Lung Cancer

Artificial intelligence (AI) is expected to one day affect the entire continuum of cancer care – from screening and risk prediction to diagnosis, risk stratification, treatment selection, and follow-up, according to an expert in the field.

Hugo J.W.L. Aerts, PhD, director of the AI in Medicine Program at Brigham and Women’s Hospital in Boston, described studies using AI for some of these purposes during a presentation at the AACR Virtual Special Conference: Artificial Intelligence, Diagnosis, and Imaging (Abstract IA-06).

In one study, Dr. Aerts and colleagues set out to determine whether a convolutional neural network (CNN) could extract prognostic information from chest radiographs. The researchers tested this theory using patients from two trials – the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial and the National Lung Screening Trial (NLST).

The team developed a CNN, called CXR-risk, and tested whether it could predict the longevity and prognosis of patients in the PLCO (n = 52,320) and NLST (n = 5,493) trials over a 12-year time period, based only on chest radiographs. No clinical information, demographics, radiographic interpretations, duration of follow-up, or censoring were provided to the deep-learning system.

CXR-risk output was stratified into five categories of radiographic risk scores for probability of death, from 0 (very low likelihood of mortality) to 1 (very high likelihood of mortality).

The investigators found a graded association between radiographic risk score and mortality. The very-high-risk group had mortality rates of 53.0% (PLCO) and 33.9% (NLST). In both trials, this was significantly higher than for the very-low-risk group. The unadjusted hazard ratio was 18.3 in the PCLO data set and 15.2 in the NLST data set (P < .001 for both).

This association was maintained after adjustment for radiologists’ findings (e.g., a lung nodule) and risk factors such as age, gender, and comorbid illnesses like diabetes. The adjusted HR was 4.8 in the PCLO data set and 7.0 in the NLST data set (P < .001 for both).

In both data sets, individuals in the very-high-risk group were significantly more likely to die of lung cancer. The aHR was 11.1 in the PCLO data set and 8.4 in the NSLT data set (P < .001 for both).

This might be expected for people who were interested in being screened for lung cancer. However, patients in the very-high-risk group were also more likely to die of cardiovascular illness (aHR, 3.6 for PLCO and 47.8 for NSLT; P < .001 for both) and respiratory illness (aHR, 27.5 for PLCO and 31.9 for NLST; P ≤ .001 for both).

With this information, a clinician could initiate additional testing and/or utilize more aggressive surveillance measures. If an oncologist considered therapy for a patient with newly diagnosed cancer, treatment choices and stratification for adverse events would be more intelligently planned.
 

Using AI to predict the risk of lung cancer

In another study, Dr. Aerts and colleagues developed and validated a CNN called CXR-LC, which was based on CXR-risk. The goal of this study was to see if CXR-LC could predict long-term incident lung cancer using data available in the EHR, including chest radiographs, age, sex, and smoking status.

The CXR-LC model was developed using data from the PLCO trial (n = 41,856) and was validated in smokers from the PLCO trial (n = 5,615; 12-year follow-up) as well as heavy smokers from the NLST trial (n = 5,493; 6-year follow-up).

Results showed that CXR-LC was able to predict which patients were at highest risk for developing lung cancer.

CXR-LC had better discrimination for incident lung cancer than did Medicare eligibility in the PLCO data set (area under the curve, 0.755 vs. 0.634; P < .001). And the performance of CXR-LC was similar to that of the PLCO_M2012 risk score in both the PLCO data set (AUC, 0.755 vs. 0.751) and the NLST data set (AUC, 0.659 vs. 0.650).

When they were compared in screening populations of equal size, CXR-LC was more sensitive than Medicare eligibility criteria in the PLCO data set (74.9% vs. 63.8%; P = .012) and missed 30.7% fewer incident lung cancer diagnoses.
 

AI as a substitute for specialized testing and consultation

In a third study, Dr. Aerts and colleagues used a CNN to predict cardiovascular risk by assessing coronary artery calcium (CAC) from clinically obtained, readily available CT scans.

Ordinarily, identifying CAC – an accurate predictor of cardiovascular events – requires specialized expertise (manual measurement and cardiologist interpretation), time (estimated at 20 minutes/scan), and equipment (ECG-gated cardiac CT scan and special software).

In this study, the researchers used a fully end-to-end automated system with analytic time measured in less than 2 seconds.

The team trained and tuned their CNN using the Framingham Heart Study Offspring and Third Generation cohorts (n = 1,636), which included asymptomatic patients with high-quality, cardiac-gated CT scans for CAC quantification.

The researchers then tested the CNN on two asymptomatic and two symptomatic cohorts:

• Asymptomatic Framingham Heart Study participants (n = 663) in whom the outcome measures were cardiovascular disease and death.
• Asymptomatic NLST participants (n = 14,959) in whom the outcome measure was atherosclerotic cardiovascular death.
• Symptomatic PROMISE study participants with stable chest pain (n = 4,021) in whom the outcome measures were all-cause mortality, MI, and hospitalization for unstable angina.
• Symptomatic ROMICAT-II study patients with acute chest pain (n = 441) in whom the outcome measure was acute coronary syndrome at 28 days.

Among 5,521 subjects across all testing cohorts with cardiac-gated and nongated chest CT scans, the CNN and expert reader interpretations agreed on the CAC risk scores with a high level of concordance (kappa, 0.71; concordance rate, 0.79).

There was a very high Spearman’s correlation of 0.92 (P < .0001) and substantial agreement between automatically and manually calculated CAC risk groups, substantiating robust risk prediction for cardiovascular disease across multiple clinical scenarios.

Dr. Aerts commented that, among the NLST participants who had the highest risk of developing lung cancer, the risk of cardiovascular death was as high as the risk of death from lung cancer.
 

Using AI to assess patient outcomes

In an unpublished study, Dr. Aerts and colleagues used AI in an attempt to determine whether changes in measurements of subcutaneous adipose tissue (SAT), visceral adipose tissue (VAT), and skeletal muscle mass would provide clues about treatment outcomes in lung cancer patients.

The researchers developed a deep learning model using data from 1,129 patients at Massachusetts General and Brigham and Women’s Hospitals, measuring SAT, VAT, and muscle mass. The team applied the measurement system to a population of 12,128 outpatients and calculated z scores for SAT, VAT, and muscle mass to determine “normal” values.

When they applied the norms to surgical lung cancer data sets from the Boston Lung Cancer Study (n = 437) and TRACERx study (n = 394), the researchers found that smokers had lower adiposity and lower muscle mass than never-smokers.

More importantly, over time, among lung cancer patients who lost greater than 5% of VAT, SAT, and muscle mass, those patients with the greatest SAT loss (P < .0001) or VAT loss (P = .0015) had the lowest lung cancer–specific survival in the TRACERx study. There was no significant impairment of lung cancer-specific survival for patients who experienced skeletal muscle loss (P = .23).

The same observation was made for overall survival among patients enrolled in the Boston Lung Cancer Study, using the 5% threshold. Overall survival was significantly worse with increasing VAT loss (P = .0023) and SAT loss (P = .0082) but not with increasing skeletal muscle loss (P = .3).

The investigators speculated about whether the correlation between body composition and clinical outcome could yield clues about tumor biology. To test this, the researchers used the RNA sequencing–based ORACLE risk score in lung cancer patients from TRACERx. There was a high correlation between higher ORACLE risk scores and lower VAT and SAT, suggesting that measures of adiposity on CT were reflected in tumor biology patterns on an RNA level in lung cancer patients. There was no such correlation between ORACLE risk scores and skeletal muscle mass.
 

Wonderment ... tempered by concern and challenges

AI has awe-inspiring potential to yield actionable and prognostically important information from data mining the EHR and extracting the vast quantities of information from images. In some cases (like CAC), it is information that is “hiding in plain sight.” However, Dr. Aerts expressed several cautions, some of which have already plagued AI.

He referenced the Gartner Hype Cycle, which provides a graphic representation of five phases in the life cycle of emerging technologies. The “innovation trigger” is followed by a “peak of inflated expectations,” a “trough of disillusionment,” a “slope of enlightenment,” and a “plateau of productivity.”

Dr. Aerts noted that, in recent years, AI has seemed to fall into the trough of disillusionment, but it may be entering the slope of enlightenment on the way to the plateau of productivity.

His research highlighted several examples of productivity in radiomics in cancer patients and those who are at high risk of developing cancer.

In Dr. Aerts’s opinion, a second concern is replication of AI research results. He noted that, among 400 published studies, only 6% of authors shared the codes that would enable their findings to be corroborated. About 30% shared test data, and 54% shared “pseudocodes,” but transparency and reproducibility are problems for the acceptance and broad implementation of AI.

Dr. Aerts endorsed the Modelhub initiative (www.modelhub.ai), a multi-institutional initiative to advance reproducibility in the AI field and advance its full potential.

However, there are additional concerns about the implementation of radiomics and, more generally, data mining from clinicians’ EHRs to personalize care.

Firstly, it may be laborious and difficult to explain complex, computer-based risk stratification models to patients. Hereditary cancer testing is an example of a risk assessment test that requires complicated explanations that many clinicians relegate to genetics counselors – when patients elect to see them. When a model is not explainable, it undermines the confidence of patients and their care providers, according to an editorial related to the CXR-LC study.

Another issue is that uptake of lung cancer screening, in practice, has been underutilized by individuals who meet current, relatively straightforward Medicare criteria. Despite the apparently better accuracy of the CXR-LC deep-learning model, its complexity and limited access could constitute an additional barrier for the at-risk individuals who should avail themselves of screening.

Furthermore, although age and gender are accurate in most circumstances, there is legitimate concern about the accuracy of, for example, smoking history data and comorbid conditions in current EHRs. Who performs the laborious curation of the input in an AI model to assure its accuracy for individual patients?

Finally, it is unclear how scalable and applicable AI will be to medically underserved populations (e.g., smaller, community-based, free-standing, socioeconomically disadvantaged or rural health care institutions). There are substantial initial and maintenance costs that may limit AI’s availability to some academic institutions and large health maintenance organizations.

As the concerns and challenges are addressed, it will be interesting to see where and when the plateau of productivity for AI in cancer care occurs. When it does, many cancer patients will benefit from enhanced care along the continuum of the complex disease they and their caregivers seek to master.

Dr. Aerts disclosed relationships with Onc.AI outside the presented work.

Dr. Lyss was a community-based medical oncologist and clinical researcher for more than 35 years before his recent retirement. His clinical and research interests were focused on breast and lung cancers, as well as expanding clinical trial access to medically underserved populations. He is based in St. Louis. He has no conflicts of interest.

Artificial intelligence (AI) is expected to one day affect the entire continuum of cancer care – from screening and risk prediction to diagnosis, risk stratification, treatment selection, and follow-up, according to an expert in the field.

Hugo J.W.L. Aerts, PhD, director of the AI in Medicine Program at Brigham and Women’s Hospital in Boston, described studies using AI for some of these purposes during a presentation at the AACR Virtual Special Conference: Artificial Intelligence, Diagnosis, and Imaging (Abstract IA-06).

In one study, Dr. Aerts and colleagues set out to determine whether a convolutional neural network (CNN) could extract prognostic information from chest radiographs. The researchers tested this theory using patients from two trials – the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial and the National Lung Screening Trial (NLST).

The team developed a CNN, called CXR-risk, and tested whether it could predict the longevity and prognosis of patients in the PLCO (n = 52,320) and NLST (n = 5,493) trials over a 12-year time period, based only on chest radiographs. No clinical information, demographics, radiographic interpretations, duration of follow-up, or censoring were provided to the deep-learning system.

CXR-risk output was stratified into five categories of radiographic risk scores for probability of death, from 0 (very low likelihood of mortality) to 1 (very high likelihood of mortality).

The investigators found a graded association between radiographic risk score and mortality. The very-high-risk group had mortality rates of 53.0% (PLCO) and 33.9% (NLST). In both trials, this was significantly higher than for the very-low-risk group. The unadjusted hazard ratio was 18.3 in the PCLO data set and 15.2 in the NLST data set (P < .001 for both).

This association was maintained after adjustment for radiologists’ findings (e.g., a lung nodule) and risk factors such as age, gender, and comorbid illnesses like diabetes. The adjusted HR was 4.8 in the PCLO data set and 7.0 in the NLST data set (P < .001 for both).

In both data sets, individuals in the very-high-risk group were significantly more likely to die of lung cancer. The aHR was 11.1 in the PCLO data set and 8.4 in the NSLT data set (P < .001 for both).

This might be expected for people who were interested in being screened for lung cancer. However, patients in the very-high-risk group were also more likely to die of cardiovascular illness (aHR, 3.6 for PLCO and 47.8 for NSLT; P < .001 for both) and respiratory illness (aHR, 27.5 for PLCO and 31.9 for NLST; P ≤ .001 for both).

With this information, a clinician could initiate additional testing and/or utilize more aggressive surveillance measures. If an oncologist considered therapy for a patient with newly diagnosed cancer, treatment choices and stratification for adverse events would be more intelligently planned.
 

Using AI to predict the risk of lung cancer

In another study, Dr. Aerts and colleagues developed and validated a CNN called CXR-LC, which was based on CXR-risk. The goal of this study was to see if CXR-LC could predict long-term incident lung cancer using data available in the EHR, including chest radiographs, age, sex, and smoking status.

The CXR-LC model was developed using data from the PLCO trial (n = 41,856) and was validated in smokers from the PLCO trial (n = 5,615; 12-year follow-up) as well as heavy smokers from the NLST trial (n = 5,493; 6-year follow-up).

Results showed that CXR-LC was able to predict which patients were at highest risk for developing lung cancer.

CXR-LC had better discrimination for incident lung cancer than did Medicare eligibility in the PLCO data set (area under the curve, 0.755 vs. 0.634; P < .001). And the performance of CXR-LC was similar to that of the PLCO_M2012 risk score in both the PLCO data set (AUC, 0.755 vs. 0.751) and the NLST data set (AUC, 0.659 vs. 0.650).

When they were compared in screening populations of equal size, CXR-LC was more sensitive than Medicare eligibility criteria in the PLCO data set (74.9% vs. 63.8%; P = .012) and missed 30.7% fewer incident lung cancer diagnoses.
 

AI as a substitute for specialized testing and consultation

In a third study, Dr. Aerts and colleagues used a CNN to predict cardiovascular risk by assessing coronary artery calcium (CAC) from clinically obtained, readily available CT scans.

Ordinarily, identifying CAC – an accurate predictor of cardiovascular events – requires specialized expertise (manual measurement and cardiologist interpretation), time (estimated at 20 minutes/scan), and equipment (ECG-gated cardiac CT scan and special software).

In this study, the researchers used a fully end-to-end automated system with analytic time measured in less than 2 seconds.

The team trained and tuned their CNN using the Framingham Heart Study Offspring and Third Generation cohorts (n = 1,636), which included asymptomatic patients with high-quality, cardiac-gated CT scans for CAC quantification.

The researchers then tested the CNN on two asymptomatic and two symptomatic cohorts:

• Asymptomatic Framingham Heart Study participants (n = 663) in whom the outcome measures were cardiovascular disease and death.
• Asymptomatic NLST participants (n = 14,959) in whom the outcome measure was atherosclerotic cardiovascular death.
• Symptomatic PROMISE study participants with stable chest pain (n = 4,021) in whom the outcome measures were all-cause mortality, MI, and hospitalization for unstable angina.
• Symptomatic ROMICAT-II study patients with acute chest pain (n = 441) in whom the outcome measure was acute coronary syndrome at 28 days.

Among 5,521 subjects across all testing cohorts with cardiac-gated and nongated chest CT scans, the CNN and expert reader interpretations agreed on the CAC risk scores with a high level of concordance (kappa, 0.71; concordance rate, 0.79).

