- Email: [email protected]
DIFFERENTIATION OF BENIGN PROSTATIC HYPERPLASIA FROM PROSTATE CANCER USING PROSTATE SPECIFIC ANTIGEN DYNAMIC PROFILE AFTER TRANSRECTAL PROSTATE BIOPSY
0022-5347/04/1716-2226/0 THE JOURNAL OF UROLOGY® Copyright © 2004 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 171, 2226 –2229, June 2004 Printed in U.S.A.
DOI: 10.1097/01.ju.0000123988.27122.cb
DIFFERENTIATION OF BENIGN PROSTATIC HYPERPLASIA FROM PROSTATE CANCER USING PROSTATE SPECIFIC ANTIGEN DYNAMIC PROFILE AFTER TRANSRECTAL PROSTATE BIOPSY TZU-PING LIN, WILLIAM J. S. HUANG* , †
AND
KUANG-KUO CHEN
From the Division of Urology, Department of Surgery, Taipei Veterans General Hospital and Department of Urology, National Yang-Ming University, School of Medicine, Taipei, Taiwan, Republic of China
ABSTRACT
Purpose: Serum prostate specific antigen (PSA) level is increased after needle biopsy (Bx) of the prostate. This study tested the hypothesis that the prostate harboring malignant lesions demonstrates less leakage of PSA after prostate biopsy and this phenomenon can be helpful in discriminating benign from cancer diagnosis. Materials and Methods: This prospective study was divided into 3 separate phases. Sextant prostate biopsy was done with transrectal ultrasound guidance, and the change and PSA values after biopsy were evaluated. Phases 1 and 2 had 20 and 41 patients, respectively. PSA dynamic profiles were plotted. We defined the most appropriate timing for blood sampling and the cutoff value of the PSA ratio (post-Bx total PSA-to-pre-Bx total PSA) to be applied for further assessment. Phase 3 recruited 97 cases, of which 66 satisfied the end point criteria in which a diagnosis such as malignancy, or 3 successive benign biopsies or benign transurethral resection of prostate was obtained. Results: From phases 1 and 2 the cutoff value for the PSA ratio was 2.0, and the timing of blood sampling was 60 minutes after the biopsy. In phase 3 of those whose PSA ratio was less than 2.0, 92.6% (25) had cancer. For those whose PSA ratio was greater than 2.0, 82.1% (32) were benign. Of the 4 cases with a PSA ratio less than 2.0 and an initial benign biopsy, 3 (75%) were proven to have cancer later. Conclusions: Evaluating the 1-hour PSA ratio might be helpful for clinicians to select the high risk patients who might have cancer in the prostate. Repeat biopsy should be suggested for cases with a smaller PSA ratio in spite of initial benign results. KEY WORDS: prostatic neoplasms, prostate-specific antigen, biopsy
Recently using prostate specific antigen (PSA) with related parameters free PSA, PSA density, PSA velocity and PSA acceleration enables urologists to diagnose cancer of prostate (CaP) at an earlier stage or to obtain better treatment outcome.1–3 However, PSA increase can also be noted in several situations such as benign prostatic hyperplasia (BPH), prostatic infection,4 prostate infarction and prostatic manipulation in prostatic massage or transrectal biopsy itself.5 The efficacy of transrectal ultrasound guided biopsy (TRUSBx) to detect CaP has been well illustrated in the literature.6, 7 The most popular indications for TRUS-Bx are an increased serum PSA or an abnormal digital rectal examination. However, in clinical practice there has been a significant number of negative biopsies according to these criteria. Effort has been made to increase specificity in diagnosing CaP. However, a 75% negative biopsy rate can be generally expected.1 Having a repeat biopsy is a tremendous psychological pressure for patients and physicians. It would be highly appreciated if a reliable guideline were available to indicate when to proceed and when to stop performing biopsies. Charrie et al first stated the phenomenon of significant increase of the serum prostatic acid phosphatase and PSA
after a prostate aspiration biopsy.8 The PSA increase was significantly greater in patients with BPH than in those with CaP. Furthermore, Stamey et al addressed the findings of serum PSA changes after transurethral resection of prostate (TURP).9 He described the initial increase in serum PSA after TURP as significantly higher in those with benign disease than in those with CaP.9 Recently Yuan et al5 and Ornstein et al10 reported that prostate biopsy caused a transient increase in free and total PSA. Besides, PSA after biopsy was higher in patients with adenoma than in patients with adenocarcinoma.8 Lechevallier et al established a PSA kinetic profile regarding levels after biopsy, and concluded that PSA increased significantly 1 hour after prostate biopsy and returned to baseline in 30 days.11 In particular free form PSA increased dramatically 1 hour after biopsy. It is interesting to evaluate whether we can use this characteristic of PSA differentiation to estimate the final pathology in patients undergoing TRUS-Bx.
