DOI: 10.1111/jdv.18256 JEADV ORIGINAL ARTICLE Clinicopathological features, MCPyV status and outcomes of Merkel cell carcinoma in solid-organ transplant recipients: a retrospective, multicentre cohort study ndiz-Pulido,1,* C. Ferra A. Go mez-Toma s,1 B. Llombart,2 D. Mendoza,3 J. Marcoval,4 S. Piaserico,5 C. Baykal,6 J.N. Bouwes-Bavinck,7 E. Racz,8 J. Kanitakis,9 C.A. Harwood,10 P. Cetkovska, 11 A. Geusau, 12 13 V.del Marmol, E. Masferrer, 14 C. Orte Cano, 13 J. Ricar,11 15 W.R.de Oliveira, R. Salido-Vallejo, 16 9 10 E. Ducroux, M.A. Gkini, J.A. Lo pez-Guerrero, 17,18,19 H. Kutzner,20 W. Kempf,21 D. Sec ß kin22 1 Department of Dermatology, Hospital Universitari Vall d’Hebron, Universitat Auto noma de Barcelona, Barcelona, Spain 2 Servicio de Dermatologıa, Instituto Valenciano de Oncologıa, Valencia, Spain 3 Department of Dermatology, Fundacio n Jimenez Dıaz, Madrid, Spain 4 Department of Dermatology, Hospital de Bellvitge, IDIBELL, University of Barcelona, Barcelona, Spain 5 Dermatology Unit, Department of Medicine, Universita  di Padova, Padova, Italy 6 Department of Dermatology, Istanbul University, Istanbul Medical Faculty, Istanbul, Turkey 7 Department of Dermatology, Leiden University Medical Center, Leiden, The Netherlands 8 Department of Dermatology, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands 9 Department of Dermatology, Edouard Herriot Hospital Group, Hospices Civils de Lyon, Lyon, France 10 Centre for Cell Biology and Cutaneous Research, Blizard Institute, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London, UK 11 Department of Dermatovenereology, Faculty of Medicine, Charles University, Pilsen, The Czech Republic 12 Department of Dermatology, Medical University of Vienna, Vienna, Austria 13 Service de Dermatologie, Ho ^pital Erasme, Universite Libre de Bruxelles, Brussels, Belgium 14 Department of Dermatology, Hospital Universitari Mu tua de Terrassa, Barcelona, Spain 15 Department of Dermatology, University of Sa ~o Paulo, Sa ~o Paul, Brazil 16 Department of Dermatology, University Clinic of Navarra, School of Medicine, University of Navarra, Pamplona, Spain 17 Laboratory of Molecular Biology, Fundacio n Instituto Valenciano de Oncologıa, Valencia, Spain 18 IVO-CIPF Joint Research Unit of Cancer, Prıncipe Felipe Research Center (CIPF), Valencia, Spain 19 Department of Pathology, School of Medicine, Catholic University of Valencia ‘San Vicente Martir’, Valencia, Spain 20 Dermatopathologie, Friedrichshafen, Germany 21 Kempf und Pfaltz Histologische Diagnostik and Department of Dermatology, University Hospital Zurich, Zu €rich, Switzerland 22 Department of Dermatology, Basßkent University Faculty of Medicine, Ankara, Turkey *Correspondence: C. Ferrandiz-Pulido. E-mail: caferrandiz@vhebron.net Abstract Background The proportion of Merkel cell carcinomas (MCCs) in solid-organ transplant recipients (SOTR) harbouring Merkel cell polyomavirus (MCPyV) is unknown, as are factors affecting their outcomes. Objective To describe clinicopathological features of MCC in SOTR, investigate the tumoral MCPyV-status and iden- tify factors associated with tumour outcomes. Methods Retrospective, international, cohort-study. MCPyV-status was investigated by immunohistochemistry and polymerase chain reaction. Results A total of 30 SOTR and 44 consecutive immunocompetent patients with MCC were enrolled. SOTR were younger at diagnosis (69 vs. 78 years, P < 0.001). Thirty-three percent of SOTR MCCs were MCPyV-positive vs. 91% of immunocompetent MCCs (P = 0.001). Solid-organ transplantation was associated with an increased cumulative inci- dence of progression (SHR: 3.35 [1.57–7.14], P = 0.002), MCC-specific mortality (SHR: 2.55 [1.07–6.06], P = 0.034) and overall mortality (HR: 3.26 [1.54–6.9], P = 0.002). MCPyV-positivity and switching to an mTOR inhibitor (mTORi) after MCC diagnosis were associated with an increased incidence of progression (SHR: 4.3 [1.5–13], P = 0.008 and SHR: 3.6 [1.1–12], P = 0.032 respectively) in SOTR. Limitations Retrospective design and heterogeneity of SOTR cohort. JEADV 2022 Ó 2022 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. 2 ndiz-Pulido et al. Ferra Conclusions MCPyV appears to play a less prominent role in the aetiopathogenesis of MCC in SOTR. SOTR have a worse prognosis than their immunocompetent counterparts and switching to an mTORi after the diagnosis of MCC does not improve progression. Received: 19 December 2021; Accepted: 13 April 2022 Conflicts of interest ndiz-Pulido has received consulting fees from Almirall and Sanofi, honoraria for lectures or educational events Carla Ferra from Almirall, Isdin, Sanofi, Novartis, Leo-Pharma and Sunpharma, and support for attending meetings from Galderma, Isdin, Almirall, Sanofi and Sunpharma. Alvaro Go  mez-Toma s has received support for attending meetings and/or travel from Pierre Fabre, Leo-Pharma, AbbVie, Cantabria Labs and Janssen. Beatriz Llombart has received honoraria for lectures or educational events from Roche, Sanofi and Sunpharma. Stefano Piaserico received consultation fees from Abbie, Almirall, Celgene, Janssen, LeoPharma, Eli Lilly, Novartis, Sandoz and UCB as a speaker and/or participant in advisory boards. Jan-Nico Bouwes-Bavinck has received a grant from the EADV. Catherine Harwood has received grants or contracts from Barts Charity, Medical Research Council, UK and British Skin Foundation, consulting fees from AMLo Biosciences, Leo Pharma, Sanofi, L’Oreal, Almirall and Incanthera, honoraria for lectures from Sanofi and Merck, support for attending meetings from Pellepharm and received equipment materials from MEDA. Petra Cetkovska has received consulting fees from Abbvie, UCB, Lilly and Pfizer and honoraria for lectures from Abbvie, Sanofi, Novartis, Leo-Pharma. Alexandra Geusau has received grants from Jubilee Grant of City of Vienna and Buergermeister Fond of the City of Vienna. Emili Masferrer has received honoraria for lectures or educational events from Isdin and Sunpharma and support for attending meetings and/or travel