Evaluation of prostate-specific antigen in untreated prostatic carcinoma

A study was performed on 290 men to compare the level of serum prostate-specific antigen (PSA) in controls, patients with benign prostatic hyperplasia (BPH) and patients with prostatic cancer. The upper limit of normal was 5.0 micrograms/l as determined in 110 elderly hospitalized males (mean age 62 years) without urological complaints. Of the 106 patients with BPH, 33% had raised values above 5.0 micrograms/l. Values above 10 micrograms/l were found in 18 BPH patients. A positive correlation was found between prostate volume (grams of tissue removed during transurethral resection) and preoperative PSA levels (r = 0.55, n = 106, p less than 0.001). PSA levels above 10 micrograms/l were found in 4% of BPH patients with a prostate volume of less than 20 g (n = 54), in contrast with 45% of patients with a prostate volume above 40 g (n = 20). The sensitivity of this PSA assay (cutoff level 10 micrograms/l) as established in 74 prostate carcinoma patients was 31% for category T0 (n = 13), 56% for category T1-2 (n = 16), 75% for category T3-4 (n = 20) and 100% for category M1 or N1-4 (n = 25). In an earlier study prostatic acid phosphatase (PAP) was measured in these same samples. PSA appeared to be much more sensitive than PAP. Seventeen of the 74 prostatic carcinoma patients (23%) had normal PAP levels but their PSA values were raised above 10 micrograms/l, while in only 2 patients an increased PAP level was combined with a normal PSA.(ABSTRACT TRUNCATED AT 250 WORDS)     

Prostate tumour markers and differentiation grade in prostatic cancer.

Serum acid phosphatase activity, prostate specific phosphatase and prostate specific antigen were measured in 100 patients with prostatic cancer. The patients were divided according to the differentiation grade into 3 groups: G1 (well), G2 (moderately) and G3 (poorly differentiated) carcinoma. Bone metastases were identified by scintigraphy. Among the 76 M0 patients the mean levels of all 3 markers were slightly higher in patients with moderately differentiated prostatic carcinoma. Among the 24 M1 patients the primary tumour was either G2 (18 patients) or G3 (6 patients); none had G1 lesions. Significantly higher serum ACP and PAP levels were found in patients with G2 tumours than in those with G3 lesions. It was concluded that the histological differentiation grade of prostatic carcinoma did affect serum levels of prostatic tumour markers; the tendency towards higher levels in the G2 group was noticeable in both non-metastatic and metastatic cases despite the limited number of patients in the latter category. In clinical practice this information may be an important additional tool in staging prostatic cancer.     

Tumour markers in prostatic carcinoma. A comparison of prostate-specific antigen with acid phosphatase

A study was performed on 130 men to compare the level of serum prostate-specific antigen (PSA) in controls, patients with benign prostatic hyperplasia (BPH) and patients with prostatic carcinoma. The results showed that all 30 normal controls below 40 years of age had values less than 10 ng/ml. Of the 40 patients with BPH, all aged over 40 years, 13 (32.5%) had raised levels above 10 ng/ml. In the 60 patients with prostatic carcinoma, all over 40 years, 24 had localised disease (MO) and 36 had metastatic spread (M1), as judged by isotope bone scan. In patients with MO disease, 16 (66.6%) had raised PSA levels compared with 34 (94.5%) of those with M1 disease. The corresponding figures for raised prostatic acid phosphatase (PAP) values were 4% in the MO group and 52.7% in the M1 group. PSA levels reflected neither the histological grade nor the local stage of the tumour and were of no value in estimating tumour burden. PSA was found to be a valuable index in the management of prostatic cancer because of this sensitivity. Stable disease not requiring hormonal manipulation was reflected by unchanging levels of PSA, whereas progressive disease requiring hormonal therapy was reflected by an alteration in the PSA levels corresponding to the patients' response. The same group of progressive disease patients showed only a 50% rise in serum PAP levels, confirming the greater sensitivity of PSA as a measure of prostate cancer. PSA measurements should be included in any further trials on prostatic carcinoma and should be regarded as a standard marker for evaluating response to therapy.     

