Concomitant COVID-19 Vaccination Tied to Improved OS with Immunotherapy

Co-exposure to a COVID-19 mRNA vaccine within 90 days of initiating a PD-(L)1 inhibitor was associated with a modest improvement in overall survival (OS), according to findings from a French nationwide real-world cohort study published in Annals of Oncology.

The primary efficacy analysis demonstrated a significant OS advantage for patients who received a COVID-19 mRNA vaccine within the specified window compared with those who did not. Kaplan-Meier estimates revealed a longer median OS of 21.6 months (95% CI, 21.1-22.2) in the vaccine co-exposed cohort compared with 18.5 months (95% CI, 18.0-19.0) in the non–co-exposed group. In propensity score-weighted Cox models, mRNA vaccine co-exposure was associated with a 9% reduction in the risk of all-cause mortality, translating to a weighted HR (wHR) of 0.91 (95% CI, 0.88-0.93).

The survival benefit remained largely consistent when investigators stratified the study population by distinct cancer localizations:

• Non–small cell lung cancer (NSCLC; wHR, 0.91; 95% CI, 0.88-0.95)
• Melanoma (wHR, 0.88; 95% CI, 0.78-0.98)
• Head and neck cancer (wHR, 0.84; 95% CI, 0.77-0.91)
• Kidney cancer (wHR, 0.85; 95% CI, 0.75-0.95)
• Small cell lung cancer (SCLC; wHR, 0.95; 95% CI, 0.87-1.04)
• Liver cancer (wHR, 0.94; 95% CI, 0.85-1.05)
• Bladder cancer (wHR, 0.92; 95% CI, 0.82-1.03)

“In this large nationwide study, COVID-19 mRNA vaccine within 3 months of anti–PD-(L)1 initiation was associated with a modest improvement in survival,” wrote lead study author Hugo Jourdain, PhD, of EPI-PHARE, French National Agency for Medicine and Health Product Safety and the French National Health Insurance Center, and coauthors, in the study. “While the magnitude of the association does not support modification in vaccination timing relative to anti–PD-(L)1 treatments, these findings provide population level evidence that may guide future studies to clarify the full potential implications in clinical practice.”

How was the nationwide cohort study structured?

The real-world study utilized individual data from the French National Health Data System, which records longitudinal healthcare information for nearly the entire French population. Across the system, all patients with characterized cancer types and initiation of a PD-(L)1 inhibitor between January 1, 2021, and December 31, 2022, were included in the study; patients were tracked from the date of first anti–PD-(L)1 delivery to death, last care reimbursement date, or July 31, 2025.

The evaluable population included 30,898 patients who were assigned via 1:1 matching into either the co-exposed group (n=15,449) or the non–co-exposed group (n=15,449). Patient sorting for matching was based on age in 3-year increments, sex, calendar quarter of anti–PD-(L1) initiation, and medical indication.

Co-exposure was defined as receiving at least 1 dose of a COVID-19 mRNA vaccine within 90 days before or 90 days after the first delivery of an anti–PD-(L)1 agent. For the co-exposed group, vaccination occurred at an average of 42.0 days (SD, 26.9) before or after treatment initiation. Notably, in the non–co-exposed group, 79.9% of patients received at least 1 COVID-19 dose outside of this timeframe.

Across the entire sample, pembrolizumab (Keytruda) was the most common treatment (54.2%), followed by nivolumab (Opdivo; 20.9%), atezolizumab (Tecentriq; 13.1%), durvalumab (Imfinzi; 8.8%), and avelumab (Bavencio; 3.0%). Regarding treatment regimens, 48.6% received anti–PD-(L)1 monotherapy, 38.9% received anti–PD-(L)1 agents plus chemotherapy, and 3.3% received an anti–PD-(L)1 agent plus an anti-CTLA4 agent, with 9.2% receiving other combinations.

The most common disease states in the study were:

• NSCLC: 54.8%
• Melanoma: 10.4%
• Head and neck cancer: 8.7%
• Extensive-stage SCLC: 7.2%
• Kidney cancer: 6.8%
• Liver cancer: 6.1%
• Bladder cancer: 6.0%

What safety parameters and vaccine timelines were noted during follow-up?

The study tracked baseline clinical interventions and health events during the critical 180-day co-exposure window. In the co-exposed cohort, 28.0% of patients received an influenza vaccination, whereas only 15.4% of non–co-exposed patients received an influenza vaccine during the same period. Positive COVID-19 tests were recorded in 4.5% of the co-exposed group and 10.3% of the non–co-exposed group during this timeframe. Furthermore, COVID-19–related hospitalization occurred in 0.6% of co-exposed patients compared with 2.0% of non–co-exposed individuals.

The median duration of clinical follow-up reached 21.4 months in the co-exposed group and 18.2 months in the non–co-exposed group, during which death from any cause occurred in 67.3% and 70.5% of patients, respectively.

Reference

Jourdain H, Haddy N, Singier A, et al. Overall survival according to COVID-19 mRNA vaccination co-exposure in patients treated with PD-(L)1 inhibitors: a French real-world cohort study. Ann Oncol. Published online May 18, 2026. doi:10.1016/j.annonc.2026.05.002