Higher lung dose and shorter immunotherapy-to-radiotherapy intervals predict symptomatic pneumonitis

After immune checkpoint inhibitor exposure, higher EQD2-adjusted lung dose and shorter intervals before thoracic RT were associated with symptomatic pneumonitis.

KEY POINTS

  • This retrospective single-center study included 98 patients with lung cancer receiving thoracic radiotherapy after exposure to immune checkpoint inhibitors between October 2024 and October 2025. Non-small cell lung cancer accounted for 84.7%, small cell lung cancer for 13.3%, and all patients received VMAT with heterogeneous fractionation schedules.
  • The primary endpoint was CTCAE grade ≥2 symptomatic pneumonitis within 90 days of radiotherapy completion, accounting for death as a competing event. Pneumonitis occurred in 26 patients, while 70 remained event-free and two died before developing pneumonitis, corresponding to a 90-day cumulative incidence of 26.5% (95% CI 17.7–35.3%).
  • Archived treatment plans were reconstructed voxel by voxel and converted to EQD2 using α/β = 3 Gy, rather than applying a single conversion to organ-level mean lung dose. Fraction sizes ranged from 1.5 to 10 Gy, making the biological conversion relevant for comparing heterogeneous schedules.
  • Patients developing pneumonitis had a higher mean EQD2-adjusted lung dose than event-free patients (8.06 ± 2.00 versus 6.41 ± 2.28 Gy). In the primary Fine–Gray regression model, each additional Gy was associated with a 33.9% higher subdistribution hazard of symptomatic pneumonitis (SHR 1.339, 95% CI 1.113–1.612; p=0.002).
  • The median interval between the last immunotherapy administration and the first RT fraction was 32 days. Longer intervals were associated with lower pneumonitis incidence: SHR 0.784 per additional 10 days (95% CI 0.639–0.963; p=0.020), after adjustment for EQD2-adjusted mean lung dose.
  • Physical mean lung dose showed a consistent association with symptomatic pneumonitis (SHR 1.197 per Gy, 95% CI 1.036–1.382; p=0.015). An exploratory prediction model combining EQD2-adjusted lung dose and treatment interval produced a bootstrap-corrected AUC of 0.760, but neither the proposed 7.18 Gy dose cutoff nor a 30-day interval threshold was externally validated.
  • Interpretation is limited by the retrospective ascertainment of pneumonitis, heterogeneous treatment intent and fractionation, incomplete characterization of immune checkpoint inhibitor exposure, and difficulty distinguishing radiation pneumonitis from immune-mediated or mixed pulmonary inflammation. The absence of an ICI-naive comparator prevents conclusions about immunotherapy-specific radiosensitization.

CLINICAL TAKEAWAY

In patients receiving thoracic RT after immune checkpoint inhibitors, both biologically adjusted lung dose and the interval since the last immunotherapy administration may be relevant to pneumonitis risk assessment. However, these observational associations do not establish a safe treatment interval or new lung dose constraint, and prospective validation with standardized pneumonitis adjudication is required before changing clinical schedules.

SOURCE

Frontiers in Oncology

Browse more research Suggest a correction