. Pupim noted. This includes patients with rheumatoid arthritis and those with giant cell arteritis.

“It was hypothesized that GM-CSF receptor–alpha blockade may reduce infiltration of pathogenic cells into the lung and may suppress inflammation in COVID-19 pneumonia in hyperinflammation,” she explained.

Study details and other outcome results

The study presented by Dr. Pupim was a phase 2/3 double-blind, placebo-controlled trial predominantly conducted in Brazil, the United States, and South Africa, with some participation in Peru and Chile.

Patients were eligible for inclusion if they had had a positive COVID-19 test within 14 days of randomization and had been hospitalized but not ventilated. Evidence of bilateral pneumonia on chest x-ray or CT scan and clinical laboratory evidence indicative of hyperinflammation were also prerequisites for study enrollment.

The ongoing study comprised two cohorts, Dr. Pupim explained: patients who have not been ventilated and those who have recently been ventilated. Dr. Pupim presented the data on the nonventilated cohort, noting that there was a total of 116 patients aged a mean of 57 years.

Patients were randomized to one of three treatment arms: two groups received a single intravenous infusion of mavrilimumab, either 6 mg/kg or 10 mg/kg, and the third group got a placebo.

“Using a time-to-event approach, looking at mechanical ventilation-free survival, mavrilimumab recipients experienced a 65% reduction in the risk of mechanical ventilation or death,” Dr. Pupim said (P = .0175).

“Separation in the Kaplan-Meier curves was evident very early after study drug administration,” she added.

There were trends toward a faster benefit with mavrilimumab than placebo in two other key secondary endpoints: the median time to achieving a two-point clinical improvement (7 vs. 11 days) and the median time to room air (7 vs. 9 days).

Timing of mavrilimumab administration and safety

Study coauthor and chief clinical development officer at Kiniksa, Arian Pano, MD, answered questions on the presentation. When asked about the timing of giving mavrilimumab, he said: “Based on these data it is before they go to ventilation, as soon as you have symptoms of hyperinflammation and a need for oxygen.”

Mavrilimumab is given as a single infusion “and has been well tolerated; virtually no interruptions occurred in this study.”

No serious adverse events related to mavrilimumab were seen, and adverse events, including secondary infections, which are known complications of COVID-19, occurred less frequently in mavrilimumab recipients, compared with placebo.

Dr. Pupim reported that there was a case of tuberculosis in one patient treated with mavrilimumab (10 mg/kg). That case had occurred in an “endemic area for tuberculosis,” and the patient had been screened before entry but only via a sputum sample.

“Prior to these events, the patient received high-dose corticosteroids, a known risk factor for reactivation of TB, and thus the potential additive contribution of mavrilimumab, if any, is uncertain.” Dr. Pupim said.

“Thrombotic events, another known complication of COVID-19, occurred in the placebo arm only,” she added.

Dr. Pano commented that the study has now “seamlessly continued to phase 3. So, basically, we did not stop the study. At the end of phase 2, we just locked the database and collected the data.” Both the 6 mg/kg and 10 mg/kg are being studied, but it’s “very likely [that] 6 mg/kg could be the dose that we may bring forward to the clinic in terms of registration, but that’s at this point in time. We will need to wait for the phase 3 data,” he observed. Those findings will hopefully be available later this year.

Kiniksa funded the study. Dr. Pupim, Dr. Pano, and multiple study coinvestigators are employees of the company.

Dr. Schulze-Koops was not involved in the study and had no specific disclosures. Dr. Conway had no financial disclosures to make in relation to his comments.

Inhibiting granulocyte/macrophage–colony stimulating factor (GM-CSF) with mavrilimumab prevented some patients with severe COVID-19 pneumonia and hyperinflammation from needing mechanical ventilation and reduced their risk of dying versus placebo in a phase 2 study.

