The Prostate 28:295-299 (I 996) Prostatic Acid Phosphatase Levels (Enzymatic Method) From Completely Sectioned, Clinically Benign, Whole Prostates P.N. Brawn, D.W. Jay, D.M. Foster, D. Kuhl, V.O. Speights, E.H. Johnson, M. Riggs, M.L. Lind, K.S. Coffield, and B. Weaver Department of Pathology, Veterans Administration Medical Center, University of Michigan, Ann Arbor (P.N.6.); Department of Pathology, Veterans Administration Medical Center (D.WJ., D.K, E.H.1.) and Departments of Pathology (V.O.S.), Statistics (M.R), and Urology (KS.C.), Scott & White Memorial Hospital, Texas A M University School of Medicine, and Department of Pathology (6.W.), Veterans Administration Medical Center, Templejunior College, Temple, Department of Nuclear Medicine (D.M.F.), Veterans Administration Medical Center, Institute of Biomedical Studies, Baylor University, Waco, and Department of Research and Education (M.LL), Kelsey-Seybold Clinic, Houston, Texas ABSTRACT: Clinically benign, whole, untrimmed prostates were obtained from 104 patients at autopsy, completely sectioned, and examined microscopically. The histological and gross findings of the prostate were correlated with premortem prostatic acid phosphatase levels (PAP, enzymatic method, ACA, Dupont Co.) to determine how often carcinoma of the prostate (CAP) affected PAP levels and to identify other findings within the prostate associated with elevated PAP levels. Sixty (58%)prostates did not have CAP, 34 (33%)had CAP smaller than 1 ml in volume, and 10 (10%)had CAP larger than 1 ml in volume. PAP levels were elevated (greater than 1 U/L) in 8 of the 60 (13%)prostates without CAP, in 2 of the 34 (6%) prostates with CAP smaller than 1 ml, and in 1 of the 10 (10%) prostates with CAP larger than 1 ml. These differences were not statistically significant. Likewise, a statistically significant correlation between PAP levels and patient age, patient race, severe inflammation, or high grade prostatic intraepithelial neoplasia (PIN) was not found. However, there was a statistically significant correlation between PAP levels and prostate weight (P < 0.OOOl).This study suggests that PAP cannot distinguish between patients with clinically undetected CAP and patients without CAP. Furthermore, elevated PAP levels are often not due to metastatic CAP and additional evidence should be present, even in patients with known CAP, before an elevated PAP level is considered to be conclusive evidence of metastatic CAP. 0 1996 Wiley-Liss, h c . * KEY WORDS: prostate carcinoma, acid phosphatase, atypical hyperplasidadenosis, prostatic acid phosphatase, prostatic intraepithelial neoplasia INTRODUCTION levels are prostate-specific antigen valuable but have limitations, i.e., the inability to detect most carcinoma Of the prostate (CAP) -less than 1 d in volume and lack of specificity for CAP [l]. Although another serum test, prostatic acid @ 1996 Wiley-Lss, Inc. *This article is a US Government work and, as such, is in the public domain in the United States of America Received for publication December 24, 1994; accepted April 27, ,mE Il7.I. Address reprint requests to Peter N. Brawn, Department of Pathology, Veterans Administration Medical Center, University of Michigan, Ann Arbor, MI 48105. 296 Brawn et al. phosphatase (PAP, enzymatic method) has been available for decades, it is unknown how often PAP is elevated in patients with clinically undetected CAP or in patients without CAP. Furthermore, the frequency and etiology of elevated PAP levels in patients without CAP, using fully sectioned prostates, have not been reported. Since most patients, at one time or another, receive at least one serum PAP determination, it remains important to be able to interpret normal, borderline, or elevated PAP levels, regardless of the PSA level. To further our understanding of PAP, this study correlated serum PAP levels with the histological and gross findings from completely sectioned, clinically benign, whole prostates. MATERIALS AND METHODS Whole, untrimmed prostates were obtained from 104 consecutive autopsies which fulfilled the following criteria: 1) They