The in vitro colony assay: a predictor of clinical outcome.

PURPOSE

To determine the predictive power of an in vitro colony assay on the clinical normal-tissue complication rate.

MATERIAL AND METHODS

Primary skin fibroblasts from 88 individuals were generated from the skin biopsies of patients who received a standardized radiotherapy. Tissue was cultured for three to six passages, irradiated with doses between 1 and 8 Gy under defined conditions, seeded and finally the colonies were stained and counted after 10-14 days. The survival curves were fitted by the L-Q model and the SF2, alpha/beta and plating efficiency were calculated.

RESULTS

The parameters SF2 and plating efficiency were stable throughout the 4-year test period. Intra-individual differences between repeated experiments were significantly lower than inter-individual test results. For the observed acute skin and late normal-tissue reactions other than skin the in vitro parameter SF2 correlated significantly (p<0.005). For late skin reactions this correlation was not found.

DISCUSSION

In contrast to other publications, a clear correlation was found between the in vitro test results and clinically observed early reactions. The lack of correlation for late skin reactions suggests that the combination of intrinsic radiation sensitivity and exogenous factors may alter the clinically observed reaction of certain tissues to a different extent.