MATERIALS AND METHODS
Patients were included in this prospective study due to an increase PSA (from 4 to 30 ng/ml) and/or an abnormal digital rectal examination. This study was divided into 3 phases. The phase 1 (pilot) study was from January to March 1996 and included a total of 20 patients. The purpose of this pilot study was to evaluate the dynamics of serum PSA levels after biopsy. After the first biopsy core was taken, serial serum PSA was examined at 5, 10, 30, 60, 180 and 1,080 minutes (fig. 1). All biopsies were performed with a spring-loaded, 18
Accepted for publication December 12, 2003. Presented at annual meeting of American Urological Association, New Orleans, Louisiana, April 12–17, 1997. * Correspondence: Division of Urology, Department of Surgery, Taipei Veterans General Hospital, 201, Sec. 2, Shih-pai Rd., Taipei, Taiwan 11217 TAIWAN Republic of China (telephone: ⫹886-228757519; FAX: ⫹886-2-28757540; e-mail: [email protected]). † Recipient of Grant VGH-89-98 from Taipei Veterans General Hospital. 2226
PROSTATE SPECIFIC ANTIGEN DYNAMICS AFTER BIOPSY TO SUSPECT PROSTATE CANCER
FIG. 1. PSA ratio change after prostate biopsy. Sampling time at 30 or 60 minutes is most discriminative between benign and malignant lesions. Asterisks indicate Mann-Whitney U test p ⬍0.001.
gauge Trucut biopsy needle (BPI Co., Berlin, Germany) under ultrasound guidance using a 7 MHz biplanar endorectal transducer (B&K Co., Herlev, Denmark). The peripheral zone was biopsied using random sextant technique and at least 6 cores of tissue were obtained as described by Hodge et al.12 Additional biopsy for the hypoechoic lesion was also taken when present. Serum PSA was measured with the Hybritech Tandem-R monoclonal radioimmunoassay (Hybritech, Inc., San Diego, California). In this phase we defined the most appropriate timing for blood sampling and the cutoff value of the PSA ratio (post-Bx total PSA to-pre-Bx total PSA) to be applied for further assessment. The phase 2 study was from September to December 1998 and 41 patients were enrolled. These patients had a symptom score of more than 20 and were willing to receive future TURP to relieve obstructive symptoms. In this phase we further confirmed the results from the phase 1 pilot study. Based on the findings of phases 1 and 2 we started a phase 3 survey to confirm further the validity of the PSA ratio. Another 97 patients were recruited from December 1998 to October 1999. In this phase when a benign result was obtained after biopsy the patient was advised to have another serum PSA evaluation 3 months later. If a followup PSA was also increased a further repeat biopsy was suggested. If the patient was symptomatic as well, a TURP was performed when indicated. A diagnosis end point of this phase was defined as either confirmed cancer or benign after 3 repeat prostate biopsies or after TURP. Statistical analysis was performed using commercially available computer software, SPSS version 8.0 (SPSS, Inc., Chicago, Illinois). The paired sample t test was used to compare the pre-Bx and post-Bx PSA between the benign and malignant groups, respectively.
2227
test the most significant time point to discriminate among the different diagnoses was 60 minutes after biopsy. Assessed with the receiver operating characteristics (ROC) method, the PSA ratio at 2.0 (doubling of pre-Bx level) was found to be an effective cutoff value (fig. 2).13 From phase 2 (41 patients), the initial pathological results revealed cancer in 20 patients. The other 21 with benign biopsies subsequently underwent TURP. On examining the TURP specimens, 2 more patients were confirmed to have cancer in the prostate. The PSA ratio was 1.09 and 1.92, respectively. The PSA dynamic change stratified by diagnosis was plotted in each case in figure 3. There were 19 benign cases (mean age 71.9) with mean PSA ratio 3.52. Only 1 patient had a PSA ratio less than 2.0 (1.48). The other 22 patients were diagnosed with cancer (mean age 73.5) and the mean PSA ratio was 1.67. In this group 3 patients had a PSA ratio greater than 2.0 (3.5, 3.7 and 4.2). Taking 2.0 as a PSA ratio cutoff sensitivity was 95%, specificity 85.7%, positive predictive value (PPV) was 86.3% and negative predictive value (NPV) was 94.7%, with p ⬍0.001 (chi-square test) for the 41 patients in the phase 2 study. The significance for this proposed cutoff was for the 61 patients in phases 1 and 2. Of the 97 cases in phase 3, 66 reached the end point of diagnosis. The 31 patients excluded from study were lost to followup after 1 (4) or 2 TRUS-Bx (27), or unwilling to have a third biopsy although it was indicated. However, there was no particular difference found in the demographic data between the 66 included and the 31 excluded cases. Of the 66 patients 20 had 3 negative biopsies, 23 had 2 negative biopsies then TURP and 23 had 1 negative biopsy then TURP. Of the 27 men whose PSA ratio was less than 2.0, 92.6% (25) were proven to have CaP. In the other 39 cases with a PSA ratio greater than 2.0, 82.1% (32) satisfied the criteria of a benign lesion (see table). The difference