Original Paper

Urol Int 2003;71:150–153DOI: 10.1159/000071837Internationalis

Urologia

Detection of Prostate Cancer and Changes inProstate-Specific Antigen (PSA) Six Monthsafter Surgery for Benign Prostatic Hyperplasiain Patients with Elevated PSA

Cüneyt Özden Gürdal Inal Öztug Adsan Sertaç Yazıcı Bulent ÖzturkMesut Çetinkaya

Second Urology Clinic of Ankara Numune Education and Research Hospital, Ankara, Turkey

Dr. Cüneyt ÖzdenKucukesat Yaprak Sok. 25/4TR–6660 Ankara (Turkey)Tel. + 90 312 419 26 85, Fax +90 312 419 83 33E-Mail cuneytozden@hotmail.com

ABCFax + 41 61 306 12 34E-Mail karger@karger.chwww.karger.com

© 2003 S. Karger AG, Basel0042–1138/03/0712–0150$19.50/0

Accessible online at:www.karger.com/uin

Key WordsProstate cancer W High prostate-specific antigen W

Prostate biopsy W Benign prostatic hyperplasia

AbstractObjective: To evaluate early postoperative results ofpatients with elevated prostate-specific antigen (PSA)levels who underwent surgery due to benign prostatichyperplasia (BPH). Patients and Methods: 64 patientswho had lower urinary tract symptoms (LUTS), normaldigital rectal examinations (DRE), elevated PSA levelsand prostate biopsies reported as being benign patho-logically in specimens obtained by transrectal ultra-sound (TRUS)-guided biopsies, were included in thestudy. Patients were assessed in accordance with PSAdensity, free/total PSA ratio and uroflowmetric studies.Patients had no cancer pre- and postoperatively (accord-ing to operative specimens). Six months postopera-tively, 32 patients were accepted for re-evaluation for allPSA parameters, routine tests and prostatic biopsies.Results: 64 patients with a mean age of 66.8 (SD 6.72)were included in the study. Total PSA average value was14.38 (SD 7.49) ng/ml. Free PSA average value was 2.11(SD 1.43) ng/ml. Average PSA density and free/total PSA

ratio were 25.19 SD (14.12) ng/ml/cm3 and 14.53% (SD5.35%) respectively. 56 patients had BPH, 7 had chronicprostatitis and 1 had prostatic intraepithelial neoplasia(PIN) preoperatively with biopsies. Re-biopsy of the pa-tient with PIN was reported as BPH. In pathologic exami-nation with resected tissues postoperatively, 49 patientshad BPH, 14 had chronic prostatitis and 1 had PIN. In thesixth month, average values of free/total PSA were 0.45(SD 0.26) and 3.71 (SD 4.96) ng/ml respectively. AveragePSA density and free/total PSA ratio were 12.41 (SD 13.8)ng/ml/cm3 and 19.59% (SD 10.33%) respectively. Therewere significant decreases in PSA densities (p ! 0.001)and increases in free/total PSA ratios (p = 0.004). Sevenpatients still had elevated PSA levels 6 months postoper-atively. Three of 7 patients were reported as chronicprostatitis. One of them was indicated as prostatic carci-noma who was reported as PIN preoperatively. All otherpatients were stated as BPH at re-biopsies. Conclusion: Ifpretreatment biopsies are negative and operative speci-mens are also benign in patients with high PSA values,these patients can be followed up like usual BPH pa-tients, but long-term follow-up is still unclear.

Copyright © 2003 S. Karger AG, Basel

Detection of Prostate Cancer after Surgeryfor BPH in Patients with Elevated PSA

Urol Int 2003;71:150–153 151

Introduction

Prostate-specific antigen (PSA) was first demonstratedin prostatic tissue in 1970 and in seminal plasma in 1971.It was purified from prostatic tissue in 1979 and mea-sured in the serum of men in 1980. Since 1988, it has beenwidely used as a clinical marker of prostate cancer (PCa)[1].

