Pathology (April 2014) 46(3), pp. 205–210
ANATOMICAL PATHOLOGY
Merkel cell polyomavirus and p63 status in Merkel cell carcinoma by immunohistochemistry: Merkel cell polyomavirus positivity is inversely correlated with sun damage, but neither is correlated with outcome MARCUS DABNER1,*, ROBERT J. MCCLURE1,*, NATHAN T. HARVEY1, CHARLEY A. BUDGEON2,3, TREVOR W. BEER4, BENHUR AMANUEL1 AND BENJAMIN A. WOOD1 1Anatomical Pathology, PathWest Laboratory Medicine, QEII Medical Centre and School of Pathology and Laboratory Medicine, University of Western Australia, 2Centre for Applied Statistics, University of Western Australia, 3Department of Research, Sir Charles Gairdner Hospital, Hospital Avenue, Nedlands, 4Clinipath Pathology, WA, Australia; *these authors contributed equally to the work
Summary The aim of this study was to determine the frequency of Merkel cell polyomavirus (MPyV) and p63 positivity by immunohistochemistry in a large cohort of primary Merkel cell carcinoma (MCC) from a region with high rates of actinic damage. We also aimed to determine whether there is any relationship between these markers and histological correlates of chronic sun exposure and to identify whether these markers have prognostic significance in our population. Ninety-five cases of primary cutaneous MCC were identified and stained with immunohistochemical markers for MPyV and p63. The presence of solar elastosis and squamous dysplasia in the overlying/adjacent skin were recorded as markers of actinic damage. Follow up data were obtained from the Western Australian Cancer Registry. MPyV was detected by immunohistochemistry in 23% of cases. There was a statistically significantly lower rate of positivity in tumours associated with markers of chronic sun damage as assessed by the presence of solar elastosis and squamous dysplasia. There was no association with overall or disease specific survival. p63 positivity was detected in 17% of cases. There was no association with markers of actinic damage or with overall or disease specific survival. Our data demonstrate a significant difference in rates of immunohistochemical positivity for MPyV between MCC in sun-damaged and non-sundamaged sites. This may go some way to explaining previously identified geographical differences. When compared with a number of studies from Europe and North America, p63 positivity is less common in our population and does not show the strong prognostic significance that has been found in these other regions. Key words: Merkel cell carcinoma, Merkel cell polyomavirus, p63. Received 25 July, revised 3 September, accepted 8 September 2013
INTRODUCTION Merkel cell carcinoma (MCC) is a rare but aggressive cutaneous neuroendocrine tumour. The incidence of this tumour appears to be rising, with some reports indicating a tripling of reported cases over the last two decades.1 Among the aetiological factors which have been implicated in the development of MCC are ultraviolet radiation exposure and a novel polyomavirus, Merkel cell polyomavirus (MPyV).2 MCC is Print ISSN 0031-3025/Online ISSN 1465-3931 DOI: 10.1097/PAT.0000000000000069
#
typically found in sun-exposed skin and the incidence of MCC in Western Australia, a region with high levels of ultraviolet radiation, is the highest reported in the world with a concomitantly significant mortality.3 Although varying rates have been reported, MPyV has been detected in over 80% of MCC samples in several studies.2,4,5 The initial description of MPyV by Feng et al. described viral DNA integrated within the tumour genome in a clonal pattern, suggesting that integration preceded clonal expansion of the tumour cells.2 In addition, subsequent in vitro work has demonstrated that MPyV positive MCC cell lines require the expression of MPyV T antigen for ongoing proliferation and survival.6 Taken together, these lines of evidence support an aetiological role for MPyV in the development of at least a significant proportion of MCC. However, there are conflicting data regarding possible geographic variation in the rate of MPyV positivity in MCC. Some studies have suggested a lower rate of infection in Australia,7 and have suggested a more prominent role for sun exposure rather than viral aetiology in an Australian population. Other studies have failed to reproduce this difference.5 There are also conflicting data regarding the association of MPyV status with prognosis. While some studies have shown better overall survival in patients with MPyV positive tumours or higher viral copy numbers,8,9 other studies have failed to confirm this.5,10,11 Further controversy has also emerged in regard to the significance of positive immunohistochemical staining for the transcription factor p63. p63 is a member of the p53 family and is essential for the formation and normal functioning of the epidermis.12 In 2007 Asioli et al. published data on a small series of patients suggesting that p63 expression occurs in approximately 50% of cases of MCC and is associated with an aggressive clinical course.13 Subsequently, further work from the same group14 and from the USA10 has supported these findings. However, despite using the same methodology, a low rate of p63 staining and a lack of prognostic significance were found in a recent and relatively large study from Australia.15 Given these discrepant results for both MPyV and p63 status from around the globe, we sought to investigate these markers in a large Western Australian cohort of primary MCC, to determine the frequency of detection by immunohistochemistry and to assess whether either showed prognostic significance in our population.
2014 Royal College of Pathologists of Australasia
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
206
DABNER et al.
MATERIALS AND METHODS Institutional approval Ethics approval for this study was obtained from the Sir Charles Gairdner Hospital Human Research Ethics Committee and the Western Australian Department of Health Human Research Ethics Committee. Case selection and histological review The files of PathWest Laboratory Medicine (QEII Medical Centre) were searched for cases of primary cutaneous MCC diagnosed between 1990 and 2012. All available sections were retrieved and were reviewed by one of the authors (BAW). To meet inclusion criteria, cases were required to fulfil all of the following: an apparent primary tumour compatible with MCC on the basis of available clinical and pathological data; tumour in the skin or subcutis; tumour morphologically compatible with MCC on haematoxylin and eosin (H&E) appearances (Fig. 1A); cytoplasmic positivity for CK20 (Fig. 1B) or, in the absence of CK20 positivity, negative staining with TTF1 and no evidence of extracutaneous primary disease on re-evaluation of the available clinical and pathological data. Cases which did not meet these criteria (e.g., salivary gland or nodal disease with unknown primary, clinical suspicion of possible pulmonary or mediastinal primary) were excluded. A total of 67 cases met these inclusion criteria. An additional 28 cases, comprising a subset of cases with available blocks selected as previously described by one of the authors (TWB)16 were obtained from the files of Clinipath Pathology (formerly Cutaneous Pathology, Nedlands WA) and also reviewed by one author (BAW). The presence of solar elastosis in skin adjacent to the tumour was recorded, as was the presence of squamous dysplasia, including actinic keratosis and Bowen’s disease.