Comparison of prostate secretory protein with prostate specific antigen and prostatic acid phosphatase as a serum biomarker for diagnosis and monitoring patients with prostate carcinoma

Serum prostate secretory protein (PSP) levels were measured in 49 patients with benign prostatic hyperplasia (BPH), 144 patients with various stages of prostatic carcinoma (CaP), and 82 CaP patients who were followed serially. PSP values were compared with serum levels of prostate specific antigen (PSA) and prostatic acid phosphatase (PAP). In the BPH group, PSP was elevated (> 10 ng/ml) in 41% of patients, whereas PSA (> 4 ng/ml) and PAP (> 3.3 ng/ml) were elevated in 39% and 23% of the cases, respectively. PSP levels were elevated in 48% of the CaP pretreatment specimens, compared to 79% for PSA and 40% for PAP. PSP levels in cancer patients who had intracapsular disease were about two to three times higher than those observed for PAP. PSP was found to be the only marker elevated in eight (6%) pretreatment CaP patient serum specimens, while PAP was never found to be elevated when PSA was normal. PSP serum concentrations correlated with the clinical course of the disease in 79% of patients, compared with 90% for PSA and 66% for PAP. In certain patients, monitored over time, disease correlation was reflected in serum values with only a single biomarker, i.e., 1% with PAP, 8% with PSP, and 10% with PSA. This study has shown that PSP is a less sensitive serum biomarker than PSA, but more sensitive than PAP for detection and monitoring the early stages of prostate cancer. This suggests that PSP as a biomarker may be a useful adjunct for the management of a subpopulation of low-stage and -grade CaP. © 1993 Wilcy-Liss, Inc.     

Gamma-seminoprotein--a new tumour marker in prostatic cancer? Results of a pilot study.

Serum levels of gamma-seminoprotein (GSM), prostate specific antigen (PSA) and prostate specific acid phosphatase (PAP) were examined, using enzyme immunoassay, in 250 patients with prostatic disease. The results indicated that the highest specificity was obtained with GSM (94%) and the lowest with PSA (60%). In contrast, the highest sensitivity in newly detected carcinomas (n = 41) was obtained with PSA (71%), whereas that of GSM (51%) was comparable to that of PAP (44%). Of 41 patients with newly detected prostatic cancer, 35 (85%) showed a significant increase in at least 1 of the tumour markers. Five of 6 patients whose markers were within normal limits had incidental carcinomas. During follow-up, PSA was raised in 88%, GSM in 66% and PAP in 55% within 12 months prior to clinical progression. Our results suggest that the determination of GSM may be of value in the serological detection and monitoring of prostatic cancer. These findings must be confirmed by further studies with larger numbers of patients and longer follow-up.     

Daily variability in human serum prostate-specific antigen and prostatic acid phosphatase: a comparative evaluation

The daily variation of serum levels of prostate-specific antigen (PSA) and prostatic acid phosphatase (PAP) was investigated simultaneously in 10 patients with osseous metastatic prostatic cancer, 10 patients with benign prostatic hyperplasia, and 10 volunteers without prostatic disease. Duplicate serum samples were obtained from all patients on the same day at 8 AM, 12 PM, 4 PM, and 8 PM. Statistical analysis (two-factor analysis of variance comparing time period to disease group) of the mean PSA and PAP levels at the four sampling times on all patient groups demonstrated no evidence of circadian rhythmic variation or any other distinct pattern for the observed sample times. Overall, the variability in PSA levels was significantly less than that observed for PAP. There was no significant difference in mean percent variation between patient groups (cancer, benign, and normal prostate glands) for both the PSA and PAP assays. Our data reveal that serum PSA measurements fluctuate unpredictably over the course of a day in patients with and without prostatic disease, but to a lesser extent than that seen for serum PAP values. These findings illustrate the potential inaccuracy of single determinations of serum PAP or PSA levels for monitoring disease recurrence and treatment response in patients with prostate cancer.     

Prediction of pelvic lymph node metastases by a prostate-specific antigen and prostatic acid phosphatase in clinical T3/T4M0 prostatic cancer.