Nowadays, PSA measurement in serum is the mostuseful test for distinguishing benign and malignant pros-tatic diseases [2]. Therefore, PSA is specific for prostategland; it is elevated in serum by benign prostatic hyper-plasia (BPH), prostatitis, prostatic infarct, prostatic trau-ma, prostatic manipulations and prostatic carcinoma [2].Although PSA values are elevated in most cases with pros-tatic carcinoma, it is !4 ng/ml in 12–30% organ-confinedcarcinomas and 14 ng/ml in 20–30% of patients withBPH [3]. However, the PSA assay is not a perfect test forPCa detection. Similar elevations of PSA can be found inmen with PCa or BPH, especially in the gray zone of totalserum PSA level (between 4 and 10 ng/ml). To improvethe specificity of PSA and to reduce the number of unnec-essary prostate biopsies, many derivations of serum PSAtests such as PSA density, age-specific PSA referenceranges, PSA velocity, transitional zone PSA density anddifferent molecular forms of PSA have been suggested.

Patients who have already undergone prostatic surgerydue to BPH need to be closely observed by the surgeon,because of the probability of PCa detection in the postop-erative period [4, 5].

The purpose of our study was to evaluate the earlypostoperative results of patients with elevated PSA levelswho underwent prostatic surgery due to BPH.

Patients and Methods

From March 1998 to March 2000, 64 Caucasian men of Turkishorigin who had LUTS were included in this prospective study. Allpatients came from the vicinity of Ankara, Turkey, and had allundergone evaluation and surgery at Numune Training Hospital. Asinclusion criteria they had severe LUTS, benign digital rectal exami-nation (DRE), PSA levels 14 ng/ml and benign pathologic examina-tion results. The patients were questioned for detailed history espe-cially about prostatic and/or urethral manipulations and medica-tions, which could affect serum PSA levels. Their LUTS complaintswere determined with International Prostate Symptom Scoring(IPSS). On DRE, nodule formation, hard nodule palpation, irregular-ity and fixation findings were considered as malignancy criteria.Routine biochemistry, blood counts and urine analyses were per-formed. Maximum urinary flow rates were measured by uroflowmet-ric studies with MMS-5000 and UD-2000 measurement systems.Serum free and total PSA values were measured by the monoclonal

Tandem R testing kit (Hybritech Inc., San Diego, Calif., USA). PSAcut-off value was accepted as 4 ng/ml.

The same physician performed TRUS-guided prostatic biopsieswith a Hitachi EUB-400 and 6.5-MHz biplane transrectal probe.During the TRUS procedure, prostatic morphologies were appraised,prostatic volumes were measured and PSA densities were calculated.At the same session, six-core sextant prostate biopsies were obtainedfrom all patients. Fifty-four of the 64 patients underwent TURPoperations and open prostatectomy operation was performed in theremaining 10 patients.

In the sixth postoperative month, 32 of 64 (50%) patients came tothe visit and were re-evaluated. Physical and rectal examinationswere performed. Serum free and total PSA values were determined,TRUS were performed and PSA densities were calculated. Afterobtaining ethical committee permission, even if PSA values were notelevated, all patients underwent another biopsy again.

Results were compared to preoperative values. The paired t testwas used for statistical analysis and the results are described with thestandard deviation (SD).

Results

The mean age of the patients was 66.8 (SD 6.72) years.Serum free and total PSA average values were 2.11 (SD1.43) and 14.38 (SD 7.49) ng/ml respectively. AveragePSA density and free/total PSA ratio were 25.19 (SD14.12) ng/ml/cm3 and 14.53% (SD 5.35%) respectively.Pathologic examinations of the prostate biopsies werereported as BPH in 56 patients (87.4%), chronic prostati-tis in 7 patients (10.9%) and prostatic intraepithelial neo-plasia (PIN) in 1 patient (1.5%). Repeated biopsy wasdone in a patient with PIN and was reported as BPH.

Postoperatively, pathologic examination of the re-sected tissues was reported as BPH in 49 patients (76.5%),chronic prostatitis in 14 patients (21.8%) and PIN in 1patient (1.5%). 32 patients were accepted for re-evalua-tion in the sixth postoperative month (mean 6.81, SD2.55 months). Serum free and total PSA average valueswere 0.45 (SD 0.26) and 3.71 (SD 4.96) ng/ml respective-ly. Average PSA density and free/total PSA ratio were cal-culated as 12.41 (SD 13.8) ng/ml/cm3 and 19.59% (SD10.33%) respectively for postoperative evaluation. Theaverage decrease in free and total PSA values after surgerywas 64% (SD 29.52%) and 71.34% (SD 31.08%) respec-tively. Those calculations were statistically different (p !0.001). We found statistically significant decreases in PSAdensities (p ! 0.001) and increases in free/total PSA ratios(p = 0.004) when postoperative results were comparedwith preoperative values (table 1).