was statistically significant (chi-square test p ⫽ 0.046). Of the patients 10 had a prostatitis lesion in the biopsy tissue, but no particular difference was found in the PSA change compared with the BPH counterpart in this study. Basically the serum PSA dynamic follows a benign prototype. In the CaP group mean pre-Bx and 1-hour post-Bx PSA levels were 14.9 ng/ml (range 1.9 to 29) and 18.4 ng/ml (range 2.9 to 29), respectively. In the benign group mean pre-Bx and 1-hour post-Bx PSA levels were 9.1 ng/ml (range 1.6 to 24.5) and 49.4 ng/ml (range 5.7 to 681). The mean PSA ratio for the CaP group (32 patients) was 1.85 (range 0.50 to 7.82) while the mean PSA ratio of the benign group (34 patients) was 5.31 (range 1.00 to 60.37). Of the 66 patients 4 had an initial benign biopsy with a PSA ratio less than 2 and during followup 3 of them were found to have CaP (2 by TURP and the other by the third TRUS-Bx). For the 66 cases using 2.0 as a cutoff, sensitivity
RESULTS
From phase 1 of the study (20 patients) we found serum PSA had increased abruptly after the biopsy, and the peak value appeared as early as 5 minutes after the first core was taken. Interestingly, on analyzing the PSA dynamic curves 2 different patterns could be identified. Pattern 1 was characterized by an acute increase in serum PSA followed by a constant decline. Pattern 2 was a flat curve with only mild fluctuations of pre-Bx levels (fig. 1). Surprisingly all patients who demonstrated pattern 1 had benign pathological findings. However, of the patients with pattern 2 83% (5 of 6) were positive for malignancy. The other patient with pattern 2 and initial benign pathology was finally confirmed to have CaP after TURP 2 weeks later. Using the Mann-Whitney U
FIG. 2. ROC curve using multiple cutoff values of PSA ratio including 1.0, 1.5, 2.0, 2.5, 3.0, 3.5 and 4.0 in differentiating benign from malignant pathological diagnosis. Figure demonstrates that cutoff value at 2.0 offers largest area under ROC curve.
2228
PROSTATE SPECIFIC ANTIGEN DYNAMICS AFTER BIOPSY TO SUSPECT PROSTATE CANCER
FIG. 3. PSA value kinetic change of phase 1 and part of phase 2 cases after prostate biopsy.
Demographic data from phase 3 study Mean ⫾ SD
Pt age Baseline PSA (ng/ml) 1-Hr PSA (ng/ml)* Prostatic vol (ml) PSA ratio (60-min)† * PSA 1 hour after biopsy. † Ratio of PSA change 1 hour after
BPH Group
CaP Group
71.8 ⫾ 6.0 9.1 ⫾ 5.9 49.4 ⫾ 111.7 53.4 ⫾ 28.7 5.31
73.8 ⫾ 5.5 14.9 ⫾ 6.9 18.4 ⫾ 11.5 52.5 ⫾ 10.6 1.85
biopsy to baseline PSA.
was 78.1%, specificity 94.1%, PPV 92.5%, and NPV 82.1% (chi-square test p ⬍0.0001). DISCUSSION
Why needle biopsies entering cancerous tissue should liberate less PSA into the serum than needle biopsies entering BPH tissue is still unknown. We hypothesize that the cancerous tissue has been chronically leaking PSA, therefore, the clearance and conjugating mechanism has been on alert. For the benign tissue once the PSA is released, it will take more time to clear. Additionally, the amount of PSA production per unit tissue is higher in benign tissue than in cancerous tissue. The ratio of PSA increase helps to define risk for patients with only mild to moderately increased PSA. We included only patients with serum PSA less than 30 ng/ml in the study because we wished to establish a useful differentiation tool particularly helpful for those with PSA at this range. When the ratio is less than 2.0 it is rather specific to the diagnosis of cancer. In other words, if a patient had initial benign results a repeat biopsy or a TURP may be warranted. By controlling PSA density we found the PSA ratio was valid to differentiate between the 2 diagnoses. This finding is supported by the fact that of the 4 patients with ratios less than 2.0 and initial benign results, 3 (75%) were later diagnosed with CaP. However, for the patient with a ratio of less than 2 but with 3 consecutive negative biopsies a close followup protocol or TURP is still recommended because of the possibility of tumor in the transitional zone. On the other hand, there were 3 patients who had initial negative biopsy and PSA ratio greater than 2.0, then had cancer on the subsequent biopsy. Therefore, a further prostate biopsy is still recommended if clinically indicated. It should be emphasized that the PSA ratio evaluation is more a specific test than a sensitive one. If a low PSA ratio occurs
