Prognostic impact of post-prostatectomy PSA slope determined with a novel, nucleic acid detection immunoassay (NADiA ProsVue) for total PSA.
Moul JW., Semmes OJ., Vessella R., McDermed JE., Lilja H.
47 Background: ProsVue is an investigational PSA immunoassay whose reporter antibody is labeled with a synthetic DNA sequence. RT-PCR detects the DNA signal indicating the PSA concentration. Pilot studies showed ProsVue slope (least-squares linear slope of 3 post-RP PSA values) to correctly classify prostate cancer (PC) patients (pts) with no evidence of disease (NED) and clinical progression (CP) using a 2 pg/mL/month (mo) decision threshold. In a retrospective, multicenter clinical trial, we evaluated potential prognostic value of ProsVue slope at this decision threshold. METHODS: 392 PC pts having RPs 11/1991 to 8/2001 were studied. Eligibility required first post-RP PSA <100 pg/mL, full pathologic and radiographic data and 3 frozen serum samples drawn 6 weeks to 19.4 mo post-RP. Adjuvant radiotherapy (RT) and/or hormone therapy (HT) was not allowed. CP was documented by positive imaging, biopsy results or PC-related death. Efficacy of ProsVue slope as a prognostic test for NED/CP at a 2 pg/mL/month decision threshold was determined and also examined with regards to Gleason score (GS), pre-RP PSA and final pathology stage. RESULTS: Median (range) of pre-RP PSA was 6.3 (0-60.6) ng/mL and post-RP GS was 7.0 (4-10). 73 pts received neoadjuvant HT. Pathologic stage was pT0-2 (228), pT3 (147), pT4 (17); 116 pts had positive margins and 8 had positive lymph nodes. The 3 PSA values were drawn after a median of 4.9, 8.6 and 12.8 mo and showed median values of 10.7, 23.0 and 50.7 pg/mL, respectively. Calculated sensitivity, specificity, PPV and NPV for a 2 pg/mL/mo ProsVue slope were 75.0%, 96.6%, 81.4% and 95.2%, respectively. Median follow-up (f/u) was 10.5 years. There were 40 deaths (14 from PC). CONCLUSIONS: ProsVue provides information previously unknown in post-RP pts. ProsVue slope ≤2pg/mL/mo in the first year is highly associated with NED over long-term f/u. Potential clinical utility may include predicting pts not requiring long-term oncologic f/u and predicting a need for post-RP adjuvant RT. ProsVue slope may become a new paradigm for identifying those patients at reduced risk for recurrence of prostate cancer post-RP. Further studies are planned to address these questions. [Table: see text].