2233 COMMUNICATION International Union Against Cancer Workshop on Prostate Markers Gerald P. Murphy, M.D., Alan Partin, M.D., Ph.D.2 1 D.Sc. 1 Pacific Northwest Cancer Foundation, Northwest Hospital, Seattle, Washington. 2 James Buchanan Brady Urological Institute, The Johns Hopkins Hospital, Baltimore, Maryland. Neuroendocrine Cells Data presented at the Workshop on Prostate Markers, Vancouver, British Columbia, Canada, July 11–13, 1998. This meeting was sponsored by the Tumour Biology Committee of the International Union Against Cancer and supported in part by educational grants from Hybritech-Beckman, San Diego, California, and Northwest Biotherapeutics, Seattle, Washington. The following is a list of workshop participants: Per-Anders Abrahamsson, M.D., Ph.D., Lund, Sweden; Georg Bartsch, M.D., Innsbruck, Austria; David Bostwick, M.D., Rochester, MN; Alton Boynton, Ph.D., Seattle, WA; Michael K. Brawer, M.D., Seattle, WA; A. T. K. Cockett, M.D., Rochester, NY; Louis Denis, M.D., Antwerp, Belgium; Lana Grauer, Ph.D., San Diego, CA; Eric Holmes, Ph.D., Seattle, WA; Jari Leinonen, Ph.D., Helsinki, Finland; Curtis Mettlin, Ph.D., Buffalo, NY; Gerald P. Murphy, M.D., D.Sc., Seattle, WA; Alan Partin, M.D., Ph.D., Baltimore, MD; Harry Rittenhouse, Ph.D., San Diego, CA; Peter Snow, Ph.D., Colorado Springs, CO; Paula Southwick, Ph.D., San Diego, CA; Sai Su, Ph.D., Seattle, WA; Donald Tindall, M.D., Rochester, MN; and Robert Wolfert, Ph.D., San Diego, CA. Address for reprints: Gerald P. Murphy, M.D., D.Sc., Pacific Northwest Cancer Foundation, Northwest Hospital, 120 Northgate Plaza, Suite 205, Seattle, WA 98125-7001. Received July 27, 1998; accepted August 28, 1998. © 1998 American Cancer Society Prostatic neuroendocrine cells were first demonstrated in 1944. These granule-containing cells have functional paracrine, autocrine, and endocrine capacity. The neuroendocrine cell secretory activity produces products that stimulate prostate smooth muscle cells, benign prostate epithelial cells, and prostate carcinoma cells. The materials produced by these cells have been demonstrated to induce focal differentiation in prostate carcinoma. Prostate specific antigen (PSA) is coexpressed in some neuroendocrine prostate cells. There are a number of things that remain unknown regarding such cell types. For example: 1) the expression of neuroendocrine receptors, 2) correlation between receptors and prognostic factors, 3) relation to androgen dependence of prostate cells, and 4) correlation with PSA serum values. Neuroendocrine cells communicate with peripheral nerves and with each other. They are involved in the normal regulation of prostate cell growth and the regulation of prostatic secretions. There also is an interaction between the products of the prostatic stroma and neuroendocrine cells. In prostate carcinoma, neuroendocrine differentiation is present in 7–100% of cases, varying according to fixative and method of identification employed. These cells are found in association with nerves and also in such locations as metastatic lymph nodes. Although it is believed that neuroendocrine cells are androgen receptor negative, this has not been established definitely. There currently are a diversity of opinions and limited available information. It has been observed, after antiandrogen therapy, that prostate tumors may have an increased number of neuroendocrine cells. This situation also is disputed. In a few limited reports, the levels of neuroendocrine secretory products in the blood have been found to be higher in patients with androgen independence. It also is believed that these serum products are not suppressed by androgen ablation therapy or radiation therapy. Whether these products in the tissue have value as an independent indicator for poor prognosis is unresolved. At the current time, staining prostate carcinoma for neuroendocrine cells is believed to be an important research project. The future role of neuroendocrine substances as practical markers are unresolved. Data were presented at the meeting in Sweden to suggest that in 103 cases, neuroendocrine staining was related to the Gleason grade of the primary tumor and, furthermore, that this difference also was related to survival. 