Prostate-specific antigen (PSA) and prostatic acid phosphatase (PAP) were determined in the serum of 69 patients with clinical T3/T4M0 prostatic cancer before staging lymphadenectomy. In principle, high-dose radiotherapy was given only to patients of pathological N0 category. Seventeen patients had a prelymphadenectomy PSA level below the normal upper reference limit (10 micrograms/l) and only 3 of them had pelvic lymph node metastases. Fifteen of 52 patients with a preoperative PSA level > or = 10 micrograms/l were of N0 category. Only 8 of the 41 evaluable patients had PAP values above the normal range, and 6 of these 8 patients had pelvic lymph node metastases. Preoperative PSA values, but not preoperative PAP levels, assist the clinician in predicting regional lymph node metastases in patients with clinical T3/T4M0 prostatic cancer. Two-thirds of the patients with T3/T4 tumours and PSA values between 10 and 50 micrograms/l have regional lymph node metastases. About 80% of the patients with PSA levels < 10 micrograms/l belong to the N0 category. About 75% of the patients with PSA > 50 micrograms/l have N+ disease. Taking into account the individual preoperative PSA values, the indication for preradiotherapy staging lymphadenectomy should be balanced between the chance of demonstrating N+ disease, the expected postoperative morbidity and the benefit for the patient found to be of N0 category.     

Carcinoembryonic antigen and carbohydrate antigen 19-9-producing adenocarcinoma of the prostate: report of an autopsy case

Prostate-specific antigen (PSA) and prostatic acid phosphatase (PAP) are well known as specific tumor markers of prostate cancer, but carcinoembryonic antigen (CEA)- and carbohydrate antigen 19-9 (CA19-9)-producing adenocarcinoma originating in the prostate is rare. We report here a case of prostatic adenocarcinoma positive for these 4 tumor markers in a 50-year-old man who had initially complained about chest pain due to metastatic bone tumor. In spite of the extensive treatment involving hormone and radiation therapy, the patient died of rapid tumor extension only 4 months after initial diagnosis. Autopsy revealed multiple metastases to the bone, liver, lungs and lymph nodes. Histologically, two types of adenocarcinoma were involved in both primary prostate and metastatic sites: one was a poorly differentiated adenocarcinoma positive for PSA and PAP but not CEA or CA19-9, and the other one was a less differentiated adenocarcinoma partially positive for CEA and CA19-9 but not for PSA or PAP. Based on this case and previous cases by review of the literature, CEA- and CA19-9-producing adenocarcinoma of the prostate was suggested to rapidly progress with multiple metastases and to show poor prognosis with strong resistance to any treatment.     

Immunoperoxidase localization of prostatic antigens. Comparison of primary and metastatic sites

Immunoperoxidase staining for prostate specific antigen (PSA) and prostatic acid phosphatase (PAP) help to identify patients with prostatic carcinoma presenting as metastatic disease from an occult primary source. To clarify further the reliability of these prostatic tissue antigens, we have examined the primary tumor and metastatic sites in 16 autopsy cases. Eleven of these had diffusely positive findings for PSA and PAP in the primary and all metastatic sites, and 1 case lacked both antigens in all locations. Four cases demonstrated variability between these antigens and among various sites. Prostatic primary lesions contained PAP and PSA in 13 (81%) and 12 (75%) cases, respectively. The most reliable metastatic sites were lymph nodes, seminal vesicles, lung, bone, and kidney; while liver, adrenal, and colorectal sites were less reliable. No relationship existed between serum PAP levels and tissue detectability of PAP. The use of both PAP and PSA increases the likelihood of properly identifying the prostate as the organ of origin of metastatic disease. In spite of the use of both markers, however, three primary lesions would have been misdiagnosed, and 1 case lacked both antigens in all metastatic sites as well. In poorly differentiated lesions, the lack of both antigens does not unequivocally eliminate the possibility of prostatic carcinoma.     

Prostatic acid phosphatase and prostate specific antigen in liver disease

Serum concentrations of prostate specific antigen (PSA) and prostatic acid phosphatase (PAP) were measured in 51 liver cirrhosis, 37 chronic active hepatitis (CAH) patients and 26 healthy individuals. Elevated PSA levels have been found in 2 of cirrhotic patients while no increase has been detected in CAH and controls. Serum PAP levels have been observed slightly increased in 2 patients with cirrhosis, 2 patients with CAH and 1 control case. Mean PSA and mean PAP values showed no significant difference between groups (p>0.05). Serum PSA and PAP levels are reliable in diagnosing and monitoring prostate cancer in chronic liver patients and maintain their specificity in this situation.     

