Treatment-related pneumonitis after thoracic radiotherapy/chemoradiotherapy combined with anti-PD-1 monoclonal antibodies in advanced esophageal squamous cell carcinoma.

Abstract

This study aims to evaluate the risk factors of treatment-related pneumonitis (TRP) following thoracic radiotherapy/chemoradiotherapy combined with anti-PD‑1 monoclonal antibodies (mAbs) in patients with advanced esophageal squamous cell carcinoma (ESCC). We retrospectively reviewed 97patients with advanced ESCC who were treated with thoracic radiotherapy/chemoradiotherapy combined with anti-PD‑1 mAbs. Among them, 56patients received concurrent radiotherapy with anti-PD‑1 mAbs and 41patients received sequential radiotherapy with anti-PD‑1 mAbs. The median prescribed planning target volume (PTV) dose was 59.4 Gy (range from 50.4 to 66 Gy, 1.8-2.2 Gy/fraction). Clinical characteristics, the percentage of lung volume receiving more than 5-50 Gy in increments of 5 Gy (V5-V50, respectively) and the mean lung dose (MLD) were analyzed as potential risk factors for TRP. 46.4% (45/97), 20.6% (20/97), 20.6% (20/97), 4.1% (4/97), and 1.0% (1/97) of the patients developed any grade of TRP, grade1 TRP, grade2 TRP, grade3 TRP, and fatal (grade5) TRP, respectively. Anti-PD‑1 mAbs administered concurrently with radiotherapy, V5, V10, V15, V25, V30, V35, V40 and MLD were associated with the occurrence of grade2 or higher TRP. Concurrent therapy (P = 0.010, OR = 3.990) and V5 (P = 0.001, OR = 1.126) were independent risk factors for grade2 or higher TRP. According to the receiver operating characteristic (ROC) curve analysis, the optimal V5 threshold for predicting grade2 or higher TRP was 55.7%. The combination of thoracic radiotherapy/chemoradiotherapy with anti-PD‑1 mAbs displayed atolerable pulmonary safety profile. Although the incidence of TRP was high, grade1-2 TRP accounted for the majority. Anti-PD‑1 mAbs administered concurrently with radiotherapy and the lung V5 were significantly associated with the occurrence of grade2 or higher TRP. Therefore, it seems safer to control V5 below 55% in clinical, especially for the high-risk populations receiving concurrent therapy.