Comprehensive histopathological comparison of epidermotropic/dermal metastatic melanoma and primary nodular melanoma.

AIMS: Metastatic melanoma involving the epidermis and/or upper dermis may show significant histological overlap with primary cutaneous melanoma, especially the nodular subtype. Proper histopathological classification is crucial to appropriate staging and management, but is often challenging. The aim of this study was to identify helpful histopathological features for differentiating epidermotropic/dermal metastatic melanoma (EDMM) and primary nodular melanoma (PNM). METHODS AND RESULTS: A cohort of EDMMs (n = 74) and PNMs (n = 75) was retrospectively reviewed for various histopathological features, and the data were compared between groups by the use of univariate analysis. Features significantly associated with EDMM included a tumour size of <2 mm, an absence of tumour-infiltrating lymphocytes and plasma cells, monomorphism, and involvement of adnexal epithelium. Features associated with PNM included a polypoid (exophytic) configuration, prominent tumour-infiltrating plasma cells (TIPs), a tumour size of >10 mm, ulceration, epidermal collarettes, a higher mitotic rate, necrosis, multiple phenotypes, significant pleomorphism, and lichenoid inflammation. In multivariate analysis, a logistic regression model including large tumour size, ulceration, prominent TIPs, lichenoid inflammation and epidermal collarettes was highly predictive of PNM. Six (8%) EDMMs from three patients showed an 'epidermal-only' or 'epidermal-predominant' pattern closely simulating in-situ or microinvasive melanoma. Two of these cases were tested by fluorescence in-situ hybridisation, which confirmed clonal relationships with their corresponding primary melanomas. CONCLUSIONS: This is the first comprehensive histopathological comparison of EDMM and PNM. Recognition of the above histopathological associations should aid in the correct classification and staging of cutaneous melanoma. Epidermotropic metastatic melanomas may occasionally show an epidermal-only/epidermal-predominant pattern; accurate diagnosis requires prudent clinical correlation and, when necessary, ancillary molecular tests.

Histopathological features of molluscum contagiosum other than molluscum bodies.

AIMS: Classic histopathological features of molluscum contagiosum (MC) include a crateriform, acanthotic epidermis containing intracytoplasmic molluscum bodies (MBs). In our experience, a subset of cases lack these features on initial haematoxylin and eosin-stained sections. We aimed to describe the histopathological features of MC other than those classically described. METHODS AND RESULTS: Sixty-seven biopsies diagnosed as MC from January 2011 to October 2012 were retrospectively reviewed. Keratinocytes peripheral to the diagnostic cells with MBs had prominent nucleoli (67; 100%), amphophilic cytoplasm (54; 81%), and in many instances clear cytoplasmic vacuolization (38; 57%). Stroma surrounding MC lesions showed fibroedematous to fibromyxoid changes in many cases (36; 54%), with a subset (13; 19%) showing abundant dermal mucin. In eight of 67 cases (12%), initial sections did not possess MBs or crateriform epidermis of MC. In these cases, initial sections revealed only the epithelial and/or perilesional stromal changes described above. Additional sections contained MBs in all of these cases. CONCLUSIONS: Perilesional fibroedematous to fibromyxoid stroma and keratinocyte changes, including prominent nucleoli and amphophilic cytoplasm with clear vacuolization, are common in MC. Recognizing these features may prove helpful in reaching the diagnosis of MC in cases lacking classic histopathological features on initial sections.

Primary cutaneous cribriform carcinoma: report of six cases with clinicopathologic data and immunohistochemical profile.