There was no difference in outcomes between the two doses of mavrilimumab used in the trial (6 mg/kg or 10 mg/kg) and combined data showed a higher percentage of patients achieving the primary endpoint of being alive and free of mechanical ventilation at 29 days, at 87%, versus placebo, at 74%.

The P value was 0.12, “which achieved the prespecified evidentiary standard of 0.2,” according to Lara Pupim, MD, vice president of clinical research and development at Kiniksa Pharmaceuticals in Lexington, Mass.

Importantly, there was a 61% reduction in the risk of dying if patients had received mavrilimumab rather than placebo, she reported at the annual European Congress of Rheumatology. Mortality at day 29 was 21% in the placebo arm but just 8% in the combined mavrilimumab arms (P = .07).

Hendrik Schulze-Koops, MD, called it a “surprising study” and that “the outcome is very spectacular” in his short appraisal of the study during the Clinical Highlights session on the final day of the congress.

Mavrilimumab was “a compound that we would not have thought that would have such an impact on the outcome of COVID-19 infected patients,” Dr. Schulze-Koops of Ludwig Maximilian University of Munich added.

In this small study, “there was a consistent suggestion of a biological effect across key endpoints,” Richard Conway, MBChB, PhD, a consultant rheumatologist at St. James’s Hospital in Dublin, pointed out in an interview.

“Similar to tocilizumab, the benefits with mavrilimumab appear to be in addition to those seen with glucocorticoids, as 96% of patients received dexamethasone,” Dr. Conway observed. Furthermore, nearly one-third received antiviral or remdesivir treatment.

“This study was likely underpowered to assess a clinically meaningful benefit,” he said, adding that “there is insufficient evidence at present to begin using mavrilimumab as an alternative to currently available agents.” That said, “these results are promising for future studies.”

Rationale for GM-CSF inhibition with mavrilimumab in COVID-19 pneumonia

“The cytokine GM-CSF is vital to both lung homeostasis and regulation of inflammation in autoimmunity,” Dr. Pupim explained.

She added that “GM-CSF is implicated in the mechanism of aberrant immune cell infiltration and activation in the lungs, and it may contribute to respiratory failure and death in patients with severe COVID-19 pneumonia and systemic hyperinflammation.”

The efficacy and safety of blocking GM-CSF with mavrilimumab have been shown previously in phase 2 studies in other diseases, Dr. Pupim noted. This includes patients with rheumatoid arthritis and those with giant cell arteritis.

“It was hypothesized that GM-CSF receptor–alpha blockade may reduce infiltration of pathogenic cells into the lung and may suppress inflammation in COVID-19 pneumonia in hyperinflammation,” she explained.

Study details and other outcome results

The study presented by Dr. Pupim was a phase 2/3 double-blind, placebo-controlled trial predominantly conducted in Brazil, the United States, and South Africa, with some participation in Peru and Chile.

Patients were eligible for inclusion if they had had a positive COVID-19 test within 14 days of randomization and had been hospitalized but not ventilated. Evidence of bilateral pneumonia on chest x-ray or CT scan and clinical laboratory evidence indicative of hyperinflammation were also prerequisites for study enrollment.

The ongoing study comprised two cohorts, Dr. Pupim explained: patients who have not been ventilated and those who have recently been ventilated. Dr. Pupim presented the data on the nonventilated cohort, noting that there was a total of 116 patients aged a mean of 57 years.

Patients were randomized to one of three treatment arms: two groups received a single intravenous infusion of mavrilimumab, either 6 mg/kg or 10 mg/kg, and the third group got a placebo.

“Using a time-to-event approach, looking at mechanical ventilation-free survival, mavrilimumab recipients experienced a 65% reduction in the risk of mechanical ventilation or death,” Dr. Pupim said (P = .0175).

“Separation in the Kaplan-Meier curves was evident very early after study drug administration,” she added.

There were trends toward a faster benefit with mavrilimumab than placebo in two other key secondary endpoints: the median time to achieving a two-point clinical improvement (7 vs. 11 days) and the median time to room air (7 vs. 9 days).