2234 CANCER November 15, 1998 / Volume 83 / Number 10 There is a cell line called CRL-5813 that is derived from a lymph node metastasis from a prostate carcinoma patient. The cDNA fragments found were positive for estrogen receptor-␣, and for androgen receptor in this particular cell line. All three chromogranin secretory products also have been identified. Future goals relate to the measurement of neuroendocrine receptors and the mitogenic effects of the neuropeptides produced by these cells. It was further reiterated that these cells do coexpress PSA. The source of neuroendocrine cells in the prostate is unknown. Although they previously were considered to be stem cell products, evidence was reported at the meeting that suggests that these cells may be of ectodermal origin. Data are required from prospective randomized trials to evaluate whether the expression of these cells is an independent indicator of disease progression. There may be preliminary information forthcoming from a Radiation Therapy Oncology Group trial recently completed in North America. However, further details are unavailable at the current time. Free Prostate Specific Antigen A major prospective multiinstitutional study conducted by Hybritech/Beckman of San Diego, California, examined patients who had a serum PSA level between 4.0 –10.0 ng/mL and a % free PSA (fPSA) determination. Particular emphasis was made to examine men age ⬎ 50 years (in whom the prevalence of prostate carcinoma was 4%). In this particular study, these men also had a negative digital rectal examination. It was noted that this group would equal 10% of the population of men in a previously conducted total PSA trial. The proportion of men in this group who were found to be negative by digital rectal examination with a PSA between 4 –10 ng/mL was 9%. This group also represented 33% of all the biopsies performed in the initial Hybritech report. A total of 773 men were reported. A cutoff point of ⬍ 25% for abnormal fPSA in this group, as shown, would detect 95% of the tumors. Approximately 20% of the benign prostatic hyperplasia (BPH) cases would not be biopsied, sparing these patients an unnecessary procedure. The percentage of fPSA levels was similar in African-American and white patients. As patient age increased, the percentage of fPSA also increased. Using the 25% fPSA level obviates any need in this study for age specific cutoff levels because such a level, when employed, would detect nearly 98% of younger men with prostate carcinoma. This study found that the tumors that would not be detected by this test were located in: 1) older men; 2) men with larger glands (benign); and 3) men with less aggressive tumors. The percent fPSA could be used in two different ways: 1) as a single cutoff point of 25%; and 2) in individual patient risk assessment, determining that the lower the percent fPSA, the higher probability of prostate carcinoma. These levels and the degree of risk are determined by the prevalence of prostate carcinoma in any particular group. In addition, for the prediction of favorable pathology in patients with biopsy-proven carcinoma, a fPSA value ⬎ 15% was used as an effective discriminator. Additional data were obtained from a multiinstitutional program collected from different countries. However, the information has not yet been defined fully. In men with larger prostates, not necessarily in the screening population, different data may be obtained subsequently. It is possible that the application of the Hybritech report to a general population may result in differing observations. This situation is unresolved and further information from this particular report will be needed. In the chairman’s summary it was noted that there are reports that fPSA is low in men with chronic prostatitis, in contrast to other reports that state that acute prostatitis does not affect fPSA. There are limited reports suggesting that fPSA may predict capsular penetration and that the advantage of fPSA determinations may be reduced in patients with prostates of smaller size compared, for example, in patients with prostates ⬎ 40 cc. It also was noted that the five ␣-reductase acting agents might equally lower total PSA, as well as fPSA. In the European screening trial currently underway, as a result of preliminary experiences, digital rectal