By David Douglas
NEW YORK (Reuters Health) - 1/5/2019
"Our comprehensive meta-analysis of adverse events from PD-1 and PD-L1 inhibitors in clinical trials will be an important guide for clinical practice," Dr. Michael L. Wang of The University of Texas MD Anderson Cancer Center, in Houston, told Reuters Health by email.
Dr. Wang and colleagues examined data from 125 clinical trials involving more than 20,000 patients. The studies were categorized as covering seven different groups: melanoma, lung cancer, gastrointestinal cancer, genitourinary cancer, hematologic malignant neoplasm, other cancers, and mixed cancer types.
In studies with relevant data, the overall incidence of all-grade adverse events was 66.0%, the most common being fatigue, pruritus and diarrhea. For adverse events of grade 3 or higher, including anemia and aspartate aminotransferase increase, the incidence was 14.0%.
The most frequent endocrine immune-related adverse events were hypothyroidism (6.07%) and hyperthyroidism (2.82%), the researchers report in JAMA Oncology, online April 25.
The highest mean all-grade adverse events incidence (1.72%) was seen in melanoma, but this did not differ much from the lowest incidence which was seen in lung cancer (1.55%). These and other data, say the researchers, "suggest that the mean incidences of all-grade and grade 3 or higher adverse events were similar across different cancer types."
However, PD-1 inhibitors were associated with a greater mean incidence of grade 3 or higher adverse events compared with PD-L1 inhibitors (odds ratio, 1.58; 95% confidence interval, 1.00 to 2.54). Comparison of PD-1 inhibitors showed this was also the case for nivolumab versus pembrolizumab (OR, 1.28; 95% CI, 0.97 to 1.79).
These findings, concluded Dr. Wang, "will help hematologists and oncologists throughout the world who use checkpoint inhibitors - both within and outside of clinical trials - to be able to predict, monitor, and mitigate side effects of these treatments. Further, it will enable clinicians to better inform their patients about potential side effects of checkpoint inhibitors."
In an editorial, Drs. John M. Kirkwood and Diwakar Davar of the UPMC Hillman Cancer Center, in Pittsburgh, observe that ultimately this and other studies "raise the question of whether the identification of immunogenic tumor antigens may enable more effective immunotherapy while avoiding self-directed immune responses."
Dr. Cary Gross, a professor of medicine at Yale University, in New Haven, Connecticut, who was not involved in the study, told Reuters Health by email, "These new agents have shown to be remarkably effective, and it's critical to understand their toxicity profile."
"While it's helpful to consolidate data from clinical trials, as these authors have done, it's important to understand that trial participants are frequently younger and healthier than patients in real world clinical practice," he added.
"Hence," Dr. Gross concluded, "what I want to know is what happens in the community - outside of the research setting - how are patients doing? Is the likelihood of experiencing side effects similar to what we're seeing in the studies?"
Dr. Benjamin Besse, a professor of medical oncology at Paris-Sud University, in Orsay, who also was not involved in the study, told Reuters Health by email, "The choice between PD-1 and PD-L1 inhibitors is based on an efficacy/toxicity ratio, and this meta-analysis focused only on toxicity. It gives a very nice overview of the immune-related adverse events. Because of the learning curve, one can wonder if the toxicity management improved with time."
JAMA Oncol 2019.