The authors discuss the current standard of care and review the new data for immunotherapy in advanced merkel cell carcinoma (MCC).
Ragini Kudchadkar, MD
Merkel cell carcinoma (MCC) is a rare but potentially lethal skin cancer that has traditionally been treated with surgery, radiation, and in advanced cases, chemotherapy. Patients with even localized disease have a high risk of recurrence and those who develop metastatic disease have had a poor overall survival. Several factors have been associated with the development of MCC including immunosuppression, ultraviolet (UV) exposure, and infection with the Merkel cell polyomavirus (MCPyV). Randomized-controlled trials are a paucity in this disease; therefore, historic treatment algorithms are mostly based on retrospective case series. Historically, the National Comprehensive Cancer Network guidelines have recommended participation in a clinical trial for patients with metastatic disease. If patients were not able to participate in or access a trial, then they would often receive a combination of platinum and etoposide chemotherapy. However, recent years have seen trials of immunotherapies showing benefit for patients with stage IV disease. In 2017, avelumab became the first immunotherapy to be approved for advanced stage MCC. This paper will summarize the basis of the current standard of care for this disease as well as review the new data for biomarkers and immunotherapy in advanced MCC.
Merkel cell carcinoma (MCC) is a rare and aggressive cutaneous malignancy with an estimated incidence in the United States of approximately 1600 cases per year. Unfortunately, the number of cases has been steadily increasing over the last few decades. Patients with stage I or II disease have a 5-year disease specific survival of 78% while those with stage III disease have only a 54%.1The 5-year overall survival (OS) for patients with metastatic disease has historically averaged between 0% to 18%.2-5There is a higher incidence of MCC in men and it predominantly affects those with lighter skin. MCC typically affects older individuals with the mean age for diagnosis of 74 years for men and 76 years for women.6The risk of MCC is higher in immunosuppressed individuals such as those with a history of organ transplant, HIV, or other malignancy.
Due to the rarity of this disease, the data on natural history and treatment of MCC are lacking in quality; also, there are sparse data from clinical trials. The data are gathered primarily from retrospective case series, although a handful of randomized trials are available. This should be taken into account while reading this review.
Primary lesions of MCC generally appear as painless, rapidly growing, and nontender lesions. The color can vary from flesh colored to red or bluish. The most common sites are in sun-exposed areas with 43% occurring in the head and neck area.7Diagnosis of MCC requires a high index of suspicion. The acronym AEIOU was developed to help identify clinical characteristics of lesions suggestive of MCC: asymptomatic, expanding rapidly, immune suppression, older than 50 years of age, and UV-exposed area in a fair-skinned individual.8 Traditionally, MCC is thought to arise from Merkel cells located in the basal layer of the dermis. However, there are alternative hypotheses regarding the cell of origin. On immunohistochemistry examination, MCC expresses both epithelial and neuroendocrine markers. CK20 is consistently positive with a paranuclear staining pattern.
Several factors have been associated with the development of MCC including immunosuppression, UV exposure, and infection with the Merkel cell polyomavirus (MCPyV). MCPyV is a ubiquitous double stranded DNA virus. Approximately 88% of older individuals with MCC will have positive antibody titers specific to the capsid viral protein 1.9 The virus is present at tumor initiation and the viral proteins are felt to be oncogenic drivers. There is a subset of MCCs with low or absent MCPyV antigen expression. These tumors are felt to be related more to UV exposure and have higher rates of mutations in tumor suppressor genes such as TP53 and RB1.10,11The increase in mutational burden is felt to increase expression of neoantigens and immunogenicity.
Systemic Therapy Options
Historically, the National Comprehensive Cancer Network guidelines have recommended participation in a clinical trial for patients with metastatic disease. If patients were not able to participate in or access a trial, then they would often receive a combination of platinum-based chemotherapy and etoposide in the first-line setting. MCC is a chemosensitive disease with response rates of 51% to 63% with chemotherapy.3,5,12-14However, these responses are short in duration with a median progression-free survival (PFS) of 94 days. Additionally, chemotherapy has not demonstrated an improvement in OS.14Given the older age of many patients and their comorbidities, the toxicity of treatment could severely impact quality of life. However, until recently the majority of patients were treated with chemotherapy.
In March 2017, the FDA approved avelumab, a programmed death receptor ligand 1 (PD-L1) inhibitor. Approval was based on the results of a phase II study that enrolled 88 patients. All patients had progressed on first-line chemotherapy. They were treated with avelumab 10 mg/kg every 2 weeks. The objective response rate (ORR) was 32% with a complete response rate (CRC) of 9%. Ten percent of patients had stable disease. Median PFS was 2.7 months and median OS was 11.3 months. The duration of response was greater than or equal to 6 months in 86% of patients. Treatment was overall well tolerated with no grade 4 or 5 toxicities and only 4.5% of the patients experienced grade 3 toxicities. Most common toxicities included fatigue and infusion reactions. Additionally, responses were seen regardless of PD-L1 expression or polyomavirus status.15
Although not yet approved at the writing of this article, the PD-1 inhibitor pembrolizumab has shown promise in clinical trials. A phase II study of pembrolizumab enrolled 25 patients. Patients had distant metastatic or recurrent locoregional disease that was not amenable to surgery or radiation. Patients could not have received prior systemic therapy for advanced disease. An objective response rate was 56% with 4 patients achieving a CR and 10 patients out of 25 achieving a partial response. Twelve of the 14 patients with a response had an ongoing response at follow-up with a median follow-up of 33 weeks. Toxicity was similar to other trials and disease types with pembrolizumab; 4 of 25 patients had grade 3-4 toxicity, including 1 case of myocarditis from supplement of this article.16
The approval of avelumab and promising results seen in a clinical trial with pembrolizumab will dramatically shift the way patients are treated as part of standard of care. Although it is still too early for long- term survival data, the information available demonstrates that these responses are more durable than those obtained with chemotherapy. Additionally, responses are seen in both virus positive and negative tumors with expected rates of toxicity. Although not yet approved in the first-line setting immunotherapy is poised to replace chemotherapy as standard treatment.