from Novartis, Vin ~as, Galenicum and Cantabria. Rafael Salido-Vallejo has received honoraria for lectures or educational events from Almirall, Biofrontera, Leo-Pharma, SunPharma and Roche,  Antonio and support for attending meetings and/or travel from Almirall, Biofrontera, Leo-Pharma, Isdin and Avene. Jose   ~ Lopez-Guerrero has received grants from Generalitat Valenciana, EU-H2020 and Asociacion Espanola de Urologia, consulting fees from AstraZeneca-MSD, honoraria for lectures from AstraZeneka-MSD and GSK and has the patent Software Mamapred (2107208395155). Werner Kempf has received honoraria for lectures, from Takeda and Stemline. The other authors have no conflict of interest to declare. Funding source None Introduction prevalence in tumours from SOTR and immunocompetent Merkel cell carcinoma (MCC) is a rare, aggressive neuroen- patients, and evaluate the role of SOT in the risk of progression, docrine cutaneous cancer with a high propensity for recurrence, MCC-specific and overall mortality. metastasis and mortality.1,2 Risk factors include chronic ultravi- olet radiation (UVR) exposure, age over 50 years and immuno- Methods suppression, including HIV/AIDS, haematological malignancies, autoimmune diseases and iatrogenic immunosuppression as Study design and population after solid organ transplantation (SOT).1,3–5 This was a retrospective, international, multicentre cohort study The link between post-transplant immunosuppression and involving 13 European (within the ‘Skin Care in Organ Transplant MCC was first reported in 1999.6 The initial solid-organ trans- Patients, Europe’ [SCOPE] Network) and one Brazilian (within the plant recipients (SOTR) with MCC were younger, and outcomes ‘International Transplant Skin Cancer Collaborative’ [ITSCC]) spe- were worse, compared with the general population,6 findings cialist SOTR dermatology clinics. All SOTR diagnosed with MCC that were subsequently confirmed.3,7–10 Risk estimates for post- at the collaborating centres between 1990 and 2019, and 44 consec- SOT MCC incidence vary; in the largest series to date, a 23.8- utive immunocompetent patients from two Spanish Oncoderma- fold increased risk was reported.3 Merkel cell polyomavirus tology centres between 2002 and 2018 were included. (MCPyV) is detected in 60%–80% of MCCs in the general pop- In all cases, immunohistopathological material was reviewed ulation.11–15 The presence of MCPyV in MCC arising in SOTR by experienced dermatopathologists. Diagnosis of MCC required has, however, not been well studied.8,14,16,17 the presence of dermal and/or subcutaneous proliferation of We aimed to characterize the clinicopathological features of small, basophilic cells expressing keratins (especially keratin 20) MCC in SOTR, identify prognostic factors, compare MCPyV and neuroendocrine markers (Fig. S1A–D). MCCs were JEADV 2022 Ó 2022 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology. Merkel cell carcinoma and transplantation 3 classified by histological subtypes, namely intermediate, small- P-values <0.05 were considered statistically significant. All cell and trabecular.18,19 tests were two-tailed. Statistical analyses were performed using R Clinical data included sex, age, race, skin phototype, skin can- version 3.6.1 (R Core Team, Vienna, Austria, 2019). cer history, immunosuppressive regimen, location, and diameter This study was approved by the local ethics committee in Bar- of MCC, stage at diagnosis, treatment modality, and outcome celona, Spain (PR(AG)274/2018) and conducted in accordance including progression, date and cause of death or disease status with the guidelines of the Helsinki Declaration of 1975, as at last follow-up. revised in 1983. All patients were restaged according to the American Joint Committee on Cancer staging system 7th ed.20 Progression was Results defined as recurrence in patients disease-free after treatment, increased tumour burden during treatment or MCC-specific Baseline characteristics of the SOTR cohort death. A total of 30 SOTR developing MCC after transplantation were identified (10 from Spain, 4 each from the Netherlands and Italy, MCPyV status 3 each from Turkey and France, 2 from the United Kingdom Available biopsy specimens obtained and handled in a similar and 1 each from Brazil, Czech Republic, Austria, and Belgium). manner in both cohorts were analysed for MCPyV status. SOTR were transplanted between 1974–2014; 63% were kidney Immunohistochemistry (IHC) was performed using the CM2B4 transplants. The median duration of immunosuppression before antibody (Fig. S1)21 and polymerase chain reaction (PCR) using MCC diagnosis was 8 years (range, 6–16) (Table 1). Thirty per- the primer sets LT1, LT3 and VP1.12 Protocols are provided in cent of SOTR were receiving an mTOR inhibitor (mTORi) at Supplemental Material. Tumour MCPyV status was considered the time of MCC diagnosis: compared with SOTR not receiving positive if at least one of these two tests was positive. Tumour mTORi at diagnosis, these patients had undergone more than cell morphology was also assessed and correlated with MCPyV one SOT (50% vs. 10.5%, P = 0.044), had a previous history of status. keratinocyte carcinoma (KC) (100% vs. 53%, P = 0.026) and a tendency towards a higher TNM stage (Table S1). Statistical analysis Ulceration and bleeding were present in 6/21 (29%) and 3/21 Descriptive and univariate statistics were computed as custom- (14%) of tumours with available information respectively ary.22 Clinical and demographic characteristics of the two (Fig. S2). Histologically, the small-cell subtype was predominant cohorts were compared using the Fisher’s exact test and the Wil- (9/14, 64%), followed by trabecular (3/14, 21%) and intermedi- coxon rank-sum test, as appropriate, unless stated otherwise. ate subtypes (2/14, 14%) (Table 1). Disease-progression and disease-specific mortality risks were evaluated with multivariable competing risk regression analysis. Comparison of demographic, clinical and treatment Disease progression and deaths from MCC were considered as characteristics events of interest, while deaths from other causes as competing SOTR were significantly younger than immunocompetent events. Surviving patients were censored on the date of last patients (median age at diagnosis: 69 and 78 years, respectively, follow-up. Fine & Gray’s subdistribution hazards model (Sub- P = 0.001) (Table 2). Men accounted for 70% of SOTR vs. 36% Hazard Ratio [SHR] [95% CI]) was used to determine the effect of the immunocompetent patients, (P = 0.009). Although most of SOT and selected covariates on tumour progression and MCCs were located on the head and neck in both groups, trun- MCC-specific mortality rates. Cumulative incidence functions cal location was more frequent in SOTR (P = 0.044). Tumour (CIF) were used to graphically assess the cumulative incidence of diameter was smaller in SOTR (16 vs. 25 mm, P = 0.028), with outcomes of interest in the presence of competing events, and more SOTR being diagnosed with T1 tumours. Gray’s test to assess differences in CIF between groups. Most patients were treated surgically. Sentinel lymph node Cox proportional-hazards models (Hazard Ratio [HR] [95% biopsy (SLNB) was performed in 11 SOTR (39%) and 10 CI]) were used to estimate the risk of overall mortality when the immunocompetent patients (24%) (P = 0.191), with similar proportional hazards assumption was met. Kaplan–Meier curves rates of positivity. Adjuvant radiotherapy was given more often were generated to compare survival between groups, and log- to immunocompetent patients (70% vs. 23%, respectively, rank tests were used to assess differences between groups. Age, P < 0.001). There was no difference in use of conventional sex, MCPyV status, stage at presentation and year at diagnosis chemotherapy (immunotherapy was not available for metastatic were included in all models featuring both cohorts. MCC during the study period). After MCC diagnosis, immuno- Fine & Gray models were used to assess the impact of differ- suppression was tapered or discontinued in nine (33%) SOTR ent demographic, clinical, histological and treatment variables and five (19%) were switched to an mTORi. on the probability of progression in SOTR adjusting by age, sex, Compared with immunocompetent patients, SOTR had year of diagnosis and stage at diagnosis. higher rates of local recurrence (39% vs. 11%), lymph-node JEADV 2022 Ó 2022 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology. 4 ndiz-Pulido et al. Ferra Table 1 Clinical and histological characteristics of SOTR with involvement (59% vs. 23%), distant metastasis (48% vs. 21%) posttransplant MCC and overall mortality (86% vs. 57%). There were no significant Clinical characteristics no./total no. (%) differences in overall mortality between allograft types. Allograft type Kidney 19/30 (63) MCPyV status Heart 5/30 (17) MCPyV was detected in 5/15 (33%, 95% CI [15%–58%]) post- Liver 3/30 (10) SOT MCCs vs. 21/23 (91%, 95% CI [73%–98%]) MCCs from Kidney and pancreas 2/30 (7) immunocompetent patients (P = 0.001) (Table 2). Eight PCR- Lung 1/30 (3) positive MCCs were MCPyV-negative by IHC (two in SOTR IS drugs taken at diagnosis and six in the immunocompetent cohort); conversely, one PCR- Tacrolimus 10/27 (37) negative (immunocompetent) MCC was IHC-positive. The cel- Cyclosporine 10/27 (37) lular morphology observed in H&E histological sections was sig- Mycophenolate Mofetil 6/27 (22) nificantly associated with MCPyV-status: MCPyV-negative Azathioprine 7/27 (26) MCCs had more pleomorphic and irregular nuclei (Odds ratio Prednisone 20/27 (74) (OR): 31.17, 95% CI [3.29–295.36], P = 0.0027) and more Sirolimus 7/27(26) abundant cytoplasm (OR: 18.0, 95% CI [2.94–110.31], Everolimus 1/27 (4) P = 0.0018) compared with MCPyV-positive MCC, which had Number of IS drugs at MCC diagnosis mostly monomorphous round nuclei and scant cytoplasm One 2/27 (7) (Table S2). The delay from the time of biopsy to MCPyV analy- Two 17/27 (63) sis measured in years was not associated with MCPyV-status Three 8/27 (30) (OR: 1.03, 95% CI [0.89–1.18], P = 0.65) nor was patient’s age Caucasian 30/30 (100) at diagnosis (OR: 1.01, 95% CI [0.96–1.08], P = 0.56) or coun- Fitzpatrick skin phototype† try of origin (Spain vs. others). MCPyV status by clinical and I 1/23 (4) demographic characteristics is shown in Table S2. II 11/23 (48) III 11/23 (48) Survival analysis Previous skin cancer Seventy-two patients with available data contributed 264.25 Keratinocyte carcinoma (SCC or BCC) 18/26 (69) person-years of follow-up. At the end of the study, 4 (14%) Cutaneous SCC 17/26 (65) SOTR and 19 (43%) immunocompetent patients were alive. BCC 10/26 (39) Progression and MCC-specific mortality usually occurred History of previous internal malignancy 4/26 (15) within 2.5 years of diagnosis in both cohorts, but all three Concomitant cutaneous neoplasm at the time of MCC diagnosis main outcome measures occurred earlier in SOTR (Fig. 1, Cutaneous SCC 5/26 (19) Table 3). BCC 2/26 (8) Figure 1 shows multivariable-adjusted SHRs and HRs for No 19/26 (73) MCC risk factors. SOT had a strong effect on the cumulative Histological features incidence of progression (SHR: 3.35 [1.57–7.14], P = 0.002), Histological classification even after adjusting for TNM stage. SOT also had a strong effect Small-cell 9/14 (64) on the cumulative incidence of MCC-specific mortality (SHR: Trabecular 3/14 (21) 2.55 [1.07–6.06], P = 0.034), as did disease stage III/IV at diag- Intermediate 2/14 (14) nosis. In adjusted analyses, SOT remained significantly associ- Tumour growth pattern ated with increased overall mortality (HR: 3.26 [1.54–6.9], Infiltrative 8/18 (44) P = 0.002), as did male sex, age > 75 years, and disease stage IV Nodular 10/18 (56) at presentation. MCPyV status was not significantly associated Mitotic index – high 14/14 (100) with any of these outcomes. Lymphovascular invasion 7/19 (37) To assess a possible source of confounding bias by country Perineural invasion 1/17 (6) of origin, we initially adjusted for Spanish