Background: This study aimed to explore adjuvant chemotherapy (ACT) candidates based on a recurrence risk-scoring model in completely lobectomized stage I patients with lung adenocarcinoma (LAD).
Methods: A retrospective study was performed on 4606 patients (non-ACT group: n = 3514; ACT group: n = 1092) who underwent complete lobectomy for LAD at Shanghai Chest Hospital from 2008 to 2014. The nomogram predicting recurrence-free survival (RFS) was developed in the non-ACT group using Cox proportional hazards regression. The nomogram-based risk score was calculated in the entire cohort. Differences of RFS between the non-ACT and ACT groups were compared as stratified by the risk score. The score cut-off points were determined using the X-tile software.
Results: Six independent predictors, including age, sex, tumor size, pathological subtype, visceral pleural invasion (VPI), and lymphovascular invasion (LVI) were associated with RFS. The nomogram more accurately predicted RFS than the 8th TNM staging {C-index: 0.784 [95% confidence interval (CI) 0.756-0.812] vs. 0.719 (95% CI 0.689-0.749), p = 0.0017}. A significant RFS difference was observed among the low-, intermediate- and high-risk groups (p < 0.0001), as divided by the optimal cut-points of risk score (203 and 244). ACT did not improve RFS for patients at intermediate-risk, or was even detrimental for low-risk patients; however, improved RFS was observed in ACT receivers at high-risk (p = 0.0416). ACT candidates with a risk score ≥ 245 constituted 2.6% of stage I patients.
Conclusions: The nomogram provided an individual prediction of RFS for stage I LAD following lobectomy. High-risk patients (score ≥ 245) may benefit from postoperative ACT.