There was a very high Spearman’s correlation of 0.92 (P < .0001) and substantial agreement between automatically and manually calculated CAC risk groups, substantiating robust risk prediction for cardiovascular disease across multiple clinical scenarios.

Dr. Aerts commented that, among the NLST participants who had the highest risk of developing lung cancer, the risk of cardiovascular death was as high as the risk of death from lung cancer.
 

Using AI to assess patient outcomes

In an unpublished study, Dr. Aerts and colleagues used AI in an attempt to determine whether changes in measurements of subcutaneous adipose tissue (SAT), visceral adipose tissue (VAT), and skeletal muscle mass would provide clues about treatment outcomes in lung cancer patients.

The researchers developed a deep learning model using data from 1,129 patients at Massachusetts General and Brigham and Women’s Hospitals, measuring SAT, VAT, and muscle mass. The team applied the measurement system to a population of 12,128 outpatients and calculated z scores for SAT, VAT, and muscle mass to determine “normal” values.

When they applied the norms to surgical lung cancer data sets from the Boston Lung Cancer Study (n = 437) and TRACERx study (n = 394), the researchers found that smokers had lower adiposity and lower muscle mass than never-smokers.

More importantly, over time, among lung cancer patients who lost greater than 5% of VAT, SAT, and muscle mass, those patients with the greatest SAT loss (P < .0001) or VAT loss (P = .0015) had the lowest lung cancer–specific survival in the TRACERx study. There was no significant impairment of lung cancer-specific survival for patients who experienced skeletal muscle loss (P = .23).

The same observation was made for overall survival among patients enrolled in the Boston Lung Cancer Study, using the 5% threshold. Overall survival was significantly worse with increasing VAT loss (P = .0023) and SAT loss (P = .0082) but not with increasing skeletal muscle loss (P = .3).

The investigators speculated about whether the correlation between body composition and clinical outcome could yield clues about tumor biology. To test this, the researchers used the RNA sequencing–based ORACLE risk score in lung cancer patients from TRACERx. There was a high correlation between higher ORACLE risk scores and lower VAT and SAT, suggesting that measures of adiposity on CT were reflected in tumor biology patterns on an RNA level in lung cancer patients. There was no such correlation between ORACLE risk scores and skeletal muscle mass.
 

Wonderment ... tempered by concern and challenges

AI has awe-inspiring potential to yield actionable and prognostically important information from data mining the EHR and extracting the vast quantities of information from images. In some cases (like CAC), it is information that is “hiding in plain sight.” However, Dr. Aerts expressed several cautions, some of which have already plagued AI.

He referenced the Gartner Hype Cycle, which provides a graphic representation of five phases in the life cycle of emerging technologies. The “innovation trigger” is followed by a “peak of inflated expectations,” a “trough of disillusionment,” a “slope of enlightenment,” and a “plateau of productivity.”

Dr. Aerts noted that, in recent years, AI has seemed to fall into the trough of disillusionment, but it may be entering the slope of enlightenment on the way to the plateau of productivity.

His research highlighted several examples of productivity in radiomics in cancer patients and those who are at high risk of developing cancer.

In Dr. Aerts’s opinion, a second concern is replication of AI research results. He noted that, among 400 published studies, only 6% of authors shared the codes that would enable their findings to be corroborated. About 30% shared test data, and 54% shared “pseudocodes,” but transparency and reproducibility are problems for the acceptance and broad implementation of AI.

Dr. Aerts endorsed the Modelhub initiative (www.modelhub.ai), a multi-institutional initiative to advance reproducibility in the AI field and advance its full potential.

However, there are additional concerns about the implementation of radiomics and, more generally, data mining from clinicians’ EHRs to personalize care.

Firstly, it may be laborious and difficult to explain complex, computer-based risk stratification models to patients. Hereditary cancer testing is an example of a risk assessment test that requires complicated explanations that many clinicians relegate to genetics counselors – when patients elect to see them. When a model is not explainable, it undermines the confidence of patients and their care providers, according to an editorial related to the CXR-LC study.

Another issue is that uptake of lung cancer screening, in practice, has been underutilized by individuals who meet current, relatively straightforward Medicare criteria. Despite the apparently better accuracy of the CXR-LC deep-learning model, its complexity and limited access could constitute an additional barrier for the at-risk individuals who should avail themselves of screening.

Furthermore, although age and gender are accurate in most circumstances, there is legitimate concern about the accuracy of, for example, smoking history data and comorbid conditions in current EHRs. Who performs the laborious curation of the input in an AI model to assure its accuracy for individual patients?

Finally, it is unclear how scalable and applicable AI will be to medically underserved populations (e.g., smaller, community-based, free-standing, socioeconomically disadvantaged or rural health care institutions). There are substantial initial and maintenance costs that may limit AI’s availability to some academic institutions and large health maintenance organizations.

As the concerns and challenges are addressed, it will be interesting to see where and when the plateau of productivity for AI in cancer care occurs. When it does, many cancer patients will benefit from enhanced care along the continuum of the complex disease they and their caregivers seek to master.

Dr. Aerts disclosed relationships with Onc.AI outside the presented work.

Dr. Lyss was a community-based medical oncologist and clinical researcher for more than 35 years before his recent retirement. His clinical and research interests were focused on breast and lung cancers, as well as expanding clinical trial access to medically underserved populations. He is based in St. Louis. He has no conflicts of interest.

Artificial intelligence (AI) is expected to one day affect the entire continuum of cancer care – from screening and risk prediction to diagnosis, risk stratification, treatment selection, and follow-up, according to an expert in the field.

Hugo J.W.L. Aerts, PhD, director of the AI in Medicine Program at Brigham and Women’s Hospital in Boston, described studies using AI for some of these purposes during a presentation at the AACR Virtual Special Conference: Artificial Intelligence, Diagnosis, and Imaging (Abstract IA-06).

In one study, Dr. Aerts and colleagues set out to determine whether a convolutional neural network (CNN) could extract prognostic information from chest radiographs. The researchers tested this theory using patients from two trials – the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial and the National Lung Screening Trial (NLST).

The team developed a CNN, called CXR-risk, and tested whether it could predict the longevity and prognosis of patients in the PLCO (n = 52,320) and NLST (n = 5,493) trials over a 12-year time period, based only on chest radiographs. No clinical information, demographics, radiographic interpretations, duration of follow-up, or censoring were provided to the deep-learning system.

CXR-risk output was stratified into five categories of radiographic risk scores for probability of death, from 0 (very low likelihood of mortality) to 1 (very high likelihood of mortality).

The investigators found a graded association between radiographic risk score and mortality. The very-high-risk group had mortality rates of 53.0% (PLCO) and 33.9% (NLST). In both trials, this was significantly higher than for the very-low-risk group. The unadjusted hazard ratio was 18.3 in the PCLO data set and 15.2 in the NLST data set (P < .001 for both).

This association was maintained after adjustment for radiologists’ findings (e.g., a lung nodule) and risk factors such as age, gender, and comorbid illnesses like diabetes. The adjusted HR was 4.8 in the PCLO data set and 7.0 in the NLST data set (P < .001 for both).

In both data sets, individuals in the very-high-risk group were significantly more likely to die of lung cancer. The aHR was 11.1 in the PCLO data set and 8.4 in the NSLT data set (P < .001 for both).

This might be expected for people who were interested in being screened for lung cancer. However, patients in the very-high-risk group were also more likely to die of cardiovascular illness (aHR, 3.6 for PLCO and 47.8 for NSLT; P < .001 for both) and respiratory illness (aHR, 27.5 for PLCO and 31.9 for NLST; P ≤ .001 for both).

With this information, a clinician could initiate additional testing and/or utilize more aggressive surveillance measures. If an oncologist considered therapy for a patient with newly diagnosed cancer, treatment choices and stratification for adverse events would be more intelligently planned.
 

Using AI to predict the risk of lung cancer

In another study, Dr. Aerts and colleagues developed and validated a CNN called CXR-LC, which was based on CXR-risk. The goal of this study was to see if CXR-LC could predict long-term incident lung cancer using data available in the EHR, including chest radiographs, age, sex, and smoking status.

The CXR-LC model was developed using data from the PLCO trial (n = 41,856) and was validated in smokers from the PLCO trial (n = 5,615; 12-year follow-up) as well as heavy smokers from the NLST trial (n = 5,493; 6-year follow-up).

Results showed that CXR-LC was able to predict which patients were at highest risk for developing lung cancer.

CXR-LC had better discrimination for incident lung cancer than did Medicare eligibility in the PLCO data set (area under the curve, 0.755 vs. 0.634; P < .001). And the performance of CXR-LC was similar to that of the PLCO_M2012 risk score in both the PLCO data set (AUC, 0.755 vs. 0.751) and the NLST data set (AUC, 0.659 vs. 0.650).

When they were compared in screening populations of equal size, CXR-LC was more sensitive than Medicare eligibility criteria in the PLCO data set (74.9% vs. 63.8%; P = .012) and missed 30.7% fewer incident lung cancer diagnoses.
 

AI as a substitute for specialized testing and consultation

In a third study, Dr. Aerts and colleagues used a CNN to predict cardiovascular risk by assessing coronary artery calcium (CAC) from clinically obtained, readily available CT scans.

Ordinarily, identifying CAC – an accurate predictor of cardiovascular events – requires specialized expertise (manual measurement and cardiologist interpretation), time (estimated at 20 minutes/scan), and equipment (ECG-gated cardiac CT scan and special software).

In this study, the researchers used a fully end-to-end automated system with analytic time measured in less than 2 seconds.

The team trained and tuned their CNN using the Framingham Heart Study Offspring and Third Generation cohorts (n = 1,636), which included asymptomatic patients with high-quality, cardiac-gated CT scans for CAC quantification.

The researchers then tested the CNN on two asymptomatic and two symptomatic cohorts:

• Asymptomatic Framingham Heart Study participants (n = 663) in whom the outcome measures were cardiovascular disease and death.
• Asymptomatic NLST participants (n = 14,959) in whom the outcome measure was atherosclerotic cardiovascular death.
• Symptomatic PROMISE study participants with stable chest pain (n = 4,021) in whom the outcome measures were all-cause mortality, MI, and hospitalization for unstable angina.
• Symptomatic ROMICAT-II study patients with acute chest pain (n = 441) in whom the outcome measure was acute coronary syndrome at 28 days.

Among 5,521 subjects across all testing cohorts with cardiac-gated and nongated chest CT scans, the CNN and expert reader interpretations agreed on the CAC risk scores with a high level of concordance (kappa, 0.71; concordance rate, 0.79).

There was a very high Spearman’s correlation of 0.92 (P < .0001) and substantial agreement between automatically and manually calculated CAC risk groups, substantiating robust risk prediction for cardiovascular disease across multiple clinical scenarios.

Dr. Aerts commented that, among the NLST participants who had the highest risk of developing lung cancer, the risk of cardiovascular death was as high as the risk of death from lung cancer.
 

Using AI to assess patient outcomes

In an unpublished study, Dr. Aerts and colleagues used AI in an attempt to determine whether changes in measurements of subcutaneous adipose tissue (SAT), visceral adipose tissue (VAT), and skeletal muscle mass would provide clues about treatment outcomes in lung cancer patients.

The researchers developed a deep learning model using data from 1,129 patients at Massachusetts General and Brigham and Women’s Hospitals, measuring SAT, VAT, and muscle mass. The team applied the measurement system to a population of 12,128 outpatients and calculated z scores for SAT, VAT, and muscle mass to determine “normal” values.

When they applied the norms to surgical lung cancer data sets from the Boston Lung Cancer Study (n = 437) and TRACERx study (n = 394), the researchers found that smokers had lower adiposity and lower muscle mass than never-smokers.

More importantly, over time, among lung cancer patients who lost greater than 5% of VAT, SAT, and muscle mass, those patients with the greatest SAT loss (P < .0001) or VAT loss (P = .0015) had the lowest lung cancer–specific survival in the TRACERx study. There was no significant impairment of lung cancer-specific survival for patients who experienced skeletal muscle loss (P = .23).

The same observation was made for overall survival among patients enrolled in the Boston Lung Cancer Study, using the 5% threshold. Overall survival was significantly worse with increasing VAT loss (P = .0023) and SAT loss (P = .0082) but not with increasing skeletal muscle loss (P = .3).

The investigators speculated about whether the correlation between body composition and clinical outcome could yield clues about tumor biology. To test this, the researchers used the RNA sequencing–based ORACLE risk score in lung cancer patients from TRACERx. There was a high correlation between higher ORACLE risk scores and lower VAT and SAT, suggesting that measures of adiposity on CT were reflected in tumor biology patterns on an RNA level in lung cancer patients. There was no such correlation between ORACLE risk scores and skeletal muscle mass.
 

Wonderment ... tempered by concern and challenges

AI has awe-inspiring potential to yield actionable and prognostically important information from data mining the EHR and extracting the vast quantities of information from images. In some cases (like CAC), it is information that is “hiding in plain sight.” However, Dr. Aerts expressed several cautions, some of which have already plagued AI.

He referenced the Gartner Hype Cycle, which provides a graphic representation of five phases in the life cycle of emerging technologies. The “innovation trigger” is followed by a “peak of inflated expectations,” a “trough of disillusionment,” a “slope of enlightenment,” and a “plateau of productivity.”

Dr. Aerts noted that, in recent years, AI has seemed to fall into the trough of disillusionment, but it may be entering the slope of enlightenment on the way to the plateau of productivity.

His research highlighted several examples of productivity in radiomics in cancer patients and those who are at high risk of developing cancer.

In Dr. Aerts’s opinion, a second concern is replication of AI research results. He noted that, among 400 published studies, only 6% of authors shared the codes that would enable their findings to be corroborated. About 30% shared test data, and 54% shared “pseudocodes,” but transparency and reproducibility are problems for the acceptance and broad implementation of AI.

Dr. Aerts endorsed the Modelhub initiative (www.modelhub.ai), a multi-institutional initiative to advance reproducibility in the AI field and advance its full potential.

However, there are additional concerns about the implementation of radiomics and, more generally, data mining from clinicians’ EHRs to personalize care.

Firstly, it may be laborious and difficult to explain complex, computer-based risk stratification models to patients. Hereditary cancer testing is an example of a risk assessment test that requires complicated explanations that many clinicians relegate to genetics counselors – when patients elect to see them. When a model is not explainable, it undermines the confidence of patients and their care providers, according to an editorial related to the CXR-LC study.

Another issue is that uptake of lung cancer screening, in practice, has been underutilized by individuals who meet current, relatively straightforward Medicare criteria. Despite the apparently better accuracy of the CXR-LC deep-learning model, its complexity and limited access could constitute an additional barrier for the at-risk individuals who should avail themselves of screening.

Furthermore, although age and gender are accurate in most circumstances, there is legitimate concern about the accuracy of, for example, smoking history data and comorbid conditions in current EHRs. Who performs the laborious curation of the input in an AI model to assure its accuracy for individual patients?

Finally, it is unclear how scalable and applicable AI will be to medically underserved populations (e.g., smaller, community-based, free-standing, socioeconomically disadvantaged or rural health care institutions). There are substantial initial and maintenance costs that may limit AI’s availability to some academic institutions and large health maintenance organizations.

As the concerns and challenges are addressed, it will be interesting to see where and when the plateau of productivity for AI in cancer care occurs. When it does, many cancer patients will benefit from enhanced care along the continuum of the complex disease they and their caregivers seek to master.

Dr. Aerts disclosed relationships with Onc.AI outside the presented work.

Dr. Lyss was a community-based medical oncologist and clinical researcher for more than 35 years before his recent retirement. His clinical and research interests were focused on breast and lung cancers, as well as expanding clinical trial access to medically underserved populations. He is based in St. Louis. He has no conflicts of interest.

