Molecular Subtypes and Outcomes in a Multi-Institutional Review of Rectal Cancer Patients

Abstract

Colorectal cancer (CRC) is a heterogeneous malignancy associated with a variety of genetic mutations. More recent data indicates that CRC can be broken down into unique biologically distinct consensus molecular subtypes (CMS) based on different pathological and genetic signatures. The purpose of this study is to evaluate the outcomes of rectal cancer patients in a multi-institutional network based on tumor mutational assessment and CMS group. Patient exome and transcriptome sequencing data and clinical outcomes were collected under the Total Cancer Care Protocol and Avatar® project within the Oncology Research Information Exchange Network (ORIEN). A total of 101 patients with demographic and outcomes information had data for microsatellite instability (MSI), tumor mutation burden (TMB), transcriptome, and whole exome sequencing (WES). Molecular subclasses (CMS1, CMS2, CMS3, CMS4, CMS-Mixed) were assigned based on transcriptional signatures by the R package "CMS Caller." Survival analysis was performed with the R packages "Survival" and "Survminer." A total of 101 rectal cancer patients, with a median age of 56.8, had a median follow up of 3.5 years (range 0.26-23.5). 78 patients were treated with curative intent for clinically localized disease and 35% of these patients developed metastatic disease. The remaining 23 patients had synchronous metastatic disease at presentation. There were 5 (5%) CMS1, 29 (29%) CMS2, 13 (13%) CMS 3, 49 (49%) CMS 4, and 5 (5%) CMS-Mixed patients in our cohort respectively. The cohort included 5 (5%) BRAF, 51 (50%) KRAS, and 63 (62%) TP53 mutated patients and 5 (5%) MSI high patients. Median survival was 18.8, 117.2, 125.7 and 119 months for CMS1, CMS2, CMS3, and CMS4 patients respectively, with insufficient events in CMS-mixed for calculation (p = 0.15). CMS1 patients had a significantly shorter survival compared to the other cohorts (p = 0.02), with 2 of 5 of these patients having received immunotherapy. 40% (2,0) CMS1, 52% (7,8) of CMS2, 15% (1,1) of CMS3, 59% (13,15) of CMS4, and 40% (1,1) of CMS-Mixed presented with or developed metastatic disease respectively. When divided into mutation groups, median survival was 43 versus 119, 119 versus 117, and 126 versus 119 months for BRAF, KRAS, and TP53 mutated and wild type patients respectively (p = 0.18, p = 0.48, p = 0.93). Evaluation of TMB and MSI status did not reveal significant differences in outcomes (p = 0.54, p = 0.7), with median survival of 126 months versus 117 in TMB high versus low patients and unreached versus 119 months in MSI versus MSS patients respectively. Of note, 3 of the 5 MSI patients were also CMS1, with the other two coming from CMS4 and CMS-mixed cohorts. CMS classification and tumor mutation status are associated with differential outcomes for rectal cancer patients, with some groups having a large likelihood of developing metastatic disease. Further work on optimizing and personalizing treatments for these high-risk populations is necessary.