Changes in tumor markers following cryosurgery of prostate carcinoma

Cryosurgery is performed in poor risk cases of prostate carcinoma with dysuria. This modality has been reported to reduce the metastatic lesion postoperatively in cases of prostate carcinoma accompanied by metastasis and is employed as an adjuvant therapy of prostate carcinoma. However, many cases are already at an advanced stage and have undergone other therapeutic modalities and as a result the exact role of cryosurgery in prostate carcinoma is not clear. The present investigation was undertaken to clarify the effectiveness of cryosurgery in prostate carcinoma. The patients consisted of 21 untreated cases of histologically confirmed prostate carcinoma admitted our hospital during the 5-year period from December, 1982 to December, 1987, in all of whom treatment by cryosurgery alone was indicated, i.e., up stage B, and in whom changes in prostate carcinoma tumor markers, alkaline phosphatase (ALP), acid phosphatase (ACP), prostatic ACP detected enzymatically (PACP), and by radioimmunoassay (PAP), gamma-seminoprotein (gamma-Sm), and prostate specific antigen (PSA) were measured. During the same period, changes in tumor makers in 11 cases of prostate hypertrophy treated by transurethral resection of prostate (TUR-P) were also examined. The tumor markers were measured prior to cryosurgery and 1, 3, 7 and 14 days postoperatively as well as at 1, 3 and 6 months. Following TUR-P, in the cases of prostate hypertrophy, no postoperative changes in ALP, ACP or PACP were observed but there was elevation of PAP and gamma-Sm at day 1 and elevation of PSA until day 3, but none of these were statistically significant differences.(ABSTRACT TRUNCATED AT 250 WORDS)     

Comparative study of the clinical usefulness of prostate specific antigen and prostatic acid phosphatase in prostatic disease

The clinical usefulness of prostate specific antigen (PSA) and prostatic acid phosphatase (PAP) activity measurements has been compared in 45 patients with benign prostatic hyperplasia (BPH) and 132 patients with prostatic carcinoma (PC), 21 of whom had metastatic disease (MPC) and 111 of whom had intracapsular cancer. No BPH patient had increased PAP but 47% had increased PSA. Of the PC patients only 27% had increased PAP and 70% increased PSA. All of the MPC patients had increased PSA but only 62% had increased PAP. Increased PAP was found only in MPC but increased PSA was also found in BPH. In identifying PC, the predictive value of an increased PSA concentration is 83% and an increased PAP activity is 100%. On the other hand, the predictive value of a normal PSA concentration is 51% and of a normal PAP activity only 34%. As the PAP test is much less efficient than the PSA test, it should be discontinued.     

Relationship between Prostate-Specific Antigen, Clinical Stage, and Degree of Bone Metastasis in Patients with Prostate Cancer: Comparison with Prostatic Acid Phosphatase and Alkaline Phosphatase

Background:
The study was designed to examine the relation of the levels of prostate-specific antigen (PSA), prostatic acid phosphatase (PAP), and alkaline phosphatase (ALP) to clinical stage and bone metastasis in prostate cancer patients.
Methods:
Serum PSA, PAP, and ALP levels were evaluated in 272 patients with prostate cancer. The relation of the level of PSA, PAP, and ALP to clinical stage and to degree of bone metastasis were examined by a multiple comparison method using ranks. The superiority of a marker in the rate of detection of bone metastasis was evaluated with receiver operating characteristic (ROC) curves. The correlation coefficients of the order of the extent of bone metastasis with PSA, PAP, and ALP were examined with Spearman's rank order correlation coefficient test.
Results:
The levels of PSA showed significant differences among 8 pairs of clinical stages, in contrast, the levels of PAP showed significant differences among 6 pairs, and the levels of ALP showed significant differences among only 4 pairs. The area under the ROC curves of PSA, PAP, and ALP for revealing bone metastasis was 84.9%, 81.4%, and 77.3%, respectively. The correlation coefficients of the order of extent of disease (EOD) with log (PSA), log (PAP), and log (ALP) were 0.346, 0.394, and 0.618, respectively, and the levels of ALP showed the most significant differences regarding the extent of bone metastasis.
Conclusion:
PSA was the best marker for differentiating clinical stages, but showed limited reliability for stratifying the extent of bone metastasis.     

Predicting radionuclide bone scan findings in patients with newly diagnosed, untreated prostate cancer: prostate specific antigen is superior to all other clinical parameters.

