Preoperative CT-based Deep Learning Model for Predicting Disease-Free                     Survival in Patients with Lung Adenocarcinomas

8550 Background: Lung cancer is one of the leading causes of cancerous deaths globally. High mutation burden is a special character in lung adenocarcinoma patients. Mutation burden is usually based on the number of non-synonymous mutations implying the instability of genome. We hypothesize genome-wide mutation burden indicates mutation degree and is correlated with prognostic in lung adenocarcinoma. Methods: Whole-exome sequencing was performed on 98 Chinese lung adenocarcinoma patients with tumor and normal tissue to a mean depth of 49.6ⅹ. The total number of non-synonymous somatic mutations was calculated from the sequencing data of each patient. Patients were divided into high mutation burden and low mutation burden groups in accordance with the mean mutation burden and Kaplan-Meier analysis was performed for survival analysis between these two groups. The association between mutation burden and age or smoking status was analyzed by Wilcoxon rank-sum test. Results: Among these 98 patients, the values of mutation burden varied from 5 to 1121 with mean value 161.8, 36 (36.7%) patients with smoking history and 34 (34.7%) patients were older than 65 years; the numbers of patients in I, II, III stage were 19 (19.4%), 16 (16.3%) and 63 (64.3%) respectively. 32 patients were classified into high mutation burden group, the other 66 patients classified into low mutation burden group. Survival analysis showed a significantly longer disease free survival (DFS) in low mutation burden group (p-value = 0.0133).Mutation burden was significantly associated with age ( < 65 vs ≥65, p-value = 0.0208) and smoking status (p-value = 8.67ⅹ10-4). Conclusions: The association between mutation burden and age or smoking status suggested the high risk for mutation burden accumulation. The significant difference of DFS between high mutation burden and low mutation burden groups reveals the potential of mutation burden as one of the prognostic factors in patients with lung adenocarcinomas.

Download Full-text

Development of a deep learning model for classifying thymoma as Masaoka-Koga stage I or II via preoperative CT images

Annals of Translational Medicine ◽

10.21037/atm.2020.02.183 ◽

2020 ◽

Vol 8 (6) ◽

pp. 287-287

Author(s):

Lei Yang ◽

Wenjia Cai ◽

Xiaoyu Yang ◽

Haoshuai Zhu ◽

Zhenguo Liu ◽

...

Keyword(s):

Deep Learning ◽

Learning Model ◽

Ct Images ◽

Stage I ◽

Preoperative Ct ◽

Deep Learning Model

Download Full-text

Impact on disease-free survival of adjuvant erlotinib or gefitinib in patients with resected lung adenocarcinomas that harbor epidermal growth factor receptor (EGFR) mutations

Journal of Clinical Oncology ◽

10.1200/jco.2009.27.15_suppl.7523 ◽

2009 ◽

Vol 27 (15_suppl) ◽

pp. 7523-7523

Author(s):

Y. Y. Janjigian ◽

B. J. Park ◽

M. G. Kris ◽

V. A. Miller ◽

G. J. Riely ◽

...

Keyword(s):

Lung Adenocarcinoma ◽

Egfr Mutation ◽

Disease Free Survival ◽

Stage I ◽

Egfr Mutations ◽

Free Survival ◽

Resected Stage ◽

Lung Adenocarcinomas ◽

Disease Free ◽

Exon 19

7523 Background: Patients with stage IV adenocarcinoma whose tumors harbor EGFR mutations have high rates of response (∼ 75%) and prolonged progression free survival after EGFR tyrosine kinase inhibitor (TKI) treatment. Adjuvant cisplatin-based chemotherapy improves disease free survival (DFS) and overall survival (OS) in patients with resected stages IB-IIIA NSCLC. To see if adjuvant treatment with EGFR TKI (gefitinib or erlotinib) improves DFS in patients with EGFR mutation NSCLC, we conducted a retrospective review of patients with resected lung adenocarcinoma harboring EGFR mutations, some of whom received EGFR TKIs postoperatively. Methods: With Institutional Review Board approval, clinical information was obtained on all patients with stage I-III lung adenocarcinoma harboring EGFR exon 19 or 21 mutations that underwent resection at MSKCC between May 2002 and August 2008. Age, gender, type of surgery, histology, EGFR mutation status (exon 19 deletions and exon 21 L858R), stage, perioperative therapy and survival were recorded. Kaplan-Meier analysis and Cox regression analysis were performed. Results: We studied 150 patients (112 women, 38 men) with completely resected stage I-III lung adenocarcinoma whose resection specimens contained EGFR activating mutations in exon 19 or 21. Median age was 69. Forty two patients (28%) received cytotoxic chemotherapy. Forty eight (32%) received either erlotinib (n=26) or gefitinib (n=22) postoperatively. The median time on TKI was 16 months. The median DFS was 43 months in the group that received a TKI vs. 31 months for those that did not. After controlling for stage, individuals who received adjuvant gefitinib or erlotinib had a better DFS (HR=0.38, 95%CI: 0.16–0.90) than the non-TKI group (p=0.03). The median overall survival has not been reached. Conclusions: These data indicate that the adjuvant use of either gefitinib or erlotinib improves DFS in patients with completely resected stage I -III lung adenocarcinomas with mutations in EGFR exons 19 and 21. These data justify a randomized trial in similar patients. [Table: see text]

Download Full-text