This article is mentioned in:

Introduction

Melanoma is a neoplasm with numerous chromosomal aberrations that can be difficult to differentiate from atypical nevi, particularly during the early stages of its development (1,2). Chromosomal instability (CIN) is a hallmark of all neoplasms and is highly prevalent in melanoma, primarily being caused by exposure to ultraviolet radiation, genetics or other factors (angiogenesis, stress and smoking, among others) (2-4). CIN is characterized by an increased rate of chromosomal missegregation leading to aneuploidy, which may also occur in benign tissues (particularly in Spitz nevi) (2,5).

Differentiating between melanoma in its early stages from atypical nevi presents a pathological challenge that can potentially result in misdiagnosis due to the similarities in their characteristics (2,6). It has been demonstrated that melanomas can be differentiated from nevi based on the loss or gain of certain chromosomal regions (1). The gold standard in melanoma diagnosis is histopathological examination; however, additional tests can also be used to obtain a more accurate diagnosis. One of these complementary diagnostic techniques is fluorescence in situ hybridization (FISH), which can be used to assist in the diagnosis of melanoma using skin biopsy specimens. FISH facilitates the visualization of nucleic acid sequences and involves precise annealing of a single-stranded, fluorophore-labeled DNA probe to complementary target sequences. The hybridization of the DNA probe be visualized using an epi-illumination fluorescence microscope. FISH can identify various chromosomal aberrations that may have contributed to the occurrence of melanoma; it is used to detect the duplication/amplification, deletion or translocation of certain chromosomes [for example, gains of 6p25 (RREB1) and 11q13 (CCND1), and losses of 9p21 (CDKN2A)], making it a valuable tool for diagnosis and prognosis, predicting the response to therapy or monitoring patients to assess treatment efficacy (2). In the present study, signal gains for RREB1 and CCND1 were frequently observed. Additionally, MYB and CEP6 aberrations were also identified, supporting the utility of FISH for detecting diagnostically relevant chromosomal changes.

Lai et al (7) reported that the Vysis Melanoma FISH Probe Kit (Abbott Pharmaceutical Co. Ltd.) was a valuable ancillary tool and highly sensitive in distinguishing early-stage acral and cutaneous melanomas from dysplastic nevi in the Chinese population. Also, according to Vergier et al (8), the Vysis Melanoma FISH Probe kit achieved a sensitivity of 85%, a specificity of 90%, a positive predictive value of 89.5% and a negative predictive value of 86%. While the sensitivity and specificity of FISH can be high, it is essential to rely on clinicopathological examinations to confirm the final diagnosis and treatment decision, particularly when assessing ambiguous melanocytic lesions (9). Therefore, although FISH has proven to be a helpful tool in assisting clinicians to establish a diagnosis it should be used as a complementary diagnostic technique due to the potential for false-positive or false-negative results (10).

In the present study, based on these findings, the aim was to assess the reliability of the FISH test, through a small-scale investigation conducted on 18 melanocytic lesions previously diagnosed as either atypical nevi or melanoma by dermoscopy (clinical examination) and histopathology. A 4-probe set targeting 6p25 [Ras responsive element binding protein 1 (RREB1)], 6q23 [v-myb myeloblastosis viral oncogene homologue (MYB)], centromere 6 (CEP6) and 11q13 [cyclin D1 (CCND1)] was used (2). This kit was chosen due to the various aforementioned studies that indicated that it was reliable and produced good results.

Materials and methods

Sample collection and processing

A total of 20 melanocytic lesions that had previously been diagnosed through dermoscopy and histopathological examination between January 2017 and December 2022 at a private dermatology office in Sibiu (Romania) and at the Emergency Hospital of Sibiu County (Sibiu, Romania) were initially obtained. Probes were collected from 20 patients (9 men and 11 women), aged between 30 and 84 years (mean age, 56.2 years). Patients were included in the study based on the following criteria: Adult patients diagnosed with atypical nevi or melanoma. Patients were excluded if they were minors, patients with non-melanocytic tumors and patients with common (non-atypical) nevi. However, two lesions (cases 15 and 16) were excluded in the FISH pre/post-treatment phase of the study due to primary processing artifacts (tissue quality/age and fixation peculiarities). The study was conducted in 2023 at a private analysis center with the help of two anatomopathologists and a dermatologist. Of the 18 remaining lesions, five were histopathologically confirmed as melanomas, eight were ambiguous melanocytic lesions that were clinically diagnosed as melanomas, but later proved to be atypical nevi on the histopathological examination and the last five lesions were from patients with atypical mole syndrome and were used for research purposes. Therefore, the study was divided into two categories: FISH testing on melanocytic tumors and FISH-histopathological associations. The lesions were histopathologically analyzed according to the American Joint Committee on Cancer 8th edition TNM classification for melanoma (11) for tumoral thickness (T1, ≤1 mm; T2, >1- ≤2 mm; T3, >2- ≤4 mm; and T4, >4 mm in thickness), differentiation (G-), cell infiltrates, the presence of emboli and mitoses, which are important prognostic factors for localized primary melanomas where a higher mitotic count is indicative of a higher likelihood of the tumor having metastasized, as well as the presence of pigment. Histopathology is considered the gold standard in melanoma diagnosis; however, it should also be used in association to clinical history and examination of the patient (12,13). Typically, FISH is used solely on ambiguous melanocytic lesions to confirm their melanoma nature; however, in the present study, FISH was also used to confirm atypical nevi to assess its reliability for research-related reasons. In the present study, the Vysis Melanoma FISH Probe kit (catalogue number 30-608220/R6; Abbott Pharmaceutical Co. Ltd.) was used to test the 18 melanocytic lesions. This kit can detect the copy numbers of the RREB1 (6p25), MYB (6q23) and CCND1 (11q13) genes and of CEP6 in formalin-fixed, paraffin-embedded human skin biopsy specimens. The RREB1 (6p25) probe is labeled with SpectrumRed and covers a 638 kb region that contains the RREB1 gene. The MYB (6q23) probe is labeled with SpectrumGold and covers a 743 kb region that contains the MYB gene. The CCND1 (11q13) probe is labeled with SpectrumGreen and covers a 378 kb region that contains the CCND1 gene. The CEP6 probe, labeled with SpectrumAqua, hybridizes with the α satellite DNA located at the centromere of chromosome 6 (6p11.1-q11). The specimens were prepared by two anatomopathologists; this included sectioning of formalin-fixed, paraffin-embedded tissue specimens and preparing slides. The tissue samples were fixed in 10% neutral buffered formalin at room temperature for 24 h. Sections were cut at a thickness of 4 µm. All specimens were histopathologically re-verified to confirm their diagnosis. The specimens were subjected to DNA denaturation (the DNA was denatured into single-stranded DNA) at 80˚C for 5 min and then hybridized using the RREB1/MYB/CCND1/CEP probes at 37˚C for 24 h (pretreatment). After hybridization, the slides were washed with a washing buffer solution (post-treatment), wash buffer 2X SSC, pH 7.0, 0.3% Nonidet P-40, to remove the unbound probes, and subsequently the slides were dried and the nuclei were stained with DAPI [at room temperature (20-25˚C) for 10-15 min]. Then, the slides were covered using lamellas and stored in a freezer according to the manufacturer's instructions (-20˚C). Slides were examined using a fluorescence microscope (Olympus BX61; Olympus Corporation). Each slide was analyzed by an anatomopathologist and a dermatologist who manually counted all the FISH signals. The medical personnel who performed the FISH test had received formal training in the FISH technique as part of their medical specialty and professional development. They were experienced in performing FISH, including sample preparation, hybridization and fluorescence signal interpretation. Quality control steps were implemented to ensure accuracy and consistency, including supervision and validation of the procedures by experienced personnel, adherence to established protocols, use of control samples [previously diagnosed benign lesions (e.g., atypical nevi) within the sample cohort] to differentiate true signals from artifacts and review of results to minimize subjective bias.

