Prostate
Acid phosphatase-biochemical. Prostatic carcinoma often may be detected chemically because normal prostatic tissue is rich in the enzyme acid phosphatase, and adenocarcinomas arising from the prostate often retain the ability to produce this enzyme. Acid phosphatase is actually a group of closely related enzymes that share certain biochemical characteristics. Members of this group are found in several tissues (e.g., kidney, bone, prostate, platelets, and spleen). For almost 30 years, standard biochemical methods were the only assay systems available for acid phosphatase. Since enzyme quantity cannot be measured directly by these systems, it is estimated indirectly by the amount of change the enzyme produces in a measured quantity of substrate (a substance that can be changed by action of the enzyme). Several different assay systems have evolved, each of which differs somewhat in sensitivity and specificity for acid phosphatase of prostate origin. This accounts for some (although not all) of the conflicting reports in the literature regarding ability of biochemical enzyme system assays for acid phosphatase to detect prostate carcinoma.
Studies have shown that about 5%–10% (literature range, 5%–15%) of patients with prostate adenocarcinoma confined to the prostate have elevated serum acid phosphatase levels. Elevation occurs in about 20%–25% (10%–56%) of patients with extension of prostate tumor outside the prostate capsule without distant metastases, and in about 75%–80% (47%–92%) of those with bone metastases.
False positive results. Since acid phosphatase measurement is usually undertaken to detect prostate carcinoma, elevation in other conditions is usually considered to be a false positive result. Prostatic infarcts can temporarily elevate serum acid phosphatase levels. Acid phosphatase levels are reported to be elevated in about 5%–10% (0%–19%) of patients with benign prostatic hypertrophy without any evidence of carcinoma. Elevations have been reported in patients with several other diseases; patients with the most substantial percentage of abnormality include those with nonprostatic metastatic carcinomas to bone, certain metabolic bone diseases (Paget’s disease, primary hyperparathyroidism), Gaucher’s disease, and certain platelet disorders (thrombocytosis, disorders of platelet destruction). Alkaline and acid phosphatase are similar enzymes that differ predominantly in the pH at which they work best. Therefore, any condition that produces greatly increased alkaline phosphatase values may induce some elevation of acid phosphatase values because the ordinarily negligible action of the alkaline phosphatase at lower pH is magnified by a greatly increased quantity of enzyme.
There is controversy in the literature whether a rectal examination will temporarily (up to 24 hours) increase serum acid phosphatase levels. More investigators have reported no change than those who did find significant elevation, but the size of the two groups is not far apart. Therefore, the test probably should be repeated if elevated levels occur after a rectal examination. There is some evidence that elevated serum acid phosphatase values can fluctuate 25%–50% or even more during a 48-hour period.
Various methods have been used to decrease false positive results, some of which are more specific than others for prostatic acid phosphatase. However, none is completely specific. For example, it was found that various substances could inhibit either prostate or nonprostate acid phosphatase; the most widely used of these substances is L-tartrate, which inhibits prostatic acid phosphatase. Thus, acid phosphatase elevation that persists after patient serum is incubated with L-tartrate (“tartrate resistant”) is likely to be of nonprostatic origin. Unfortunately, although tartrate inhibition does increase specificity, it does not produce complete specificity. Fortunately, in most situations where prostatic carcinoma is suspected, the majority of the nonprostatic conditions that produce elevated acid phosphatase levels would be unlikely (except for occult nonprostatic tumor involving bone).
False negative results. Biochemical enzyme assays for acid phosphatase are heat and pH sensitive. Serum left at room temperature after exposure to air may exhibit significantly decreased activity after as little as 1 hour. The use of a preservative with prompt refrigeration is strongly recommended. Immunoassays can tolerate exposure of the sample to room temperature for 4-5 days.
Acid phosphatase immunoassay. In the last half of the 1970s, methods were found to obtain antibodies against prostatic acid phosphatase. This enabled the development of immunoassays (radioimmunoassay [RIA], enzyme-linked immunosorbent assay [ELISA], and counterimmunoelectrophoresis) for determining prostatic acid phosphatase levels. In general, the immunoassays have greater specificity for prostatic acid phosphatase than biochemical enzyme procedures have. The immunoassays also detect more patients with prostate carcinoma. Unfortunately, the tests uncover only about 10%–20% more patients with prostate carcinoma than the biochemical tests at comparable stages of tumor spread (Table 33-7). In addition, the immunoassays are not completely specific; for example, they become elevated in some cases of nonprostate carcinoma that is metastatic to bone. Many of the elevations of test kits found that a certain number of patients with benign prostatic hypertrophy, without evidence of tumor, had elevated results. Moreover, there has been a very disturbing variation in sensitivity and specificity of the immunoassay kits. Another drawback is that immunoassays are 3 to 5 times more costly to perform than the biochemical enzyme methods.
