Key PointsQuestionÌý
What are the oncologic outcomes following lobar or sublobar resection in patients with peripheral small (≤2 cm) non–small cell lung cancer (NSCLC) with visceral pleural invasion (VPI)?
FindingsÌý
In this secondary analysis of a randomized clinical trial of 697 patients with NSCLC, patients who had small (≤2cm) peripheral NSCLC with VPI (pT2) had unexpectedly higher recurrence rates and worse survival compared with patients with tumors without VPI, regardless of the extent of parenchymal resection.
MeaningÌý
The results of this study suggest that unexpectedly high recurrence rates, including distant recurrences, and worse survival of small peripheral NSCLCs with VPI are not mitigated by larger parenchymal resection.
ImportanceÌý
The randomized clinical trial Cancer and Leukemia Group B (CALGB) 140503 showed that for patients with clinically staged T1N0 non–small cell lung cancer (NSCLC; ≤2 cm), sublobar resections were associated with similar oncological outcomes to those after lobar resection. The association of the extent of parenchymal resection with recurrence and survival in patients with tumors pathologically upstaged to T2 based on visceral pleural invasion (VPI) is controversial.
ObjectiveÌý
To determine survival and recurrence rates in patients with small peripheral pT2 NSCLC (≤2 cm) that was treated by either lobar or sublobar resection in CALGB 140503.
Design, Participants, and SettingÌý
CALGB 140503, a randomized multicenter noninferiority trial, included 697 patients with small peripheral NSCLC that was clinically staged as T1N0. Enrollment was from June 2007 through March 2017 at 83 participating institutions, and after a median follow-up of 7 years, the primary outcome of disease-free survival after sublobar resection was noninferior to that after lobar resection.
InterventionÌý
Lobar or sublobar resection.
Main Outcomes and MeasuresÌý
Survival end points were estimated by the Kaplan-Meier estimator. Hazard ratios and 95% CIs were estimated using stratified Cox proportional hazard models.
ResultsÌý
Of 679 participants, 390 (57.4%) were female, and the median (range) age was 67.8 (37.8-89.7) years. Among 697 patients randomized, 566 (81.2%) had pT1 tumors (no VPI) and 113 (16.2%) had pT2 tumors (VPI). Five-year disease-free survival was 65.9% (95% CI, 61.9%-70.2%) in patients with pT1 compared with 53.3% (95% CI, 44.3%-64.1%) in patients with pT2 tumors (stratified log-rank: P = .02). Disease recurrence developed in 27.6% of patients with pT1 (locoregional only: 60 [10.8%]; distant only: 81 [14.6%]) and 41.6% of those with pT2 (locoregional only: 17 [15.0%]; distant only: 27 [23.9%]). Five-year recurrence-free survival was 73.1% (95% CI, 69.2%-77.1%) for pT1 tumors and 58.2% (95% CI, 49.2%-68.8%) for pT2 tumors (stratified log-rank: P = .01). There were no intergroup differences in disease-free or recurrence-free survival based on the extent of parenchymal resection.
Conclusions and RelevanceÌý
The results of this secondary analysis suggest that compared with patients with tumors without VPI, patients who had tumors with VPI had worse disease-free and recurrence-free survival and a higher rate of local and distant disease recurrence. These high rates of recurrence were independent of the extent of parenchymal resection, and these data support the inclusion of these patients in adjuvant therapy trials.
Trial RegistrationÌý
ClinicalTrials.gov Identifier: