Abstract

The molecular heterogeneity of KRAS is well established, with a pool of variants comprising >75% of all known mutations; this pool includes mutations in classic codons 12, 13, and 61, as well as 146 and 117. In addition, there are rare variants that are more frequently encountered clinically due to the advances in next-generation sequencing and more widespread implementation of All-RAS sequencing over the past five years. We have previously identified a missense variant of KRAS, A59T, in a patient with CRC that was associated with a response to an epidermal growth factor inhibitor when added to chemotherapy, supporting the hypothesis that distinct biochemical impacts of different KRAS mutations may produce varied responses to targeted therapy. In this study, we explored a large genomic database comprising 17,909 cases of CRC to determine the prevalence of the A59T mutation and characterized the concurrent genomic alterations associated with this variant in more detail, particularly in relation to the expanding set of potential predictive immuno-oncologic biomarkers. We identified 14 cases of A59 mutations in this dataset (0.08% prevalence). We evaluated the prevalence of high tumor mutation burden (TMB), positive PD-L1 expression, and microsatellite instability-high/mismatch repair-deficiency (MSI-H/dMMR) using both next generation sequencing (NGS) and immunohistochemistry (IHC). The genomic features of pertinent signaling pathways were also described, including RAS pathway, chromatin remodeling, DDR, hedgehog signaling, PI3K, receptor tyrosine kinases, signal transduction, TGF-beta, TP53, and WNT. We uncovered a high level of association of predictive markers of responsiveness to checkpoint inhibition and potentially other forms of immunotherapy, with nearly half of all cases harboring microsatellite instability as assessed using NGS. A59T was also detected in 11 additional cancer types, most prominently in cases of gynecologic or other gastrointestinal sites of origin. This study provides supportive evidence that A59T, and possibly other similarly rare KRAS variants, co-occur with predictive biomarkers of response to immunotherapy.

Highlights

  • The implementation of generation sequencing that routinely assesses the full coding sequence of K, N, and H-RAS genes has demonstrated a larger, more diverse spectrum of cancer-associated RAS mutations than initially characterized in colon and other solid cancers with more focused molecular approaches

  • Of 17,909 cases of colorectal cancers (CRC) reviewed, we identified 14 cases of A59T mutations (0.08%), a prevalence that is consistent with other sources [7], including the American Association for Cancer Research (AACR) GENIE database

  • It has been established that oncogenic KRAS mutations stimulate a tumor-promoting inflammatory microenvironment that differentiates these tumors from their RAS wild-type counterparts [8]

Read more

Summary

Introduction

The implementation of generation sequencing that routinely assesses the full coding sequence of K-, N-, and H-RAS genes has demonstrated a larger, more diverse spectrum of cancer-associated RAS mutations than initially characterized in colon and other solid cancers with more focused molecular approaches. The observation of significant clinical response despite the presence of a KRAS mutation with oncogenic biochemical features [2] was counterintuitive and led us to postulate that the mechanistic impact of this missense variant may be distinct from other common KRAS mutations and permit tumor response to EGFR inhibition. This hypothesis was countered by an accompanying editorial to our report by Loree and Kopetz [4], who posited that the intratumoral heterogeneity of KRAS, in which the patient’s tumors harbored both wild-type KRAS-expressing cells and cells expressing the detected A59T mutation, may have led to the response. We sought to further characterize this unusual variant by exploring a large dataset of CRC with available allRAS sequencing as well as extensive genomic profiling, including MSI-H status and tumor mutational burden (TMB)

Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call