Can the tumor-agnostic evaluation of MSI/MMR status be the common denominator for the immunotherapy treatment of patients with several solid tumors?

Alterations in short-repetitive DNA sequences, known as microsatellite instability (MSI), can reflect deficiencies in Mismatch Repair (MMR) system which represents a major player in DNA integrity maintenance. The incidence of MSI-H/dMMR has been shown to be variable depending on the tumor type. Several studies confirmed that dMMR/MSI status, although less frequent than PD-L1 expression, may better predict response to immune-checkpoint inhibitors (ICIs) in patients with solid tumors. In October 2016, the FDA granted pembrolizumab as breakthrough therapy for the treatment of non-CRC, MSI-H/dMMR tumors, providing, for the first time, a tumor-agnostic indication. In the next future, the tissue-agnostic evaluation of MSI-H/dMMR could become the common denominator for the immunotherapy treatment of patients with different advanced solid tumors, in order to select patient subgroups which may benefit from this therapy. In this Review we provided an overview of the main clinical studies describing the association between MSI-H/dMMR tumors and immunotherapy response.

Intratumor heterogeneity: the hidden barrier to immunotherapy against MSI tumors from the perspective of IFN-γ signaling and tumor-infiltrating lymphocytes

In this era of precision medicine, with the help of biomarkers, immunotherapy has significantly improved prognosis of many patients with malignant tumor. Deficient mismatch repair (dMMR)/microsatellite instability (MSI) status is used as a biomarker in clinical practice to predict favorable response to immunotherapy and prognosis. MSI is an important characteristic which facilitates mutation and improves the likelihood of a favorable response to immunotherapy. However, many patients with dMMR/MSI still respond poorly to immunotherapies, which partly results from intratumor heterogeneity propelled by dMMR/MSI. In this review, we discuss how dMMR/MSI facilitates mutations in tumor cells and generates intratumor heterogeneity, especially through type II interferon (IFN-γ) signaling and tumor-infiltrating lymphocytes (TILs). We discuss the mechanism of immunotherapy from the perspective of dMMR/MSI, molecular pathways and TILs, and we discuss how intratumor heterogeneity hinders the therapeutic effect of immunotherapy. Finally, we summarize present techniques and strategies to look at the tumor as a whole to design personalized regimes and achieve favorable prognosis.

Worldwide variation in lynch syndrome screening: case for universal screening in low colorectal cancer prevalence areas

To perform a systematic assessment of universal Lynch syndrome (LS) screening yield in colorectal cancer (CRC) patients around the world. Universal screening for LS is recommended in all CRC patients. However, the variation in yield of LS screening in the setting of significant global variation in CRC prevalence is unknown. A systematic review of articles in the MEDLINE database was performed to identify studies performing universal screening for LS. All cases with microsatellite instability (MSI-H) or missing one or more proteins on immunohistochemistry (IHC) were considered screening positive. The overall pooled yield of universal LS screening in 97 study arms from 89 identified studies was 11.9% (5649/47545) and the overall pooled percentage of confirmed LS patients was 1.8% (682/37220). LS screening positivity varied significantly based on geographic region (Kruskal Wallis test, p < 0.001) and reported 5-year CRC prevalence in the country (Fisher's exact, p < 0.001). Significant inverse correlation was found between LS screening positivity and 5-year CRC prevalence (Pearson correlation, r =  - 0.56, p < 0.001). The overall yield of LS screening was 15.00% (382/2553) and rate of confirmed LS was 7.7% (113/1475) in LS screening done in patients ≤ 50 years (16 studies). There is significant geographic variation in LS screening positivity with higher yield in countries with lower prevalence of CRC. Our results highlight the importance of universal LS screening in younger patients and low CRC prevalence countries.

Frequency of Mismatch Repair Protein (MMRP) Deficiency among Young Jordanians Diagnosed with Colorectal Carcinoma (CRC)

Purpose

Microsatellite instability (MSI) caused by mismatch repair protein (MMRP) deficiency is detected in 15% of sporadic colorectal cancers (CRCs). Our aim is to investigate the frequency of MMRP deficiency in young CRC patients, using immunohistochemical analysis.

