DNA ploidy and proliferation heterogeneity in human prostate cancers.

DNA ploidy determinations have been shown to have clinical application in predicting disease progression, survival, or response to anti-androgen therapies in prostate carcinomas. Since intra-tumor heterogeneity may have a profound effect on DNA measurements, we determined the frequency of DNA ploidy and proliferation (here S-phase fraction) heterogeneity in early prostatic carcinomas, and estimated the potential impact of heterogeneity on predicting disease course, survival, or response to therapy. Using image and flow cytometric analysis of archival, paraffin-embedded prostate tumors, we measured DNA ploidy in individual foci of prostatic carcinoma in stage T1a, T1b and T1c disease. Image analysis studies included the use of Feulgen stained tissue sections, and a comparison of these results with flow cytometric DNA ploidy determinations on nuclei isolated from the same tumor foci. Flow cytometry was also used to measure DNA Index and tumor S-phase fraction, in some cases using multiparameter analysis of isolated nuclei to determine DNA content and the level of the proliferation-associated antigen, p105. Our results indicate that DNA aneuploid foci of prostate carcinoma are infrequently seen in stage T1a disease (13% of the individuals studied), and that the presence of both DNA diploid and aneuploid foci in the same sample is seen in less than 10% of these individuals. Stage T1b and T1c tumors containing only DNA diploid nuclei are seen, though these are likely most common in low volume, low Gleason grade tumors. By using flow cytometry to compare these results with those using image analysis of the same tumor foci, we demonstrated that the majority (> 75%) of these aneuploid tumors are DNA tetraploid. Our data on prostate tumor S-phase fractions indicate that DNA diploid tumors generally have a lower S-phase than DNA aneuploid foci (including comparisons of DNA diploid and aneuploid foci in the same prostate tumor). These results support the model that early prostate tumors are DNA diploid and have a low S-phase, and that these tumors likely evolve to DNA tetraploid tumors with a similar low S-phase fraction.