in any given patient the physician needs to pay extra attention to the possibility of tumor foci in the prostate. Although the current standard of biopsy techniques suggests more cores in each biopsy, we believe that using techniques to obtain more cores or adding cores from a transitional zone might improve the sensitivity of prostate biopsy. However, it might lower the specificity of the PSA ratio. It will be interesting to observe the role of the PSA ratio in the current practice scenario. For example, we would recommend another, more complete biopsy or a TURP biopsy in a patient with a completely adequate, 18-core benign biopsy and a PSA ratio that does not increase. Using 2.0 as a cutoff value in evaluating the PSA ratio after prostate biopsy, we obtained different statistical data from the phase 2 and phase 3 cohorts. Sensitivity decreased from 95% in phase 2 to 78% in phase 3, however, specificity increased from 85.7% to 94.1%. The PPV also increased from 86.3% to 92.5%, while the NPV decreased from 94.7% to 82.1%. At phase 2 all patients with initial benign biopsy were treated with TURP. At phase 3 only 70% of patients with initial negative biopsy were treated with TURP. Although all other cases remained negative after 3 repeat biopsies, it could be easily understood that a few cases of CaP could be detected at a 4th biopsy and so on. Many previous reports also suggest obtaining cases of more tissue cores from the transitional zone at repeat biopsy.14, 15 Therefore, defining pathological results based on TURP tissue would increase sensitivity and NPV. By the definition of this study, the PSA ratio was obtained only at the first prostate biopsy. We did not collect the entire PSA ratio at every repeat biopsy. Therefore, we cannot compare the difference among biopsies. Based on the available data (20 cases) for cases repeat biopsy, it is interesting to note that most had a constant dynamic pattern. However, it is necessary to conduct another prospective observation to verify the validity of this finding. CONCLUSIONS
In terms of applying these findings to office practice, it could be recommended that patients have a simultaneous PSA checkup 60 minutes after biopsy. Repeat biopsy might be indicated if a lower PSA ratio (less than 2.0) was obtained. However, a higher PSA ratio cannot guarantee permanent benignancy. Close followup is still necessary for higher risk cases. REFERENCES
1. Catalona, W. J., Partin, A. W., Slawin, K. M., Brawer, M. K., Flanigan, R. C., Patel, A. et al: Use of the percentage of free prostate-specific antigen to enhance differentiation of prostate cancer from benign prostatic disease: a prospective multicenter clinical trial. JAMA, 279: 1542, 1998 2. Djavan, B., Zlotta, A. R., Byttebier, G., Shariat, S., Omar, M., Schulman, C. C. et al: Prostate specific antigen density of the transition zone for early detection of prostate cancer. J Urol, 160: 411, 1998 3. Kamoi, K. and Babaian, R. J.: Advances in the application of prostate-specific antigen in the detection of early-stage prostate cancer. Semin Oncol, 26: 140, 1999 4. Nadler, R. B., Humphrey, P. A., Smith, D. S., Catalona, W. J. and Ratliff, T. L.: Effect of inflammation and benign prostatic hyperplasia on elevated serum prostate specific antigen levels. J Urol, 154: 407, 1995 5. Yuan, J. J. J., Coplen, D. E., Petros, J. A., Figenshau, R. S., Ratliff, T. L., Smith, D. S. et al: Effects of rectal examination, prostatic massage, ultrasonography and needle biopsy on serum prostate specific antigen levels. J Urol, 147: 810, 1992 6. Abe, M., Hashimoto, T., Matsuda, T., Saitoh, M. and Watanabe, H.: Prostatic biopsy guided by transrectal ultrasonography using real-time linear scanner. Urology, 29: 567, 1987 7. Devonec, M., Fendler, J. P., Monsallier, M., Mouriquand, P., Mestas, J. L., Dutrieux-Berger, N. et al: Prostatic cancer di-
PROSTATE SPECIFIC ANTIGEN DYNAMICS AFTER BIOPSY TO SUSPECT PROSTATE CANCER
8.
9. 10.
11.
agnosis with ultrasonically guided prostatic biopsy: results in 226 cases. Prog Clin Biol Res, 303: 35, 1989 Charrie, A., Fleury-Goyon, M. C., Dutey, P. and Perrin, P.: The effect of prostate aspiration biopsy on serum levels of prostatic acid phosphatase and prostate-specific antigen. JAMA, 256: 474, 1986 Stamey, T. A., Yang, N., Hay, A. R., McNeal, J. E., Freiha, F. S. and Redwine, E.: Prostate-specific antigen as a serum marker for adenocarcinoma of the prostate. N Engl J Med, 317: 909, 1987 Ornstein, D. K., Rao, G. S., Smith, D. S., Ratliff, T. L., Basler, J. W. and Catalona, W. J.: Effect of digital rectal examination and needle biopsy on serum total and percentage of free prostate specific antigen levels. J Urol, 157: 195, 1997 Lechevallier, E., Echazarian, C., Ortega, J. C., Daniel, L., Roux, F., Thirion, X. et al: Kinetics of postbiopsy levels of serum free prostate-specific antigen and percent free prostate-specific an-
2229
tigen. Urology, 53: 731, 1999 12. Hodge, K. K., McNeal, J. E., Terris, M. K. and Stamey, T. A.: Random systematic versus directed ultrasound guided transrectal core biopsies of the prostate. J Urol, 142: 71, 1989 13. Lin, H., Huang, W. J. S., Chen, K.-K. and Chang, L. S.: The immediate dynamic of serum PSA after transrectal prostate biopsy. J Urol, suppl., 157: 59, abstract 223, 1997 14. Mai, K. T., Moazin, M., Morash, C. and Collins, J. P.: Transitional zone and anterior peripheral zone of the prostate. A correlation of small-volume cancer in the biopsy cores and high PSA with positive anterior margins in radical prostatectomy specimens. Urol Int, 66: 191, 2001 15. Deliveliotis, C., Varkarakis, J., Albanis, S., Argyropoulos, V. and Skolarikos, A.: Biopsies of the transitional zone of the prostate. Should it be done on a routine basis, when and why? Urol Int, 68: 113, 2002
Vol. 171, 2226 –2229, June 2004 Printed in U.S.A.
