PET-Guided Dose Escalation for Stage III Non-Small Cell Lung Cancer: A Randomized Clinical Trial

Uniform radiation dose escalation has not improved outcomes in locally advanced non-small cell lung cancer (NSCLC), and locoregional failure remains common despite multimodal therapy. We evaluated whether fluorodeoxyglucose positron emission tomography (FDG-PET)-guided dose escalation to metabolically active tumor subvolumes reduces locoregional failure compared with standard chemoradiation.

PET-BOOST was a multi-institutional, open-label, phase 2 randomized trial conducted at 7 Canadian centers (January 2017-May 2022). Eligible patients had inoperable, FDG-avid stage IIIA/IIIB NSCLC and an achievable boost dose of at least 65 Gy. Patients were randomized 1:1 to standard chemoradiation with 60 Gy in 30 fractions or metabolically adaptive radiotherapy with an integrated boost to FDG-avid subvolumes to a maximum of 85 Gy in 30 fractions. Concurrent platinum-based chemotherapy was planned, and consolidation durvalumab was permitted after 2018. The primary endpoint was 2-year locoregional failure; secondary endpoints included overall survival (OS), progression-free survival (PFS), grade ≥3 treatment-related toxicity, and quality of life (QOL).

Seventy-eight patients were randomized: 40 to standard radiotherapy and 38 to dose escalation, with a median follow-up of 5.0 years. Median age was 64.4 years; 53% were male, 62% had stage IIIA disease, and 76% received consolidation durvalumab. Median prescription boost dose was 75 Gy. Two-year locoregional failure was 27.5% with standard radiotherapy versus 18.4% with dose escalation (P=.13). Two-year OS (72.5% vs. 73.7%; P=.91) and PFS (52.5% vs. 57.9%; P=.74) did not differ between arms. Grade ≥3 treatment-related toxicities occurred in 25% versus 37% of patients (P=.26), with one grade 5 pulmonary hemorrhage in the dose-escalation arm. QOL did not differ between arms.

Metabolically adaptive PET-guided radiotherapy dose escalation was feasible in the context of consolidation immunotherapy and resulted in a numerically lower rate of locoregional failure, but did not significantly improve locoregional control or survival. The lower-than-anticipated event rate may have limited statistical power, and larger studies would be required to determine whether selective dose escalation provides a clinically meaningful improvement in locoregional control.