Grant Number:	1R21CA318798-01 Interpret this number
Primary Investigator:	Hu, Chen
Organization:	Johns Hopkins University
Project Title:	Defining the Role of Twice-Daily Thoracic Radiotherapy in Limited-Stage Sclc in the Immunotherapy Era: a Trial Integration Approach
Fiscal Year:	2026

Limited-stage small cell lung cancer (LS-SCLC) remains one of the few thoracic malignancies in which the optimal thoracic radiotherapy schedule is uncertain. The landmark INT-0096 trial supported twice-daily (BID) therapy, but subsequent randomized trials using modern once-daily (QD) regimens showed comparable outcomes. As immunotherapy, radiotherapy delivery techniques, and patient selection have evolved, the clinical relevance of the historic BID benefit is unclear. The recently completed NRG-LU005 trial, which incorporated chemoradiation with concurrent and consolidative immunotherapy, provides a unique opportunity to revisit this long- standing question in the contemporary era. However, LU005 was not randomized by schedule, and the influence of treatment feasibility, geography, and socioeconomic context on real-world schedule selection remains unknown. This project will use causal inference, modern machine-learning methods, and individual-level data from NRG-LU005, CONVERT, and CALGB-30610 to determine when and for whom BID therapy remains beneficial. Aim 1 will establish how the historical randomized evidence for BID translates to the immunotherapy era by estimating the adjusted schedule effect within LU005 and transporting randomized effects from CONVERT and CALGB-30610 to the LU005 population. Harmonized datasets will be integrated through a causal-triangulation framework to identify concordance or divergence across eras. Aim 2 will identify patient, tumor, and delivery- context subgroups most likely to benefit from BID therapy. We will use debiased machine- learning estimators and causal-discovery methods to estimate individualized treatment effects, followed by cross-era validation to distinguish robust modifiers from context-dependent or hypothesis-generating factors. This project directly addresses a major evidence gap in LS-SCLC, where new randomized trials of fractionation are infeasible or readily available soon. It introduces an innovative analytic framework that combines internal validity from randomized trials with the external relevance of contemporary immunotherapy-era data. By quantifying both treatment benefit and delivery feasibility, the study will produce actionable, patient-centered evidence to inform NCCN guidelines, cooperative-group trial development, and real-world radiotherapy decision making. The resulting analytic pipeline will serve as a generalizable blueprint for integrating legacy trials with modern datasets to evaluate treatment-intensification strategies in oncology.





None