Differentiating True Lynch Syndrome in Patients With dMMR Rectal Cancer

Yi-Qian Nancy You, MD, MHSc

The proper management of subgroups of patients with DNA mismatch repair deficiency (dMMR) rectal cancer depends on a thorough molecular and clinical understanding of the causes and risks of these patients, an endeavor that is still underway in the treatment of patients with dMMR rectal cancer and still has further to go.

A genetic understanding of dMMR in rectal cancer could lead the way to precision medicine for these patients, said Yi-Qian Nancy You, MD, MHSc, one of the study authors on a report published in the Journal of Clinical Oncology addressing gene alterations and treatment for patients with dMMR rectal cancer.¹

For colorectal cancer, identification of dMMR triggers the need for genetic testing for the detection of a potential heritable germline de ciency in the DNA mismatch repair system. The implications of dMMR status are still unknown in rectal cancer, however, meaning that genetic testing has rarely occurred.

When genetic testing is performed on dMMR rectal cancer patients, a separation occurs between patients who present with mutations in their tumors in the mismatch repair system as well as the same mutations in the germline, and those who do not show the same mutations in the germline as can be found in their tumors.

Historically, this group has been clinically considered to have Lynch syndrome because there’s been no evidence to the contrary, meaning that they, and their family members, would be considered to have a high lifetime risk of developing cancer and would undergo extensive screening.

“At some point when we gure out what’s going on with these people—what led to their tumor mutation if it’s not in the germline...then maybe we won’t need to keep them in this high-risk screening program and test them for all of these other Lynch-associated tumors,” said You, associate professor of Surgical Oncology, Division of Surgery, University of Texas MD Ander- son Cancer Center, in an interview with Targeted Therapies in Oncology.

However, this group, often termed “Lynch-like syndrome” by many reports, stands apart from those with Lynch syndrome. These patients have nonsporadic high microsatellite instability tumors, abnormal immunohistochemistry of the MMR proteins in the tumor, and no evidence of MLH1 hypermethylation or BRAF mutation. Also, genetic testing in these patients show various unclassied variants that are likely nonpathogenic, or results are inconclusive, whereas patients with true Lynch syndrome have identifiable pathogenic mutations.²It is possible that these patients may not require as extensive screening as those with Lynch syndrome would.

One population-based study examined the risk of cancer in families of patients with &ldquo;Lynch-like syndrome&rdquo; in comparison with families of Lynch syndrome patients.³The study found that the familial risk is reduced with Lynch-like syndrome compared with Lynch syndrome. The standardized incidence ratio (SIR) for colorectal cancer (CRC) for Lynch-like syndrome was 2.12 (95% CI, 1.16-3.56) compared with 6.04 for classic Lynch syndrome (95% CI, 3.58-9.54; P <.001). However, the SIR for Lynch-like syndrome was higher than the SIR for sporadic CRC of 0.48 (95% CI, 0.27-0.79; P <.001), confirming the need for surveillance of these families, although not to the extent of Lynch syndrome families.

The community does not yet have a consensus on not only how to manage these patients, but also regarding what to call this subgroup, commented You. She has proposed that these patients be termed &ldquo;mutation-negative Lynch syndrome.&rdquo; This phrase would differentiate these patients from the classic Lynch syndrome and nod to the fact that genetic testing shows unknown MMR genes in these patients.

The basis for &ldquo;mutation-negative Lynch syndrome&rdquo; needs to be understood in order to differentiate cancer treatment and surveillance for these patients and their family members. You and colleagues&rsquo; study of dMMR in rectal cancer aimed to begin to identify the molecular basis of this patient population compared with Lynch syndrome patients, and to begin to identify which treatment would most benefit these patients.

In the study a selection of 62 patients over the age of 18 years old with dMMR rectal adenocarcinoma, who were diagnosed between 1992 and 2012, were tested for tumor MMR status, microsatellite instability, and immunohistochemistry of the MLH1, MSH2, MSH6, and PMS2 proteins.¹

Patients with detected deficiency in MLH1, MSH2, MSH6, or PMS2 underwent genetic counseling for con rmatory germline mutation testing. Forty-two patients had true Lynch syndrome and 15 patients had mutation-negative Lynch syndrome,⁷of which had presented with a variant of unknown signi - cance (VUS), all of which were missense mutations, and 8 had uninformative negative results. The remaining 5 patients had declined genetic testing.

Ninety-eight percent of the patients met the revised Bethesda criteria while only 15% met Amsterdam criteria I or II. The median PREMM 1,2,6 score for all of the patients was 25.6 (range, 12.1-51.0).