examination and transrectal ultrasound have been dropped. All patients are given a PSA determination and every screening patient with a total PSA ⬎ 3.0 ng/mL is recommended for biopsy. Other potential cutoff levels for % fPSA and total PSA were discussed. Complexed Prostate Specific Antigen Two published studies are available in the U. S. from the Bayer Company and from Hybritech. A number of previous reports from the 1991 era measuring complexed PSA (C-PSA) are of limited value. C-PSA is the serum moiety that occurs in greater proportion in men with carcinoma. In 1997, the Hybritech group published their assay. The one question concerned how to measure C-PSA: 1) the sandwich assay, binding epitopes on PSA and ␣1-antichymotrypsin (ACT); and 2) a competition assay utilizing unlabeled cold antiPSA antibodies (E-Epitope). The assay results, comparing C-PSA with total PSA and fPSA, were reviewed from recent data. This provided a sensitivity of 94% and a specificity of 23%. The conclusion of the Hybritech report was that C-PSA did not add to total PSA UICC Workshop on Prostate Markers/Murphy and Partin for the detection of tumor. %fPSA was superior to C-PSA or total PSA within the 4.0 –10.0 ng/mL total PSA range. A recent study on the Bayer assay was completed in Seattle. This study reported the amount of the CPSA. The % C-PSA in this particular data set had the best receiver operating characteristic (ROC). Comparing the percentage of C-PSA and the percentage fPSA also may be more useful than C-PSA alone. C-PSA also is useful for monitoring and staining prostate tumor cells. In another study in the 4.0 –10.0 ng/mL range from Seattle, the best ROC curve was noted using C-PSA (Bayer assay). Additional combination studies currently are being considered to evaluate further the diagnostic potential of C-PSA. Additional data were presented regarding 35 cases. The conclusions from these studies were: 1) that C-PSA by itself offers a 10% improvement in specificity overall and a 17% improvement in the 4.0 –10 ng/mL total PSA range; and 2) C-PSA may play a role in obviating unnecessary biopsies in men with a total PSA ⬍ 4.0 ng/mL. The Helsinki group reported at this meeting on a PSA-API ␣-1 protease inhibitor (PSA-API) in the serum. It is possible to combine the proportion of PSAAPI and PSA-ACT or fPSA in the serum over PSA levels of 4 –20 ng/mL. This may improve the diagnostic accuracy. The group also is measuring nonimmunoreactive PSA. They have measured ␣-2 macroglobulin (A2M) C-PSA. They have devised a novel immunoflurometric assay for the measurement of PSA-A2M based on the removal of the immunoreactive PSA in the serum by immunoabsorption, denaturation of PSA-A2M at high pH, and the measurement of the released PSA immunoreactivity by a conventional PSA immunoassay. ROC analysis suggested the measurement of the ratio of PSA-A2M to total PSA in the serum improves the diagnostic accuracy for prostate carcinoma compared with an assay for total PSA only. However, these results are preliminary. The group also has developed a specific immunoflurometric assay for PSA-ACT based on one monoclonal antibody against PSA and one monoclonal antibody specific for complexed ACT. A monoclonal antibody specific to the complexed ACT can be developed and used with anti-PSA monoclonal antibodies in a sandwich assay for PSA-ACT. These monoclonal antibodies strongly reduced binding with the native ACT, suggesting they recognize ACT modified during the complexation. The reduction of the nonspecific interference due to ACT and due to CG-ACT (a monoclonal antibody that reacts with cathepsin G with ACT) should allow more reliable use of PSA-ACT assays in 2235 the diagnosis of prostate carcinoma. There were several other interesting observations, namely that the proportion of PSA-A2M was lower in the serum in prostate carcinoma patients than in those with benign disease. Perhaps the sum of the proportions of PSAA2M plus fPSA can be used to reduce the false-positive results due to BPH. This work is extremely interesting but at a very early stage in its development. Human Kallikrein The human kallikrein family contains three very homologous serine proteases including PSA. The hK2 gene was discovered in 1987. The hK2 and hK1 kallikreins show approximately 80% and 