origin. However, this Epidermotropism 3/18 (17) variable was non-significant, suggesting that immune status, Collision with SCC 3/30 (10) not country of origin, was relevant for MCC outcomes and SOTR, solid organ transplant recipients; IS, immunosuppressive; MCC, Merkel MCPyV-status (data not shown). Moreover, comparing the cell carcinoma; SCC, squamous cell carcinoma; BCC, basal cell carcinoma. management between SOTR and immunocompetent patients Differences in total number of cases for different characteristics are due to missing data. disclosed a lower use of adjuvant radiotherapy in SOTR. We †Determined by the treating dermatologist. found a positive association, albeit statistically not significant, JEADV 2022 Ó 2022 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology. Merkel cell carcinoma and transplantation 5 Table 2 Comparison of demographic and clinical characteristics of SOTR and immunocompetent patients with MCC Demographics SOTR Immunocompetent P-value patients Follow-up period, median (IQR), months 11 (7–30) 40 (14–80) <0.001*** Age at diagnosis of MCC, median (IQR), years 69 (61–76) 78 (71–83) 0.001** Male sex, no./total no. (%) 21/30 (70) 16/44 (36) 0.009** Tumour location, no./total no. (%) 0.011* Head and neck 18/30 (60) 23/44 (52) 1.00 Trunk 5/30 (17) 0 (0) 0.044* Upper limbs 5/30 (17) 8/44 (18) 1.00 Lower limbs 2/30 (7) 11/44 (25) 0.245 Unknown primary 0/30 (0) 2/44 (5) 1.00 Tumour diameter, mm Median (range) 16 (5–40) 25 (5–100) 0.028* Quartiles (Q1-Q3) 11–21 15–30 Staging AJCC 7th ed. T, no./total no. (%) 0.005** T1 19/29 (66) 14/44 (32) 0.057 T2 5/29 (17) 20/44 (46) 0.063 T3 0 (0) 4/44 (9) 0.57 T4 4/29 (14) 2/44 (4) 0.64 Tx 1/29 (3) 4/44 (9) 0.81 N, no./total no. (%) <0.001*** N0 11/29 (38) 28/44 (64) 0.2 N1a 2/29 (7) 6/44 (14) 0.38 N1b 7/29 (24) 4/44 (9) 0.48 N2 2/29 (7) 5/44 (11) 0.81 Nx 7/29 (24) 1/44 (2) 0.04* M, no./total no. (%) 0.551 M0 26/29 (90) 36/44 (82) M1 2/29 (7) 7/44 (16) Mx 1/29 (3) 1/44 (2) Stage at diagnosis, no./total no. (%) 0.073 I 12/28 (43) 12/43 (28) II 3/28 (11) 15/43 (35) III 10/28 (36) 9/43 (21) IV 3/28 (11) 7/43 (16) MCPyV-status, no./total no. (%) By immunohistochemistry 3/15 (20) 15/23 (65) By polymerase-chain reaction 2/7 (29) 19/22 (86) Total 5/15 (33) 21/23 (91) 0.001** Treatment Surgery, no./total no. (%) 29/30 (97) 39/44 (89) 0.391 SLNB performed, no./total no. (%) 11/28 (39) 10/42 (24) 0.191 SLNB positivity, no./total no. (%) 4/11 (36) 3/10 (30) 0.416 Adjuvant radiation therapy, no./total no. (%) 7/30 (23) 28/40 (70) <0.001*** Lymph node dissection, no./total no. (%) 7/28 (25) 12/42 (29) 0.790 Radiation therapy, no./total no. (%) 11/27 (41) 7/41 (17) 0.077 Chemotherapy, no./total no. (%) 6/30 (20) 11/41 (27) 0.508 Reduction of IS, no./total no. (%) 9/27 (33) not applicable Switch to mTORi after diagnosis, no./total no. (%) 5/27 (19) not applicable Received acitretin after diagnosis, no./total no. (%) 4/27 (15) not applicable JEADV 2022 Ó 2022 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology. 6 ndiz-Pulido et al. Ferra Table 2 Continued Demographics SOTR Immunocompetent P-value patients Observed outcomes Disease progression, no./total no. (%) Overall progression 19/28 (68) 19/44 (43) 0.054 Local recurrence 11/28 (39) 5/44 (11) 0.007** Lymph node involvement 16/27 (59) 10/43 (23) 0.004** Solid organ metastasis 13/27 (48) 9/44 (21) 0.005** Mortality, no./total no. (%) Overall mortality 24/28 (86) 25/44 (57) 0.01* MCC-specific mortality 15/28 (54) 14/44 (32) 0.067 Death by allograft type, no./total no. (%) not applicable 0.432† Kidney 15/17 (88) Kidney and pancreas 2/2 (100) Liver 2/3(67) Heart 4/5 (80) Lung 1/1 (100) SOTR, solid organ transplant recipients; MCC, Merkel cell carcinoma; MCPyV, Merkel cell polyomavirus; IQR, interquartile range; AJCC, American Joint Committee on Cancer; SLNB, sentinel lymph node biopsy; mTORi, mTOR inhibitor; IS, immunosuppression. P-values from Wilcoxon rank-sum test for continuous variables and from Fisher’s exact test for categorical variables. When the omnibus test was statistically significant, post-hoc tests were carried out adjusting for multiple comparisons using Holm’s method. Differences in total number of cases for different clinical, therapeutic or outcome characteristics are due to missing data. †P-value from Fisher’s exact test for count data for SOTR and death counts by allograft type. P < 0.1, *P < 0.05, **P < 0.01, ***P < 0.001. between adjuvant radiotherapy and MCC progression after Discussion adjusting for age, sex and year of diagnosis in both IC and To our knowledge, this is the largest study evaluating MCC out- SOTR cohorts (SHR: 1.63 [0.54–4.89], P = 0.38 and 1.93 comes, prognostic factors and MCPyV status in SOTR compared [0.71–5.23], P = 0.2 respectively). After further adjusting for with immunocompetent patients. the TNM stage, this association was almost null in IC (SHR: The age at diagnosis was lower in SOTR, suggesting that iatro- 1.04, P = 0.94) and was attenuated in SOTR (SHR:1.7 [0.48– genic immunosuppression favours an earlier onset of 6.4], P = 0.4). MCC.3,6,23,24 Even though male sex has been reported as an inde- pendent risk factor for developing MCC,25 there was a female pre- Impact of histopathological subtype, viral status and dominance in our immunocompetent cohort. This finding is treatment characteristics on progression of MCC in SOTR consistent with other European studies,26,27 and is explained by Disease progression occurred in 19/28 (68%) SOTR in whom different age-adjusted rates between sexes.26 There were more relevant information was available. In multivariate analysis, males in our SOTR-MCC cohort, consistent with the male pre- MCPyV positivity (SHR: 4.3 [1.5–13], P = 0.008) and non- dominance in the general SOTR population.3,7,24,28 Head and tumour-free margins after tumour resection (SHR: 23 [2.1–250], neck location was common in both groups, consistent with an P = 0.01) were associated with an increased risk of progression. aetiological role of UVR.18,29 The median interval between the The small-cell histological subtype was associated with a lower first SOT and MCC diagnosis was 8 years, similar to previous risk of progression (SHR: 0.06 [0.01–0.33], P = 0.001) reports.6,28,30 Although the clinical significance of histological sub- (Table 4). Reduction of immunosuppression did not affect the typing is uncertain,18 the small-cell subtype predominated in risk of progression but switching to an mTORi (SHR: 3.6 [1.1– SOTR and was associated with a lower risk of progression. 