A novel drug that offers multilineage protection from chemotherapy-induced myelosuppression has been approved by the Food and Drug Administration.

The drug, trilaciclib (Cosela, G1 Therapeutics) is administered intravenously as a 30-minute infusion within 4 hours prior to the start of chemotherapy. It is indicated specifically for use in adults with extensive-stage small-cell lung cancer (ES-SCLC) who are receiving chemotherapy.

Trilaciclib is a CDK4/6 inhibitor, and this action appears to protect normal bone marrow cells from the harmful effects of chemotherapy.

“For patients with extensive-stage small-cell lung cancer, protecting bone marrow function may help make their chemotherapy safer and allow them to complete their course of treatment on time and according to plan,” Albert Deisseroth, MD, PhD, of the FDA’s Center for Drug Evaluation and Research, said in an FDA press release.
 

First drug of its type

Trilaciclib “is the first and only therapy designed to help protect bone marrow (myeloprotection) when administered prior to treatment with chemotherapy,” according to the drug’s manufacturer.

Myelosuppression is one of the most severe adverse effects of chemotherapy, and it can be life-threatening. It can increase the risk of infection and lead to severe anemia and/or bleeding.

“These complications impact patients’ quality of life and may also result in chemotherapy dose reductions and delays,” Jeffrey Crawford, MD, of Duke Cancer Institute, Durham, N.C., said in a company press release.

“To date, approaches have included the use of growth factor agents to accelerate blood cell recovery after the bone marrow injury has occurred, along with antibiotics and transfusions as needed. By contrast, trilaciclib provides the first proactive approach to myelosuppression through a unique mechanism of action that helps protect the bone marrow from damage by chemotherapy.”
 

Approval based on randomized, placebo-controlled trials

The approval of trilaciclib is based on data from three randomized, double-blind, placebo-controlled studies, involving a total of 245 patients with ES-SCLC.

These patients were being treated with chemotherapy regimens that were based on the combination of carboplatin and etoposide (with or without the immunotherapy atezolizumab) or regimens that were based on topotecan.

Before receiving the chemotherapy, patients were randomly assigned to receive trilaciclib or placebo.

Results showed that patients who had received an infusion of trilaciclib before receiving chemotherapy had a lower chance of developing severe neutropenia compared with patients who received a placebo, the FDA noted. In addition, among the patients who did develop severe neutropenia, this had a shorter duration among patients who received trilaciclib than among those who received placebo.

The most common side effects of trilaciclib were fatigue; low levels of calcium, potassium, and phosphate in the blood; increased levels of aspartate aminotransferase; headache; and pneumonia.

The FDA noted that patients should also be advised about injection site reactions, acute drug hypersensitivity, interstitial lung disease/pneumonitis, and embryo-fetal toxicity.

The approval received a priority review, based on the drug’s breakthrough therapy designation. As is common for such products, the company plans postmarketing activities that will assess the effects of trilaciclib on disease progression or survival with at least a 2-year follow up. This clinical trial is scheduled to start in 2022.

A version of this article first appeared on Medscape.com.

A novel drug that offers multilineage protection from chemotherapy-induced myelosuppression has been approved by the Food and Drug Administration.

The drug, trilaciclib (Cosela, G1 Therapeutics) is administered intravenously as a 30-minute infusion within 4 hours prior to the start of chemotherapy. It is indicated specifically for use in adults with extensive-stage small-cell lung cancer (ES-SCLC) who are receiving chemotherapy.

Trilaciclib is a CDK4/6 inhibitor, and this action appears to protect normal bone marrow cells from the harmful effects of chemotherapy.

“For patients with extensive-stage small-cell lung cancer, protecting bone marrow function may help make their chemotherapy safer and allow them to complete their course of treatment on time and according to plan,” Albert Deisseroth, MD, PhD, of the FDA’s Center for Drug Evaluation and Research, said in an FDA press release.
 

First drug of its type

Trilaciclib “is the first and only therapy designed to help protect bone marrow (myeloprotection) when administered prior to treatment with chemotherapy,” according to the drug’s manufacturer.

Myelosuppression is one of the most severe adverse effects of chemotherapy, and it can be life-threatening. It can increase the risk of infection and lead to severe anemia and/or bleeding.

“These complications impact patients’ quality of life and may also result in chemotherapy dose reductions and delays,” Jeffrey Crawford, MD, of Duke Cancer Institute, Durham, N.C., said in a company press release.

“To date, approaches have included the use of growth factor agents to accelerate blood cell recovery after the bone marrow injury has occurred, along with antibiotics and transfusions as needed. By contrast, trilaciclib provides the first proactive approach to myelosuppression through a unique mechanism of action that helps protect the bone marrow from damage by chemotherapy.”
 

Approval based on randomized, placebo-controlled trials

The approval of trilaciclib is based on data from three randomized, double-blind, placebo-controlled studies, involving a total of 245 patients with ES-SCLC.

These patients were being treated with chemotherapy regimens that were based on the combination of carboplatin and etoposide (with or without the immunotherapy atezolizumab) or regimens that were based on topotecan.

Before receiving the chemotherapy, patients were randomly assigned to receive trilaciclib or placebo.

Results showed that patients who had received an infusion of trilaciclib before receiving chemotherapy had a lower chance of developing severe neutropenia compared with patients who received a placebo, the FDA noted. In addition, among the patients who did develop severe neutropenia, this had a shorter duration among patients who received trilaciclib than among those who received placebo.

The most common side effects of trilaciclib were fatigue; low levels of calcium, potassium, and phosphate in the blood; increased levels of aspartate aminotransferase; headache; and pneumonia.

The FDA noted that patients should also be advised about injection site reactions, acute drug hypersensitivity, interstitial lung disease/pneumonitis, and embryo-fetal toxicity.

The approval received a priority review, based on the drug’s breakthrough therapy designation. As is common for such products, the company plans postmarketing activities that will assess the effects of trilaciclib on disease progression or survival with at least a 2-year follow up. This clinical trial is scheduled to start in 2022.

A version of this article first appeared on Medscape.com.

A novel drug that offers multilineage protection from chemotherapy-induced myelosuppression has been approved by the Food and Drug Administration.

The drug, trilaciclib (Cosela, G1 Therapeutics) is administered intravenously as a 30-minute infusion within 4 hours prior to the start of chemotherapy. It is indicated specifically for use in adults with extensive-stage small-cell lung cancer (ES-SCLC) who are receiving chemotherapy.

Trilaciclib is a CDK4/6 inhibitor, and this action appears to protect normal bone marrow cells from the harmful effects of chemotherapy.

“For patients with extensive-stage small-cell lung cancer, protecting bone marrow function may help make their chemotherapy safer and allow them to complete their course of treatment on time and according to plan,” Albert Deisseroth, MD, PhD, of the FDA’s Center for Drug Evaluation and Research, said in an FDA press release.
 

First drug of its type

Trilaciclib “is the first and only therapy designed to help protect bone marrow (myeloprotection) when administered prior to treatment with chemotherapy,” according to the drug’s manufacturer.

Myelosuppression is one of the most severe adverse effects of chemotherapy, and it can be life-threatening. It can increase the risk of infection and lead to severe anemia and/or bleeding.

“These complications impact patients’ quality of life and may also result in chemotherapy dose reductions and delays,” Jeffrey Crawford, MD, of Duke Cancer Institute, Durham, N.C., said in a company press release.

“To date, approaches have included the use of growth factor agents to accelerate blood cell recovery after the bone marrow injury has occurred, along with antibiotics and transfusions as needed. By contrast, trilaciclib provides the first proactive approach to myelosuppression through a unique mechanism of action that helps protect the bone marrow from damage by chemotherapy.”
 

Approval based on randomized, placebo-controlled trials

The approval of trilaciclib is based on data from three randomized, double-blind, placebo-controlled studies, involving a total of 245 patients with ES-SCLC.

These patients were being treated with chemotherapy regimens that were based on the combination of carboplatin and etoposide (with or without the immunotherapy atezolizumab) or regimens that were based on topotecan.

Before receiving the chemotherapy, patients were randomly assigned to receive trilaciclib or placebo.

Results showed that patients who had received an infusion of trilaciclib before receiving chemotherapy had a lower chance of developing severe neutropenia compared with patients who received a placebo, the FDA noted. In addition, among the patients who did develop severe neutropenia, this had a shorter duration among patients who received trilaciclib than among those who received placebo.

The most common side effects of trilaciclib were fatigue; low levels of calcium, potassium, and phosphate in the blood; increased levels of aspartate aminotransferase; headache; and pneumonia.

The FDA noted that patients should also be advised about injection site reactions, acute drug hypersensitivity, interstitial lung disease/pneumonitis, and embryo-fetal toxicity.

The approval received a priority review, based on the drug’s breakthrough therapy designation. As is common for such products, the company plans postmarketing activities that will assess the effects of trilaciclib on disease progression or survival with at least a 2-year follow up. This clinical trial is scheduled to start in 2022.

A version of this article first appeared on Medscape.com.

Longer-term survival with immunotherapy for patients with non–small cell lung cancer (NSCLC) is once again being applauded by experts in the field.

This time, the data come from trials that tested immunotherapy in the second-line setting for patients who had experienced disease progression with platinum-based chemotherapy. The latest 5-year follow-up from two landmark trials, one with pembrolizumab, the other with nivolumab, show that the survival benefit can persist for years after treatment is stopped.

“These are unprecedented data,” Fred R. Hirsch, MD, PhD, executive director of the Center for Thoracic Oncology at the Tisch Cancer Institute, New York, said in an interview. He was not involved in either trial and was approached for comment.
 

Pembrolizumab survival data

The new longer-term data on pembrolizumab come from the KEYNOTE-010 trial, which included more than 1,000 patients with advanced NSCLC who had previously undergone treatment with platinum-based chemotherapy. The patients were randomly assigned to receive either pembrolizumab or docetaxel for 2 years.

This is the latest update on data from this trial, which has been described as “really extraordinary.”

The 5-year overall survival rates were more than doubled in the pembrolizumab groups, compared with the docetaxel group, reported Roy Herbst, MD, PhD, department of medical oncology, Yale Comprehensive Cancer Center, New Haven, Conn.. He was presenting the new data at the recent World Conference on Lung Cancer 2020.

Overall results for patients with programmed death-ligand 1 (PD-L1) Tumor Proportion Score (TPS) expression greater than 1% show that 15.6% of the pembrolizumab group were still alive at 5 years versus 6.5% of the docetaxel group.

The results were even better among patients who had high PD-L1 TPS expression (>50%): in this subgroup, 25% of the patients who received pembrolizumab were still alive versus 8.2% of those who received docetaxel.

In addition, at 5 years, 9.4% of patients who received pembrolizumab were disease free versus 0.7% of the patients who received docetaxel, Dr. Herbst reported.

Dr. Hirsch commented that the 5-year survival rate of 25% among patients with high PD-L1 expression who underwent treatment with pembrolizumab is “great progress in lung cancer treatment, there is no doubt about it.”

He noted that the results also show that “numerically,” it matters whether patients have low PD-L1 expression. “We know from first-line studies that pembrolizumab monotherapy is effective in high PD-L1–expressing tumors, so these data fit very well,” he said.

At the meeting, Dr. Herbst summarized his presentation on pembrolizumab for patients with NSCLC who had previously undergone treatment, saying that, “with 5 years of follow-up, we continue to see a clinically meaningful improvement in overall survival and PFS [progression-free survival].

“Pembrolizumab monotherapy is a standard of care in patients with immunotherapy-naive or previously treated PD-L1–positive advanced non–small cell lung cancer,” Herbst stated.

Dr. Hirsch was largely in agreement. He believes that, for patients with a PD-L1 TPS of at least 50%, the standard of care “is practically pembrolizumab monotherapy, unless there are certain circumstances where you would add chemotherapy,” such as for patients with a high tumor volume, “where you want to see a very quick response.”

Dr. Hirsch pointed out, however, that currently most patients with high PD-L1–expressing tumors are given pembrolizumab in the first line, which begs the question as to what to give those who experience disease progression after immunotherapy.

“That is an open space,” he said. “There is a lot of studies going on in what we call the immunotherapy-refractory patients.

“We don’t have clear guidance for clinical practice yet,” he commented. He noted that there are several options: “Do you continue with chemotherapy? Do you continue with chemotherapy plus another immunotherapy? Do you switch to another immunotherapy?”

Commenting on Twitter, Stephen V. Liu, MD, director of thoracic oncology at Georgetown University, Washington, said the results were “very exciting.”

However, he wondered whether the results suggest that patients with high PD-L1 expression “may be able to stop” receiving pembrolizumab, whereas those with disease of lower expression “may need longer therapy.”

H. Jack West, MD, medical director of the thoracic oncology program, Swedish Cancer Institute, Seattle, said on Twitter that, to him, the “most impressive” aspect was the “new insight about patients stopping pembro after 2 years but still having two-thirds with sustained response.”

He added that he would “love to learn which patients can stop therapy and when, or whether we can do infrequent maintenance IO [immunotherapy].”

 

Nivolumab survival data

The data on nivolumab come from a pooled analysis of 5-year data on 854 patients from CheckMate 057 and CheckMate 017. The analysis was published in the Journal of Clinical Oncology on Jan. 15, 2021.

Both of these trials compared nivolumab with docetaxel for patients with NSCLC who had experienced disease progression with platinum-based chemotherapy.

The pooled analysis showed that the 5-year overall survival rate was more than fivefold greater with nivolumab than with docetaxel, at 13.4% versus 2.6%.

Moreover, more than 80% of patients who had not experienced progression with the immunotherapy at 2 years were still alive at 5 years. The percentage rose to more than 90% among those who had not experienced progression at 3 years.

Lead author Julie R. Brahmer, MD, Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University, Baltimore, and colleagues said the results “demonstrate that nivolumab can provide long-term survival benefit with durable responses and a tolerable safety profile in patients with previously treated, advanced NSCLC.

“Furthermore, some patients appear to maintain prolonged disease control even after stopping systemic therapy,” they noted.

Dr. Hirsch commented that, although the survival rates with nivolumab were slightly lower than reported with pembrolizumab in KEYNOTE-010, they could still be “within the range.” He added that “I wouldn’t conclude that pembrolizumab is better than nivolumab.”

Many factors may account for these differences, he suggested, including differences in the patient populations or simply differences in the numbers of patients included.

For him, the “main point” of the new data from both trials is that immunotherapy has shown “tremendous progress, compared to chemotherapy.”

KEYNOTE-010 was sponsored by Merck Sharp & Dohme. CheckMate 017 and CheckMate057 were sponsored by Bristol-Myers Squibb. Dr. Herbst has relationships with Jun Shi Pharmaceuticals, AstraZeneca, Genentech, Merck, Pfizer, AbbVie, Biodesix, Bristol-Myers Squibb, Eli Lilly, EMD Serono, Heat Biologics, Loxo, Nektar, NextCure, Novartis, Sanofi, Seattle Genetics, Shire, Spectrum Pharmaceuticals, Symphogen, Tesaro, Neon Therapeutics, Infinity Pharmaceuticals, Armo Biosciences, Genmab, Halozyme, and Tocagen. Dr. Brahmer has relationships with Roche/Genentech, Bristol-Myers Squibb, Lilly, Celgene, Syndax, Janssen Oncology, Merck, Amgen, Genentech, AstraZeneca, Incyte, Spectrum Pharmaceuticals, Revolution, and Roche/Genentech.

A version of this article first appeared on Medscape.com.

Longer-term survival with immunotherapy for patients with non–small cell lung cancer (NSCLC) is once again being applauded by experts in the field.

This time, the data come from trials that tested immunotherapy in the second-line setting for patients who had experienced disease progression with platinum-based chemotherapy. The latest 5-year follow-up from two landmark trials, one with pembrolizumab, the other with nivolumab, show that the survival benefit can persist for years after treatment is stopped.

