BRAF refines clinical interpretation of mismatch repair deficiency in colorectal cancer

Reliable TNM classification is mandatory in order to advise colorectal cancer patients regarding prognosis and optimal choice of treatment strategy. Despite the revolution in molecular biology, molecular markers have had a minor role in the initial characterization of colorectal cancer. It is now time to not only use immunohistochemical analysis of the mismatch repair (MMR) system, but also BRAF as the standard screening strategy in every new colorectal cancer.

It is well known that the MMR system is clinically relevant for prognosis, heredity and treatment choices, but it is now becoming clearer that knowledge of BRAF status further supports all of these important questions. Other prerequisites for carrying out BRAF testing in the daily clinic – availability, fair price and reliability – have also been sufficiently satisfied.

BRAF

BRAF is a proto-oncogene in the RAS–RAF–MEK–ERK–MAP kinase pathway that affects cell division, differentiation and secretion. A missense mutation (V600E) in the kinase domain activates the serine/threonine kinase and causes RAS-independent downstream signaling [1].

Until recently, the BRAF V600E mutation could only reliably be analyzed using gene sequencing. Therefore, analysis has been restricted to molecular biology laboratories and academic-based clinical departments of pathology. A new mutation-specific immunohistochemical approach has now been validated against sequencing [2] and tested in several settings with promising results [3,4].

Mismatch repair

There are four important proteins involved in DNA MMRs: MLH1, PMS2, MSH2 and MSH6. They work in pairs detecting and correcting base-to-base mispairing and loop formation. The system dysfunctions in Lynch syndrome, which is caused by hereditary mutations in any of the four genes. MLH1 and MSH2 mutations are the commonest, while MSH6 and especially PMS2 mutations are rare [5]. The only known somatic (i.e., nonhereditary) cause of MMR deficiency (dMMR) is MLH1 promoter hypermethylation [6].

Immunohistochemistry is a reliable way to detect dMMR and is a standard analysis in most pathology departments [7]. The advantages of this test are the positive internal control from protein expression in non-neoplastic cells, the fast procedure, low cost and information gained about the affected gene. PMS2 and MSH6 are lost in the absence of MLH1 and MSH2, respectively, while isolated loss of PMS2 or MSH6 is typical for a mutation in the corresponding gene. dMMR causes accumulation of length mutations in tumor DNA compared with normal DNA. This is detected as microsatellite instability [5].

Prognosis

It is well known that dMMR is a positive prognostic factor [8] and it has also been suggested as a marker for deselecting stage II patients for adjuvant chemotherapy [101]. More data are now accumulating, indicating that BRAF holds additional prognostic information. In a recent study, patients with proficient MMR (pMMR)/BRAF mutant tumors had a worse outcome, while dMMR/BRAF mutant and especially dMMR/BRAF wild-type patients had a favorable outcome. Toon and colleagues conclude that combined BRAF and MMR status is a marker for prognostic risk stratification [9].

Heredity

It can be extracted from the description of MMR above that immunohistochemistry can readily detect tumors probably caused by hereditary mutations in MSH2, MSH6 or PMS2. In the case of MLH1 deficiency, the sporadic phenomenon promoter hypermethylation is seen in nine out of ten tumors. Methylation analysis is available in specialized laboratories and it makes good sense to perform it in order to identify the vast majority of dMMR patients with sporadic disease. Interestingly, BRAF V600E is a carcinogenic mutation not seen in Lynch syndrome [10]. It is, therefore, an alternative to MLH1 promoter methylation in the identification of sporadic MLH1-negative cases. Some laboratories have found BRAF sequencing easier than methylation assays and its clinical value has been validated [11]. As mentioned above, immunohistochemical detection of BRAF mutations is now available and can be performed by the pathologist together with MMR. This is a major step forward in the so-called reflex screening for Lynch syndrome. Every new colorectal cancer should be characterized with respect to MMR and BRAF. In dMMR cases without BRAF mutations, the patient should be offered genetic counselling. Approximately 2–4% of all colorectal cancers can be identified by this method and each proband will have, on average, three family members with the same mutation [12]. With regular colonoscopies and endometrial ultrasound, cancer deaths can be prevented.

Treatment

Surgery

If immunohistochemistry for MMR and BRAF mutations are performed at the time of diagnosis on the diagnostic biopsies [13], Lynch syndrome can be detected before surgery. This allows the surgeon to take the high risk of new cancers into account and the patient may be advised to have a total or subtotal colectomy [7]. Therefore, clinical geneticists should participate in the multidisciplinary assessment of all colorectal cancer patients to allow timely counseling.

Adjuvant chemotherapy

MMR and BRAF status identify subgroups where the favourable prognosis does not justify adjuvant chemotherapy and a subgroup of pMMR/BRAF mutations with a high risk of relapse and, thus, a strong indication for adjuvant chemotherapy. This should be discussed with the patient.

Prognostic markers identify patients at risk of relapse, while predictive markers identify treatments likely to kill tumor cells irrespective of the risk of relapse. The predictive effect of dMMR has been a matter of great debate and conflicting results. A common interpretation is that dMMR tumors are less sensitive to adjuvant monotherapy with 5-fluorouracil and it should be avoided [14]. The standard of care is 5-fluorouracil and oxaliplatin, and it should be offered to patients eligible for combination chemotherapy. Another interpretation is that a positive effect of chemotherapy in dMMR patients might be driven by the subgroup of patients with hereditary disease and not sporadic dMMR [15], rendering wild-type BRAF as a marker of sporadicity even more important.

Chemotherapy for metastatic disease

In metastatic colorectal cancer, MMR status does not seem to be a predictor of the effectiveness of available systemic therapies. BRAF V600E on the other hand is very important. BRAF and RAS mutations are mutually exclusive and are downstream of EGFR. In theory, this renders the tumor insensitive to extracellular blockage with the EGFR inhibitors cetuximab and panitumumab. In practice, tumors harboring the BRAF V600E do not respond to these drugs. There may be a small effect in terms of progression-free survival and overall survival [16], but the potential minor benefit does not justify the use of EGFR inhibitors. The study also confirmed that patients with BRAF V600E have a very bad outcome in the metastatic setting, with survival times halved. In early-phase trials, the selected subgroup of patients with the BRAF V600E mutation seemed to benefit from more intense chemotherapy with triplet cytotoxics combined with bevacizumab [17]. It would be of major clinical interest if a negative prognosis can be repealed by a positive predictive value for a certain treatment.

Conclusion

In my opinion, immunohistochemistry for MLH1, PMS2, MSH2, MSH6 and BRAF should be performed in every new colorectal cancer, preferably on the diagnostic biopsies. Based on the results, the patients can be informed more precisely about prognosis, the treating doctors have a more solid background to suggest the optimal treatment, whether it is surgery, chemotherapy or a combination, and the specific hereditary disease Lynch syndrome can be identified with the potential to prevent cancer death in the patient and their family.

Future perspective

We are not likely to see any large randomized studies giving absolute evidence that MMR and BRAF immunohistochemistry will improve survival and be cost effective, but we can choose a scientific approach based on biology and the key clinical data highlighted above. Immunohistochemical data should be collected in comprehensive databases and, thus, allow comparative effectiveness research of BRAF in clinics. Other research areas include BRAF as a marker for effect of adjuvant chemotherapy and, in the metastatic setting, randomized trials should confirm the benefit from intense first-line chemotherapy in metastatic BRAF-mutated cases. With the implementation of BRAF analysis as an integrated part of reflex screening for Lynch syndrome, a basis is founded for cost–benefit analyses, general adoption and lives being saved.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.