12], P = 0.032) after diagnosis was associated with an increased No treatment guidelines exist specifically for post-SOT MCC, risk of progression. and management usually follows that recommended for the gen- Patients receiving acitretin chemoprevention either before eral population.31 Minimization of immunosuppression is and/or after diagnosis (4/27, 15%) had a lower cumulative inci- advised whenever possible,32 but the role of switching to an dence of progression (SHR: 0.029 [0.0005–0.19], P < 0.001). mTORi is unknown. We found an increased incidence of pro- However, 50% of them had a competing event (death by other gression in SOTR switched to an mTORi after MCC diagnosis, cause), making progression less likely to be observed. similar to recent observations in post-SOT Kaposi sarcoma.33 JEADV 2022 Ó 2022 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology. Merkel cell carcinoma and transplantation 7 Figure 1 (a) Cumulative incidence functions (CIF) from a competing-risks model exploring Merkel cell carcinoma (MCC) progression and MCC-specific mortality by immune status (SOTR, solid organ transplant recipient; IC; immunocompetent). P-value from Gray’s test for equality of CIFs across groups. Kaplan–Meier plot for overall-cause mortality by immune status. P-value from log-rank test; (b) Cause- specific subhazard ratios (SHR) and their 95% confidence intervals (CI) from a multivariate Fine & Gray’s subdistribution of hazards model for MCC progression and MCC-specific mortality, and from a multivariate Cox proportional hazard (HR) model for overall mortality. All models are further adjusted by year of diagnosis. This association persisted even when we minimized the risk of MCC progression, but there was certainly no evidence that it confounding by indication by adjusting for TNM stage at diag- reduced this progression. Further studies are needed to assess nosis (although patients switched to an mTORi did not signifi- the effects of first-generation mTORi in post-SOT MCC. cantly have more advanced tumours). In terms of a possible Comparing MCC management between SOTR and immuno- explanation for the negative effect of mTORi switch, although competent patients, we found a lower use of adjuvant radiother- activation of the PI3K/Akt/mTOR pathway occurs frequently in apy in SOTR. This may reflect differences in national practices, MCCs,33 first-generation mTORi (such as rapamycin) does not concerns regarding second malignancies post-radiotherapy in suppress MCC-cell growth in vitro.34 In contrast, agents target- the context of pre-existing field cancerization in SOTRs and – ing both mTOR complex (mTORC)1 and mTORC2 seem to be probably more importantly – an era effect, in that SOTR were promising for treatment.35 Due to the observational nature of diagnosed over a longer time span during which adjuvant radio- our study, we cannot be certain that conversion to a first- therapy was probably less widely used. Moreover, we did not generation mTORi after diagnosis per se increased the risk of observe the known positive impact of adjuvant radiotherapy on JEADV 2022 Ó 2022 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology. 8 ndiz-Pulido et al. Ferra Table 3 MCC progression, MCC-specific and overall mortality rates and their 95% confidence intervals at 1, 2 and 5 years in SOTR and immunocompetent patients 1-year 2-year 5-year P-value† Progression rate 0.002** Immunocompetent patients 37% (23–51) 45% (29–59) 45% (29–59) SOTR 64% (43–79) 68% (47–82) 68% (47–82) MCC-specific mortality 0.05 Immunocompetent patients 18% (8–31) 26% (14–40) 33% (29–59) SOTR 34% (17–52) 50% (29–67) 58% (36–75) Overall mortality <0.001*** Immunocompetent patients 20% (8–32) 30% (15–43) 48% (30–62) SOTR 48% (25–64) 68% (44–82) 84% (61–94) SOTR, solid organ transplant recipients; MCC, Merkel cell carcinoma. †P-value for progression rate and MCC-specific mortality from Gray’s test for equality of cumulative incidence functions across groups and P-value for overall mortality from log-rank test. P < 0.1, *P < 0.05, **P < 0.01, ***P < 0.001. the risk of progression.36 This finding could be explained by abundant cytoplasm, a finding usually linked to MCPyV nega- indication bias as radiotherapy was more frequently prescribed tive tumours and more complex genetic aberrations.43–45 Con- in advanced stages vs. localized stages in our patients. This could sistent with this possibility, we observed that all virus-negative have biased our results towards the null, not allowing us to MCCs arose in patients with previous KCs, who were therefore observe a beneficial effect of adjuvant radiotherapy in MCC. The likely to have high cumulative sun-exposure and UV-mutational lower rates of SLNB compared with current practice is also likely burden. These results reinforce the need for close surveillance to reflect an era effect. with radiological imaging in SOTR with MCC as determination SOTR with MCC had higher rates of progression and MCC- of oncoprotein antibody titre may be less useful in the clinical specific and overall mortality compared with immunocompetent follow-up of this population.46 Although MCPyV-negative MCCs patients, despite being younger and having smaller tumours at are reportedly more aggressive,14 we found that virus-positive diagnosis. Although we noticed a trend towards higher MCC- SOTR-MCCs had an increased risk of progression. This observa- specific mortality in men, this did not reach statistical signifi- tion, together with the higher risk of progression among SOTR cance, as