AI-generated research brief — verify at source
Adding ICIs to Radiotherapy Improves Survival, But Sequencing and Tumor Type Are Critical
Across 41 randomized trials enrolling more than 15,000 patients, adding immune checkpoint inhibitors to a radiotherapy backbone produced approximately 20% lower risk of disease progression (PFS pooled HR 0.80, 95% CI 0.68–0.93) and, after excluding glioblastoma trials, about 17% lower risk of death (OS pooled HR 0.83, 95% CI 0.70–0.99). The survival signal was strongest when ICIs were delivered in the adjuvant setting rather than concurrently or as induction therapy, underscoring that how and when these treatments are combined matters as much as whether they are combined at all.
What Was Studied
This analysis examined whether adding immune checkpoint inhibitors to radiotherapy — or radiotherapy to an ICI backbone — improves overall survival, progression-free survival, and event-free survival in patients with solid tumors. Despite preclinical evidence of synergy between these two modalities, randomized trials have produced conflicting results, creating a need for a comprehensive, continuously updated evidence synthesis.
How It Was Studied
The investigators conducted a living systematic review and meta-analysis, searching PubMed and EMBASE for randomized controlled trials published between January 2010 and January 2026. Two distinct strategies were evaluated: trials that added ICIs to a radiotherapy backbone (35 trials) and trials that added radiotherapy to an ICI backbone (6 trials). In total, 41 trials enrolling 15,049 patients across multiple solid tumor types were included. Hazard ratios for OS, PFS, and EFS were pooled using random-effects models, risk of bias was assessed using the RoB-2 tool, and meta-regression was applied to explore subgroup differences including ICI sequencing relative to radiotherapy.
What Was Observed
- Overall survival across ICI trials showed a trend toward benefit that did not reach statistical significance in the full dataset — approximately 12% lower risk of death (pooled HR 0.88, 95% CI 0.75–1.02; 28 trials). However, after excluding glioblastoma trials, OS benefit became statistically significant, corresponding to about 17% lower mortality risk (pooled HR 0.83, 95% CI 0.70–0.99; 23 trials), revealing that glioblastoma outcomes meaningfully diluted the overall signal.
- Progression-free survival improved significantly with the addition of ICIs, with roughly 20% lower risk of disease progression or death (pooled HR 0.80, 95% CI 0.68–0.93; 27 trials). Event-free survival showed an even larger relative improvement of approximately 27% (pooled HR 0.73, 95% CI 0.55–0.99; 7 trials), though the EFS estimate spans a wide confidence interval, reflecting fewer contributing trials.
- Treatment sequencing was identified as a critical determinant of benefit. Adjuvant ICI delivery conferred substantially greater OS benefit than concurrent or induction approaches (pooled HR 0.81 versus 0.98; interaction p = 0.045), a pattern that was also evident for PFS (pooled HR 0.69 versus 0.94; interaction p = 0.013), suggesting that the timing of immunotherapy relative to radiation shapes the magnitude of clinical gain.
Why This Matters
This meta-analysis synthesizes the broadest randomized evidence base assembled to date on the radiotherapy-ICI combination, providing a field-wide overview that individual trials cannot offer. The finding that tumor type — particularly glioblastoma — and ICI sequencing substantially modify the treatment effect has direct implications for how future trials should be designed: stratification by tumor histology and careful specification of treatment timing are essential. The living meta-analysis framework, updated as new trial data emerge, is intended to support ongoing evidence surveillance in this rapidly evolving area.
How to Read This Result
This high-quality synthesis of randomized trial data supports a positive direction of effect for radiotherapy-ICI combinations, particularly for PFS and OS outside of glioblastoma, but the subgroup dependency of the OS signal and substantial tumor-specific heterogeneity across trials mean that pooled estimates should be interpreted cautiously and not applied uniformly across all solid tumor types or treatment settings.
Limitations
The results across included randomized trials were inconsistent, and tumor-specific heterogeneity — most notably the divergent behavior of glioblastoma — limits the interpretability of any single pooled estimate. Variability in how ICIs were sequenced relative to radiotherapy (adjuvant versus concurrent versus induction) introduces additional heterogeneity across trials. Critically, the OS benefit only reached statistical significance after removing glioblastoma trials, indicating that the primary pooled result is sensitive to subgroup inclusion and should not be treated as a universal finding.