DOI: 10.1097/01.ju.0000123988.27122.cb
DIFFERENTIATION OF BENIGN PROSTATIC HYPERPLASIA FROM PROSTATE CANCER USING PROSTATE SPECIFIC ANTIGEN DYNAMIC PROFILE AFTER TRANSRECTAL PROSTATE BIOPSY TZU-PING LIN, WILLIAM J. S. HUANG* , †
AND
KUANG-KUO CHEN
From the Division of Urology, Department of Surgery, Taipei Veterans General Hospital and Department of Urology, National Yang-Ming University, School of Medicine, Taipei, Taiwan, Republic of China
ABSTRACT
Purpose: Serum prostate specific antigen (PSA) level is increased after needle biopsy (Bx) of the prostate. This study tested the hypothesis that the prostate harboring malignant lesions demonstrates less leakage of PSA after prostate biopsy and this phenomenon can be helpful in discriminating benign from cancer diagnosis. Materials and Methods: This prospective study was divided into 3 separate phases. Sextant prostate biopsy was done with transrectal ultrasound guidance, and the change and PSA values after biopsy were evaluated. Phases 1 and 2 had 20 and 41 patients, respectively. PSA dynamic profiles were plotted. We defined the most appropriate timing for blood sampling and the cutoff value of the PSA ratio (post-Bx total PSA-to-pre-Bx total PSA) to be applied for further assessment. Phase 3 recruited 97 cases, of which 66 satisfied the end point criteria in which a diagnosis such as malignancy, or 3 successive benign biopsies or benign transurethral resection of prostate was obtained. Results: From phases 1 and 2 the cutoff value for the PSA ratio was 2.0, and the timing of blood sampling was 60 minutes after the biopsy. In phase 3 of those whose PSA ratio was less than 2.0, 92.6% (25) had cancer. For those whose PSA ratio was greater than 2.0, 82.1% (32) were benign. Of the 4 cases with a PSA ratio less than 2.0 and an initial benign biopsy, 3 (75%) were proven to have cancer later. Conclusions: Evaluating the 1-hour PSA ratio might be helpful for clinicians to select the high risk patients who might have cancer in the prostate. Repeat biopsy should be suggested for cases with a smaller PSA ratio in spite of initial benign results. KEY WORDS: prostatic neoplasms, prostate-specific antigen, biopsy
Recently using prostate specific antigen (PSA) with related parameters free PSA, PSA density, PSA velocity and PSA acceleration enables urologists to diagnose cancer of prostate (CaP) at an earlier stage or to obtain better treatment outcome.1–3 However, PSA increase can also be noted in several situations such as benign prostatic hyperplasia (BPH), prostatic infection,4 prostate infarction and prostatic manipulation in prostatic massage or transrectal biopsy itself.5 The efficacy of transrectal ultrasound guided biopsy (TRUSBx) to detect CaP has been well illustrated in the literature.6, 7 The most popular indications for TRUS-Bx are an increased serum PSA or an abnormal digital rectal examination. However, in clinical practice there has been a significant number of negative biopsies according to these criteria. Effort has been made to increase specificity in diagnosing CaP. However, a 75% negative biopsy rate can be generally expected.1 Having a repeat biopsy is a tremendous psychological pressure for patients and physicians. It would be highly appreciated if a reliable guideline were available to indicate when to proceed and when to stop performing biopsies. Charrie et al first stated the phenomenon of significant increase of the serum prostatic acid phosphatase and PSA
after a prostate aspiration biopsy.8 The PSA increase was significantly greater in patients with BPH than in those with CaP. Furthermore, Stamey et al addressed the findings of serum PSA changes after transurethral resection of prostate (TURP).9 He described the initial increase in serum PSA after TURP as significantly higher in those with benign disease than in those with CaP.9 Recently Yuan et al5 and Ornstein et al10 reported that prostate biopsy caused a transient increase in free and total PSA. Besides, PSA after biopsy was higher in patients with adenoma than in patients with adenocarcinoma.8 Lechevallier et al established a PSA kinetic profile regarding levels after biopsy, and concluded that PSA increased significantly 1 hour after prostate biopsy and returned to baseline in 30 days.11 In particular free form PSA increased dramatically 1 hour after biopsy. It is interesting to evaluate whether we can use this characteristic of PSA differentiation to estimate the final pathology in patients undergoing TRUS-Bx.