60%, respectively, sequence identity with PSA. Like PSA, hK2 is localized to the prostate and hormonally regulated. In contrast to PSA, hK2 is a potent enzyme with approximately 20,000-fold more enzyme activity than PSA on small molecular weight substrates. The current physiologic function of hK2 is not known but its enzyme specificity may provide important clues. hK2 selectively cleaves arginine residues and readily cleaves the zymogen form of PSA (proPSA) to produce active PSA, at least in vitro. One function of hK2 may be to activate and regulate PSA. hK2 forms complexed with ACT. In seminal fluid approximately 50% of the hK2 is complexed with protein C inhibitor whereas PSA is nearly entirely in the free form. Thus far, research has confirmed that hK2 is expressed selectively in prostate epithelial cells. Androgens for the androgen receptor can regulate this gene. There also is an interaction between the androgen receptor and the C-JUN or C-FOS oncogenes. Vitamin D3 and thyroxine also play a role in the regulation of this gene. A recent report at the American Urological Association meeting in San Diego, California in 1998 by Dr. Kevin Slawin of Baylor College of Medicine, described reverse transcriptasepolymerase chain reaction (RT-PCR) hK2 results in 293 men after radical prostatectomy. At a follow-up time of 24 months, there was a difference in the recurrence rates between those men who subsequently were believed to be positive (14%) versus those men who were negative. Multivariate analysis with other parameters, including Gleason score, currently is in progress. The status of the pelvic lymph nodes measured at radical prostatectomy in 219 patients also suggested that preoperative RT-PCR hK2 predicts final lymph node positivity and may predict recurrence. These initial results will be watched very closely to observe how follow-up of clinical status correlates with RT-PCR hK2 and available multiparametric clinical data. Immunohistochemical studies at the Mayo Clinic in Rochester, Minnesota have confirmed that the hK2 2236 CANCER November 15, 1998 / Volume 83 / Number 10 protein is expressed specifically in prostate epithelial cells. In contrast to PSA, hK2 has a lower expression in BPH than in prostate tumors. However, there was considerable overlap in the intervals between all three groups. Basal cells in the prostate virtually were negative. Stroma and urothelium also were negative. Data with two different antibodies to hK2 were similar in performance and superior to staining with PSA or prostate acid phosphatase (PAP). PSA and PAP cell staining generally was decreased in both prostatic intraepithelial neoplasia (PIN) and prostate tumors. No prognostic value for hK2 staining was found for patients with lymph node metastasis although virtually all lymph node metastatic cells stained intensely for hK2. Hybritech/Beckman then reviewed their hK2 total assay results. The various forms present in seminal plasma, prostate tissue, and serum were recounted. The cross-reactivity of PSA and hK2 antibodies to the homologous antigen remains a potential problem for antibody selection. The Hybritech/Beckman assay for total hK2 does not cross-react with PSA in the serum. In general, monoclonal antibodies to conformational epitopes for PSA are less likely to cross-react with hK2 and vice versa. Polyclonal antibodies to PSA or hK2 cross-react with the homologous kallikreins. Monoclonal antibodies to linear epitopes on PSA or hK2 are more likely to cross-react with the other kallikreins, but can be selected to be specific to either PSA or hK2. The epitope maps for PSA and hK2 are complex and the data from various groups will be reported shortly from recent international workshops. A multiinstitutional study showed that the ratio of fPSA to total hK2 may significantly enhance discrimination between prostate carcinoma and BPH, especially in the range of 2.0 – 6.0 ng/mL. Prostate Specific Membrane Antigen Prostate specific membrane antigen (PSMA) was identified by the monoclonal antibody 7E11 in 1987. The cDNA sequence for the coding region of the membrane bound PSMA antigen reactive with 7E11 was revealed in research completed in 1993. The extracellular domain contains at least two homologies: one to transferrin receptor and one to NAALADase, an