BACKGROUND/AIMS For the evaluation of response to cytostatic agents, determination of changes in tumor growth speed seems to be essential. RECIST, which lacks the dimension of time, may not be particularly suitable for these agents. The aim of this study is to elucidate the usefulness of tumor marker doubling time (DT) in the evaluation of response to a molecular target agent. PATIENTS AND METHODS A total of 25 patients with advanced HCC participated in the phase I/II clinical trial of TSU-68, a novel tyrosine kinase inhibitor of VEGF. We enrolled 15 patients whose alpha-fetoprotein (AFP) or des-gamma-carboxyprothrombin (DCP) showed exponential increase during wash-out phase before TSU administration. We analyzed the relationship between tumor marker DT and tumor volume DT measured on two CT examinations before the commencement of TSU-68 administration. Next, tumormarker DT in the first 8 weeks of TSU-68 administration was similarly calculated, and compared with DT before TSU-68 administration. DT should be elongated, or turn negative, if TSU-68 was effective. The evaluation based on tumor marker DT was compared with RECIST. Lastly, tumor marker DT after the cessation of TSU-68 was calculated and compared with DT in 4 weeks immediately before the cessation. RESULTS DT of tumor marker level and that of tumor volume before TSU-68 administration were almost identical (R2=0.94, P<0.001). The evaluation based on tumor marker DT comparison between before and during TSU-68 administration and that based on RECIST were compatible in most cases (12/15); discordance was observed in 3 cases judged as disease progression by RECIST due to substantial tumor necrosis without volume shrinkage or appearance of new lesions in spite of apparent effects on target lesions. In these cases DT elongation was observed, which suggested cytostatic effect of TSU-68. Tumor marker DT after the cessation of TSU86 administration was evaluated in 4 cases, where DT was shortened after the cessation. CONCLUSION Serum tumor marker levels can be used to evaluate viable tumor burden irrespective of the presence of tumor necrosis, which may compromise radiographic evaluation. This may be applicable to the evaluation of response to chemotherapy with cytostatic agents in particular.