Expression of activation antigens by T cells infiltrating basal cell carcinomas

The association between non-melanoma skin cancer and a young dimorphic Alu element within the major histocompatibility complex class I genomic region

A non-melanoma skin cancer (NMSC) susceptibility locus within the major histocompatibility complex (MHC) class I region was previously identified telomeric of the HLA-C gene using high-density microsatellite markers. Here, we have extended the previous microsatellite study by using the same DNA samples obtained from 154 NMSC patients and 213 normal controls from the town of Busselton in Western Australia and examined the relationship between five polymorphic Alu insertions (POALINs) within the MHC class I region and their association with NMSC. The genotype distribution of the AluyTF insertion that is located within the NMSC susceptibility region telomeric of the HLA-C gene was significantly increased according to the Fisher's exact test in the NMSC patients, and it was not in Hardy-Weinberg equilibrium in the control group. There was no difference between the cancer patients and controls for the genotypes of the AluyMICB locus within intron 1 of the MICB gene and the other three POALINs (AluyHJ, AluyHG and AluyHF) that are located within the genomic region of the HLA-A, -G and -F gene cluster. The test for significant linkage disequilibrium for 10 pairs of POALIN loci and estimations of two locus POALIN haplotype frequencies also revealed AluyTF differences between the cases and controls. In conclusion, the MHC class I POALIN, AluyTF, that is located within the NMSC susceptibility locus and near the HLA-C gene was strongly associated with NMSC. This finding, using five different polymorphic Alu insertion markers, supports the previous microsatellite association study that one or more genes located in close proximity to the AluyTF insertion has a potential role in NMSC.

Immunoprevention of basal cell carcinomas with recombinant hedgehog-interacting protein

Basal cell carcinomas (BCCs) are driven by abnormal hedgehog signaling and highly overexpress several hedgehog target genes. We report here our use of one of these target genes, hedgehog-interacting protein (Hip1), as a tumor-associated antigen for immunoprevention of BCCs in Ptch1^+/− mice treated with ionizing radiation. Hip1 mRNA is expressed in adult mouse tissues at levels considerably lower than those in BCCs. Immunization with either of two large recombinant Hip1 polypeptides was well tolerated in Ptch1^+/− mice, induced B and T cell responses detectable by enzyme-linked immunosorbent assay, Western blot, delayed type hypersensitivity, and enzyme-linked immunospot assay, and reduced the number of BCCs by 42% (P < 0.001) and 32% (P < 0.01), respectively. We conclude that immunization with proteins specifically up-regulated by hedgehog signaling may hold promise as a preventive option for patients such as those with the basal cell nevus syndrome who are destined to develop large numbers of BCCs.

Dense inflammation does not mask residual primary basal cell carcinoma during Mohs micrographic surgery

BACKGROUND

Areas of dense inflammation are commonly removed during Mohs micrographic surgery for basal cell carcinoma because of the concern that they may mask areas of tumor.

OBJECTIVE

Our purpose was to determine whether inflammation masks tumor during Mohs surgery for primary basal cell carcinoma.

METHODS

Twenty-five consecutive cases of primary basal cell carcinoma with areas of dense inflammation encountered during Mohs surgery were sectioned and stained with hematoxylin and eosin and Ber-EP4.

RESULTS

In no cases did the dense inflammation mask residual tumor.

CONCLUSION

Dense inflammation does not mask primary basal cell carcinoma during Mohs surgery and should be carefully evaluated before additional surgery is performed.

Cytokine profiles in spontaneously regressing basal cell carcinomas

BACKGROUND

Basal cell carcinomas (BCCs) can cause considerable morbidity due to their ability to enlarge progressively and to destroy underlying tissues. However, some BCCs may undergo spontaneous regression in the absence of therapy capable of inducing antineoplastic effects. Histological criteria for this process have been described, and previous studies have suggested that it may be mediated by infiltrating activated CD4-positive T cells.

OBJECTIVES

The purpose of this study was to compare the expression of cytokines in actively regressing and non-regressing BCCs, to ascertain if active regression is associated with a particular cytokine profile.

METHODS

Reverse transcriptase-polymerase chain reaction, a sensitive, quantitative technique allowing analysis of multiple cytokines from small tumour samples, was used.

RESULTS

Interferon (IFN)-gamma was significantly elevated in actively regressing BCCs compared with non-regressing BCCs. Furthermore, interleukin (IL)-2, tumour necrosis factor (TNF)-beta and CD3 delta tended to be elevated in actively regressing tumours, although not to statistically significant levels. IFN-gamma, IL-2, IL-10, TNF-beta, granulocyte-macrophage colony-stimulating factor and Fas ligand showed strong positive correlations with CD3 delta, indicating an association between infiltrating T cells and these cytokines.

