Spitz Nevi Display Allelic Deletions

Background  Spitz nevi are acquired benign melanocytic lesions that occur in childhood and adolescence. Histologically, they resemble malignant melanoma and were first termed benign juvenile melanoma. Several studies have attempted the difficult task of establishing diagnostic criteria to differentiate between Spitz nevi and malignant melanoma.

Objective  To elucidate sets of diagnostic criteria for differentiation between the 2 lesions.

Design  We aimed to search for allelic deletions in Spitz nevi and to evaluate whether loss of heterozygosity (LOH) or microsatellite instability (MSI) would be a valuable diagnostic tool to differentiate between Spitz nevi and malignant melanoma.

Setting  Two areas within each of 5 lesions were microdissected, and LOH and MSI were evaluated at chromosomes 6q (using polymorphic DNA marker D6S305), 9p21 (D9S171, IFNA, D9S265, and D9S270), 10q (D10S185), and 14q (D14S53).

Patients  Five Swiss patients with Spitz nevi.

Interventions  None.

Main Outcome Measure  Allelic deletions may serve as a diagnostic tool to distinguish Spitz nevi from melanoma.

Results  All lesions were informative, displaying LOH or MSI with at least one marker. No LOHs were found at 14q. At 6q, MSI was found in 2 dissected areas from the same lesion; the remaining lesions were noninformative. Loss of heterozygosity was found in 2 of 6 areas at D9S171, 2 of 6 at IFNA, 3 of 6 at D9S270, 3 of 4 at D9S265, and 1 of 4 at D10S185. Microsatellite instability was found in 1 of 4 areas at D9S265.

Conclusions  With the markers used in our study, Spitz nevi display LOH and MSI similar to those in melanoma. Analysis of LOH or MSI is therefore not a suitable diagnostic tool in distinguishing Spitz nevi from melanoma.

SPITZ NEVI are melanocytic lesions that most commonly occur in childhood and adolescence and behave in a benign way. However, they show histopathological features that are sometimes impossible to distinguish from those of malignant melanoma, which can present difficulties regarding diagnosis and subsequent therapy. For that reason, several studies^1-4 have tried to establish diagnostic criteria to differentiate between the 2 lesions. Loss of heterozygosity (LOH) has, to our knowledge, thus far only been addressed by Healy et al,⁵ who reported LOH at chromosome 9p in a small percentage of Spitz nevi studied.

Our aim was to examine Spitz nevi for possible genetic differences relative to those of melanoma and to present a genetic diagnostic method to distinguish between the 2 lesions. The molecular events we focused on were LOH and microsatellite instability (MSI), which occur in melanoma.^5-8 We evaluated chromosomes 6q, 9p21, 10q, and 14q, which are also involved in genetic changes in melanoma.^5,9

We analyzed archived material from 5 patients (3 female and 2 male patients; mean age, 27.2 years, range 10-57 years). In all patients, there was no record of a subsequent malignant tumor or metastasis 5 years or longer after excision of the Spitz nevi. All existing paraffin-embedded tissue sections were histopathologically examined by an experienced dermatologist and subsequently analyzed for LOH and MSI.

We looked at 10 areas from 5 lesions. Histologic criteria used for Spitz nevi are those described by Barnhill.¹⁰ Briefly, nests of large nevus cells are spindle-shaped or epithelioid, with centrally located nuclei and smooth and regular contours. Typically, the appearance of the cells and nuclei is uniform. Furthermore, Spitz nevi have few deeply located mitoses, or none, and there is little cytologic atypia. Overall, the nevus cells are regular and symmetric, with epidermal hyperplasia and hyperkeratosis. The lesions were excised from the face in 2 patients and from the back, buttock, and lower extremity in 1 patient each.

In each case, a 5-µm paraffin-embedded tissue section was obtained for hematoxylin-eosin staining and microdissection. Microdissection was performed under light microscope (original magnification ×200). In each case, 2 areas of 10 to 50 nevus cells each were dissected from the same slide (Figure 1). Nevus cells were selectively removed with a disposable 30-gauge needle. In addition, normal epidermal cells from the same slide were procured as a control.

Procured cells were immediately suspended in a 20-µL solution containing 0.05M solution of Tris hydrochloride, 1mM solution of EDTA, 1% Tween 20, and 2.5 mg/mL proteinase K (pH 8.0) and incubated overnight at 37°C. The mixture was boiled for 10 minutes at 94°C to inactivate the proteinase K, and 1.5 µL of this solution was used as template DNA in the polymerase chain reaction amplification.

Seven polymorphic DNA markers (Research Genetics, Huntsville, Ala) were used: D9S171, IFNA, D9S265, and D9S270 for chromosome 9p21; D6S305 for 6q; D10S185 for 10q; and D14S53 for 14q. The polymerase chain reaction was performed in 10-µL volumes that contained 1.5 µL of template DNA; 50 pmol of each primer per liter; 20 nmol/L each of deoxyadenosine triphosphate, deoxycytidine triphosphate, deoxyguanosine triphosphate, and deoxythymidine triphosphate; 0.2 µL (32P) of deoxycytidine triphosphate (22 200 × 10¹⁰ Bq/mmol); 0.1 U of Taq DNA polymerase; and 1 µL of 10+ buffer (100mM solution of Tris hydrochloride [pH 8.3], 500mM solution of potassium chloride, 15mM solution of magnesium chloride, and 0.1% weight-volume gelatin). Each sample was overlaid with mineral oil and amplified in a thermal cycler (Gene Amp PCR System 9600; Perkin Elmer, Zurich, Switzerland) for 35 cycles by denaturing at 94°C for 1 minute, annealing 1 minute at the temperature already described for each primer, and extending 1 minute at 72°C, with a final 10-minute extension at 72°C.