Timing of mavrilimumab administration and safety

Study coauthor and chief clinical development officer at Kiniksa, Arian Pano, MD, answered questions on the presentation. When asked about the timing of giving mavrilimumab, he said: “Based on these data it is before they go to ventilation, as soon as you have symptoms of hyperinflammation and a need for oxygen.”

Mavrilimumab is given as a single infusion “and has been well tolerated; virtually no interruptions occurred in this study.”

No serious adverse events related to mavrilimumab were seen, and adverse events, including secondary infections, which are known complications of COVID-19, occurred less frequently in mavrilimumab recipients, compared with placebo.

Dr. Pupim reported that there was a case of tuberculosis in one patient treated with mavrilimumab (10 mg/kg). That case had occurred in an “endemic area for tuberculosis,” and the patient had been screened before entry but only via a sputum sample.

“Prior to these events, the patient received high-dose corticosteroids, a known risk factor for reactivation of TB, and thus the potential additive contribution of mavrilimumab, if any, is uncertain.” Dr. Pupim said.

“Thrombotic events, another known complication of COVID-19, occurred in the placebo arm only,” she added.

Dr. Pano commented that the study has now “seamlessly continued to phase 3. So, basically, we did not stop the study. At the end of phase 2, we just locked the database and collected the data.” Both the 6 mg/kg and 10 mg/kg are being studied, but it’s “very likely [that] 6 mg/kg could be the dose that we may bring forward to the clinic in terms of registration, but that’s at this point in time. We will need to wait for the phase 3 data,” he observed. Those findings will hopefully be available later this year.

Kiniksa funded the study. Dr. Pupim, Dr. Pano, and multiple study coinvestigators are employees of the company.

Dr. Schulze-Koops was not involved in the study and had no specific disclosures. Dr. Conway had no financial disclosures to make in relation to his comments.

Patients with rheumatoid arthritis who were using rituximab at the time of COVID-19 onset had a fourfold higher risk of being hospitalized, needing mechanical ventilation, or dying, compared with patients taking a tumor necrosis factor inhibitor (TNFi), according to a report given at the annual European Congress of Rheumatology.

The use of Janus kinase inhibitors (JAKi) also was associated with a twofold higher risk for these COVID-19 outcomes, said Jeffrey A. Sparks, MD, of Brigham and Women’s Hospital and Harvard Medical School, Boston, in presenting the analysis from the COVID-19 Global Rheumatology Alliance (GRA) Physician Registry.

“The strong association of rituximab and JAK inhibitor use with poor COVID-19 outcomes highlights the prioritization of risk mitigation strategies for these patients,” Dr. Sparks said at the meeting.

The full findings have now been published in Annals of the Rheumatic Diseases.
 

JAKi association questioned

These findings provide “an important understanding for the risk of our patients in times before vaccination,” said Hendrik Schulze-Koops, MD, of Ludwig Maximilian University of Munich, who chaired the session in which the study was presented.

However, “recently, baricitinib was licensed to prevent particular aspects of severe COVID. What’s the explanation for this discrepancy?” he asked.

“Certainly, the JAK inhibitor finding deserves further study,” Dr. Sparks acknowledged, adding that the data were analyzed by class rather than for individual drugs.

“One possible explanation could be when JAK inhibitors are used,” he suggested. “It might be different for patients who [have been] just infected – that might have different biologic effects – as opposed to choosing to treat patients right when there’s a hyperinflammatory cascade, or there’s oxygen need.”

Regarding the JAK inhibitor finding, Ronald van Vollenhoven, MD, PhD, of the University of Amsterdam, pointed out during the online Q&A that “JAKi have a very short half-life compared to biologics.”

Dr. van Vollenhoven asked: “Could the practice of stopping these drugs upon COVID infection have a negative impact on the course?” To which Dr. Sparks responded: “The different half-life of drugs would be a promising avenue to look at, to see whether increases in disease activity might have imparted some of the effects we saw.”
 