Pathology (2014), 46(3), April
studies of this marker in MCC.13 The H-score was calculated as 0 % of negative cells þ 1 % of weakly staining cells þ 2 % of moderately staining cells þ 3 % of strongly staining cells, giving a possible maximum score of 300. A total score of more than 10 was considered positive for the purposes of statistical evaluation (Fig. 1D). In addition to external controls, strong positive staining within the epidermis and adnexal structures served as an internal control. Follow up Follow up information, including survival, cause of death and development of metastatic disease to 15 January 2013 was obtained from the Western Australian Cancer Registry. The Western Australian Cancer Registry is a population based registry which receives and collates information in accordance with the (WA) Health (Notification Of Cancer) Regulations 1981. Notification of MCC is mandatory in Western Australia and the registry maintains information regarding outcome for notified cases.3 Statistical analysis Data were analysed using the R environment for statistical computing.17 Fisher’s exact p values were calculated to identify significant associations between the histological and immunohistochemical variables. Cox proportional hazards regression analysis was conducted to determine the factors associated with the survival outcomes, including all cause mortality, development of metastasis, disease specific survival and disease free survival. The specific variables investigated were sex, age, MPyV status (0 or 1) and p63 status (10, >10). Kaplan–Meier plots were used to display survival outcomes. Hazard ratios and 95% confidence intervals were obtained and statistical significance was set at p < 0.05.
Immunohistochemistry Immunohistochemical staining for MPyV (Clone cm2b4; 1/100 dilution; Santa Cruz, USA) and p63 (Clone 7jul; 1/50 dilution; Leica, Germany) was performed according to standard laboratory protocols. The results were assessed by one author (BAW). MPyV immunohistochemistry was considered positive if any nuclear staining was identified (Fig. 1C), with positive cases graded in a semiquantitative fashion on a three point scale (1, weak staining; 2, moderate staining; 3, strong staining). A score of 1 or more was considered positive for the purposes of statistical evaluation. Positive and negative controls were run in parallel. p63 staining was recorded as an H-score as previously described in
RESULTS A total of 95 patients with primary MCC were studied. The patients included 63 men and 30 women, with a median age of 73.6 years (range 33–94 years). Age and gender were unavailable in two cases. Follow-up information to either death or censoring date was available in 86 patients, with 28 (33%) dying of disease and 58 (67%) either alive or dead of unrelated causes. Most cases occurred in the head and neck region (55%),
Fig. 1 (A) A typical example of a Merkel cell carcinoma from our study. This tumour occurred on the leg of a 77-year-old female. It shows classical morphological features, with a sheet-like growth of relatively small cells showing hyperchromatic nuclei, minimal cytoplasm and a finely granular chromatin pattern. Mitotic figures are readily identified (see inset for cytological detail). (B) Immunohistochemistry for CK20 on the same tumour demonstrates the typical paranuclear staining pattern. (C) Another tumour from the study, from the knee of a 69-year-old male, showing positive immunohistochemical labelling for Merkel cell polyomavirus. (D) Another tumour from the study, from the forehead of a 49-year-old male, showing positive immunohistochemical labelling for p63.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
MERKEL CELL POLYOMAVIRUS AND P63 IN MERKEL CELL CARCINOMA
Table 1
Clinical features of 95 cases of primary Merkel cell carcinoma
Clinical feature
n (%)
Age (years) (n ¼ 93) 0–20 21–40 41–60 61–80 81þ Gender (n ¼ 93) Male Female Primary site (n ¼ 95) Head and neck Upper limb Lower limb Trunk Other/unknown
0 4 13 47 29
(0%) (4%) (14%) (51%) (31%)
63 (68%) 30 (32%) 52 15 19 5 4
(55%) (16%) (20%) (5%) (4%)
with the lower and upper limbs accounting for the bulk of the remainder (Table 1). Solar elastosis within the dermis was able to be assessed in 89 cases and was present in 73 cases and absent in 16 cases. Adjacent or overlying squamous dysplasia was detected in 23 of 89 cases in which this feature could be assessed. MPyV status by immunohistochemistry was assessable in 93 cases, with 23% showing positive staining and 77% showing no staining (Table 2). MPyV positivity was less common in tumours where the adjacent dermis showed evidence of solar elastosis: 15% of cases in elastotic skin were positive versus 44% of cases on non-elastotic skin (Fisher’s exact p value 0.018). Similarly, MPyV positivity was less common in lesions with adjacent squamous dysplasia: 5% of cases with squamous dysplasia were positive versus 27% of cases without adjacent squamous dysplasia (Fisher’s exact p value 0.034). Immunohistochemical analysis for p63 expression was also able to be assessed in 93 cases. Of these, 17% were positive (Table 2). There were no cases which showed positivity for both MPyV and p63. There was no significant association between p63 status and solar elastosis or the presence of squamous dysplasia. Divergent differentiation within the carcinoma was identified in 10 cases, in all of these represented by a squamous component. All cases with divergent differentiation were negative for MPyV. Table 2 Staining results for Merkel cell polyomavirus and p63 immunohistochemistry
Positive
Negative
MPyV total cases 21 (23%) 72 (77%) MPyV status by histological markers of actinic damage Elastosis present 11 61 Elastosis absent 7 9 Squamous dysplasia 1 21 No squamous dysplasia 18 48 p63 total cases 16 (17%) 77 (83%) p63 status by histological markers of actinic damage Elastosis present 13 59 Elastosis absent 0 16 Squamous dysplasia 3 19 No squamous dysplasia 10 56
Fisher’s exact p value
0.018 0.034
0.114 1.000
Not all parameters were able to be assessed in all cases, thus totals for each category are not equivalent.
207
We used univariate analysis to test for associations between MPyV status and p63 status and all-cause mortality, development of metastasis and disease specific mortality. The results are summarised in Table 3, while the relevant Kaplan–Meier curves are presented in Fig. 2. In our series there was no significant correlation on univariate analysis between immunohistochemical positivity for either MPyV or p63 and any of the outcomes investigated.
