Methodologic problems in clonogenic assays of spontaneous human tumors.

Colony formation in soft agar was used to investigate growth properties and drug sensitivity in 102 tumor specimens from 91 patients. Sufficient colony growth for sensitivity testing with various drugs was obtained in 36 of 67 specimens (54%) with adequate cell yield and pathologically documented malignancy. Room temperature (20-24 degrees C) is superior to both 4 degrees C and 37 degrees C for 12-36 h storage and transport of malignant effusions. By contrast, fine mincing in sterile saline or balanced salt solution, and refrigerated storage (4 degrees C) appear optimal in experiments with three solid tumors. The use of buffered NH4Cl to lyse red blood cells markedly reduced plating efficiencies, and also reduced the percentage of tumors in which drug sensitivities could be tested from 64% to 38%. Several combinations of potential growth factors and culture media have been tested. Insulin enhanced plating efficiency (PE) in all six adenocarcinomas tested. Drug sensitivity of tumors was not affected by varying plating efficiency up to five-fold in two tumors. In eleven cases tumor cells were exposed to combinations of two or more drugs, and results assessed for evidence of drug interactions. In almost all cases, these two-drug combinations produced additive cell killing than either antagonistic or greater-than-additive effects.U