BioSothis

For scientists, by scientists

PMS1 loss defines distinct mismatch repair complex deficiencies across dog and human cancers.

2026-07-22, Molecular Cancer Therapeutics (10.1158/1535-7163.MCT-26-0249) (online)
Deborah F Wilsker, Ralph E Parchment, James H Doroshow, Asma Begum, and Alan Brooks (?)
Tumors in pet dogs provide a valuable comparative model for evaluating immune checkpoint inhibitor (ICI) therapies in humans. Whether mismatch repair deficiency (MMRd) predicts response to ICI in dogs, as it does in humans, remains unknown. To enable biomarker-driven clinical trials, development of a validated canine mismatch repair (MMR) assay is required. Mass spectrometry identified PMS1 as the binding partner of MLH1 in dogs, rather than PMS2. A novel monoclonal antibody was developed to detect PMS1 in both canine and human tissues by immunohistochemistry (IHC). A multiplex immunofluorescence assay was established to simultaneously detect MLH1, MSH2, MSH6, and PMS2 in parallel to PMS1. MMR protein expression profiles were characterized across 39 human colorectal cancers, 25 patient-derived xenografts (PDXs), and 43 canine tumor samples. PMS1 is frequently lost in conjunction with MLH1 and PMS2 in human colorectal and endometrial tumors. In canine cancers, lymphomas and mast cell tumors demonstrated a high incidence of MMRd typified by the loss of two or more markers, frequently including MLH1/PMS1. Loss of PMS1 alone occurred in three of 19 canine soft tissue sarcomas. A standardized definition of MMRd in dogs is essential to advance biomarker-driven ICI trials. The multiplex MMR assay described here enables detection of MMRd across species and tumor types. MMRd is a frequent event in common canine cancers, establishing dogs as a relevant model for evaluating ICI therapies. These MMR assays are now suitable for implementation in canine biomarker-guided clinical studies.
This article has not yet been included in any curations.
 
 
0
   

Comments

There are no comments on this article yet.


You need to login or register to comment.
FAQ | Manual | Privacy Policy | Contact