did not have a history of CAP, 2) digital examinations of the prostate had been done within 1 year of death and were not suspicious for malignancy, and 3) stored serum had been drawn within 72 hr prior to death and frozen at the time of autopsy. These 104 prostates include 100 prostates that had not been used in a study before and 4 prostates that had been included in a prior study of PSA [11. These whole, untrimmed prostates included the entire prostate, both seminal vesicles, approximately 1 cm of bladder neck, and a small amount of tissue beyond the surgical capsule. The prostates were untrimmed because the adjacent tissue was thought to be essential for evaluating whether CAP or other histological findings extended beyond the prostate. The prostates accounted for 75% of the weight of the untrimmed specimens as previously described [l]. The untrimmed prostates were weighed, fixed in formalin for at least a week, completely cross sectioned perpendicular to the urethra at 3 mm intervals, cut with a microtome, placed on large slides, and stained as previously described . Full cross sections were examined histologically for the presence of CAP, inflammation, prostatic intraepithelial neoplasia (PIN), and atypical hyperplasia/adenosis (AHA). Unequivocal evidence was required for the diagnosis of CAP. CAP was strictly separated from AHA, atypical atrophy, sclerosing adenosis, atypia caused by inflammation, mesonephric hyperplasia, and cribriform hyperplasia [3-91. Inflammation was divided into three categories: minimal to none, moderate, and severe. Minimal to no inflammation was defined as occasional collections of inflammatory cells, usually lymphocytes, pri- marily in prostatic stroma with minimal involvement of prostatic glands. Moderate inflammation involved a larger portion of the prostate and could be acute, chronic, or both. Moderate inflammation, in any location, did not completely fill one low-power (5 mm) microscopic field. Severe inflammation could be acute, chronic, or both and was defined as inflammation which completely filled at least one low-power microscopic field. PIN was categorized as low grade (PIN 1) and high grade (PIN 2-3) and was characterized by cytological abnormalities of prostatic glands without major architectural disturbances [lo]. AHA was categorized as mild, moderate, and severe as has been described in numerous studies [3,4,6]. AHA is characterized by architectural disturbances of prostatic glands without significant cytological abnormalities. PAP levels were determined by the enzymatic method (ACA, Dupont Co., normal 0-1 UL).Sample stability has been previously described [11,12]. Necessary precautions were taken to ensure sample integrity. Validity of data was assured through repetitive testing, either initially or longitudinally, or both. Statistical analyses were performed using the Kruskal-Wallis test and Pearson product-moment correlations. P values < 0.05 were considered to be statistically sigruficant. RESULTS The 104 patients included 89 white (including Hispanic) and 15 black patients with an average age of 69 years (range 38-96 years). Ninety-three had PAP levels between 0-1 U/L and 11 had PAP levels above 1 UL.The characteristics of these latter 11 patients are presented in Table I. Patients Without CAP Sixty of the 104 patients did not have CAP. Fifty were white and 10 were black. These patients had an average age of 66 years (range 38-95 years) and an average untrimmed prostate weight of 35.3 g (range 15-152 g). Flfty-two of these 60 (87%) patients had PAP levels between 0-1 U/L while 8 (13%)had PAP levels above 1 U/L (mean and median PAP levels for these 60 patients were, respectively, 1.58 U/L and 0.37 U/Lwith a range from 0.01 to 45 UL). Patients With CAP Fourty-four of the 104 (43%)had CAP. Thirty-nine were white and five were black. These patients had an average age of 73 years (range 55-96 years) and an average untrimmed prostate weight of 39.3 g (range 16-103 g). Thirty-four of these patients had CAP less than 1 PAP Levels in Whole Prostates 297 TABLE 1. Characteristics of the I I Patients With