MOLECULAR STAGING OF PROSTATE-CANCER .2. A COMPARISON OF THE APPLICATION OF AN ENHANCED REVERSE-TRANSCRIPTASE POLYMERASE CHAIN-REACTION ASSAY FOR PROSTATE-SPECIFIC ANTIGEN VERSUS PROSTATE-SPECIFIC MEMBRANE ANTIGEN

Current imaging modalities used to stage prostate cancer clinically fa il to detect extracapsular disease in a significant subset of patients . A molecular based peripheral blood assay using the reverse transcrip tase polymerase chain reaction has recently been shown to be a highly sensitive staging modality for detecting extraprostatic disease preope ratively. The assay uses primers that are specific for prostate specif ic antigen (PSA). We compare the application of the reverse transcript ase polymerase chain reaction assay using primers specific for the hum an prostate specific membrane antigen with results obtained from the s ame specimens by reverse transcriptase polymerase chain reaction for P SA. Prostate specific membrane antigen, a recently cloned prostatic an tigen, is a transmembrane glycoprotein that has been described as pros tate specific. These assays were applied to ribonucleic acids extracte d from the peripheral blood lymphocyte fraction of 80 patients with cl inically localized prostate cancer. In addition, blood specimens from 20 female patients, 20 young male patients, 25 age-matched control men under treatment for benign prostatic hypertrophy and 20 men with esta blished, untreated metastatic prostate cancer were tested.All 3 groups of noncancer patients had negative polymerase chain reactions for PSA as well as prostate specific membrane antigen. Of 20 metastatic prost ate cancer patients 16 (80%) had positive polymerase chain reactions f or PSA, while only 10 (50%) had positive results for prostate specific membrane antigen. Among the 80 patients with clinically localized dis ease (stages T1 to T2cN0M0), 27 and 19 had positive polymerase chain r eaction for PSA and prostate specific membrane antigen, respectively, from blood specimens obtained preoperatively. Analyzing the final path ology in each patient with the reverse transcriptase polymerase chain reaction assay identified a significantly stronger correlation with tu mor invasion using the results of the PSA test rather than the results of the prostate specific membrane antigen reverse transcriptase polym erase chain reaction test (67% versus 34% sensitivity for detecting ca psular penetration, 87% versus 46% sensitivity for detecting disease t o the surgical margin and 83% versus 16% sensitivity for detecting sem inal vesicle invasion). In contrast to the reverse transcriptase polym erase chain reaction assay for PSA, a similar assay done for prostate specific membrane antigen did not correlate with pathological stage of prostate cancer.