DISCUSSION Since its identification in 2008,2 MPyV has been identified in MCC from numerous geographical centres, with many studies reporting that up to 80% of MCC are positive.2,4,5,9,14 While some studies report a similar incidence in Australian cohorts,5 others have documented much lower rates in this country,7 raising the interesting possibility of a geographical variation in contributions of causative factors for MCC. One of the predisposing factors for the development of MCC is exposure to solar UV radiation, reflected in the fact that a high proportion of tumours occur in sun-exposed sites. It has been suggested that in Australia, UV radiation may be a more prevalent factor in the development of MCC than viral causes.7 Australia has high levels of solar UV radiation, as it lies in a latitude range of 108S to 438S, placing it closer to the equator than Europe and much of North America. In addition, the sun-earth separation is 3% less in southern hemisphere summers than in northern hemisphere summers,18 and the major population centres of Western Australia have higher levels of solar exposure than major population centres in the south-east of the continent.19 Thus, if there was a difference in the relative contribution of MPyV and UV radiation, it is expected that UV radiation would be significantly involved in our population. In this context, it is noteworthy that Western Australia has the highest documented global incidence of MCC.3 By immunohistochemistry, we found a relatively low rate of MPyV positivity in our tumours (23%). This is in keeping with the results of Garneski et al.,7 who documented a rate of 24% in Australian tumours using a polymerase chain reaction (PCR) based methodology. It is probable that immunohistochemistry represents a less sensitive technique for detection of MPyV than PCR. In this regard, a recent study using multiple detection methods has identified higher rates of MPyV positivity than most previous work, raising the possibility that MPyV is present in all cases of MCC.20 In that study, a correlation between immunohistochemical positivity and MPyV copy number per cell was noted. In contrast, others have recently identified distinct expression profiles between MPyV and MPyV negative cases in a series of cases in which 46% were positive for MPyV by PCR based testing.21 These results suggest that differences may exist between cases with low and high levels of MPyV expression, which straddle the thresholds of detection of the various techniques employed in different studies. The correlation of MPyV positivity with markers of sun damage in our series suggests that immunohistochemical positivity represents a biologically significant parameter in MCC, although immunohistochemical negativity should not be construed as indicating viral absence. Given the suggestion that MPyV may be less prevalent in tumours resulting from UV exposure, we compared the rate of MPyV positivity seen in tumours from sun-damaged and nonsun-damaged skin, using dermal elastosis and the presence of epidermal dysplasia as markers of actinic damage. This
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
208 Table 3
DABNER et al.
Pathology (2014), 46(3), April
Results of univariate analysis
Variable
All cause mortality
Development of metastases
Disease specific mortality
MPyV (present vs absent)
HR 1.22 (0.49–3.00) p ¼ 0.67 HR 2.14 (0.51–9.01) p ¼ 0.30
HR 1.63 (0.68–3.91) p ¼ 0.28 HR 1.56 (0.69–3.58) p ¼ 0.29
HR 1.24 (0.41–3.75) p ¼ 0.70 HR 1.40 (0.33–6.09) p ¼ 0.65
p63 (positive vs negative)
HR, hazard ratio; 95% confidence intervals are provided in parentheses.
HR 1.217 95% CI (0.494, 3.003) p = 0.6693
0.6
HR 2.138 95% CI (0.507, 9.011) p = 0.3006
p63 Negative Positive
0.0
0.0
Negative Positive
0.4
Survival probability MPyV
0.2
0.6 0.4 0.2
Survival probability
0.8
0.8
1.0
1.0
All cause mortality
5
0
A
10
0
15
5
Time (years)
10
15
Time (years)
1.0
HR 1.564 95% CI (0.685, 3.575) p = 0.2886
p63
0.0
Negative Positive
0.0
Negative Positive
0.6
0.8 HR 1.628 95% CI (0.677, 3.914) p = 0.2762
MPyV
0.4
METS probability
0.2
0.6 0.4 0.2
METS probability
0.8
1.0
Development of metastases
0
5
10
B
15
0
20
5
10
Time (years)
20
15
Time (years)
0.6
0.8 0
C
p63
HR 1.405 95% CI (0.325, 6.085) p = 0.6491
Negative Positive 0.0
0.0
Negative Positive
0.4
Survival probability HR 1.244 95% CI (0.413, 3.751) p = 0.6976
MPyV
0.2
0.6 0.4 0.2
Survival probability
0.8
1.0
1.0
Disease specific mortality
5
10
Time (years)
15
0
5
10
15
Time (years)
Fig. 2 (A) Kaplan–Meier plots comparing all cause mortality, (B) development of metastases and (C) disease specific mortality between tumours based on either Merkel cell polyomavirus or p63 status (n ¼ 93).
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
MERKEL CELL POLYOMAVIRUS AND P63 IN MERKEL CELL CARCINOMA
analysis did reveal a difference, with a significantly greater proportion of tumours on non-sun-damaged skin showing MPyV positivity and lower rates of associated squamous dysplasia in MPyV positive cases. The rate of MPyV positivity in tumours arising in non-elastotic skin in our population (44%) is close to the rate of 47% reported by Hall et al. in a North American population using immunohistochemical detection.10 It is also noteworthy that MPyV has been found to be absent in cases of MCC with squamous differentiation.22 Squamous differentiation in MCC is relatively common in our practice, being present in 11% of cases in this series. While similar rates have been reported in other series, many of these have included consultation cases, which are likely to over-represent the frequency of this phenomenon in other populations.22 These data may lend some further weight to the hypothesis that there is geographic heterogeneity in MCC. The significance of MPyV status in determining tumour outcome has been controversial, with some studies reporting improved survival in patients with positive tumours,8,9 while others have refuted this.5,10,11 In our cohort, there was no relationship between MPyV status and overall survival, time to development of metastasis or disease specific survival. Countless previous studies on prognostic factors in MCC have been published, many of which have conflicting outcomes. These have been recently summarised by Lim et al.15 It is uncertain whether this variability is a result of the relatively small number of cases in some of the studies (due primarily to the rarity of MCC), or whether there is true heterogeneity in MCC. The larger published studies have documented worse outcomes for patients with higher stage,23–25 lymphovascular invasion,24 male gender,23 and increased tumour size23,25 and a positive impact for adjuvant radiotherapy.26 Recent