Molecular genetic analysis excludes implantation metastasis of basal cell carcinomas

Archives of Pathology & Laboratory Medicine, Sep 2003 by Hafner, Christian, Hartmann, Arndt, Knuechel, Ruth, Dietmaier, Wolfgang, Et al

* Basal cell carcinoma (BCC) of the skin is the most common tumor in the white population. A 66-year-old man developed 2 BCCs at the left parietal region of the head and at the helix of the left ear. The 2 lesions matched exactly when pressing the ear against the head, suggesting an implantation metastasis mechanism. Molecular genetic techniques were used to confirm or exclude such a mechanism in this rare clinical constellation. Tumor tissues were precisely microdissected for DNA isolation. Exons 5-9 of the p53 tumor suppressor gene were directly sequenced. In addition, loss of heterozygosity analysis of chromosome 9q was performed using 5 polymorphic microsatellite markers. The BCC of the ear revealed a p53 mutation at codon 273, whereas the other one lacked this mutation. In addition, the smaller BCC of the ear showed loss of heterozygosity at 9q33.3, in contrast with the larger BCC. Interestingly, histologically normal skin of the ear distant from the small BCC had the same deletion, indicating a field defect of this skin patch at 9q33.3. Molecular genetic analysis clearly demonstrated different genetic alterations of the two BCCs and therefore most likely excludes a mechanism of implantation metastasis.

Basal cell carcinoma (BCC) of the skin is the most common type of cancer in the white population. It is characterized by local invasive and destructive growth, but only occasionally a metastatic course. To our knowledge, implantation metastasis by direct inoculation of malignant cells to neighbored sites in multiple intraindividual BCC has never been reported yet. Some studies describe experimental implantation of human BCCs to immunosuppressed nude mice subcutaneously or by grafting technique, but the efficiency was low.1 Implantation metastasis is known from other tumor entities like multifocal urothelial carcinoma, where this mechanism is called intraluminal seeding.2 Microscopic lesions of the urothelial tract favor the implantation of shed malignant cells. Hence, it is thinkable that direct contact implantation of malignant cells works in BCC as well, perhaps facilitated by microlesions of the epidermis.

We report a case of a man with 2 BCCs, of which we were suspicious of implantation metastasis. To investigate whether the 2 BCCs had developed by an implantation metastasis mechanism or independently of each other, molecular genetic techniques were used to investigate a possible clonal relationship between the two lesions. Sequencing of the p53 tumor suppressor gene was performed because p53 mutations are frequent events in the carcinogenesis especially of late-onset BCC.3 However, a functional role of p53 in the carcinogenesis of BCC still remains questionable because most BCCs do not exhibit genomic instability.4 Additionally, loss of heterozygosity (LOH) of chromosome 9q was analyzed. Chromosome 9q bears the PTCH1 gene (9q22), which has been reported to be frequently deleted and mutated in BCC.5 The PTCH1 gene encodes for a transmembrane receptor, which binds the ligand Sonic hedgehog (Shh). Ptch1 inhibits signaling by suppressing the cell-surface molecule, Smoothened (Smo). Upon binding of Shh to Ptch1, the suppression of Smo is antagonized and, finally, Shh target genes like GH1 are activated. Shh signaling is important for the regulation of proliferation and patterning of multiple tissues and organs during embryogenesis and is also required for the control of hair follicle keratinocyte growth. In human BCC, loss of PTCH1 or gain of function of Smo leads to permanent signaling independent of Shh.4

Detection of molecular genetic alterations provides specific fingerprints characterizing each tumor and is a suitable method for clonality analysis of multifocal intraindividual tumors, as demonstrated for other tumor entities like multifocal urothelial carcinoma.2 It is essential that the alterations occur at early stages of the lesions and are stable during tumor progression. However, for the purpose of clonality analysis, it is not necessary that the molecular genetic markers be functionally involved in carcinogenesis. In the present case, we could show that the 2 BCCs are characterized by different genetic alterations. Therefore, they are most likely not related to each other, making a possible implantation metastasis mechanism very unlikely.

REPORT OF A CASE

A 66-year-old male patient developed simultaneously 2 BCCs. One large ulcerated BCC (B1) was localized at the left parietal region above the ear. The other one (B2) was quite small and grew at the helix of the left ear (Figure 1). The 2 lesions were not contiguous, but at locations opposed to each other. The BCCs matched exactly when pressing the ear against the head, a situation that physiologically may have occurred for months when the patient was sleeping. This rare clinical constellation suggests a possible implantation metastasis mechanism of the malignant basal cells from the BCC on the head to the opposite helix of the ear or vice versa.

MATERIALS AND METHODS DNA Isolation and Whole Genome Amplification