Methods

This study targeted cases of CRC at King Hussein Cancer Center from 2004 until 2012 in patients 45 years of age or younger at the time of diagnosis. Clinicopathological data was obtained from 155 patients' records. Immunohistochemistry for MLH1, MSH2, PMS2, and MSH6 proteins was performed on paraffin-embedded tissue containing carcinoma.

Results

The median age of patient at diagnosis was 38 years. A total of 29 (19%) cases showed deficient MMRP(dMMRP)expression. Loss of expression of PMS2 was seen in 17 cases, 12 cases of which showed loss of MLH1 expression. Loss of expression of MSH6 was seen in 10 cases, 9 of which showed loss of MSH2 expression. One case (3.4%) showed loss of all four MMR proteins, and another case (3.4%) showed loss of PMS2/MLH1 and MSH6. There was a significant association between abnormal MMR protein expression and tumor location proximal to splenic flexure (p value 0.000), pathologic features suggestive of microsatellite instability (p value 0.000), P53 negativity (p value 0.000), and stage (p value 0.02). Patients with dMMRP CRC appeared to have a significantly better overall survival compared to patients with proficient MMRP(pMMRP)(p value 0.02). Loss of MSH2/MSH6 was significantly associated with positive family history of cancer (p value = 0.020).

Conclusions

The prevalence of dMMRP tumors in this age group appears to be similar to international literature. dMMRP tumors tends to be associated with earlier stages and better outcomes compared to pMMRP cases. dMMRP can serve as a biomarker for better prognosis. These results are of value in directing the clinical management of young patients with CRC.

Prevalence and clinicopathological characteristics of mismatch repair-deficient colorectal carcinoma in early onset cases as compared with late-onset cases: a retrospective cross-sectional study in Northeastern Iran

OBJECTIVES

Lynch syndrome (LS), a genetically inherited autosomal disorder, increases the incidence of colorectal carcinoma (CRC). We aimed to perform a universal strategy to assess the prevalence and clinicopathological characteristics of early onset CRCs at high risk of LS versus late-onset ones in the Iranian population.

SETTING

A local population-based study from Northeastern Iran.

PARTICIPANTS

321 consecutive CRCs and pathology specimen screened between 2013 and 2016.

PRIMARY AND SECONDARY OUTCOME MEASURES

Retrospectively, information regarding the clinical criteria was obtained by interviewing the patients with CRC or, their families. Pathologists tested tumours with immunohistochemistry (IHC) staining of four mismatch repair (MMR) proteins (MLH1, MSH2, MSH6 and PMS2). Tumours with absent IHC staining of MLH1 were tested for BRAF mutations to exclude sporadic CRCs. Prevalence of early onset CRCs at high risk of LS and familial CRC type X were assessed as primary and secondary outcome measures, respectively.

RESULTS

Of 321 CRCs (13/123 (10.57%), early onset vs 21/198 (10.6%) late-onset) were detected to be MMR-deficient (dMMR). Nine early onset cases and 14 late-onset ones with a loss of MLH1 underwent testing for the BRAF mutation, none of the early onset and four (2.02%) late-onset were recognised as sporadic. The difference in the outcome of IHC-analysis between early and late-onset CRCs at high risk of LS was not statistically significant (p=0.34). Majority of the suspected LS tumours from early onset patients had arisen in distal part (8/11 (72.72%) vs 8/14 (57.14%)), all of which were occurred in the rectum or sigmoid.

CONCLUSION

Clinically, these findings suggest that in case of limitation for BRAF testing, the practitioner in Iran may consider managing early onset dMMR cases like LS until access to BRAF testing becomes available to them, before germline testing to accurately diagnose LS.

Prevalence and clinicopathologic/molecular characteristics of mismatch repair-deficient colorectal cancer in the under-50-year-old Japanese population

PURPOSE

To clarify the prevalence and clinicopathologic/molecular characteristics of mismatch repair (MMR)-deficient colorectal cancer in the young Japanese population.

METHODS

Immunohistochemical analyses for MMR proteins (MLH1, MSH2, MSH6, and PMS2) were performed in formalin-fixed paraffin-embedded sections prepared from the resected CRC specimens of 119 consecutive patients aged <50 years old, who underwent resection of the primary tumor at our institution between 1996 and 2015. Analyses for somatic BRAF V600E mutation, somatic hypermethylation of the MLH1 promoter, and germline MMR gene mutations were undertaken where indicated.