Following testing, it was found that a majority of the patients had a mutation in either MSH2 (53%) or MSH6 (23%). Expectations had favored the MLH1 gene for the majority of defects in MMR, study authors noted in a statement, as this is commonly found in patients with dMMR colon cancer. However, only 16% of patients in the study had a mutated MLH1 gene and 8% had a mutation in PMS2.

Thirty of 40 patients (75%) with stage II or III rectal cancer were treated with neoadjuvant uoropyrimidine-based chemotherapy and long-course pelvic radiation. All but one of these patients underwent surgical resection. Eight patients (27.6%) experienced a pathological complete response and 16 patients (55.5%) were downstaged following treatment.

For the entire patient population, the 5-year overall survival (OS) rate was 87% and the rectal cancer-specific survival (RC-SS) rate was 90.6% overall. For patients with stage I and II disease, the RC-SS rate was 100%, 85.1% for stage III, and 60% for stage IV (P = .03). Recurrence occurred in 18.6% (11 of 59) of the patients who had curative-intent resection.&nbsp;Eleven of the patients died, 6 due to rectal cancer and 5&nbsp;due to extracolorectal cancers after a median of 17.3 years. Although all of the patients with VUS mutation-negative Lynch syndrome were still alive as of last follow-up.

Fourteen patients (22.6%) developed a total of 22 extra-colorectal cancers. Study authors noted that the MSH2 and MSH6 MMR genes have been associated with higher incidences of extracolorectal cancers.

&ldquo;Our data basically established what the results are with current therapy. That provides a baseline for developing trials that include new drugs because you have to have something to compare it against to see whether the new drugs are going to add any benefit,&rdquo; You said.

Future trials may examine the effect of immunotherapy and targeted therapy agents on patients with dMMR rectal cancer. You noted that clinical trials are already underway examining PD-1 blockade in tumors with DNA dMMR, which have already shown substantial results.⁴It is suggested that dMMR is predictive of response to immune checkpoint blockade treatment.

The trial also provides implications for the underlying genetics of dMMR rectal cancers. &ldquo;Identification of a dMMR rectal cancer should trigger germline testing, followed by lifelong surveillance for both colorectal and extracolorectal malignancies,&rdquo; the study authors emphasized in their abstract.¹

In dMMR colon cancers, genetic testing is more common right off the bat to identify patients with Lynch syndrome. However, dMMR is different in rectal cancer than in colon cancer, You pointed out.

&ldquo;We think if you find rectal cancer, the likelihood that someone is actually Lynch is much, much higher than if it&rsquo;s a colon cancer because there are other ways of causing it in colon cancer. The other known mechanisms that cause tumor dMMR we can&rsquo;t find in rectal cancer.&rdquo;

Subsets of colon cancers that develop DNA dMMR are caused by other mechanisms, such as hypermethylation of a gene, typically MLH1. On the other hand, there is no methylation or other known mechanisms causing the deficiency in rectal cancer patients. The causes are still unknown for rectal cancer.

In the meantime, there is a better indication of the long-term prognosis of these patients. The long-term multiorgan follow-up of patients with dMMR rectal cancers is very important, You urged, especially since many will develop extracolonic cancers as their germline mutations put them at a lifelong risk for developing other cancers.

This follow-up extends beyond the patient to members of their family as well. Once a patient has been diagnosed with a genetic mutation it can affect the entire family as they will all need to be continually screened for associated cancers.

&ldquo;Precision medicine is good for this one patient, of course, but it&rsquo;s so much more powerful if you think of the implications for all of the generations,&rdquo; You said.

References:

1. de Rosa N, Rodriguez-Bigas MA, Chang GJ, et al. DNA Mismatch Repair De ciency in Rectal Cancer: Benchmarking Its Impact on Prognosis, Neoadjuvant Response Prediction, and Clinical Cancer Genetics. [Published online July 28, 2016] J Clin Oncol. doi: 10.1200/ JCO.2016.66.6826.
2. You YN, Vilar E: Classifying MMR variants: Time for revised nomenclature in Lynch syndrome. Clin Cancer Res. 2013;19(9):2280-2282.
3. Rodrígues-Soler M, Pérez-Carbonell L, Guarinos C, et al. Risk of cancer in cases of suspected lynch syndrome without germline mutation. Gastroenterology. 2013;144(5):926-932.
4. Le DT, Uram JN, Wang H, et al. PD-1 Blockade in Tumors with Mismatch-Repair De ciency. N Engl J Med. 2015;372(26):2509-2520.