enzyme found in the brain. The antigen exists in at least two forms: the so-called membrane bound PSMA antigen, and PSM (PSM Prime [PSM⬘]), which has an alternate start site, lacks the transmembrane region, and has fewer amino acids as a result. This is found in the cytoplasm. The discussion focused on the advantages of PSMA as a marker given that it is an integral membrane protein. This property makes it ideal for cell harvesting, imaging, and therapeutics. Other ques- tions focused around the mechanism of 7E11 binding and its expression outside the prostate. Two reports were reviewed from the Mayo Clinic that used the 7E11 antibody for staining of radical prostatectomy cases as well as lymph nodes. The confidence limits for the staining results were quite narrow in contrast to those reported the previous day for hK2. There was no background staining in contrast to previous reports published elsewhere. This is believed to be due to technical differences. The Mayo Clinic investigators use overnight incubation for the antibody in contrast to other more arduous and perhaps artifact-creating endeavors. The basal cells were not stained. Immunoreactivity was enhanced on the luminal surface of the prostate epithelial cells again in contrast to the diffuse staining with hK2. High grade prostate carcinoma showed enhanced expression. There is heterogeneity observed in low grade prostate carcinoma. Again, this is in contrast with results published elsewhere. The antibody is a reliable staining agent and has the ability to differentiate between BPH, PIN, and adenocarcinoma. Another report dealt with the characteristics of the entire PSMA molecule and monoclonal antibodies formed to different regions. There are at least ten potential glycosylation sites on the antigen. 7E11 recognizes the first six amino acids from the N-terminal of the protein that are distributed on the cytosolic side of the membrane. Pacific Northwest Cancer Foundation/Northwest Hospital in collaboration with Northwest Biotherapeutics and Hybritech/Beckman has created more than 32 monoclonal antibodies to the extracellular domain. These have unique properties and provide multiple opportunities for diagnosis and therapy. PSMA expressing 7E11 positive cells also are found in semen, and are increased in amount relative to those staining for cytokeratin 8-18 in the presence of malignancy and increased tumor size. Alternate splicing as a factor in the propagation of PSMA also was reviewed. RT-PCR with PSMA and PSM⬘ can be performed. The PSM⬘ from normal prostate is slightly different from that derived from the LNCAP cell line material. The PSM⬘ is higher in normal prostate tissue compared with that found in prostate carcinoma. It is further down-regulated by androgen in LNCAP cells. The clinical value of the use of RT-PCR for detecting hematogenous circulating prostate cells in prostate carcinoma patients remains unresolved. There are three main problems: 1) modes of protocol (e.g., choice of primers) and PCR conditions such as the number of cycles; 2) diffuse expression of genes under various conditions affects the performance of the assay; and 3) technologic problems that relate to sampling issues also are present. UICC Workshop on Prostate Markers/Murphy and Partin Further research has been done by the Hybritech/ Beckman group to identify PSM⬘ in the cytosol fraction from LNCAP cells. PSMA as a specific molecule has a great number of properties and currently is being used for visualization of lymph node metastases in nuclear medicine tests. It doubtless is a matter of time until a serum assay is available to test the ability of PSMA to detect prognostic factors at both early and late events in the patient with prostate carcinoma. Artificial Neural Networks Preliminary data from artificial neural networks involving diagnosis, screening data, prostatectomy results, and recurrence rates were presented. The data were preliminary and no conclusion was reached. However, the potential ability of artificial neural networks to provide a means for further discrimination was discussed and presented. Several early data sets from patients with prostate carcinoma at various stages of early diagnosis and advanced disease had some interesting features. Throughout the presentation, two