Growth of cell colonies in soft agar from biopsies of different human solid tumors.

Abstract

Abstract Colony growth in soft agar of cells disaggregated from 87 solid tumor specimens was evaluated. Tumors were disaggregated by an enzymatic method consisting of microtome slicing of tissues and incubation at 37° for 2 hr in Roswell Park Memorial Institute Tissue Culture Medium 1640 containing 10% fetal bovine serum, 0.8% collagenase II, and 0.002% DNase I. Tumor cells were cultured in a two-layer agar system (0.5% agar feeder, 0.3% agar plating layer) in enriched medium with or without addition of different factors. Fifty-eight of 87 malignant tumors (16 of 22 melanomas, 11 of 16 sarcomas, 8 of 16 pulmonary carcinomas, 8 of 11 colon carcinomas, 5 of 8 breast carcinomas, 2 of 3 carcinomas of unknown site, 2 of 2 ovarian carcinomas, 1 of 1 urinary bladder carcinoma, 0 of 1 kidney carcinoma, 1 of 1 epiglottic carcinoma, 1 of 1 pancreatic carcinoma, 1 of 1 pleural mesothelioma, 1 of 1 lung hemangiopericytoma, 0 of 1 malignant schwannoma, 0 of 1 malignant giant cell tumor, and 1 of 1 neuroblastoma) grew in the soft-agar system yielding an overall cloning success rate of 67%. The number of colonies (30 or more cells) after plating 5 × 105 cells ranged from 50 to 3774 per plate, yielding plating efficiency from 0.01 to 0.8%. Considerable variations in plating efficiency were noticed among tumors of the same type. A linear relationship was obtained between the number of cells plated and the number of colonies formed. Tumor cells washed after enzymatic disaggregation yielded more colonies per plate than did tumor cells plated without washing. Refrigeration of tumor tissue for 24 hr in Roswell Park Memorial Institute Tissue Culture Medium 1640 containing 10% fetal bovine serum did not decrease the number of colonies. The number of colonies formed was not increased with two different cell-conditioned media. Murine and rat red blood cells did not affect tumor colony growth. The cells plucked from individual colonies and stained by the Wright-Giemsa and Papanicolaou methods had the same morphological characteristics as did tumor cells in the original cell suspension. The histological pattern of the original tumors was maintained after passage from agar methyl cellulose into the nude mouse.