Determining the risks associated with atypical Spitz tumors is a continuing challenge for dermatopathologists. Since traditional prognosticators have been shown to be of limited utility in assessing these tumors and due to the lack of consensus among dermatopathologists, molecular techniques may help improve assessing the probability of aggressive behavior from these enigmatic tumors. A recent study by our group examined a series of 75 atypical Spitz tumors by FISH, and demonstrated that chromosomal copy number aberrations detected by FISH can be used to improve the classification and prognostication of atypical Spitz tumors. Cases with homozygous 9p21 deletion had the highest likelihood of developing an adverse event, while cases with 6p25 gain and 11q13 gain were of intermediate risk. Cases with a loss of 6q23 posed the lowest risk, which was similar to those atypical Spitz tumors without copy number changes.
Since the original description of Spitz nevi in 1948, assessing the risk for aggressive clinical behavior from an atypical Spitz tumor has been an ongoing challenge facing dermatopathologists Citation[1–3]. While the vast majority of atypical Spitz tumors will follow a benign clinical course, some cases fitting a histologic classification as atypical Spitz tumors will result in distant metastasis and death of the patient Citation[4–6]. Multiple studies have shown that histologic analysis alone is limited in its ability to identify those cases at greatest risk. Furthermore, inter-observer studies have shown a lack of consensus by expert dermatopathologists in evaluating atypical Spitz tumors. In fact, in an inter-observer study by Barnhill et al., some cases in which an expert panel favored a diagnosis of benign Spitz nevus resulted in distant metastasis and death of disease Citation[2].
In the last two decades, a number of molecular techniques have been developed to improve the classification and prognostication of melanocytic neoplasms. Using comparative genomic hybridization, it has been well established that unequivocally benign nevi and unequivocally malignant melanomas have clear differences at the chromosomal level Citation[7–10]. Specifically, more than 95% of melanomas have chromosomal copy number aberrations, which are mostly absent in nevi. There are a few exceptions such as 11p gains Citation[11], including the HRAS gene in a subset of Spitz nevi, and 3p deletions Citation[12] including the BAP1 gene in some cutaneous epithelioid neoplasms occasionally resembling Spitz nevi. However, because of the numerous challenges involved in performing a study for prognostication of atypical Spitz tumors, until recently studies addressing this question have been highly limited.
Challenges involved in performing a high-quality study addressing this issue include the following. First, in order to show that a given marker or chromosomal aberration discriminates aggressive from nonaggressive cases, the study needs to be sufficiently powered with a significant number of cases with and without aggressive behavior. Considering the rarity and infrequent result of distant metastasis and death from atypical Spitz tumors, this is truly a challenge and probably requires participation from multiple academic centers. Those cases without adverse events may require a minimum of 5–10 years follow-up since clinical follow-up is the optimal gold standard for such a study as histology is indeterminate as a prerequisite. Furthermore, it has not been clearly established what the minimum period of time is to conclude that most atypical Spitzoid neoplasms are unlikely to metastasize or prove fatal, since this has occurred as long as 15 years after diagnosis. Second, the study design needs well-defined and meaningful end points. Other than the presence of distant metastasis and death, there remains some uncertainty regarding other end points for atypical Spitz tumors. Only recently, in a critical study by Ludgate et al. at the University of Michigan (MI, USA), was it shown that while nearly half of atypical Spitz tumors may show involvement of sentinel lymph nodes, this finding is not predictive of further disease progression Citation[5]. Currently, it is presumed that atypical Spitz tumors with more advanced locoregional disease such as involvement of nonsentinel lymph nodes, palpable lymph nodes or in-transit disease have a high likelihood for distant metastasis and death. However, further follow-up studies of such patients are needed. Third, a panel of dermatopathologists experienced in melanocytic lesion pathology should reach a consensus about the consistent classification of lesions entered into the study. The cases should be judged as atypical Spitz tumors by dermatopathologists at the expert level, as the term may be used more liberally by some less experienced dermatopathologists, decreasing the homogeneity of the study group. Finally, if the study involves CGH, then sufficient tissue needs to be available for this; or if the study involves FISH, testing all the most critical loci, in particular 9p21 in the case of Spitz tumors, should be evaluated.