Adjuvant Standards and Future Directions
Surgery is the primary treatment modality for localized disease. Surgical margins of 1 to 2 cm are the standard of care for excision of the primary lesion. Sentinel lymph node biopsy (SLNB) is recommended prior to excision and is important in staging patients. One third of early stage patients have been documented to have sentinel lymph node (SLN) micrometasteses.4In patients with SLNB, the nodal basin can be observed or radiation can be considered in high-risk patients. If a positive SLN has been identified, patients are recommended to be discussed in a multidisciplinary tumor board about both nodal dissection and adjuvant radiation. MCC is notable for a high-risk of recurrence of up to 56% in SLNB-positive patients and 39% on SLNB-negative patients. Radiation is often used in the adjuvant setting to minimize locoregional recurrence. A meta-analysis showed a lowered rate of regional recurrence with radiotherapy (RT) when compared with surgery alone.17In patients with no lymph node involvement there is an improvement in OS. However, this has not been demonstrated in lymph node-positive disease.18
MCC is a chemosensitive disease in the metastatic setting. However, there are no prospective trials showing a benefit of adjuvant chemotherapy. A retrospective review of 4815 patients with MCC of the head and neck showed an improvement in survival with adjuvant RT and chemo-radiotherapy. However, adjuvant chemotherapy alone had a worse OS compared with surgery alone.19Given the retrospective nature of this paper, it is difficult to make conclusions as treating physicians likely would choose to give adjuvant chemotherapy in the younger, healthier patients. National Cancer Database review of 6908 patients did not show an improvement or worsening of OS with chemotherapy or chemoradiotherapy.18It should be noted that there are no high-quality randomized studies with chemotherapy and/or radiation in this disease. Neither has been prospectively shown to improve survival. Because of this fact, it is very difficult to know the true benefit or lack of benefit these interventions have on MCC.
Given the high risk of recurrent disease and the promise of immune checkpoint blockade in the metastatic setting, several trials are ongoing looking at both neoadjuvant and adjuvant treatment for MCC. NCT02196961 is an adjuvant trial for resected Merkel cell that administers 4 doses of ipilimumab at 3 mg/kg vs. observation. Additionally, nivolumab is being given in the neoadjuvant setting prior to resection (NCT02488759). Studies with avelumab are currently in development as well. Results are not yet available, but are anxiously awaited given the high risk of recurrence in patients.
The prognosis of patients diagnosed with MCC is currently based on TNM staging. Intratumoral CD8 positivity may play an additional role in determining prognosis at diagnosis for this immunogenic malignancy. CD8+ lymphocytes mediate antiviral immune responses. One hundred and thirty-seven MCC cases were classified based on the presence of intra-tumoral CD8. Samples were categorized as having absent, low, or moderate/strong CD8 expression. Intra-tumoral CD8+ presence was found to have a direct correlation with MCC-specific survival. A similar trend was found for OS, but was not statistically significant.20
Patients who undergo resection of MCC have a high risk of recurrence. Current surveillance recommendations are for periodic imaging with no clear scientifically defined frequency. However, in patients who have developed a positive antibody titer to the MCPyV oncoprotein, recent research has shown promise for detecting recurrence based on monitoring of the antibody titers. In a cohort of 219 prospectively followed patients, an increase in oncoprotein antibody titers were associated with an increase in disease recurrence. Among seropositive patients, an increase in oncoprotein titer had a positive predictive value of 66% while a decrease had a negative predictive value of 97%.21 This represents a promising new tool for monitoring recurrence in high-risk patients. It should be noted that is a single institution study and much additional research is needed to define the optimal screening using both a combination of imaging and antibody titers.
Although immunotherapy represents an improvement in treatment options for patients with metastatic disease, not everybody responds to therapy. Several exploratory biomarker analyses have investigated patterns of response. An exploratory biomarker analysis of the phase II trial that led to the approval of avelumab looked at the role of tumor PD-L1 expression. An ORR of 34.5% was found in PD-L1positive versus only 18.8% in PDL-1–negative patients at the 1% expression level cutoff. At the 5% expression level cutoff 52.6% of PD-L1 positive patients and 23.6% of PD-L1 negative patients had an overall response. Additionally, 26.1% of patients with MCPyV–positive tumors compared with 35.5% in MCPyV–negative tumors had a response. When high versus low CD8+ baseline T cell density at the margin was compared, the ORR was 44.4% versus 19.2% respectively.22 In the trial of patients receiving first-line treatment with pembrolizumab, 62% of patients who responded had virus-positive tumors while 44% had virus-negative tumors.16
MCC is a rare and aggressive malignancy. Patients with localized disease have a high risk of recurrence and those who develop metastatic disease have a poor OS. In recent years much has been learned about the roles that MCPyV and mutational burden play in the pathogenesis of the disease. This has translated into potential options for monitoring high-risk patients with antibody titers though further study is needed.