Optimal timing of PD-1 inhibitor immunotherapy administration to stereotactic radiosurgery in malignant gliomas: a propensity score–matched multi-institutional cohort study

OBJECTIVE: The optimal order of treatment with programmed cell death protein 1 (PD-1) inhibitor immunotherapy and stereotactic radiosurgery (SRS) in malignant gliomas remains unclear. This study examined whether the timing of PD-1 inhibitor administration relative to SRS affects patient neurological outcomes and survival in a multi-institutional propensity score-matched cohort. METHODS: A retrospective analysis was conducted using the TriNetX Research Network database (2005-2022), identifying adults with malignant gliomas who received SRS and at least one PD-1 inhibitor. Patients were stratified into two cohorts based on whether PD-1 inhibition occurred 8 weeks before SRS (pre-SRS) versus after SRS (post-SRS). Propensity score matching (1:1) was performed for demographics, comorbidities, and relevant medications. Clinical outcomes, including mortality, cognitive impairments, cranial nerve deficits, seizures, hemorrhagic complications, and functional impairments, were assessed over a 3-year follow-up period. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated, and Kaplan-Meier (KM) survival analyses were used to compare cumulative incidence rates. RESULTS: Before matching, the pre-SRS (n = 429) versus post-SRS (n = 381) cohorts differed significantly in sex (female, unknown), ethnicity (Hispanic/Latino, unknown), race (Asian, American Indian, other), overweight/obesity, type 2 diabetes mellitus, lipidemia, anticoagulant use, durvalumab use, and atezolizumab use (all p < 0.05). After propensity score matching (n = 233 per group), PD-1 inhibition pre- versus post-SRS showed no significant differences in baseline characteristics except for age (mean 63.50 [SD 11.84] years vs 63.36 [SD 10.82] years, p = 0.012) and tremelimumab use (n = 0 [0%] vs n = ≤ 10 [≤ 4.29%], p = 0.001). In propensity score-matched patients, PD-1 inhibitor use before SRS was associated with significantly higher odds of cognitive impairments (OR 1.75, 95% CI 1.04-2.98, p = 0.034), cranial nerve deficits (OR 2.00, 95% CI 1.11-3.69, p = 0.021), and mortality (OR 1.53, 95% CI 1.06-2.20, p = 0.023), while other outcomes including behavioral/mood changes (OR 0.91, p = 0.726), gait/coordination disorders (OR 1.28, p = 0.421), hemorrhagic complications (OR 0.97, p = 0.930), motor deficits (OR 0.86, p = 0.590), seizures/epilepsy (OR 0.99, p = 0.971), and sensory disturbances (OR 0.91, p = 0.737) showed no significant differences. KM analyses confirmed elevated risks of cognitive impairment (p = 0.0076), cranial nerve deficits (p = 0.0104), and mortality (p = 0.0109) in the pre-SRS group. CONCLUSIONS: Pre-SRS PD-1 inhibitor use was associated with more neurological deficits and lower survival, highlighting the need for prospective studies on optimal SRS immunotherapy timing in malignant gliomas.