- Email: [email protected]
Saturation Technique Does Not Improve Cancer Detection as an Initial Prostate Biopsy Strategy
Oncology: Prostate/Testis/Penis/Urethra Saturation Technique Does Not Improve Cancer Detection as an Initial Prostate Biopsy Strategy J. Stephen Jones,*,† Amit Patel, Lynn Schoenfield, John C. Rabets, Craig D. Zippe‡ and Cristina Magi-Galluzzi From the Glickman Urological Institute, the Cleveland Clinic Foundation, Cleveland, Ohio
Purpose: We reported on the results of a sequential cohort study comparing office based saturation prostate biopsy to traditional 10-core sampling as an initial biopsy. Materials and Methods: Based on improved cancer detection of office based saturation prostate biopsy repeat biopsy, we adopted the technique as an initial biopsy strategy to improve cancer detection. Two surgeons performed 24-core saturation prostate biopsies in 139 patients undergoing initial biopsy under periprostatic local anesthesia. Indication for biopsy was an increased PSA of 2.5 ng/dl or greater in all patients. Results were compared to those of 87 patients who had previously undergone 10-core initial biopsies. Results: Cancer was detected in 62 of 139 patients (44.6%) who underwent saturation biopsy and in 45 of 87 patients (51.7%) who underwent 10-core biopsy (p ⬎0.9). Breakdown by PSA level failed to show benefit to the saturation technique for any degree PSA increase. Men with PSA 2.5 to 9.9 ng/dl were found to have cancer in 53 of 122 (43.4%) saturation biopsies and 26 of 58 (44.8%) 10-core biopsies. Complications included 3 cases of prostatitis in each group. Rectal bleeding was troublesome enough to require evaluation only in 3 men in the saturation group and 1 in the 10-core group. Conclusions: Although saturation prostate biopsy improves cancer detection in men with suspicion of cancer following a negative biopsy, it does not appear to offer benefit as an initial biopsy technique. These findings suggest that further efforts at extended biopsy strategies beyond 10 to 12 cores are not appropriate as an initial biopsy strategy. Key Words: prostate, biopsy, ambulatory surgical procedures, prostatic neoplasms
a repeat biopsy strategy,9 we adopted its use as an initial biopsy strategy to enhance cancer detection in a single setting. We now report on the prostate cancer detection rate in men with PSA 2.5 ng/dl or greater.
andom systematic ultrasound guided transrectal prostate biopsy has become the mainstay in the diagnosis of prostate cancer.1 Studies have demonstrated that a traditional sextant technique may miss substantial numbers of cancers and that additional sampling of the lateral peripheral zone may increase the diagnostic yield.2,3 Despite such modifications and strategies using 10 to 14 cores, the false-negative rate remains substantial.4 As a result saturation biopsy has been adopted in several centers resulting in cancer detection rates approaching a third when extended biopsy schemes of 14 to 45 cores were used even following multiple negative biopsies.5 This procedure was initially performed in the operating room. The application of local anesthesia has greatly reduced the pain and discomfort associated with prostate biopsy.6,7 Our group demonstrated that even saturation biopsy can be performed routinely in the office with the patient under local anesthesia.8 With our experience that cancer detection was improved with saturation biopsy as
R
MATERIALS AND METHODS As part of an ongoing institutional review board approved prostate biopsy database, 409 consecutive patients underwent prostate biopsy from September 2001 through June 2004. Patients were counseled to have biopsy only if they had a predicted life expectancy of at least 10 years or if symptomatic and palliative therapy would be appropriate if this were not the case. Excluding from analysis men who had previously undergone biopsy and men with PSA less than 2.5 ng/dl, 226 patients were available for evaluation of initial biopsy strategy. Patient characteristics are shown in table 1. Before August 2003 our initial biopsy technique was to perform a 10-core laterally focused biopsy as described by Gore et al.10 These 87 patients compromise group 2. Around that time we reported improved cancer detection during repeat biopsy using an office based saturation biopsy strategy with of the patient under local anesthesia. As a result of this successful endeavor we began performing 24-core biopsy as an initial biopsy technique. These 139 men are the patients of interest for this report and constitute group 1. Informed consent was obtained from all patients. No enemas were used. Each patient received a single dose of fluoroquinolone before biopsy, and based on surgeon prefer-
Submitted for publication January 6, 2005. Study received institutional review board approval. * Correspondence: Urological Institute, the Cleveland Clinic Foundation, 9500 Euclid Ave., St. A100, Cleveland, Ohio 44195 (telephone: 216-839-3666; FAX: 216-839-3660; e-mail: [email protected]). † Financial interest and/or other relationship with Pfizer, Cook, Photocure, Endocare and Abbott. ‡ Financial interest and/or other relationship with Pfizer, AstraZeneca, TAP and Lilly-Icos.