Labeled amplified DNA was mixed with an equal volume of formamide loading dye (95% formamide, 20mM solution of EDTA, 0.05% bromophenol blue, and 0.05% xylene cyanol). Samples were then denatured for 5 minutes at 94°C, followed by rapid cooling, and loaded onto a commercially available 8% acrylamide gel (Gel-Mix8; GIBCO BRL, Gaithersburg, Md). The gel was run at 1600 V for 2 hours. After electrophoresis, the gel was transferred to 3-mm Whatman paper (Merck & Co Inc, Zurich) and dried. Autoradiography was performed with X-Ray DX-41 film (Typon, Burgdorf, Switzerland). A case was considered informative for a polymorphic marker if normal tissue DNA showed 2 different alleles. Loss of heterozygosity of informative polymorphic loci was visually evaluated by comparing allele band intensity of normal epidermal layer relative to that of nevus cells from the same patient. Complete loss or more than 50% reduction in a single band intensity in an informative locus was designated as LOH by 2 independent observers. A lesion was considered positive for MSI when allele bands from the patients showed a mobility shift compared with those from the controls.

We analyzed 5 Spitz nevi from 5 patients. From each slide, we procured 10 to 50 cells from 2 different areas by microdissection and evaluated the cells for LOH and MSI, thereby examining 10 areas from 5 Spitz nevi. Using polymorphic DNA markers, we searched for LOH at chromosomes 9p21, 6q, 10q, and 14q.

All lesions were heterozygous with at least one marker and were included in our analysis. Overall, we found LOH in 2 of 6 areas at D9S171 and IFNA (4 homozygous); 3 of 4 areas at D9S265 (6 noninformative), and 3 of 6 areas at D9S270 (4 noninformative). At 6q, MSI was found in 2 areas from the same lesion (the rest were noninformative), and at D9S265, MSI was found in 1 of 4 areas. Also, in 1 of 4 areas, an allelic deletion was found with the marker D10S185, while no LOH or MSI was found at 14q.

Genetic heterogeneity was demonstrated, with 1 of 5 Spitz nevi showing genetic changes (LOH or MSI) at 3 different DNA markers examined, and 3 of 5 Spitz nevi showing genetic changes at 2 markers. In summary, 3 of 5 Spitz nevi showed genetic heterogeneity with at least 2 polymorphic markers, as demonstrated by the finding of LOH or MSI (Table 1 and Figure 2).

Spitz nevi were first described in 1948 by Sophie Spitz¹¹ as "melanoma of childhood" or "juvenile melanoma" and were considered to be a malignant neoplasm of childhood. A few years later, however, it was recognized that "juvenile melanoma" also occurred in adults and exhibited an excellent prognosis, indicating its benign nature.¹² In the years that followed, several case reports on this topic were published, and histologic criteria were reevaluated. Subsequently, Spitz nevus was renamed spindle-cell nevus, and finally was termed Spitz nevus and classified as a benign melanocytic lesion.^13-17 Spitz nevi account for about 1% of surgically removed nevi.¹⁸ They show histopathological features that are similar to those of malignant melanoma, and, in some cases, the 2 lesions are impossible to distinguish. Difficulties in differentiating between Spitz nevi and malignant melanoma prompted studies^19,20 that focused on the definition of clear histopathological criteria to distinguish between the 2 lesions. Histopathological criteria were evaluated, accompanied by immunohistochemical, ultrastructural, and molecular studies.^1-4,21-24 The only study, to our knowledge, that searched for LOH in Spitz nevi was by Healy et al,⁵ who reported interstitial deletions at chromosome 9p in fewer than 10% of Spitz nevi.

We evaluated the significance of LOH and MSI in Spitz nevi because these 2 genetic changes frequently occur in melanoma and, in some cases, in atypical nevi.^5,6,25,26 We focused on chromosomes 6q, 9p21, 10q, and 14q because they often display genetic alterations in melanoma.^5-9,26 Specifically, 9p21, which harbors the tumor suppressor gene p16, is of major interest because it plays an important role in the development of melanoma.^5,27

Another question we addressed was whether Spitz nevi are genetically heterogeneous. Previous studies^9,28 demonstrated that primary melanomas and melanoma metastases display some genetic heterogeneity. Therefore, we dissected 2 different areas within each lesion to test for heterogeneity.

In all Spitz nevi examined, we found LOH at 9p21 with at least one marker. Furthermore, we found MSI at 6q in one Spitz nevus. However, because of the small sample, conclusions cannot be drawn about the frequency of such genetic changes in Spitz nevi. In addition, the pattern of LOH using different DNA markers demonstrates genetic heterogeneity within the Spitz nevi, which has already been shown in primary melanoma and melanoma metastases.²⁸

In summary, by showing genetic changes in Spitz nevi that are similar to those in melanoma, we conclude that neither LOH nor MSI analysis is a suitable method to differentiate between Spitz nevi and malignant melanoma.

Accepted for publication June 26, 2001.

Corresponding author and reprints: Roland Böni, MD, Department of Dermatology, University Hospital Zurich, Gloriastr 31, 8091 Zurich, Switzerland (e-mail: rboeni@derm.unizh.ch).