Analysis of recurrence and risk factors in resectable early-stage adenocarcinoma with uncommon <i>EGFR</i> mutations: A comprehensive cohort study in South Korea.
Abstract
8038 Background: This study aimed to identify clinicopathologic factors associated with recurrence in patients with surgically resected lung adenocarcinoma harboring uncommon EGFR mutations. Methods: We conducted a multicenter retrospective cohort study. The study included patients with stage I–III lung adenocarcinoma harboring uncommon EGFR mutations (G719X, S768I, or L861Q) who underwent curative-intent resection between 2010 and 2017 in South Korea. Recurrence-free survival (RFS) and overall survival (OS) were analyzed using Cox proportional hazards models. Results: A total of 48 patients were analyzed (stage I: n=33, stage II: n=6, stage III: n=9). The median age was 65 years; 56.3% were female and 66.7% were never-smokers. Mutation distribution was G719X (72.9%), S768I (16.7%), and L861Q (10.4%). Visceral pleural invasion (VPI) and lymphovascular invasion (LVI) were present in 37.5% and 16.7% of cases, respectively. During follow-up, 26 patients (54.2%) experienced recurrence and 21 (43.8%) died. Multivariable analysis identified ECOG performance status ≥1 (HR 4.55, 95% CI 1.226–16.899; P=0.024), stage III vs. I (HR 5.07, 1.181–21.771; P=0.029), and VPI (HR 3.67, 1.030–13.058; P=0.045) as independent predictors of recurrence. EGFR subtype (non-G719X vs. G719X) was not significantly associated with recurrence (P=0.428). Conclusions: Patients with surgically resected lung adenocarcinoma harboring uncommon EGFR mutations face a substantial risk of postoperative recurrence. Conventional clinicopathologic factors—particularly ECOG status, advanced stage, and VPI—outperform EGFR subtype as prognostic indicators. These findings support the need for risk-adapted surveillance and prospective evaluation of adjuvant strategies in these high-risk subsets. Univariate and multivariate Cox regression analysis for the risk factors associated with recurrence. Variables Recurrence rate* Univariate analysis,HR (95% CI) P -value Multivariate analysis,HR (95% CI) P -value ECOG 0 50.0 (19/38) - - 1≤ 70.0 (7/10) 1.64 (0.678-3.955) 0.273 4.55 (1.226-16.899) 0.024 EGFR mutation G719X 54.3 (19/35) - - Non-G719X 53.8 (7/13) 0.91 (0.377–2.210) 0.840 0.58 (0.155–2.207) 0.428 Tumor size † < 2.0 cm 38.5 (5/13) - - ≥ 2.0 cm 60.0 (21/35) 1.59 (0.599–4.214) 0.353 1.75 (0.434–7.068) 0.431 Clinical stage I 48.5 (16/33) - - II 66.7 (4/6) 1.65 (0.549–4.950) 0.373 1.33 (0.368–4.801) 0.663 III 66.7 (6/9) 2.11 (0.820-5.436) 0.121 5.07 (1.181-21.771) 0.029 Lymphovascular invasion Absent 52.5(21/40) - - Present 62.5(5/8) 1.11 (0.420–2.958) 0.827 0.03 (0.001–0.841) 0.040 Visceral pleural invasion Absent 46.7 (14/30) - - Present 66.7 (12/18) 1.97 (0.893–4.334) 0.093 3.67 (1.030–13.058) 0.045 *Presented as number of recurrences / number of patients (%). † ROC curve and Youden’s index analyses were used to determine the cut-off value of primary tumor size.
Article Details
Journal Info
Journal of Clinical Oncology
Lippincott Williams & Wilkins
Authors (2)
Insu Kim
Jung Seop Eom
Lung Cancer Center, Pusan National University Hospital, Busan, South Korea