Search

The present study was performed to determine the frequency of p53 protein immunoreactivity in classical Kaposi's sarcoma (KS) as a whole and in relation to the histological subtypes which are considered to correspond to the developmental stages of the tumor. The accumulation of p53 protein was studied immunohistochemically using monoclonal antibody BP53-12 on formalin-fixed paraffin-embedded sections of 36 KS lesions, of which 14 were classified histologically as early type and 22 as spindle-cell or mixed type. No positive immunoreactivity was detected in any of the 14 early-type lesions. Among the 22 spindle-cell and mixed variants, positive staining was detected in 5-10% of the tumor cells in one lesion, 1-5% of the cells in six lesions, and in < 1% of the cells in two lesions. These very small percentages of positively stained cells in less than half of the cases of the spindle-cell and mixed variants do not support a significant role for p53 in tumor progression and evolution in KS.

We studied the staining pattern of HMB-45 antibody in 29 compound Spitz nevi (SNs) of the epithelioid cell variety, 17 of which showed extension of nevus cells into the reticular dermis (i.e., "deep"); 20 ordinary compound nevi (CNs), all with a deep dermal component; and 22 primary cutaneous invasive malignant melanomas (MMs) (excluding the desmoplastic and spindle cell types), 12 of which extended into Clark level IV or V. Of the 29 SNs, eight (28%) stained negatively; five (17%), including two deep SNs, stained in the epidermal component only; and 16 (55%), including 10 deep SNs, stained in both the epidermal and dermal components. Of the latter 10 deep SNs, eight stained in the upper dermis only, and in the remaining two lesions, a smaller number of positively stained nevus cells were detectable in the lower dermis as well; these two SNs were not atypical histologically. Of the 20 CNs, four (20%) stained negatively, two (10%) stained in the epidermal component only, and 14 (70%) stained in the epidermal component and the upper dermis only. Of the 22 MMs, one stained negatively, and 21 (95%) stained positively in both the epidermal and dermal components. The pattern was variable in frequency of both staining and distribution, but showed no stratification. We conclude that the majority of our positively stained deep compound SNs showed a stratified pattern of HMB-45 staining, similar to ordinary CNs and different from MMs, and that this pattern might be used as an adjunct in the histopathologic differential diagnosis of compound SN and MM, in the proper clinicopathological context.

The accumulation of p53 protein was studied immunohistochemically on paraffin-embedded sections of 26 Spitz nevi (SNs), 26 primary invasive cutaneous malignant melanomas (MMs), 20 metastases of MM, and 17 ordinary compound nevi (CNs), using monoclonal antibody BP53-12. Positive reactivity was detected in some of the tumor cells in seven (35%) metastatic MMs, all exhibiting strong nuclear staining; eight (31%) primary MMs, of which seven showed strong nuclear staining; two (7%) SNs, of which only one showed strong nuclear staining; and none of the CNs. The frequencies of the positively stained lesions in general, and the strongly positively stained lesions in particular, in the MM and metastatic MM groups were each statistically significantly higher than the respective frequencies in the SN and CN groups. We believe that the immunohistochemical detection of p53 protein with the use of monoclonal antibodies such as BP53-12 on paraffin sections, especially when strong nuclear reactivity is demonstrated, may prove to be an adjunctive tool in the histopathologic differentiation of MM from SN.

Several of the methods of treating plaque-stage mycosis fungoides (MF) advocate treatment of the whole skin, although relatively little information exists on the histologic appearance of clinically uninvolved MF skin. Our study was performed to elucidate the histologic appearance of the clinically uninvolved skin of plaque-stage MF patients. Biopsies were taken from the clinically uninvolved skin of 18 untreated patients with plaque-stage MF, 1 cm (near) and > or = 10 cm (far) from the MF plaques, and from 23 normal, healthy controls. Eight to 12 serial sections from each specimen were examined. The most common histologic finding, which was observed in six (33%) biopsies of near skin and four (22%) biopsies of far skin compared with only one (4%) biopsy of normal control (p = 0.02 and 0.1, respectively), consisted of mononuclear-cell infiltrate around a blood vessel in the papillary dermis, with extension of some of these cells into the overlying epidermis. More diffuse and epidermotropic mononuclear cell infiltrates were seen in an additional three (17%) biopsies of near skin, whereas mild nonepidermotropic superficial perivascular or perifollicular mononuclear cell infiltrates were observed in an additional four (17%) biopsies of the normal controls. In conclusion, the most common histologic finding in our study, which was observed predominantly in the MF group, might represent an earlier stage and clinically undetectable involvement of the normal-looking skin in MF patients.