Statistical analysis

Data were handled using Microsoft Excel (Office 365 version 2412; Microsoft Corporation) and are presented as frequency (%). SPSS version 23 (IBM Corp.) was used for statistical analysis. For the comparison of two qualitative variables, an association table was used (Crosstabs). The significance level of the Likelihood ratio test was considered. The results were interpreted based on the instructions for melanoma status determination provided by the manufacturer of the kit as follows: i) The number of RREB1, MYB, CCND1 and CEP6 counts in each of 30 nuclei (out of 90 nuclei); ii) the number of RREB1 nuclei with aberrant signals (<2 or >2) and the percentage of abnormal nuclei for RREB1 signals; iii) the sum and median of MYB signals per 90 nuclei; iv) the number and median of CCND1 signals per 90 nuclei; and v) the percentage and number of nuclei that showed a loss for MYB relative to CEP6. The diagnosis (FISH+/FISH-) was based on either a median value of CCND1 signals per nuclei or a median of MYB signals per nuclei ≥2.5, a percentage loss of MYB relative to CEP6 of ≥31% or a percentage of abnormal nuclei for RREB1 of ≥63% (FISH+ diagnosis). A FISH- diagnosis was considered if none of the diagnostic criteria were met. Histopathological association was performed for the lesions that had ambiguous FISH results. P<0.05 was considered to indicate a statistically significant difference.

Results

FISH testing on melanocytic tumors

A total of 18 melanocytic lesions were analyzed based on the manufacturer's instructions for the Vysis Melanoma FISH Probe kit. The results for each examined gene are provided below on a case-by-case basis.

The diagnosis of the cases was as follows: Case 1, nodular melanoma; case 2, lentigo maligna melanoma; case 3, melanoma; case 4, lentigo simplex; case 5, junctional atypical nevus; case 6, lentiginous junctional atypical nevus; case 7, junctional atypical nevus; case 8, Spitz nevus; case 9, lentiginous melanocytic proliferation; case 10, mixed atypical nevus; case 11, junctional atypical nevus; case 12, mixed atypical nevus; case 13, mixed atypical nevus; case 14, mixed atypical nevus; case 17, melanoma; case 18, mixed atypical nevus; case 19, atypical nevus; and case 20, lentigo maligna melanoma.

RREB1 (6p25). The number of signals in each of 30 out of 90 nuclei were counted and the lowest number of counts was 52, found in case 14 (compound atypical nevus). The highest number of signals was 94 in case 17 (melanoma) (Table I). The sum of the signals for RREB1 in each case was calculated and the lowest sum of signals value (163 signals) was found in case 14, while 4 cases had >200 signals: Case 2 (lentigo maligna melanoma; 206 signals), case 3 (melanoma; 229 signals), case 9 (lentiginous melanocytic proliferation; 213 signals) and case 17 (melanoma; 242 signals) (Table II). There was a mean of 183.55 signals/case for RREB1 (median, 172.00 signals/case; IQR, 11.25 signals/case).

Table I Number of RREB1, MYB, CCND1 and CEP6 counts in each of 30 out of 90 nuclei.

Table I

Number of RREB1, MYB, CCND1 and CEP6 counts in each of 30 out of 90 nuclei.

