Skip to main content
Premium Trial:

Request an Annual Quote

CAP Weighs in on dMMR, MSI Assays for Immunotherapy Patient Selection


NEW YORK – When it comes to measuring DNA mismatch repair deficiency or microsatellite instability to guide immunotherapy decisions for patients, the College of American Pathologists (CAP) now recommends healthcare providers carefully choose which assay to perform and, in certain cancers, prioritize immunohistochemistry and PCR approaches over next-generation sequencing.

The guidelines, published last week in the Archives of Pathology and Laboratory Medicine, are meant to help oncologists and pathologists accurately determine their patients' eligibility for immune checkpoint inhibitors, including Merck's Keytruda (pembrolizumab), which in 2017 garnered tissue-agnostic US Food and Drug Administration approval for advanced, refractory solid tumors that are MSI-high or mismatch repair deficient (dMMR).

While the agency often approves biomarker-defined treatments simultaneously with a companion diagnostic that can identify the intended use population, in granting tumor-agnostic approval to Keytruda, the agency left it up to oncologists to decide how they wanted to identify patients with MSI-high or dMMR tumors. This has caused substantial confusion and inconsistency across the field over the last five years, according to Russell Broaddus, chair of pathology and laboratory medicine at the University of North Carolina at Chapel Hill.

"It's been all over the place," Broaddus said of the diagnostic landscape for determining MSI and dMMR for immune checkpoint inhibitor eligibility. "People were equating all of the different assays, almost saying, 'Oh, it doesn't matter what assay you use; it's all microsatellite instability, and any assay will detect it.'"

According to Broaddus, lead author of the new CAP guidelines, the idea that assays are interchangeable is "pretty close to being true" in colorectal cancer and certain adenocarcinomas of the gastrointestinal tract. "But when you look at cancer types outside of the GI tract, that statement starts falling apart pretty quickly," he said. In particular, he hammered home CAP's position that, beyond colorectal cancer, the evidence just isn't there yet to support using broad-based NGS tests to detect dMMR.

"It may be dissatisfying to many people, but one of the big messages to come out of this guideline is that these assays are not all interchangeable," Broaddus said.

IHC vs. PCR vs. NGS: Which to use?

The new CAP guidelines are the culmination of a yearslong review process in which a multidisciplinary panel of pathologists, clinicians, oncologists, genetic counselors, guidelines methodologists, and patient advocates met eight times and defined the scope of the guideline, drafted the recommendations, responded to feedback and comments, and evaluated the evidence from a systematic review of the published literature.

A total of 103 articles were included in the data extraction and qualitative analysis, and hundreds of other studies were considered as background references for the panelists' discussions. The Association of Molecular Pathologists (AMP), the American Society of Clinical Oncology (ASCO), and the patient advocacy group Fight Colorectal Cancer collaborated with CAP on the guidelines.

In evaluating the available assays, the panel considered IHC for the mismatch repair proteins MLH1, MSH2, MSH6, and PMS2 as well as PCR-based MSI testing and MSI testing by NGS for the detection of DNA mismatch repair defects.

Ultimately, the group came up with six evidence-based guideline statements surrounding assay choice and three good practice statements to help the field evaluate patients' dMMR and MSI status. The good practice statements are key suggestions based on studies and data outside the scope of the formal systematic evidence review.

The first guideline strongly recommends that, for patients with colorectal cancer, pathologists should use IHC to determine MMR and/or PCR to determine MSI. Though these methods are preferred, an NGS assay that has been validated for dMMR could also be used to determine MSI status for checkpoint inhibitor eligibility in this cancer type. In this case, the expert panelists wrote that the NGS assay "must be validated against MMR-IHC or MSI by PCR and must show equivalency."

The second guideline focuses on patients with gastroesophageal or small bowel cancer. In this case, the experts said MMR-IHC or MSI by PCR should be used over MSI by NGS. Of note, esophageal squamous cell carcinoma was not included in this recommendation.

In the third guideline, the experts focused on endometrial cancer patients, strongly recommending the use of MMR-IHC as opposed to MSI by PCR or NGS.

The fourth guideline deals with cancer types other than colorectal, gastroesophageal, small bowel, and endometrial cancers. For these other cancer types, the guideline acknowledges that the optimal approach has not been established and that, in the process of considering the best test to use, oncologists and pathologists should note that "assays must be adequately validated for the specific cancer type being tested with careful consideration of performance characteristics of MMR-IHC and MSI by NGS or PCR for the detection of DNA mismatch repair defects."

This recommendation, Broaddus acknowledged, is more of a spotlight on a large knowledge gap than it is a specific recommendation. Broad NGS panels might be appealing for oncologists trying to assess as much molecular information from a tumor as possible at one time, but that convenience shouldn't outweigh the fact that, for a lot of cancer types, there isn't enough evidence to show that NGS accurately determines dMMR or MSI status.

"You need to recognize that if you do NGS, you may not capture high levels of MSI for other cancer types, especially outside the GI tract," he said. "But if you're capturing other cancer types, you may miss patients who would otherwise be eligible for immunotherapy."