studies have suggested that p63 positivity may portend a very significantly worse outcome in MCC, and might be particularly useful for prognosticating earlier-staged tumours.10,13,14 We found a relatively low rate (17%) of p63 positivity in our cases. This is much lower than in previous studies from both the USA and Europe, which have reported p63 positivity in between 50% and 80% of cases.10,13,14 However, this rate is comparable to another recent study from Australia, which found positive p63 staining in only 9% of primary MCC.15 We found no correlation between patient survival and p63 status. Given that p63 is a technically robust immunohistochemical marker in routine diagnostic use, that internal positive controls can be visualised in most sections of primary MCC and that the clone used in this study was identical to that used in previous work, we consider it unlikely that technical variation can explain these discrepant results. Our results add to previous studies which have demonstrated that in an Australian population there is a relatively low incidence of both MPyV and p63 positivity by immunohistochemistry and that neither of these markers appear to correlate with survival. Given the differing results from studies in different geographic locations and an apparent inverse association between MPyV immunohistochemical status and markers of chronic sun damage, it remains possible that there is more heterogeneity within MCC than has previously been recognised. Indeed, early work with in vitro MCC cell lines demonstrated some morphological variability,27 and later gene expression profiling studies were able to demonstrate distinct expression profiles for different subtypes.28 On a practical level, our data suggest that p63 expression cannot be introduced
209
as a prognostic marker in routine practice at this stage. Ongoing molecular characterisation of these tumours, including homogenised sample and in situ PCR detection of MPyV as well as next generation sequencing may yield more information to explain the apparent heterogeneity observed in this aggressive skin tumour. Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Address for correspondence: Dr B. A. Wood, Department of Anatomical Pathology, PathWest Laboratory Medicine, QEII Medical Centre, Hospital Avenue, Nedlands, WA 6009, Australia. E-mail: [email protected]. gov.au
References 1. Hodgson NC. Merkel cell carcinoma: changing incidence trends. J Surg Oncol 2005; 89: 1–4. 2. Feng H, Shuda M, Chang Y, Moore PS. Clonal integration of a polyomavirus in human Merkel cell carcinoma. Science 2008; 319: 1096– 100. 3. Girschik J, Thorn K, Beer TW, Heenan PJ, Fritschi L. Merkel cell carcinoma in Western Australia: a population-based study of incidence and survival. Br J Dermatol 2011; 165: 1051–7. 4. Becker JC, Houben R, Ugurel S, Trefzer U, Pfohler C, Schrama D. MC polyomavirus is frequently present in Merkel cell carcinoma of European patients. J Invest Dermatol 2009; 129: 248–50. 5. Schrama D, Peitsch WK, Zapatka M, et al. Merkel cell polyomavirus status is not associated with clinical course of Merkel cell carcinoma. J Invest Dermatol 2011; 131: 1631–8. 6. Houben R, Shuda M, Weinkam R, et al. Merkel cell polyomavirus-infected Merkel cell carcinoma cells require expression of viral T antigens. J Virol 2010; 84: 7064–72. 7. Garneski KM, Warcola AH, Feng Q, Kiviat NB, Leonard JH, Nghiem P. Merkel cell polyomavirus is more frequently present in North American than Australian Merkel cell carcinoma tumors. J Invest Dermatol 2009; 129: 246–8. 8. Bhatia K, Goedert JJ, Modali R, Preiss L, Ayers LW. Immunological detection of viral large T antigen identifies a subset of Merkel cell carcinoma tumors with higher viral abundance and better clinical outcome. Int J Cancer 2010; 127: 1493–6. 9. Sihto H, Kukko H, Koljonen V, Sankila R, Bohling T, Joensuu H. Clinical factors associated with Merkel cell polyomavirus infection in Merkel cell carcinoma. J Natl Cancer Inst 2009; 101: 938–45. 10. Hall BJ, Pincus LB, Yu SS, Oh DH, Wilson AR, McCalmont TH. Immunohistochemical prognostication of Merkel cell carcinoma: p63 expression but not polyomavirus status correlates with outcome. J Cutan Pathol 2012; 39: 911–7. 11. Handschel J, Muller D, Depprich RA, et al. The new polyomavirus (MCPyV) does not affect the clinical course in MCCs. Int J Oral Maxillofac Surg 2010; 39: 1086–90. 12. Vanbokhoven H, Melino G, Candi E, Declercq W. p63, a story of mice and men. J Invest Dermatol 2011; 131: 1196–207. 13. Asioli S, Righi A, Volante M, Eusebi V, Bussolati G. p63 expression as a new prognostic marker in Merkel cell carcinoma. Cancer 2007; 110: 640–7. 14. Asioli S, Righi A, de Biase D, et al. Expression of p63 is the sole independent marker of aggressiveness in localised (stage I-II) Merkel cell carcinomas. Mod Pathol 2011; 24: 1451–61. 15. Lim CS, Whalley D, Haydu LE, et al. Increasing tumor thickness is associated with recurrence and poorer survival in patients with Merkel cell carcinoma. Ann Surg Oncol 2012; 19: 3325–34. 16. Beer TW, Ng LB, Murray K. Mast cells have prognostic value in Merkel cell carcinoma. Am J Dermatopathol 2008; 30: 27–30. 17. Team RC. R: A Language and Environment for Statistical Computing. Vienna: R Foundation for Statistical Computing, 2012. 18. Gies P, Roy C, Javorniczky J, Henderson S, Lemus-Deschamps L, Driscoll C. Global Solar UV Index: Australian measurements, forecasts and comparison with the UK. Photochem Photobiol 2004; 79: 32–9. 19. Australian Government, Bureau of Meteorology. Average daily solar exposure. 26 Feb 2013; cited 19 Jun 2013. http://www.bom.gov.au/jsp/ ncc/climate_averages/solar-exposure/index.jsp.). 20. Rodig SJ, Cheng J, Wardzala J, et al. Improved detection suggests all Merkel cell carcinomas harbor Merkel polyomavirus. J Clin Invest 2012; 122: 4645–53.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
210
DABNER et al.
21. Harms P, Patel R, Verhaegen M, et al. Distinct gene expression profiles of viral- and nonviral-associated Merkel cell carcinoma revealed by transcriptome analysis. J Invest Dermatol 2013; 133: 936–45. 22. Martin B, Poblet E, Rios JJ, et al. Merkel cell carcinoma with divergent differentiation: histopathological and immunohistochemical study of 15 cases with PCR analysis for Merkel cell polyomavirus. Histopathology 2013; 62: 711–22. 23. Albores-Saavedra J, Batich K, Chable-Montero F, Sagy N, Schwartz AM, Henson DE. Merkel cell carcinoma demographics, morphology, and survival based on 3870 cases: a population based study. J Cutan Pathol 2009; 37: 20–7. 24. Fields RC, Busam KJ, Chou JF, et al. Five hundred patients with Merkel cell carcinoma evaluated at a single institution. Ann Surg 2011; 254: 465– 73.