CONCLUSIONS

These findings support a role for T-helper 1 type cytokines in mediating spontaneous regression of BCCs.

Perineural spread of basal cell carcinomas treated with Mohs micrographic surgery

BACKGROUND

Perineural spread is a well-documented feature of cutaneous tumors and may portend a more aggressive course. The incidence of perineural invasion in basal cell carcinoma (BCC) is reportedly 1%. The authors sought to determine whether perineural spread occurs more commonly than previously thought.

METHODS

The authors prospectively evaluated 434 patients with BCC treated with Mohs surgery, assessing the presence or absence of perineural inflammation and invasion in tumors requiring more than one stage of surgery. They also documented the demographic features, clinical characteristics, histologic subtype, and operative data in each case.

RESULTS

Seventy-eight BCCs required more than one stage of Mohs surgery. Perineural inflammation, perineural tumor invasion, or both were present in 29 of the 78 tumors (37%), or 6.7% of all 434 prospectively evaluated cases. Twenty-one of the 78 tumors (26.9%) exhibited perineural inflammation, 3 (3.8%) demonstrated perineural invasion, and 5 (6.4%) exhibited both. Tumors with perineural invasion required 5.3 surgical stages on average for clearance, in contrast to tumors without perineural invasion, which required 2.2 stages. Tumors with perineural inflammation, inflammation plus tumor invasion, and invasion alone were, respectively, 138%, 149%, and 194% greater in area preoperatively than tumors without perineural involvement, and their mean defect areas after Mohs surgery were, respectively, 151%, 121%, and 605% larger than those of tumors without perineural involvement.

CONCLUSIONS

The incidence of perineural invasion among cases of BCC appears higher than previously recognized. Tumor aggressiveness appears to correlate with the presence of perineural invasion. Surgery with horizontal frozen-section margin control enables easy detection of perineural involvement and should therefore be strongly considered for the treatment of high risk BCC patients.

Immune-associated cells in basal cell carcinomas of skin

Increased numbers of mast cells (MCs) and lymphocytes infiltrating in basal cell carcinomas (BCCs) have been observed. The presence of these infiltrating cells has been considered a sign of an immunologic anti-tumor response in the host, but the relationship of these two cell populations has not been examined. To elucidate this possible relationship, 30 non-ulcerated BCCs were analyzed. Frozen sections of the tumors were stained with monoclonal antibodies for Langerhans' cells, lymphocyte subsets and natural killer cells. Fluorescein isothiocynate (FITC)-avidin as well as anti-tryptase and anti-CD45RO monoclonal antibodies were used on formalin-fixed, paraffin-embedded sections for mast cell and T cell identification, respectively. B cells and natural killer cells were rarely observed in these tumors. MCs and T cells were quantified by direct enumeration and expressed as number of cells per high power field (hpf). FITC-avidin and anti-tryptase antibodies were equivalent in their ability to identify MCs. MC content in BCCs ranged from 1.0 to 31 cells/hpf. The number of T cells ranged from 0 to 50 cells/hpf with helper/suppressor cell ratios of 0.2 to 10. There was no correlation between helper/suppressor ratios and mast cell numbers; however, an inverse relationship was observed between the numbers of T cells and the number of mast cells in these tumors. These studies indicate that T cells and MCs are the primary immune cell populations responding to BCCs, and that decreased numbers of T cells are associated with more aggressive tumors.

Regression in basal cell carcinoma: an immunohistochemical analysis

Spontaneous regression of some cutaneous tumours is well recognized, and is thought to result from an immunological response to the tumour. Regression has previously been noted in basal cell carcinomas, but no studies defining the role of the immune response in the regression of this malignancy have been performed. We have examined 45 primary basal cell carcinomas (BCCs) (20 nodular, 25 superficial) and identified the cellular phenotypes and activation states of the cells infiltrating primary regressing and non-regressing BCCs, by immunocytochemistry. We have found a significantly increased number of CD3+ and CD4+ T cells infiltrating regressing compared with non-regressing tumours, and the expression of interleukin-2 receptor (an early activation marker for T cells) was also increased. There were no significant differences in class II major histocompatibility complex (MHC), CD1, or macrophage antigen expression in these groups. These findings suggest that activated CD4+ cytokine-secreting cells are important in the regression of BCCs.