In order to determine epithelial markers in malignant melanoma in routinely processed paraffin sections and to compare the staining of primary (cutaneous) malignant melanomas and their metastases, we stained formalin-fixed paraffin sections of 13 primary and 18 metastatic malignant melanomas using the streptavidin-biotin peroxidase method by antibodies to S-100, vimentin, HMB-45, polyclonal carcinoembryonic antigen (CEA), monoclonal CEA, cytokeratins (CAM 5.2 and broad-spectrum CKKES), and epithelial membrane antigen (EMA). All primary and most metastatic malignant melanomas showed positive staining with anti-S-100, HMB-45, and anti-vimentin. Reactivity with polyclonal CEA was observed in 15 (48%) of the 31 lesions; 14 of them were metastatic. No lesion was reactive with monoclonal CEA. Significant cytokeratin (CK) staining was evident in only three (9.7%) lesions (all metastatic), which also stained specifically with anti-CK 18. EMA was observed only focally in two (6.5%) lesions. There was no correlation between epithelial markers staining of the primary tumours and their metastases. All lesions with CK or EMA staining showed concomitant extensive staining for S-100, HMB-45, and vimentin. We conclude that (a) polyclonal CEA staining in malignant melanoma is not rare and is probably due to CEA-related molecules; (b) significant CK reactivity is rare and related to simple CK, such as CK 18; (c) epithelial marker reactivity is more common in metastases of malignant melanomas and is not correlated to the reactivity in their primary tumors. Considering our results and reports of positive S-100, vimentin, and HMB-45 in epithelial tumors, a wide panel of antibodies is recommended for the study of undifferentiated tumors.

The histological findings in the first diagnostic biopsy of 50 patients with Kaposi's sarcoma (KS) were classified and were related to the clinical course. Forty-one patients had the "classic KS," of whom 31 had a benign course and 10 an aggressive disease. Nine patients had iatrogenically induced KS. Histologically the following subtypes were found: mixed type in 18 patients, spindle type in 16, early type in 14, and hemangiomatous and lymphangiomatous types in one each. We did not find any statistical relationship between the histological subtypes, degree of nuclear atypia, number of mitoses, and density of the inflammatory cell infiltrates in the KS lesions, and the clinical types and course of the disease. The results of this study do not support the possibility of a prognostic importance regarding the histological features of the early lesions in patients with classic or iatrogenic Kaposi's sarcoma.

We studied 10 biopsies of early (< 1 week) localized granuloma annulare (LGA) lesions and 10 specimens of the normal-looking skin adjacent to actively spreading LGA lesions for signs of vasculitis, using histological and direct immunofluorescent (DIF) techniques. In the early LGA lesions, the collagen fibers showed various forms of alterations, with hyalinization and fragmentation being the most common; in half of these lesions, neutrophils and nuclear fragments in various numbers and densities were found among the altered collagen fibers. Some of the small blood vessels in the areas of granulomatous inflammation showed endothelial cell hypertrophy in four cases, and in one case also endothelial cell proliferation and luminal occlusion. In only one case, a single small blood vessel, which was situated in the center of a palisading granuloma, showed fibrinoid necrosis of its walls. The DIF study of all 20 specimens and the histological study of the 10 specimens of normal-looking adjacent skin did not reveal immune deposits in the vessels' walls, or histological evidence of vasculitis, respectively. We believe that these findings do not support a role of vasculitis in the formation of LGA lesions, but suggest that neutrophils might play a primary or secondary role in their pathogenesis.

Porokeratotic eccrine ostial and dermal duct nevus (PEODDN) has been said to represent a widely dilated, keratin-plugged acrosyringium and dermal duct. We have observed in a case of congenital PEODDN a normal-appearing, acrosyringium-like duct that traverses vertically the entire length of the parakeratotic column. Also, in its lower course, it stained positively for carcinoembryonic antigen, while the inner borders of the invagination from which the parakeratotic column arose stained negatively. This leads us to suggest that the epithelial structure in PEODDN is an abnormally keratinizing epidermal invagination through which an acrosyringium-like duct traverses, rather than an abnormally dilated, parakeratotically plugged acrosyringium and dermal duct.