See Editorial on page 419.
0022-5347/06/1752-0485/0 THE JOURNAL OF UROLOGY® Copyright © 2006 by AMERICAN UROLOGICAL ASSOCIATION
485
Vol. 175, 485-488, February 2006 Printed in U.S.A. DOI:10.1016/S0022-5347(05)00211-9
486
SATURATION TECHNIQUE NOT ADVANTAGEOUS FOR INITIAL PROSTATE BIOPSY TABLE 1. Patient characteristics
No. pts Mean age (range) Abnormal digital rectal examination
Group 1 (saturation biopsy)
Group 2 (10-core biopsy)
139 63 (38–89) 19
87 63 (43–85) 7
No statistically significant differences.
ence a small number (less than 20) received additional doses empirically. All men were instructed to discontinue anticoagulants 5 days before biopsy but biopsy ensued if they were taking aspirin or nonsteroidal anti-inflammatory drugs. Patients were placed in the left lateral decubitus position and the ultrasound probe was inserted transrectally. A 22 gauge 7 inch spinal needle (Becton Dickinson, Franklin Lakes, New Jersey) was placed at the base of the prostate where the prostatic sensory nerves enter the prostate. This area was identified laterally where fat in the notch between prostate and seminal vesicle creates a white hyperechoic pyramid we call the “Mount Everest sign.”11 The needle was directed into this notch and 10 cc of a local anesthetic was injected to infiltrate from prostatic base to apex on each side. Successful placement was confirmed by observing the injectate cause separation of the seminal vesicles and prostate from the rectal wall. Ultrasound examination and volume calculations were then performed. A total of 24 biopsies were taken without delay using a spring-loaded biopsy gun. Biopsies were evenly distributed throughout the prostate and included any visible abnormalities. The 5 sectors (and number of cores) biopsied on each side for group 1 were lateral base (2), lateral mid zone (3), apex (3), parasagittal mid zone (2) and parasagittal base (2) as shown in the figure. Care was taken to assure that the 3 apical biopsies adequately sampled the anterior horn of peripheral zone tissue. Group 2 had 1 core taken from each of the same 5 sectors. A board certified pathologist interpreted all slides. The uropathology team reviewed any equivocal results during regularly scheduled sessions. Statistics were recorded on an Excel® spreadsheet and reviewed on an intermittent basis, at least once per quarter. Any complications were noted and addressed to assure patient safety.