“These are unprecedented data,” Fred R. Hirsch, MD, PhD, executive director of the Center for Thoracic Oncology at the Tisch Cancer Institute, New York, said in an interview. He was not involved in either trial and was approached for comment.
 

Pembrolizumab survival data

The new longer-term data on pembrolizumab come from the KEYNOTE-010 trial, which included more than 1,000 patients with advanced NSCLC who had previously undergone treatment with platinum-based chemotherapy. The patients were randomly assigned to receive either pembrolizumab or docetaxel for 2 years.

This is the latest update on data from this trial, which has been described as “really extraordinary.”

The 5-year overall survival rates were more than doubled in the pembrolizumab groups, compared with the docetaxel group, reported Roy Herbst, MD, PhD, department of medical oncology, Yale Comprehensive Cancer Center, New Haven, Conn.. He was presenting the new data at the recent World Conference on Lung Cancer 2020.

Overall results for patients with programmed death-ligand 1 (PD-L1) Tumor Proportion Score (TPS) expression greater than 1% show that 15.6% of the pembrolizumab group were still alive at 5 years versus 6.5% of the docetaxel group.

The results were even better among patients who had high PD-L1 TPS expression (>50%): in this subgroup, 25% of the patients who received pembrolizumab were still alive versus 8.2% of those who received docetaxel.

In addition, at 5 years, 9.4% of patients who received pembrolizumab were disease free versus 0.7% of the patients who received docetaxel, Dr. Herbst reported.

Dr. Hirsch commented that the 5-year survival rate of 25% among patients with high PD-L1 expression who underwent treatment with pembrolizumab is “great progress in lung cancer treatment, there is no doubt about it.”

He noted that the results also show that “numerically,” it matters whether patients have low PD-L1 expression. “We know from first-line studies that pembrolizumab monotherapy is effective in high PD-L1–expressing tumors, so these data fit very well,” he said.

At the meeting, Dr. Herbst summarized his presentation on pembrolizumab for patients with NSCLC who had previously undergone treatment, saying that, “with 5 years of follow-up, we continue to see a clinically meaningful improvement in overall survival and PFS [progression-free survival].

“Pembrolizumab monotherapy is a standard of care in patients with immunotherapy-naive or previously treated PD-L1–positive advanced non–small cell lung cancer,” Herbst stated.

Dr. Hirsch was largely in agreement. He believes that, for patients with a PD-L1 TPS of at least 50%, the standard of care “is practically pembrolizumab monotherapy, unless there are certain circumstances where you would add chemotherapy,” such as for patients with a high tumor volume, “where you want to see a very quick response.”

Dr. Hirsch pointed out, however, that currently most patients with high PD-L1–expressing tumors are given pembrolizumab in the first line, which begs the question as to what to give those who experience disease progression after immunotherapy.

“That is an open space,” he said. “There is a lot of studies going on in what we call the immunotherapy-refractory patients.

“We don’t have clear guidance for clinical practice yet,” he commented. He noted that there are several options: “Do you continue with chemotherapy? Do you continue with chemotherapy plus another immunotherapy? Do you switch to another immunotherapy?”

Commenting on Twitter, Stephen V. Liu, MD, director of thoracic oncology at Georgetown University, Washington, said the results were “very exciting.”

However, he wondered whether the results suggest that patients with high PD-L1 expression “may be able to stop” receiving pembrolizumab, whereas those with disease of lower expression “may need longer therapy.”

H. Jack West, MD, medical director of the thoracic oncology program, Swedish Cancer Institute, Seattle, said on Twitter that, to him, the “most impressive” aspect was the “new insight about patients stopping pembro after 2 years but still having two-thirds with sustained response.”

He added that he would “love to learn which patients can stop therapy and when, or whether we can do infrequent maintenance IO [immunotherapy].”

 

Nivolumab survival data

The data on nivolumab come from a pooled analysis of 5-year data on 854 patients from CheckMate 057 and CheckMate 017. The analysis was published in the Journal of Clinical Oncology on Jan. 15, 2021.

Both of these trials compared nivolumab with docetaxel for patients with NSCLC who had experienced disease progression with platinum-based chemotherapy.

The pooled analysis showed that the 5-year overall survival rate was more than fivefold greater with nivolumab than with docetaxel, at 13.4% versus 2.6%.

Moreover, more than 80% of patients who had not experienced progression with the immunotherapy at 2 years were still alive at 5 years. The percentage rose to more than 90% among those who had not experienced progression at 3 years.

Lead author Julie R. Brahmer, MD, Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University, Baltimore, and colleagues said the results “demonstrate that nivolumab can provide long-term survival benefit with durable responses and a tolerable safety profile in patients with previously treated, advanced NSCLC.

“Furthermore, some patients appear to maintain prolonged disease control even after stopping systemic therapy,” they noted.

Dr. Hirsch commented that, although the survival rates with nivolumab were slightly lower than reported with pembrolizumab in KEYNOTE-010, they could still be “within the range.” He added that “I wouldn’t conclude that pembrolizumab is better than nivolumab.”

Many factors may account for these differences, he suggested, including differences in the patient populations or simply differences in the numbers of patients included.

For him, the “main point” of the new data from both trials is that immunotherapy has shown “tremendous progress, compared to chemotherapy.”

KEYNOTE-010 was sponsored by Merck Sharp & Dohme. CheckMate 017 and CheckMate057 were sponsored by Bristol-Myers Squibb. Dr. Herbst has relationships with Jun Shi Pharmaceuticals, AstraZeneca, Genentech, Merck, Pfizer, AbbVie, Biodesix, Bristol-Myers Squibb, Eli Lilly, EMD Serono, Heat Biologics, Loxo, Nektar, NextCure, Novartis, Sanofi, Seattle Genetics, Shire, Spectrum Pharmaceuticals, Symphogen, Tesaro, Neon Therapeutics, Infinity Pharmaceuticals, Armo Biosciences, Genmab, Halozyme, and Tocagen. Dr. Brahmer has relationships with Roche/Genentech, Bristol-Myers Squibb, Lilly, Celgene, Syndax, Janssen Oncology, Merck, Amgen, Genentech, AstraZeneca, Incyte, Spectrum Pharmaceuticals, Revolution, and Roche/Genentech.

A version of this article first appeared on Medscape.com.

Longer-term survival with immunotherapy for patients with non–small cell lung cancer (NSCLC) is once again being applauded by experts in the field.

This time, the data come from trials that tested immunotherapy in the second-line setting for patients who had experienced disease progression with platinum-based chemotherapy. The latest 5-year follow-up from two landmark trials, one with pembrolizumab, the other with nivolumab, show that the survival benefit can persist for years after treatment is stopped.

“These are unprecedented data,” Fred R. Hirsch, MD, PhD, executive director of the Center for Thoracic Oncology at the Tisch Cancer Institute, New York, said in an interview. He was not involved in either trial and was approached for comment.
 

Pembrolizumab survival data

The new longer-term data on pembrolizumab come from the KEYNOTE-010 trial, which included more than 1,000 patients with advanced NSCLC who had previously undergone treatment with platinum-based chemotherapy. The patients were randomly assigned to receive either pembrolizumab or docetaxel for 2 years.

This is the latest update on data from this trial, which has been described as “really extraordinary.”

The 5-year overall survival rates were more than doubled in the pembrolizumab groups, compared with the docetaxel group, reported Roy Herbst, MD, PhD, department of medical oncology, Yale Comprehensive Cancer Center, New Haven, Conn.. He was presenting the new data at the recent World Conference on Lung Cancer 2020.

Overall results for patients with programmed death-ligand 1 (PD-L1) Tumor Proportion Score (TPS) expression greater than 1% show that 15.6% of the pembrolizumab group were still alive at 5 years versus 6.5% of the docetaxel group.

The results were even better among patients who had high PD-L1 TPS expression (>50%): in this subgroup, 25% of the patients who received pembrolizumab were still alive versus 8.2% of those who received docetaxel.

In addition, at 5 years, 9.4% of patients who received pembrolizumab were disease free versus 0.7% of the patients who received docetaxel, Dr. Herbst reported.

Dr. Hirsch commented that the 5-year survival rate of 25% among patients with high PD-L1 expression who underwent treatment with pembrolizumab is “great progress in lung cancer treatment, there is no doubt about it.”

He noted that the results also show that “numerically,” it matters whether patients have low PD-L1 expression. “We know from first-line studies that pembrolizumab monotherapy is effective in high PD-L1–expressing tumors, so these data fit very well,” he said.

At the meeting, Dr. Herbst summarized his presentation on pembrolizumab for patients with NSCLC who had previously undergone treatment, saying that, “with 5 years of follow-up, we continue to see a clinically meaningful improvement in overall survival and PFS [progression-free survival].

“Pembrolizumab monotherapy is a standard of care in patients with immunotherapy-naive or previously treated PD-L1–positive advanced non–small cell lung cancer,” Herbst stated.

Dr. Hirsch was largely in agreement. He believes that, for patients with a PD-L1 TPS of at least 50%, the standard of care “is practically pembrolizumab monotherapy, unless there are certain circumstances where you would add chemotherapy,” such as for patients with a high tumor volume, “where you want to see a very quick response.”

Dr. Hirsch pointed out, however, that currently most patients with high PD-L1–expressing tumors are given pembrolizumab in the first line, which begs the question as to what to give those who experience disease progression after immunotherapy.

“That is an open space,” he said. “There is a lot of studies going on in what we call the immunotherapy-refractory patients.

“We don’t have clear guidance for clinical practice yet,” he commented. He noted that there are several options: “Do you continue with chemotherapy? Do you continue with chemotherapy plus another immunotherapy? Do you switch to another immunotherapy?”

Commenting on Twitter, Stephen V. Liu, MD, director of thoracic oncology at Georgetown University, Washington, said the results were “very exciting.”

However, he wondered whether the results suggest that patients with high PD-L1 expression “may be able to stop” receiving pembrolizumab, whereas those with disease of lower expression “may need longer therapy.”

H. Jack West, MD, medical director of the thoracic oncology program, Swedish Cancer Institute, Seattle, said on Twitter that, to him, the “most impressive” aspect was the “new insight about patients stopping pembro after 2 years but still having two-thirds with sustained response.”

He added that he would “love to learn which patients can stop therapy and when, or whether we can do infrequent maintenance IO [immunotherapy].”

 

Nivolumab survival data

The data on nivolumab come from a pooled analysis of 5-year data on 854 patients from CheckMate 057 and CheckMate 017. The analysis was published in the Journal of Clinical Oncology on Jan. 15, 2021.

Both of these trials compared nivolumab with docetaxel for patients with NSCLC who had experienced disease progression with platinum-based chemotherapy.

The pooled analysis showed that the 5-year overall survival rate was more than fivefold greater with nivolumab than with docetaxel, at 13.4% versus 2.6%.

Moreover, more than 80% of patients who had not experienced progression with the immunotherapy at 2 years were still alive at 5 years. The percentage rose to more than 90% among those who had not experienced progression at 3 years.

Lead author Julie R. Brahmer, MD, Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University, Baltimore, and colleagues said the results “demonstrate that nivolumab can provide long-term survival benefit with durable responses and a tolerable safety profile in patients with previously treated, advanced NSCLC.

“Furthermore, some patients appear to maintain prolonged disease control even after stopping systemic therapy,” they noted.

Dr. Hirsch commented that, although the survival rates with nivolumab were slightly lower than reported with pembrolizumab in KEYNOTE-010, they could still be “within the range.” He added that “I wouldn’t conclude that pembrolizumab is better than nivolumab.”

Many factors may account for these differences, he suggested, including differences in the patient populations or simply differences in the numbers of patients included.

For him, the “main point” of the new data from both trials is that immunotherapy has shown “tremendous progress, compared to chemotherapy.”

KEYNOTE-010 was sponsored by Merck Sharp & Dohme. CheckMate 017 and CheckMate057 were sponsored by Bristol-Myers Squibb. Dr. Herbst has relationships with Jun Shi Pharmaceuticals, AstraZeneca, Genentech, Merck, Pfizer, AbbVie, Biodesix, Bristol-Myers Squibb, Eli Lilly, EMD Serono, Heat Biologics, Loxo, Nektar, NextCure, Novartis, Sanofi, Seattle Genetics, Shire, Spectrum Pharmaceuticals, Symphogen, Tesaro, Neon Therapeutics, Infinity Pharmaceuticals, Armo Biosciences, Genmab, Halozyme, and Tocagen. Dr. Brahmer has relationships with Roche/Genentech, Bristol-Myers Squibb, Lilly, Celgene, Syndax, Janssen Oncology, Merck, Amgen, Genentech, AstraZeneca, Incyte, Spectrum Pharmaceuticals, Revolution, and Roche/Genentech.

A version of this article first appeared on Medscape.com.

New research suggests that proposed lung cancer screening guidelines could inadvertently increase racial and ethnic disparities, but adding in a risk prediction model could reduce some of these disparities by identifying people with high predicted benefit, regardless of race or ethnicity.

The draft United States Preventive Services Task Force (USPSTF) 2020 guidelines recommend annual lung cancer screening for individuals aged 50-80 who currently smoke or quit in the last 15 years, and who have a smoking history equivalent to at least one pack of cigarettes per day for 20 years or more.

This expands the age range and smoking history requirement compared to the 2013 USPSTF recommendations in an attempt to partially ameliorate racial disparities in screening eligibility. The 2013 guidelines recommend screening ever-smokers aged 55-80 with 30 or more pack-years and 15 or fewer quit-years.

However, neither the 2013 nor the 2020 USPSTF recommendations consider the higher risk of lung cancer and younger ages at diagnosis among African Americans, despite their smoking less than Whites, according to Rebecca Landy, PhD, of the National Cancer Institute in Bethesda, Md.

“For the same age and smoking history as Whites, minorities have substantially different lung cancer risk,” Dr. Landy said. “Incorporating individualized prediction models into USPSTF guidelines may reduce racial/ethnic disparities in lung cancer screening eligibility.”

Dr. Landy and colleagues set out to test that theory, and she presented the results at the 2020 World Congress on Lung Cancer (Abstract 3564), which was rescheduled for January 2021. The results were published in the Journal of the National Cancer Institute.
 

Study details

Dr. Landy and colleagues modeled the performance of National Lung Screening Trial–like screening (three annual CT screens, 5 years of follow-up) among three cohorts of ever-smokers aged 50-80 using the 2015 National Health Interview Survey.

One group was eligible by USPSTF 2013 guidelines, another by draft USPSTF 2020 guidelines, and yet another by augmenting the USPSTF 2020 guidelines using risk prediction to include individuals with 12 or more days of life gained according to the Life-Years From Screening–CT (LYFS-CT) model.

“Among each race/ethnicity, we calculated the number eligible for screening, proportion of preventable lung cancer deaths prevented, proportion of gainable life-years gained, and screening effectiveness, as well as the relative disparities in lung cancer deaths prevented and life-years gained,” Dr. Landy said.
 

Results

Under the 2013 guidelines, 8 million ever-smokers were eligible. The disparities in lung cancer death sensitivity, compared to Whites, were 15% for African Americans, 15% for Asian Americans, and 24% for Hispanic Americans. Disparities for life-year gained sensitivity were 15%, 13%, and 24%, respectively.

Under the 2020 draft guidelines, 14.5 million ever-smokers were eligible, but racial/ethnic disparities persisted. Disparities in lung cancer death sensitivity were 13% for African Americans, 19% for Asian Americans, and 27% for Hispanic Americans. Disparities for life-year gained sensitivity were 16%, 19%, and 27%, respectively.

Using the LYFS-CT predictive-risk model added an additional 3.5 million people and “nearly eliminated” disparities for African Americans, Dr. Landy noted. However, disparities persisted for Asian Americans and Hispanic Americans.

