Canadian Pharmacy: Prostate-specific antigen

2015-07-29 / disease / 0 Comments

Unfractionated

Serum PSA measurement has revolutionized the detection of prostate cancer in the general population, and has produced a significant downward stage migration in the epidemiology of the disease. Benign disease, however, can also elevate the PSA: 28% of men with histologically proven BPH have a PSA over 4.0 ng/ml. In fact, the similarity in the epidemiology of the two conditions and their frequent coexistence constitutes one of the greatest impediments to the specificity of PSA testing for prostate cancer. BPH confounds PSA screening via a number of mechanisms. Unfractionated PSA correlates with prostate volume in a log-linear fashion; the relationship between the two variables becomes stronger with advancing patient age30. Conversely, finasteride, a common treatment for BPH, reduces PSA levels by roughly 50% after 6 months of therapy. While ‘correcting’ the PSA by doubling the reported value has been suggested for men taking finasteride, the actual percentage change may range from −81 to +20%. (Of note, alpha blockers and saw palmetto33, two other commonly used medications among BPH patients, do not significantly affect PSA.) Urinary retention, a frequent complication of BPH, can produce further increases in PSA of up to six-fold; levels have been reported to drop by 50% within 48 h of relief of obstruction.
In the general screening population, unfractionated PSA has a sensitivity of 67.5–80% and specificity of 60–70%, given a threshold for normal of 4.0 ng/ml35. PSA levels above 4.0 ng/ml are found in approximately 8–13% of men with neither BPH nor prostate cancer, and in more than 30% of men with BPH but not prostate cancer.

Among patients

with PSA 4–10 ng/ml, only about 25% with normal DRE results in fact have prostate cancer. The specificity of PSA in men with PSA 4–10 ng/ml falls to as low as 50%, owing to the broad overlap in PSA levels between men with BPH and prostate cancer in this range. In particular, Monda and colleagues found that PSA could not reliably differentiate BPH from T1a prostate cancer. In the series of BPH patients of Lepor and colleagues, PSA and DRE screening had a sensitivity of 86.7 and 80.0%, specificity of 80.9 and 86.3%, and PPV of 25 and 30%, respectively.
Digital rectal examination still detects up to 25% of prostate cancers that present in the setting of PSA less than 4 ng/ml, and therefore remains an essential part of prostate cancer screening, in the context of BPH or otherwise. Certainly, however, the PPV of DRE goes up with increasing PSA, ranging from 4 to 11% in men with PSA 0–2.9 ng/ml, and from 33 to 83% in those with PSA 3–9.9 ng/ml. Owing to the inability of unfractionated PSA to discriminate reliably between prostate cancer and BPH, a number of modifications and refinements have been proposed over the past decade, and currently stand at various levels of development. This review focuses on those that have been specifically evaluated in the context of differentiating BPH from prostate cancer in a screening population. It should be noted that none of these tests has yet shown consistent enough benefit to win the endorsement of the American Urological Association’s PSA best practices policy