recently reported.37 Our findings are consistent with on mTORi after diagnosis, could also be partly explained by the those of prior single-institution registry studies.2,7,28,38–41 A pre- fact that mTORi promotes MCPyV replication by inhibiting the vious study28 including 8 SOTR and 89 immunocompetent Skp2 E3 ligase, which targets phospho-serine 220 on the MCPyV patients with MCC also concluded that SOTR were younger at Large-T molecule, and increases the expression of Large-T pro- diagnosis and had increased (4.1-fold) hazards for progression, teins.47 Moreover, a recent publication also showed that mTORi MCC-specific mortality (11.9-fold) and all-cause mortality increases the expression of Large-T proteins for MCPyV and that (10.5-fold).28 A more recent study reported a 25% 5-year sur- these drugs are highly activated in a dose-dependent fashion at vival for SOTR-MCC,7 compared with 16% in our study. Pro- therapeutic levels for increasing MCPyV DNA replication in gression in our SOTR occurred mainly during the first 2 years SOTR.48 Further investigation is needed to clarify the impact of after diagnosis, highlighting the importance of intense surveil- viral status on the risk of progression and survival in SOTR.49 lance during this period. The main limitations of our study are the retrospective design MCPyV is detected in up to 80% of MCCs.11–15,42 Virus- and the heterogeneity of the SOTR group, with cases provided negative tumours usually harbour prominent UVR-signature by 14 institutions across many years and countries, and the use mutations and chromosomal aberrations.18 Viral carcinogenesis of a Spanish-only immunocompetent cohort which, however, is common in immunosuppression-associated cancers, and we enabled us to include consecutive cases, thereby avoiding selec- therefore anticipated a high rate of virus positivity in SOTR- tion bias. The recruitment periods for the SOTR and immuno- MCC.32 Surprisingly, only 33% of SOTR-MCC harboured competent group differed. However, there was a substantial detectable MCPyV, a rate significantly lower than in MCC from overlap between immunocompetent and SOTR groups regard- immunocompetent counterparts. A recent study also reported ing year of diagnosis (Fig. S3). Nonetheless, we adjusted for the this unexpected finding.17 These data are consistent with UVR variable ‘year of diagnosis’ in all analyses to minimize the possi- playing a more important role than MCPyV in the aetiopatho- bility of confounding bias due to an era effect. Different genesis of SOTR-MCC and are also supported by the fact that immunosuppressive regimens and MCC treatments in SOTR SOTR-MCC tumours had mostly pleomorphic nuclei and across centres might have also affected MCC outcomes. Tumour JEADV 2022 Ó 2022 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology. Merkel cell carcinoma and transplantation 9 Table 4 Subdistribution hazard ratios (SHR) (and 95% confidence intervals [CI]) from Fine & Gray’s subdistribution hazard model for MCC progression in SOTR Crude model† Model 1‡ Model 2§ SHR 95% CI P-value SHR 95% CI P- value SHR 95% CI P-value Demographics Previous history of KC 1.18 (0.5–3) 0.730 1.14 (0.5–2.9) 0.780 0.67 (0.21–2.1) 0.500 Cancer history 2.77 (1.2–6.3) 0.016* 2.45 (1.03–5.8) 0.042* 1.7 (0.45–6.2) 0.450 Number of transplants 1.57 (0.6–3.8) 0.320 1.69 (0.7–4.3) 0.270 1.4 (0.55–3.8) 0.450 Type of transplant (thoracic vs. abdominal) 0.74 (0.3–1.8) 0.500 0.88 (0.32–2.3) 0.800 1.4 (0.49–3.9) 0.530 Time since transplantation (>5 years) 2.1 (0.5–10) 0.34 2.1 (0.25–17) 0.5 2.1 (0.42–10) 0.370 Tumour characteristics Location: upper limbs (vs. trunk) 0.42 (0–4.4) 0.470 0.46 (0–5.8) 0.550 0.1 (0.002–6.8) 0.280 Location: head & neck (vs. trunk) 2.90 (0.9–9.5) 0.079 3.60 (0.7–17.5) 0.110 2.4 (0.66–8.6) 0.190 Location: lower limbs (vs. trunk) 8.39 (2–34.4) 0.003** 7.30 (1.5–34.7) 0.012* 0.89 (0.06–13) 0.930 Tumour diameter: 0–10 mm Ref. - - Ref. - - Ref. - - Tumour diameter: 11–20 mm 2.18 (0.6–7.8) 0.230 2.60 (0.6–12.1) 0.230 0.89 (0.14–5.8) 0.900 Tumour diameter: 21–30 mm 1.62 (0.3–8.1) 0.550 1.35 (0.3–7.2) 0.730 0.15 (0.004–5.2) 0.290 Tumour diameter: >31 mm 6.67 (1.7–26.8) 0.007** 7.87 (1.2–53.2) 0.034* 0.41 (0.01–14) 0.620 Painful tumour 0.91 (0.3–2.8) 0.870 1.32 (0.4–4.7) 0.670 0.24 (0.05–1.3) 0.091 Tumour growth pattern (vs. infiltrative) 2.2 (0.69–7.2) 0.180 1.4 (0.31–6.6) 0.650 2.9 (0.37–23) 0.310 Small-cell (vs. non small-cell tumour) 0.73 (0.18–3) 0.670 0.38 (0.1–1.4) 0.150 0.06 (0.01–0.33) 0.001** Positive MCPyV status (vs. negative) 2.44 (0.8–7.2) 0.110 3.79 (0.9–16.3) 0.074 4.3 (1.5–13) 0.008** Immunosuppressants at diagnosis Tacrolimus at diagnosis 2.32 (0.9–6.1) 0.088 2.03 (0.7–5.9) 0.190 3.5 (0.95–13) 0.059 Cyclosporine at diagnosis 0.35 (0.1–0.95) 0.040* 0.35 (0.1–1.2) 0.087 0.60 (0.2–2.4) 0.470 Azathioprine at diagnosis 1.04 (0.4–2.5) 0.940 1.40 (0.5–4.4) 0.560 0.95 (0.2–4.1) 0.940 Mycophenolate mofetil at diagnosis 0.91 (0.4–2.4) 0.850 0.78 (0.3–2.1) 0.620 0.75 (0.2–2.3) 0.610 Prednisone at diagnosis 1.15 (0.4–3.5) 0.800 1.19 (0.4–3.7) 0.760 0.43 (0.1–1.4) 0.170 mTORi at diagnosis 1.70 (0.6–4.5) 0.290 1.47 (0.5–4.2) 0.480 1.20 (0.4–3.7) 0.750 Number of immunosuppressants at diagnosis (2 vs. 1) 1.30 (0.14,12) 0.820 0.91 (0.059,14) 0.950 0.43 (0.0039,48) 0.730 Number of immunosuppressants at diagnosis (3 vs. 1) 1.40 (0.15,12) 0.790 1.20 (0.075,18) 0.910 0.36 (0.0028,46) 0.680 Management Histological non-tumour-free margins 28.97 (3.4–249.1) 0.002** 44.02 (3.2–596.5) 0.004** 23 (2.1–250) 0.010* Sentinel lymph node biopsy 2.12 (1.1–4.2) 0.033* 2.55 (1.2–5.6) 0.020* 2 (0.44–9.2) 0.360 Adjuvant radiotherapy 2.03 (0.8–5.1) 0.130 1.94 (0.7–5.2) 0.190 1.7 (0.48–6.4) 0.400 Reduction of immunosuppression after diagnosis 2.43 (0.9–6.4) 0.072 2.46 (0.9–6.7) 0.078 1.2 (0.39–3.7) 0.750 mTORi initiated after MCC diagnosis (vs. no 2.60 (0.79,8.8) 0.120 2.60 (0.67,10) 0.170 3.60 (1.1,12) 0.032* mTORi before or after diagnosis) Being on an mTORi after diagnosis (vs. no 2.35 (1–5.8) 0.061 2.18 (0.8–5.7) 0.110 2.6 (0.85–7.7) 0.095 mTORi after diagnosis) Acitretin use after diagnosis 0.69 (0.2–3.1) 0.630 0.61 (0.1–3.3) 0.560 0.029 (0.0005–0.19) <0.001*** MCC, Merkel cell carcinoma; SOTR, solid organ transplant recipients; MCPyV, Merkel cell polyomavirus; KC, keratinocyte carcinoma; mTORi, mTOR inhibitor. †Only one predictor variable is included in the model. ‡Crude model adjusted by age (continuous), sex, and year of diagnosis. §Model 1 further adjusted by TNM Stage (I,II,III,IV)  P < 0.1, *P < 0.05, **P < 0.01, ***P < 0.001. material was not always available for MCPyV-status testing. In conclusion, compared with MCCs in immunocompetent Although we tried to minimize technical differences between the individuals, MCCs in SOTR develop at an earlier age and have a two laboratories, the possibility that this may have affected the worse prognosis, with higher risk of progression, disease-specific results cannot be entirely excluded. Finally, the influence of and overall mortality. Conversion to an mTORi after MCC diag- mTORi on outcome, although difficult to analyse because of the nosis did not reduce the risk of progression. MCPyV is less low number of patients, was adjusted on initial risk factors such prevalent in MCC from SOTR compared to immunocompetent as TNM at diagnosis to avoid bias by indication. individuals, suggesting that it plays a less important role in its JEADV 2022 Ó 2022 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology. 10 ndiz-Pulido et al. Ferra aetiopathogenesis. Because MCC is a rare tumour, multicentre, 12 Feng H, Shuda M, Chang Y, Moore PS. Clonal integration of a poly- retrospective studies such as ours are important in providing omavirus in human Merkel cell carcinoma. Science 2008; 319: 1096–1100. 13 Garneski KM, Warcola AH, Feng Q, Kiviat NB, Leonard JH, Nghiem P. information on epidemiology, clinical features, treatment strate- Merkel cell polyomavirus is more frequently present in North American gies and outcomes in SOTR. Our findings have significant trans- than Australian Merkel cell carcinoma tumors. J Invest Dermatol 2009; lational and clinical implications and should now be confirmed 129: 246–248. 14 Moshiri AS, Doumani R, Yelistratova L et al. Polyomavirus-Negative in larger, prospective cohorts. Merkel Cell Carcinoma: A More Aggressive Subtype Based on Analysis of 282 Cases Using Multimodal Tumor Virus Detection. J Invest Dermatol Consent for publication 2017; 137: 819–827. The patients in this study have given written informed consent 15 Schrama D, Peitsch WK, Zapatka M et al. Merkel cell polyomavirus status is not associated with clinical course of Merkel cell carcinoma. J Invest to publication of their case details. Dermatol 2011; 131: 1631–1638. 16 Rizzo JM, Harms PW, Harms KL, Plaska A, Brenner C, Durham AB. RB approval status Unknown primary Merkel cell carcinoma in the immunosuppressed Reviewed and approved by ‘Comite de etica de investigaci on de patient: Case series and review of literature. J Am Acad Dermatol 2020 Dec 1; 8: 19–22. medicamentos Hospital Universitari Vall d’Hebron’; approval 17 Starrett GJ, Thakuria M, Chen T et al. Clinical and molecular characteri- PR(AG)274/2018. zation of virus-positive and virus-negative Merkel cell carcinoma. Gen- ome Med 2020; 12: 30. Data availability statement 18 Becker JC, Stang A, DeCaprio JA et al. Merkel cell carcinoma. Nat Rev Dis Primer 2017; 3: 17077. The data presented in this manuscript are available from the cor- 19 Trenkic Bozinovic M, Krasic D, Katic V et al. Comparative analysis of responding author upon reasonable request. clinicopathological and immunohistochemical characteristics of Merkel cell carcinoma. J BUON Off J Balk Union Oncol 2014; 19: 530–534. Acknowledgement 20 Amin MB, Edge SB, Greene FL. Merkel cell carcinoma (Chapter 46). AJCC cancer staging manual. 8th edn. Springer, Switzerland, 2017: 549– Open access funding enabled and organized by ProjektDEAL. 61. 21 Mitteldorf C, Berisha A, Tronnier M, Pfaltz MC, Kempf W. PD-1 and References PD-L1 in neoplastic cells and the tumor microenvironment of Merkel cell 1 Hodgson NC. Merkel cell carcinoma: changing incidence trends. J Surg carcinoma. J Pathol 2017; 44: 740–746. Oncol 2005; 89: 1–4. 22 Daniel WW, Cross CL. Biostatistics: A Foundation for Analysis in the 2 Lemos BD, Storer BE, Iyer JG et al. Pathologic nodal evaluation improves Health Sciences, 10th edn. Wiley, Hoboken, NJ, 2013. prognostic accuracy in Merkel cell carcinoma: analysis of 5823 cases as 23 Cook M, Baker K, Redman M et al. Differential Outcomes Among the basis of the first consensus staging system. J Am Acad Dermatol 2010; Immunosuppressed Patients with Merkel Cell Carcinoma: Impact of 63: 751–761. Immunosuppression Type on Cancer-specific and Overall Survival. Am J 3 Clarke CA, Robbins HA, Tatalovich Z et al. Risk of Merkel Cell Carci- Clin Oncol 2019; 42: 82–88. noma After Solid Organ Transplantation. J Natl Cancer Inst 2015; 107: 24 Koljonen V, Sahi H, B€ ohling T, M€akisalo H. Post-transplant Merkel Cell 107. Carcinoma. Acta Derm Venereol 2016; 96: 442–447. 4 Engels EA, Frisch M, Goedert JJ, Biggar RJ, Miller RW. Merkel cell carci- 25 Bj€orn Andtback H, Bj€ ornhagen-S€afwenberg V, Shi H, Lui W-O, Masucci noma and HIV infection. Lancet 2002; 359: 497–498. GV, Villabona L. Sex Differences in Overall Survival and the Effect of 5 Tadmor T, Liphshitz I, Aviv A, Landgren O, Barchana M, Polliack A. Radiotherapy in Merkel Cell Carcinoma-A Retrospective Analysis of a Increased incidence of chronic lymphocytic leukaemia and lymphomas in Swedish Cohort. Cancers 2021; 13: 13(2). patients with Merkel cell carcinoma - a population based study of 335 26 Zaar O, Gillstedt M, Lindel€ of B, Wennberg-Lark€ o A-M, Paoli J. Merkel cases with neuroendocrine skin tumour. Br J Haematol 2012; 157: 457– cell carcinoma incidence is increasing in Sweden. J Eur Acad Dermatol 462. Venereol 2016; 30: 1708–1713. 6 Penn I, First MR. Merkel’s cell carcinoma in organ recipients: report of 27 Sahi H, Their J, Gissler M, Koljonen V. Merkel cell carcinoma treatment 41 cases. Transplantation 1999; 68: 1717–1721. in Finland in 1986-2016-a real-world data study. Cancers 2020; 12: 12. 