Primary cutaneous adenoid cystic carcinoma: a clinicopathologic, immunohistochemical, and fluorescence in-situ hybridisation study of 13 cases.

AIMS: This study aimed to investigate the clinical, histological, immunohistochemical and chromosomal features of primary cutaneous adenoid cystic carcinoma (PCACC). METHODS AND RESULTS: We retrospectively analysed 13 cases identified on their clinicopathological features and performed fluorescence in-situ hybridisation (FISH) on six available cases. Head and neck (46.2%) were most commonly involved. The median age was 53 years, with a male predilection. Histologically, tumours were classified as grades 1 (eight), 2 (four) and 3 with high-grade transformation (HGT) (one). The HGT component was demonstrated as poorly differentiated carcinoma with multifocal necrosis and myoepithelial differentiation. Patients with one of the following factors: longest diameter of the lesion (≥ 1 cm), involvement of subcutaneous fat tissue and widely infiltrative border had a relatively higher rate of local recurrence, distant metastasis and death. Five of six cases were confirmed to have MYB translocation, while nuclear staining for MYB proto-oncogene, transcription factor (MYB) protein was found in four cases. During the follow-up (median = 64 months), two patients experienced local recurrences. One patient, who was classified as grade III PCACC with HGT, developed multiple metastases and died of disease. Another patient was alive with multiple metastases. CONCLUSIONS: This is the largest single-institution study, to our knowledge, of PCACC in an Asian population. We describe the first case of scalp PCACC with HGT, which is the only death case in our series. PCACC tends to recur locally and has metastatic potential. PCACC with HGT has a poor prognosis.

Clinicopathological diversity and outcome of longitudinal melanonychia in children and adolescents: analysis of 35 cases identified by excision specimens.

AIMS: Longitudinal melanonychia in paediatric patients often represents a difficult diagnostic challenge, and studies emphasising its clinical and histopathological features are limited due to its low incidence in childhood. METHODS AND RESULTS: We retrospectively analysed 35 paediatric cases identified by excision specimens on their clinicopathological features, and performed fluorescence in-situ hybridisation on 13 available cases. Fingernails (77.1%) were more likely to be affected. Total melanonychia and Hutchinson's sign were observed in 10 (28.6%) and 14 (40.0%) cases, respectively. Nail dystrophy at diagnosis was present in five cases. After complete excision of the lesions, four patients relapsed during follow-up (mean = 38 months). Seventeen cases were diagnosed as lentigines and 18 as naevi, among which 11 cases were categorised as lentigines/naevi with atypical melanocytic hyperplasia. Mild-to-moderate nuclear atypia, confluency of melanocytes, focal pagetoid spread and peri-ungual skin involvement were found in 25.7% (9 of 35), 40.0% (14 of 35), 40.0% (14 of 35) and 40.0% (14 of 35) of cases, respectively. Thirteen cases tested by fluorescence in-situ hybridisation showed no copy number aberration at the probed loci. There was a statistically significant difference in the following features between patients aged less and more than 10 years (P < 0.05): cytomorphology, mild-to-moderate nuclear atypia, confluency of melanocytes, focal pagetoid spread and melanocyte count. CONCLUSIONS: Some concerning clinicopathological characteristics, which are signs indicative of melanoma in adults, are not uncommon in paediatric longitudinal melanonychia, especially in patients aged ≤ 10 years. Owing to the extremely low incidence of melanoma in paediatric longitudinal melanonychia, in most circumstances a more conservative clinical management strategy should be adopted.

Nevus cell aggregates massively occupying parenchyma of an external iliac lymph node: A case report and review of the literature.

We report a case of nevus cell aggregates (NCAs) in an external iliac lymph node from a patient with a compound congenital nevus in the corresponding drainage skin. Melanocytes in parenchyma were in band, nest-like or nodular fashion, and partly continuous with those in capsule and trabeculae. The largest nodule in parenchyma measured 6.5 mm. Melanocytes mostly exhibited benign appearance identical to cutaneous nevus. A few regions abundant in cells displayed atypical features, including increased nucleo-cytoplasmic ratio, small nucleoli, and occasional mitotic figures. Immunohistochemistry showed that melanocytes stained positive for p16, but negative for HMB-45 and nestin. Ki-67 labeling was less than 1% and reticulin mainly surrounded individual melanocytes. Besides, Vysis melanoma fluorescence in situ hybridization (FISH) plus another 2 probes targeting 9p21(CDKN2A) and 8q24(MYC) showed normal results. The patient is alive without malignant tumor after 52-month follow up. Our case provides a new evidence for the existence of intraparenchymal NCAs in deep lymph node and indicates that melanocytes with some atypical features can occur in nodal nevi. Nevus cells in parenchyma connected to those in capsule and trabeculae are a significant clue to distinguish nodal nevi from metastatic melanomas. Additionally, immunohistochemistry and FISH assay are useful in differential diagnosis.

PD-L1 expression in tumour-infiltrating lymphocytes is a poor prognostic factor for primary acral melanoma patients.