Relative importance of prior basal cell carcinomas, continuing sun exposure, and circulating T lymphocytes on the development of basal cell carcinoma

This 36-month prospective study of a group of 61 people at high risk to develop multiple basal cell carcinomas (BCC) examined the circulating lymphocyte subsets of the population, patterns of sun exposure, and the longitudinal development of basal cell carcinoma. Sun exposure status was highly correlated with immune status defined by the CD4/CD8 T-lymphocyte ratio. There were significantly more BCC at 18 and 36 months in the 35 patients with high sun exposure and low CD4/CD8 ratio than in the 20 patients with low sun exposure and high CD4/CD8 ratio. A multivariate analysis assessed the relative importance of prior basal cell carcinoma, sun exposure, and immune status on the development of the skin cancer. Basal cell carcinoma developing in the previous 18 months and sun exposure during those 18 months were the first and second most important variables in determining development of basal cell carcinoma during the next 18 months. CD4/CD8 ratio had no additional predictive ability once prior skin cancers and sun exposure were accounted for. A low ratio of CD4/CD8 cells correlated with high sun exposure during the preceding 18 months.

Biology of basal cell carcinoma (Part II)

Host-tumor relationships involve several factors that can enhance or suppress neoplastic growth. This second part of a review of basal cell carcinoma biology examines the role that hormones, cytokines, local and systemic immunity, congenital and genetic syndromes, and environmental factors play in the development of this neoplasm. Theories of etiology and pathogenesis are discussed, and transplantation and cell culture techniques used to study this cancer are explored. Valuable second-line therapies for treatment of multiple tumors are reviewed, and important areas of present and future research are emphasized.

Biology of basal cell carcinoma (Part I)

Basal cell carcinoma is the most common malignancy in humans. Although rarely metastatic, it is capable of significant local destruction and disfigurement. This two-part article reviews the current understanding of basal cell carcinoma biology. Part I examines significant clinical, histologic, and ultrastructural features that relate to invasive potential. Genetic characteristics, including tumor growth rate, chromosomal abnormalities, and oncogene presence, are discussed, and expression of important cell and matrix proteins, including keratin, fibronectin, and HLA antigens, are reviewed. Further topics to be explored in Part II include host immunologic responses, theories of pathogenesis, and valuable second-line therapeutic regimens for treatment of multiple cancers.

Activation and inducer subset phenotype of the lymphocytic infiltrate around epidermally derived tumors

An in situ analysis of the mononuclear cell infiltrate found in association with a range of benign, premalignant, and malignant epidermal tumors is described. The predominant cell phenotype was that of the recently described immunoregulatory helper/inducer T lymphocyte. A large number of lymphocytes expressed antigens associated with cellular activation, suggesting an ongoing immunologic response by the host against the tumor, although evidence of in situ proliferation of these cells was lacking. These findings suggest that the infiltrate found in association with cutaneous tumors does not represent passive accumulation of lymphocytes from the circulation but rather an active antitumor response.

Occurrence of Langerhans cells and expression of class II antigens on keratinocytes in malignant and benign epithelial tumors of the skin: an immunohistopathologic study with monoclonal antibodies

We used an avidin-biotin complex immunoperoxidase technique with various monoclonal antibodies to determine Langerhans cell densities, class II antigen expression on keratinocytes, and phenotypes of other infiltrating cells in several malignant and benign epithelial tumors of the skin. Our observations indicate (1) there are few Langerhans cells in nests of basal cell carcinoma and squamous cell carcinoma; (2) there are increased Langerhans cell densities in seborrheic keratoses, verrucous epidermal nevus, and Bowen's disease; (3) there is an expression of class II molecules on the keratinocytes and cancer cells of basal cell carcinoma, squamous cell carcinoma, Bowen's disease, seborrheic keratosis, and verrucous epidermal nevus; and (4) there is a netlike staining of the keratinocyte surface with OKM5 in the epidermal lesion of seborrheic keratosis, verrucous epidermal nevus, and Bowen's disease, as well as in the epidermis adjacent to the basal cell carcinoma and squamous cell carcinoma nests.