Table 33-7 Comparison of biochemical and immunologic assays for acid phosphatase*
Prostate-specific antigen (PSA). Prostate-specific antigen (PSA) is a glycoprotein enzyme that was isolated from prostate gland epithelial cells in 1979. There are claims that this enzyme is specific for prostate origin, although a few isolated instances of elevation in nonprostate tumors have been reported. Three manufacturers now have kits commercially available. Reports thus far agree that PSA detects more patients with prostate carcinoma than acid phosphatase by immunoassay or biochemical methods (Table 33-7). Prostate carcinoma in stage A is detected in about 55% of cases. Overall prostate carcinoma detection is about 80%–90% (range, 75%–96%). Unfortunately, detection rate for patients with benign prostatic hypertrophy is about 50%–60% (range, 10%–83%). Since prostatic glandular hypertrophy is common in the same age group as prostate carcinoma, there has been much dispute whether PSA should be used as a screening test for prostate carcinoma. Although screening with PSA detects a significant number of patients with early stage cancer, there has not yet been proof that a significantly greater number of patients will be cured. Also, about 20% (range, 16%–33%) of patients with cancer have PSA values within the usual PSA population reference range of 0.1-4.0 ng/ml (µg/L). Several ways have been suggested to improve PSA usefulness. It has been shown that PSA levels correlate reasonably well on the average with prostate weight due to benign prostate glandular hyperplasia (BPH) and prostate cancer produces about 10 times the amount of PSA on a tissue volume basis than does BPH. Studies have resulted in several formulas that at least partially correct PSA levels for effect of BPH. This is the most well established of these formulas at present: Predicted PSA level (PSA “serum density”, in ng/ml or µg/L = 0.12 Ч gland volume (in cubic centimeters [cc]) by transrectal ultrasound [TRUS]; when TRUS gland volume = prostate height Ч width Ч length Ч 0.523). A PSA serum density greater than the predicted value suggests increased possibility of carcinoma. Another parameter is the height of the PSA level. Values within reference range suggest relatively low probability of cancer (although, as noted previously, some 20% of cancers have normal PSA); values over 10 ng/ml suggest a relatively high probability of carcinoma; and values between 4-10 ng/ml suggest an intermediate possibility (this being the zone in which there is greatest overlap with elevation from BPH). Another parameter, especially for values below 10 ng/ml, is the trend of PSA values repeated in 3- or 6-month intervals. A definite upward trend increases the likelihood of cancer. Since this may also occur due to BPH, there have been efforts to find a trend formula that differentiates the two entities. However, none of the articles proposing a trend formula had sufficient number of patients to make the formula statistically valid and none have had adequate independent evaluation.
Several investigators have proposed age-adjusted PSA reference ranges, although their values differ somewhat. A Mayo Clinic study proposes a PSA upper limit of 2.5 ng/ml (PSA density [PSAD] upper limit 0.08) for ages 40–49 years, PSA 3.5 (PSAD 0.10) for ages 50–59, PSA 4.5 (PSAD 0.11) for ages 60–69, and PSA 6.5 (PSAD 0.13) for ages 70–79.
These diagnostic problems have occurred because prostate carcinoma in general is a slow-growing tumor that most often clinically becomes evident in older persons; and although there are tumor subsets that are aggressive, the majority of patients die from other causes than prostate cancer. In addition, the three current modes of therapy (radical prostatectomy, radiation therapy, and hormonal therapy or androgen deprivation) all have a significant amount of both somewhat unpleasant morbidity and also of treatment failure. Nevertheless, since it has been calculated that about 10% of U.S. males will develop clinically evident prostate carcinoma and about 3% of U.S. males will die from it, screening will continue in the hope of detecting occult disease in a curable stage.
The most common prostate screening recommendation is to obtain both PSA and direct rectal examination (DRE). Although PSA misses about 20%–30% and DRE misses about 50% (range, 14%–64%) of prostate cancer, the two together detect an additional 15%–20% or more over results from either one alone. If DRE is positive, or if DRE is negative but clinical suspicion is high, transrectal ultrasound is advised to help find a target for biopsy and to estimate prostate volume for the PSA serum density value if no TRUS area suggestive of cancer is seen. Abnormal PSA serum density could be a reason for biopsies without a clearly defined TRUS target. Incidence of prostate carcinoma is increased in African Americans (about twice the risk of European descendants). Previous vasectomy may create some increased risk.
There are some points regarding PSA that are worth mentioning. There is no diurnal variation. No preservatives are needed for the specimens. There has been some controversy whether DRE will temporarily elevate PSA; most investigators found 2% or fewer patients above 4.0 after DRE. A post-DRE specimen could be drawn less than 1 hour after the examination to minimize any examination effect. Prostate biopsy or TUR elevate PSA to varying degree in most persons. This effect varies in duration between persons, the amount of prostate injury, or the height of the PSA response, but is usually back to baseline or below by 6 weeks (in one study, 63% of persons took less than 2 weeks and 15% took over 4 weeks). The serum half-life of PSA is about 2 days. Finasteride therapy of BPH in dosage of 5 mg/day decreases the PSA to 50% of pretherapy baseline value after 6-12 months of therapy. Patient compliance can be monitored by serum dihydrotestosterone assay, which should be suppressed by more than 60% from pretherapy baseline values. One report recommends that serum PSA should be checked 6 months after beginning therapy, and if the value (in a compliant patient) is not 50% of baseline, the patient should have a workup for possible prostate carcinoma.
Serum alkaline phosphatase. Stage D prostate carcinoma metastasizes to bone in about 70% of patients (range, 33%–92%). More than 90% of these metastases have an osteoblastic component. Serum alkaline phosphatase levels are elevated in 70%–90% of patients with bone metastoses.