BACKGROUND: Primary cutaneous cribriform carcinoma (PCCC) is a rare and under-recognized variant of sweat gland carcinoma, characterized by anastomosing tubules and solid nests producing a sieve-like appearance. METHODS: Six cases of PCCC were clinically, histopathologically and immunophenotypically studied. RESULTS: All cases showed an unencapsulated, nodular dermal tumor composed of solid and cribriform nests, cords and tubules embedded within a desmoplastic stroma. Some tubules showed micropapillary projections reminiscent of tubular adenoma, and intraluminal thin bridges resembling adenomatoid tumor. One case showed a predominantly solid component. Only two cases showed very focal decapitation secretion. Tumor cells expressed CK7, CK5/6, EMA, CEA, S-100, BerEP4 and c-kit and were negative for cytokeratin 20, estrogen receptor/progesterone receptor (ER/PR), androgen receptor and GCDFP-15. Calponin, p63 and smooth muscle actin (SMA) showed absence of a myoepithelial layer. CONCLUSIONS: Contrary to the original reports, our series shows that PCCC may be predominantly solid with only a minor cribriform component, and expresses c-kit and S-100 which may potentially lead to the misdiagnosis of adenoid cystic carcinoma. Immunohistochemical stains may aid in distinction from tubular adenoma. Current data suggest a favorable outcome, as metastatic disease has not been reported to date.

Atypical umbilical naevi: histopathological analysis of 20 cases.

AIMS: Melanocytic naevi on the umbilicus have been described as a form of flexural naevi, with the most common feature being a 'nested and dyshesive pattern'. We have encountered a distinct group of umbilical naevi with more significant atypia and prominent fibrosis, not reported previously. This study aimed to characterize these naevi more clearly. METHODS AND RESULTS: Eighty-one umbilical naevi from 2000 to 2013 were reviewed retrospectively, 20 cases of which showed lamellar fibrosis and atypia and were designated as atypical umbilical naevi (AUN). Lamellar fibrosis in AUN was extensive and frequently entrapped dermal melanocytes, resulting in low-grade cytological atypia (74% of cases) and impaired maturation (47%) of the dermal component. Other common features included bridging (95%), shoulder architecture (94%), lentiginous growth (85%) and high-grade junctional cytological atypia (85%). 'Nested and dyshesive pattern' was observed in only 20% of AUN. Ki-67 immunostaining performed on 12 AUN revealed a consistently low proliferation index of <1%. All AUN in this series lacked junctional confluence, florid pagetoid spread and dermal mitoses. CONCLUSION: A subset of special site naevi in the umbilical region demonstrate characteristic lamellar fibrosis, architectural disorder and cytological atypia. Familiarity with these site-related atypical features would avoid overdiagnosis of melanoma.

Selected inflammatory imitators of mycosis fungoides: histologic features and utility of ancillary studies.

Mycosis fungoides is the most common primary cutaneous lymphoma; however, it remains a significant diagnostic challenge, in part because of the overlap with several inflammatory dermatoses. Despite advances in immunohistochemistry and molecular diagnostics, false-positive, false-negative, and indeterminate diagnoses are not uncommon. In most cases, the overall balance of morphologic, immunophenotypic, and genetic features must be considered carefully because there are few sensitive and specific clues to the diagnosis. Moreover, an appropriate clinical presentation is essential to the diagnosis and helps to favor or exclude inflammatory/reactive processes. Herein, we discuss 3 important inflammatory dermatoses that may closely simulate mycosis fungoides, and we review the use of ancillary studies in these challenging cases.

Expression of the p40 isoform of p63 has high specificity for cutaneous sarcomatoid squamous cell carcinoma.

Cutaneous spindle cell malignancies such as sarcomatoid squamous cell carcinoma (SCC), leiomyosarcoma, desmoplastic melanoma (DM) and atypical fibroxanthoma (AFX) may be morphologically indistinguishable, yet accurate diagnosis is important for appropriate clinical management. The distinction among these entities relies on immunohistochemical evaluation for epidermal, muscle or melanocytic differentiation. Epidermal differentiation markers include cytokeratins and p63. p63 is expressed as two distinct isoforms, ΔNp63 (p40) and TAp63. p40 positivity is highly specific for pulmonary SCC and head and neck sarcomatoid SCC. We examined the utility of p40 vs. p63 immunostaining in the differentiation of a variety of cutaneous spindle cell malignancies, including sarcomatoid SCC (n = 27), AFX (n = 34) and DM (n = 10). p40 was less sensitive than p63 for detecting sarcomatoid SCC (56% and 81%, respectively). p63 and p40 were comparably specific for sarcomatoid SCC relative to AFX, with only rare weak staining of tumor cells for p63 and/or p40 in a minority of AFX cases, including one case with approximately 10% of cells staining weakly for p40. All cases of DM were negative for p40 and p63. Our results support continued use of p63 for diagnosis of cutaneous sarcomatoid SCC because of greater sensitivity relative to p40.

