Abstract
Background: Drugs targeting raf proteins like sorafenib or PLX4032 can induce skin neoplasms : keratoacanthomas (KA) and KA-like squamous cell carcinomas (SCC) in about 7% of the patients (pts) with sorafenib and 23% of pts with PLX4032. The occurrence of these borderline or malignant skin tumors in pts treated with anticancer agents is intriguing and has no clear biological explanation.
Methods: A prospective ancillary dermatological study was performed on pts treated with sorafenib or placebo for renal cell cancer (RCC) during a randomized double blind study. Biopsies of normal skin were studied for keratinocyte proliferation and expression of proteins involved in the MAP-kinase and PI3-Kinase pathways by immunohistochemistry. 30 skin lesions from additional pts treated with sorafenib for RCC, hepatocarcinoma or melanoma were studied with the same methods. DNA was extracted from these tumors and sequenced for mutation hotspots of BRAF, NRAS, HRAS, KRAS, AKT, PI3KCA genes.
Results: Normal skin biopsies from 12 pts treated with sorafenib and 8 pts with placebo showed a moderate increase of keratinocyte proliferation index in normal skin of pts with sorafenib compared to placebo. P-ERK expression was significantly increased in normal skin of sorafenib treated pts vs placebo treated pts. Pathology analysis of 30 proliferating skin lesions occurring on 21 pts during sorafenib tt showed 15 KA or KA-like SCC, 12 cystic lesions and 3 pre-neoplastic keratoses. P-ERK was expressed in all of them and p-AKT in 26/30 (87%). HRAS mutations were found in two benign cystic lesions and one KA. No BRAF mutation was found.
Conclusions: Although sorafenib is a potent anticancer agent, it stimulates keratinocyte proliferation and activates ERK phosphorylation in normal skin. There is a continuous spectrum of skin proliferation from cystic lesions to KA and SCC that may involve mutational events like raf protein mutations. Paradoxical activation of the MAP-kinase pathway and tumor induction in the skin by raf inhibition has to be further studied and understood as it might be an issue for more potent raf inhibitors presently in development.
Methods: A prospective ancillary dermatological study was performed on pts treated with sorafenib or placebo for renal cell cancer (RCC) during a randomized double blind study. Biopsies of normal skin were studied for keratinocyte proliferation and expression of proteins involved in the MAP-kinase and PI3-Kinase pathways by immunohistochemistry. 30 skin lesions from additional pts treated with sorafenib for RCC, hepatocarcinoma or melanoma were studied with the same methods. DNA was extracted from these tumors and sequenced for mutation hotspots of BRAF, NRAS, HRAS, KRAS, AKT, PI3KCA genes.
Results: Normal skin biopsies from 12 pts treated with sorafenib and 8 pts with placebo showed a moderate increase of keratinocyte proliferation index in normal skin of pts with sorafenib compared to placebo. P-ERK expression was significantly increased in normal skin of sorafenib treated pts vs placebo treated pts. Pathology analysis of 30 proliferating skin lesions occurring on 21 pts during sorafenib tt showed 15 KA or KA-like SCC, 12 cystic lesions and 3 pre-neoplastic keratoses. P-ERK was expressed in all of them and p-AKT in 26/30 (87%). HRAS mutations were found in two benign cystic lesions and one KA. No BRAF mutation was found.
Conclusions: Although sorafenib is a potent anticancer agent, it stimulates keratinocyte proliferation and activates ERK phosphorylation in normal skin. There is a continuous spectrum of skin proliferation from cystic lesions to KA and SCC that may involve mutational events like raf protein mutations. Paradoxical activation of the MAP-kinase pathway and tumor induction in the skin by raf inhibition has to be further studied and understood as it might be an issue for more potent raf inhibitors presently in development.
Original language | English |
---|---|
Article number | 9130 |
Journal | Journal of Clinical Oncology |
Volume | 28 |
Issue number | 15_Supplement |
DOIs | |
Publication status | Published - 20 May 2010 |
Bibliographical note
No significant financial relationships to disclose.© 2010 by American Society of Clinical Oncology