Case no.	RREB1 0-30 nuclei	RREB1 31-60 nuclei	RREB1 61-90 nuclei	MYB 0-30 nuclei	MYB 31-60 nuclei	MYB 61-90 nuclei	CEP6 0-30 nuclei	CEP6 31-60 nuclei	CEP6 61-90 nuclei	CCND1 0-30 nuclei	CCND1 31-60 nuclei	CCND1 61-90 nuclei
Case 1	58	62	65	57	66	67	60	56	56	56	75	66
Case 2	68	66	72	61	62	72	60	56	57	71	69	72
Case 3	83	73	73	64	63	73	62	63	64	74	77	77
Case 4	55	58	54	53	57	59	52	56	55	56	53	56
Case 5	59	57	57	65	57	56	56	60	57	62	60	54
Case 6	56	60	60	55	52	49	56	57	55	56	55	56
Case 7	61	59	60	55	54	54	55	56	54	54	49	53
Case 8	54	56	54	52	60	57	50	53	54	52	58	56
Case 9	68	72	73	76	67	71	57	54	55	79	89	63
Case 10	59	63	60	58	64	60	54	50	58	67	61	58
Case 11	57	61	57	61	56	56	56	58	56	55	58	57
Case 12	58	58	55	55	55	56	59	55	58	54	57	61
Case 13	57	55	58	58	54	57	60	58	57	56	60	54
Case 14	54	57	52	58	56	54	57	52	58	53	58	54
Case 17	68	80	94	62	72	58	59	56	55	59	57	57
Case 18	57	59	54	55	57	53	56	53	58	56	53	54
Case 19	61	54	55	57	56	53	57	56	55	54	56	56
Case 20	57	56	55	53	56	54	58	55	57	56	53	56

[i] RREB1, Ras responsive element binding protein 1; MYB, v-myb myeloblastosis viral oncogene homologue; CEP6, centromere 6; CCND1, cyclin D1.

Table II Sum of RREB1, MYB, CEP6 and CCND1 signals per 90 nuclei.

Table II

Sum of RREB1, MYB, CEP6 and CCND1 signals per 90 nuclei.

Case no.	RREB1	MYB	CEP6	CCND1
Case 1	185	190	172	197
Case 2	206	195	173	212
Case 3	229	200	189	228
Case 4	167	169	163	165
Case 5	173	178	173	176
Case 6	176	156	168	167
Case 7	180	163	165	156
Case 8	164	169	157	166
Case 9	213	214	166	231
Case 10	182	182	162	186
Case 11	175	173	170	170
Case 12	171	166	172	172
Case 13	170	169	175	170
Case 14	163	168	167	165
Case 17	242	192	170	173
Case 18	170	165	167	163
Case 19	170	166	168	166
Case 20	168	163	170	165

[i] RREB1, Ras responsive element binding protein 1; MYB, v-myb myeloblastosis viral oncogene homologue; CEP6, centromere 6; CCND1, cyclin D1.

The number of abnormal nuclei for RREB1 (with <2 or >2 signals per nuclei) was counted and the percentage was next calculated. The lowest number of abnormal nuclei for RREB1 was found in case 12 (mixed atypical nevus; 9 abnormal nuclei) and the highest number was encountered in case 17 (melanoma; 45 abnormal nuclei out of 90; Fig. 1). Concerning the percentage of abnormal nuclei for RREB1, the highest percentages were as follows: 50.0% (Case 17; melanoma), 42.2% (case 3; melanoma) and 31.1% (case 2; lentigo maligna melanoma) (Table III).

Figure 1 Ras responsive element binding protein 1 (red) signals in a patient with melanoma (case 17).

Table III Percentage of abnormal nuclei for RREB1.

Table III

Percentage of abnormal nuclei for RREB1.

Case no.	Number of abnormal nuclei for RREB1 signals	Percentage of abnormal nuclei for RREB1
Case 1	17	18.8
Case 2	28	31.1
Case 3	38	42.2
Case 4	25	27.7
Case 5	17	18.8
Case 6	12	13.3
Case 7	13	14.4
Case 8	21	23.3
Case 9	24	26.7
Case 10	12	13.3
Case 11	12	13.3
Case 12	9	10.0
Case 13	10	11.1
Case 14	14	15.5
Case 17	45	50.0
Case 18	12	13.3
Case 19	14	15.5
Case 20	12	13.3

[i] RREB1, Ras responsive element binding protein 1.

MYB (6q23) and CEP6. The lowest number of counts (49 signals) was found in case 6 (atypical lentiginous junctional nevus) and the highest number of signals (76 signals) was found in case 9 (lentiginous melanocytic proliferation) (Table I). The sum of MYB signals was also calculated and the lowest sum (156 signals) was found in case 6, while 2 cases had ≥200 signals: Case 3 (melanoma; 200 signals) and case 9 (melanocytic proliferation; 214 signals) (Table II). There was a mean of 176.55 signals/case for MYB (median, 169.00 signals/case; IQR, 24.75 signals/case).

The sum of MYB signals for each case was counted. The lowest number of signals was found in case 6 (junctional atypical nevus; 156 signals) and the highest number of signals was observed in cases 9 (lentiginous melanocytic proliferation; 214 signals) and 3 (melanoma; 200 signals). As for the median values of MYB signals, there were 6 cases with a median of >2 signals per nuclei: Case 1 (nodular melanoma; 2.11 signals/nuclei), case 2 (lentigo maligna melanoma; 2.16 signals/nuclei), case 3 (melanoma; 2.22 signals/nuclei), case 9 (lentiginous melanocytic proliferation; 2.37 signals/nuclei), case 10 (mixed atypical nevus; 2.02 signals/nuclei), and case 17 (melanoma; 2.13 signals/nuclei; Fig. 2) (Table IV).

Figure 2 V-myb myeloblastosis viral oncogene homologue (gold) signals in a patient with melanoma (case 17).

Table IV Sum and median of MYB and CCND1 signals.

Table IV

Sum and median of MYB and CCND1 signals.

Case no.	Sum of MYB signals	Median of MYB signals	Sum of CCND1 signals	Median of CCND1 signals
Case 1	190	2.11	197	2.18
Case 2	195	2.16	212	2.35
Case 3	200	2.22	228	2.53
Case 4	169	1.87	165	1.88
Case 5	178	1.97	176	1.95
Case 6	156	1.73	167	1.85
Case 7	163	1.81	156	1.73
Case 8	169	1.87	166	1.84
Case 9	214	2.37	231	2.56
Case 10	182	2.02	186	2.06
Case 11	173	1.92	170	1.88
Case 12	166	1.84	172	1.91
Case 13	169	1.87	170	1.88
Case 14	168	1.86	165	1.83
Case 17	192	2.13	173	1.92
Case 18	165	1.83	163	1.81
Case 19	166	1.84	166	1.84
Case 20	163	1.81	165	1.83

[i] MYB, v-myb myeloblastosis viral oncogene homologue; CCND1, cyclin D1.

