Defining recurrence of nonmelanoma skin cancer after Mohs micrographic surgery: Report of the American College of Mohs Surgery Registry and Outcomes Committee.

BACKGROUND: Standardized definitions and methods of surveillance for local recurrence of nonmelanoma skin cancer are critical in determining cure rates attributed to treatment modalities. OBJECTIVE: We sought to offer a standard definition of local recurrence after surgical treatment of nonmelanoma skin cancer and to propose an acceptable surveillance period and tracking methods. METHODS: A literature search was performed for background definitions of local recurrence and tracking methods. The American College of Mohs Surgery (ACMS) Registry and Outcomes Committee then conducted a modified Delphi process to arrive at consensus definitions. RESULTS: We define local recurrence as a tumor with comparable histology, with contiguity to the surgical scar after treatment, and that arises within the area of the previously treated tumor. LIMITATIONS: This project reports the results of a modified Delphi method process involving members of the ACMS. The model described may not be useful for nonexcision type treatments such as topical chemotherapy, electrodessication and curettage, or radiation treatment. CONCLUSIONS: Previous definitions of recurrence and surveillance methods after surgical treatment of nonmelanoma skin cancer are variable and nonstandard. We describe consensus standards for defining and tracking recurrence that should allow for consistent scientific evaluation and development of performance data in skin cancer outcomes registries.

Several human PATCHED1 mutations block protein maturation.

The tumor suppressor gene PATCHED1 (PTCH1) is mutated in sporadic and inherited forms of basal cell carcinoma. PTCH1 binds Hedgehog proteins and inhibits signaling in the absence of ligand. Although PTCH1 mutations are proposed to reduce or abolish protein function, few mutations have been tested for activity. We introduced six PTCH1 missense mutations into mouse patched1 and tested them in murine cells deficient for patched1 function. Three mutants retained significant activity. Three other mutants had little or no function, and of these, two were retained in the secretory pathway. These studies indicate that missense mutations can abolish PTCH1 function by blocking protein maturation.

Comparative gene expression profile analysis of GLI and c-MYC in an epithelial model of malignant transformation.

Transcription factor oncogenes such as GLI and c-MYC are central to the pathogenesis of human tumors. GLI encodes a zinc finger protein that is activated by Sonic Hedgehog signaling. Mutations in this pathway induce GLI expression in basal cell carcinoma, and expression of GLI in mice is sufficient to induce these skin tumors. We used microarrays to identify transcripts regulated by GLI or c-MYC after retroviral transduction and short-term culture of epithelial RK3E cells. Although each of these oncogenes induces malignant transformation of RK3E, two distinct sets of genes were identified. Of approximately 17,500 transcripts represented on the microarrays, GLI up-regulated the expression of 158 and repressed the expression of 52. In contrast, transcripts regulated by c-MYC were mainly repressed (424 of 682 regulated transcripts). Transcripts induced by the GLI transgene are likewise expressed in association with endogenous GLI in Ptch-deficient murine fibroblasts or in human skin tumors, but are not up-regulated in RK3E cells transformed by c-MYC, KLF4, or HRAS1. Unlike these other oncogenes, GLI induced the expression of mesenchymal cell markers including Snail, a zinc finger protein implicated in epithelial-mesenchymal transition in development and during tumor progression. A novel GLI-estrogen receptor fusion protein rapidly induced Snail mRNA expression in a manner like Ptch, a known direct transcriptional target gene. Induction of Snail expression and epithelial-mesenchymal transition by GLI may account for certain histopathological features of basal cell carcinoma, such as the absence of a well-defined, intraepithelial precursor lesion. In addition, consistent expression of the newly identified GLI-induced transcripts within GLI-expressing tumors in vivo indicates that oncogene-specific transcriptional profiles may be useful diagnostic tools for analysis of human tumors.

Distinct consequences of sterol sensor mutations in Drosophila and mouse patched homologs.

The membrane protein Patched (Ptc) is a critical regulator of Hedgehog signaling. Ptc is among a family of proteins that contain a sterol sensor motif. The function of this domain is poorly understood, but some proteins that contain sterol sensors are involved in cholesterol homeostasis. In the SREBP cleavage-activating protein (SCAP), sterols inhibit the protein's activity through this domain. Mutations in two highly conserved residues in the SCAP sterol sensor have been identified that confer resistance to sterol regulation. We introduced the analogous mutations in the sterol sensor motif of fly Ptc and mouse Ptc1 and examined their effect on protein activity. In contrast to SCAP, the sterol sensor mutations had different affects on Drosophila Ptc; Ptc Y442C retained function, while Ptc D584N conferred dominant negative activity. In the wing imaginal disc, Ptc D584N overexpression induced Hedgehog targets by stabilizing Cubitus interruptus and inducing decapentaplegic. However, Ptc D584N did not induce collier, a gene that requires high levels of Hedgehog signaling. In mouse Ptc1, the Y438C and D585N mutations did not stimulate signaling in Shh-responsive cell lines but did complement murine ptc1(-/-) cells. The results suggest that mutations in sterol sensor motifs alter function differently between sterol sensor family members.

Several PATCHED1 missense mutations display activity in patched1-deficient fibroblasts.

Mutations in mouse and human patched1 (ptc1) genes are associated with birth defects and cancer. Ptc1 is a receptor for Hedgehog (Hh) signaling proteins. Hh proteins activate transcription of target genes, including ptc1, and Ptc1 represses those genes, both by regulating the activity of Gli transcription factors. We have established mammalian cell lines with reduced Ptc1 function and a lacZ reporter to investigate Hh signal transduction. Embryonic fibroblasts were derived from mice, heterozygous or homozygous for a ptc1 mutation that inserts lacZ under the control of the ptc1 promoter (ptc1-lacZ). In heterozygous ptc1 cells, ptc1-lacZ was expressed at low levels but could be induced by Sonic Hedgehog (Shh) and Gli-1. Homozygous ptc1 cells expressed high levels of ptc1-lacZ without Hh stimulation. ptc1-lacZ expression was dependent on cell density in ptc1 homozygotes and Hh-stimulated heterozygotes but was independent of density when Gli1 was used to activate ptc1-lacZ. A wild-type ptc1 transgene introduced into homozygous ptc1 cells greatly reduced ptc1-lacZ expression. Expression of either half of Ptc1 alone resulted in improper maturation of the protein and a failure to complement the ptc1(-/-) cells. When co-expressed, both Ptc1 halves matured and had an activity similar to that of the intact protein. Three missense PTCH1 mutations exhibited significant functions in homozygous ptc1 cells. The missense mutants retained activity when expressed at about 10-fold lower levels and appeared as stable as wild-type Ptc1. These studies suggest that some tumors and disease phenotypes may arise from small reductions in PTCH1 activity.