MATERIALS AND METHODS
Patients were included in this prospective study due to an increase PSA (from 4 to 30 ng/ml) and/or an abnormal digital rectal examination. This study was divided into 3 phases. The phase 1 (pilot) study was from January to March 1996 and included a total of 20 patients. The purpose of this pilot study was to evaluate the dynamics of serum PSA levels after biopsy. After the first biopsy core was taken, serial serum PSA was examined at 5, 10, 30, 60, 180 and 1,080 minutes (fig. 1). All biopsies were performed with a spring-loaded, 18
Accepted for publication December 12, 2003. Presented at annual meeting of American Urological Association, New Orleans, Louisiana, April 12–17, 1997. * Correspondence: Division of Urology, Department of Surgery, Taipei Veterans General Hospital, 201, Sec. 2, Shih-pai Rd., Taipei, Taiwan 11217 TAIWAN Republic of China (telephone: ⫹886-228757519; FAX: ⫹886-2-28757540; e-mail: [email protected]). † Recipient of Grant VGH-89-98 from Taipei Veterans General Hospital. 2226
PROSTATE SPECIFIC ANTIGEN DYNAMICS AFTER BIOPSY TO SUSPECT PROSTATE CANCER
FIG. 1. PSA ratio change after prostate biopsy. Sampling time at 30 or 60 minutes is most discriminative between benign and malignant lesions. Asterisks indicate Mann-Whitney U test p ⬍0.001.
gauge Trucut biopsy needle (BPI Co., Berlin, Germany) under ultrasound guidance using a 7 MHz biplanar endorectal transducer (B&K Co., Herlev, Denmark). The peripheral zone was biopsied using random sextant technique and at least 6 cores of tissue were obtained as described by Hodge et al.12 Additional biopsy for the hypoechoic lesion was also taken when present. Serum PSA was measured with the Hybritech Tandem-R monoclonal radioimmunoassay (Hybritech, Inc., San Diego, California). In this phase we defined the most appropriate timing for blood sampling and the cutoff value of the PSA ratio (post-Bx total PSA to-pre-Bx total PSA) to be applied for further assessment. The phase 2 study was from September to December 1998 and 41 patients were enrolled. These patients had a symptom score of more than 20 and were willing to receive future TURP to relieve obstructive symptoms. In this phase we further confirmed the results from the phase 1 pilot study. Based on the findings of phases 1 and 2 we started a phase 3 survey to confirm further the validity of the PSA ratio. Another 97 patients were recruited from December 1998 to October 1999. In this phase when a benign result was obtained after biopsy the patient was advised to have another serum PSA evaluation 3 months later. If a followup PSA was also increased a further repeat biopsy was suggested. If the patient was symptomatic as well, a TURP was performed when indicated. A diagnosis end point of this phase was defined as either confirmed cancer or benign after 3 repeat prostate biopsies or after TURP. Statistical analysis was performed using commercially available computer software, SPSS version 8.0 (SPSS, Inc., Chicago, Illinois). The paired sample t test was used to compare the pre-Bx and post-Bx PSA between the benign and malignant groups, respectively.
2227
test the most significant time point to discriminate among the different diagnoses was 60 minutes after biopsy. Assessed with the receiver operating characteristics (ROC) method, the PSA ratio at 2.0 (doubling of pre-Bx level) was found to be an effective cutoff value (fig. 2).13 From phase 2 (41 patients), the initial pathological results revealed cancer in 20 patients. The other 21 with benign biopsies subsequently underwent TURP. On examining the TURP specimens, 2 more patients were confirmed to have cancer in the prostate. The PSA ratio was 1.09 and 1.92, respectively. The PSA dynamic change stratified by diagnosis was plotted in each case in figure 3. There were 19 benign cases (mean age 71.9) with mean PSA ratio 3.52. Only 1 patient had a PSA ratio less than 2.0 (1.48). The other 22 patients were diagnosed with cancer (mean age 73.5) and the mean PSA ratio was 1.67. In this group 3 patients had a PSA ratio greater than 2.0 (3.5, 3.7 and 4.2). Taking 2.0 as a PSA ratio cutoff sensitivity was 95%, specificity 85.7%, positive predictive value (PPV) was 86.3% and negative predictive value (NPV) was 94.7%, with p ⬍0.001 (chi-square test) for the 41 patients in the phase 2 study. The significance for this proposed cutoff was for the 61 patients in phases 1 and 2. Of the 97 cases in phase 3, 66 reached the end point of diagnosis. The 31 patients excluded from study were lost to followup after 1 (4) or 2 TRUS-Bx (27), or unwilling to have a third biopsy although it was indicated. However, there was no particular difference found in the demographic data between the 66 included and the 31 excluded cases. Of the 66 patients 20 had 3 negative biopsies, 23 had 2 negative biopsies then TURP and 23 had 1 negative biopsy then TURP. Of the 27 men whose PSA ratio was less than 2.0, 92.6% (25) were proven to have CaP. In the other 39 cases with a PSA ratio greater than 2.0, 82.1% (32) satisfied the criteria of a benign lesion (see table). The