major prognostic factors were identified. These were the Gleason grade and PSMA serum values as determined by Western blot analysis. This was in marked contrast to other previously reported factors that have been enumerated in various studies, such as PSA, total PSA, fPSA, age, race, and other pathologic findings. It is too early to reach conclusions from such tests, but the results raise important questions for current and future study. Total Prostate Specific Antigen, Free Prostate Specific Antigen, Complexed Prostate Specific Antigen, and hK2 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. SUGGESTED READINGS Neuroendocrine 1. 2. 3. 4. 5. 6. 7. Deftos LJ, Granin A. Parathyroid hormone-related protein and calcitonin gene products in neuroendocrine prostate cancer. Prostate 1998;(Suppl):23–31. Diaz M, Abdul M, Hoosein N. Modulation of neuroendocrine differentiation in prostate cancer by interleukin-1 and -2. Prostate 1998;(Suppl):32– 6. Abrahamsson PA, Cockett TK, di Sant’Agnese. Prognostic significance of neuroendocrine differntiation in clinically localized prostatic carcinoma. Prostate 1998;(Suppl):37– 42. Cussenot O, Villette JM, Cochand-Priollet B, Berthon P. Evaluation and clinical value of neuroendocrine differentiation in human prostatic tumors. Prostate 1998;(Suppl):43– 51. Xue Y, Smedts F, Verhofstad A, Debruyne F, de la Rosette J, Schalken JK. Cell kinetics of prostate exocrine and neuroendocrine epithelium and their differential interrelationship: new perspectives. Prostate 1998;(Suppl):62–73. di Sant’Agnese PA. Neuroendocrine differentiation in prostatic carcinoma: an update. Prostate 1998;(Suppl):74 –9. Ischia R, Culig Z, Eder U, Bartsch G, Winkler H, FischerColbrie R, et al. Presence of chromogranins and regulation of their synthesis and processing in a neuroendocrine prostate tumor cell line. Prostate 1998;(Suppl):80 –7. 2237 12. 13. 14. 15. 16. 17. Wang MC, Valenzuela LA, Murphy GP, Chu TM. Purification of a human prostate specific antigen. Invest Urol 1979;17: 159 – 63. Brawer MK, Lange PH. Prostate-specific antigen: its role in early detection, staging, and monitoring patients with prostatic carcinoma. J Endourol 1989;3:227–36. Cooner WH, Mosley BR, Rutherford CL, et al. Prostate cancer detection. I. A clinical urological practice by ultrasonography, digital rectal examination and prostate specific antigen. J Urol 1990;143:1146. Catalona WJ, Smith DS, Ratliff TL, et al. Measurement of prostate-specific antigen in serum as a screening test for prostate cancer. N Engl J Med 1991;324:1156. Labrie F, Dupont A, Suburu R, et al. Serum prostate specific antigen as pre-screening test for prostate cancer. J Urol 1992;147:846. Partin AW, Pound CR, Clemens JQ, et al. Serum PSA after anatomic radical prostatectomy. The Johns Hopkins experience after 10 years. Urol Clin North Am 1993;20:713–25. Elgamal AA, Cornillie FJ, Van Poppel HP, et al. Free-to-total prostate specific antigen ratio as a single test for detection of significant stage T1c prostate cancer. J Urol 1996;156: 1042–9. Elgamal AA, Baert L. RE: The free-to-total prostate specific antigen ratio improves the specificity of prostate specific antigen in screening for prostate cancer in the general population. J Urol 1998;159:994 –5. Douglas TH, Morgan TO, McLeod DG, et al. Comparison of serum prostate specific membrane antigen, prostate specific antigen, and free prostate specific antigen levels in radical prostatectomy patients. Cancer 1997;80:107–14. Woodrum DL, Brawer MK, Partin AW, et al. Interpretation of free prostate specific antigen clinical research studies for detection of prostate cancer. J Urol 1998;159:5–12. Pannek J, Rittenhouse HG, Chan DW, et al. The use of percent free prostate specific antigen for staging clinically localized prostate cancer. J Urol 1998;159:1238 – 42. Stenman U-H, Leinonen J, Alfthan H, et al. A complex between prostate-specific antigen and alpha 1-antichympotrypsin is the major form of prostate-specific antigen in serum of patients with prostatic cancer: assay of complex improves clinical sensitivity for cancer. Cancer Res 1991;51: 222. Lilja H, Christensson A, Dahlén U, et al. Prostate-specific antigen in serum occurs predominantly in complex with ␣1-antichymotrypsin. Clin Chem 1991;37:1618. Christensson A, Bjôk