Is chemosensitivity testing for peri-operative treatment planning in gastro-intestinal cancer by the human tumour colony assay worthwhile?

Based on the concept of a combined modality cancer treatment in surgical oncology, the use of the human tumour colony assay for routine chemosensitivity testing and prospective treatment planning was investigated in 204 surgical biopsies of primary human solid tumours. The majority of the tumours (135/204) were of gastro-intestinal (GI) origin. Sufficient growth for drug testing occurred in 29-67% of all tumours depending on the tumour type, with a mean of 36% in GI-carcinomas. Chemosensitivity testing in vitro against standard anti-cancer agents correlated well with clinical experience, 5-FU and FUDR being the most active drugs (27% respectively 24% sensitive tumours in vitro) in GI-carcinomas. Relatively good agreement of in vitro/in vivo correlations was seen with an overall of 25/32 correct predictions in GI and other tumours. Predictivity was particularly good for loco-regional chemotherapy. Nevertheless, the limited in vitro growth rate of gastro-intestinal tumour specimens and their chemoresistance restrict the use of this method-in particular with respect to individual treatment planning.