Disparities in lung cancer death sensitivity were 0% for African Americans, 19% for Asian Americans, and 23% for Hispanic Americans. Disparities for life-year gained sensitivity were 1%, 19%, and 24%, respectively.
 

More and widening disparity

The results showed that augmenting USPSTF criteria to include high-benefit people selected significantly more African Americans than Whites and could therefore reduce or even eliminate disparities between Whites and African Americans.

“The 2020 USPSTF draft recommendations would make 6.5 million more people eligible to be screened, in addition to the 8 million from the 2013 criteria,” said Gerard Silvestri, MD, of the Medical University of South Carolina, Charleston, who was not involved in this study.

“But there will be more White people than African American people added, and the disparity between them may widen. Using the risk prediction model outlined in this well-researched study could close the gap in disparity. It’s important to identify individual risk and life expectancy.”

Dr. Silvestri pointed out that, compared to Whites, African Americans develop lung cancer at an earlier age with fewer pack-years history of smoking and have worse outcomes.

“We can’t just focus on one aspect of disparity,” he said. “African Americans are much less likely to be insured or to identify a primary care provider for integrated care. We know that screening works. The 2020 USPSTF draft recommendations will enlarge the pool of eligible African Americans and reduce disparities if the other part of the equation holds; that is, they get access to care and screening.”

This study was funded by the National Institutes of Health/National Cancer Institute. Dr. Landy and Dr. Silvestri have no disclosures.

New research suggests that proposed lung cancer screening guidelines could inadvertently increase racial and ethnic disparities, but adding in a risk prediction model could reduce some of these disparities by identifying people with high predicted benefit, regardless of race or ethnicity.

The draft United States Preventive Services Task Force (USPSTF) 2020 guidelines recommend annual lung cancer screening for individuals aged 50-80 who currently smoke or quit in the last 15 years, and who have a smoking history equivalent to at least one pack of cigarettes per day for 20 years or more.

This expands the age range and smoking history requirement compared to the 2013 USPSTF recommendations in an attempt to partially ameliorate racial disparities in screening eligibility. The 2013 guidelines recommend screening ever-smokers aged 55-80 with 30 or more pack-years and 15 or fewer quit-years.

However, neither the 2013 nor the 2020 USPSTF recommendations consider the higher risk of lung cancer and younger ages at diagnosis among African Americans, despite their smoking less than Whites, according to Rebecca Landy, PhD, of the National Cancer Institute in Bethesda, Md.

“For the same age and smoking history as Whites, minorities have substantially different lung cancer risk,” Dr. Landy said. “Incorporating individualized prediction models into USPSTF guidelines may reduce racial/ethnic disparities in lung cancer screening eligibility.”

Dr. Landy and colleagues set out to test that theory, and she presented the results at the 2020 World Congress on Lung Cancer (Abstract 3564), which was rescheduled for January 2021. The results were published in the Journal of the National Cancer Institute.
 

Study details

Dr. Landy and colleagues modeled the performance of National Lung Screening Trial–like screening (three annual CT screens, 5 years of follow-up) among three cohorts of ever-smokers aged 50-80 using the 2015 National Health Interview Survey.

One group was eligible by USPSTF 2013 guidelines, another by draft USPSTF 2020 guidelines, and yet another by augmenting the USPSTF 2020 guidelines using risk prediction to include individuals with 12 or more days of life gained according to the Life-Years From Screening–CT (LYFS-CT) model.

“Among each race/ethnicity, we calculated the number eligible for screening, proportion of preventable lung cancer deaths prevented, proportion of gainable life-years gained, and screening effectiveness, as well as the relative disparities in lung cancer deaths prevented and life-years gained,” Dr. Landy said.
 

Results

Under the 2013 guidelines, 8 million ever-smokers were eligible. The disparities in lung cancer death sensitivity, compared to Whites, were 15% for African Americans, 15% for Asian Americans, and 24% for Hispanic Americans. Disparities for life-year gained sensitivity were 15%, 13%, and 24%, respectively.

Under the 2020 draft guidelines, 14.5 million ever-smokers were eligible, but racial/ethnic disparities persisted. Disparities in lung cancer death sensitivity were 13% for African Americans, 19% for Asian Americans, and 27% for Hispanic Americans. Disparities for life-year gained sensitivity were 16%, 19%, and 27%, respectively.

Using the LYFS-CT predictive-risk model added an additional 3.5 million people and “nearly eliminated” disparities for African Americans, Dr. Landy noted. However, disparities persisted for Asian Americans and Hispanic Americans.

Disparities in lung cancer death sensitivity were 0% for African Americans, 19% for Asian Americans, and 23% for Hispanic Americans. Disparities for life-year gained sensitivity were 1%, 19%, and 24%, respectively.
 

More and widening disparity

The results showed that augmenting USPSTF criteria to include high-benefit people selected significantly more African Americans than Whites and could therefore reduce or even eliminate disparities between Whites and African Americans.

“The 2020 USPSTF draft recommendations would make 6.5 million more people eligible to be screened, in addition to the 8 million from the 2013 criteria,” said Gerard Silvestri, MD, of the Medical University of South Carolina, Charleston, who was not involved in this study.

“But there will be more White people than African American people added, and the disparity between them may widen. Using the risk prediction model outlined in this well-researched study could close the gap in disparity. It’s important to identify individual risk and life expectancy.”

Dr. Silvestri pointed out that, compared to Whites, African Americans develop lung cancer at an earlier age with fewer pack-years history of smoking and have worse outcomes.

“We can’t just focus on one aspect of disparity,” he said. “African Americans are much less likely to be insured or to identify a primary care provider for integrated care. We know that screening works. The 2020 USPSTF draft recommendations will enlarge the pool of eligible African Americans and reduce disparities if the other part of the equation holds; that is, they get access to care and screening.”

This study was funded by the National Institutes of Health/National Cancer Institute. Dr. Landy and Dr. Silvestri have no disclosures.

New research suggests that proposed lung cancer screening guidelines could inadvertently increase racial and ethnic disparities, but adding in a risk prediction model could reduce some of these disparities by identifying people with high predicted benefit, regardless of race or ethnicity.

The draft United States Preventive Services Task Force (USPSTF) 2020 guidelines recommend annual lung cancer screening for individuals aged 50-80 who currently smoke or quit in the last 15 years, and who have a smoking history equivalent to at least one pack of cigarettes per day for 20 years or more.

This expands the age range and smoking history requirement compared to the 2013 USPSTF recommendations in an attempt to partially ameliorate racial disparities in screening eligibility. The 2013 guidelines recommend screening ever-smokers aged 55-80 with 30 or more pack-years and 15 or fewer quit-years.

However, neither the 2013 nor the 2020 USPSTF recommendations consider the higher risk of lung cancer and younger ages at diagnosis among African Americans, despite their smoking less than Whites, according to Rebecca Landy, PhD, of the National Cancer Institute in Bethesda, Md.

“For the same age and smoking history as Whites, minorities have substantially different lung cancer risk,” Dr. Landy said. “Incorporating individualized prediction models into USPSTF guidelines may reduce racial/ethnic disparities in lung cancer screening eligibility.”

Dr. Landy and colleagues set out to test that theory, and she presented the results at the 2020 World Congress on Lung Cancer (Abstract 3564), which was rescheduled for January 2021. The results were published in the Journal of the National Cancer Institute.
 

Study details

Dr. Landy and colleagues modeled the performance of National Lung Screening Trial–like screening (three annual CT screens, 5 years of follow-up) among three cohorts of ever-smokers aged 50-80 using the 2015 National Health Interview Survey.

One group was eligible by USPSTF 2013 guidelines, another by draft USPSTF 2020 guidelines, and yet another by augmenting the USPSTF 2020 guidelines using risk prediction to include individuals with 12 or more days of life gained according to the Life-Years From Screening–CT (LYFS-CT) model.

“Among each race/ethnicity, we calculated the number eligible for screening, proportion of preventable lung cancer deaths prevented, proportion of gainable life-years gained, and screening effectiveness, as well as the relative disparities in lung cancer deaths prevented and life-years gained,” Dr. Landy said.
 

Results

Under the 2013 guidelines, 8 million ever-smokers were eligible. The disparities in lung cancer death sensitivity, compared to Whites, were 15% for African Americans, 15% for Asian Americans, and 24% for Hispanic Americans. Disparities for life-year gained sensitivity were 15%, 13%, and 24%, respectively.

Under the 2020 draft guidelines, 14.5 million ever-smokers were eligible, but racial/ethnic disparities persisted. Disparities in lung cancer death sensitivity were 13% for African Americans, 19% for Asian Americans, and 27% for Hispanic Americans. Disparities for life-year gained sensitivity were 16%, 19%, and 27%, respectively.

Using the LYFS-CT predictive-risk model added an additional 3.5 million people and “nearly eliminated” disparities for African Americans, Dr. Landy noted. However, disparities persisted for Asian Americans and Hispanic Americans.

Disparities in lung cancer death sensitivity were 0% for African Americans, 19% for Asian Americans, and 23% for Hispanic Americans. Disparities for life-year gained sensitivity were 1%, 19%, and 24%, respectively.
 

More and widening disparity

The results showed that augmenting USPSTF criteria to include high-benefit people selected significantly more African Americans than Whites and could therefore reduce or even eliminate disparities between Whites and African Americans.

“The 2020 USPSTF draft recommendations would make 6.5 million more people eligible to be screened, in addition to the 8 million from the 2013 criteria,” said Gerard Silvestri, MD, of the Medical University of South Carolina, Charleston, who was not involved in this study.

“But there will be more White people than African American people added, and the disparity between them may widen. Using the risk prediction model outlined in this well-researched study could close the gap in disparity. It’s important to identify individual risk and life expectancy.”

Dr. Silvestri pointed out that, compared to Whites, African Americans develop lung cancer at an earlier age with fewer pack-years history of smoking and have worse outcomes.

“We can’t just focus on one aspect of disparity,” he said. “African Americans are much less likely to be insured or to identify a primary care provider for integrated care. We know that screening works. The 2020 USPSTF draft recommendations will enlarge the pool of eligible African Americans and reduce disparities if the other part of the equation holds; that is, they get access to care and screening.”

This study was funded by the National Institutes of Health/National Cancer Institute. Dr. Landy and Dr. Silvestri have no disclosures.

Tailoring adjuvant chemotherapy based on the expression of two molecular markers did not confer a survival advantage in patients with completely resected stage II-III non–small cell lung cancer (NSCLC) in a phase 3 trial.

The patients were randomized to receive investigator’s choice of platinum-based chemotherapy or treatment tailored according to messenger RNA (mRNA) expression of two molecular markers – excision repair cross complementation 1 (ERCC1) and thymidylate synthase (TS).

There was no significant difference in overall survival or recurrence-free survival between the treatment approaches. However, toxicity was less common among patients who received customized treatment.

These results, from the phase 3 ITACA trial, were presented at the 2020 World Conference on Lung Cancer (Abstract 1820), which was rescheduled to January 2021.

“There is a clear need to define patients most likely to derive survival benefit from adjuvant therapy and spare patients who do not need adjuvant chemotherapy due to the toxicity of such therapy,” said presenter Silvia Novello, MD, PhD, of the University of Turin in Italy. “mRNA expression of different genes has been correlated with the sensitivity or resistance to specific anticancer agents.”

With this in mind, Dr. Novello and colleagues conducted the ITACA trial. The researchers’ primary goal was to determine whether an adjuvant pharmacogenomic-driven approach was able to improve overall survival in completely resected NSCLC.
 

Patients and treatment

The researchers randomized 773 NSCLC patients within 5-8 weeks after radical surgery. Genomic analyses were performed soon after surgery, and patients were randomly assigned to investigator’s choice of platinum-based chemotherapy or to tailored treatments defined by mRNA levels of ERCC1 and TS.

Patients with high ERCC1 mRNA expression who were randomized to tailored treatment received single-agent docetaxel if their TS level was high or pemetrexed monotherapy if their TS level was low.

Patients with low ERCC1 mRNA expression who were randomized to tailored treatment received cisplatin-gemcitabine if their TS level was high or cisplatin-pemetrexed if their TS was low.

The most frequent doublets used in control patients were cisplatin-gemcitabine and cisplatin-vinorelbine.

The demographic characteristics of the 384 patients randomized to tailored therapy and the 389 control subjects were well-balanced, Dr. Novello said. Two-thirds of patients had stage II disease, 11% were never smokers, and the vast majority had a lobectomy as the resection method.
 

Results

At a median follow-up of 28.2 months, the median overall survival was 96.4 months in the tailored therapy arm and 83.5 months in the control arm. The median recurrence-free survival was 64.4 months and 41.5 months, respectively.

“Adjuvant chemotherapy customization based on the primary tumor tissue mRNA expression of ERCC1 and TS did not significantly improve overall survival or recurrence-free survival,” Dr. Novello said. “There was a non–statistically significant trend for overall survival favoring the customized arm.”

Dr. Novello noted that, when the final analysis was performed, the study was underpowered, as only 46% of expected events were collected. Assuming the same hazard ratio point estimate and that the expected 336 events were collected, the hazard ratio estimate would be 0.76 (P = .012).

Grade 3/4 toxicities occurred in 32.6% of patients in the tailored therapy arm and 45.9% of those in the control arm (P < .001).

“It is important to underline that the treatment customization significantly improved the toxicity profile without compromising the efficacy,” Dr. Novello said.

She added that “more comprehensive and high-throughput diagnostic techniques will be needed in order to tailor adjuvant chemotherapy, with or without immunotherapy, in completely resected NSCLC.”

“The ITACA study is the largest adjuvant study tailored to ERCC1/TS status, and the results have been long-awaited,” said Tetsuya Mitsudomi, MD, a professor at Kindai University in Japan and president of the International Association for the Study of Lung Cancer.

“This trial should be praised for the mandated genomic analysis that was accomplished within a reasonably short time frame before random assignment. In addition, this trial confirmed that there is no biomarker strong enough to predict the efficacy of cytotoxic chemotherapy. However, the concept of customizing adjuvant therapy according to the genomic status of patients’ tumors is valid, leading to the recent demonstration in the ADAURA study of the superiority of osimertinib in delaying the postoperative recurrence of disease in patients with EGFR-mutated NSCLC.”

The ITACA study was funded by University of Turin and Eli Lilly. Dr. Novello disclosed relationships with Eli Lilly, Amgen, AstraZeneca, Bohringer Ingelheim, Beigene, Pfizer, Roche, Merck, Bristol-Myers Squibb, Takeda, and Sanofi. Dr. Mitsudomi disclosed relationships with Eli Lilly, AstraZeneca, Boehringer-Ingelheim, Chugai, Pfizer, Merck, Ono Pharmaceutical, Bristol-Myers Squibb, Novartis, ThermoFisher, Guardant, Eisai, Amgen, and Johnson & Johnson.

Tailoring adjuvant chemotherapy based on the expression of two molecular markers did not confer a survival advantage in patients with completely resected stage II-III non–small cell lung cancer (NSCLC) in a phase 3 trial.

The patients were randomized to receive investigator’s choice of platinum-based chemotherapy or treatment tailored according to messenger RNA (mRNA) expression of two molecular markers – excision repair cross complementation 1 (ERCC1) and thymidylate synthase (TS).

There was no significant difference in overall survival or recurrence-free survival between the treatment approaches. However, toxicity was less common among patients who received customized treatment.

These results, from the phase 3 ITACA trial, were presented at the 2020 World Conference on Lung Cancer (Abstract 1820), which was rescheduled to January 2021.

“There is a clear need to define patients most likely to derive survival benefit from adjuvant therapy and spare patients who do not need adjuvant chemotherapy due to the toxicity of such therapy,” said presenter Silvia Novello, MD, PhD, of the University of Turin in Italy. “mRNA expression of different genes has been correlated with the sensitivity or resistance to specific anticancer agents.”