RESULTS

MMR protein loss was found in 10 patients (8.4%), 7 (5.9%) of whom were subsequently identified to have Lynch syndrome (LS). The remaining 3 patients were categorized as having sporadic MMR-deficient CRC (n = 2) or "possible LS (n = 1)". In multivariate logistic regression analysis, the presence of tumor-infiltrating lymphocytes (P < 0.01), right-sided location of the tumor (P = 0.01), and a history of LS-associated tumors in the first-degree relatives (P < 0.01) were identified as independent factors predictive of MMR-deficient CRC.

CONCLUSION

These results are of value in the clinical management of patients with the early onset CRC under circumstances where universal tumor screening approaches for LS are still not available, like in Japan.

Systematic immunohistochemistry screening for Lynch syndrome in early age-of-onset colorectal cancer patients undergoing surgical resection

BACKGROUND

Lynch syndrome (LS), defined by a deleterious (pathogenic) germline mutation in a mismatch repair (MMR) gene, is characterized by early age-of-onset colorectal cancer (CRC). Because clinical criteria for LS, such as the Amsterdam II Criteria, may miss cases, reflex tumor tissue testing of all CRC patients for LS has been proposed. Our study describes the impact of routine immunohistochemistry (IHC) analysis of tumor tissue for loss of MMR protein expression in early age-of-onset CRC patients undergoing resection.

STUDY DESIGN

A prospective institutional program was established to perform IHC analysis on all early age-of-onset (≤50 years) CRC patients undergoing resection. Patients with abnormal IHC analysis were referred to the Clinical Genetics Service for further evaluation. The study cohort excluded patients with other polyposis syndromes and inflammatory bowel disease.

RESULTS

IHC was performed on 198 patients from July 2006 to June 2010. The median age was 42.8 years (range 23.1 to 50.6 years). Abnormal IHC was reported in 38 (19.1%) patients, and 22 (57.8%) with abnormal IHC analysis had germline genetic testing. Seventeen (77.2%) had an alteration detected in an MMR gene: 10 were known to be deleterious mutations and 7 were variants of uncertain significance. Overall, LS was detected in 5.1% of patients. Only 2 of the 10 (20%) with a deleterious mutation actually met the Amsterdam II Criteria.

CONCLUSIONS

Reflex IHC testing for LS on early age-of-onset CRC patients undergoing resection is feasible at the institutional level. This strategy identifies a substantial number of LS patients who would have been missed if genetic testing was based on the Amsterdam II Criteria alone.

Immunohistochemistry to detect hereditary nonpolyposis colorectal cancer in young patients: the 7-year Auckland experience

BACKGROUND

In accordance with the Bethesda Guidelines, Auckland's metropolitan hospitals routinely perform immunohistochemistry for mismatch repair proteins on the tumor specimens of all patients with colorectal cancer aged 50 years and younger. When loss of expression is evident, patients are offered genetic counseling and gene mutation analysis.

OBJECTIVES

This study aimed to determine the completeness of young patient capture over the first 7 years of routine testing, to find whether patients were referred for genetic testing, and to determine the proportion of patients found to have a mismatch repair gene mutation.

DESIGN

This study retrospectively reviewed clinical, pathological, and genetic data.

SETTINGS

The study was conducted at 3 public hospitals in Auckland, New Zealand.

PATIENTS

All patients aged 50 years and younger treated for colorectal cancer at Auckland's metropolitan hospitals between January 2001 and December 2007 (n = 243) were included.

MAIN OUTCOME MEASURES

The loss of expression of mismatch repair proteins by immunohistochemistry, referral for genetic testing, and proportion with mismatch repair gene mutation were the main outcome measures.

RESULTS

Two hundred fourteen (88%) eligible patients had immunohistochemical analysis of their tumor and 33 (14%) had loss of expression of one or more mismatch repair proteins. Twenty-six patients were referred for genetic counseling, of whom 22 underwent genetic testing. A mismatch repair gene mutation was identified in 10 patients.

LIMITATIONS

Seven patients with loss of expression of mismatch repair proteins by immunohistochemistry were not referred for genetic assessment.

CONCLUSIONS

We have identified a mismatch repair gene mutation diagnostic of hereditary nonpolyposis colorectal cancer in 5% of all patients with colorectal cancer who were aged 50 years and younger. Routine immunohistochemical prescreening has important clinical benefit for these patients and their relatives.