Surgery after neoadjuvant immunochemotherapy versus immuno-chemoradiotherapy in patients with cN3 non-small cell lung cancer: A cohort study in two academic centers.

BACKGROUND

Clinically N3 (cN3) stage non-small cell lung cancer (NSCLC) has historically been considered unresectable. While neoadjuvant immunochemotherapy demonstrates unprecedented pathological responses in earlier-stage disease, its role in enabling surgical resection versus definitive immuno-chemoradiotherapy for cN3 NSCLC remains undefined.

METHODS

This multicenter retrospective study (2017-2024) analyzed 196 patients with stage IIIB-IIIC (cT1-4 N3) NSCLC from two academic centers. Eighty-two patients received neoadjuvant immunochemotherapy followed by surgery (surgery group), while 114 underwent definitive immuno-chemoradiotherapy (radiotherapy group). Propensity score matching (PSM) balanced baseline characteristics. Primary endpoint was progression-free survival (PFS); secondary endpoints included overall survival (OS) and pathological complete response (pCR).

RESULTS

After PSM (74 per group), no overall PFS difference was observed (surgery: 32.2 months vs. radiotherapy: 21.3 months, P = 0.094). But surgery was associated with a significant OS benefit (median not reached vs 71.3 months, P = 0.012). Furthermore, only surgical patients achieving pCR (PFS: not reached vs. 21.3 months, P = 0.001; OS: not reached vs. 71.3 months, P = 0.019) or nodal clearance (ypN0) (PFS: not reached vs. 21.3 months, P = 0.010, OS: not reached vs. 71.3 months, P = 0.009) showed superior PFS and OS versus radiotherapy. Recurrence pattern differed between the two groups: Among locoregional failure patterns, all events in the surgery group were regional lymph node recurrences, whereas radiotherapy failures primarily involved primary site progression (59.3%), followed by regional nodal recurrence (39.7%).

CONCLUSION

Surgery after neoadjuvant immunochemotherapy demonstrated survival advantage over immuno-chemoradiotherapy in cN3 NSCLC. Significant survival benefits were confined to surgical patients attaining pCR or ypN0, highlighting the need for biomarker-driven patient selection.