With this in mind, Dr. Novello and colleagues conducted the ITACA trial. The researchers’ primary goal was to determine whether an adjuvant pharmacogenomic-driven approach was able to improve overall survival in completely resected NSCLC.
 

Patients and treatment

The researchers randomized 773 NSCLC patients within 5-8 weeks after radical surgery. Genomic analyses were performed soon after surgery, and patients were randomly assigned to investigator’s choice of platinum-based chemotherapy or to tailored treatments defined by mRNA levels of ERCC1 and TS.

Patients with high ERCC1 mRNA expression who were randomized to tailored treatment received single-agent docetaxel if their TS level was high or pemetrexed monotherapy if their TS level was low.

Patients with low ERCC1 mRNA expression who were randomized to tailored treatment received cisplatin-gemcitabine if their TS level was high or cisplatin-pemetrexed if their TS was low.

The most frequent doublets used in control patients were cisplatin-gemcitabine and cisplatin-vinorelbine.

The demographic characteristics of the 384 patients randomized to tailored therapy and the 389 control subjects were well-balanced, Dr. Novello said. Two-thirds of patients had stage II disease, 11% were never smokers, and the vast majority had a lobectomy as the resection method.
 

Results

At a median follow-up of 28.2 months, the median overall survival was 96.4 months in the tailored therapy arm and 83.5 months in the control arm. The median recurrence-free survival was 64.4 months and 41.5 months, respectively.

“Adjuvant chemotherapy customization based on the primary tumor tissue mRNA expression of ERCC1 and TS did not significantly improve overall survival or recurrence-free survival,” Dr. Novello said. “There was a non–statistically significant trend for overall survival favoring the customized arm.”

Dr. Novello noted that, when the final analysis was performed, the study was underpowered, as only 46% of expected events were collected. Assuming the same hazard ratio point estimate and that the expected 336 events were collected, the hazard ratio estimate would be 0.76 (P = .012).

Grade 3/4 toxicities occurred in 32.6% of patients in the tailored therapy arm and 45.9% of those in the control arm (P < .001).

“It is important to underline that the treatment customization significantly improved the toxicity profile without compromising the efficacy,” Dr. Novello said.

She added that “more comprehensive and high-throughput diagnostic techniques will be needed in order to tailor adjuvant chemotherapy, with or without immunotherapy, in completely resected NSCLC.”

“The ITACA study is the largest adjuvant study tailored to ERCC1/TS status, and the results have been long-awaited,” said Tetsuya Mitsudomi, MD, a professor at Kindai University in Japan and president of the International Association for the Study of Lung Cancer.

“This trial should be praised for the mandated genomic analysis that was accomplished within a reasonably short time frame before random assignment. In addition, this trial confirmed that there is no biomarker strong enough to predict the efficacy of cytotoxic chemotherapy. However, the concept of customizing adjuvant therapy according to the genomic status of patients’ tumors is valid, leading to the recent demonstration in the ADAURA study of the superiority of osimertinib in delaying the postoperative recurrence of disease in patients with EGFR-mutated NSCLC.”

The ITACA study was funded by University of Turin and Eli Lilly. Dr. Novello disclosed relationships with Eli Lilly, Amgen, AstraZeneca, Bohringer Ingelheim, Beigene, Pfizer, Roche, Merck, Bristol-Myers Squibb, Takeda, and Sanofi. Dr. Mitsudomi disclosed relationships with Eli Lilly, AstraZeneca, Boehringer-Ingelheim, Chugai, Pfizer, Merck, Ono Pharmaceutical, Bristol-Myers Squibb, Novartis, ThermoFisher, Guardant, Eisai, Amgen, and Johnson & Johnson.

Tailoring adjuvant chemotherapy based on the expression of two molecular markers did not confer a survival advantage in patients with completely resected stage II-III non–small cell lung cancer (NSCLC) in a phase 3 trial.

The patients were randomized to receive investigator’s choice of platinum-based chemotherapy or treatment tailored according to messenger RNA (mRNA) expression of two molecular markers – excision repair cross complementation 1 (ERCC1) and thymidylate synthase (TS).

There was no significant difference in overall survival or recurrence-free survival between the treatment approaches. However, toxicity was less common among patients who received customized treatment.

These results, from the phase 3 ITACA trial, were presented at the 2020 World Conference on Lung Cancer (Abstract 1820), which was rescheduled to January 2021.

“There is a clear need to define patients most likely to derive survival benefit from adjuvant therapy and spare patients who do not need adjuvant chemotherapy due to the toxicity of such therapy,” said presenter Silvia Novello, MD, PhD, of the University of Turin in Italy. “mRNA expression of different genes has been correlated with the sensitivity or resistance to specific anticancer agents.”

With this in mind, Dr. Novello and colleagues conducted the ITACA trial. The researchers’ primary goal was to determine whether an adjuvant pharmacogenomic-driven approach was able to improve overall survival in completely resected NSCLC.
 

Patients and treatment

The researchers randomized 773 NSCLC patients within 5-8 weeks after radical surgery. Genomic analyses were performed soon after surgery, and patients were randomly assigned to investigator’s choice of platinum-based chemotherapy or to tailored treatments defined by mRNA levels of ERCC1 and TS.

Patients with high ERCC1 mRNA expression who were randomized to tailored treatment received single-agent docetaxel if their TS level was high or pemetrexed monotherapy if their TS level was low.

Patients with low ERCC1 mRNA expression who were randomized to tailored treatment received cisplatin-gemcitabine if their TS level was high or cisplatin-pemetrexed if their TS was low.

The most frequent doublets used in control patients were cisplatin-gemcitabine and cisplatin-vinorelbine.

The demographic characteristics of the 384 patients randomized to tailored therapy and the 389 control subjects were well-balanced, Dr. Novello said. Two-thirds of patients had stage II disease, 11% were never smokers, and the vast majority had a lobectomy as the resection method.
 

Results

At a median follow-up of 28.2 months, the median overall survival was 96.4 months in the tailored therapy arm and 83.5 months in the control arm. The median recurrence-free survival was 64.4 months and 41.5 months, respectively.

“Adjuvant chemotherapy customization based on the primary tumor tissue mRNA expression of ERCC1 and TS did not significantly improve overall survival or recurrence-free survival,” Dr. Novello said. “There was a non–statistically significant trend for overall survival favoring the customized arm.”

Dr. Novello noted that, when the final analysis was performed, the study was underpowered, as only 46% of expected events were collected. Assuming the same hazard ratio point estimate and that the expected 336 events were collected, the hazard ratio estimate would be 0.76 (P = .012).

Grade 3/4 toxicities occurred in 32.6% of patients in the tailored therapy arm and 45.9% of those in the control arm (P < .001).

“It is important to underline that the treatment customization significantly improved the toxicity profile without compromising the efficacy,” Dr. Novello said.

She added that “more comprehensive and high-throughput diagnostic techniques will be needed in order to tailor adjuvant chemotherapy, with or without immunotherapy, in completely resected NSCLC.”

“The ITACA study is the largest adjuvant study tailored to ERCC1/TS status, and the results have been long-awaited,” said Tetsuya Mitsudomi, MD, a professor at Kindai University in Japan and president of the International Association for the Study of Lung Cancer.

“This trial should be praised for the mandated genomic analysis that was accomplished within a reasonably short time frame before random assignment. In addition, this trial confirmed that there is no biomarker strong enough to predict the efficacy of cytotoxic chemotherapy. However, the concept of customizing adjuvant therapy according to the genomic status of patients’ tumors is valid, leading to the recent demonstration in the ADAURA study of the superiority of osimertinib in delaying the postoperative recurrence of disease in patients with EGFR-mutated NSCLC.”

The ITACA study was funded by University of Turin and Eli Lilly. Dr. Novello disclosed relationships with Eli Lilly, Amgen, AstraZeneca, Bohringer Ingelheim, Beigene, Pfizer, Roche, Merck, Bristol-Myers Squibb, Takeda, and Sanofi. Dr. Mitsudomi disclosed relationships with Eli Lilly, AstraZeneca, Boehringer-Ingelheim, Chugai, Pfizer, Merck, Ono Pharmaceutical, Bristol-Myers Squibb, Novartis, ThermoFisher, Guardant, Eisai, Amgen, and Johnson & Johnson.

An artificial intelligence (AI) diagnostic system based on neural networks may assist in the diagnosis of COVID-19, according to a pilot study.

The system, called CXR-Net, was trained to differentiate SARS-CoV-2 chest x-rays (CXRs) from CXRs that are either normal or non–COVID-19 lung pathologies, explained Abdulah Haikal, an MD candidate at Wayne State University, Detroit.

Mr. Haikal described CXR-Net at the AACR Virtual Meeting: COVID-19 and Cancer (Abstract S11-04).

CXR-Net is a two-module pipeline, Mr. Haikal explained. Module I is based on Res-CR-Net, a type of neural network originally designed for the semantic segmentation of microscopy images, with the ability to retain the original resolution of the input images in the feature maps of all layers and in the final output.

Module II is a hybrid convolutional neural network in which the first convolutional layer with learned coefficients is replaced by a layer with fixed coefficients provided by the Wavelet Scattering Transform. Module II inputs patients’ CXRs and corresponding lung masks quantified by Module I, and generates as outputs a class assignment (COVID-19 or non–COVID-19) and high-resolution heat maps that detect the severe acute respiratory syndrome–-associated lung regions.

“The system is trained to differentiate COVID and non-COVID pathologies and produces a highly discriminative heat map to point to lung regions where COVID is suspected,” Mr. Haikal said. “The Wavelet Scattering Transform allows for fast determination of COVID versus non-COVID CXRs.”
 

Preliminary results and implications

CXR-Net was piloted on a small dataset of CXRs from non–COVID-19 and polymerase chain reaction–confirmed COVID-19 patients acquired at a single center in Detroit.

Upon fivefold cross validation of the training set with 2,265 images, 90% accuracy was observed when the training set was tested against the validation set. However, once 1,532 new images were introduced, a 76% accuracy rate was observed.

The F1 scores were 0.81 and 0.70 for the training and test sets, respectively.

“I’m really excited about this new approach, and I think AI will allow us to do more with less, which is exciting,” said Ross L. Levine, MD, of Memorial Sloan Kettering Cancer Center in New York, who led a discussion session with Mr. Haikal about CXR-Net.

One question raised during the discussion was whether the technology will help health care providers be more thoughtful about when and how they image COVID-19 patients.

“The more data you feed into the system, the stronger and more accurate it becomes,” Mr. Haikal said. “However, until we have data sharing from multiple centers, we won’t see improved accuracy results.”

Another question was whether this technology could be integrated with more clinical parameters.

“Some individuals are afraid that AI will replace the job of a professional, but it will only make it better for us,” Mr. Haikal said. “We don’t rely on current imaging techniques to make a definitive diagnosis, but rather have a specificity and sensitivity to establish a diagnosis, and AI can be used in the same way as a diagnostic tool.”

Mr. Haikal and Dr. Levine disclosed no conflicts of interest. No funding sources were reported in the presentation.

An artificial intelligence (AI) diagnostic system based on neural networks may assist in the diagnosis of COVID-19, according to a pilot study.

The system, called CXR-Net, was trained to differentiate SARS-CoV-2 chest x-rays (CXRs) from CXRs that are either normal or non–COVID-19 lung pathologies, explained Abdulah Haikal, an MD candidate at Wayne State University, Detroit.

Mr. Haikal described CXR-Net at the AACR Virtual Meeting: COVID-19 and Cancer (Abstract S11-04).

CXR-Net is a two-module pipeline, Mr. Haikal explained. Module I is based on Res-CR-Net, a type of neural network originally designed for the semantic segmentation of microscopy images, with the ability to retain the original resolution of the input images in the feature maps of all layers and in the final output.

Module II is a hybrid convolutional neural network in which the first convolutional layer with learned coefficients is replaced by a layer with fixed coefficients provided by the Wavelet Scattering Transform. Module II inputs patients’ CXRs and corresponding lung masks quantified by Module I, and generates as outputs a class assignment (COVID-19 or non–COVID-19) and high-resolution heat maps that detect the severe acute respiratory syndrome–-associated lung regions.

“The system is trained to differentiate COVID and non-COVID pathologies and produces a highly discriminative heat map to point to lung regions where COVID is suspected,” Mr. Haikal said. “The Wavelet Scattering Transform allows for fast determination of COVID versus non-COVID CXRs.”
 

Preliminary results and implications

CXR-Net was piloted on a small dataset of CXRs from non–COVID-19 and polymerase chain reaction–confirmed COVID-19 patients acquired at a single center in Detroit.

Upon fivefold cross validation of the training set with 2,265 images, 90% accuracy was observed when the training set was tested against the validation set. However, once 1,532 new images were introduced, a 76% accuracy rate was observed.

The F1 scores were 0.81 and 0.70 for the training and test sets, respectively.

“I’m really excited about this new approach, and I think AI will allow us to do more with less, which is exciting,” said Ross L. Levine, MD, of Memorial Sloan Kettering Cancer Center in New York, who led a discussion session with Mr. Haikal about CXR-Net.

One question raised during the discussion was whether the technology will help health care providers be more thoughtful about when and how they image COVID-19 patients.

“The more data you feed into the system, the stronger and more accurate it becomes,” Mr. Haikal said. “However, until we have data sharing from multiple centers, we won’t see improved accuracy results.”

Another question was whether this technology could be integrated with more clinical parameters.

“Some individuals are afraid that AI will replace the job of a professional, but it will only make it better for us,” Mr. Haikal said. “We don’t rely on current imaging techniques to make a definitive diagnosis, but rather have a specificity and sensitivity to establish a diagnosis, and AI can be used in the same way as a diagnostic tool.”

Mr. Haikal and Dr. Levine disclosed no conflicts of interest. No funding sources were reported in the presentation.

An artificial intelligence (AI) diagnostic system based on neural networks may assist in the diagnosis of COVID-19, according to a pilot study.

The system, called CXR-Net, was trained to differentiate SARS-CoV-2 chest x-rays (CXRs) from CXRs that are either normal or non–COVID-19 lung pathologies, explained Abdulah Haikal, an MD candidate at Wayne State University, Detroit.

Mr. Haikal described CXR-Net at the AACR Virtual Meeting: COVID-19 and Cancer (Abstract S11-04).

CXR-Net is a two-module pipeline, Mr. Haikal explained. Module I is based on Res-CR-Net, a type of neural network originally designed for the semantic segmentation of microscopy images, with the ability to retain the original resolution of the input images in the feature maps of all layers and in the final output.

Module II is a hybrid convolutional neural network in which the first convolutional layer with learned coefficients is replaced by a layer with fixed coefficients provided by the Wavelet Scattering Transform. Module II inputs patients’ CXRs and corresponding lung masks quantified by Module I, and generates as outputs a class assignment (COVID-19 or non–COVID-19) and high-resolution heat maps that detect the severe acute respiratory syndrome–-associated lung regions.

“The system is trained to differentiate COVID and non-COVID pathologies and produces a highly discriminative heat map to point to lung regions where COVID is suspected,” Mr. Haikal said. “The Wavelet Scattering Transform allows for fast determination of COVID versus non-COVID CXRs.”
 

Preliminary results and implications

CXR-Net was piloted on a small dataset of CXRs from non–COVID-19 and polymerase chain reaction–confirmed COVID-19 patients acquired at a single center in Detroit.

Upon fivefold cross validation of the training set with 2,265 images, 90% accuracy was observed when the training set was tested against the validation set. However, once 1,532 new images were introduced, a 76% accuracy rate was observed.

The F1 scores were 0.81 and 0.70 for the training and test sets, respectively.

