Single biopsy, tumor kinetic analyses: a comparison of methods and an extension to shorter sampling intervals

There is emerging and established clinical and laboratory evidence that proliferation of tumor clonogens during radiation therapy can impair local tumor control. The pre-treatment, tumor potential doubling time, T(pot), estimated with in situ bromodeoxyuridine (BrdUrd) labeling, followed by a single biopsy and flow cytometry, may be a predictor of a given tumor's ability to undergo such intra-treatment proliferation. Recent studies have found a strong similarity between T(pot)'s determined in this fashion and the effective doubling times of surviving tumor cells during radiotherapy, as estimated from tumor control versus treatment duration data. Furthermore, several preliminary clinical studies have indicated that T(pot) may be a predictor of outcome, with faster tumors doing worse. Accelerated fractionation might overcome such proliferation, but is more acutely toxic and is unlikely to benefit patients with slowly proliferating tumors. Thus, the BrdUrd/single biopsy method may offer the possibility of selecting between accelerated and conventional or hyperfractionated treatment. Several approaches to the analysis of data generated by this method have been described, but there has been little documentation of the validity of methods in experimental systems, particularly in human experimental tumors. This study explores various analytic methods employed with the BrdUrd, delayed, single-biopsy technique used in determining the potential doubling time, T(pot), of tumors. It compares methods of analysis in three experimental systems and in 40 in situ-labeled human tumors, and proposes a method for shortening the required labeling-biopsy interval to a clinically more convenient range of 3 to 4 hours.