PROSPECTIVE ANALYSIS OF PROSTATE-SPECIFIC MARKERS IN PELVIC LYMPH-NODES OF PATIENTS WITH HIGH-RISK PROSTATE-CANCER

Background: Pathologic evidence of pelvic lymph node involvement is ob tained in 12%-20% of patients with localized prostate cancer that exhi bits high-risk features (defined on the basis of tumor size, serum pro state-specific antigen [PSA] level, or Gleason score), The rate of sys temic failure (i.e., relapse) in patients with this type of prostate c ancer and no pathologic evidence of regional lymph node involvement is 55%-92% within 5 years of definitive local therapy, Since reverse tra nscription-polymerase chain reaction (RT-PCR) methods are likely to be more sensitive than routine pathologic examination in detecting metas tatic tumor cells, we compared the ability of the two approaches to de tect prostate cells in the pelvic lymph nodes of patients with localiz ed, high-risk disease, Methods: Fifty-eight lymph node specimens isola ted from 33 patients before definitive local therapy were examined, Ex pression of PSA and prostate-specific membrane antigen (PSM) messenger RNAs in the specimens was assessed by means of nested RT-PCR, Results : Pathologic examination identified tumor cells in the lymph nodes of four (12%) of the 33 patients, and PSA and/or PSM expression was posit ive in specimens from 27 (82%) of the patients (two-sided P<.0001), Th e four patients with positive pathologic findings also had positive RT -PCR results, Among the 29 patients with no pathologic evidence of lym ph node involvement, 23 (79%) tested positive by means of RT-PCR, In t hese 23 patients, PSM expression was detected more frequently than PSA expression; however, in two patients, only PSA expression was detecte d, Conclusions: Expression of prostate-specific markers in the pelvic lymph nodes of patients with localized, high-risk prostate cancer may indicate the presence of metastatic tumor cells, Such cells may be res ponsible for the high rate of systemic failure seen in these patients, Additional studies are required to determine the prognostic relevance of our findings.