Postoperatively, 7 of 32 patients still had PSA levels14 ng/ml. No patients had any abnormality in urine anal-ysis and in DRE. Three of 7 patients (9.3%) who had ele-

152 Urol Int 2003;71:150–153 Özden/Inal/Adsan/Yazıcı/Özturk/Çetinkaya

Table 1. Pre- and postoperative comparison of PSA parameters

n Total PSA Free PSA PSAD Free/total Decline ofng/ml ng/ml ng/ml/cm3 PSA, % PSA, %

Preoperative 32 13.50B6.09 1.84B1.06 23.97B13.15 13.50B4.79Postoperative 32 3.71B4.96 0.45B0.26 12.41B13.8 19.59B10.33 71.34B31.08p value !0.001 !0.001 !0.001 0.004

Table 2. Pathological assessments of the patients

n BPH (%) Chronicprostatitis (%)

PIN (%) PCa

Preoperative 64 56 (87.4) 7 (10.9) 1 (1.56) –Operative 64 49 (76.5) 14 (21.8) 1 (1.56) –Postoperative 32 28 (87.5) 3 (9.37) 1 (3.12)

vated PSA after surgery had chronic prostatitis; 1 (1.5%)had PIN in preoperative prostate biopsies and in theresected tissues. TRUS-guided biopsies of 32 patientswere performed in the sixth month postoperative periodand prostatic carcinoma was detected in 1 patient whosebiopsy was reported as PIN previously (table 2).

Discussion

PSA is the most useful marker for PCa and it is usedwidely with DRE for early detection of PCa [3]. With theuse of PSA values alone, 70% of patients with organ-con-fined PCa are detected, but with the use of DRE alone thisratio decreases to 30% [6]. Although PSA is accepted as aspecific marker for prostate gland, elevation of PSA is notsufficient enough to distinguish PCa from other benignprostatic diseases [2]. Approximately 25% of patientswith BPH have PSA levels 14 ng/ml [3].

Previous studies have stated that chronic prostatitisand enlarged prostatic volumes are the main reasons forelevated serum PSA levels in patients with BPH [7, 8].Bare et al. [9] indicated that the degree of elevation ofPSA in an inflamed prostate is between the levels in BPHand PCa. According to Irani et al. [10], there is a correla-tion between PSA elevation in inflammation and epithe-lial destruction in prostatic tissue. This can explain theelevation of PSA levels when we consider that inflamma-tion is found in 98% of patients with BPH. In our study,prostatitis was found in 10.9% of preoperative biopsies

and in 21.8% of postoperative resected tissues. This couldbe one of the reasons to explain the high PSA levels.

As shown previously, enlargement of the prostate isone of the main causes of PSA elevation [8]. Some param-eters of PSA were investigated to improve the sensitivityand specificity of PSA. Benson et al. [11] defined a con-cept of PSA density that is calculated by total PSA level/prostatic volume ratio and determined that lower degreesof density (!0.15) support BPH. Stenman et al. [12]defined that PSA bound to ·1-antichymotrypsin is foundin greater levels in serum than in the free form of PSA inpatients with PCa. Previous studies aim to prevent pa-tients in the gray zone undergoing prostatic biopsies byenhancing the accuracy of PSA as a marker of PCa.According to our findings, if we consider free/total PSAratio !0.10 and PSA density 10.15, 84.3 and 21.8% ofour patients would not undergo prostate biopsies respec-tively.

According to some retrospective studies, 3% of pa-tients develop PCa who had prostatic surgery due to BPHpreviously [4]. On the other hand, Schwartz et al. [13]reported no significant difference (2.9 vs. 3.2%) betweenthe group of patients that were operated due to BPH andthe control group. In our study, we found 1 patient withPCa in 32 patients (3.1%) although they all had elevatedPSA levels preoperatively. Postoperatively, 7 of 32 pa-tients (21.8%) still had high PSA levels (14 ng/ml) after6 months. One of them was reported to have PIN and thelast patient was reported as prostate carcinoma subse-quently in the postoperative biopsies in the sixth month.In cases with PIN, the probability of PCa at re-biopsy is33–100% [6]. We had a case in this study with PIN andreported as PCa subsequently. All other patients with highPSA levels had benign diseases according to pre- and post-operative pathologic examinations (table 2).