Pathology (2014), 46(3), April
25. Lemos BD, Storer BE, Iyer JG, et al. Pathologic nodal evaluation improves prognostic accuracy in Merkel cell carcinoma: analysis of 5823 cases as the basis of the first consensus staging system. J Am Acad Dermatol 2010; 63: 751–61. 26. Mojica P, Smith D, Ellenhorn JD. Adjuvant radiation therapy is associated with improved survival in Merkel cell carcinoma of the skin. J Clin Oncol 2007; 25: 1043–7. 27. Leonard JH, Bell JR, Kearsley JH. Characterization of cell lines established from Merkel-cell (’small-cell’) carcinoma of the skin. Int J Cancer 1993; 55: 803–10. 28. Van Gele M, Boyle GM, Cook AL, et al. Gene-expression profiling reveals distinct expression patterns for Classic versus Variant Merkel cell phenotypes and new classifier genes to distinguish Merkel cell from small-cell lung carcinoma. Oncogene 2004; 23: 2732–42.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
ANATOMICAL PATHOLOGY
Merkel cell polyomavirus and p63 status in Merkel cell carcinoma by immunohistochemistry: Merkel cell polyomavirus positivity is inversely correlated with sun damage, but neither is correlated with outcome MARCUS DABNER1,*, ROBERT J. MCCLURE1,*, NATHAN T. HARVEY1, CHARLEY A. BUDGEON2,3, TREVOR W. BEER4, BENHUR AMANUEL1 AND BENJAMIN A. WOOD1 1Anatomical Pathology, PathWest Laboratory Medicine, QEII Medical Centre and School of Pathology and Laboratory Medicine, University of Western Australia, 2Centre for Applied Statistics, University of Western Australia, 3Department of Research, Sir Charles Gairdner Hospital, Hospital Avenue, Nedlands, 4Clinipath Pathology, WA, Australia; *these authors contributed equally to the work
Summary The aim of this study was to determine the frequency of Merkel cell polyomavirus (MPyV) and p63 positivity by immunohistochemistry in a large cohort of primary Merkel cell carcinoma (MCC) from a region with high rates of actinic damage. We also aimed to determine whether there is any relationship between these markers and histological correlates of chronic sun exposure and to identify whether these markers have prognostic significance in our population. Ninety-five cases of primary cutaneous MCC were identified and stained with immunohistochemical markers for MPyV and p63. The presence of solar elastosis and squamous dysplasia in the overlying/adjacent skin were recorded as markers of actinic damage. Follow up data were obtained from the Western Australian Cancer Registry. MPyV was detected by immunohistochemistry in 23% of cases. There was a statistically significantly lower rate of positivity in tumours associated with markers of chronic sun damage as assessed by the presence of solar elastosis and squamous dysplasia. There was no association with overall or disease specific survival. p63 positivity was detected in 17% of cases. There was no association with markers of actinic damage or with overall or disease specific survival. Our data demonstrate a significant difference in rates of immunohistochemical positivity for MPyV between MCC in sun-damaged and non-sundamaged sites. This may go some way to explaining previously identified geographical differences. When compared with a number of studies from Europe and North America, p63 positivity is less common in our population and does not show the strong prognostic significance that has been found in these other regions. Key words: Merkel cell carcinoma, Merkel cell polyomavirus, p63. Received 25 July, revised 3 September, accepted 8 September 2013
INTRODUCTION Merkel cell carcinoma (MCC) is a rare but aggressive cutaneous neuroendocrine tumour. The incidence of this tumour appears to be rising, with some reports indicating a tripling of reported cases over the last two decades.1 Among the aetiological factors which have been implicated in the development of MCC are ultraviolet radiation exposure and a novel polyomavirus, Merkel cell polyomavirus (MPyV).2 MCC is Print ISSN 0031-3025/Online ISSN 1465-3931 DOI: 10.1097/PAT.0000000000000069
#
typically found in sun-exposed skin and the incidence of MCC in Western Australia, a region with high levels of ultraviolet radiation, is the highest reported in the world with a concomitantly significant mortality.3 Although varying rates have been reported, MPyV has been detected in over 80% of MCC samples in several studies.2,4,5 The initial description of MPyV by Feng et al. described viral DNA integrated within the tumour genome in a clonal pattern, suggesting that integration preceded clonal expansion of the tumour cells.2 In addition, subsequent in vitro work has demonstrated that MPyV positive MCC cell lines require the expression of MPyV T antigen for ongoing proliferation and survival.6 Taken together, these lines of evidence support an aetiological role for MPyV in the development of at least a significant proportion of MCC. However, there are conflicting data regarding possible geographic variation in the rate of MPyV positivity in MCC. Some studies have suggested a lower rate of infection in Australia,7 and have suggested a more prominent role for sun exposure rather than viral aetiology in an Australian population. Other studies have failed to reproduce this difference.5 There are also conflicting data regarding the association of MPyV status with prognosis. While some studies have shown better overall survival in patients with MPyV positive tumours or higher viral copy numbers,8,9 other studies have failed to confirm this.5,10,11 Further controversy has also emerged in regard to the significance of positive immunohistochemical staining for the transcription factor p63. p63 is a member of the p53 family and is essential for the formation and normal functioning of the epidermis.12 In 2007 Asioli et al. published data on a small series of patients suggesting that p63 expression occurs in approximately 50% of cases of MCC and is associated with an aggressive clinical course.13 Subsequently, further work from the same group14 and from the USA10 has supported these findings. However, despite using the same methodology, a low rate of p63 staining and a lack of prognostic significance were found in a recent and relatively large study from Australia.15 Given these discrepant results for both MPyV and p63 status from around the globe, we sought to investigate these markers in a large Western Australian cohort of primary MCC, to determine the frequency of detection by immunohistochemistry and to assess whether either showed prognostic significance in our population.
2014 Royal College of Pathologists of Australasia
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
206
DABNER et al.
MATERIALS AND METHODS Institutional approval Ethics approval for this study was obtained from the Sir Charles Gairdner Hospital Human Research Ethics Committee and the Western Australian Department of Health Human Research Ethics Committee. Case selection and histological review The files of PathWest Laboratory Medicine (QEII Medical Centre) were searched for cases of primary cutaneous MCC diagnosed between 1990 and 2012. All available sections were retrieved and were reviewed by one of the authors (BAW). To meet inclusion criteria, cases were required to fulfil all of the following: an apparent primary tumour compatible with MCC on the basis of available clinical and pathological data; tumour in the skin or subcutis; tumour morphologically compatible with MCC on haematoxylin and eosin (H&E) appearances (Fig. 1A); cytoplasmic positivity for CK20 (Fig. 1B) or, in the absence of CK20 positivity, negative staining with TTF1 and no evidence of extracutaneous primary disease on re-evaluation of the available clinical and pathological data. Cases which did not meet these criteria (e.g., salivary gland or nodal disease with unknown primary, clinical suspicion of possible pulmonary or mediastinal primary) were excluded. A total of 67 cases met these inclusion criteria. An additional 28 cases, comprising a subset of cases with available blocks selected as previously described by one of the authors (TWB)16 were obtained from the files of Clinipath Pathology (formerly Cutaneous Pathology, Nedlands WA) and also reviewed by one author (BAW). The presence of solar elastosis in skin adjacent to the tumour was recorded, as was the presence of squamous dysplasia, including actinic keratosis and Bowen’s disease.
