Palliative Systemic Treatment of Advanced Merkel Cell Carcinoma in the Pre-Immunotherapy Era: A Retrospective, Single-Center Analysis of Patients with an Orphan Neuroendocrine Malignancy

Results: MCC is a chemotherapy-sensitive tumour with an objective response rate of 63% to first-line chemotherapy. The combination of cisplatin or carboplatin with etoposide was the most frequently used regimen (n=13) with responses seen in 69% of patients. The median progression-free survival after first-line chemotherapy was 8 months. Eight patients received second line chemotherapy with gemcitabine, taxanes or vinca alkaloids with a response rate of 25%. The median overall survival since start of first line chemotherapy was 13 months. Conclusions: A considerable proportion of patients with MCC fails local treatments and requires systemic therapy. Advanced MCC is a chemotherapy-sensitive disease with high response rates. The poor overall survival achieved with chemotherapy supports the need for novel systemic strategies, such as the routine implementation of immunologic treatment approaches. Immune checkpoint modulation is complementary to chemotherapy, and should be further developed as single agent, in sequence or in combination with other biological or cytotoxic therapy.


INTRODUCTION
it gives rise to the production of oncoproteins and induces mutations [11]. Since MCC occurs predominantly in sunexposed skin in the head and neck in older individuals, UVinduced mutations may trigger viral DNA integration in the tumour genome. MCC is most commonly staged according the following classification: stage I, localized disease (IA ≤ 2 cm, IB > 2 cm); stage II, regional lymph node involvement; stage III, distant metastasis [7]. In 2010 a new staging system was proposed which divides patients into 4 groups [6,7].
Most MCC patients present with a localized tumour (Stage I).
The 5-year survival of these patients is in the range of 64% [6]. Surgery with wide local excision is the generally accepted standard treatment. Mohs micrographic surgery is an option with comparable local recurrence rates [12]. Some authors advocate postoperative radiotherapy, since MCC is known to be radiotherapy-sensitive [6]. Involvement of lymph nodes (Stage II) is associated with worse prognosis (5-year survival around 47%), since nodal spread is associated with a risk of developing distant metastasis [6]. Although only retrospective data are available and results are conflicting, a database analysis showed an improvement in overall survival with adjuvant radiotherapy compared to surgery alone [13]. Treatment standards for stage II MCC usually include surgery and radiotherapy to the affected sites [7]. Adjuvant chemotherapy is a hypothetical option for patients at high risk of recurrence, although this has never been studied prospectively and most studies were using chemo-radiation, making the effect of adjuvant chemotherapy even more difficult to assess [14]. Until recently there was no real established treatment standard for metastatic MCC. Theoretical options included chemotherapy, radiotherapy and surgery in individual cases. MCC is generally perceived by oncologists as a chemotherapysensitive neuroendocrine tumour. The reported evidence for the use of cytotoxic agents in advanced MCC is very limited, and a variety of chemotherapeutic agents are used in clinical routine. The choice of treatment is commonly extrapolated from the experience in other neuroendocrine malignancies, such as in small cell lung cancer [3,8]. The true role and efficacy of chemotherapy in MCC remains elusive due to the absence of prospective controlled trials. This lack of evidence with regard to efficacy and outcome is adding to the fact that most chemotherapy combinations currently applied for MCC can pose considerable and even irreversible toxicity to the patient.
Knowledge over the role of chemotherapy in MCC is therefore the basis for further improving treatment concepts in this rare aggressive disease.
Our report summarizes the retrospective experience of the University Hospitals in Leuven with the systemic treatment of advanced, incurable MCC over a period of 13 years prior to the immunotherapy era. The purpose of this report is to document the role of cytotoxic therapy in the absence of historical prospective evidence, and to underline the unmet medical need for novel, safe and effective treatments for this orphan disease.

After approval of this project by the independent Ethics
Committee of the University Hospitals in Leuven, we We supplemented our analysis with a literature search on this orphan malignancy. We reviewed articles published between 1990 and 2018, and discuss the emerging role of immune checkpoint modulation in this disease.     Table 1.  due to septic shock. One patient developed thoracic zoster.

Patient characteristics
One case of ototoxicity was reported, likely related to the treatment with cisplatin (Table 3).  [19]. In another retrospective study, a combination of cisplatin and doxorubicin was found to be superior to other regimens in terms of response rate [15].
A higher response rate was also seen with regimens that included 5-fluorouracil as compared to schedules without a fluoropyrimidine [15]. Especially among older patients and those with poorer performance or organ dysfunction treatment with a single agent can be considered, although possibly less effective. As an example, orally administered etoposide is well-tolerated and can be applied on an outpatient basis. Complete responses were reported in 3 out of 4 patients in a small series [23].
Our data highlight once again the "responding resistance" pattern seen in clinical routine in different neuroendocrine malignancies. While initial response rates are high, the median duration of response in our series was short and the overall survival far from being satisfactory. We observed a median progression-free survival after first-line chemotherapy of 8 months (range 1-146), which again is comparable to published results [15,17]. The median overall survival since the start of chemotherapy was 13 months. The 2-and 5-year overall survival rates were only 59% and 29%, respectively.
This matches the experience of Tai et al., [16], who reported survival of 59% at 2 years and 22% at 5 years. Chemotherapy results reported from our other series are similar to what can be achieved in other neuroendocrine carcinomas, such as small cell lung cancer or high grade gastroenteropancreatic neuroendocrine carcinomas. In these tumor types, response rates vary between 33 and 67% [16] and progressionfree survival ranges between 1 and 21 months [24]. Our monocentric experience provides supportive evidence for the effectiveness of chemotherapy in the treatment of MCC, but also highlights the short duration of clinical benefit in this rare disease.
In receptor, platelet derived growth factor receptor and KIT, which all belong to the same family of tyrosine kinase receptors [8,25].
In a clinical trial with pazopanib, a multi-targeted tyrosine kinase inhibitor, a patient with metastatic MCC showed an impressive tumor regression, but progression-free survival was short [18]. A trial with imatinib in MCC had to be discontinued due to the lack of activity [26-28], even though some case reports suggested therapeutic effects [29][30][31]. clinical presentation, prognosis and management of this uncommon cancer [33].

Conclusion and Outlook
MCC is an aggressive neuroendocrine tumor of the skin with  [35]. Remarkable responses have been observed with PD-1 antibodies as well [36]. A phase II trial with pembrolizumab as first line treatment showed a response rate of 56%, with a PFS beyond 6 months [37].