We established statistically significant decreases in se-rum free PSA, total PSA values and PSA densities and astatistically significant increase in free/total PSA ratioswhen we compared the results of 32 of 64 (50%) patients6 months postoperatively with preoperative values.

Detection of Prostate Cancer after Surgeryfor BPH in Patients with Elevated PSA

Urol Int 2003;71:150–153 153

Aus et al. [14] reported that the mean PSA value wasdecreased from 6 to 1.9 ng/ml in 70% of patients whounderwent TURP in the third month postoperatively.Marks et al. [15] established a decrease in mean PSA val-ues from 4.6 to 0.7 ng/ml 6 months postoperatively andthey also reported that PSA stayed steady for 5 years. Inour study, the mean decrease in PSA level was 71.34%(SD 31.08%) 6 months postoperatively.

According to our results 6 months postoperatively, wecan conclude that if pretreatment biopsies are negativeand also operative specimens are benign in patients withhigh PSA values, in spite of PCa suspicion, those patientscan be followed up like usual BPH patients, but long-termfollow-up results are still unclear.

References

1 Partin AW, Oesterling JE: The clinical useful-ness of prostate-specific antigen: Update 1994.J Urol 1994;152:1358–1368.

2 Oesterling JE: Prostate-specific antigen: A criti-cal assessment of the most useful tumor markerfor adenocarcinoma of the prostate. J Urol1991;145:907–923.

3 Arcangeli CG, Ornstein DK, Keetch DW, An-driole GL: Prostate-specific antigen as a screen-ing test for prostate cancer. The United Statesexperience. Urol Clin North Am 1997;2:299–306.

4 David GB, Williams HC, Louis D, George WJ,Peter TS, Gerald PM, et al: The association ofbenign prostatic hyperplasia and cancer of theprostate. Cancer 1992;70:291–301.

5 Wilfred SK, Thomas MS, Ian MT, et al: Thelong-term risk of development of prostate can-cer in patients with benign prostatic hyperpla-sia: Correlation with stage A1 disease. J Urol1993;150:1746–1748.

6 Brawer MK, Bigler SA, Sohlberg OE, et al: Sig-nificance of prostatic intraepithelial neoplasiaon prostate needle biopsy. Urology 1991;38:103–107.

7 Nadler RB, Humprey PA, Smith DS, et al:Effect of inflammation and benign prostatichyperplasia on elevated serum prostate-specificantigen levels. J Urol 1995;154:407–413.

8 Hammerer PG, McNeal JE, Stamey TA, et al:Correlation between serum prostate-specificantigen levels and the individual glandularzones of the human prostate. J Urol 1995;153:111–114.

9 Bare R, Hart L, McCullough DL: Correlationof prostate-specific antigen and prostate-spe-cific antigen density with outcome of prostatebiopsy. Urology 1994;43:191–196.

10 Irani J, Levillain P, Goujon JM, et al: Inflam-mation in benign prostatic hyperplasia: Corre-lation with prostate-specific antigen value. JUrol 1997;157:1301–1303.

11 Benson MC, Whang IS, Pantuk A, et al: Pros-tate-specific antigen density: A means of distin-guishing benign prostatic hypertrophy andprostate cancer. J Urol 1992;147:815–817.

12 Stenman UH, Hakama M, Knekt P, et al:Serum concentrations of prostate-specific anti-gen and its complex with ·1-antichymotrypsinbefore diagnosis of prostate cancer. Lancet1994;344:1594–1598.

13 Schwartz I, Wein JA, Malloy TR, et al: Prostat-ic cancer after prostatectomy for benign dis-ease. Cancer 1986;58:994–996.

14 Aus G, Bergdahl S, Frosing R, Lodding P, et al:Reference range of PSA after transurethral re-section of prostate. Urology 1996;47:529–531.

15 Marks LS, Dorey FJ, Rhods T, et al: SerumPSA levels after transurethral resection of pros-tate: A longitudinal characterization in menwith benign prostatic hyperplasia. J Urol 1996;156:1035–1039.