“I’m really excited about this new approach, and I think AI will allow us to do more with less, which is exciting,” said Ross L. Levine, MD, of Memorial Sloan Kettering Cancer Center in New York, who led a discussion session with Mr. Haikal about CXR-Net.

One question raised during the discussion was whether the technology will help health care providers be more thoughtful about when and how they image COVID-19 patients.

“The more data you feed into the system, the stronger and more accurate it becomes,” Mr. Haikal said. “However, until we have data sharing from multiple centers, we won’t see improved accuracy results.”

Another question was whether this technology could be integrated with more clinical parameters.

“Some individuals are afraid that AI will replace the job of a professional, but it will only make it better for us,” Mr. Haikal said. “We don’t rely on current imaging techniques to make a definitive diagnosis, but rather have a specificity and sensitivity to establish a diagnosis, and AI can be used in the same way as a diagnostic tool.”

Mr. Haikal and Dr. Levine disclosed no conflicts of interest. No funding sources were reported in the presentation.

Neoadjuvant atezolizumab prior to lung cancer surgery was well tolerated by patients with stage IB-IIIB lung cancer and produced a 21% major pathologic response rate, according to the primary analysis of the Lung Cancer Mutation Consortium (LCMC) 3 study.

Small pilot studies previously suggested that preoperative immune checkpoint inhibitor (ICI) therapy may benefit patients with resectable non–small cell lung cancer (NSCLC).

The LCMC3 study is “unique” because it is the largest monotherapy trial of checkpoint inhibition in resectable NSCLC, and it’s “a landmark study” because it validated results from smaller trials and can serve as a benchmark for future ones, said Jay M. Lee, MD, of the University of California, Los Angeles.

Dr. Lee presented results from LCMC3 at the 2020 World Congress on Lung Cancer (Abstract PS01.05), which was rescheduled for January 2021.

The study included 181 patients, median age 65 years, with stage IB-IIIB NSCLC. The vast majority (90%) of patients were current/former smokers, and two-thirds had a nonsquamous histology. Patients were categorized in the following stages: 17 patients were staged at IB, 20 were IIA, 55 were IIB, 72 were IIIA, and 17 were IIIB.

Patients received 1,200 mg of neoadjuvant atezolizumab intravenously every 3 weeks for two cycles followed by resection between 30 and 50 days from the first cycle. Patients who benefited from the therapy continued adjuvant atezolizumab for 12 months.

The primary endpoint was major pathological response, defined as no more than 10% viable tumor cells at surgery, in patients without epidermal growth factor receptor or anaplastic lymphoma kinase mutations.
 

Results

Following atezolizumab treatment, 43% of patients were down-staged, and 19% were up-staged. Some degree of pathological regression was observed in all but 3 of the 159 patients who underwent resection.

Among the 144 patients included in the efficacy analysis, the major pathological response rate was 21%, with 7% of patients achieving a complete pathological response.

“We demonstrated that more than half of patients resected with a minimally invasive operation. Remarkably, only 15% required thoracotomy. The 92% complete resection rate is comparable, if not superior to, preoperative chemotherapy trials,” Dr. Lee said.

The majority (88%) of patients underwent surgical resection within a 20-day protocol window. The median time from end of neoadjuvant therapy to surgery was 22 days.

“Historically, the neoadjuvant chemotherapy window is much later for surgery, 3 weeks from neoadjuvant therapy, and that can be stretched to up to 56 days,” Dr. Lee said.

In an exploratory analysis, the 1.5-year overall survival rate was 91% for stage I and II disease and 87% for stage III disease. The survival in both cohorts was superior to that expected historically, Dr. Lee noted.

Intraoperative complications were rare (3%). Postoperative adverse reactions correlated with fewer viable tumor cells in the resected specimen.

One patient died following surgery after the first 30 days, which was deemed unrelated to treatment. Another patient died between 30 and 90 days from treatment-related pneumonitis.

“The LCMC3 study successfully met its primary endpoint of achieving major pathological response,” Dr. Lee concluded. “Neoadjuvant atezolizumab monotherapy was well tolerated, and resection was performed with low perioperative morbidity and mortality, usually within a narrow protocol window and with a short time frame from completion of atezolizumab and with a correspondingly high complete resection rate.”

The study’s results suggest that “neoadjuvant atezolizumab monotherapy is effective, well tolerated, and surgically acceptable,” said study discussant Shinichi Toyooka, MD, of Okayama (Japan) University Hospital.

“I would consider single-agent ICI neoadjuvant therapy for patients with early-stage disease and poor performance status, and an ICI plus chemotherapy for more advanced resectable cases, like locally advanced disease,” Dr. Toyooka said.

The LCMC3 study is sponsored by Genentech. Dr. Lee disclosed relationships with Genentech/Roche, AstraZeneca, Bristol-Myers Squibb, Merck, and Novartis. Dr. Toyooka disclosed relationships with AstraZeneca, Chugai, Taiho Pharmaceutical Group, and Ono Pharmaceutical.

Neoadjuvant atezolizumab prior to lung cancer surgery was well tolerated by patients with stage IB-IIIB lung cancer and produced a 21% major pathologic response rate, according to the primary analysis of the Lung Cancer Mutation Consortium (LCMC) 3 study.

Small pilot studies previously suggested that preoperative immune checkpoint inhibitor (ICI) therapy may benefit patients with resectable non–small cell lung cancer (NSCLC).

The LCMC3 study is “unique” because it is the largest monotherapy trial of checkpoint inhibition in resectable NSCLC, and it’s “a landmark study” because it validated results from smaller trials and can serve as a benchmark for future ones, said Jay M. Lee, MD, of the University of California, Los Angeles.

Dr. Lee presented results from LCMC3 at the 2020 World Congress on Lung Cancer (Abstract PS01.05), which was rescheduled for January 2021.

The study included 181 patients, median age 65 years, with stage IB-IIIB NSCLC. The vast majority (90%) of patients were current/former smokers, and two-thirds had a nonsquamous histology. Patients were categorized in the following stages: 17 patients were staged at IB, 20 were IIA, 55 were IIB, 72 were IIIA, and 17 were IIIB.

Patients received 1,200 mg of neoadjuvant atezolizumab intravenously every 3 weeks for two cycles followed by resection between 30 and 50 days from the first cycle. Patients who benefited from the therapy continued adjuvant atezolizumab for 12 months.

The primary endpoint was major pathological response, defined as no more than 10% viable tumor cells at surgery, in patients without epidermal growth factor receptor or anaplastic lymphoma kinase mutations.
 

Results

Following atezolizumab treatment, 43% of patients were down-staged, and 19% were up-staged. Some degree of pathological regression was observed in all but 3 of the 159 patients who underwent resection.

Among the 144 patients included in the efficacy analysis, the major pathological response rate was 21%, with 7% of patients achieving a complete pathological response.

“We demonstrated that more than half of patients resected with a minimally invasive operation. Remarkably, only 15% required thoracotomy. The 92% complete resection rate is comparable, if not superior to, preoperative chemotherapy trials,” Dr. Lee said.

The majority (88%) of patients underwent surgical resection within a 20-day protocol window. The median time from end of neoadjuvant therapy to surgery was 22 days.

“Historically, the neoadjuvant chemotherapy window is much later for surgery, 3 weeks from neoadjuvant therapy, and that can be stretched to up to 56 days,” Dr. Lee said.

In an exploratory analysis, the 1.5-year overall survival rate was 91% for stage I and II disease and 87% for stage III disease. The survival in both cohorts was superior to that expected historically, Dr. Lee noted.

Intraoperative complications were rare (3%). Postoperative adverse reactions correlated with fewer viable tumor cells in the resected specimen.

One patient died following surgery after the first 30 days, which was deemed unrelated to treatment. Another patient died between 30 and 90 days from treatment-related pneumonitis.

“The LCMC3 study successfully met its primary endpoint of achieving major pathological response,” Dr. Lee concluded. “Neoadjuvant atezolizumab monotherapy was well tolerated, and resection was performed with low perioperative morbidity and mortality, usually within a narrow protocol window and with a short time frame from completion of atezolizumab and with a correspondingly high complete resection rate.”

The study’s results suggest that “neoadjuvant atezolizumab monotherapy is effective, well tolerated, and surgically acceptable,” said study discussant Shinichi Toyooka, MD, of Okayama (Japan) University Hospital.

“I would consider single-agent ICI neoadjuvant therapy for patients with early-stage disease and poor performance status, and an ICI plus chemotherapy for more advanced resectable cases, like locally advanced disease,” Dr. Toyooka said.

The LCMC3 study is sponsored by Genentech. Dr. Lee disclosed relationships with Genentech/Roche, AstraZeneca, Bristol-Myers Squibb, Merck, and Novartis. Dr. Toyooka disclosed relationships with AstraZeneca, Chugai, Taiho Pharmaceutical Group, and Ono Pharmaceutical.

Neoadjuvant atezolizumab prior to lung cancer surgery was well tolerated by patients with stage IB-IIIB lung cancer and produced a 21% major pathologic response rate, according to the primary analysis of the Lung Cancer Mutation Consortium (LCMC) 3 study.

Small pilot studies previously suggested that preoperative immune checkpoint inhibitor (ICI) therapy may benefit patients with resectable non–small cell lung cancer (NSCLC).

The LCMC3 study is “unique” because it is the largest monotherapy trial of checkpoint inhibition in resectable NSCLC, and it’s “a landmark study” because it validated results from smaller trials and can serve as a benchmark for future ones, said Jay M. Lee, MD, of the University of California, Los Angeles.

Dr. Lee presented results from LCMC3 at the 2020 World Congress on Lung Cancer (Abstract PS01.05), which was rescheduled for January 2021.

The study included 181 patients, median age 65 years, with stage IB-IIIB NSCLC. The vast majority (90%) of patients were current/former smokers, and two-thirds had a nonsquamous histology. Patients were categorized in the following stages: 17 patients were staged at IB, 20 were IIA, 55 were IIB, 72 were IIIA, and 17 were IIIB.

Patients received 1,200 mg of neoadjuvant atezolizumab intravenously every 3 weeks for two cycles followed by resection between 30 and 50 days from the first cycle. Patients who benefited from the therapy continued adjuvant atezolizumab for 12 months.

The primary endpoint was major pathological response, defined as no more than 10% viable tumor cells at surgery, in patients without epidermal growth factor receptor or anaplastic lymphoma kinase mutations.
 

Results

Following atezolizumab treatment, 43% of patients were down-staged, and 19% were up-staged. Some degree of pathological regression was observed in all but 3 of the 159 patients who underwent resection.

Among the 144 patients included in the efficacy analysis, the major pathological response rate was 21%, with 7% of patients achieving a complete pathological response.

“We demonstrated that more than half of patients resected with a minimally invasive operation. Remarkably, only 15% required thoracotomy. The 92% complete resection rate is comparable, if not superior to, preoperative chemotherapy trials,” Dr. Lee said.

The majority (88%) of patients underwent surgical resection within a 20-day protocol window. The median time from end of neoadjuvant therapy to surgery was 22 days.

“Historically, the neoadjuvant chemotherapy window is much later for surgery, 3 weeks from neoadjuvant therapy, and that can be stretched to up to 56 days,” Dr. Lee said.

In an exploratory analysis, the 1.5-year overall survival rate was 91% for stage I and II disease and 87% for stage III disease. The survival in both cohorts was superior to that expected historically, Dr. Lee noted.

Intraoperative complications were rare (3%). Postoperative adverse reactions correlated with fewer viable tumor cells in the resected specimen.

One patient died following surgery after the first 30 days, which was deemed unrelated to treatment. Another patient died between 30 and 90 days from treatment-related pneumonitis.

“The LCMC3 study successfully met its primary endpoint of achieving major pathological response,” Dr. Lee concluded. “Neoadjuvant atezolizumab monotherapy was well tolerated, and resection was performed with low perioperative morbidity and mortality, usually within a narrow protocol window and with a short time frame from completion of atezolizumab and with a correspondingly high complete resection rate.”

The study’s results suggest that “neoadjuvant atezolizumab monotherapy is effective, well tolerated, and surgically acceptable,” said study discussant Shinichi Toyooka, MD, of Okayama (Japan) University Hospital.

“I would consider single-agent ICI neoadjuvant therapy for patients with early-stage disease and poor performance status, and an ICI plus chemotherapy for more advanced resectable cases, like locally advanced disease,” Dr. Toyooka said.

The LCMC3 study is sponsored by Genentech. Dr. Lee disclosed relationships with Genentech/Roche, AstraZeneca, Bristol-Myers Squibb, Merck, and Novartis. Dr. Toyooka disclosed relationships with AstraZeneca, Chugai, Taiho Pharmaceutical Group, and Ono Pharmaceutical.

Asymptomatic screening of cancer patients receiving anticancer therapy detected a very low rate of COVID-19 in a retrospective study.

Of more than 2,000 patients, less than 1% were found to be COVID-19 positive on asymptomatic screening, an investigator reported at the AACR Virtual Meeting: COVID-19 and Cancer (Abstract S09-04).

While several models have been proposed to screen for COVID-19 among cancer patients, the optimal strategy remains unknown, said investigator Justin A. Shaya, MD, of the University of California, San Diego.

The most commonly used approach is symptom/exposure-based screening and testing. However, other models have combined this method with polymerase chain reaction (PCR) testing for asymptomatic high-risk patients (such as those undergoing bone marrow transplant, receiving chemotherapy, or with hematologic malignancies) or with PCR testing for all asymptomatic cancer patients.

Dr. Shaya’s institution implemented a novel COVID-19 screening protocol for cancer patients receiving infusional anticancer therapy in May 2020.

The protocol required SARS-CoV-2 PCR testing for asymptomatic patients 24-96 hours prior to infusion. However, testing was only required before the administration of anticancer therapy. Infusion visits for supportive care interventions did not require previsit testing.

The researchers retrospectively analyzed data from patients with active cancer receiving infusional anticancer therapy who had at least one asymptomatic SARS-CoV-2 PCR test between June 1 and Dec. 1, 2020. The primary outcome was the rate of COVID-19 positivity among asymptomatic patients.

Results

Among 2,202 patients identified, 21 (0.95%) were found to be COVID-19 positive on asymptomatic screening. Most of these patients (90.5%) had solid tumors, but two (9.5%) had hematologic malignancies.

With respect to treatment, 16 patients (76.2%) received cytotoxic chemotherapy, 2 (9.5%) received targeted therapy, 1 (4.7%) received immunotherapy, and 2 (9.5%) were on a clinical trial.

At a median follow-up of 174 days from a positive PCR test (range, 55-223 days), only two patients (9.5%) developed COVID-related symptoms. Both patients had acute leukemia, and one required hospitalization for COVID-related complications.

In the COVID-19–positive cohort, 20 (95.2%) patients had their anticancer therapy delayed or deferred, with a median delay of 21 days (range, 7-77 days).

In the overall cohort, an additional 26 patients (1.2%) developed symptomatic COVID-19 during the study period.

“These results are particularly interesting because they come from a high-quality center that sees a large number of patients,” said Solange Peters, MD, PhD, of the University of Lausanne (Switzerland), who was not involved in this study.

“As they suggest, it is still a debate on how efficient routine screening is, asking the question whether we’re really detecting COVID-19 infection in our patients. Of course, it depends on the time and environment,” Dr. Peters added.

Dr. Shaya acknowledged that the small sample size was a key limitation of the study. Thus, the results may not be generalizable to other regions.

“One of the most striking things is that asymptomatic patients suffer very few consequences of COVID-19 infection, except for patients with hematologic malignancies,” Dr. Shaya said during a live discussion. “The majority of our patients had solid tumors and failed to develop any signs/symptoms of COVID infection.

“Routine screening provides a lot of security, and our institution is big enough to allow for it, and it seems our teams enjoy the fact of knowing the COVID status for each patient,” he continued.

