Combining clinical factors, history with PSA could avoid unnecessary biopsies
MedWire News: Combining the prostate-specific antigen (PSA) test with other factors such as prostate size, body mass index (BMI), and family history could improve prostate cancer screening, potentially avoiding a quarter of unnecessary biopsies, researchers say.
They found that considering these other factors along with testing PSA levels substantially improved upon the ability of current PSA-based screening methods to predict when patients need aggressive treatment - while avoiding detection of indolent cancers best left undiagnosed.
The findings come just after a US Preventive Services Task Force expert panel concluded that the benefits of PSA-based prostate cancer screening do not outweigh the harms.
Lead author of the present study, Martin Sanda (Harvard Medical School, Boston, Massachusetts, USA), commented in a press statement that the Task Force "threw the baby out with the bathwater by their blanket recommendation against prostate cancer screening."
Sanda and team developed their model for predicting histologically aggressive prostate cancers in 635 men undergoing a first prostate biopsy who were enrolled in the multicenter National Cancer Institute Early Detection Research Network. The researchers then used data for 3833 participants in the control arm of the Prostate Cancer Prevention Trial to validate the model.
As reported in the journal Cancer, 361 (57%) of the derivation cohort had no cancer, while 88 (14%) had indolent cancer and 186 (29%) had histologically aggressive cancer.
Multivariate analysis showed that age, BMI, family history of prostate cancer, abnormal digital rectal examination, and PSA density (PSAD) significantly predicted aggressive cancer. Receiver operating characteristic analysis demonstrated that this model predicted the risk with an area under the curve (AUC) of 0.81.
This compared with AUCs for PSA only or PSAD only of 0.71 and 0.75, respectively.
When compared with PSA alone at a sensitivity of 90%, the multivariate model improved specificity from 32% to 42%. And, the authors note, use of the model would have avoided 24.6% of biopsies among 236 patients enrolled in the Early Detection Research Network from January 2008 through April 2009.
Sanda and colleagues further report that the model "performed well" in predicting aggressive cancer in the validation cohort, at an AUC of 0.78.
They say that, although predicted and observed probabilities of prostate cancer diverged at higher risk for aggressive cancer, overpredicting men at such risk, the model was robust at risk levels below 10%, "confirming the predictive accuracy of this model in identifying men who can forego prostate biopsy while retaining 90% sensitivity in detection of aggressive cancer."
In a separate study led by Sanda, published in the journal Urologic Oncology, further improvement in PSA-based screening was demonstrated by including a urine test to detect presence of the genetic biomarkers TMPRSS2:ERG and PCA3.
However, Sanda cautioned that "urine testing for prostate cancer is in its infancy," adding that research funded by the National Institutes of Health is ongoing to explore this further.
By Caroline Price