Alteration of the p53 gene is one of the most frequent events in human tumorigenesis. The inactivation of p53 tumor suppressor function can be caused by chromosome deletion, gene deletion, or mainly by point mutations. p53 mutations occur moderately often in hematopoietic malignancies. A significantly higher frequency of p53 alterations in cell lines vs primary samples has been observed for all types of malignant hematopoietic cell lines. It has been postulated that p53 gene abnormalities arise in cell lines during in vitro establishment of the culture or prolonged culture; but it is also conceivable that those cases that carry p53 mutations may be more suitable for in vitro establishment as permanent cell lines. We analyzed data on the p53 gene status in a panel of matched primary hematopoietic tumor cells and the respective cell lines derived from this original material. In 85% (53/62) of the pairs of matched primary cells and cell lines, the in vivo and in vitro data were identical (both with p53 wild-type or both with the same p53 mutation). In some instances, serial clinical samples (eg at presentation and relapse) and serial sister cell lines were available. These cases showed that a clinical sample at presentation often had a p53 wild-type configuration whereas the derived cell line and a relapse specimen carried an identical p53 point mutation. These findings suggest that a minor clone, at first undetectable by standard analysis, represents a reservoir for the outgrowth of resistant cells in vivo and also a pool of cells with a growth advantage in vitro, providing a significantly higher chance of immortalization in culture. This was further supported by studies employing mutant allele-specific gene amplifications, a technique which is significantly more sensitive (100- to 1000-fold) than the commonly applied SSCP assay with a sensitivity threshold of about 10% mutated cells within a pool of wild-type cells. Taken together, this analysis confirms the usefulness of human hematopoietic cell lines as in vitro model systems for the study of the biology of hematopoietic malignancies. It further underlines the notion that p53 gene alterations confer a survival advantage to, at least some, malignant cells in vitro and presumably also in vivo; however, it is highly unlikely that a p53 mutation alone would suffice for the immortalization of a cell line in vitro or tumor development in vivo. Leukemia (2000) 14, 198-206.