Objective: We tested the hypothesis that p53 frameshift mutations in ovarian cancer occur as a result of genomic instability rather than as a proximal cause of this process.
Study design: Sequencing of the p53 tumor suppressor gene has been carried out on 305 ovarian, fallopian tube, and peritoneal cancers. Two groups of p53 null mutations were identified: (1) those caused by frameshift insertion or deletion mutations (n = 31) and (2) those caused by nonsense mutations (n = 28). As a control group 59 tumors with p53 missense mutations were selected by matching with the p53 null tumors on the basis of patient age at diagnosis, stage and grade of cancer, cancer site, and year of diagnosis. Microsatellite instability was determined from paired normal and tumor tissue deoxyribonucleic acid by means of the following different markers: D2S123, D5S346, D17S250, BAT25, and BAT26. Amplimers from polymerase chain reactions were evaluated on 7% polyacrylamide gels.
Results: The p53 null tumors were more likely to be of higher stage and grade. Fallopian tube cancers were more common (P =.02) in the p53 frameshift group. The overall incidence of microsatellite instability was 39%, 36%, and 25% for tumors with p53 frameshift nonsense and missense mutations (P =.30). Microsatellite instability was seen almost exclusively with ovarian cancer (P =.04).
Conclusions: Microsatellite instability is a relatively common event in ovarian cancer and is dependent on marker selection. The p53 frameshift mutations do not appear to occur as a consequence of genomic instability.