statistically significant, detection rates were actually higher in the 10-core biopsy group. The likelihood of cancer did not appear to be affected by digital rectal examination findings in this population. The breakdown by PSA level is shown in table 2 and Gleason scores are listed in table 3. Symptoms of prostatitis including fever developed in 3 patients in each group. Two patients in the 10-core group required hospitalization and recovered with a few days of intravenous antibiotic therapy. One of these men had undergone cystoscopy for bladder cancer surveillance on the same day, and also became septic (despite broad-spectrum antibiotic prophylaxis) following a subsequent cystoscopy a year later for bladder tumor surveillance as well, suggesting a propensity for infection. Rectal bleeding was common, but troublesome enough to require evaluation only in 3 men in the saturation group and 1 in the 10-core group. Two men, including the 1 from group 2, were observed expectantly as outpatients and bleeding stopped spontaneously on the day of the biopsy. A tampon was placed intrarectally in a second man in group 1, but he passed it almost immediately and bleeding stopped at that point. One man was admitted elsewhere and underwent transfusion by the internist for unclear indications of a hematocrit reported to be in the range of 30%. None of these men was taking any anticoagulant. A total of 19 men in group 1 and 15 men in group 2 underwent radical retropubic prostatectomy. No residual cancer was identified on the final surgical specimen from 1 patient in each group. Acute prostatitis had occurred either before (1) or following biopsy (1) in each case. The original diagnosis was confirmed correct in each by pathological review of the biopsy specimen. Upon examination of the specimen in men who underwent radical prostatectomy, cancer was clinically significant by published definitions12 in 16 of 19 patients in group 1 (84.2%) and 14 of 15 men in group 2 (93.3%). These differences were not statistically significant, and include the single patient in each group whose biopsy showed clear
RESULTS All patients tolerated complete biopsies obtaining the planned numbers of cores. All but 1 man in the 10-core group experienced no significant pain or discomfort. The exception was a man in whom acute prostatitis developed and who required hospitalization. Cancer was detected in 62 of 139 patients (44.6%) who underwent saturation biopsy and in 45 of 87 patients (51.7%) who underwent 10-core biopsy (p ⬎0.9). Breakdown by PSA level failed to show benefit to the saturation technique for any degree PSA increase. In PSA range 2.5 to 3.9 ng/dl, 6 of 21 (28.6%) men had cancer during saturation as did 7 of 18 (38.9%) during 10-core biopsy. In PSA range 4.0 to 9.9 ng/dl cancer was found in 47 of 101 patients who underwent saturation biopsy (46.5%) and 19 of 40 men who underwent 10-core biopsy (47.5%). Therefore, men with PSA 2.5 to 9.9 ng/dl were found to have cancer in 53 of 122 (43.4%) saturation biopsies and 26 of 58 (44.8%) 10-core biopsies. Although the differences were not
Template showing location of cores obtained during saturation biopsy. Men in group 2 (10-core biopsies) had 1 biopsy core obtained from each of same sectors.
SATURATION TECHNIQUE NOT ADVANTAGEOUS FOR INITIAL PROSTATE BIOPSY TABLE 2. Results No./Total No. (%) PSA (ng/dl)
Group 1 (24-core saturation biopsy)
Group 2 (10-core biopsy)
2.5–3.9 4.0–9.9 2.5–9.9 Greater than 10.0 All
6/21 (28.6) 47/101 (46.5) 53/122 (43.4) 10/17 (59.9) 62/139 (44.6)
7/18 (38.9) 19/40 (47.5) 26/58 (44.8) 19/29 (65.5) 45/87 (51.7)
No statistically significant difference for any level PSA.