Next, the number of CEP6 signals in each of 30 out of 90 nuclei was counted. The lowest number of counts (50 signals) was found in cases 8 (spitz nevus) and 10 (atypical compound nevus). The highest number of counts (64 signals) was observed in case 3 (melanoma) (Table I). The sum of signals per case was counted; the lowest number (157 signals) was found in case 8 (spitz nevus) and the highest number of signals (189 signals) was observed in case 3 (melanoma). A mean value of 169.27 signals/case for CEP6 was calculated (median, 169.00 signals/case; IQR, 6.5 signals/case).

The number of nuclei where the number of signals for MYB was less than the number of signals for CEP6 was counted. This was done to highlight the number of nuclei where there was a loss for MYB relative to CEP6. The lowest number of nuclei with a loss for MYB relative to CEP6 was 6: Cases 9 (lentiginous melanocytic proliferation) and 17 (melanoma). The highest number of nuclei with the aforementioned characteristics was 24 (case 7; junctional atypical nevus). The percentage loss of MYB relative to CEP6 was calculated, and the highest percentage was 26.66% for case 7 (Table V). The staining pattern for CEP6 is illustrated in Fig. 3 for cases 11 and 17.

Figure 3 Centromere 6 (aqua) signals in (A) a patient with atypical nevus (case 11) and (B) a patient with melanoma (case 17).

Table V Number and percentage of nuclei that show a loss for MYB relative to CEP6.

Table V

Number and percentage of nuclei that show a loss for MYB relative to CEP6.

Case no.	Number of nuclei that show a loss for MYB relative to CEP6	Percent loss of MYB relative to CEP6
Case 1	7	7.77
Case 2	8	8.88
Case 3	14	15.55
Case 4	16	17.77
Case 5	16	17.77
Case 6	18	20.00
Case 7	24	26.66
Case 8	15	16.66
Case 9	6	6.66
Case 10	9	10.00
Case 11	10	11.11
Case 12	11	12.22
Case 13	10	11.11
Case 14	13	14.44
Case 17	6	6.66
Case 18	13	14.44
Case 19	15	16.66
Case 20	16	17.77

[i] MYB, v-myb myeloblastosis viral oncogene homologue; CEP6, centromere 6.

CCND1 (11q13). For CCND1, the lowest number of counts (49 signals) was found in case 7 (atypical junctional nevus) and the highest number of counts (89 signals) was observed in case 9 (lentiginous melanocytic proliferation) (Table I). The sum of the CCND1 signals was calculated. The lowest sum of signals (156 signals) was found in case 7, while 3 cases had >200 signals: Case 2 (lentigo maligna melanoma; 212 signals), case 3 (melanoma; 228 signals) and case 9 (melanocytic proliferation; 231 signals) (Table II). There was a mean of 179.33 signals/case for CCND1 (median, 170.00 signals/case; IQR, 23.75 signals/case) (Table IV).

The median of CCND1 signals/nuclei was also calculated, and a median of >2 signals/nuclei was obtained for 5 cases: Case 1 (nodular melanoma; 2.18 signals/nuclei), case 2 (lentigo maligna melanoma; 2.35 signals/nuclei), case 3 (melanoma; 2.53 signals/nuclei), case 9 (lentiginous melanocytic proliferation; 2.56 signals/nuclei) and case 10 (atypical compound nevus; 2.06 signals/nuclei) (Table IV). The lowest median value was encountered in case 7 with 1.73 signals/nuclei. The staining pattern for CCND1 is illustrated in Fig. 4 for case 17.

Figure 4 Cyclin D1 (green) signals in a patient with melanoma (case 17).

FISH+/- diagnosis of the melanocytic lesions

Only one FISH+ result out of the 5 histopathologically confirmed melanoma cases was obtained, as well as a FISH+ result for a lentiginous melanocytic proliferation specimen (histopathologically confirmed benign lesion). The FISH+ specimens were cases 3 (melanoma) and 9 (lentiginous melanocytic proliferation), with an average of CCND1 signals of >2.5 signals/nuclei (2.53 signals for case 3 and 2.56 signals for case 9). The remaining of the analyzed melanomas and atypical nevi were diagnosed as FISH-, although certain results approached a FISH+ diagnosis: Case 1 (nodular melanoma), with a median of 2.18 signals/nuclei for CCND1 and 2.11 signals/nuclei for MYB; case 2 (lentigo maligna melanoma), with a median of 2.35 signals/nuclei for CCND1 and 2.16 signals/nuclei; and case 17 (melanoma), with a median of 2.13 signals/nuclei for MYB and 50.00% of abnormal nuclei for RREB1 (Fig. 5). Additionally, two atypical nevi cases also approached a FISH+ result: Case 6, with 20.00% loss of MYB relative to CEP6; and case 7, with 26.66% loss of MYB relative to CEP6. Case 10, with a median of 2.06 signals/nuclei for CCND1 and a median of 2.02 signals/nuclei for MYB, as well as case 20 (lentigo maligna melanoma) did not approach a FISH+ result (Table VI).

Figure 5 Overview of the different FISH staining patterns (RREB1, MYB, CEP6, and CCND1) in the same case (case 17) for comparative purposes.

Table VI Criteria for FISH+/- diagnosis.

Table VI

Criteria for FISH+/- diagnosis.

