By Mridula A. George, MD, and Ann W. Silk, MD
- Merkel cell carcinoma is characterized by sensitivity to chemotherapy and radiotherapy, but has a high chance of relapse due to distant metastases.
- Merkel cell polyomavirus is associated with approximately 80% of cases of Merkel cell carcinoma. Testing for antibodies to the oncoprotein can inform clinical surveillance after curative surgery and/or radiation therapy. Merkel cell polyomavirus positive tumors are associated with enhanced immune cell infiltration, high PD-L1 expression, and low tumor mutational burden.
- Virus-negative tumors are characterized by a high tumor mutational burden.
- Immune checkpoint therapy has largely replaced chemotherapy as first-line systemic treatment.
Merkel cell carcinoma (MCC) is an aggressive neuroendocrine tumor of the skin. Like other high-grade neuroendocrine tumors, MCC is characterized by sensitivity to chemotherapy and radiotherapy, but has a high chance of relapse because of distant metastases. MCC is sensitive to cisplatin or carboplatin with or without etoposide.1 MCC is often initially chemosensitive, but tumor responses are rarely durable, suggesting the rapid emergence of chemoresistance.2 Voog et al reported a 57% response rate for first-line chemotherapy in patients with metastatic disease, and Tai et al reported a 59% response rate.3,4 In a retrospective study done by Iyer et al, which studied 62 patients with metastatic MCC treated with chemotherapy, the progression-free survival (PFS) was 3 months with a median overall survival (OS) of 9.5 months.5
MCC has metastatic potential regardless of the size of the primary tumor and a 5-year disease associated mortality rate of 46%.6 Upon presentation, approximately 66% of patients have local disease, 27% have nodal involvement, and 7% have distant metastasis.7 Patients with local disease have a 64% relative survival at 5 years, as compared to 39% in patients with regional nodal disease and 18% with metastatic disease.8 As of 2013, the annual incidence of MCC in the United States is 0.7 per 100,000 persons.9 In 1986, incidence and mortality rates per 100,000 were 0.22 and 0.03, respectively, which has increased to 0.79 and 0.43 in 2011, respectively.10 There are approximately 1,600 new cases diagnosed per year in the United States.11 The age-adjusted incidence of MCC is estimated at 0.24 per 100,000 person-years, which is higher than the estimated 0.13 per 100,000 person-years found in Europe.12
Immunosuppression and immunodeficiency have been found to play a role in the pathogenesis of MCC given that there is a 50-fold increase in the incidence in patients with HIV, organ transplants, autoimmune disorders, and lymphoproliferative disorders.7,12 The association with immunosuppression and immunodeficiency led to the discovery of Merkel cell polyomavirus, an oncogenic virus that integrates into the genomic DNA leading to the expressions of oncoproteins on the tumor surface.13
Merkel cell polyomavirus is associated with approximately 80% of MCC cases.14 Merkel cell polyomavirus is a double-stranded DNA virus that consists of both early and late gene regions. The early region contains the tumor antigen locus, and the late region encodes viral capsid proteins. Merkel cell polyomavirus large tumor antigen and small tumor antigen promotes oncogenesis.15 Large tumor antigen downregulates Toll-like receptor (a mechanism that promotes immune evasion), disables retinoblastoma protein and activates survivin (an inhibitor of apoptosis).16-19 Small tumor antigen inhibits proteosomal degradation of large tumor antigen, c-Myc, NOTCH, mTOR, NF-kB2, and cyclin E, contributing to oncogenesis.20 Merkel cell polyomavirus has been shown to suppress p53 function in MCC.21
Viral oncoprotein-specific CD4 and CD8 T cells are detectable in tumor and peripheral blood in patients with virus-positive MCC. Antibodies to Merkel cell polyomavirus oncoproteins have been detected in patients with MCC and have been found to correlate with disease burden.22 A clinical test called AMERK is available from the University of Washington. Serum can be tested in the first 3 months following surgical excision of MCC to determine if antibodies to the oncoprotein are present. On serial measurements, it may drop to undetectable. Monitoring the titer over time can be useful to predict or correlate with recurrence and, presumably, it could be used to increase the interval for surveillance imaging.