Pathology (2014), 46(3), April
studies of this marker in MCC.13 The H-score was calculated as 0 % of negative cells þ 1 % of weakly staining cells þ 2 % of moderately staining cells þ 3 % of strongly staining cells, giving a possible maximum score of 300. A total score of more than 10 was considered positive for the purposes of statistical evaluation (Fig. 1D). In addition to external controls, strong positive staining within the epidermis and adnexal structures served as an internal control. Follow up Follow up information, including survival, cause of death and development of metastatic disease to 15 January 2013 was obtained from the Western Australian Cancer Registry. The Western Australian Cancer Registry is a population based registry which receives and collates information in accordance with the (WA) Health (Notification Of Cancer) Regulations 1981. Notification of MCC is mandatory in Western Australia and the registry maintains information regarding outcome for notified cases.3 Statistical analysis Data were analysed using the R environment for statistical computing.17 Fisher’s exact p values were calculated to identify significant associations between the histological and immunohistochemical variables. Cox proportional hazards regression analysis was conducted to determine the factors associated with the survival outcomes, including all cause mortality, development of metastasis, disease specific survival and disease free survival. The specific variables investigated were sex, age, MPyV status (0 or 1) and p63 status (10, >10). Kaplan–Meier plots were used to display survival outcomes. Hazard ratios and 95% confidence intervals were obtained and statistical significance was set at p < 0.05.
Immunohistochemistry Immunohistochemical staining for MPyV (Clone cm2b4; 1/100 dilution; Santa Cruz, USA) and p63 (Clone 7jul; 1/50 dilution; Leica, Germany) was performed according to standard laboratory protocols. The results were assessed by one author (BAW). MPyV immunohistochemistry was considered positive if any nuclear staining was identified (Fig. 1C), with positive cases graded in a semiquantitative fashion on a three point scale (1, weak staining; 2, moderate staining; 3, strong staining). A score of 1 or more was considered positive for the purposes of statistical evaluation. Positive and negative controls were run in parallel. p63 staining was recorded as an H-score as previously described in
RESULTS A total of 95 patients with primary MCC were studied. The patients included 63 men and 30 women, with a median age of 73.6 years (range 33–94 years). Age and gender were unavailable in two cases. Follow-up information to either death or censoring date was available in 86 patients, with 28 (33%) dying of disease and 58 (67%) either alive or dead of unrelated causes. Most cases occurred in the head and neck region (55%),
Fig. 1 (A) A typical example of a Merkel cell carcinoma from our study. This tumour occurred on the leg of a 77-year-old female. It shows classical morphological features, with a sheet-like growth of relatively small cells showing hyperchromatic nuclei, minimal cytoplasm and a finely granular chromatin pattern. Mitotic figures are readily identified (see inset for cytological detail). (B) Immunohistochemistry for CK20 on the same tumour demonstrates the typical paranuclear staining pattern. (C) Another tumour from the study, from the knee of a 69-year-old male, showing positive immunohistochemical labelling for Merkel cell polyomavirus. (D) Another tumour from the study, from the forehead of a 49-year-old male, showing positive immunohistochemical labelling for p63.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
MERKEL CELL POLYOMAVIRUS AND P63 IN MERKEL CELL CARCINOMA
Table 1
Clinical features of 95 cases of primary Merkel cell carcinoma
Clinical feature
n (%)
Age (years) (n ¼ 93) 0–20 21–40 41–60 61–80 81þ Gender (n ¼ 93) Male Female Primary site (n ¼ 95) Head and neck Upper limb Lower limb Trunk Other/unknown
0 4 13 47 29
(0%) (4%) (14%) (51%) (31%)
63 (68%) 30 (32%) 52 15 19 5 4
(55%) (16%) (20%) (5%) (4%)
with the lower and upper limbs accounting for the bulk of the remainder (Table 1). Solar elastosis within the dermis was able to be assessed in 89 cases and was present in 73 cases and absent in 16 cases. Adjacent or overlying squamous dysplasia was detected in 23 of 89 cases in which this feature could be assessed. MPyV status by immunohistochemistry was assessable in 93 cases, with 23% showing positive staining and 77% showing no staining (Table 2). MPyV positivity was less common in tumours where the adjacent dermis showed evidence of solar elastosis: 15% of cases in elastotic skin were positive versus 44% of cases on non-elastotic skin (Fisher’s exact p value 0.018). Similarly, MPyV positivity was less common in lesions with adjacent squamous dysplasia: 5% of cases with squamous dysplasia were positive versus 27% of cases without adjacent squamous dysplasia (Fisher’s exact p value 0.034). Immunohistochemical analysis for p63 expression was also able to be assessed in 93 cases. Of these, 17% were positive (Table 2). There were no cases which showed positivity for both MPyV and p63. There was no significant association between p63 status and solar elastosis or the presence of squamous dysplasia. Divergent differentiation within the carcinoma was identified in 10 cases, in all of these represented by a squamous component. All cases with divergent differentiation were negative for MPyV. Table 2 Staining results for Merkel cell polyomavirus and p63 immunohistochemistry
Positive
Negative
MPyV total cases 21 (23%) 72 (77%) MPyV status by histological markers of actinic damage Elastosis present 11 61 Elastosis absent 7 9 Squamous dysplasia 1 21 No squamous dysplasia 18 48 p63 total cases 16 (17%) 77 (83%) p63 status by histological markers of actinic damage Elastosis present 13 59 Elastosis absent 0 16 Squamous dysplasia 3 19 No squamous dysplasia 10 56
Fisher’s exact p value
0.018 0.034
0.114 1.000
Not all parameters were able to be assessed in all cases, thus totals for each category are not equivalent.
207
We used univariate analysis to test for associations between MPyV status and p63 status and all-cause mortality, development of metastasis and disease specific mortality. The results are summarised in Table 3, while the relevant Kaplan–Meier curves are presented in Fig. 2. In our series there was no significant correlation on univariate analysis between immunohistochemical positivity for either MPyV or p63 and any of the outcomes investigated.
