according to research published in .
However, the research also suggests that real-world patients who receive ICIs achieve longer survival than patients on standard-of-care medications.
“Patients receiving ICIs in real-world practice may differ from those enrolled in trials in a variety of ways, including age, race, performance status, and comorbidity burden,” said study author, of the University of Southern California, Los Angeles.
Dr. Lee noted that only 3%-4% of cancer patients participate in clinical trials. In fact, more than half of patients with melanoma and nearly three-quarters of those with non–small cell lung cancer (NSCLC) do not meet criteria for eligibility in clinical trials, he said.
To examine the discrepancies between real-world practice and clinical trials and to better understand which patients receive ICIs in clinical practice, Dr. Lee and colleagues conducted a retrospective analysis using electronic health record data from Veterans Administration (VA) facilities nationwide.
The researchers identified 11,888 cancer patients who were treated with ICIs. The cohort included patients who are underrepresented in pivotal clinical trials, including older, non-White, and/or higher disease-burdened patients.
The majority of patients were treated for NSCLC (51.1%), followed by melanoma (14.4%), renal cell carcinoma (RCC; 8.1%), squamous cell carcinoma of the head and neck (6.8%), urothelial cancer (6.4%), hepatocellular carcinoma (4.5%), and other less common cancer types (8.8%).
Overall survival by indication
In general, median overall survival (OS) in the VA cohort was inferior to median OS reported in clinical trials. However, patients treated with first-line nivolumab for melanoma and second-line pembrolizumab or nivolumab for NSCLC had similar OS in the real-world and trial data.
The researchers did not report exact OS numbers from clinical trials. However, they did report the exact numbers from the VA cohort and show OS differences between the VA cohort and clinical trials graphically.
Among patients in the VA cohort, the median OS was:
- 25.5 months in melanoma patients on first-line nivolumab
- 16.3 months in RCC patients receiving nivolumab in the second line or higher
- 14 months in RCC patients on first-line ipilimumab and nivolumab
- 10.6 months in NSCLC patients on first-line pembrolizumab
- 9.9 months in NSCLC patients receiving pembrolizumab or nivolumab in the second line or higher
- 9.1 months in NSCLC patients on first-line pembrolizumab and platinum-based chemotherapy
- 6.7 months in urothelial cancer patients receiving ICIs in the second line or higher.
A number of factors may have contributed to the shorter OS observed in the VA cohort, according to the researchers. The VA cohort is predominantly male, is older, and has a higher degree of comorbidity, compared with patients in clinical trials.
In addition, no data are available to determine the cause for discontinuation of therapy, and VA patients may have received ICIs after failing multiple lines of previous therapy, while clinical trials may limit patients to only one or two previous lines of therapy.
After stratifying VA patients by frailty status, the OS among non-frail patients was more similar to the OS reported in clinical trials.
“Real-world outcomes from the VA were more similar when adjusted for frailty, which shows the importance of patient diversity in clinical trials,” Dr. Lee said. He added that the definition of frailty among VA patients included potential injury during combat and therefore differs from a generic frailty definition.