Virus-positive tumors and virus-negative tumors have unique genomic alterations supporting the etiology of their pathogenesis. Virus-negative MCC tumors are usually seen in the sun-exposed regions of the head, neck, or limbs. Virus-negative subtypes are more prevalent in Australia compared with the Northern Hemisphere, suggesting that UV radiation is central to the pathogenesis of virus-negative MCC. Their genomic alterations have a distinctive UV damage signature of tandem CC to TT substitutions. Generally, the mutation burden is high.23-25 Recurrent mutations noted in the virus-negative MCC included RB1, TP53, NOTCH1, APC, and TET2.25 Virus-negative tumors generally have low frequency of tumor-infiltrating lymphocytes (TILs) and low PD-L1 expression in the tumor microenvironment.25,26
The persistent expression of viral proteins triggers a humoral and cellular response; however, the MCC tumor cells manage to evade the immune system. Merkel cell polyomavirus infection enhances immune cell infiltration into the tumor. Merkel cell polyomavirus-positive tumors contained a higher proportion of CD3+, CD8+, CD4+, CD68+ TILs compared to virus-negative tumors.27 In MCC, the presence of high infiltration by CD8+ T lymphocytes has been shown to be an independent predictor of survival, similar to other malignancies such as melanoma, non–small cell lung cancer, and ovarian cancer.27,28 Patients who have a high number of tumor-infiltrating T cells have a survival benefit, irrespective of the viral positivity.
In spite of the presence of tumor-specific T cells, their activation has shown to be suppressed. CD69 and CD25 are markers of T-cell activation and expression of these markers was decreased in MCC TILs.29 In the setting of PD-1 expression on the MCC tumor cells, this is a sign that the tumor cells cause T-cell exhaustion, which results in decreased cytokine production and reduced cytotoxicity, and leads immune escape.30
The major histocompatibility complex (MHC) class I on the cell surface plays a role in presenting tumor-associated antigens expressed on MCC tumor cells to CD8+ T lymphocytes. Immunohistochemical evaluation of 114 MCC tumors showed that 84% of the tumors had downregulated expression of MHC class I, with 51% having undetectable MHC class I expression.31 Downregulation of MHC class I impairs T-cell recognition of MCC-specific tumor antigens, thus able to evade the immune system. Inhibition by the T-regulatory cells is another mechanism for immune evasion in MCC.29 In mouse model studies, the tumor cells were noted to have locally high concentrations of both CD4 and CD8 T-regulatory cells causing impaired activation of T cells.29
Approximately half of MCCs express the immunosuppressive ligand PD-L1 on the tumor cells and on the TILs.26 Among tumors with brisk TILs, PD-L1 expression on the tumor cells is nearly universal. Based on the high expression of PD-L1 in the tumor microenvironment, immune checkpoint inhibitors were studied in MCC. PD-1 blockade with pembrolizumab showed an objective response rate of 56%.32 PFS at 6 months was 67% (95% CI, 49 to 86).32 Of note, tumor PD-L1 status was not associated with outcome, but response rate was numerically higher (62% vs. 44%) in tumors caused by the Merkel cell polyomavirus versus those caused by UV exposure. In CheckMate-358, 25 patients with MCC were treated with 240 mg of nivolumab every 2 weeks until progression with a median follow-up of 26 weeks. In 22 evaluable patients, the objective response rate was 68%. Responses were noted in both virus-positive and virus-negative cancer. At 3 months, PFS was 82%.33
Avelumab, a fully human anti–PD-L1 IgG1 monoclonal antibody that inhibits PD-L1 and PD-1 interactions, was also studied in patients with stage IV chemotherapy-refractory MCC. The 6-month durable response rate was 29%, and the 6-month PFS rate was 40%.34 Only 5% grade 3/4 immune-related adverse events were observed. Based on the results of this open label non-randomized trial, avelumab obtained fast-track, accelerated approval and orphan drug designations from the U.S. Food and Drug Administration. The agent received approval for the treatment of metastatic MCC, including those who had not received prior chemotherapy.35 In the most recent edition of the National Comprehensive Cancer Network Guideline for MCC, preferred first-line systemic therapies include clinical trials, avelumab, pembrolizumab, and nivolumab, which have largely replaced chemotherapy as first-line treatment.