Presently, the standard staging evaluation of prostate cancer includes digital rectal examination, measurement of serum tumor markers and a radionuclide bone scan. To evaluate the ability of local clinical stage, tumor grade, serum acid phosphatase, serum prostatic acid phosphatase (PAP) and serum prostate specific antigen (PSA) to predict bone scan findings, a retrospective review of 521 randomly chosen patients (mean age 70 years, range 44 to 92 years) with newly diagnosed, untreated prostate cancer was performed. Local clinical stage, tumor grade, acid phosphatase, PAP and PSA all correlated positively with bone scan findings (p less than 0.0001). Using receiver operating characteristic curves, however, PSA had the best over-all correlation with bone scan results. The median serum PSA concentration in patients with a positive bone scan was 158.0 ng./ml., whereas men with a negative bone scan had a median serum PSA level of 11.3 ng./ml. (p less than 0.0001). Using multivariate logistic regression analysis, local clinical stage, tumor grade, acid phosphatase and PAP were evaluated in combination with PSA to assess whether these parameters increased the ability of PSA alone to predict bone scan findings. None of these clinical parameters, irrespective of the combination used, contributed appreciably to the predictive power of PSA alone. A probability plot with 95% confidence intervals was constructed that allows the practicing urologist to estimate on an individual basis the probability of a positive bone scan for any given serum PSA value. The most significant finding of this study, however, was the negative predictive value of a low serum PSA concentration for bone scan findings. In 306 men with a serum PSA level of 20 ng./ml. or less only 1 (PSA 18.2 ng./ml.) had a positive bone scan (negative predictive value 99.7%). This finding would suggest that a staging radionuclide bone scan in a previously untreated prostate cancer patient with a low serum PSA concentration may not be necessary.     

PAP and PSA in prostatic carcinoma cell lines and aspiration biopsies: relation to hormone sensitivity and to cytological grading

Because a change from hormone-sensitive to hormone-resistant carcinoma of the prostate often occurs concomitantly with genetic changes or as a result of the latter, the markers specific for prostatic tissues might also be affected. We therefore first studied the presence of prostatic acid phosphatase (PAP) and prostate-specific antigen (PSA) in LNCaP and LNCaP-r human prostatic carcinoma cell lines. Since both markers were found in these cell lines, we proceeded to quantitate PAP and PSA in aspiration biopsies from patients with prostate tumors. The amounts of these markers were compared with cytological findings. PAP and PSA were analyzed in the biopsy material from 120 patients using commercial radioimmunoassay (RIA) kits. DNA was determined using Riedel H33258 stain. Cytological grading was performed according to the Uropathological Study Group of Prostatic Carcinoma. Significant correlations were found between PAP/DNA or PSA/DNA values and grade of differentiation of the prostate tumor. In view of earlier reports and the results presented here, the amounts of markers or the protein pattern of tumor tissue may be a useful complement to the morphological findings and for selecting optimal therapy for patients with prostatic tumors.     

Prostate specific antigen--a screening test for prostatic cancer?

A series of 287 patients referred by their family doctors with symptoms of bladder outflow obstruction were asked to attend the hospital for "pre-clinic" screening for carcinoma of prostate (CaP). Blood samples were collected from 211 patients and analysed for serum prostate specific antigen (PSA) and prostatic acid phosphatase (PAP). Thirty-six patients had a serum PSA greater than 10 micrograms/l and 7 had PAP levels greater than 5 iu/l. In no instance was the PAP elevated without an associated increase in PSA concentration. Patients with raised markers underwent further investigations which included prostatic biopsy and/or resection; 17 patients were proved to have carcinoma of the prostate, 9 of whom had distant metastases. The specificity of PSA for detecting prostate cancer in this study was 90% with a sensitivity of 89.5%, in contrast to values for PAP of 100% and 36.8%. The routine use of PAP as a marker for prostatic cancer should be abandoned. The use of PSA as a screening test in a group of patients with prostatism appears justified, but with a positive predictive value of only 47%, its use in a mass unselected screening programme is not recommended.     