evidence of Gleason score 6 adenocarcinoma but final pathological examination of the resected specimen failed to identify the cancer. DISCUSSION Sextant biopsy has been clearly shown to have an unacceptably high false-negative rate.4 Multiple authors have shown increased cancer detection in the range of approximately 40% for men with PSA between 4 and 10 ng/dl using 10 to 12-core biopsy schemes. One recent report of 107 patients showed a 49% detection rate in those with PSA between 4.0 and 10.0 ng/dl,13 consistent with our detection rate of 44.8% in patients who underwent 10-core biopsy. We recently demonstrated that office based saturation prostate biopsy increases the detection rate of prostate cancer in patients with suspicious clinical findings following negative standard biopsy. Based on our findings that even such extended initial biopsy schemes miss almost a third of cancers, we attempted to improve the detection rate using a saturation technique for initial biopsy. In contrast to our theory that this practice would increase cancer detection, we found that increasing the number of cores to 24 does not yield a higher detection rate. In fact, our detection rate was less than our own experience using 10 cores, although this difference was not statistically significant. One explanation could be that this population was tested during a later time when screening might have decreased the total number of cancers in the study population. This seems unlikely as it immediately followed the comparison group. Based on our recent data for patients with 10core biopsy, we chose to compare results in these patients to a sequential cohort in lieu of randomization. These data cannot justify randomization in further study. We did not investigate use of more than 24 cores, but the trend observed makes it unlikely that doing so would change our findings. Regardless, there is clearly no increased detection rate in our patient population using 24 cores. The falsenegative rate remains undefined. We believe a cancer small enough to avoid detection by a 24-core technique would be unlikely to be clinically significant. Moreover, we focused on the lateral and anteroapical areas known to be the most likely site of malignancy. These patients will be followed to determine how many are eventually found to have cancer. The rates of clinically insignificant cancer detection from both groups are consistent with the literature using fewer cores,14 suggesting that most cancers detected as a result of increased PSA are significant regardless of biopsy scheme used. There were more clinically insignificant cancers in the saturation group, but the difference was not statistically significant.
487
The cases in which no cancer was identified on final pathological examination of radical retropubic prostatectomy specimens were concerning, although this occurred in a single case using both techniques. The original biopsy slides and the resected specimen were reexamined by at least 2 board certified pathologists, and the diagnosis was agreed upon in each circumstance. Goldstein et al described “vanishing cancer syndrome” for this phenomenon.15 This has recently been reported to occur in less than 1% of radical prostatectomy specimens.16 We believe the most likely explanation to be that a small cancer can be overlooked even by a team of experienced urological pathologists. Its prognostic significance is reportedly similar to any clinically insignificant cancer. Our urological pathology team reviewed these 2 cases and took our standard approach that the effort required for an exhaustive search to find such a small tumor was an unwise use of resources, and would not ultimately change disease management. We agree with this assessment. A concern regarding biopsy schemes obtaining such widespread sampling is the possibility of overdiagnosis. The question is whether cancers detected in such schemes are clinically significant or potentially harmful. If not, it might be preferable that the patients never learn of their existence to avoid the costs and risks of unnecessary treatment. The concern of overdetection must be weighed against the risk of clinically significant malignancy being missed by inadequate biopsy strategies that have considerable false-negative rates. We believe the best approach to balance these risks at present involves the concept espoused by Carroll, “unlinking detection and treatment, as they are separate processes.”17 This series demonstrates our commitment to this concept, as more than a fourth of the men with cancer in each group are currently being treated with watchful waiting/active surveillance (table 4). Many cases of low volume, low grade disease may be managed expectantly, and treatment can be safely delayed until clinical indicators such as sequential PSA, repeat biopsy or physical examination indicates disease