AIMS: Programmed cell death protein 1-programmed death-ligand 1 (PD-L1) blockade immunotherapy has shown notable therapeutic benefit in metastatic melanoma, but the clinical relevance of PD-L1 expression remains unclear in melanoma, especially in acral melanoma, which is the most common subtype in Asians. The aim of this study was to evaluate the clinical effect of PD-L1 expression in primary acral melanoma. METHODS AND RESULTS: We used immunohistochemistry to evaluate PD-L1 expression in tumour cells and tumour-infiltrating lymphocytes (TILs), and analysed its associations with clinicopathological features and survival in 78 primary acral melanoma patients. We found that expression of PD-L1 in tumour cells and TILs occurred exclusively in a tumour-stroma interface pattern, consistent with the predominant pattern of TIL distribution. The presence of peritumoral TILs was also associated with high PD-L1 expression in tumour cells. Furthermore, PD-L1 expression in tumour cells and that in TILs showed a close relationship (Spearman's rho = 0.381, P = 0.001). However, neither PD-L1 expression in tumour cells nor that in TILs was significantly correlated with clinicopathological features. In univariate analysis, cases with PD-L1-positive TILs had significantly poorer survival than those with PD-L1-negative TILs (median disease-specific survival of 40.7 months versus 78.0 months; P = 0.008). In multivariate analysis, PD-L1 expression in TILs was an independent factor for poor prognosis (P = 0.032), whereas PD-L1 expression in tumour cells was not significantly correlated with survival in univariate analysis (P = 0.378) and multivariate analysis (P = 0.354). CONCLUSION: This is the first study to demonstrate that PD-L1 expression in TILs, but not that in tumour cells, is an independent predictor of poor prognosis in acral melanoma.

Primary leptomeningeal melanocytic neoplasms: A clinicopathologic, immunohistochemical, and molecular study of 12 cases.

This study investigated the clinicopathological, immunohistochemical, and molecular features of primary leptomeningeal melanocytic neoplasms (LMNs). Twelve LMN cases were retrospectively reviewed. We performed Fluorescence in-situ hybridization (including a 4-probe FISH assay with CDKN2A and MYC assay) and Next-Generation sequencing analyses on available cases. Histologically, 2 tumours were classified as melanocytomas (MC), 2 as intermediate-grade melanocytomas (IMC), and 8 as leptomeningeal melanomas (LMM). Two rare cases of LMM were associated with large plaque-like blue nevus. One MC case was associated with Ota. Ten cases (83.3%) showed melanocytic cells with benign features diffusely proliferating within the meninges. The Ki-67 in three categories differed (MC 0-1%, IMC 0-3%, LMM 3-10%). 57.1% of LMM cases (4/7) were positive for FISH. Nine of 10 tumours harboured activating hotspot mutations in GNAQ, GNA11, or PLCB4. Additional mutations of EIF1AX, SF3B1, or BAP1 were found in 40%, 30%, and 10% of tumours, respectively. During the follow-up (median = 43 months), 5 LMM patients experienced recurrence and/or metastasis, 3 of them died of the disease and the other 2 are alive with the tumour. Our study is by far the first cohort of LMN cases tested by FISH. In addition to morphological indicators including necrosis and mitotic figures, using a combination of Ki-67 and FISH helps to differentiate between IMC and LMM, especially in LMM cases with less pleomorphic features. SF3B1 mutation is first described in 2 cases of plaque-type blue nevus associated with LMM. Patients with SF3B1 mutation might be related to poor prognosis in LMN.

Preferentially expressed antigen in melanoma immunohistochemistry as an adjunct for differential diagnosis in acral lentiginous melanoma and acral nevi.

Preferentially expressed antigen in melanoma (PRAME) has shown promising utility in distinguishing benign melanocytic lesions from melanomas, but knowledge of its expression pattern in acral lentiginous melanoma (ALM) and acral nevi (ANs) is limited. Immunohistochemical expression of PRAME was examined in 75 ALMs and 34 ANs. The clinical and histopathologic characteristics of patients with ALM were collected. PRAME was immunoreactive in 89.3% (67/75) of ALMs, but entirely negative in 94.1% (32/34) of ANs. When staining at least 50% of lesional melanocytes was determined as positivity, the sensitivity and specificity of PRAME for distinguishing ALM from ANs were 69.3% and 100%, respectively. Seventy-one cases of ALMs had tumor cells in the epidermis; 71.8% (51/71) of them showed positive for PRAME. By contrast, 61 ALMs had tumor cells in the dermis; 65.6% (40/61) exhibited positive expression. Twenty-nine of 39 (74.4%) epithelioid cell ALMs were observed to be positive for PRAME. By comparison, 63.8% (23/36) of ALMs with spindle tumor cells were positive for PRAME. However, PRAME positive expression was not associated with any clinical and histopathologic characteristics of patients with ALM, including Breslow thickness, ulcer, cytomorphology, lymph node metastasis, or tumor-infiltrated lymphocytes (TILs). Nevertheless, we observed that 82.6% (19/23) of ALMs with lymph node involvement at diagnosis expressed PRAME, compared with 57.6% (20/35) of those without. In summary, PRAME immunohistochemistry can serve as a helpful adjunct in the differential diagnosis of ALMs and ANs with good sensitivity and high specificity. Additionally, PRAME tends to have a higher positive rate in epidermal melanocytes than in the dermis and is inclined to express in epithelioid cells than in spindle cells of ALMs.