COVICI-SURV: COVID-19 Vaccination Around ICI Start and Survival

Overview

COVICI-SURV is a nationwide, retrospective observational cohort study using the French national health data system (SNDS) to evaluate whether COVID-19 vaccination, particularly mRNA vaccination, received within 100 days before or after immune checkpoint inhibitor (ICI) initiation is associated with survival among adults with cancer. The cohort includes 95,015 adults initiating anti-PD-1, anti-PD-L1 or anti-CTLA-4 therapy between 28 December 2020 and 22 March 2023. Pre-ICI vaccination is analysed using inverse probability of treatment weighting, and post-ICI vaccination among patients without COVID-19 vaccination in the preceding 100 days is analysed using a cloning-censoring-weighting target-trial emulation. The primary outcome is all-cause mortality; secondary outcomes include cancer-specific mortality, non-COVID-19 mortality and real-world progression-free survival. Follow-up extends to 31 December 2024 for all-cause mortality and to 31 December 2023 for cause-specific mortality.

"This study evaluates the association between peri-ICI COVID-19 vaccination and survival in a nationwide population of adults with cancer initiating immune checkpoint inhibitor therapy in France. The exposure of interest is receipt of at least one COVID-19 mRNA vaccine dose within a prespecified 100-day window before ICI initiation (pre-ICI analysis) or within 100 days after ICI initiation (post-ICI analysis). The source population is derived from the SNDS and includes adults aged 18 years or older with a cancer diagnosis who initiated a first ICI administration (anti-PD-1, anti-PD-L1 or anti-CTLA-4) between 28 December 2020, corresponding to first COVID-19 vaccine availability in France, and 22 March 2023. ICI initiation is the index date/time zero. Patients are required to have at least 3 years of observable SNDS history before ICI initiation. Patients with solid organ or bone marrow transplantation or HIV infection are excluded. Two complementary causal contrasts are prespecified. First, in the pre-ICI window, patients receiving at least one mRNA COVID-19 vaccine dose within 100 days before ICI initiation are compared with patients receiving no COVID-19 vaccine in that window; mixed or only non-mRNA schedules in that window are excluded from the primary pre-ICI analysis. Baseline confounding is addressed using inverse probability of treatment weighting. Second, in the post-ICI window, a target trial is emulated among patients without COVID-19 vaccination in the 100 days before ICI initiation. Eligible patients are cloned into strategies of receiving at least one mRNA COVID-19 vaccine dose within 100 days after ICI initiation versus receiving no COVID-19 vaccine in that window. Protocol-deviation censoring and inverse probability of censoring weights are used to address immortal-time bias, time-dependent vaccination and informative censoring. The primary endpoint is all-cause mortality. Secondary endpoints include cancer-specific mortality, non-COVID-19 mortality and real-world progression-free survival defined from ICI discontinuation or death. Analyses include time-varying hazard ratios and absolute survival differences at clinically relevant time points, cancer-site subgroup analyses, sensitivity analyses using shorter exposure windows and ICI-continuation definitions, comparisons by vaccine platform, analyses of other adult vaccines as control exposures, documented COVID-19 infection, and positive/negative control outcomes to explore residual confounding."