progression and the need for deferred intervention.18 If this approach is accepted for men whose biopsies suggest clinically insignificant disease, careful observation will usually allow detection of progression while still curable. This report shows that the complication rate with 24 cores is consistent with published rates for biopsy schemes using fewer cores, although there was a trend (not statistically significant) for increased bleeding compared to what we experienced in our 10-core biopsy population. There appeared to be no greater risk of infection. However, even if there is no increase in complications, the use of saturation biopsy cannot be justified as an initial biopsy technique based on these data. If is feasible that a larger series might reveal a small advantage to saturation biopsy as an initial biopsy strategy, but our findings in this pilot study make that appear unlikely and preclude us from further exploration of the concept. TABLE 3. Gleason score (%) No./Total No. (%)
Gleason Gleason Gleason Gleason
6 7 8 9
Group 1 (24-core saturation biopsy)
Group 2 (10-core biopsy)
39/62 (62.9) 18/62 (29.0) 3/62 (4.8) 2/62 (3.2)
27/45 (60.0) 13/45 (28.9) 1/45 (2.2) 4/45 (8.8)
488
SATURATION TECHNIQUE NOT ADVANTAGEOUS FOR INITIAL PROSTATE BIOPSY TABLE 4. Treatments chosen
Radical prostatectomy External beam radiation Brachytherapy Hormonal ablation Watchful waiting or no decision made yet
Group 1 (24-core saturation biopsy)
Group 2 (10-core biopsy)
19 7 17 1 18
15 2 10 4 14
Based on these findings combined with those of our previous report on the use of saturation biopsy as a repeat biopsy strategy, we believe the literature supports use of a 10 to 12-core laterally based biopsy technique for initial biopsy and an office based saturation biopsy for subsequent biopsies. Our saturation technique obtains 20 laterally based cores with a focus on the apex and anterior horn based on our experience that we have never identified cancer on a medial biopsy in the absence of a positive lateral core (and no longer separate the cores based on this).19 Due to our observation that the false-negative rate approaches 30% even for extended biopsy schemes involving 10 to 12 cores, we recommend repeat biopsy within the following year for men who have persistence of suspicion for prostate cancer either due to persistently increased PSA or palpable abnormality. The timing during that year is individualized based on the suspicion of the physician and the input of the patient. The presence of high grade prostatic intraepithelial neoplasia on initial biopsy does not lead to the automatic recommendation of immediate repeat biopsy based on reports showing that the likelihood of finding cancer based on this indication alone is low, but we recommend that men with high grade prostatic intraepithelial neoplasia undergo repeat (office based saturation) biopsy 3 years after the initial biopsy regardless of PSA as recommended by Lefkowitz et al.20 Our experience has been that a positive biopsy following a negative repeat saturation biopsy is rare and has occurred in less than 5% of men who underwent subsequent repeat saturation biopsy due to rapidly increasing PSA values. However, until we gather enough experience with this population, we continue to biopsy on a judgment basis if PSA increases more than 20% from a previous saturation biopsy level. CONCLUSIONS Although office based saturation prostate biopsy improves cancer detection in men with suspicion of cancer following a negative biopsy, it does not appear to offer benefit as an initial biopsy technique. These findings suggest that further efforts at extended biopsy strategies beyond 10 to 12 cores are not appropriate as an initial biopsy strategy. We believe the literature supports our recommendation for a 10 to 12core laterally based biopsy technique for initial biopsy and office based saturation biopsy for subsequent biopsies.
Abbreviations and Acronyms PSA ⫽ prostate specific antigen REFERENCES 1. 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
2. Keetch, D. W., Catalona, W. J. and Smith, D. S.: Serial prostate biopsies in men with persistently elevated serum prostate specific antigen values. J Urol, 151: 1571, 1994 3. Epstein, J. I., Walsh, P. C. and Carter, H. B.: Importance of posterolateral needle biopsies in the detection of prostate cancer. Urology, 57: 1112, 2001 4. Applewhite, J. C., Matagla, B. R. and McCullough, D. L.: Results of the 5 region prostate biopsy method: the repeat biopsy population. J Urol, 168: 500, 2002 5. Borboroglu, P. G., Comer, S. W., Riffenburgh, R. H. and Amling, C. L.: Extensive repeat transrectal ultrasound guided prostate biopsy in patients with previous benign sextant biopsies. J Urol, 163: 158, 2000 6. Nash, P. A., Bruce, J. E., Indudhara, R. and Shinohara, K.: Transrectal ultrasound guided prostatic nerve blockade eases