Toxicity Within 6 Months of Heterogeneous Fluorodeoxyglucose-Guided Radiotherapy Dose Escalation for Locally Advanced Non-Small Cell Lung Cancer in the Scandinavian Randomized Phase III NARLAL2 Trial.

Abstract

Radiation dose escalation for locally advanced non-small cell lung cancer (LA-NSCLC) has been challenged by toxicity concerns. The Scandinavian phase III multicenter dose-escalation trial NARLAL2 (ClinicalTrials.gov identifier: NCT02354274) used a novel approach to dose escalation: heterogeneous escalation driven by the fluorodeoxyglucose positron emission tomography-avid region, with strict normal tissue dose constraints. We report early toxicity within 6 months of random assignment. Patients were recruited from seven institutions in Scandinavia. Eligibility criteria included performance status 0-1, NSCLC stage IIB-IIIB, and feasibility of delivering 66 Gy/33 fraction treatment plan. Patients were randomly assigned between standard (66 Gy) and heterogeneously dose-escalated radiotherapy. Two treatment plans were made for each patient before random assignment with matched mean lung dose and V_20Gy, and strict dose constraints for all normal tissues. Toxicity was evaluated weekly during radiotherapy, and every 3 months after random assignment. Concurrent chemotherapy was cisplatin/carboplatin and vinorelbine. Between January 2015 and March 2023, 350 patients were randomly assigned. The as-treated analysis included 178 patients in the standard and 172 in dose-escalated (mean tumor dose 88 Gy) arms. Median gross tumor and planning target volumes were, respectively, 54 cm³ and 321 cm³ (standard arm) and 61 cm³ and 339 cm³ (escalated arm). No difference in early toxicity between the two arms was observed. Grade 2 esophagitis during radiotherapy was 28.1% and 25.6%, grade 3 esophagitis 7.3% and 4.1%, grade 2 pneumonitis 15.7% and 20.3%, and grade 3 pneumonitis 3.9% and 5.8% in standard and escalated arms, respectively. For both arms, the maximum grade of early toxicity aggregated over all toxicities was 35% and 1% for grades ≥3 and 5, respectively. Four patients died from potential treatment-related toxicity. Heterogeneous dose escalation did not increase early toxicity despite delivery of 88 Gy mean dose to the primary tumor, demonstrating this as an attractive strategy for LA-NSCLC radiotherapy dose escalation.