• Luca Roncati.
• Dr. Luca Roncati, MD, PhD, Department of Medical and Surgical Sciences Institute of Pathology, University Hospital of Modena, Policlinico, I-41124 Modena (MO), Italy; emailmedical@gmail.com.
• Acta Dermatovenerol Croat. 2018 Dec 1; 26 (4): 341-343.

AbstractDear Editor, Immune-checkpoint blockade is a type of passive immunotherapy aimed at enhancing preexisting anti-tumor responses of the organism, blocking self-tolerance molecular interactions between T-lymphocytes and neoplastic cells (1,2). Despite a significant increase in progression-free survival, a large proportion of patients affected by metastatic melanoma do not show durable responses even after appropriate diagnostic categorization and shared therapeutic choices (3-9). Therefore, predictive biomarkers of clinical response are urgently needed, and predictive immunohistochemistry (IHC) meets these requirements. Strong evidence suggests that DNA mismatch repair (MMR) deficiency is a frequent condition in malignant melanoma, as well as in other tumors (10). As is known, DNA MMR is a safeguard system for the detection and repair of DNA errors, which can randomly occur in the phase of DNA replication inside the cell. In humans, seven DNA MMR proteins (Mlh1, Mlh3, Msh2, Msh3, Msh6, Pms1, and Pms2) work in a coordinated and sequential manner to repair DNA mismatches. When this system is defective, the cell accumulates a series of replication errors in terms of new microsatellites; therefore, a condition of genetic hypermutability and microsatellite instability (MSI) takes place inside the cell itself (11). For this reason, my working group has started to search for MMR protein deficiency in melanoma biopsies from patients of both sexes and of all ages with metastatic spread, correlating the data with the response to pembrolizumab, the well-known anti-programmed cell death protein 1 (PD1) human monoclonal immunoglobulin G4, capable of blocking the interaction between PD1, the surface receptor of activated T-lymphocytes, and its ligand, the programmed death-ligand 1 (PD-L1), favoring melanoma cell attack by T-lymphocytes (1) rather than its depression (12). PD-L1 is highly expressed in about half of all melanomas and thus the role of PD1 in melanoma immune evasion is now well established (13). Surprisingly, the best therapeutic results to pembrolizumab, in terms of progression-free survival and overall survival, occur precisely in those patients, approximately 7% in my database, affected by deficient MMR (dMMR) melanomas. In particular, the most important benefits to pembrolizumab-based treatment have occurred in a female patient, who developed a subungual melanoma in the second finger of the left hand at the age of 41 years, together with lymph node metastases to ipsilateral axilla at the onset. The patient was promptly submitted to amputation of the first phalanx and emptying of the axillary cable. The primary tumor was a vertical growth phase melanoma with a Breslow's depth of 1.4 mm; three mitotic figures for 1 mm2 were ascertained. There was no evidence of ulceration, regression, microsatellitosis, or lymphocytic infiltration; moreover, the surgical margins tested free of disease. Further molecular analyses did not show rearrangements in B-RAF and C-KIT genes. After four years, metastases appeared in the brain and ileum; however, at present the patient is still alive and in complete pembrolizumab response with progression-free survival and overall survival of 956 days and 2546 days, respectively. The tumor was afterwards identified as a dMMR melanoma for an exclusive loss of Msh6 expression on IHC (Figure 1). This finding is in line with the fact that the U.S. Food and Drug Administration has approved the use of pembrolizumab in 2017 for unresectable or metastatic solid tumors with MMR deficiency (14). In conclusion, dMMR melanoma seems to be a particular subset of disease that can be identified with high sensibility and specificity by predictive IHC as a complete loss of one or more DNA MMR proteins and that deserves targeted therapy.

18 keywords...

hide…