systematic needle biopsy of the prostate. J Urol, 155: 607, 1996 7. Soloway, M. S. and Obek, C.: Periprostatic local anesthesia before ultrasound guided prostate biopsy. J Urol, 163: 172, 2000 8. Jones, J. S., Oder, M. and Zippe, C. D.: Saturation biopsy with periprostatic block can be performed in the office. J Urol, 168: 2108, 2002 9. Rabets, J. C., Jones, J. S., Patel, A. and Zippe, C. D.: Prostate cancer detection with office based saturation biopsy in a repeat biopsy population. J Urol, 172: 94, 2004 10. Gore, J. L., Shariat, S. F., Miles, B. J., Kadmon, D., Jiang, N., Wheeler, T. M. et al: Optimal combination of systematic sextant and laterally directed biopsies for the detection of prostate cancer. J Urol, 165: 1554, 2001 11. Jones, J. S., Ulchaker, J. C., Nelson, D., Kursh, E. D., Kitay, R., Angie, S. et al: Periprostatic local anesthesia eliminates pain of office-based transrectal prostate biopsy. Prostate Cancer Prostatic Dis, 6: 53, 2003 12. Jack, G. S., Cookson, M. S., Coffey, C. S., Vader, V., Roberts, R. L., Chang, S. S. et al: Pathological parameters of radical prostatectomy for clinical stages T1c versus T2 prostate adenocarcinoma decreased pathological stage and increased detection of transition zone tumors. J Urol, 168: 519, 2002 13. Emiliozzi, P., Scarpone, P., DePaula, F., Pizzo, M., Federico, G., Pansadoro, A. et al: The incidence of prostate cancer in men with prostate specific antigen greater than 4.0 ng/ml: a randomized study of 6 versus 12 core transperineal prostate biopsy. J Urol, 171: 197, 2004 14. Chan, T. Y., Chan, D. Y., Stutzman, K. L. R. E. and Epstein, J. I.: Does increased needle biopsy sampling of the prostate detect a higher number of potentially insignificant tumors? J Urol, 166: 2181, 2001 15. Goldstein, N. S., Begin, L. R., Grody, W. W., Novak, J. M., Qian, J. and Bostwick, D. G.: Minimal or no cancer in radical prostatectomy specimens. Report of 13 cases of the “vanishing cancer phenomenon”. Am J Surg Pathol, 19: 1002, 1995 16. Herkommer, K., Kuefer, R., Gschwend, J. E., Hautmann, R. E. and Volkmer, B. G.: Pathological T0 prostate cancer without neoadjuvant therapy clinical presentation and follow-up. Eur Urol, 45: 36, 2004 17. Carroll, P. R.: Early stage prostate cancer— do we have a problem with over-detection, overtreatment or both? J Urol, 173: 1061, 2005 18. Patel, M. I., DeConcini, D. T., Lopez-Corona, E., Ohori, M., Wheeler, T. and Scardino, P. T.: An analysis of men with clinically localized prostate cancer who deferred definitive therapy. J Urol, 171: 1520, 2004 19. Patel, A. R., Jones, J. S., Rabets, J., DeOreo, G. and Zippe, C. D.: Parasagittal biopsies add minimal information in repeat saturation prostate biopsy. Urology, 63: 87, 2004 20. Lefkowitz, G. K., Sidhu, G. S., Torre, P., Lepor, H. and Taneja, S. S.: Is repeat prostate biopsy for high-grade prostatic intraepithelial neoplasia necessary after routine 12-core sampling? Urology, 58: 999, 2001
Purpose: We reported on the results of a sequential cohort study comparing office based saturation prostate biopsy to traditional 10-core sampling as an initial biopsy. Materials and Methods: Based on improved cancer detection of office based saturation prostate biopsy repeat biopsy, we adopted the technique as an initial biopsy strategy to improve cancer detection. Two surgeons performed 24-core saturation prostate biopsies in 139 patients undergoing initial biopsy under periprostatic local anesthesia. Indication for biopsy was an increased PSA of 2.5 ng/dl or greater in all patients. Results were compared to those of 87 patients who had previously undergone 10-core initial biopsies. Results: Cancer was detected in 62 of 139 patients (44.6%) who underwent saturation biopsy and in 45 of 87 patients (51.7%) who underwent 10-core biopsy (p ⬎0.9). Breakdown by PSA level failed to show benefit to the saturation technique for any degree PSA increase. Men with PSA 2.5 to 9.9 ng/dl were found to have cancer in 53 of 122 (43.4%) saturation biopsies and 26 of 58 (44.8%) 10-core biopsies. Complications included 3 cases of prostatitis in each group. Rectal bleeding was troublesome enough to require evaluation only in 3 men in the saturation group and 1 in the 10-core group. Conclusions: Although saturation prostate biopsy improves cancer detection in men with suspicion of cancer following a negative biopsy, it does not appear to offer benefit as an initial biopsy technique. These findings suggest that further efforts at extended biopsy strategies beyond 10 to 12 cores are not appropriate as an initial biopsy strategy. Key Words: prostate, biopsy, ambulatory surgical procedures, prostatic neoplasms
a repeat biopsy strategy,9 we adopted its use as an initial biopsy strategy to enhance cancer detection in a single setting. We now report on the prostate cancer detection rate in men with PSA 2.5 ng/dl or greater.
