Cancerline

Search

Sign in

email address
password
Remember Me

Information for

This is an AstraZeneca International website for healthcare professionals. The site is not intended for the following audiences who should use the links below:

ProstateLine

BreastCancerSource

23 July 2008

Percentage of [-2]proenzyme PSA predicts prostate cancer diagnosis

Related news
MedWire News: Prostate cancer detection is better predicted by the percentage of [-2]proenzyme prostate-specific antigen (PSA) than by the percentage of free PSA, particularly in the 2-10 ng/ml range, conclude US scientists.

In both prostate tissue and serum there are disease-specific isoforms of free PSA that may help to overcome some of the current limitations of measuring PSA in the early detection of prostate cancer, explain Lori Sokoll, from Johns Hopkins Medical Institutions in Baltimore, Maryland, and colleagues.

The team collected serum samples before biopsy from 123 men with no prior biopsy or prostate cancer history, of whom 51% had prostate cancer and the remainder were noncancer controls. Levels of PSA, free PSA, [-2]proenzyme PSA, benign PSA, and testosterone were measured.

The average age of the overall group was 62.2 years. Among cancer patients, 52% had a biopsy Gleason score of 6, 38% a score of 7, and 10% a score of 8 or 9. There were no significant difference between patients and controls in terms of age or total PSA concentrations, at 62.6 years versus 61.7 years, and 6.94 ng/ml versus 6.80 ng/ml, respectively.

Cancer patients had significantly lower percentage free PSA than controls, and had significantly higher [-2]proenzyme PSA and percentage [-2]proenzyme PSA levels, the team notes in the Journal of Urology.

Receiver operating characteristics analysis revealed that the greatest area under the curve (AUC) was seen with percentage [-2]proenzyme PSA, at 0.69 and a sensitivity and specificity of 60% and 70%, respectively, followed by percentage free PSA, at an AUC of 0.61.

The AUC for a model combining PSA, benign PSA, percentage free PSA, percentage [-2]proenzyme PSA, [-2]proenzyme PSA/benign PSA, and testosterone was 0.73.

At a PSA range of 2-10 ng/ml, percentage [-2]proenzyme PSA and the model had the largest AUC, at 0.73, while that for percentage free PSA was 0.53. At 90% sensitivity, the specificities for percentage [-2]proenzyme PSA, the model, and percentage free PSA were 41%, 32%, and 18%, respectively, compared with 31%, 26%, and 16%, respectively, at a sensitivity of 95%.

The team writes: "These findings provide a rationale for broader validation studies to determine whether percentage [-2]proenzyme PSA alone supplants other multiple molecular PSA assays for improving the accuracy of prostate cancer screening."



J Urol 2008; 180: 539-543

http://dx.doi.org/10.1016/j.juro.2008.04.015