Association of Survival With Adjuvant Chemotherapy Among Patients With Early-Stage Non–Small Cell Lung Cancer With vs Without High-Risk Clinicopathologic Features
Ranjan Pathak,Sarah B. Goldberg,Maureen Canavan,Jeph Herrin,Jessica R. Hoag,Michelle C. Salazar,Marianna Papageorge,Theresa Ermer,Daniel J. Boffa
DOI: https://doi.org/10.1001/jamaoncol.2020.4232
IF: 33.006
2020-11-01
JAMA Oncology
Abstract:<span><strong>Question</strong> <span>Is adjuvant chemotherapy associated with survival in patients with node-negative, early-stage non–small cell lung cancer (NSCLC) with vs without high-risk clinicopathologic features?</span><strong>Findings</strong> <span>In this cohort study of 50 814 patients with NSCLC, adjuvant chemotherapy use was associated with a survival benefit if the tumor was larger than 3 to 4 cm and sublobar resection was performed, the tumor was larger than 4 to 5 cm and at least 1 high-risk pathologic feature was present, or the tumor was larger than 5 cm irrespective of high-risk pathologic features.</span><strong>Meaning</strong> <span>The findings suggest that high-risk pathologic features, extent of resection, and tumor size should be simultaneously considered when evaluating patients with early-stage NSCLC for adjuvant chemotherapy.</span><strong>Importance</strong> <span>Tumor size larger than 4 cm is accepted as an indication for adjuvant chemotherapy in patients with node-negative non–small cell lung cancer (NSCLC). Treatment guidelines suggest that high-risk features are also associated with the efficacy of adjuvant chemotherapy among patients with early-stage NSCLC, yet this association is understudied.</span><strong>Objective</strong> <span>To assess the association between adjuvant chemotherapy and survival in the presence and absence of high-risk pathologic features in patients with node-negative early-stage NSCLC.</span><strong>Design, Setting, and Participants</strong> <span>This retrospective cohort study using data from the National Cancer Database included 50 814 treatment-naive patients with a completely resected node-negative NSCLC diagnosed between January 1, 2010, and December 31, 2015. The study was limited to patients who survived at least 6 weeks after surgery (ie, estimated median time to initiate adjuvant chemotherapy after surgery) to mitigate immortal time bias. Statistical analysis was performed from December 1, 2018, to February 29, 2020.</span><strong>Exposures</strong> <span>Adjuvant chemotherapy use vs observation, stratified according to the presence or absence of high-risk pathologic features (visceral pleural invasion, lymphovascular invasion, and high-grade histologic findings), sublobar surgery, and tumor size.</span><strong>Main Outcomes and Measures</strong> <span>The association of high-risk pathologic features with survival after adjuvant chemotherapy vs observation was evaluated using Cox proportional hazards regression models.</span><strong>Results</strong> <span>Overall, 50 814 eligible patients with NSCLC (27 365 women [53.9%]; mean [SD] age, 67.4 [9.5] years]) were identified, including 4220 (8.3%) who received adjuvant chemotherapy and 46 594 (91.7%) who did not receive adjuvant chemotherapy. Among patients with tumors 3 cm or smaller, chemotherapy was not associated with improved survival (hazard ratio [HR], 1.10; 95% CI, 0.96-1.26; <i>P</i> = .17). For patients with tumors larger than 3 cm to 4 cm, adjuvant chemotherapy was associated with a survival benefit among patients who underwent sublobar surgery (HR, 0.72; 95% CI, 0.56-0.93; <i>P</i> = .004). For tumors larger than 4 cm to 5 cm, a survival benefit was associated with adjuvant chemotherapy only in patients with at least 1 high-risk pathologic feature (HR, 0.67; 95% CI, 0.56-0.80; <i>P</i> = .02). For tumors larger than 5 cm, adjuvant chemotherapy was associated with a survival benefit irrespective of the presence of high-risk pathologic features (HR, 0.75; 95% CI, 0.61-0.91; <i>P</i> = .004).</span><strong>Conclusions and Relevance</strong> <span>In this cohort study, tumor size alone was not associated with improved efficacy of adjuvant chemotherapy in patients with early-stage (node-negative) NSCLC. High-risk clinicopathologic features and tumor size should be considered simultaneously when evaluating patients with early-stage NSCLC for adjuvant chemotherapy.</span></span>
oncology