Performing the analysis

As of April 12, 2021, the GRA Physician Registry contained the records of more than 15,000 patients. Dr. Sparks, collaborator Zachary Wallace, MD, of Massachusetts General Hospital, Boston, and associates limited their analysis to 2,869 patients with RA who had been treated with either a biologic or targeted synthetic disease-modifying antirheumatic drug (b/tsDMARD) at the time they were diagnosed with COVID-19.

“We wanted to limit it to a single disease and also limit it to drugs that are considered for that disease,” Dr. Sparks explained in an interview.

“Because patients with rheumatoid arthritis are often treated sequentially, we wanted to further limit the analysis to patients who were on advanced therapies so that they were at a similar disease state, and also had the opportunity to receive advanced therapies.”

This approach hopefully minimizes the possibility of confounding by indication, Dr. Sparks said.

Most of the patients included in the analysis had received a TNFi (n = 1,388), and they were used as the control arm of the analysis. Outcomes associated with treatment with the other b/tsDMARDs, which included abatacept (n = 237), rituximab (n = 364), interleukin-6 inhibitors (IL-6i; n = 317), and JAKi (n = 563), were then compared with TNFi.

Baseline characteristics of patients were broadly similar across the groups. The mean age was 56.7 years and 80.8% of the study population was female. There were a few expected differences among users of rituximab versus TNFi, notably a higher percentage of patients with interstitial lung disease (11% vs. 1.4% of TNFi users) or cancer (7.4% vs. 0.9%) among patients treated with rituximab since it is commonly used in these patients, Dr. Sparks said.

“We did perform a sensitivity analysis where we restricted the population to not having ILD or cancer and we actually found really similar findings,” he added.

Four COVID-19 outcomes assessed

The researchers used a four-point ordinal scale modeled after one set by the World Health Organization to assess four COVID-19 outcomes: not hospitalized, hospitalized without oxygenation, hospitalized with oxygenation or ventilation, and death.

Odds ratios (ORs) comparing rituximab to TNFi for these four COVID-19 outcomes were a respective 4.53, 2.87, 4.05, and 4.57. The ORs for JAKi versus TNFi were a respective 2.4, 1.55, 2.03, and 2.04.

“We found no consistent associations of abatacept or interleukin-6 inhibitors with COVID-19 severity, compared to TNF inhibitors,” which is reassuring, Dr. Sparks said.

ORs for the four COVID-19 outcomes with abatacept were a respective 1.18, 1.12, 1.41, and 1.46, and for IL-6i were 0.84, 0.72, 0.75, and 1.13.

Rituximab use in patients with RA who develop COVID-19

So, should rituximab be stopped in patients with RA if they develop COVID-19? “This is an important question and one that would be decided on a case-by-case basis,” Dr. Sparks said. “Of course, the drug has a very long half-life, so risk mitigation strategies are still of utmost importance,” he added.

“I think everyone’s a bit reticent to want to start rituximab in this environment, but it might also make me pause about starting a JAK inhibitor,” Dr. Sparks added. “Given that this is a first finding, I’m not sure I would necessarily change patients who are doing well on these medications. I think what it really makes me want to do is to try to obviously vaccinate the patients on JAK inhibitors as they do have a short half-life.”

More observational studies would be helpful, Dr. Sparks said, adding that “the most pressing need is to try to figure out how to protect our patients with rituximab.”

The COVID-19 Global Rheumatology Alliance Physician Registry is supported by the American College of Rheumatology and the European Alliance of Associations for Rheumatology. Dr. Sparks disclosed serving as a consultant for Bristol Myers Squibb, Gilead, Inova, Optum, and Pfizer for work unrelated to this study. Dr. Wallace disclosed receiving grant support from Bristol Myers Squibb and Principia/Sanofi and serving as a consultant for Viela Bio and Medpace for work unrelated to this study.