Recently, the authors orchestrated a study with the goal of fulfilling all of these requisites. In a multicenter case-controlled retrospective study involving multiple sites in the USA including Northwestern University (IL, USA), the University of Michigan, the Memorial Sloan–Kettering Cancer Center (NY, USA), MD Anderson (TX, USA), the University of San Francisco (CA, USA), the University of Pennsylvania (PA, USA) and the Sydney Melanoma Unit (NSW, Australia), the authors identified 75 atypical Spitz tumors with either 5 years follow-up with no evidence of adverse events (64 cases) or atypical Spitz tumors with advanced locoregional disease such as tumor spread to nonsentinel lymph nodes, palpable lymphadenopathy or in-transit metastasis (eight cases) or distant metastasis with death of disease (three cases) Citation[1]. All cases were histologically diagnosed as atypical Spitz tumors by dermatopathologists at the expert level involved in the study. Two unstained slide sections cut at 5 µm were obtained for each case allowing for hybridization with two different probe sets, one targeting 6p25, 6q23, 11q13 and Cep6, and the other targeting 6p25, 9p21, 11q13 and 8q24 was performed on each case Citation[13,14].
In addition to the molecular data, all of the American Joint Committee on Cancer data points typically used for prognostication of melanoma were also collected for all cases including age, sex, site, Breslow depth, Clark level, ulceration status and mitotic count Citation[15]. In addition, features such as the presence or absence of Kamino bodies, expansile nodule formation, epidermal effacement or consumption, and epithelioid versus spindle morphology were also recorded for each case. The data were then evaluated using Fisher’s exact test, logistic regression analysis and multivariate analysis to compare all the variables in the cases resulting in more advanced locoregional disease or death to those patients without these findings.
Remarkably, nine out of 11 atypical Spitz tumor cases with more advanced disease had homozygous 9p21 deletion detected by FISH. Alternatively, only three out of 64 patients with disease limited to sentinel lymph node involvement after 5 or more years had homozygous 9p21 deletion. Hence, there was a highly statistically significant relationship between homozygous 9p21 deletion and the group of patients with more advanced tumor spread, with a p-value of <0.0001 and for death from disease, a p-value of 0.003. Other copy number aberrations including 6p25 gain and 11q13 gain also showed a statistically significant relationship with more advanced tumor spread, but with a p-value of 0.02 in both cases. In multivariate analysis, homozygous 9p21 deletions and mitotic count were the only variables maintaining a statistically significant relationship with advanced tumor spread, with p-values of <0.0001 and 0.03, respectively. Only homozygous 9p21 deletion had a statistically significant relationship with death of disease in multivariate analysis – p = 0.01.
A number of highly significant findings were revealed by this study. First, the relationship between homozygous 9p21 deletion identified by FISH and the adverse prognosis of atypical Spitz tumors. Second, among the patients with a more advanced disease, a sentinel lymph node biopsy was positive in eight out of eight patients where it was performed. In the patient group without tumor spread beyond the sentinel lymph node, a sentinel lymph node biopsy was positive in 13 out of 19 cases. What this suggests is that most atypical Spitz tumors that result in aggressive clinical behavior are highly likely to have a positive sentinel lymph node. However, the high rate of sentinel lymph node positivity in the group without tumor spread beyond the sentinel lymph node expunges the value of the test as a prognosticator in atypical Spitz tumors.
Moreover, interestingly, four patients, all from separate institutions, showed a markedly similar course involving the presence of an atypical spitzoid melanocytic neoplasm with many homozygous 9p21 deletions and persistent in-transit disease/satellitosis around the primary tumor site, with four out of four cases also showing involvement of sentinel lymph nodes. Three of these four patients had completion dissection, and all three also had evidence of tumor involvement of nonsentinel lymph nodes. Hence, a common pattern seen in atypical Spitz tumors with an aggressive clinical course appears to be progression through a stage of advanced locoregional disease. Finally, none of the patients with isolated 6q23 deletions showed any evidence of tumor spread beyond a sentinel lymph node. This further emphasizes the fact that not all chromosomal copy number aberrations are of equal value in predicting clinical behavior for atypical Spitz tumors.
