Association of Survival With Adjuvant Chemotherapy Among Patients With Early-Stage Non-Small Cell Lung Cancer With vs Without High-Risk Clinicopathologic Features

Abstract

Importance: 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.

Objective: 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.

Design, setting, and participants: 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.

Exposures: 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.

Main outcomes and measures: The association of high-risk pathologic features with survival after adjuvant chemotherapy vs observation was evaluated using Cox proportional hazards regression models.

Results: 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; P = .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; P = .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; P = .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; P = .004).

Conclusions and relevance: 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.

Figure 1.. Diagram of the Primary Study Cohort Selection Steps

NSCLC indicates non–small cell lung cancer.

Figure 2.. Association of Survival With Adjuvant Chemotherapy Based on the Presence of High-Risk Pathologic Features or Sublobar Resection in Patients Stratified by the 4 Tumor Size Categories or Stages

Overall refers to all comers in the size category irrespective of high-risk features; high-grade indicates poor differentiation or undifferentiated. HR indicates hazard ratio; LVI, lymphovascular invasion; and VPI, visceral pleural invasion. ^aFor sublobar surgery, a larger cohort from 2004 to 2015 was used to increase power.

Figure 3.. Association of Survival With Adjuvant Chemotherapy in Patients With Tumors 4 cm or Smaller Who Underwent Sublobar Resection

HR indicates hazard ratio.