The value of acid phosphatase measurements in predicting extraprostatic cancer growth before radical prostatectomy

Acid phosphatase levels were determined using both an enzymatic method (32 cases) and radioimmunoassay (35 cases) in 35 patients with clinically localised prostatic cancer. All patients underwent total prostatectomy and pelvic lymphadenectomy. In cases of intracapsular prostatic cancer the level of prostatic acid phosphatase (PAP) measured by radioimmunoassay was 1.4 +/- 0.8 micrograms/l. In patients with either local extraprostatic disease or pelvic lymph node metastases the mean level of PAP was 3.5 +/- 2.8 micrograms/l. The difference was statistically significant. The specificity, sensitivity and accuracy of an elevated PAP (greater than 3.0 micrograms/l) in revealing extraprostatic extension of clinically localised prostatic cancer were 100, 37 and 66% respectively. When the enzymatic method was used, the level of acid phosphatase was elevated (greater than 13 u/l) in only 1 case. The specificity, sensitivity and accuracy of the enzymatic method were 100, 6 and 47% respectively. Elevation of PAP predicts, with a high degree of probability, either local extension outside the prostate or lymph node metastases. A normal PAP does not exclude extraprostatic extension of prostatic cancer.     

Clinical and pathological study of tumor marker in benign prostatic hypertrophy and incidental prostatic cancer]

To determine the value of prostatic markers for prostate cancer, serum prostatic acid phosphatase (PAP), prostate specific antigen (PSA) and gamma-Seminoprotein (gamma-Sm) were measured in 81 patients with benign prostatic hypertrophy and in 12 patients with incidental prostatic cancer. gamma-Sm was the most sensitive but the least specific of the three markers. Large prostate glands, especially hyper-glandular type tended to be associated with high gamma-Sm levels in our study. Patients with acute urinary retention, acute prostatitis and necrosis also showed positive markers. Out of 12 patients with incidental cancer, 5 patients had more than 2 elevated markers. Four patients with poorly differentiated adenocarcinoma failed to show increased markers.     

Comparative experimental study of the serum prostate specific antigen and prostatic acid phosphatase in serially transplantable human prostatic carcinoma lines in nude mice

Three different human lines of prostate carcinoma were successively transplanted on Balb/c nude mice and the serum values of the prostate specific antigen (PSA) and prostatic acid phosphatase (PAP) were determined simultaneously. In a group of the tumor-bearing animals the influence of endocrine manipulation (castration, estradiol) on the serum concentration of these tumor markers was studied. As far as untreated tumor-bearing mice are concerned, the serum values of both PAP and PSA proved to be strictly dependent on the tumor volume measured. In the exponential growth phase of the grafted tumors, linear growth was linked with a correspondingly increasing PAP and PSA serum level of the test animals. A close correlation was found to exist between the two tumor markers; however, the indicator value of PSA was 20 to 50 times higher than that of PAP under the test conditions. PSA determination yielded no false-negative results, if PAP was elevated. PAP determination was false-negative in 21 per cent of cases with measurable tumors, although the serum level of PSA already showed marked elevation. In treated animals both markers were found to decrease. Arrested growth and tumor regression was associated with falling PAP and PSA serum levels or with levels within normal range. The results of this experimental study support the conclusion that prostate specific antigen represents a substantially more sensitive tumor marker than prostatic acid phosphatase.     

Multicenter, randomized, double-blind, placebo controlled study to investigate the effect of finasteride (MK-906) on stage D prostate cancer.

A total of 28 untreated patients with asymptomatic, stage D prostate cancer was randomized in a double-blinded fashion to receive finasteride (10 mg. per day), a 5 alpha-reductase inhibitor or placebo. Patients were evaluated at 3-week intervals by rectal examination, and serum prostate specific antigen (PSA) and prostatic acid phosphatase (PAP) levels, and at 6-week intervals by bone scan and transrectal ultrasound determinations of prostatic volume. Patients stopped the medication at week 6 at the discretion of the investigator when PSA levels increased from baseline. After 12 weeks all patients were reevaluated. Of the patients 13 received finasteride and 15 received placebo. The 2 groups did not differ statistically with respect to patient age, initial PSA and PAP level, or the extent of metastases on initial bone scan. A statistically significant decrease in the median percentage change from baseline in PSA at weeks 3 and 6 occurred in the finasteride group compared to the placebo group (-22.9% versus -2.9% and -15.1% versus +11.7%, respectively, p less than 0.05). Finasteride had no effect upon PAP, serum testosterone, prostatic volume or appearance of bone scans. A decrease in serum PSA in the finasteride treatment group suggests that finasteride exerts a minor effect in patients with prostate cancer. This effect does not approach that seen with medical or surgical castration yet because of the potency preserving feature and the lack of toxicity finasteride may warrant further study in the treatment of prostate cancer.