Aberrant subcellular localization of mismatch repair protein MLH1 dysregulates the cell cycle to create new therapeutic opportunities

Read the full article See related articles

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Mismatch repair loss is a commonly used prognostic and predictive clinical diagnostic across several cancer types. Currently, mismatch repair loss is assessed using bulk tumor approaches to assess genome instability (e.g. tumor mutation burden, microsatellite instability) or by immunohistochemistry for total protein level of individual mismatch repair proteins. Here, we identify a novel mechanism by which mismatch repair proteins are dysregulated that cannot be detected by either of these clinical diagnostic approaches. Sub-clonal mutations in MLH1 , a principal mismatch repair gene, detected in tumors from estrogen receptor positive breast cancer patients do not impact abundance of MLH1 protein, nor do they contribute sufficiently to genomic instability to enable detection at the bulk tumor level. However, as we demonstrate here, some of these sub-clonal mutations aberrantly localize MLH1 protein to the cytoplasm, preventing MLH1-mediated activation of the G1 cell cycle checkpoint in response to standard targeted endocrine therapies. The lack of G1 checkpoint activity results in treatment resistance of a sub-clonal persister-like population that remains viable even after long-term exposure to treatment. The dependence on G1 progression, however, renders these sub-clonal populations acutely sensitive to CDK4/6 inhibitors such as palbociclib. Overall, data presented here strongly suggests that aberrant cytoplasmic localization of MLH1 can serve as a predictive biomarker to more accurately predict resistance to endocrine therapy, and sensitivity to combinatorial treatment with CDK4/6 inhibitors. This is particularly important in the adjuvant setting where the risk-benefit ratio for untargeted CDK4/6 inhibitor intervention is narrow. It is possible that these findings have important implications for treatment guidance in other non-breast cancer types as well, where MLH1 mutation is even more frequent.

Article activity feed