Dr. Shaya and Dr. Peters disclosed no conflicts of interest. No funding sources were reported in the presentation.

Asymptomatic screening of cancer patients receiving anticancer therapy detected a very low rate of COVID-19 in a retrospective study.

Of more than 2,000 patients, less than 1% were found to be COVID-19 positive on asymptomatic screening, an investigator reported at the AACR Virtual Meeting: COVID-19 and Cancer (Abstract S09-04).

While several models have been proposed to screen for COVID-19 among cancer patients, the optimal strategy remains unknown, said investigator Justin A. Shaya, MD, of the University of California, San Diego.

The most commonly used approach is symptom/exposure-based screening and testing. However, other models have combined this method with polymerase chain reaction (PCR) testing for asymptomatic high-risk patients (such as those undergoing bone marrow transplant, receiving chemotherapy, or with hematologic malignancies) or with PCR testing for all asymptomatic cancer patients.

Dr. Shaya’s institution implemented a novel COVID-19 screening protocol for cancer patients receiving infusional anticancer therapy in May 2020.

The protocol required SARS-CoV-2 PCR testing for asymptomatic patients 24-96 hours prior to infusion. However, testing was only required before the administration of anticancer therapy. Infusion visits for supportive care interventions did not require previsit testing.

The researchers retrospectively analyzed data from patients with active cancer receiving infusional anticancer therapy who had at least one asymptomatic SARS-CoV-2 PCR test between June 1 and Dec. 1, 2020. The primary outcome was the rate of COVID-19 positivity among asymptomatic patients.

Results

Among 2,202 patients identified, 21 (0.95%) were found to be COVID-19 positive on asymptomatic screening. Most of these patients (90.5%) had solid tumors, but two (9.5%) had hematologic malignancies.

With respect to treatment, 16 patients (76.2%) received cytotoxic chemotherapy, 2 (9.5%) received targeted therapy, 1 (4.7%) received immunotherapy, and 2 (9.5%) were on a clinical trial.

At a median follow-up of 174 days from a positive PCR test (range, 55-223 days), only two patients (9.5%) developed COVID-related symptoms. Both patients had acute leukemia, and one required hospitalization for COVID-related complications.

In the COVID-19–positive cohort, 20 (95.2%) patients had their anticancer therapy delayed or deferred, with a median delay of 21 days (range, 7-77 days).

In the overall cohort, an additional 26 patients (1.2%) developed symptomatic COVID-19 during the study period.

“These results are particularly interesting because they come from a high-quality center that sees a large number of patients,” said Solange Peters, MD, PhD, of the University of Lausanne (Switzerland), who was not involved in this study.

“As they suggest, it is still a debate on how efficient routine screening is, asking the question whether we’re really detecting COVID-19 infection in our patients. Of course, it depends on the time and environment,” Dr. Peters added.

Dr. Shaya acknowledged that the small sample size was a key limitation of the study. Thus, the results may not be generalizable to other regions.

“One of the most striking things is that asymptomatic patients suffer very few consequences of COVID-19 infection, except for patients with hematologic malignancies,” Dr. Shaya said during a live discussion. “The majority of our patients had solid tumors and failed to develop any signs/symptoms of COVID infection.

“Routine screening provides a lot of security, and our institution is big enough to allow for it, and it seems our teams enjoy the fact of knowing the COVID status for each patient,” he continued.

Dr. Shaya and Dr. Peters disclosed no conflicts of interest. No funding sources were reported in the presentation.

Asymptomatic screening of cancer patients receiving anticancer therapy detected a very low rate of COVID-19 in a retrospective study.

Of more than 2,000 patients, less than 1% were found to be COVID-19 positive on asymptomatic screening, an investigator reported at the AACR Virtual Meeting: COVID-19 and Cancer (Abstract S09-04).

While several models have been proposed to screen for COVID-19 among cancer patients, the optimal strategy remains unknown, said investigator Justin A. Shaya, MD, of the University of California, San Diego.

The most commonly used approach is symptom/exposure-based screening and testing. However, other models have combined this method with polymerase chain reaction (PCR) testing for asymptomatic high-risk patients (such as those undergoing bone marrow transplant, receiving chemotherapy, or with hematologic malignancies) or with PCR testing for all asymptomatic cancer patients.

Dr. Shaya’s institution implemented a novel COVID-19 screening protocol for cancer patients receiving infusional anticancer therapy in May 2020.

The protocol required SARS-CoV-2 PCR testing for asymptomatic patients 24-96 hours prior to infusion. However, testing was only required before the administration of anticancer therapy. Infusion visits for supportive care interventions did not require previsit testing.

The researchers retrospectively analyzed data from patients with active cancer receiving infusional anticancer therapy who had at least one asymptomatic SARS-CoV-2 PCR test between June 1 and Dec. 1, 2020. The primary outcome was the rate of COVID-19 positivity among asymptomatic patients.

Results

Among 2,202 patients identified, 21 (0.95%) were found to be COVID-19 positive on asymptomatic screening. Most of these patients (90.5%) had solid tumors, but two (9.5%) had hematologic malignancies.

With respect to treatment, 16 patients (76.2%) received cytotoxic chemotherapy, 2 (9.5%) received targeted therapy, 1 (4.7%) received immunotherapy, and 2 (9.5%) were on a clinical trial.

At a median follow-up of 174 days from a positive PCR test (range, 55-223 days), only two patients (9.5%) developed COVID-related symptoms. Both patients had acute leukemia, and one required hospitalization for COVID-related complications.

In the COVID-19–positive cohort, 20 (95.2%) patients had their anticancer therapy delayed or deferred, with a median delay of 21 days (range, 7-77 days).

In the overall cohort, an additional 26 patients (1.2%) developed symptomatic COVID-19 during the study period.

“These results are particularly interesting because they come from a high-quality center that sees a large number of patients,” said Solange Peters, MD, PhD, of the University of Lausanne (Switzerland), who was not involved in this study.

“As they suggest, it is still a debate on how efficient routine screening is, asking the question whether we’re really detecting COVID-19 infection in our patients. Of course, it depends on the time and environment,” Dr. Peters added.

Dr. Shaya acknowledged that the small sample size was a key limitation of the study. Thus, the results may not be generalizable to other regions.

“One of the most striking things is that asymptomatic patients suffer very few consequences of COVID-19 infection, except for patients with hematologic malignancies,” Dr. Shaya said during a live discussion. “The majority of our patients had solid tumors and failed to develop any signs/symptoms of COVID infection.

“Routine screening provides a lot of security, and our institution is big enough to allow for it, and it seems our teams enjoy the fact of knowing the COVID status for each patient,” he continued.

Dr. Shaya and Dr. Peters disclosed no conflicts of interest. No funding sources were reported in the presentation.

Best practices for managing cancer outpatients continue to evolve during the COVID-19 pandemic, with recent innovations in technology, operations, and communication.

“We’ve tried a lot of new things to ensure optimal care for our patients,” said Tiffany A. Traina, MD, of Memorial Sloan Kettering Cancer Center (MSKCC) in New York. “We need to effectively utilize all resources at our disposal to keep in touch with our patients during this time.”

Dr. Traina described the approach to outpatient management used at MSKCC during a presentation at the AACR Virtual Meeting: COVID-19 and Cancer.
 

Four guiding principles

MSKCC has established four guiding principles on how to manage cancer patients during the pandemic: openness, safety, technology, and staffing.

Openness ensures that decisions are guided by clinical priorities to provide optimal patient care and allow for prioritization of clinical research and education, Dr. Traina said.

The safety of patients and staff is of the utmost importance, she added. To ensure safety in the context of outpatient care, several operational levers were developed, including COVID surge planning, universal masking and personal protective equipment guidelines, remote work, clinical levers, and new dashboards and communications.

Dr. Traina said data analytics and dashboards have been key technological tools used to support evidence-based decision-making and deliver care remotely for patients during the pandemic.

Staffing resources have also shifted to support demand at different health system locations.
 

Screening, cohorting, and telemedicine

One measure MSKCC adopted is the MSK Engage Questionnaire, a COVID-19 screening questionnaire assigned to every patient with a scheduled outpatient visit. After completing the questionnaire, patients receive a response denoting whether they need to come into the outpatient setting.

On the staffing side, clinic coordinators prepare appointments accordingly, based on the risk level for each patient.

“We also try to cohort COVID-positive patients into particular areas within the outpatient setting,” Dr. Traina explained. “In addition, we control flow through ambulatory care locations by having separate patient entrances and use other tools to make flow as efficient as possible.”

On the technology side, interactive dashboards are being used to model traffic through different buildings.

“These data and analytics are useful for operational engineering, answering questions such as (1) Are there backups in chemotherapy? and (2) Are patients seeing one particular physician?” Dr. Traina explained. “One important key takeaway is the importance of frequently communicating simple messages through multiple mechanisms, including signage, websites, and dedicated resources.”

Other key technological measures are leveraging telemedicine to convert inpatient appointments to a virtual setting, as well as developing and deploying a system for centralized outpatient follow-up of COVID-19-positive patients.

“We saw a 3,000% increase in telemedicine utilization from February 2020 to June 2020,” Dr. Traina reported. “In a given month, we have approximately 230,000 outpatient visits, and a substantial proportion of these are now done via telemedicine.”

Dr. Traina also noted that multiple organizations have released guidelines addressing when to resume anticancer therapy in patients who have been COVID-19 positive. Adherence is important, as unnecessary COVID-19 testing may delay cancer therapy and is not recommended.

During a live discussion, Louis P. Voigt, MD, of MSKCC, said Dr. Traina’s presentation provided “a lot of good ideas for other institutions who may be facing similar challenges.”

Dr. Traina and Dr. Voigt disclosed no conflicts of interest. No funding sources were reported.

Best practices for managing cancer outpatients continue to evolve during the COVID-19 pandemic, with recent innovations in technology, operations, and communication.

“We’ve tried a lot of new things to ensure optimal care for our patients,” said Tiffany A. Traina, MD, of Memorial Sloan Kettering Cancer Center (MSKCC) in New York. “We need to effectively utilize all resources at our disposal to keep in touch with our patients during this time.”

Dr. Traina described the approach to outpatient management used at MSKCC during a presentation at the AACR Virtual Meeting: COVID-19 and Cancer.
 

Four guiding principles

MSKCC has established four guiding principles on how to manage cancer patients during the pandemic: openness, safety, technology, and staffing.

Openness ensures that decisions are guided by clinical priorities to provide optimal patient care and allow for prioritization of clinical research and education, Dr. Traina said.

The safety of patients and staff is of the utmost importance, she added. To ensure safety in the context of outpatient care, several operational levers were developed, including COVID surge planning, universal masking and personal protective equipment guidelines, remote work, clinical levers, and new dashboards and communications.

Dr. Traina said data analytics and dashboards have been key technological tools used to support evidence-based decision-making and deliver care remotely for patients during the pandemic.

Staffing resources have also shifted to support demand at different health system locations.
 

Screening, cohorting, and telemedicine

One measure MSKCC adopted is the MSK Engage Questionnaire, a COVID-19 screening questionnaire assigned to every patient with a scheduled outpatient visit. After completing the questionnaire, patients receive a response denoting whether they need to come into the outpatient setting.

On the staffing side, clinic coordinators prepare appointments accordingly, based on the risk level for each patient.

“We also try to cohort COVID-positive patients into particular areas within the outpatient setting,” Dr. Traina explained. “In addition, we control flow through ambulatory care locations by having separate patient entrances and use other tools to make flow as efficient as possible.”

On the technology side, interactive dashboards are being used to model traffic through different buildings.

“These data and analytics are useful for operational engineering, answering questions such as (1) Are there backups in chemotherapy? and (2) Are patients seeing one particular physician?” Dr. Traina explained. “One important key takeaway is the importance of frequently communicating simple messages through multiple mechanisms, including signage, websites, and dedicated resources.”

Other key technological measures are leveraging telemedicine to convert inpatient appointments to a virtual setting, as well as developing and deploying a system for centralized outpatient follow-up of COVID-19-positive patients.

“We saw a 3,000% increase in telemedicine utilization from February 2020 to June 2020,” Dr. Traina reported. “In a given month, we have approximately 230,000 outpatient visits, and a substantial proportion of these are now done via telemedicine.”

Dr. Traina also noted that multiple organizations have released guidelines addressing when to resume anticancer therapy in patients who have been COVID-19 positive. Adherence is important, as unnecessary COVID-19 testing may delay cancer therapy and is not recommended.

During a live discussion, Louis P. Voigt, MD, of MSKCC, said Dr. Traina’s presentation provided “a lot of good ideas for other institutions who may be facing similar challenges.”

Dr. Traina and Dr. Voigt disclosed no conflicts of interest. No funding sources were reported.

Best practices for managing cancer outpatients continue to evolve during the COVID-19 pandemic, with recent innovations in technology, operations, and communication.

“We’ve tried a lot of new things to ensure optimal care for our patients,” said Tiffany A. Traina, MD, of Memorial Sloan Kettering Cancer Center (MSKCC) in New York. “We need to effectively utilize all resources at our disposal to keep in touch with our patients during this time.”

Dr. Traina described the approach to outpatient management used at MSKCC during a presentation at the AACR Virtual Meeting: COVID-19 and Cancer.
 

Four guiding principles

MSKCC has established four guiding principles on how to manage cancer patients during the pandemic: openness, safety, technology, and staffing.

Openness ensures that decisions are guided by clinical priorities to provide optimal patient care and allow for prioritization of clinical research and education, Dr. Traina said.

The safety of patients and staff is of the utmost importance, she added. To ensure safety in the context of outpatient care, several operational levers were developed, including COVID surge planning, universal masking and personal protective equipment guidelines, remote work, clinical levers, and new dashboards and communications.

Dr. Traina said data analytics and dashboards have been key technological tools used to support evidence-based decision-making and deliver care remotely for patients during the pandemic.

Staffing resources have also shifted to support demand at different health system locations.
 

Screening, cohorting, and telemedicine

One measure MSKCC adopted is the MSK Engage Questionnaire, a COVID-19 screening questionnaire assigned to every patient with a scheduled outpatient visit. After completing the questionnaire, patients receive a response denoting whether they need to come into the outpatient setting.

On the staffing side, clinic coordinators prepare appointments accordingly, based on the risk level for each patient.

“We also try to cohort COVID-positive patients into particular areas within the outpatient setting,” Dr. Traina explained. “In addition, we control flow through ambulatory care locations by having separate patient entrances and use other tools to make flow as efficient as possible.”

On the technology side, interactive dashboards are being used to model traffic through different buildings.

“These data and analytics are useful for operational engineering, answering questions such as (1) Are there backups in chemotherapy? and (2) Are patients seeing one particular physician?” Dr. Traina explained. “One important key takeaway is the importance of frequently communicating simple messages through multiple mechanisms, including signage, websites, and dedicated resources.”

Other key technological measures are leveraging telemedicine to convert inpatient appointments to a virtual setting, as well as developing and deploying a system for centralized outpatient follow-up of COVID-19-positive patients.

“We saw a 3,000% increase in telemedicine utilization from February 2020 to June 2020,” Dr. Traina reported. “In a given month, we have approximately 230,000 outpatient visits, and a substantial proportion of these are now done via telemedicine.”

Dr. Traina also noted that multiple organizations have released guidelines addressing when to resume anticancer therapy in patients who have been COVID-19 positive. Adherence is important, as unnecessary COVID-19 testing may delay cancer therapy and is not recommended.

During a live discussion, Louis P. Voigt, MD, of MSKCC, said Dr. Traina’s presentation provided “a lot of good ideas for other institutions who may be facing similar challenges.”

Dr. Traina and Dr. Voigt disclosed no conflicts of interest. No funding sources were reported.