Immunohistochemical staining of sebaceous cell carcinoma of the eyelid

We characterized the inflammatory infiltrate of two sebaceous cell carcinomas of the eyelid with immunohistochemical staining to determine the functional class of the mononuclear cells associated with the tumor. The results were compared with the inflammatory infiltrate associated with basal cell carcinomas. The subepithelial spaces and the area immediately surrounding the sebaceous cell neoplasms were free of mononuclear inflammatory cells, in contrast to the basal cell tumors, which had large numbers of subepithelial inflammatory cells and inflammatory cells in intimate contact with the neoplastic cells as previously reported. Inflammatory reaction in the sebaceous cell tumor was limited to a T-cell infiltrate surrounding the vessels adjacent to the tumor. The predominant mononuclear inflammatory cell in both the sebaceous cell and the basal cell carcinomas was the T helper cell. The apparent difference in mononuclear cell infiltrate may be a significant factor in the clinical behavior of the tumors.

Gamma interferon-induced expression of class II major histocompatibility complex antigens by human keratinocytes. Effects of conditions of culture

Normal human keratinocytes grown in MCDB 153 plus bovine pituitary extract and treated with recombinant gamma interferon (rIFN-gamma) express HLA-DR, DP and DQ antigens. The expression of these class II MHC antigens is time dependent: DR and DP appear before DQ. The delay in HLA-DQ expression is not due to the effects of trypsinization of cultures prior to analysis. Increasing the calcium ion concentration from 0.1 to 1.8 mM does not alter the expression of these antigens. Keratinocytes grown without serum proteins or bovine pituitary extract exhibited markedly delayed expression of DR. By contrast, keratinocytes grown in Dulbecco's modified Eagle's medium (DMEM) plus 10% fetal calf serum express DR and DP but only very small amounts of DQ after treatment with rIFN-gamma. Expression of HLA-DR occurs at doses of rIFN-gamma that are too low to cause growth inhibition. The cells of the squamous cell carcinoma cell line SCL-1, whether grown in MCDB 153 plus bovine pituitary extract or DMEM plus 10% fetal calf serum, express HLA-DQ and DP on only small numbers of cells after treatment with the lymphokine. Thus, the conditions of culture, possibly the presence of a serum factor(s), influence the expression of class II antigens in normal keratinocytes. Furthermore, rIFN-gamma does not induce DP and DQ antigens readily in transformed squamous cells cultured in either serum-containing or serum-free medium.

Expression of human lymphocyte antigen (HLA)-DR on tumor cells in basal cell carcinoma

Immunohistologic studies of eight patients with basal cell carcinoma were undertaken using a series of monoclonal antibodies. In all of the patients, the majority of dermal infiltrates reacted with OKT3 and OKIa1 (HLA-DR), with a slight predominance of OKT4+ helper/inducer T cells (the mean OKT4/OKT8 ratio was 1.8). Both OKT4+ and OKT8+ cells were seen infiltrating the tumor masses. In addition, in five cases, human lymphocyte antigen (HLA)-DR was demonstrated on some tumor cells close to a vast number of HLA-DR+ infiltrates surrounding the carcinoma, but not on epidermal keratinocytes and tumor cells devoid of the HLA-DR+ infiltrates. A considerable number of OKT6+ dendritic cells were also observed surrounding the carcinoma. Staining with OKB7 and OKM1 revealed negligible reactive cells, and virtually none of the dermal infiltrates reacted with Leu-7 (HNK-1). These findings suggest that in addition to varied immunologically competent cells, expression of HLA-DR antigen on tumor cells may participate in a cellular immune reaction, a defense mechanism against tumor cell proliferation in basal cell carcinoma.

Quantitative analysis of T6-positive Langerhans cells in human skin cancers

Epidermal Langerhans cells (LC) are required for antigen-presentation and for stimulating antigen-specific T cell activation. Similar functions may be important in the immune response to malignant skin tumours. Monoclonal anti-T6 antibody was used to examine LC population in basal and squamous cell carcinomas. Positive control labeling was performed with monoclonal anti-HLA-DR antibody. The number of T6-positive LC per mm2 of section was significantly decreased (p less than 0.01) in the tumour group in comparison with a sex and age-matched control group. The number of sun-exposed and covered regions was taken into consideration in each respective group. Within the tumours, LC were found more frequently in the tumour periphery and in most differentiated tumour areas (horn pearls) than in the rest of the tumour mass. T6-positive LC were rarely found in the dermis. Moreover, LC exhibited morphological changes in specimens from tumours. Staining with anti-HLA-DR antibody revealed less numerous positive cells within tumour nests than labeling with OKT6. A relationship between T6-positive LC quantities and extent of HLA-DR-positive infiltrates around tumours could not be established. These results suggest that immunological surveillance of neoantigen-bearing tumour cells may be impaired in skin cancer. A reason for the reduced LC number may be an altered microenvironment in tumour tissue.