Who benefit from adjuvant chemotherapy in stage I lung adenocarcinoma? A multi-dimensional model for candidate selection
Meng-Qi Jiang,Li-Qiang Qian,Yu-Jia Shen,Yuan-Yuan Fu,Wen Feng,Zheng-Ping Ding,Yu-Chen Han,Xiao-Long Fu,Meng-qi Jiang,Li-qiang Qian,Yu-jia Shen,Yuan-yuan Fu,Zheng-ping Ding,Yu-chen Han,Xiao-long Fu
DOI: https://doi.org/10.1016/j.neo.2024.100979
IF: 6.218
2024-02-22
Neoplasia
Abstract:Background Despite promising overall survival of stage I lung adenocarcinoma (LUAD) patients, 10-25 % of them still went through recurrence after surgery. [1] While it is still disputable whether adjuvant chemotherapy is necessary for stage I patients. [2] IASLC grading system for non-mucinous LUAD shows that minor high-grade patterns are significant indicator of poor prognosis. [3] Other risk factors, such as, pleura invasion, lympho-vascular invasion, STAS, etc. are also related to poor prognosis. [4-6] There still lack evidence whether IASLC grade itself or together with other risk factors can guide the use of adjuvant therapy in stage I patients. In this article, we tried to establish a multi-variable recurrence prediction model for stage I LUAD patients that is able to identify candidates of adjuvant chemotherapy. Methods We retrospectively collected patients who underwent lung surgery from 2018.8.1 to 2018.12.31 at our institution and diagnosed with lung adenocarcinoma pT1-2aN0M0 (stage I). Clinical data, manifestation on CT scan, pathologic features, driver gene mutations and follow-up information were collected. Cox proportional hazards regression analyses were performed utilizing the non-adjuvant cohort to predict disease free survival (DFS) and a nomogram was constructed and applied to the total cohort. Kaplan-Meier method was used to compare DFS between groups. Statistical analysis was conducted by R version 3.6.3. Findings A total of 913 stage I LUAD patients were included in this study. Median follow-up time is 48.1 months.4-year and 5-year DFS are 92.9 % and 89.6 % for the total cohort. 65 patient experienced recurrence or death. 4-year DFS are 97.0 %,94.6 % and 76.2 %, and 5-year DFS are 95.5 %, 90.0 % and 74.1 % in IASLC Grade1, 2 and 3, respectively(p < 0.0001). High-risk patients defined by single risk factors, such as, IASLC grade 3, pleura invasion, STAS, less LN resected could not benefit from adjuvant therapy. A LASSO-COX regression model was built and patients are divided into high-risk and low-risk groups. In the high-risk group, patients underwent adjuvant chemotherapy have longer DFS than those who did not (p = 0.024), while in the low-risk group, patients underwent adjuvant chemotherapy have inferior DFS than those who did not ( p < 0.001). Interpretation IASLC grading is a significant indicator of DFS, however it could not guide adjuvant therapy in our stage I LUAD cohort. Growth patterns and T indicators together with other risk factors could identify high-risk patients that are potential candidate of adjuvant therapy, including some stage IA LUAD patients.
oncology