Clinical trials are exploring earlier initiation of checkpoint blockade, combinations with a checkpoint inhibitor backbone, and novel agents. There are clinical trials ongoing with ipilimumab (anti–CTLA-4) in the adjuvant setting after resection of local disease (NCT02196961). There is a phase II clinical trial underway studying the combination nivolumab and ipilimumab versus nivolumab, ipilimumab and stereotactic body radiation therapy for metastatic MCC (NCT03071406).
Novel agents include oncolytic viral therapy and adoptive cell therapies. Talimogene laherparepvec is a genetically modified herpes simplex virus, type I, which has been attenuated by the deletion of the herpes neuro-virulence viral genes, and enhanced for immunogenicity.36 Blackmon et al published a case report of two elderly patients with regionally advanced, rapidly progressing MCC who were treated with intratumoral talimogene laherparepvec.37 Both patients did not have antibody against Merkel cell polyomavirus oncoproteins.37 Talimogene laherparepvec was administered intratumorally into all palpable metastases with minimal toxicity. One patient was noted to have complete response 9 weeks after the first injection, with sustained response 5 months post-treatment. The second patient had partial response lasting 7 months since the initiation of treatment. There is currently a phase II clinical trial (NCT02978625) underway studying talimogene laherparepvec in combination with nivolumab in the metastatic setting.
Adoptive cell therapy (ACT) or autologous transfer of tumor-reactive T cells has been shown to have clinical responses in some patients with EBV-positive B-cell malignancies, HPV-positive cervical cancer, and metastatic melanoma.38-41 In ACT, TILs against viral or tumor-specific antigens are harvested from the patient, then expanded in vitro and then reinfused into their original host. CD8-negative T cells recognize at least 17 unique epitopes of the persistently expressed T antigens of Merkel cell polyomavirus.
Chapuis et al published a case report where Merkel cell polyomavirus–specific T cells were harvested from a patient age 67 with Merkel cell polyomavirus–positive MCC who presented with metastatic disease after wide local excision and radiation of original disease. The patient received three infusions of Merkel cell polyomavirus–specific CD8 cells in conjunction with interleukin-2 and human leukocyte antigen (HLA)-I upregulating agents (single-dose radiation and interferon-beta injections). The patient had complete response in two of three metastatic lesions and a prolonged period without development of additional distant metastases (535 days compared to historic median of 200 days).
A phase I/II trial studied Merkel cell polyomavirus–specific T cells in combination HLA upregulation (radiation or interferon) with (i.e., triple therapy) and without (i.e., double therapy) avelumab.42 Four patients received triple therapy and three of the four patients had sustained complete response. One out of the four patients who received double therapy had complete response before progression at 14 months, but the other three patients had progression of disease.42
Studies done on pathogenesis of the tumor, viral oncogenesis, and immunogenicity have shed light on the PD-1 axis blockade, as well as other immune-mediated pathways. Immunotherapy combinations that have shown benefit in other tumor types may be beneficial in MCC. Studies on predictive biomarkers will help identify patients who are more likely to benefit from a particular treatment and improve on immunotherapy in the first-line setting.
About the Authors: Dr. George is a Hematology/Oncology Fellow at Rutgers Cancer Institute of New Jersey. Dr. Silk is a medical oncologist with Rutgers Cancer Institute of New Jersey.