We studied keratin expression in the involved and uninvolved skin of six benign familial chronic pemphigus (BFCP) patients, using monoclonal antibodies specific for various keratin polypeptides and immunohistopathologic techniques. Normal and psoriatic (i.e., hyperproliferative) skin specimens served as controls. The uninvolved BFCP epidermis showed keratin profiles identical to that of normal epidermis. In the acantholytic epidermal segments of the involved BFCP skin, some of the lower suprabasal acantholytic cells failed to express keratin polypeptides 10 and 11 (Moll's catalog). This delay in expression of suprabasal keratins was not accompanied by an expression of hyperproliferative keratin polypeptide 16. Also, some of the lower suprabasal acantholytic cells of the involved BFCP epidermis retained staining by the antikeratin KS-1A3 antibody, which in the normal, psoriatic, and uninvolved epidermis was limited to the basal cell layer. Staining for keratin polypeptide 18 was negative in the epidermis of all four types of specimens. We believe that the delay in suprabasal keratin expression in the involved BFCP epidermis was more likely secondary to the acantholysis (i.e., "arrest of differentiation" due to acantholysis) rather than due to a primary defect in keratin expression.

We performed an immunohistochemical study that compared a primary adenoid cystic carcinoma (ACC) of the skin with two salivary gland ACC. All three tumors stained positively and in identical fashion for epithelial membrane antigen (EMA), carcinoembryonic antigen (CEA), broad-spectrum keratins, and low-molecular-weight keratins. Both EMA and CEA were localized to the luminal surfaces and the secreted contents of the tubular structures and the ductlike structures of the cribriform formations. The staining reactions for both types of keratin were more intense in the cells lining the tubular structures and the ductlike structures of the cribriform formations. One of the two salivary ACCs stained positively for S-100 protein; the other was positive for vimentin. The cutaneous ACC was negative for both antigens. Leu-7 antigen was not detected in either type of ACC. These results show that primary cutaneous ACC and salivary ACC have similar immunohistochemical staining patterns for a number of antigens. We believe this similarity is due to the fact that these antigens are shared by the sweat glands and salivary glands, which are considered to be the respective sites of origin for these two types of tumors.

We have reviewed the morphologic findings in 41 histologic slides of granuloma annulare (GA), 15 from localized (LGA) and 26 from generalized granuloma annulare (GGA). The most common pattern was the histiocytic infiltrative type, more so in cases with LGA than with GGA. The palisading granuloma pattern was present in 21.9% of patients and its prevalence was almost equal in both clinical types. The least prevalent histologic type in all patients taken together was the epithelioid nodule type. However, the prevalence of this type in LGA was equal to that of the palisading type, in contrast to that in GGA where it accounted for only 11.5%. In addition, a mixed histologic pattern was found in greater than 25% of sections from GGA and in only one section from LGA. Periodic acid-Schiff stain revealed hyalinized capillary basement membrane in most of GGA and only in approximately 50% of LGA sections. The variation in prevalence of the different histologic patterns between sections from LGA and GGA, and the difference between our findings and those observed in other series, are discussed.

The histological picture in granuloma annulare (GA) is well defined. The types of infiltrate and the changes in collagen and elastic tissue have been repeatedly described. This is a preliminary study focused on the elastic tissue changes in lesions of generalized GA. Biopsies from four patients were studied. Specimens of involved and uninvolved skin from both exposed and unexposed parts of the body were compared. Changes of actinic elastosis in the papillary dermis in each patient in different stages were observed only in the exposed skin. The typical GA infiltrate was located beneath, in the upper mid-dermis. In the zone of the infiltrate in all specimens--exposed and unexposed--with no regard to the severity of actinic elastotic changes, absence or a marked reduction of elastic material was noted with engulfment of elastic remnants in giant cells. Giant cells with phagocytosis of elastotic material were mostly located on the periphery of the granuloma in the upper as well as in the deeper portion, where elastic fibers were still present. This might support the suggestion that degenerating elastic fibers promote the granulomatous reaction that leads to the formation of granuloma annulare.