Re-excision rates of invasive ductal carcinoma with lobular features compared with invasive ductal carcinomas and invasive lobular carcinomas of the breast.

BACKGROUND: Invasive ductal carcinoma (IDC) with lobular features (IDC-L) is not recognized as a subtype of breast cancer. We previously showed that IDC-L may be a variant of IDC with clinicopathological characteristics more similar to invasive lobular carcinoma (ILC). We sought to determine the re-excision rates of IDC-L compared with ILC and IDC, and the feasibility of diagnosing IDC-L on core biopsies. METHODS: Surgical procedure, multiple tumor foci, tumor size, and residual invasive carcinoma on re-excision were recorded for IDC-L (n = 178), IDC (n = 636), and ILC (n = 251). Re-excision rates were calculated by excluding mastectomy as first procedure cases and including only re-excisions for invasive carcinoma. Slides of correlating core biopsies for IDC-L cases initially diagnosed as IDC were re-reviewed. RESULTS: For T2 tumors (2.1-5.0 cm), re-excision rates for IDC-L (76 %) and ILC (88 %) were higher than that for IDC (42 %) (p = 0.003). Multiple tumor foci were more common in IDC-L (31 %) and ILC (26 %) than IDC (7 %) (p < 0.0001), which was a significant factor in higher re-excision rates when compared with a single tumor focus (p < 0.001). Ninety-two of 149 patients (62 %) with IDC-L were diagnosed on core biopsies. Of the 44 patients initially diagnosed as IDC, 30 were re-reviewed, of which 24 (80 %) were re-classified as IDC-L. CONCLUSIONS: Similar to ILC, re-excision rates for IDC-L are higher than IDC for larger tumors. Patients may need to be counseled about the higher likelihood of additional procedures to achieve negative margins. This underscores the importance of distinguishing IDC-L from IDC on core biopsies.

Cutaneous hypertrophic lupus erythematosus: a challenging histopathologic diagnosis in the absence of clinical information.

Hypertrophic lupus erythematosus (HLE) is rare variant of chronic cutaneous lupus characterized histologically by irregular epidermal hyperplasia associated with features of classic chronic cutaneous lupus, including interface changes. Lesions frequently demonstrate reactive squamous atypia of the basal layer and may show histopathologic overlap with other more common cutaneous atypical squamoproliferative lesions. Typical histologic features of cutaneous lupus, such as follicular plugging, angiocentric lymphocytic inflammation, and dermal mucin, are very helpful clues to the diagnosis of hypertrophic lupus erythematosus. Recently, immunohistochemistry for CD123 used to detect increased plasmacytoid dendrocytes in hypertrophic lupus erythematosus has proven to be diagnostically useful. A high index of suspicion for hypertrophic lupus erythematosus is essential to avoid overdiagnosis of squamous neoplasia, particularly in limited cutaneous biopsies in the absence of adequate clinical information.

Lupus profundus (panniculitis): a potential mimic of subcutaneous panniculitis-like T-cell lymphoma.

Lupus profundus is a rare manifestation of cutaneous lupus erythematosus, seen in 1% to 3% of patients. It most commonly presents in association with classic chronic cutaneous lesions of discoid lupus erythematosus; however, such lesions, as well as a clinical history of lupus erythematosus, may be lacking. The differential diagnosis for lymphocytic lobular panniculitides is broad; however, the consideration of subcutaneous panniculitis-like T-cell lymphoma is most critical. Recently, there have been several reports emphasizing the overlapping histomorphologic features between lupus profundus and subcutaneous panniculitis-like T-cell lymphoma. Although this is controversial, some authors suggest that patients with lupus profundus are at risk for the development of abnormal, clonal T-cell proliferations and/or overt subcutaneous panniculitis-like T-cell lymphoma. In cases of atypical lymphocytic lobular panniculitis that fail to meet diagnostic criteria for subcutaneous panniculitis-like T-cell lymphoma, patients should be clinically followed indefinitely, as future subcutaneous lymphoma cannot be excluded.