In conclusion, the findings from this study indicate that certain chromosomal copy number aberrations can be used to classify and prognosticate atypical Spitz tumors. Most importantly, homozygous 9p21 deletion clearly has a highly statistically significant relationship with disease progression in atypical Spitz tumors. Atypical Spitz tumors with homozygous 9p21 deletions are a fairly homogeneous subgroup of melanocytic tumor with a high risk for aggressive clinical behavior typically proceeding through a prolonged stage of advanced locoregional disease, often including both locoregional lymph nodes and in-transit metastasis before proceeding to distant metastasis and death. Conversely, atypical Spitz tumors with isolated 6q23 deletions have a significantly lower risk for tumor spread, with a risk level similar to atypical Spitz tumors without chromosomal copy number aberrations. Other copy number aberrations such as 6p25 gain and 11q13 gain also have a statistically significant relationship to more aggressive behavior, but to a lesser extent than homozygous 9p21 deletions. Further similar studies validating the role of cytogenetics to subclassify atypical Spitz tumors into prognostically significant subgroups are needed.
Financial & competing interests disclosure
P Gerami has received honoraria for consultancy to Abbott Molecular Laboratories, Myriad Genomics and Neogenomics. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
No writing assistance was utilized in the production of this manuscript.
References
- Gerami P, Scolyer RA, Xu X et al. Risk assessment for atypical spitzoid melanocytic neoplasms using FISH to identify chromosomal copy number aberrations. Am. J. Surg. Pathol. 37(5), 676–684 (2013).
- Barnhill RL, Argenyi ZB, From L et al. Atypical Spitz nevi/tumors: lack of consensus for diagnosis, discrimination from melanoma, and prediction of outcome. Hum. Pathol. 30(5), 513–520 (1999).
- Spitz S. Melanomas of childhood. Am. J. Pathol. 24(3), 591–609 (1948).
- Sepehr A, Chao E, Trefrey B et al. Long-term outcome of Spitz-type melanocytic tumors. Arch. Dermatol. 147(10), 1173–1179 (2011).
- Ludgate MW, Fullen DR, Lee J et al. The atypical Spitz tumor of uncertain biologic potential: a series of 67 patients from a single institution. Cancer 115(3), 631–641 (2009).
- Zembowicz A, Scolyer RA. Nevus/melanocytoma/melanoma. An emerging paradigm for classification of melanocytic neoplasms? Arch. Pathol. Lab. Med. 135(3), 300–306 (2010).
- Bastian BC, LeBoit PE, Hamm H, Bröcker EB, Pinkel D. Chromosomal gains and losses in primary cutaneous melanomas detected by comparative genomic hybridization. Cancer Res. 58(10), 2170–2175 (1998).
- Bastian BC, Wesselmann U, Pinkel D, Leboit PE. Molecular cytogenetic analysis of Spitz nevi shows clear differences to melanoma. J. Invest. Dermatol. 113(6), 1065–1069 (1999).
- Bastian BC, Xiong J, Frieden IJ et al. Genetic changes in neoplasms arising in congenital melanocytic nevi: differences between nodular proliferations and melanomas. Am. J. Pathol. 161(4), 1163–1169 (2002).
- Bastian BC, Olshen AB, LeBoit PE, Pinkel D. Classifying melanocytic tumors based on DNA copy number changes. Am. J. Pathol. 163(5), 1765–1770 (2003).
- Bastian BC, LeBoit PE, Pinkel D. Mutations and copy number increase of HRAS in Spitz nevi with distinctive histopathological features. Am. J. Pathol. 157(3), 967–972 (2000).
- Wiesner T, Obenauf AC, Murali R et al. Germline mutations in BAP1 predispose to melanocytic tumors. Nat. Genet. 43(10), 1018–1021 (2011).
- Gerami P, Jewell SS, Morrison LE et al. Fluorescence in situ hybridization (FISH) as an ancillary diagnostic tool in the diagnosis of melanoma. Am. J. Surg. Pathol. 33(8), 1146–1156 (2009).
- Gerami P, Li G, Pouryazdanparast P et al. A highly specific and discriminatory FISH assay for distinguishing between benign and malignant melanocytic neoplasms. Am. J. Surg. Pathol. 36(6), 808–817 (2012).
- Balch CM, Gershenwald JE, Soong SJ et al. Final version of 2009 AJCC melanoma staging and classification. J. Clin. Oncol. 27(36), 6199–6206 (2009).