difference was statistically significant (chi-square test p ⫽ 0.046). Of the patients 10 had a prostatitis lesion in the biopsy tissue, but no particular difference was found in the PSA change compared with the BPH counterpart in this study. Basically the serum PSA dynamic follows a benign prototype. In the CaP group mean pre-Bx and 1-hour post-Bx PSA levels were 14.9 ng/ml (range 1.9 to 29) and 18.4 ng/ml (range 2.9 to 29), respectively. In the benign group mean pre-Bx and 1-hour post-Bx PSA levels were 9.1 ng/ml (range 1.6 to 24.5) and 49.4 ng/ml (range 5.7 to 681). The mean PSA ratio for the CaP group (32 patients) was 1.85 (range 0.50 to 7.82) while the mean PSA ratio of the benign group (34 patients) was 5.31 (range 1.00 to 60.37). Of the 66 patients 4 had an initial benign biopsy with a PSA ratio less than 2 and during followup 3 of them were found to have CaP (2 by TURP and the other by the third TRUS-Bx). For the 66 cases using 2.0 as a cutoff, sensitivity
RESULTS
From phase 1 of the study (20 patients) we found serum PSA had increased abruptly after the biopsy, and the peak value appeared as early as 5 minutes after the first core was taken. Interestingly, on analyzing the PSA dynamic curves 2 different patterns could be identified. Pattern 1 was characterized by an acute increase in serum PSA followed by a constant decline. Pattern 2 was a flat curve with only mild fluctuations of pre-Bx levels (fig. 1). Surprisingly all patients who demonstrated pattern 1 had benign pathological findings. However, of the patients with pattern 2 83% (5 of 6) were positive for malignancy. The other patient with pattern 2 and initial benign pathology was finally confirmed to have CaP after TURP 2 weeks later. Using the Mann-Whitney U
FIG. 2. ROC curve using multiple cutoff values of PSA ratio including 1.0, 1.5, 2.0, 2.5, 3.0, 3.5 and 4.0 in differentiating benign from malignant pathological diagnosis. Figure demonstrates that cutoff value at 2.0 offers largest area under ROC curve.
2228
PROSTATE SPECIFIC ANTIGEN DYNAMICS AFTER BIOPSY TO SUSPECT PROSTATE CANCER
FIG. 3. PSA value kinetic change of phase 1 and part of phase 2 cases after prostate biopsy.
Demographic data from phase 3 study Mean ⫾ SD
Pt age Baseline PSA (ng/ml) 1-Hr PSA (ng/ml)* Prostatic vol (ml) PSA ratio (60-min)† * PSA 1 hour after biopsy. † Ratio of PSA change 1 hour after
BPH Group
CaP Group
71.8 ⫾ 6.0 9.1 ⫾ 5.9 49.4 ⫾ 111.7 53.4 ⫾ 28.7 5.31
73.8 ⫾ 5.5 14.9 ⫾ 6.9 18.4 ⫾ 11.5 52.5 ⫾ 10.6 1.85
biopsy to baseline PSA.
was 78.1%, specificity 94.1%, PPV 92.5%, and NPV 82.1% (chi-square test p ⬍0.0001). DISCUSSION
Why needle biopsies entering cancerous tissue should liberate less PSA into the serum than needle biopsies entering BPH tissue is still unknown. We hypothesize that the cancerous tissue has been chronically leaking PSA, therefore, the clearance and conjugating mechanism has been on alert. For the benign tissue once the PSA is released, it will take more time to clear. Additionally, the amount of PSA production per unit tissue is higher in benign tissue than in cancerous tissue. The ratio of PSA increase helps to define risk for patients with only mild to moderately increased PSA. We included only patients with serum PSA less than 30 ng/ml in the study because we wished to establish a useful differentiation tool particularly helpful for those with PSA at this range. When the ratio is less than 2.0 it is rather specific to the diagnosis of cancer. In other words, if a patient had initial benign results a repeat biopsy or a TURP may be warranted. By controlling PSA density we found the PSA ratio was valid to differentiate between the 2 diagnoses. This finding is supported by the fact that of the 4 patients with ratios less than 2.0 and initial benign results, 3 (75%) were later diagnosed with CaP. However, for the patient with a ratio of less than 2 but with 3 consecutive negative biopsies a close followup protocol or TURP is still recommended because of the possibility of tumor in the transitional zone. On the other hand, there were 3 patients who had initial negative biopsy and PSA ratio greater than 2.0, then had cancer on the subsequent biopsy. Therefore, a further prostate biopsy is still recommended if clinically indicated. It should be emphasized that the PSA ratio evaluation is more a specific test than a sensitive one. If a low PSA ratio occurs