Elevated PAP Levels* Age/ Race Prostate weight untrimmed (GMS) 65W 58W 5OW 59w 64W 77w 66B 81B 71W 67W 79w PAP U/L CAP 35 1.16 33 22 26 49 40 36 32 116 57 152 1.24 1.41 2.57 3.03 3.71 6.38 7.57 11.60 14.47 45.0 YES 0.25 ml NO NO NO NO NO NO YES 7.0 ml NO YES 0.2 ml NO Additional histology Cause of death High grade PIN ASVD None None High grade PIN High grade PIN High grade PIN Prostatic infarct Moderate inflammation Prostatic infarct None High grade PIN ASVD Esophageal carcinoma with metastases Pneumonia Post surgical hemorrhage Emphysema Perforated GI ulcer and emphysema Alzheimers Leukemia Lymphoma ASVD ‘ASVD, atherosclerotic vascular disease; GI, gastrointestinal. ml and 2 of these 34 patients had elevated PAP levels (mean and median PAP levels for these 34 patients were, respectively, 0.78 U/L and 0.32 U/L with a range from 0.01 to 14.47 U/L). Ten of these patients had CAP larger than 1ml and one of these had an elevated PAP level (mean and median PAP levels for these 10 patients were, respectively, 1.1U L and 0.43 U/L with a range from 0.16 to 7.57 UL). Additional Categories of Patients andor Specimens Three untrimmed prostates weighed 80 g or more while 101 weighed less than 80 g. Two of the three weighing more than 80 g had elevated PAP levels. Twenty-two prostates had severe acute and/or chronic inflammation and three of these had elevated PAP levels. Eighty-two did not have severe acute and/or chronic inflammation and eight of these had elevated PAP levels. Sixty-two had high grade PIN while 42 did not. Five patients with high-grade PIN had elevated PAP levels while six patients without high-grade PIN had elevated PAP levels. Statistical Analysis There were no statistical differences in the median PAP levels of prostates without CAP, prostates with CAP less than 1 ml, and prostates with CAP larger than 1 ml (P = 0.38) (Fig. 1). , There were no statistical differences in the median PAP levels of the prostates with severe acute and/or chronic inflammation compared to prostates without severe acute and/or chronic inflammation (P = 0.18). There was a sigruficant correlation between untrimmed prostate weight and PAP levels (r = 0.65; P less than 0.0001) (Fig. 1). There was no significant correlation between patient age and PAP levels (P = 0.36). There was no racial difference in PAP levels between white and black patients (P = 0.08). There was no sigruficant correlation between highgrade PIN and PAP levels (P = 0.14) (median PAP levels of 0.34 U/L for patients with high-grade PIN and 0.39 U/L for patients without high-grade PIN). DISCUSSION This study found elevated PAP levels in 11 of 104 patients with clinically benign prostates. Eight of these elevations occurred in patients without CAP while the remaining three occurred in patients with clinically unsuspected CAP. Statistically sigruficant differences in PAP levels were not found between patients without CAP, patients with CAP smaller than 1 ml, or in patients with CAP larger than 1 ml. Furthermore, patient age, patient race, severe acute and/or chronic inflammation, or high-grade PIN were not statistically correlated with PAP levels. The only statistical correlation between PAP levels and the gross or histological findings of the prostate was prostate weight (P < 0.0001). Possible etiologies for elevated PAP levels were identified in five cases. These were 1)an untrimmed prostate weighing 152 g (there is a correlation be- 298 Brawn et al. 0 X al o Y CONCLUSIONS 0 o Oo0 0 l o o NoCA x CAc 1 mi C A > l mi 0 20 40 60 mal percentages of elevated PAP levels in elderly men. 80 100 120 140 160 This study confirms that PAP (enzymatic method) is of little value in detecting CAP in clinically benign prostates. Furthermore, this study is the first to document, through complete sectioning of the prostate, that a surprisingly large percentage of elderly men have elevated PAP levels in the absence of CAP and that it is often impossible to identify with certainty the etiology of the elevated PAP levels. Consequently, additional evidence, even in patients with known CAP, should be present before an elevated PAP