Correlation between TP53, KRAS, SMAD4 and other mutations profile and neoadjuvant therapy efficacy and prognosis in locally advanced rectal cancer.

Abstract

186 Background: Locally advanced rectal cancer presents a poor prognosis due to the risk of local recurrence and distant metastasis. Neoadjuvant Chemotherapy (NAC) without radiation therapy for locally advanced rectal cancer holds the potential for both local control and reduced distant recurrence, but identifying cases where its efficacy can be expected remains challenging. To explore predictive factors for the efficacy of NAC and prognosis in locally advanced rectal cancer. Methods: We examined 43 cases of locally advanced rectal cancer (cT3 or deeper, or cN3) that underwent primary tumor resection after NAC between January 2013 and June 2021. Genomic profiles were analyzed by extracting DNA from formalin-fixed paraffin-embedded (FFPE) tumor tissues, including 42 pre-NAC biopsy samples and 39 surgical samples, excluding 4 cases with significant NAC response. Targeted sequencing was used for analysis, and mutations were considered positive if detected in either biopsy or surgical samples, with an allele frequency of 5% or higher and not classified as variants of unknown significance (VUS). Additionally, we examined 334 cases of locally advanced rectal cancer that underwent preoperative treatment using publicly available data in cBioPortal. Results: The median recurrence-free survival (RFS) was 1,415 days (range: 97-2,868) with 15 recurrences, and the median overall survival (OS) was 1,702 days (range: 97-2,868) with 8 deaths (6 from the original disease, 1 from other cancers, and 1 of unknown cause). Among the 4 cases that showed histological response to NAC (AJCC Tumor Regression Grade 0-1), there was 1 recurrence, but no deaths occurred. RFS and OS in TRG0-1 were favorable compared to TRG2-3, but the differences were not statistically significant (3-year RFS: 75% vs. 65%, P=0.557; 3-year OS: 100% vs. 92%, P=0.339). In targeted sequencing analysis, TP53 mutations were found in 32 cases, KRAS mutations in 19 cases, APC mutations in 14 cases, FBXW7 mutations in 5 cases, SMAD4 mutations in 4 cases, PIK3CA mutations in 3 cases, and NRAS mutations in 2 cases. Co-mutations were observed in 13 cases for TP53/KRAS, 8 cases for TP53/APC, and 5 cases for KRAS/APC. One case each of these co-mutations was observed in TRG0-1. FBXW7, SMAD4, PIK3CA, and NRAS mutations were not found in TRG0-1. In the publicly available database, TRG0-1 showed significantly better RFS and OS compared to TRG2-3 (RFS: P=0.002, OS: P=0.089). SMAD4 mutations were significantly more frequent in TRG3 (P=0.02). Co-mutations of TP53/SMAD4, KRAS/SMAD4, and TP53/KRAS/SMAD4 were associated with significantly worse OS compared to cases with no co-mutations (P=0.001, P=0.0002, P=0.003). Conclusions: Co-mutations in TP53, KRAS, and SMAD4 may serve as factors for predicting the effectiveness of preoperative treatment and prognosis in locally advanced rectal cancer.