DISCUSSION Since its identification in 2008,2 MPyV has been identified in MCC from numerous geographical centres, with many studies reporting that up to 80% of MCC are positive.2,4,5,9,14 While some studies report a similar incidence in Australian cohorts,5 others have documented much lower rates in this country,7 raising the interesting possibility of a geographical variation in contributions of causative factors for MCC. One of the predisposing factors for the development of MCC is exposure to solar UV radiation, reflected in the fact that a high proportion of tumours occur in sun-exposed sites. It has been suggested that in Australia, UV radiation may be a more prevalent factor in the development of MCC than viral causes.7 Australia has high levels of solar UV radiation, as it lies in a latitude range of 108S to 438S, placing it closer to the equator than Europe and much of North America. In addition, the sun-earth separation is 3% less in southern hemisphere summers than in northern hemisphere summers,18 and the major population centres of Western Australia have higher levels of solar exposure than major population centres in the south-east of the continent.19 Thus, if there was a difference in the relative contribution of MPyV and UV radiation, it is expected that UV radiation would be significantly involved in our population. In this context, it is noteworthy that Western Australia has the highest documented global incidence of MCC.3 By immunohistochemistry, we found a relatively low rate of MPyV positivity in our tumours (23%). This is in keeping with the results of Garneski et al.,7 who documented a rate of 24% in Australian tumours using a polymerase chain reaction (PCR) based methodology. It is probable that immunohistochemistry represents a less sensitive technique for detection of MPyV than PCR. In this regard, a recent study using multiple detection methods has identified higher rates of MPyV positivity than most previous work, raising the possibility that MPyV is present in all cases of MCC.20 In that study, a correlation between immunohistochemical positivity and MPyV copy number per cell was noted. In contrast, others have recently identified distinct expression profiles between MPyV and MPyV negative cases in a series of cases in which 46% were positive for MPyV by PCR based testing.21 These results suggest that differences may exist between cases with low and high levels of MPyV expression, which straddle the thresholds of detection of the various techniques employed in different studies. The correlation of MPyV positivity with markers of sun damage in our series suggests that immunohistochemical positivity represents a biologically significant parameter in MCC, although immunohistochemical negativity should not be construed as indicating viral absence. Given the suggestion that MPyV may be less prevalent in tumours resulting from UV exposure, we compared the rate of MPyV positivity seen in tumours from sun-damaged and nonsun-damaged skin, using dermal elastosis and the presence of epidermal dysplasia as markers of actinic damage. This
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
208 Table 3
DABNER et al.
Pathology (2014), 46(3), April
Results of univariate analysis
Variable
All cause mortality
Development of metastases
Disease specific mortality
MPyV (present vs absent)
HR 1.22 (0.49–3.00) p ¼ 0.67 HR 2.14 (0.51–9.01) p ¼ 0.30
HR 1.63 (0.68–3.91) p ¼ 0.28 HR 1.56 (0.69–3.58) p ¼ 0.29
HR 1.24 (0.41–3.75) p ¼ 0.70 HR 1.40 (0.33–6.09) p ¼ 0.65
p63 (positive vs negative)
HR, hazard ratio; 95% confidence intervals are provided in parentheses.
HR 1.217 95% CI (0.494, 3.003) p = 0.6693
0.6
HR 2.138 95% CI (0.507, 9.011) p = 0.3006
p63 Negative Positive
0.0
0.0
Negative Positive
0.4
Survival probability MPyV
0.2
0.6 0.4 0.2
Survival probability
0.8
0.8
1.0
1.0
All cause mortality
5
0
A
10
0
15
5
Time (years)
10
15
Time (years)
1.0
HR 1.564 95% CI (0.685, 3.575) p = 0.2886
p63
0.0
Negative Positive
0.0
Negative Positive
0.6
0.8 HR 1.628 95% CI (0.677, 3.914) p = 0.2762
MPyV
0.4
METS probability
0.2
0.6 0.4 0.2
METS probability
0.8
1.0
Development of metastases
0
5
10
B
15
0
20
5
10
Time (years)
20
15
Time (years)
0.6
0.8 0
C
p63
HR 1.405 95% CI (0.325, 6.085) p = 0.6491
Negative Positive 0.0
0.0
Negative Positive
0.4
Survival probability HR 1.244 95% CI (0.413, 3.751) p = 0.6976
MPyV
0.2
0.6 0.4 0.2
Survival probability
0.8
1.0
1.0
Disease specific mortality
5
10
Time (years)
15
0
5
10
15
Time (years)
Fig. 2 (A) Kaplan–Meier plots comparing all cause mortality, (B) development of metastases and (C) disease specific mortality between tumours based on either Merkel cell polyomavirus or p63 status (n ¼ 93).
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
MERKEL CELL POLYOMAVIRUS AND P63 IN MERKEL CELL CARCINOMA
analysis did reveal a difference, with a significantly greater proportion of tumours on non-sun-damaged skin showing MPyV positivity and lower rates of associated squamous dysplasia in MPyV positive cases. The rate of MPyV positivity in tumours arising in non-elastotic skin in our population (44%) is close to the rate of 47% reported by Hall et al. in a North American population using immunohistochemical detection.10 It is also noteworthy that MPyV has been found to be absent in cases of MCC with squamous differentiation.22 Squamous differentiation in MCC is relatively common in our practice, being present in 11% of cases in this series. While similar rates have been reported in other series, many of these have included consultation cases, which are likely to over-represent the frequency of this phenomenon in other populations.22 These data may lend some further weight to the hypothesis that there is geographic heterogeneity in MCC. The significance of MPyV status in determining tumour outcome has been controversial, with some studies reporting improved survival in patients with positive tumours,8,9 while others have refuted this.5,10,11 In our cohort, there was no relationship between MPyV status and overall survival, time to development of metastasis or disease specific survival. Countless previous studies on prognostic factors in MCC have been published, many of which have conflicting outcomes. These have been recently summarised by Lim et al.15 It is uncertain whether this variability is a result of the relatively small number of cases in some of the studies (due primarily to the rarity of MCC), or whether there is true heterogeneity in MCC. The larger published studies have documented worse outcomes for patients with higher stage,23–25 lymphovascular invasion,24 male gender,23 and increased tumour size23,25 and a positive impact for adjuvant radiotherapy.26 Recent studies have suggested that p63 positivity may portend a very significantly worse outcome in MCC, and might be particularly useful for prognosticating earlier-staged tumours.10,13,14 We found a relatively low rate (17%) of p63 positivity in our cases. This is much lower than in previous studies from both the USA and Europe, which have reported p63 positivity in between 50% and 80% of cases.10,13,14 However, this rate is comparable to another recent study from