Patients with rheumatoid arthritis who were using rituximab at the time of COVID-19 onset had a fourfold higher risk of being hospitalized, needing mechanical ventilation, or dying, compared with patients taking a tumor necrosis factor inhibitor (TNFi), according to a report given at the annual European Congress of Rheumatology.

The use of Janus kinase inhibitors (JAKi) also was associated with a twofold higher risk for these COVID-19 outcomes, said Jeffrey A. Sparks, MD, of Brigham and Women’s Hospital and Harvard Medical School, Boston, in presenting the analysis from the COVID-19 Global Rheumatology Alliance (GRA) Physician Registry.

“The strong association of rituximab and JAK inhibitor use with poor COVID-19 outcomes highlights the prioritization of risk mitigation strategies for these patients,” Dr. Sparks said at the meeting.

The full findings have now been published in Annals of the Rheumatic Diseases.
 

JAKi association questioned

These findings provide “an important understanding for the risk of our patients in times before vaccination,” said Hendrik Schulze-Koops, MD, of Ludwig Maximilian University of Munich, who chaired the session in which the study was presented.

However, “recently, baricitinib was licensed to prevent particular aspects of severe COVID. What’s the explanation for this discrepancy?” he asked.

“Certainly, the JAK inhibitor finding deserves further study,” Dr. Sparks acknowledged, adding that the data were analyzed by class rather than for individual drugs.

“One possible explanation could be when JAK inhibitors are used,” he suggested. “It might be different for patients who [have been] just infected – that might have different biologic effects – as opposed to choosing to treat patients right when there’s a hyperinflammatory cascade, or there’s oxygen need.”

Regarding the JAK inhibitor finding, Ronald van Vollenhoven, MD, PhD, of the University of Amsterdam, pointed out during the online Q&A that “JAKi have a very short half-life compared to biologics.”

Dr. van Vollenhoven asked: “Could the practice of stopping these drugs upon COVID infection have a negative impact on the course?” To which Dr. Sparks responded: “The different half-life of drugs would be a promising avenue to look at, to see whether increases in disease activity might have imparted some of the effects we saw.”
 

Performing the analysis

As of April 12, 2021, the GRA Physician Registry contained the records of more than 15,000 patients. Dr. Sparks, collaborator Zachary Wallace, MD, of Massachusetts General Hospital, Boston, and associates limited their analysis to 2,869 patients with RA who had been treated with either a biologic or targeted synthetic disease-modifying antirheumatic drug (b/tsDMARD) at the time they were diagnosed with COVID-19.

“We wanted to limit it to a single disease and also limit it to drugs that are considered for that disease,” Dr. Sparks explained in an interview.

“Because patients with rheumatoid arthritis are often treated sequentially, we wanted to further limit the analysis to patients who were on advanced therapies so that they were at a similar disease state, and also had the opportunity to receive advanced therapies.”

This approach hopefully minimizes the possibility of confounding by indication, Dr. Sparks said.

Most of the patients included in the analysis had received a TNFi (n = 1,388), and they were used as the control arm of the analysis. Outcomes associated with treatment with the other b/tsDMARDs, which included abatacept (n = 237), rituximab (n = 364), interleukin-6 inhibitors (IL-6i; n = 317), and JAKi (n = 563), were then compared with TNFi.

Baseline characteristics of patients were broadly similar across the groups. The mean age was 56.7 years and 80.8% of the study population was female. There were a few expected differences among users of rituximab versus TNFi, notably a higher percentage of patients with interstitial lung disease (11% vs. 1.4% of TNFi users) or cancer (7.4% vs. 0.9%) among patients treated with rituximab since it is commonly used in these patients, Dr. Sparks said.

“We did perform a sensitivity analysis where we restricted the population to not having ILD or cancer and we actually found really similar findings,” he added.

Four COVID-19 outcomes assessed

The researchers used a four-point ordinal scale modeled after one set by the World Health Organization to assess four COVID-19 outcomes: not hospitalized, hospitalized without oxygenation, hospitalized with oxygenation or ventilation, and death.