Case no.	Average CCND1 signals per nuclei ≥2.5	Average MYB signals per nuclei ≥2.5	Percent loss of MYB relative to CEP6 ≥31%	Percentage of abnormal nuclei for RREB1 ≥63%	FISH+/-result
Case 1	2.18	2.11	7.77	18.8	FISH-
Case 2	2.35	2.16	8.88	31.1	FISH-
Case 3	2.53	2.22	15.55	42.2	FISH+
Case 4	1.88	1.87	17.77	27.7	FISH-
Case 5	1.95	1.97	17.77	18.8	FISH-
Case 6	1.85	1.73	20.00	13.3	FISH-
Case 7	1.73	1.81	26.66	14.4	FISH-
Case 8	1.84	1.87	16.66	23.3	FISH-
Case 9	2.56	2.37	6.66	26.7	FISH+
Case 10	2.06	2.02	10.00	13.3	FISH-
Case 11	1.88	1.92	11.11	13.3	FISH-
Case 12	1.91	1.84	12.22	10.0	FISH-
Case 13	1.88	1.87	11.11	11.1	FISH-
Case 14	1.83	1.86	14.44	15.5	FISH-
Case 17	1.92	2.13	6.66	50.0	FISH-
Case 18	1.81	1.83	14.44	13.3	FISH-
Case 19	1.84	1.84	16.66	15.5	FISH-
Case 20	1.83	1.81	17.77	13.3	FISH-

[i] RREB1, Ras responsive element binding protein 1; MYB, v-myb myeloblastosis viral oncogene homologue; CEP6, centromere 6; CCND1, cyclin D1. FISH, fluorescence in situ hybridization; FISH-; FISH negative result; FISH+; FISH positive result.

The statistical analysis revealed a significant result (likelihood ratio; P=0.002) for criterion 10.a of the manufacturer's instructions for the Vysis Melanoma FISH Probe kit (catalogue number, 30-608220/R6) average CCND1 signals per nuclei or average MYB signals per nuclei ≥2.5), indicating that there was a significant association between the diagnosis of melanoma and the value of criterion 10.a. For the other two criteria (10.b and 10.c) the following results were obtained: For criterion 10.b (percent loss of MYB relative to CEP6 ≥31%) P=0.542 and for criterion 10.c (percentage of abnormal nuclei for RREB1 ≥63%) P=0.223, indicating that there was no association between these two criteria and melanoma.

FISH and histopathological associations

Next, histopathological associations with FISH for the cases that had inconclusive FISH results were performed, to examine whether further histopathological clues could be deciphered from the FISH results (Table VII). Of the 5 studied melanoma cases, 4 cases had a FISH- result, although some of the criteria analyzed to establish a FISH diagnosis approached a FISH+ result. These 4 cases are described below.

Table VII Associations of FISH with histopathology.

Table VII

Associations of FISH with histopathology.

Case no.	Histopathological diagnosis	FISH+/-	T-stage	G-grade	Cell infiltrate	Emboli	Mitoses	Pigment
Case 1	Nodular melanoma	FISH-	T2	G1	None	None	None	Melanophages present
Case 2	Lentigo maligna melanoma	FISH-	T1	G2	Score +2 (asymmetrical inflammatory infiltrate)	None	None	Present
Case 6	Junctional atypical nevus	FISH-	T2	G1	None	None	None	Present
Case 7	Junctional atypical nevus	FISH-	T1	G1	None	None	None	None
Case 9	Lentiginous melanocytic proliferation	FISH+	T2	G1	Score 1 (numerous melanophages, inflammatory infiltrate)	None	None	None
Case 10	Mixed atypical nevus	FISH-	T2	G1	Score 1 (melanophages present, inflammatory infiltrate)	None	None	None
Case 17	Melanoma	FISH-	T3	G2	Score 3	None	1 mitosis (rare)	Present
Case 20	Lentigo maligna melanoma	FISH-	T2	G2	Score +1 (low infiltrate)	None	Present	None

[i] FISH, fluorescence in situ hybridization; FISH-, FISH negative result; FISH+, FISH positive result; T1, tumor of 1 mm or less in thickness; T2, tumor >1 mm but not >2 mm in thickness; T3, tumor >2 mm but not >4 mm in thickness; G1, well-differentiated tumor; G2, moderately differentiated tumor.

i) Case 2 (lentigo maligna melanoma): This specimen had a FISH- result, although two criteria were close to a positive result, a median of 2.35 signals/nuclei for CCND1 and a median of 2.16 signals/nuclei for MYB. In the initial histopathological examination, it was a confirmed as lentigo maligna melanoma, Clark I stage, Breslow index of 0.37 mm, G2 differentiation (moderate differentiation) and a score of +2 for the cell infiltrate, with asymmetrical inflammatory infiltrate, without emboli or mitoses, but with the presence of pigment (Fig. 6).

Figure 6 Association of histopathology with fluorescence in situ hybridization in a patient with lentigo maligna melanoma (case 2). (A) Magnification, x20. (B) Magnification, x40. Pigment and cell infiltrate are observed. (C) Centromere 6 (aqua) signals. (D) V-myb myeloblastosis viral oncogene homologue (gold) signals.

ii) Case 9 (lentiginous melanocytic proliferation): This specimen exhibited a FISH+ result, with an average of 2.56 signals/nuclei for CCND1 signals (threshold of >2.5 signals/nuclei). The histopathological examination indicated that it was a benign tumor and was graded as follows: T1 tumoral thickness, G1 differentiation (well-differentiated), a score of 1 for the cell infiltrate, with several melanophages and inflammatory infiltrates, and without emboli, mitoses or pigment (Fig. 7). Initially, this lesion was diagnosed as a malignant tumor through dermoscopy (superficial spreading melanoma localized on the right shoulder of an 84-year-old female patient), but it was later shown to be a benign lesion as confirmed by the histopathological examination.