andom systematic ultrasound guided transrectal prostate biopsy has become the mainstay in the diagnosis of prostate cancer.1 Studies have demonstrated that a traditional sextant technique may miss substantial numbers of cancers and that additional sampling of the lateral peripheral zone may increase the diagnostic yield.2,3 Despite such modifications and strategies using 10 to 14 cores, the false-negative rate remains substantial.4 As a result saturation biopsy has been adopted in several centers resulting in cancer detection rates approaching a third when extended biopsy schemes of 14 to 45 cores were used even following multiple negative biopsies.5 This procedure was initially performed in the operating room. The application of local anesthesia has greatly reduced the pain and discomfort associated with prostate biopsy.6,7 Our group demonstrated that even saturation biopsy can be performed routinely in the office with the patient under local anesthesia.8 With our experience that cancer detection was improved with saturation biopsy as
R
MATERIALS AND METHODS As part of an ongoing institutional review board approved prostate biopsy database, 409 consecutive patients underwent prostate biopsy from September 2001 through June 2004. Patients were counseled to have biopsy only if they had a predicted life expectancy of at least 10 years or if symptomatic and palliative therapy would be appropriate if this were not the case. Excluding from analysis men who had previously undergone biopsy and men with PSA less than 2.5 ng/dl, 226 patients were available for evaluation of initial biopsy strategy. Patient characteristics are shown in table 1. Before August 2003 our initial biopsy technique was to perform a 10-core laterally focused biopsy as described by Gore et al.10 These 87 patients compromise group 2. Around that time we reported improved cancer detection during repeat biopsy using an office based saturation biopsy strategy with of the patient under local anesthesia. As a result of this successful endeavor we began performing 24-core biopsy as an initial biopsy technique. These 139 men are the patients of interest for this report and constitute group 1. Informed consent was obtained from all patients. No enemas were used. Each patient received a single dose of fluoroquinolone before biopsy, and based on surgeon prefer-
Submitted for publication January 6, 2005. Study received institutional review board approval. * Correspondence: Urological Institute, the Cleveland Clinic Foundation, 9500 Euclid Ave., St. A100, Cleveland, Ohio 44195 (telephone: 216-839-3666; FAX: 216-839-3660; e-mail: [email protected]). † Financial interest and/or other relationship with Pfizer, Cook, Photocure, Endocare and Abbott. ‡ Financial interest and/or other relationship with Pfizer, AstraZeneca, TAP and Lilly-Icos.
See Editorial on page 419.
