A phase II study identified the first genomic marker – mismatch repair (MMR) deficiency – to predict response to the anti-PD-1 antibody pembrolizumab.

This marker predicted responses across a range of cancers.

Among patients with colorectal cancer (CRC), 62% of those with MMR-deficient tumours experienced tumour shrinkage, while no responses were detected among those without this abnormality (“MMR-proficient”).

The response rate among patients with other MMR-deficient cancers was similar – 60%.

MMR deficiency is found in 15-20% of sporadic (non-inherited) CRCs and in nearly all CRCs associated with Lynch syndrome, which constitute up to 5% of all CRCs.

MMR deficiency is also found in other tumour types including stomach, small bowel, endometrial, prostate, and ovarian cancer.

Testing for MMR-deficiency is widely available and may enable doctors to identify a larger population of patients who might benefit from pembrolizumab and other PD-1 drugs.

“This study is really about bridging immunotherapy and genomics for the benefit of patients, and it has implications for a broad range of cancers,” said lead study author Dung T. Le, MD, an assistant professor of oncology at Johns Hopkins Kimmel Cancer Center in Baltimore.

“Opening the door to this effective new therapy would be a breakthrough for this subset of patients with metastatic colon cancer and other hard-to-treat cancers.”

MMR deficiency leads to an accumulation of genetic mutations in a tumour.

“When you have a tumour that has thousands of mutations, this increases the probability that the immune system can recognise and destroy the tumour. So, we suspected that immune checkpoint inhibitors such as pembrolizumab would work particularly well against MMR- deficient tumours,” added Dr. Le.

In this study, MMR-deficient tumours had an average of 1,782 mutations, compared to 73 mutations in MMR-proficient tumours.

Higher numbers of mutations were linked to better response to pembrolizumab.

The study included three groups of patients: MMR-proficient metastatic CRC (25 patients), MMR-deficient metastatic CRC (13 patients), and other MMR-deficient cancers (10 patients).

All patients had progressive metastatic cancer that had worsened despite prior treatment.

While researchers observed a large difference in response rates between MMR-deficient and -proficient CRCs (62 vs. 0%), the difference in disease control rates (tumour shrinkage or suppressed growth) was even greater – 92% in the MMR-deficient group and only 16% in the MMR-proficient group.

Blood marker changes such as CEA levels indicating response were seen within the first few weeks of starting treatment, and patients tended to feel better almost immediately.

In the group of other MMR-deficient cancers (excluding CRCs), the overall response rate was 60%.

Responses were detected in patients with advanced endometrial cancer and several types of advanced gastrointestinal cancers including ampullary, duodenal, cholangiocarcinoma, and gastric cancers.

Few treatment options exist for such patients.

At last analysis, responses were ongoing for all but one patient, and many responses have lasted for over a year.

Dr. Le indicated that the next step is to reproduce the findings of this prospective study in a larger group of patients to solidify the observation that MMR deficiency is a predictor of response to therapies targeting PD-1.

She noted that the durability of response with little toxicity could eventually lead to testing this approach in initial (upfront) treatment for these patients.

Pembrolizumab is currently only FDA approved to treat patients with advanced melanoma that has not responded to other standard therapies.

Another PD-1 therapy, nivolumab, is approved for the same indication, as well as in advanced squamous lung cancer.

Watch the press conference video or interview for more.

Source: ASCO