Figure 7 Histopathological images of a patient with lentiginous melanocytic proliferation (case 9). (A) Magnification, x20. (B) Magnification, x40.

iii) Case 17 (melanoma): This case was a histopathologically confirmed melanoma, but with a FISH- result. Histopathological examination revealed the following grading: T3 tumoral thickness, G2 differentiation (moderate differentiation), a score of 3 for the cell infiltrate, with rare mitoses (one), no emboli and the presence of pigment. Although the FISH test was negative, certain results approached a FISH+ diagnosis; specifically, a median of 2.13 signals/nuclei for MYB and 50.00% of abnormal nuclei for RREB1 (Fig. 8).

Figure 8 Merged image of the fluorescence in situ hybridization signals in a patient with melanoma (case 17).

iv) Case 20 (lentigo maligna melanoma): This lesion did not exhibit any FISH+ results and it obtained a FISH- diagnosis. However, it was a histopathologically confirmed lentigo maligna melanoma with T2 tumoral thickness, G2 differentiation (moderate differentiation), a score of +1 for the cell infiltrate, without emboli, present mitoses and no pigment (Fig. 9).

Figure 9 Centromere 6 (aqua) signals for a fluorescence in situ hybridization negative result in a patient with lentigo maligna melanoma (case 20).

Discussion

Melanoma is one of the most aggressive neoplasms, and patients often present with numerous chromosomal aberrations. Advances in medical technology are important for improving patient outcomes, as earlier diagnosis of melanoma can improve the prognosis of patients and long-term survival (2).

In the presnet study, the value of FISH in melanoma diagnosis was assessed, and used in accordance with the manufacturer's protocol of the kit employed. First, the number of signals for RREB1, MYB, CCND1 and CEP6 in each of 30 nuclei were counted. The highest number of signals were found in the histopathologically confirmed cases of melanoma. Then, the median of the sum of signals for each of the four probes was calculated for each case, with the highest median value of signals observed for RREB1 (183.55 signals/case). Similar to the present study, a previous study found that the aforementioned loci were predominantly observed in melanomas (87% of the cases), and rarely in congenital nevi (5%) or dysplastic nevi (5%) (14).

Next, the number and percentage of abnormal RREB1 nuclei were counted. For the diagnosis of melanoma, according to the manufacturer of the kit used, the percentage of RREB1 abnormal nuclei should be ≥63%. In the present study, the highest percentage of abnormal nuclei encountered in melanoma cases were 50.0% for case 17 (melanoma), 42.2% for case 3 (melanoma) and 31.1% for case 2 (lentigo maligna melanoma), thus not meeting the diagnostic criteria. There were no FISH+ results for RREB1 (6p25), although in the literature it is stated that RREB1 plays an important role in the tumorigenesis of melanoma, potentially serving as a crucial driver-gene in the development of melanoma by inhibiting tumor suppressors, and is one of the most sensitive criteria encountered in the melanocytic lesions analyzed using FISH (14,15). The statistical analysis did not show an association between the percentage of abnormal RREB1 nuclei and melanoma.

Concerning the sum and median of MYB and CCND1 signals, the highest sum of signals was found in melanomas, except for case 9 which had 214 signals (and a median of 2.37 signals/nuclei) for MYB and 231 signals (and a median of 2.56 signals/nuclei) for CCND1. The lowest number of signals was found in two junctional atypical nevi cases (cases 6 and 7), with 156 signals for both MYB and CCND1. For a positive melanoma diagnosis, there must be a median of ≥2.5 CCND1 or MYB signals per nuclei, a result which was only obtained for CCND1 in cases 3 (melanoma; 2.53 signals/nuclei) and 9 (lentiginous melanocytic lesion; 2.56 signals/nuclei). In the present study, there was a significant association between this criterion and melanoma. In contrast to the results of the present study where only FISH+ cases exhibited alterations in CCND1 (11q23), in a previous study, the majority of cases tested positive due to an anomaly in chromosome 6, while the probe for CCND1 did not play an important role (16).

There were no positive results for the percentage loss of MYB relative to CEP6 (≥31%); however, in 1 case the result was close to being FISH+ (26.66% in case 7; junctional atypical nevus). Newmann et al (16) also observed a small number of cases with positive FISH tests for MYB/CEP6, with only 4 out of 36 cases that met the criteria for the loss of MYB relative to CEP6. Furthermore, according to Horst et al (17), there were no cases with histopathological features typical of Spitz nevi with chromosomal abnormalities detected by FISH, similar to the present study.

Concerning the FISH+/- results, after analysis of the melanocytic lesions, one benign lesion out of all the studied nevi had a positive FISH diagnosis. A previous study found that chromosomal abnormalities were noted in 2 out of 32 cases of benign nevi, while 29 of 31 melanoma cases exhibited changes (2). In the present study, only 1 of the 5 melanoma cases had a positive FISH test, but all cases exhibited notable chromosomal changes with results close to a FISH+ diagnosis, except for 1 case (case 20; lentigo maligna melanoma). The melanoma case with a positive FISH test was due to an average of >2.5 signals/nuclei for CCND1, indicating that there were certain chromosomal anomalies found in the 11q23 region. For case 3, the result was also close to being FISH+ (2.22 signals/nuclei for MYB and 42.2% of abnormal nuclei for RREB1), indicating that there may have been anomalies in chromosome 6 (6p25 for RREB1 and 6q23 for MYB). Gerami et al (18) suggested that clonal abnormalities in chromosome 6 (short arm relative to the long arm) were common in all melanoma subtypes. For other histopathologically confirmed cases of melanoma, FISH- results were obtained, which were thus considered false-negative results. According to previous literature, there is a high number of FISH false-negative results; for example, Nijhawan et al (19), citing Kerl et al (20), found a false-negative rate of 30.7% using the criteria followed in the present study. In the present study, the false-negative rate was 80.0% (4 out of 5 melanomas had a FISH- result), whereas the false-positive percentage for FISH+ tests was 7.7% (1 out of 13 benign cases). Gerami et al (21) found that FISH correctly classified melanoma with 86.7% sensitivity and 95.4% specificity, and correctly identified all melanoma lesions (6 out of 6 cases). However, in a study by Fang et al (22), a rare nevus was incorrectly diagnosed as FISH+ (false-positive result), while 9 cases of melanoma incorrectly exhibited FISH- results, out of which 6 metastasized. Other studies have observed only modest sensitivity for FISH as a diagnostic tool. For example, 60% sensitivity was observed by Gaisier et al (23) and 50% by Tetzlaff et al (24).