PSA density

Prostate-specific antigen density (PSAD), described by Benson and colleagues in 1992, refers to the serum PSA divided by the prostate volume as calculated from transrectal ultrasound (TRUS) measurements using the prolate ellipsoid formula. This measurement attempts to improve the specificity of PSA testing for prostate cancer by accounting for the PSA changes produced by BPH. The ratio of PSA-producing epithelium to stroma is relatively preserved in BPH, and the serum PSA is thought to rise at a relatively constant rate of roughly 0.3 ng/ml per gram of hyperplastic tissue. In prostate cancer, however, the concentration of epithelial cells in a given volume of prostate tissue increases; moreover, the ‘leaky’ nature of the neoplastic endothelium allows a greater amount of PSA to enter the bloodstream, producing an increase in PSA as great as 3.5 ng/ml per gram of tumor. A study by Furuya and co-workers of patients undergoing open or transurethral prostatectomy for BPH found that each gram of BPH tissue removed reduced the PSA by an average of 0.18 ng/ml, and that after surgical treatment for BPH, PSA should return to normal levels in a patient without concurrent prostate cancer.
Benson and colleagues analyzed a cohort of 595 men with PSA levels between 4.1 and 10 ng/ml, and found mean PSAD values of 0.297 and 0.208 among men with and without prostate cancer, respectively (p<0.0001). They constructed a PSAD-based nomogram that calculated prostate cancer risk estimates ranging from 3 to 100%41. An updated analysis of 733 patients found mean PSAD values of 0.285 and 0.199 among biopsy-positive and – negative patients, and 0.165 among those with no indication for biopsy. They concluded that a PSAD>0.15 corresponded to an 18% positive biopsy probability among those with an abnormal DRE or TRUS, and a 6% probability among those with no abnormality.
A Japanese study of 63 men with histologically confirmed BPH and 234 men with prostate cancer found PSA levels between 4 and 10 ng/ml in 36 and 25 men, respectively.
The BPH patients had a mean prostate volume (determined by transabdominal ultrasound) and PSA of 17.1±8.2 ml and 6.42±1.82 ng/ml, respectively, for a mean PSAD of 0.218±0.085. The prostate cancer patients, by contrast, had a mean volume and PSA of 33.4±14.1 ml and 7.8±2.15 ng/ml, for a PSAD of 0.572±0.363. The authors found that PSAD measurement yielded>90% sensitivity and 56% specificity for distinguishing BPH from prostate cancer. Other PSAD studies have been reviewed previously by Beduschi and Oesterling.
Drawbacks to PSAD measurement include the need to perform TRUS to obtain the measurement, the operator-dependent nature of TRUS estimation of prostate volume and the variable stromal-epithelial ratio among individuals. In a large, prospective, multicenter study of nearly 5000 men screened for prostate cancer with PSA and PSAD, Catalona and associates found that TRUS-measured volume correlated poorly (r=0.61) with pathological prostate weight, and that employing a PSAD cut-off of 0.15 to guide biopsy decisions would miss 47% of cancers among patients with PSA levels between 4 and 10 ng/ml.

Transition zone-adjusted PSA

BPH is almost exclusively restricted to the transition zone of the prostate, whereas prostate cancer most often affects the peripheral zone. Kalish and colleagues therefore proposed that adjusting the PSA for the transition zone volume rather than the total prostate volume would better reflect the relative contribution to PSA from BPH tissue, particularly in the PSA ‘gray zone’ of 4–10 ng/ml46. The transition zone-adjusted PSA (PSAT) does in fact appear to distinguish BPH from prostate cancer more accurately than PSAD. Kurita and associates performed TRUS-guided biopsies on 164 consecutive patients with elevated PSA and/or abnormal DRE. They found cancer in 27.2%, and calculated ROC areas of 0.667 for PSA, 0.663 for PSAD (not significant) and 0.826 for PSAT (p<0.01).
Zlotta and coworkers evaluated PSA density parameters for histologically proven BPH (n=74) and prostate cancer (n=88) patients, finding mean PSAD values of 0.12±0.07 and 0.22±0.12, respectively, and PSAT values of 0.21±0.13 and 1.02±0.70. They estimated that using PSAT with a threshold of 0.35 would miss 10% of cancers, versus 34% using PSAD with a threshold of 0.15. Again using a threshold of 0.35, they calculated sensitivity and specificity for PSAT of 90 and 93% for patients with total PSA 0.25–10 ng/ml, and 94 and 89% for patients with PSA 4–10 ng/ml48. In a separate study, Zlotta and co-workers also found that assessment of prostate volume by TRUS was more accurate for the transition zone (correlation with pathological weight r=0.95, variability −17 to +18%) than for the whole prostate (r=0.78, variability −21 to +30%), further supporting the concept of PSAT rather than PSAD measurement.

Article by Canadian Pharmacy Mall – www.healthandcaremall.net

Read More