Odds ratios (ORs) comparing rituximab to TNFi for these four COVID-19 outcomes were a respective 4.53, 2.87, 4.05, and 4.57. The ORs for JAKi versus TNFi were a respective 2.4, 1.55, 2.03, and 2.04.

“We found no consistent associations of abatacept or interleukin-6 inhibitors with COVID-19 severity, compared to TNF inhibitors,” which is reassuring, Dr. Sparks said.

ORs for the four COVID-19 outcomes with abatacept were a respective 1.18, 1.12, 1.41, and 1.46, and for IL-6i were 0.84, 0.72, 0.75, and 1.13.

Rituximab use in patients with RA who develop COVID-19

So, should rituximab be stopped in patients with RA if they develop COVID-19? “This is an important question and one that would be decided on a case-by-case basis,” Dr. Sparks said. “Of course, the drug has a very long half-life, so risk mitigation strategies are still of utmost importance,” he added.

“I think everyone’s a bit reticent to want to start rituximab in this environment, but it might also make me pause about starting a JAK inhibitor,” Dr. Sparks added. “Given that this is a first finding, I’m not sure I would necessarily change patients who are doing well on these medications. I think what it really makes me want to do is to try to obviously vaccinate the patients on JAK inhibitors as they do have a short half-life.”

More observational studies would be helpful, Dr. Sparks said, adding that “the most pressing need is to try to figure out how to protect our patients with rituximab.”

The COVID-19 Global Rheumatology Alliance Physician Registry is supported by the American College of Rheumatology and the European Alliance of Associations for Rheumatology. Dr. Sparks disclosed serving as a consultant for Bristol Myers Squibb, Gilead, Inova, Optum, and Pfizer for work unrelated to this study. Dr. Wallace disclosed receiving grant support from Bristol Myers Squibb and Principia/Sanofi and serving as a consultant for Viela Bio and Medpace for work unrelated to this study.

Patients with rheumatoid arthritis who were using rituximab at the time of COVID-19 onset had a fourfold higher risk of being hospitalized, needing mechanical ventilation, or dying, compared with patients taking a tumor necrosis factor inhibitor (TNFi), according to a report given at the annual European Congress of Rheumatology.

The use of Janus kinase inhibitors (JAKi) also was associated with a twofold higher risk for these COVID-19 outcomes, said Jeffrey A. Sparks, MD, of Brigham and Women’s Hospital and Harvard Medical School, Boston, in presenting the analysis from the COVID-19 Global Rheumatology Alliance (GRA) Physician Registry.

“The strong association of rituximab and JAK inhibitor use with poor COVID-19 outcomes highlights the prioritization of risk mitigation strategies for these patients,” Dr. Sparks said at the meeting.

The full findings have now been published in Annals of the Rheumatic Diseases.
 

JAKi association questioned

These findings provide “an important understanding for the risk of our patients in times before vaccination,” said Hendrik Schulze-Koops, MD, of Ludwig Maximilian University of Munich, who chaired the session in which the study was presented.

However, “recently, baricitinib was licensed to prevent particular aspects of severe COVID. What’s the explanation for this discrepancy?” he asked.

“Certainly, the JAK inhibitor finding deserves further study,” Dr. Sparks acknowledged, adding that the data were analyzed by class rather than for individual drugs.

“One possible explanation could be when JAK inhibitors are used,” he suggested. “It might be different for patients who [have been] just infected – that might have different biologic effects – as opposed to choosing to treat patients right when there’s a hyperinflammatory cascade, or there’s oxygen need.”

Regarding the JAK inhibitor finding, Ronald van Vollenhoven, MD, PhD, of the University of Amsterdam, pointed out during the online Q&A that “JAKi have a very short half-life compared to biologics.”

Dr. van Vollenhoven asked: “Could the practice of stopping these drugs upon COVID infection have a negative impact on the course?” To which Dr. Sparks responded: “The different half-life of drugs would be a promising avenue to look at, to see whether increases in disease activity might have imparted some of the effects we saw.”
 

