CASE REPORT
A new generation of anticancer pharmaceuticals designed to inhibit specific malignancy-associated enzymes has become available in recent years. Vemurafenib represents one such medication and targets the V600E-substituted B-Raf protein, which may be found in 40 to 60% of melanomas.1 Though shown to be clinically efficacious in advanced melanoma treatment, vemurafenib produces cutaneous
side-effects in some patients, including squamous cell carcinomas, keratoacanthomas, palmar-plantar keratoses, keratosis-
pilaris-like eruptions, and verrucous keratoses.2 Studies into the pathogenesis of these lesions have focused on alterations
in the mitogen-activated protein kinase (MAPK) pathway, in which B-Raf participates. Biochemical and cell culture studies
have identified a mechanism by which wild-type B-Raf, in the presence of vemurafenib and an oncogenic RAS, paradoxically
upregulates the pathway.3 Such RAS mutations at hotspot codons (12, 13, and 61) have been identified in a high percentage
of squamous cell carcinomas from vemurafenib-treated patients4 but have not been demonstrated in the spectrum of benign verrucous keratoses. In addition, many of the keratoses observed in vemurafenib-treated patients appear similar to lesions
observed in cutaneous syndromes associated with RAS mutations, such as cardiofaciocutaneous syndrome, Costello syndrome, and Noonan syndrome.5 In order to further characterize these keratoses, we genotyped RAS orthologs KRAS and NRAS at hotspots associated with oncogenic activation in three of these lesions from a vemurafenib-treated melanoma patient.Three verrucous keratoses, which had been previously biopsied from the face (Figure 1) and back of a 41-year old woman with stage IV melanoma, were analyzed. The patient had begun taking vemurafenib 960 mg twice daily one month before presenting with these lesions. DNA from formalin-fixed paraffin-embedded tissue was PCR-amplified at the regions of KRAS and NRAS codons 12, 13, and 61. Pyrosequencing (PyroMark System by Qiagen) was performed on the DNA to detect mutations at these sites. Wild-type and known mutant DNA were used for controls. In addition, BRAF codon 600 was also sequenced. PyroMark software was used to calculate percentage mutation of each nucleotide when detected. This study was performed in compliance with the Western Institutional
Review Board (Olympia, WA).