T, Nilsson O, et al. Serum prostate specific antigen complexed to ␣1-antichympotrypsin as an indicator of prostate cancer. J Urol 1993;150:100. Stenman U-H, Hakama M, Knekt P, et al. Serum concentrations of prostate specific antigen and its complex with ␣1antichympotrypsin before diagnosis of prostate cancer. Lancet 1994;344:1594. Partin AW, Brawer MK, Subong ENP, et al. Prospective evaluation of percent free-PSA and complexed-PSA for early detection of prostate cancer. Prostate Cancer Prostat Dis 1998;1:197–203. Brawer MK, Meyer GE, Letran JL, et al. Measurement of complexed PSA improves specificity for early detection of prostate cancer. Urology. In press. 2238 CANCER November 15, 1998 / Volume 83 / Number 10 18. Wang TJ, Linton HJ, Payne J, et al. Clinical utility of a complexed PSA Immunoassay with a specific monoclonal antibody to PSA-ACT [abstract]. J Urol 1997;157(S):147. 19. Rittenhouse HG, Finlay JA, Mikolajczyk SD, et al. Human kallikrein 2 (hK2) and prostate specific antigen (PSA): two closely related, but distinct, kallikreins in the prostate. Crit Rev Clin Lab Sci 1998;35(4):295–368. 20. Darson MF, Pacelli A, Roche P, et al. Human glandular kallikrein 2 (hK2) expression in prostatic intraepithelial neoplasia and adenocarcinoma: a novel prostate cancer marker. Urology 1997;49:857– 62. 21. Rittenhouse H, Tindall D, Klee G, et al. Characterization and evaluation of hK2: a potential prostate cancer marker, closely related to PSA. In: Proceedings of the 1st International Consultation on Prostate Cancer. Monaco: Scientific Communication International Ltd., 1997:133– 40. 7. 8. 9. 10. Prostate Specific Membrane Antigen 1. 2. 3. 4. 5. 6. Gregorakis AK, Holmes EH, Murphy GP. Prostate-specific membrane antigen: current and future utility. Semin Urol Oncol 1998;16:2–12. Horoszewicz JS, Kawinski E, Murphy GP. Monoclonal antibodies to a new antigenic marker in epithelial prostatic cells and serum of prostatic cancer patients. Anticancer Res 1987; 7:927–36. Silver DA, Pellicer I, Fair WR, et al. Prostate-specific membrane antigen expression in normal and malignant human tissues. Clin Cancer Res 1997;3:81–5. Troyer JK, Becket ML, Wright GL. Detection and characterization of prostate-specific membrane antigen (PSMA) in tissue extract and body fluids. Int J Cancer 1995;62:552– 8. Fair WR, Israeli RS, Heston WDW. Prostate-specific membrane antigen. Prostate 1997;32:140 – 8. Liu H, Moy P, Kim S, et al. Monoclonal antibodies to the extracellular domain of prostate-specific membrane antigen 11. 12. 13. 14. also react with tumor vascular endothelium. Cancer Res 1997;57:3629 –34. Bostwick DG, Pacelli A, Blute M, et al. Prostate specific membrane antigen expression in prostatic intraepithelial neoplasia and adenocarcinoma: a study of 184 cases. Cancer 1998;82:2256 – 61. Murphy GP, Holmes EH, Boynton AL, et al. Comparison of prostate specific antigen, prostate specific membrane antigen, and LNCaP-based enzyme-linked immunosorbent assay in prostatic cancer patients and patients with benign prostatic enlargement. Prostate 1995;26:164 – 8. Murphy GP, Ragde H, Kenny G, et al. Comparison of prostate specific membrane antigen, and prostate specific antigen levels in prostatic cancer patients. Anticancer Res 1995; 15:1473– 80. Rochon YP, Horoszewicz JS, Boynton AL, et al. Western blot assay for prostate-specific membrane antigen in serum of prostate cancer patients. Prostate 1994;25:219 –23. Murphy GP, Maguire RT, Rogers B, et al. Comparison of serum PSMA, PSA levels with results of Cytogen-356 ProstaScint scanning in prostatic cancer patients. Prostate 1997; 33:281–5. Barren RJ, Holmes EH, Boynton AL, et al. Method for identifying prostate cells in semen using flow cytometry. Prostate 1998;36:181– 8. Israeli RS, Miller WH, Su SL, et al. Sensitive nested reverse transcription polymerase chain reaction detection of circulating prostatic tumor cells: comparison of prostate-specific membrane antigen and prostate-specific antigen-based assays. Cancer Res 1994;54:6306 –10. Loric S, Dumas F, Eschwege P, et al. Enhanced detection of hematogenous circulating prostatic cells in patients with prostate adenocarcinoma by using nested reverse transcription polymerase chain reaction assay based on prostatespecific membrane antigen. Clin Chem 1995;41:1698 –704.