is attributed to metastatic CAP. Weight of Untrimmed Prostates (gm) ACKNOWLEDGMENTS Fig. 1. Semi-log of PAP levels vs. untrimmed prostate weights. tween PAP levels and prostate weight), 2) an untrimmed prostate weighing 116 g which also had a prostatic infarct (both prostate weight and prostatic infarcts  are correlated with PAP levels, 3) one case with lymphoma and another case with esophageal carcinoma with metastases (Foti et al.  reported that 10 of 37 males and 14 of 53 females with nonprostatic malignancies had elevated PAP levels, and 4) one case with both lymphoma and prostatic infarct (both findings associated with elevated PAP levels). Etiologies for the elevated PAP levels could not be identified with certainty in the remaining six cases. Specifically, indwelling bladder catheters were an unlikely source of elevated PAP levels since patients with and without indwelling catheters had similar PAP levels. Likewise, digital examinations of the prostate were an unlikely source of elevated PAP levels since few patients had digital examinations of the prostate within 1 week of death and none of these examinations were prostatic massages performed by urologists. Foti et al.  found that 12%,15%,29%,and 60% of men with Stage A, B, C, and D CAP had elevated PAP levels (enzymatic method). While PAP is often elevated in Stage D CAP (metastatic CAP) and may be elevated in Stage C CAP (extension beyond the prostatic capsule), the current study suggests that PAP cannot detect Stage A or Stage B CAP (organ confined CAP). Our finding of an 11% incidence of elevated PAP levels in patients with clinically benign prostates may indicate that the 12% and 15% incidences of elevated PAP levels reported by Foti et al.  for Stage A and Stage B CAP represent the nor- The authors thank the Dupont Co. for donating the reagents used in this study. REFERENCES 1. Brawn PN, SpeightsVO, Kuhl D, Riggs M, Spiekerman AM, McCord RG, Coffield KS, Stewart DT, Lind ML: Prostate-specific antigen levels from completely sectioned, clinically benign, whole prostates. Cancer 68: 1592-1599, 1991. 2. Brawn PN, Speights VO, Contin JU, Bayardo RJ, Kuhl DL: Atypical hyperplasia in prostates of 20-40 year old men. J Clin Pathol42:383-386, 1989. 3. Helpap B: The biological significanceof atypical hyperplasia of the prostate. Virchows Arch [A] 387307-317, 1980. 4. McNeal JE: Normal histology of the prostate. Am J Surg Pathol 12:619-633, 1988. 5. Richie AW, Layfield LJ, Turcillo P, deKernion J B The sigruficance of atypia in fine needle aspiration cytology of the prostate. J Urol 140:761-765, 1988. 6. Brawn PN: Adenosis of the prostate: A dysplastic lesion that can be confused with prostate carcinoma. Cancer 49826-833, 1982. 7. Gikas PW, Del Buono EA, Epstein JI: Florid hyperplasia of mesonephric remnants involving prostate and periprostatictissue. Am J Surg Pathol17454-460,1993. 8. Young RH, Clement PB: Sclerosing adenosis of the prostate. Arch Path01 Lab Med 111:363-366, 1987. 9. Ayala AG, Srigley JR, Ro JY,Abdul-Karim MFW, Johnson DE: Clear cell cribriform hyperplasia of the prostate. Am J Surg Pathol 10:665-671, 1986. 10. Bostwick DG,de la Roza G: Intraepithelialneoplasia: A call for standarized terminology. J Urol Pathol 1:95103, 1993. 11. Panteghini M, Pagani F, Bonora R Pre-analytical and biological variability of prostatic acid phosphatase and prostate specific antigen in serum from patients with prostate pathology. Eur J Clin Chem Clin Biochem 30: 135-139, 1992. 12. Foti AG, Herschman H, Cooper JF: Comparison of human prostatic acid phosphastase by measurement of PAP Levels in Whole Prostates enzymatic activity and by radioimmunoassay. Clin Chem 23:95-99, 1977. 13. Brawn PN,Foster DM, Jay DW, Kuhl D, Speights VO, Johnson EH, Coffield KS, Lind ML, Karl R, Weaver 8: Characteristics of prostatic infarcts and their effect on serum prostate-specific antigen and prostatic acid phosphatase Urology 4471-75, 1994. 299 14. Foti AG, Cooper JF, Herschman H, Malvaez RR: Detection of prostatic cancer by solid-phase radioimmunoassay of serum prostatic acid phosphatase. N Engl J Med 2971357-1361, 1977.