Dietary immunosuppressants do not enhance UV-induced skin carcinogenesis, and reveal discordance between p53-mutant early clones and carcinomas.

Immunosuppressive drugs are thought to cause the dramatically increased risk of carcinomas in sun-exposed skin of organ transplant recipients. These drugs differ in local effects on skin. We investigated whether this local impact is predictive of skin cancer risk and may thus provide guidance on minimizing the risk. Immunosuppressants (azathioprine, cyclosporine, tacrolimus, mycophenolate mofetil, and rapamycin) were assessed on altering the UV induction of apoptosis in human skin models and of p53 mutant cell clones (putative tumor precursors) and ensuing skin carcinomas (with mutant p53) in the skin of hairless mice. Rapamycin was found to increase apoptosis (three-fold), whereas cyclosporine decreased apoptosis (three-fold). Correspondingly, a 1.5- to five-fold reduction (P = 0.07) or a two- to three-fold increase (P < 0.001) was found in cell clusters overexpressing mutant p53 in chronically UV-exposed skin of mice that had been fed rapamycin or cyclosporine, respectively. Deep sequencing showed, however, that the allelic frequency (∼5%) of the hotspot mutations in p53 (codons 270 and 275) remained unaffected. The majority of cells with mutated p53 seemed not to overexpress the mutated protein. Unexpectedly, none of the immunosuppressants admixed in high dosages to the diet accelerated tumor development, and cyclosporine even delayed tumor onset by approximately 15% (P < 0.01). Thus, in contrast to earlier findings, the frequency of p53-mutant cells was not predictive of the incidence of skin carcinoma. Moreover, the lack of any accelerative effect on tumor development suggests that immunosuppressive medication is not the sole cause of the dramatic increase in skin cancer risk in organ transplant recipients.