Independent prospective validation of a predictive test for risk of radiation induced fibrosis based on the gene expression pattern in fibroblasts irradiated in vitro.

PURPOSE: In a previously published study, we established a predictive test for the risk of radiation-induced fibrosis based on the gene expression pattern in cultured fibroblast irradiated in vitro. The present study was conducted to seek an independent prospective validation of the predictive test in a cohort of patients given curative radiotherapy for head and neck cancer.
MATERIALS AND METHODS: The study cohort comprised 160 consecutive head and neck cancer patients given curative radiotherapy between 2000 and 2004. The patients were treated according to the DAHANCA protocols. The patients were scored for subcutaneous fibrosis as part of routine follow up. Fibroblast culture was established from skin biopsies. The fibroblasts were irradiated in vitro using a fractionation scheme of 3 times 3.5 Gy. The expression of 9 genes was assessed before and after irradiation of the cells using real time PCR.
RESULTS: Based on the radiation-induced expression of the assessed genes, the material was divided into 136 patients having the 'sensitive expression profile' and 24 patients having the 'resistant expression profile'. Within the subset of patients with the 'sensitive profile', the cumulative risk of severe fibrosis was 34% at 9 years (Kaplan-Meier) whereas no patients with the 'resistant profile' developed severe fibrosis (p = 0.035).
CONCLUSION: Our study provided an independent prospective validation of the previously established predictive test for radiation induced fibrosis. In agreement with our initial findings, the classifier was able to identify a smaller subset of patients that seems to be rather radioresistant and could therefore potentially be considered for dose escalation.