Prostate cancer risk assessment tools in an unscreened population.

Pas encore traduit Pas encore traduit
Catégorie Primary study
JournalWorld journal of urology
Year 2015

Cet article est inclus dans 2 Systematic reviews Systematic reviews (2 references)

This article is part of the following publication threads:
  • ERSPC [European Randomized Study of Screening for Prostate Cancer] (44 documents)
Loading references information

OBJECTIVE:

To compare the prostate cancer prevention trial risk calculator (PCPT-RC) and European randomized study of screening for prostate cancer risk calculator (ERSPC-RC) in a unique unscreened population from the West of Ireland.

PATIENTS AND METHODS:

Data was prospectively recorded for all 556 consecutive men who underwent prostate biopsy at our institution as part of the Rapid Access Prostate Assessment Clinic program in Ireland. The estimated probabilities of detecting prostate cancer and high-grade disease were calculated using the PCPT and ERSPC risk calculators. For each calculator the discriminative ability, calibration and clinical utility was assessed.

RESULTS:

Prostate cancer was detected in 49% and high-grade prostate cancer in 34% of men. Receiver operating characteristic curve analysis demonstrated that the PCPT-RCs outperformed the ERSPC-RCs for the prediction of prostate cancer areas underneath the ROC curve (AUC 0.628 vs. 0.588, p = 0.0034) and for the prediction of high-grade prostate cancer (AUC 0.792 vs. 0.690, p = 0.0029). Both risk calculators generally over-predicted the risk of prostate cancer and high-grade disease across a wide range of predicted probabilities. Decision curve analysis suggested greater net benefit using the PCPT-RCs in this population.

CONCLUSIONS:

Multivariable nomograms can further aid patient counselling for early prostate cancer detection. In unscreened men from Western Ireland, the PCPT-RCs provided better discrimination for overall prostate cancer and high-grade disease compared to the ERSPC-RC. However, both tools overpredicted the risk of cancer detection on biopsy, and it is possible that a different set of predictive variables may be more useful in this population.
Epistemonikos ID: 8509046475facc021930300b2bd7461bb14e1c5b
First added on: Apr 22, 2019