Prostate Cancer Gene 3 (PCA3): Development And Internal Validation Of A Novel Biopsy Nomogram – In this paper, we investigated 809 prostate cancer patients subjected to >10 cores at initial or repeat prostate biopsy from two prospective, multi-center studies from Europe and North America It has been demonstrated that the urinary marker Prostate CAncer gene 3 (PCA3) represents a novel prostate cancer (PCa) detection marker capable of increasing accuracy of multivariable biopsy nomograms.[1] The paper reports the first PCA3-based nomograms which accurately identify individuals at risk of harboring PCa (AUC=0.73). If a PCA3 score in combination with established risk factors is available, this novel tool assists clinicians in deciding whether further prostatic evaluation is necessary.
Despite these promising results it must be emphasized that novel markers such as PCA3 do not replace established risk factors such as PSA and its sub-forms, digital rectal findings and/or prostate volume. However, this combination resulted in significant improvements in accuracy (between 2% to 5%) of biopsy outcome prediction. This increment of +5% related to one single marker (PCA3) is remarkable since Shariat et al. recently added 7 novel diagnostic markers to a multivariable model predicting biochemical recurrence after radical prostatectomy which were related to a gain in accuracy by “only” +15% (AUC=0.72 to AUC=0.87). [4]

Although this study involves the largest PCA 3 biopsy-verified patient cohort to date, it needs to be acknowledged that our findings are still based on a relatively small sample size. Specifically, PCA3 score cut-off analyses need further investigation in different biopsy settings (e.g. initial vs. repeat) [5, 6] or in further sub- stratification (e.g. PSA cut-offs 0-2.5 ng/ml vs. 2.6-4 ng/ml) in larger scale studies, respectively. [7] In fact, it may be argued that, similar to PSA [8], a PCA3 score is better displayed as a continuously increasing risk according to increasing scores and that cut-off values may not be indicated. For example, as suggested by the current paper, the PCA3 score used as a continuous risk variable demonstrated the highest univariable accuracy (AUC=0.68) outperforming other previously published PCA3 cut-offs (AUC=0.62- 0.63) or PSA (AUC=0.53). However, this result could not be confirmed in multivariable analysis.

In conclusion, our results are clearly encouraging – demonstrating that PCA3 is one of the novel markers alleviating PSA’s dilemma of low specificity. But larger scale studies are also clearly warranted to replicate our findings and to externally validate the first PCA3 nomogram.