Performing the analysis

As of April 12, 2021, the GRA Physician Registry contained the records of more than 15,000 patients. Dr. Sparks, collaborator Zachary Wallace, MD, of Massachusetts General Hospital, Boston, and associates limited their analysis to 2,869 patients with RA who had been treated with either a biologic or targeted synthetic disease-modifying antirheumatic drug (b/tsDMARD) at the time they were diagnosed with COVID-19.

“We wanted to limit it to a single disease and also limit it to drugs that are considered for that disease,” Dr. Sparks explained in an interview.

“Because patients with rheumatoid arthritis are often treated sequentially, we wanted to further limit the analysis to patients who were on advanced therapies so that they were at a similar disease state, and also had the opportunity to receive advanced therapies.”

This approach hopefully minimizes the possibility of confounding by indication, Dr. Sparks said.

Most of the patients included in the analysis had received a TNFi (n = 1,388), and they were used as the control arm of the analysis. Outcomes associated with treatment with the other b/tsDMARDs, which included abatacept (n = 237), rituximab (n = 364), interleukin-6 inhibitors (IL-6i; n = 317), and JAKi (n = 563), were then compared with TNFi.

Baseline characteristics of patients were broadly similar across the groups. The mean age was 56.7 years and 80.8% of the study population was female. There were a few expected differences among users of rituximab versus TNFi, notably a higher percentage of patients with interstitial lung disease (11% vs. 1.4% of TNFi users) or cancer (7.4% vs. 0.9%) among patients treated with rituximab since it is commonly used in these patients, Dr. Sparks said.

“We did perform a sensitivity analysis where we restricted the population to not having ILD or cancer and we actually found really similar findings,” he added.

Four COVID-19 outcomes assessed

The researchers used a four-point ordinal scale modeled after one set by the World Health Organization to assess four COVID-19 outcomes: not hospitalized, hospitalized without oxygenation, hospitalized with oxygenation or ventilation, and death.

Odds ratios (ORs) comparing rituximab to TNFi for these four COVID-19 outcomes were a respective 4.53, 2.87, 4.05, and 4.57. The ORs for JAKi versus TNFi were a respective 2.4, 1.55, 2.03, and 2.04.

“We found no consistent associations of abatacept or interleukin-6 inhibitors with COVID-19 severity, compared to TNF inhibitors,” which is reassuring, Dr. Sparks said.

ORs for the four COVID-19 outcomes with abatacept were a respective 1.18, 1.12, 1.41, and 1.46, and for IL-6i were 0.84, 0.72, 0.75, and 1.13.

Rituximab use in patients with RA who develop COVID-19

So, should rituximab be stopped in patients with RA if they develop COVID-19? “This is an important question and one that would be decided on a case-by-case basis,” Dr. Sparks said. “Of course, the drug has a very long half-life, so risk mitigation strategies are still of utmost importance,” he added.

“I think everyone’s a bit reticent to want to start rituximab in this environment, but it might also make me pause about starting a JAK inhibitor,” Dr. Sparks added. “Given that this is a first finding, I’m not sure I would necessarily change patients who are doing well on these medications. I think what it really makes me want to do is to try to obviously vaccinate the patients on JAK inhibitors as they do have a short half-life.”

More observational studies would be helpful, Dr. Sparks said, adding that “the most pressing need is to try to figure out how to protect our patients with rituximab.”

The COVID-19 Global Rheumatology Alliance Physician Registry is supported by the American College of Rheumatology and the European Alliance of Associations for Rheumatology. Dr. Sparks disclosed serving as a consultant for Bristol Myers Squibb, Gilead, Inova, Optum, and Pfizer for work unrelated to this study. Dr. Wallace disclosed receiving grant support from Bristol Myers Squibb and Principia/Sanofi and serving as a consultant for Viela Bio and Medpace for work unrelated to this study.