Australia, which found positive p63 staining in only 9% of primary MCC.15 We found no correlation between patient survival and p63 status. Given that p63 is a technically robust immunohistochemical marker in routine diagnostic use, that internal positive controls can be visualised in most sections of primary MCC and that the clone used in this study was identical to that used in previous work, we consider it unlikely that technical variation can explain these discrepant results. Our results add to previous studies which have demonstrated that in an Australian population there is a relatively low incidence of both MPyV and p63 positivity by immunohistochemistry and that neither of these markers appear to correlate with survival. Given the differing results from studies in different geographic locations and an apparent inverse association between MPyV immunohistochemical status and markers of chronic sun damage, it remains possible that there is more heterogeneity within MCC than has previously been recognised. Indeed, early work with in vitro MCC cell lines demonstrated some morphological variability,27 and later gene expression profiling studies were able to demonstrate distinct expression profiles for different subtypes.28 On a practical level, our data suggest that p63 expression cannot be introduced
209
as a prognostic marker in routine practice at this stage. Ongoing molecular characterisation of these tumours, including homogenised sample and in situ PCR detection of MPyV as well as next generation sequencing may yield more information to explain the apparent heterogeneity observed in this aggressive skin tumour. Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Address for correspondence: Dr B. A. Wood, Department of Anatomical Pathology, PathWest Laboratory Medicine, QEII Medical Centre, Hospital Avenue, Nedlands, WA 6009, Australia. E-mail: [email protected]. gov.au
References 1. Hodgson NC. Merkel cell carcinoma: changing incidence trends. J Surg Oncol 2005; 89: 1–4. 2. Feng H, Shuda M, Chang Y, Moore PS. Clonal integration of a polyomavirus in human Merkel cell carcinoma. Science 2008; 319: 1096– 100. 3. Girschik J, Thorn K, Beer TW, Heenan PJ, Fritschi L. Merkel cell carcinoma in Western Australia: a population-based study of incidence and survival. Br J Dermatol 2011; 165: 1051–7. 4. Becker JC, Houben R, Ugurel S, Trefzer U, Pfohler C, Schrama D. MC polyomavirus is frequently present in Merkel cell carcinoma of European patients. J Invest Dermatol 2009; 129: 248–50. 5. Schrama D, Peitsch WK, Zapatka M, et al. Merkel cell polyomavirus status is not associated with clinical course of Merkel cell carcinoma. J Invest Dermatol 2011; 131: 1631–8. 6. Houben R, Shuda M, Weinkam R, et al. Merkel cell polyomavirus-infected Merkel cell carcinoma cells require expression of viral T antigens. J Virol 2010; 84: 7064–72. 7. Garneski KM, Warcola AH, Feng Q, Kiviat NB, Leonard JH, Nghiem P. Merkel cell polyomavirus is more frequently present in North American than Australian Merkel cell carcinoma tumors. J Invest Dermatol 2009; 129: 246–8. 8. Bhatia K, Goedert JJ, Modali R, Preiss L, Ayers LW. Immunological detection of viral large T antigen identifies a subset of Merkel cell carcinoma tumors with higher viral abundance and better clinical outcome. Int J Cancer 2010; 127: 1493–6. 9. Sihto H, Kukko H, Koljonen V, Sankila R, Bohling T, Joensuu H. Clinical factors associated with Merkel cell polyomavirus infection in Merkel cell carcinoma. J Natl Cancer Inst 2009; 101: 938–45. 10. Hall BJ, Pincus LB, Yu SS, Oh DH, Wilson AR, McCalmont TH. Immunohistochemical prognostication of Merkel cell carcinoma: p63 expression but not polyomavirus status correlates with outcome. J Cutan Pathol 2012; 39: 911–7. 11. Handschel J, Muller D, Depprich RA, et al. The new polyomavirus (MCPyV) does not affect the clinical course in MCCs. Int J Oral Maxillofac Surg 2010; 39: 1086–90. 12. Vanbokhoven H, Melino G, Candi E, Declercq W. p63, a story of mice and men. J Invest Dermatol 2011; 131: 1196–207. 13. Asioli S, Righi A, Volante M, Eusebi V, Bussolati G. p63 expression as a new prognostic marker in Merkel cell carcinoma. Cancer 2007; 110: 640–7. 14. Asioli S, Righi A, de Biase D, et al. Expression of p63 is the sole independent marker of aggressiveness in localised (stage I-II) Merkel cell carcinomas. Mod Pathol 2011; 24: 1451–61. 15. Lim CS, Whalley D, Haydu LE, et al. Increasing tumor thickness is associated with recurrence and poorer survival in patients with Merkel cell carcinoma. Ann Surg Oncol 2012; 19: 3325–34. 16. Beer TW, Ng LB, Murray K. Mast cells have prognostic value in Merkel cell carcinoma. Am J Dermatopathol 2008; 30: 27–30. 17. Team RC. R: A Language and Environment for Statistical Computing. Vienna: R Foundation for Statistical Computing, 2012. 18. Gies P, Roy C, Javorniczky J, Henderson S, Lemus-Deschamps L, Driscoll C. Global Solar UV Index: Australian measurements, forecasts and comparison with the UK. Photochem Photobiol 2004; 79: 32–9. 19. Australian Government, Bureau of Meteorology. Average daily solar exposure. 26 Feb 2013; cited 19 Jun 2013. http://www.bom.gov.au/jsp/ ncc/climate_averages/solar-exposure/index.jsp.). 20. Rodig SJ, Cheng J, Wardzala J, et al. Improved detection suggests all Merkel cell carcinomas harbor Merkel polyomavirus. J Clin Invest 2012; 122: 4645–53.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
210
DABNER et al.
21. Harms P, Patel R, Verhaegen M, et al. Distinct gene expression profiles of viral- and nonviral-associated Merkel cell carcinoma revealed by transcriptome analysis. J Invest Dermatol 2013; 133: 936–45. 22. Martin B, Poblet E, Rios JJ, et al. Merkel cell carcinoma with divergent differentiation: histopathological and immunohistochemical study of 15 cases with PCR analysis for Merkel cell polyomavirus. Histopathology 2013; 62: 711–22. 23. Albores-Saavedra J, Batich K, Chable-Montero F, Sagy N, Schwartz AM, Henson DE. Merkel cell carcinoma demographics, morphology, and survival based on 3870 cases: a population based study. J Cutan Pathol 2009; 37: 20–7. 24. Fields RC, Busam KJ, Chou JF, et al. Five hundred patients with Merkel cell carcinoma evaluated at a single institution. Ann Surg 2011; 254: 465– 73.
Pathology (2014), 46(3), April
25. Lemos BD, Storer BE, Iyer JG, et al. Pathologic nodal evaluation improves prognostic accuracy in Merkel cell carcinoma: analysis of 5823 cases as the basis of the first consensus staging system. J Am Acad Dermatol 2010; 63: 751–61. 26. Mojica P, Smith D, Ellenhorn JD. Adjuvant radiation therapy is associated with improved survival in Merkel cell carcinoma of the skin. J Clin Oncol 2007; 25: 1043–7. 27. Leonard JH, Bell JR, Kearsley JH. Characterization of cell lines established from Merkel-cell (’small-cell’) carcinoma of the skin. Int J Cancer 1993; 55: 803–10. 28. Van Gele M, Boyle GM, Cook AL, et al. Gene-expression profiling reveals distinct expression patterns for Classic versus Variant Merkel cell phenotypes and new classifier genes to distinguish Merkel cell from small-cell lung carcinoma. Oncogene 2004; 23: 2732–42.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