7 Keeling E, Murray SL, Williams Y et al. Merkel cell carcinoma in kidney 28 Arron ST, Canavan T, Yu SS. Organ transplant recipients with Merkel cell transplant recipients in Ireland 1964–2018. Br J Dermatol 2019; 181: carcinoma have reduced progression-free, overall, and disease-specific 1314–1315. survival independent of stage at presentation. J Am Acad Dermatol 2014; 8 Koljonen V, Kukko H, Tukiainen E et al. Incidence of Merkel cell carci- 71: 684–690. noma in renal transplant recipients. Nephrol Dial Transplant 2009; 24: 29 Coggshall K, Tello TL, North JP, Yu SS. Merkel cell carcinoma: An update 3231–3235. and review: Pathogenesis, diagnosis, and staging. J Am Acad Dermatol 9 Krynitz B, Edgren G, Lindel€ of B et al. Risk of skin cancer and other malig- 2018; 78: 433–442. nancies in kidney, liver, heart and lung transplant recipients 1970 to 30 Kanitakis J, Euvrard S, Chouvet B, Butnaru AC, Claudy A. Merkel cell 2008--a Swedish population-based study. Int J Cancer 2013; 132: 1429– carcinoma in organ-transplant recipients: report of two cases with unu- 1438. sual histological features and literature review. J Cutan Pathol 2006; 33: 10 Na R, Grulich AE, Meagher NS, McCaughan GW, Keogh AM, Vajdic 686–694. CM. Comparison of de novo cancer incidence in Australian liver, heart 31 Schmults C, NCCN Clinical practice guidelines in Oncology: Merkel cell and lung transplant recipients. Am J Transplant 2013; 13: 174–183. carcinoma Version 2.2019, January 18, 2019. Published online January 11 Andres C, Belloni B, Puchta U, Sander CA, Flaig MJ. Prevalence of 2019. MCPyV in Merkel cell carcinoma and non-MCC tumors. J Cutan Pathol 32 Delyon J, Rabate C, Euvrard S et al. Management of Kaposi sarcoma after 2010; 37: 28–34. solid organ transplantation: A European retrospective study. J Am Acad Dermatol 2019; 81: 448–455. JEADV 2022 Ó 2022 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology. Merkel cell carcinoma and transplantation 11 33 Iwasaki T, Matsushita M, Nonaka D et al. Comparison of Akt/mTOR/4E- 43 Higaki-Mori H, Kuwamoto S, Iwasaki T et al. Association of Merkel cell BP1 pathway signal activation and mutations of PIK3CA in Merkel cell polyomavirus infection with clinicopathological differences in Merkel cell polyomavirus-positive and Merkel cell polyomavirus-negative carcino- carcinoma. Hum Pathol 2012; 43: 2282–2229. mas. Hum Pathol 2015; 46: 210–216. 44 Iwasaki T, Matsushita M, Kuwamoto S et al. Usefulness of significant 34 Shuda M, Arora R, Kwun HJ et al. Human Merkel cell polyomavirus morphologic characteristics in distinguishing between Merkel cell infection I. MCV T antigen expression in Merkel cell carcinoma, lym- polyomavirus-positive and Merkel cell polyomavirus-negative Merkel cell phoid tissues and lymphoid tumors. Int J Cancer 2009; 125: 1243– carcinomas. Hum Pathol 2013; 44: 1912–1917. 1249. 45 Kuwamoto S, Higaki H, Kanai K et al. Association of Merkel cell poly- 35 Chamcheu JC, Roy T, Uddin MB et al. Role and therapeutic targeting of omavirus infection with morphologic differences in Merkel cell carci- the PI3K/Akt/mTOR signaling pathway in skin cancer: a review of current noma. Hum Pathol 2011; 42: 632–640. status and future trends on natural and synthetic agents therapy. Cells 46 Paulson K, Lewis CW, Redman MW et al. Viral oncoprotein antibodies 2019; 8: 8(8). as a marker for recurrence of Merkel cell carcinoma: A prospective valida- 36 Tarabadkar ES, Fu T, Lachance K et al. Narrow excision margins are tion study. Cancer 2017; 123: 1464–1474. appropriate for Merkel cell carcinoma when combined with adjuvant ra- 47 Kwun HJ, Chang Y, Moore PS. Protein-mediated viral latency is a novel diation: Analysis of 188 cases of localized disease and proposed manage- mechanism for Merkel cell polyomavirus persistence. Proc Natl Acad Sci ment algorithm. J Am Acad Dermatol 2021; 84: 340–347. U S A 2017; 114: e4040–e4047. 37 Tam M, Luu M, Barker CA et al. Improved survival in women versus 48 Alvarez Orellana J, Jin Kwun H, Artusi S, Chang Y, Moore PS. Sirolimus and men with Merkel cell carcinoma. J Am Acad Dermatol 2021; 84: 321– other mechanistic target of rapamycin inhibitors directly activate latent patho- 329. genic human polyomavirus replication. J Infect Dis 2021; 13: 1160–1169. 38 Asgari MM, Sokil MM, Warton EM, Iyer J, Paulson KG, Nghiem P. 49 Schadendorf D, Lebbe C, Zur Hausen A et al. Merkel cell carcinoma: Epi- Effect of host, tumor, diagnostic, and treatment variables on outcomes demiology, prognosis, therapy and unmet medical needs. Eur J Cancer in a large cohort with Merkel cell carcinoma. JAMA Dermatol 2014; 150: 1990 2017; 71: 53–69. 716–723. 39 Jouary T, Kubica E, Dalle S et al. Sentinel node status and immunosup- Supporting information pression: recurrence factors in localized Merkel cell carcinoma. Acta Derm Venereol 2015; 95: 835–840. Additional Supporting Information may be found in the online 40 Paulson KG, Iyer JG, Blom A et al. Systemic immune suppression predicts version of this article: diminished Merkel cell carcinoma-specific survival independent of stage. J Invest Dermatol 2013; 133: 642–646. Table S1 Comparison of clinical and tumour characteristics of 41 Tarantola TI, Vallow LA, Halyard MY et al. Prognostic factors in Merkel cell carcinoma: analysis of 240 cases. J Am Acad Dermatol 2013; 68: 425– SOTR with and without an mTORi in their immunosuppressive 432. regimens at the time of MCC diagnosis. 42 Mendoza M-D, Santonja C, Gonzalez-Vela C et al. The presence of Mer- Table S2 MCPyV status (according to IHC or PCR results) by kel cell carcinoma polyomavirus is associated with a distinct phenotype in neoplastic Merkel cell carcinoma cells and their tissue microenviron- clinical and demographic characteristics of patients with MCC. ment. PLoS One 2020; 15: 15. Supporting Information S1 Supplementary Methods. JEADV 2022 Ó 2022 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology.