PHILADELPHIA — Mutational signature analysis found mismatch repair (MMR) deficient breast cancer tumors, which could potentially be targeted with PD-1 immune checkpoint inhibitors, according to a study published in Cancer Research, a journal of the American Association for Cancer Research.
“Our research demonstrates the value of using mutation patterns to identify MMR deficiency in unselected breast cancer cases,” said Serena Nik-Zainal, MD, PhD, Cancer Research UK (CRUK) Advanced Clinician Scientist and Honorary Consultant in Clinical Genetics at the University of Cambridge and Wellcome Trust Sanger Institute, Genome Research Limited in Cambridge, United Kingdom. “These findings are important because a subset of patients may be responsive to immunotherapies such as PD-1 checkpoint blockade as has been demonstrated with MMR deficiency in colorectal and other cancers.”
Cancers caused by MMR deficiency involve gene mutations that affect the ability of the cell to repair the mistakes that can happen during the DNA replication process, which the cells must perform each time they divide. MMR-deficient tumors have 10 to 100 times more mutations than tumors with intact MMR pathways.
Nik-Zainal and colleagues used previously published whole genome sequencing data of breast cancer tumors from 640 patients and analyzed patterns of mutagenesis called mutational signatures. The also looked for mutations in multiple genes known to be associated with MMR deficiency: MLH1, MSH2, MSH6, PMS2, PMS1, SETD2, MYH11, EPCAM, TGFBR2, MLH3, and MUTYH, in order to identify the defects responsible for the MMR deficiency.
The researchers discovered 11 of 640 tumors harbored the MMR deficiency in variable amounts. “We found variations of mutations, including single base substitutions and small insertions/deletions,” explained Nik-Zainal. In addition to base substitution mutational signatures specific to MMR deficient tumors, they also found an average of 20,870 small insertions/deletions in the 11 MMR-deficient breast cancer tumors compared to only a fraction, 270 on average, in the non-MMR-deficient tumors. “By looking at multiple signals, we identified a subset of tumors that would not have been detected as MMR-deficient using current clinical criteria for the assessment of breast cancer,” she said.
Due to its high degree of sensitivity and specificity to classify tumors, mutational signature analysis might be a better biomarker for use in future clinical trials, according to Nik-Zainal. “In some recent pharmaceutical trials, the protocols used single biomarkers and startlingly different cut-off points to measure PD-L1 expression for tumor classification, which may explain the different results in progression-free survival outcomes,” she explained. “With mutational signature analysis that investigates multiple mutations, we can see a direct readout of the pathway that has gone wrong. Our method might outperform the use of a single biomarker.”
The main limitation of this study is that it is a technical analysis and the findings need to be tested in clinical trials. MMR deficiency is rare in breast cancer; the investigators found that only 1 to 2 percent of breast cancer tumors had MMR deficiency, Nik-Zainal noted. She added that although this technology could be used to look for MMR deficiency in other cancers, assessing whether using these mutational signatures to tailor cancer treatment helps improve survival is important.
The previously published whole genome sequencing data was sponsored by the following organizations: the European Community’s Seventh Framework Programme; the Breast Cancer Somatic Genetics Study (BASIS); Wellcome Trust; Institut National du Cancer (the French National Cancer Institute); and the Korean Health Technology R&D Project, Ministry of Health & Welfare, Republic of Korea. The mutational signature analysis was sponsored by CRUK and Wellcome Trust.
Nik-Zainal is a consultant for Artios Pharma Ltd. She has filed five patent applications with the UK Intellectual Property Office for various clinical applications of mutational signatures.