Considering the high number of false-negative results in the present study, the FISH technique may be considered to be an unreliable tool. However, FISH is a valuable technique, known for several clinical applications, such as the screening or confirmation of melanoma diagnosis, where FISH provides information concerning the duplication/amplification, deletion or translocation of chromosomes in the studied tumors and can be used for diagnosis, prognosis and for predicting the response to therapy. Additionally, in relation to establishing a melanoma diagnosis or distinguishing between common/atypical nevi and melanoma, FISH can also be used for diagnosing rare types of melanoma (conjunctival melanoma in children) (25) and differentiating mitotically active nevi from nevoid melanoma (19). In nevi-associated melanoma, FISH can help to estimate tumoral thickness and delimitate the transition from nevi to melanoma (13). FISH can also distinguish cellular blue nevi from blue nevus-like melanomas and may confirm a diagnosis of metastatic uveal melanoma (19). Furthermore, FISH may also differentiate nodal nevi from metastatic melanoma (19), amongst other types of melanoma (26). Moreover, FISH can be used in other types of cancer, such as breast (HER2+) cancer, lung cancer, bladder cancer, acute myeloid leukemia or acute lymphocytic leukemia (19,26).

Out of all the benign lesions in the present study, one of them (5.6% of all lesions; case 9; lentiginous melanocytic proliferation) had a positive FISH result concerning the CCND1 (green) signals with >2.5 signals/nuclei (2.56 signals/nuclei). This suggests that in this lesion, there may have been a chromosomal abnormality in the 11q23 region. According to a previous study, an increase in copy number of 11q13 is most frequently seen in chronically sun-damaged melanomas, which could have been the case for the patient in the present study as well (case 9), since the lesion was in a sun-exposed area (18). This lesion was initially diagnosed through dermoscopy as a superficial spreading melanoma, located on the right shoulder area of an 84-year-old female patient. This diagnosis was later dismissed by the histopathological examination that confirmed it to be a benign lesion [lentiginous melanocytic proliferation, without emboli, mitoses or pigment, well-differentiated tumor (G1) with a small quantity of cell infiltrate]. According to the positive FISH test, this could have been diagnosed as a malignant tumor, as it had at least one criterion necessary to establish a melanoma diagnosis. Similar results were also encountered in a study by Fang et al (22), where there was 1 patient who had a benign lesion with a FISH+ result. Lai et al (7) also had similar cases with positive FISH results in 3 atypical nevi (6.0% of the studied lesions). Another study has shown that there is often a recurring theme in reporting an initial diagnosis of benign nevi that later prove to be malignant melanomas (9), which could also be the case in the present study. Zulauf et al (27) reported that they had an initial case of a benign nevus, which later was shown to be a nevoid melanoma, similar to the FISH+ case in the present study. It has been reported that if there is a FISH+ result in a melanocytic lesion, it should be monitored closely, as it is widely agreed that FISH+ lesions have a notably higher risk of progression to melanoma (14).

The most sensitive criterion in all the analyzed cases was the one regarding MYB (6q23), with 6 out of 18 cases (33.3%) having results either positive (2 cases) or close to being positive (4 cases). This was followed by the CCND1 (11q13) criterion with changes in 5 out of 18 cases (27.8%), with 2 positive cases. The least sensitive criterion was the loss of MYB relative to CEP6 which was not encountered in any cases in the present study. In a study by Abasolo et al (28), the most sensitive criterion was RREB1, observed in all cases (100%), followed by the CEP6-related MYB loss (48.1%), CCND1 gain (37%) and MYB gain (22.2%). A similar results was obtained by Gerami et al (18), who found that the gain of 6p25 had the highest sensitivity overall. In the present study, changes regarding the RREB1 criterion were found in only 2 out of 18 cases (11.1%), which were close to positive results. Another study by Hossain et al (29) showed that the gain in chromosome 11 was the most commonly encountered.

Certain FISH-histopathological associations were observed for the cases that had inconclusive results in the FISH analysis. It has been reported that FISH can distinguish mitotically active nevi from nevoid melanoma (19). In the present study, none of the nevi had mitoses, which may have helped in establishing the FISH diagnosis. However, there were 2 melanoma cases with mitoses that were diagnosed as FISH-. Tumoral thickness, emboli, the grading of the tumors, the cell infiltrate and the presence of pigment were also analyzed. None of the tumors had emboli present, although out of the 5 melanoma cases, 1 had a positive FISH result for CCND1. The presence of emboli in melanocytic lesions may suggest lymphovascular invasion, which is hypothesized as the mechanism by which melanoma cells can disseminate to regional lymph nodes or other organs (30). In general, the tumoral thickness is mostly used for malignant lesions (melanomas), it was also used for benign lesions (atypical nevi) in the present study to assess the depth of the tumors, and the following results were obtained: 2 cases with a T1 grading (one of them being a lentigo maligna melanoma), 5 cases with a T2 grading (with only 2 of them being melanomas) and 1 case with a T3 grading (this tumor also had mitoses and a higher score of cell infiltrate). The tumoral thickness was analyzed for these lesions, as it is strongly associated with survival (Breslow index) (31). Regarding the grade of the inconclusive studied specimens (with either false-negative or false-positive FISH results), 5 cases were classed as G1 (well-differentiated; with 1 case being a nodular melanoma) and 3 cases as G2 (moderatety differentiated). The G2 grade tumors were all melanomas, had a higher score of cell infiltrate, and exhibited the presence of pigment. Grade differentiation is scored from a low grade of 1, indicating a high level of cellular differentiation (well-differentiated) to a high grade of 3 (poorly differentiated or undifferentiated) (32). Vergier et al (8) reported that by combining the histopathological diagnosis with the FISH results, the diagnosis can be optimized, with an increase in specificity and sensitivity. The findings of the present study support this hypothesis, as the histopathological-FISH associations assisted in the understanding of the final diagnosis of the lesions that had an inconclusive FISH result. Moreover, Vergier et al (8) indicated that in relation to the outcome, the sensitivity and specificity of FISH were 43.0 and 80.0%, while compared with the histopathological review, the sensitivity and specificity of FISH were 34.5 and 91.0%. This shows a low sensitivity (high false-negative rate), similar to the high rate of false-negative results in the present study.