The fifth guideline says that tumor mutational burden should not be considered a surrogate for detecting dMMR status. The FDA in 2020 also approved a tissue-agnostic indication for Keytruda in advanced, refractory solid tumor patients with TMB of at least 10 mutations per megabase. "If a tumor is identified as TMB-high, pathologists may perform IHC and/or MSI by PCR to determine if high TMB is secondary to mismatch repair deficiency," the panelists wrote. 

"What people will do is say, 'Oh, TMB is the same as high levels of microsatellite instability,'" Broaddus said. "And for colorectal cancer, there's a lot of overlap. But for other cancer types, the overlap is slim."

Although patients with high TMB may now be eligible for Keytruda regardless of their dMMR or MSI status, the point that Broaddus and colleagues emphasized is that high TMB, while it may be a sign of dMMR or MSI-high status in some cases, is not the same biomarker as the other two.

Finally, in the sixth guideline, the experts noted that, if testing a patient for immune checkpoint inhibitor eligibility reveals a mismatch repair deficiency consistent with Lynch syndrome, it is strongly recommended that pathologists communicate this finding to the patient's treating physician since this is an inherited syndrome that increases the risk of other cancers.

Dealing with uncertainty

The three good practice statements in the published guideline document center on what oncologists and pathologists should do in the event of interpretation or discordance issues. This is an important point particularly with regard to IHC — the test that the experts recommended first in a number of the guidelines — since the method can, in some cases, produce varied results according to pathologists' individual interpretations.

"The good thing about IHC is that it's more universal and more clinical labs run it," Broaddus said. "But the bad thing is that there can be some subjectivity to it, and it does require some training to interpret." In the long term, Broaddus said that pathology training programs should include more formal training in how to interpret IHC assays.

In the meantime, the good practice statements included in the new CAP guidelines provide some guidance in terms of handling discrepancies. The first says that, in the event of discordant results, pathologists should interpret any evidence of either dMMR by IHC or MSI by NGS or PCR as a positive result, thus making patients eligible for the immunotherapy. "Discordant results should be reviewed to ensure that the discordance is not due to an interpretive error," the statement reads.

The second good practice statement says that, should any of the methods considered produce an indeterminate result, pathologists should perform an alternative technique or repeat that same technique on a different tumor block. "Laboratories should have a robust peer review process for indeterminate cases," it says.

Finally, the third good practice statement recommends that, in the event of a clonal loss by MMR-IHC, "pathologists should perform MSI by PCR specifically in a dissected area of tumor that has IHC loss of MMR protein."

'A living document'

While the CAP guidelines as they exist today are admittedly vague for those cancer types outside of the colorectal and GI tract tumors specifically addressed, Broaddus emphasized that the committee hopes to change that with future updates. In fact, built into the guidelines is a plan to review and revise the guideline every four years, or earlier if new high-quality evidence emerges.

"It’s kind of meant to be like the Constitution … a living, breathing document," Broaddus said. "And there are areas that I'm hoping we can update five years from now with more literature."

In the meantime, for the cancers beyond those addressed — for which the current guideline says there isn't enough evidence to recommend a specific type of assay for dMMR or MSI status at the moment — Broaddus said, "it's not entirely unreasonable to test a patient sequentially [with the different options] if you have the patience to do that."

And what will it take to address the dearth of evidence supporting a clear assay choice for these other cancer types? Better tumor-specific analyses in basket trials, Broaddus said. In particular, studies used to support NGS tests as a method of MSI or dMMR assessment often don't report results by individual cancer type, nor do they report how patients respond to immunotherapy according to the test method that was used to determine their eligibility.

"What assay did you run and what was the end result for the patient? That, to me, is the real data that's missing," he said. "A theme of this guideline is, more than anything, the call to action to start working to provide this evidence."

Unanswered questions

Broaddus was also careful to point out that liquid biopsy approaches were not among the assays considered in this CAP guideline. "To me, the liquid biopsy approaches are still on the experimental side, especially for this application," he said, noting that the approach does have clear value for other precision oncology applications, such as looking for a specific mutation in shed DNA for a patient with a tumor that's hard to reach with a biopsy. But for determining patients' immunotherapy eligibility with dMMR or MSI status, blood-based approaches are still experimental in CAP's view.

Another question that fell outside of the scope of the new CAP guidelines was whether specific assays — not just specific types of assays — were preferred over others.

"We did not get that much into the weeds," he said, adding that many clinical laboratories use very effective lab-developed tests to assess dMMR and MSI status, leaving it unclear whether FDA-approved commercial assays have a leg up for this particular use case.

Broaddus and colleagues hope that the publication will help shape the way the field approaches evidence generation to fill the gaps in knowledge that the CAP guidelines body has identified. "What I would like to see is some really organized efforts to help fill these gaps," Broaddus said. "Because these [checkpoint inhibitors] are not just incremental survival improvements. These are long-term survivals and potential cures [and] when the treatment is that good, it's really incumbent on us to find the best biomarker for it."