Classic versus type II enteropathy-associated T-cell lymphoma: diagnostic considerations.

Enteropathy-associated T-cell lymphoma is a rare type of peripheral T-cell lymphoma that characteristically involves the jejunum or ileum. Awareness of the histologic and immunophenotypic features of this subtype of lymphoma is important for accurate subclassification. Enteropathy-associated T-cell lymphoma has 2 forms, classic and type II, with different clinical and pathologic features. The 2 types will be described and discussed, with an emphasis on how to differentiate these entities in routine practice.

Rosette-like structures in the spectrum of spitzoid tumors.

BACKGROUND: Spitz nevi demonstrate a diverse spectrum of morphologies. Recently, there have been two reported examples of Spitz nevi with rosette-like structures similar to Homer-Wright rosettes. Rosettes have also been described in melanomas and in a proliferative nodule arising in a congenital nevus. METHODS: A retrospective review of 104 cases of Spitz nevi and variants (n = 51), pigmented spindle cell nevi (n = 26), combined melanocytic nevi with features of Spitz (n = 8), atypical Spitz tumor (AST, n = 9), and spitzoid melanoma (n = 10). RESULTS: Rosette-like structures were present in 3 of the 104 cases (2.9%), including a compound Spitz nevus, a desmoplastic Spitz nevus, and an AST. All three cases demonstrated several foci of small nests of epithelioid cells with peripherally palisaded nuclei arranged around a central area of fibrillar eosinophilic cytoplasm. Immunohistochemical staining of the three spitzoid lesions demonstrated that the rosette-like structures express S100 protein, Melan-A, and neuron specific enolase (NSE) and lacked expression of neurofilament, glial fibrillary acidic protein and synaptophysin. CONCLUSIONS: While uncommon, rosette-like structures can occur as a focal feature in Spitz nevi and AST. Rosette-like structures may represent a normal morphologic finding in Spitz nevi, and awareness of them may prevent misdiagnosis as a neural tumor or melanoma.

Invasive ductal carcinoma with lobular features: a comparison study to invasive ductal and invasive lobular carcinomas of the breast.

Invasive ductal carcinoma with lobular features (IDC-L) is not recognized as a distinct subtype of breast cancer, and its clinicopathologic features and outcomes are unknown. In this retrospective study, we focused on characterization of clinicopathologic features and outcomes of IDC-L and compared them to invasive ductal carcinoma (IDC) and invasive lobular carcinoma (ILC). 183 cases of IDC-L from 1996 to 2011 were compared with 1,499 cases of IDC and 375 cases of ILC. Available slides of IDC-L (n = 150) were reviewed to quantify the lobular component (≤ 20, 21-50, 51-80, >80 %), defined as small cells individually dispersed, arranged in linear cords, or in loose aggregates without the formation of tubules or cohesive nests. E-cadherin immunostain was performed to confirm ductal origin. Compared to IDC, IDC-L was more likely to have lower histologic grade (p < 0.001), be positive for estrogen receptor (96 vs. 70 %; p < 0.0001) and progesterone receptor (84 vs. 57 %; p < 0.0001), and less likely to overexpress HER-2/neu (12 vs. 23 %; p = 0.001). Despite these favorable prognostic features, IDC-L had a higher frequency of nodal metastases (51 vs. 34 %; p < 0.0001) and a worse 5-year disease-free survival than IDC (hazard ratio = 0.454; p = 0.0004). ILC and IDC-L had similar clinicopathologic features and outcomes. The proportion of the lobular component in IDC-L had no impact on the size, nodal status, stage, or outcome. Our data suggest that although IDC-L may be a variant of IDC, with >90 % of cases being E-cadherin positive, the clinical and biological characteristics are more similar to that of ILC.