Low RNA Expression of RAD51, BRCA1 Associated With Prolonged OS in HRP CRC

Daniel Walden, MD

Low expression of the homologous repair (HR) genes RAD51 and BRCA1 was associated with improved overall survival following treatment with oxaliplatin- and irinotecan-based chemotherapy in patients with homologous recombination–proficient (HRP) colorectal cancer (CRC). Daniel Walden, MD, said this finding suggests that the characterization and identification of this patient population could help inform the best use of chemotherapeutics in clinical practice.

At the 2023 Gastrointestinal Cancers Symposium, Walden and colleagues presented an analysis of survival outcomes based on HR gene expression in patients with CRC after treatment with oxaliplatin or irinotecan. In the retrospective analysis, investigators used next-generation sequencing (NGS) on tissue samples from 11,925 patients to classify them into tiers based on expression of wild-type RAD51, BRCA1, and other HR genes. Patients with pathogenic mutations in core or related homologous repair–deficient (HRD) genes and those with microsatellite instability–high (MSI-H) disease were excluded.

Data showed that patients in the bottom 10% of RAD51 RNA expression (n = 231) experienced a median OS of 80.5 months following treatment with irinotecan compared with 26.1 months for those in the top 10% (n = 255; HR, 0.55; 95% CI, 0.39-0.78; P < .001). Additionally, the median OS was 80.5 months following treatment with irinotecan for patients in the bottom 10% of BRCA1 RNA expression (n = 197) vs 25.9 months for those in the top 10% (n = 232; HR, 0.63; 95% CI, 0.44-0.89; P = .01)

“We need to validate these findings. We need to confirm that this signal is true. Once we do that, it will be imperative to think about implementing these findings for our patients [and how they could influence] the sequencing of treatments for patients who might harbor [low RNA] expression [of these genes] to see if these patients might do better with first-line irinotecan as opposed to oxaliplatin,” said Walden, a Hematology/Oncology fellow at the Mayo Clinic School of Medicine in Scottsdale, Arizona.

In an interview with OncLive®, Walden discussed the goal of analyzing RNA expression of wild-type HR genes in CRC, the importance of evaluating the benefit of chemotherapeutic approaches in patients with HRP CRC based on RNA expression levels, and the next steps for validating and implementing these findings in clinical practice.

OncLive^®: What was the rationale for investigating the RNA expression of HR genes in HRP CRC?

Walden: The impetus for the project was the data that has come out in CRC about HRD and the efficacy of drugs that we typically use to treat that [in CRC]. We have largely extrapolated [this model of HRD] from ovarian cancer and breast cancer. In breast and ovarian cancer, those tumors tend to be more sensitive to platinum-based or irinotecan-based chemotherapeutics.

[In] the context of colon cancer, we do not see the same benefit that we see in ovarian or breast cancer [with these regimens]—[outcomes are] quite dismal. A lot of the trials that look at HRD drugs, such as olaparib [Lynparza] or PARP inhibitors, show almost no response in colon cancer in those patients who harbor HRD genes. We know that physiology in colon cancer is not the same as ovarian cancer.

[This study aimed] to see if patients who do not harbor HRD genes but have low expression of RNA [of HR genes] exhibit a phenotype that might be more sensitive to traditional chemotherapeutics like oxaliplatin or irinotecan.

How were patients selected for this analysis, and what were the key objectives of the research?

This was a joint effort between the Mayo Clinic and Caris Life Science. We had the privilege of utilizing the Caris Life Science database, which included [11,925] patients who had [tissue] samples submitted. We excluded patients who harbored MSI-H [disease] because we know those patients have mutations in these genes, and they are not typical driver mutations. Then we removed patients who harbored HRD genes, so this was a true microsatellite-stable cohort of patients who did not harbor HRD genes and were not mutated in that capacity. From there, we broke patients into tiers [by RNA expression percentiles]…to see if there were differences in outcomes.

Our key objective was to see if patients who [had] low RNA expression of HR genes had a prolonged survival or significantly longer progression-free survival following exposure to oxaliplatin and irinotecan.

[We] also compared the efficacy of [oxaliplatin vs irinotecan] in patients [with low RNA expression of HR genes]. We found some interesting data [indicating] that irinotecan might be the stronger chemotherapeutic in patients who express low quantities of RNA.

Could you expand on some of the other key findings from this analysis presented at the 2023 Gastrointestinal Cancers Symposium?

We showed some real-world overall survival (OS) data that look promising [for] patients who have low [RNA] expression [of HR genes], especially [RAD51] and BRCA1, following exposure to irinotecan. OS was measured from the time of first irinotecan dose [until] death. Patients who harbored an unmutated version of [RAD51 or BRCA1] and [had RNA expression of those genes] in the bottom 10% of patients [had a median OS of 80.5 months for RAD51 and 80.5 months for BRCA1].

This is a tremendously long time in the context of metastatic CRC, where the OS from time of diagnosis is typically around 3 years. When you look at [OS] under the context of first dose of irinotecan, which is commonly given in the second line following oxaliplatin, and we see OS of 4, 5, 6, or 7 years, it is a profound finding.

What are the clinical implications of these findings, and how could they inspire continued research in this space?

Our next step in this project is to confirm these findings in prospective clinical trials, then we can retrospectively apply this construct to a clinical trial that has already been performed. We have some ongoing collaborations with our Japanese colleagues [who are integrating] these data through the [phase 3] PARADIGM trial [NCT02394795] to see if we see any of our signals present in real-world patient data.

Are other efforts being made to investigate RNA expression reported through NGS in CRC?

A [key] part of implementing this project in clinical practice would be the acquisition and quantification of RNA that currently is not done in clinical practice. Typically, we get an NGS report that tells you mutations the in the DNA. A quantified RNA approach is not currently implemented [on a large scale] or those data are not largely available to providers. The product you would get from most companies does not include quantified RNA. This project would hopefully integrate [the inclusion of quantified RNA in NGS results] into clinical practice, so that we can leverage these findings for our patients.

Editor’s note: Dr Walden had no relationships to disclose.

Reference

Walden D, Deshmukh S, Batalini F, et al. Survival of patients with colorectal cancer (CRC) with low expression of homologous recombination proficient (HRP) genes. J Clin Oncol. 2023; 41(suppl 4):225. doi:10.1200/JCO.2023.41.4_suppl.225