As for the evolution of the cases, all the patients had a good evolution and response to the specific treatment followed, except for 1 patient (case 9) who died shortly after the histopathological diagnosis, although this may have been due to old age (84 years old). A previous study indicated that survival analysis using the Kaplan-Meier method showed a trend for worse prognosis in FISH+ patients, which could also be the case for the aforementioned patient (19).

The present study has certain limitations. There were only 18 (out of 20) cases studied due to some artifacts during the pre/post-treatment phase of the study; however, the present study is intended to be a preliminary step in exploring the potential benefits of FISH in melanoma diagnosis or in distinguishing between atypical nevi and melanoma. Another reason why the number of cases was limited was that the Vysis Melanoma FISH Probe Kit could only include 20 cases, and due to the high cost of the materials needed to perform this test and the length of time required to process the results (the FISH signals had to be manually counted), the study was limited to using a single kit. Therefore, based on the findings of the present study, FISH test results should only be considered for experimental and scientific reasons and not for therapeutic or management decisions. However, although this was a small study, it highlighted the worth of FISH for further larger studies in the future. The present study demonstrated that FISH is a valuable tool; however, given the high number of false-negative results, it should only be used along with other more robust diagnostic techniques (such as histopathological examination) or for scientific and screening purposes. Aims of further studies should include the expansion of the number of cases analyzed and the conduct of larger cohort studies. Future directions include: i) Expanding the gene panel to investigate additional chromosomal regions or genes beyond those already assessed [for example, 7q34, CEP3, 3p26, 8q24 (MYC), 9q21 (CDKN2A) and CEP9]; ii) evaluating how FISH-detected aberrations correlate with the patient prognosis, metastatic risk or therapeutic response; and iii) assessing the performance of FISH compared with other molecular techniques, such as chromogenic in situ hybridization and comparative genomic hybridization, in melanoma. A larger study could improve the sensitivity and specificity of the test, and explore its full potential role in clinical decision-making.

The findings of the present study are supported by Gaiser et al (23), where FISH did not achieve clinically useful sensitivity and specificity results. Another study also reported FISH- results in melanomas, similar to the present study, and indicated that survival probability could not be concluded based on the test used (33). The accuracy of FISH may be increased when used in combination with multiple diagnostic techniques (dermoscopy, histopathology and molecular tests) and may help in improving early diagnosis (29).

The present study highlighted both the advantages and challenges of using FISH in melanoma diagnosis. While a high rate of false-negative results was observed, nearly all malignant tumors exhibited findings close to a FISH+ diagnosis, except for 1 case. Similarly, all but one benign lesion were correctly identified as FISH-, reinforcing the value of this technique. However, challenges such as manual signal counting, the high cost of reagents and potential errors due to equipment limitations should be considered. We hypothesize that similar studies are essential in presenting both the strengths and limitations of FISH as a diagnostic tool.

In conclusion, FISH has numerous clinical applications and provides information concerning the duplication/amplification, deletion or translocation of chromosomes, making it a useful tool in cancer diagnosis. FISH was found to be a valuable tool aiding clinicians in establishing a melanoma diagnosis, although it is recommended to be only used as a complementary diagnostic technique to supplement standard methods of examination. The present study focused on the chromosomal anomalies encountered in melanocytic lesions and on the importance of FISH in melanoma diagnosis. With the recent improvements in probes, including enhanced sensitivity and chromosomal target specificity, and clearer signal detection, FISH may play an increasingly central role in prognosis and early cancer diagnosis; however, future studies are required to understand the utility of FISH in the management of melanocytic lesions.

Acknowledgements

Not applicable.

Funding

Funding: The authors acknowledge Victor Babeș University of Medicine and Pharmacy Timișoara (Timișoara, Romania) for their support in covering the costs of publication for this research paper.

Availability of data and materials

The data generated in the present study may be requested from the corresponding author.

Authors' contributions

CR(J)M was involved in conceptualization, data curation, formal analysis, investigation, methodology, data validation, data visualization, writing the original draft, project administration, review and editing. ARC participated in conceptualization, supervision, data validation, data visualization, review and editing. NPG was involved in conceptualization, data visualization, supervision, review and editing. MRo participated in conceptualization, data curation, providing resources, supervision, data validation, data visualization, review and editing. MRa was involved in conceptualization, formal analysis, investigation, methodology, data validation, supervision, data visualization, review and editing. CR(J)M and MRa confirm the authenticity of all the raw data. All authors read and approved the final manuscript.

Ethics approval and consent to participate

The present study was conducted in accordance with The Declaration of Helsinki and was approved by the Ethics Committees of the Victor Babeș University of Medicine and Pharmacy (approval no. 28/23.09.2019 from 10.03.2022; Timișoara, Romania) and the Emergency Hospital of Sibiu County (approval no. 6001/09.03.2022; Sibiu, Romania). Written informed consent was obtained from the patients for the use of human tissue in research.

Patient consent for publication

Written informed consent for publication of results obtained using human tissues was obtained from the patients involved in the present study.

Competing interests

The authors that they have no competing interests.

References