in any given patient the physician needs to pay extra attention to the possibility of tumor foci in the prostate. Although the current standard of biopsy techniques suggests more cores in each biopsy, we believe that using techniques to obtain more cores or adding cores from a transitional zone might improve the sensitivity of prostate biopsy. However, it might lower the specificity of the PSA ratio. It will be interesting to observe the role of the PSA ratio in the current practice scenario. For example, we would recommend another, more complete biopsy or a TURP biopsy in a patient with a completely adequate, 18-core benign biopsy and a PSA ratio that does not increase. Using 2.0 as a cutoff value in evaluating the PSA ratio after prostate biopsy, we obtained different statistical data from the phase 2 and phase 3 cohorts. Sensitivity decreased from 95% in phase 2 to 78% in phase 3, however, specificity increased from 85.7% to 94.1%. The PPV also increased from 86.3% to 92.5%, while the NPV decreased from 94.7% to 82.1%. At phase 2 all patients with initial benign biopsy were treated with TURP. At phase 3 only 70% of patients with initial negative biopsy were treated with TURP. Although all other cases remained negative after 3 repeat biopsies, it could be easily understood that a few cases of CaP could be detected at a 4th biopsy and so on. Many previous reports also suggest obtaining cases of more tissue cores from the transitional zone at repeat biopsy.14, 15 Therefore, defining pathological results based on TURP tissue would increase sensitivity and NPV. By the definition of this study, the PSA ratio was obtained only at the first prostate biopsy. We did not collect the entire PSA ratio at every repeat biopsy. Therefore, we cannot compare the difference among biopsies. Based on the available data (20 cases) for cases repeat biopsy, it is interesting to note that most had a constant dynamic pattern. However, it is necessary to conduct another prospective observation to verify the validity of this finding. CONCLUSIONS
In terms of applying these findings to office practice, it could be recommended that patients have a simultaneous PSA checkup 60 minutes after biopsy. Repeat biopsy might be indicated if a lower PSA ratio (less than 2.0) was obtained. However, a higher PSA ratio cannot guarantee permanent benignancy. Close followup is still necessary for higher risk cases. REFERENCES
1. Catalona, W. J., Partin, A. W., Slawin, K. M., Brawer, M. K., Flanigan, R. C., Patel, A. et al: Use of the percentage of free prostate-specific antigen to enhance differentiation of prostate cancer from benign prostatic disease: a prospective multicenter clinical trial. JAMA, 279: 1542, 1998 2. Djavan, B., Zlotta, A. R., Byttebier, G., Shariat, S., Omar, M., Schulman, C. C. et al: Prostate specific antigen density of the transition zone for early detection of prostate cancer. J Urol, 160: 411, 1998 3. Kamoi, K. and Babaian, R. J.: Advances in the application of prostate-specific antigen in the detection of early-stage prostate cancer. Semin Oncol, 26: 140, 1999 4. Nadler, R. B., Humphrey, P. A., Smith, D. S., Catalona, W. J. and Ratliff, T. L.: Effect of inflammation and benign prostatic hyperplasia on elevated serum prostate specific antigen levels. J Urol, 154: 407, 1995 5. Yuan, J. J. J., Coplen, D. E., Petros, J. A., Figenshau, R. S., Ratliff, T. L., Smith, D. S. et al: Effects of rectal examination, prostatic massage, ultrasonography and needle biopsy on serum prostate specific antigen levels. J Urol, 147: 810, 1992 6. Abe, M., Hashimoto, T., Matsuda, T., Saitoh, M. and Watanabe, H.: Prostatic biopsy guided by transrectal ultrasonography using real-time linear scanner. Urology, 29: 567, 1987 7. Devonec, M., Fendler, J. P., Monsallier, M., Mouriquand, P., Mestas, J. L., Dutrieux-Berger, N. et al: Prostatic cancer di-
PROSTATE SPECIFIC ANTIGEN DYNAMICS AFTER BIOPSY TO SUSPECT PROSTATE CANCER
8.
9. 10.
11.
agnosis with ultrasonically guided prostatic biopsy: results in 226 cases. Prog Clin Biol Res, 303: 35, 1989 Charrie, A., Fleury-Goyon, M. C., Dutey, P. and Perrin, P.: The effect of prostate aspiration biopsy on serum levels of prostatic acid phosphatase and prostate-specific antigen. JAMA, 256: 474, 1986 Stamey, T. A., Yang, N., Hay, A. R., McNeal, J. E., Freiha, F. S. and Redwine, E.: Prostate-specific antigen as a serum marker for adenocarcinoma of the prostate. N Engl J Med, 317: 909, 1987 Ornstein, D. K., Rao, G. S., Smith, D. S., Ratliff, T. L., Basler, J. W. and Catalona, W. J.: Effect of digital rectal examination and needle biopsy on serum total and percentage of free prostate specific antigen levels. J Urol, 157: 195, 1997 Lechevallier, E., Echazarian, C., Ortega, J. C., Daniel, L., Roux, F., Thirion, X. et al: Kinetics of postbiopsy levels of serum free prostate-specific antigen and percent free prostate-specific an-
2229
tigen. Urology, 53: 731, 1999 12. Hodge, K. K., McNeal, J. E., Terris, M. K. and Stamey, T. A.: Random systematic versus directed ultrasound guided transrectal core biopsies of the prostate. J Urol, 142: 71, 1989 13. Lin, H., Huang, W. J. S., Chen, K.-K. and Chang, L. S.: The immediate dynamic of serum PSA after transrectal prostate biopsy. J Urol, suppl., 157: 59, abstract 223, 1997 14. Mai, K. T., Moazin, M., Morash, C. and Collins, J. P.: Transitional zone and anterior peripheral zone of the prostate. A correlation of small-volume cancer in the biopsy cores and high PSA with positive anterior margins in radical prostatectomy specimens. Urol Int, 66: 191, 2001 15. Deliveliotis, C., Varkarakis, J., Albanis, S., Argyropoulos, V. and Skolarikos, A.: Biopsies of the transitional zone of the prostate. Should it be done on a routine basis, when and why? Urol Int, 68: 113, 2002