0022-5347/06/1752-0485/0 THE JOURNAL OF UROLOGY® Copyright © 2006 by AMERICAN UROLOGICAL ASSOCIATION
485
Vol. 175, 485-488, February 2006 Printed in U.S.A. DOI:10.1016/S0022-5347(05)00211-9
486
SATURATION TECHNIQUE NOT ADVANTAGEOUS FOR INITIAL PROSTATE BIOPSY TABLE 1. Patient characteristics
No. pts Mean age (range) Abnormal digital rectal examination
Group 1 (saturation biopsy)
Group 2 (10-core biopsy)
139 63 (38–89) 19
87 63 (43–85) 7
No statistically significant differences.
ence a small number (less than 20) received additional doses empirically. All men were instructed to discontinue anticoagulants 5 days before biopsy but biopsy ensued if they were taking aspirin or nonsteroidal anti-inflammatory drugs. Patients were placed in the left lateral decubitus position and the ultrasound probe was inserted transrectally. A 22 gauge 7 inch spinal needle (Becton Dickinson, Franklin Lakes, New Jersey) was placed at the base of the prostate where the prostatic sensory nerves enter the prostate. This area was identified laterally where fat in the notch between prostate and seminal vesicle creates a white hyperechoic pyramid we call the “Mount Everest sign.”11 The needle was directed into this notch and 10 cc of a local anesthetic was injected to infiltrate from prostatic base to apex on each side. Successful placement was confirmed by observing the injectate cause separation of the seminal vesicles and prostate from the rectal wall. Ultrasound examination and volume calculations were then performed. A total of 24 biopsies were taken without delay using a spring-loaded biopsy gun. Biopsies were evenly distributed throughout the prostate and included any visible abnormalities. The 5 sectors (and number of cores) biopsied on each side for group 1 were lateral base (2), lateral mid zone (3), apex (3), parasagittal mid zone (2) and parasagittal base (2) as shown in the figure. Care was taken to assure that the 3 apical biopsies adequately sampled the anterior horn of peripheral zone tissue. Group 2 had 1 core taken from each of the same 5 sectors. A board certified pathologist interpreted all slides. The uropathology team reviewed any equivocal results during regularly scheduled sessions. Statistics were recorded on an Excel® spreadsheet and reviewed on an intermittent basis, at least once per quarter. Any complications were noted and addressed to assure patient safety.
statistically significant, detection rates were actually higher in the 10-core biopsy group. The likelihood of cancer did not appear to be affected by digital rectal examination findings in this population. The breakdown by PSA level is shown in table 2 and Gleason scores are listed in table 3. Symptoms of prostatitis including fever developed in 3 patients in each group. Two patients in the 10-core group required hospitalization and recovered with a few days of intravenous antibiotic therapy. One of these men had undergone cystoscopy for bladder cancer surveillance on the same day, and also became septic (despite broad-spectrum antibiotic prophylaxis) following a subsequent cystoscopy a year later for bladder tumor surveillance as well, suggesting a propensity for infection. Rectal bleeding was common, but troublesome enough to require evaluation only in 3 men in the saturation group and 1 in the 10-core group. Two men, including the 1 from group 2, were observed expectantly as outpatients and bleeding stopped spontaneously on the day of the biopsy. A tampon was placed intrarectally in a second man in group 1, but he passed it almost immediately and bleeding stopped at that point. One man was admitted elsewhere and underwent transfusion by the internist for unclear indications of a hematocrit reported to be in the range of 30%. None of these men was taking any anticoagulant. A total of 19 men in group 1 and 15 men in group 2 underwent radical retropubic prostatectomy. No residual cancer was identified on the final surgical specimen from 1 patient in each group. Acute prostatitis had occurred either before (1) or following biopsy (1) in each case. The original diagnosis was confirmed correct in each by pathological review of the biopsy specimen. Upon examination of the specimen in men who underwent radical prostatectomy, cancer was clinically significant by published definitions12 in 16 of 19 patients in group 1 (84.2%) and 14 of 15 men in group 2 (93.3%). These differences were not statistically significant, and include the single patient in each group whose biopsy showed clear
RESULTS All patients tolerated complete biopsies obtaining the planned numbers of cores. All but 1 man in the 10-core group experienced no significant pain or discomfort. The exception was a man in whom acute prostatitis developed and who required hospitalization. Cancer was detected in 62 of 139 patients (44.6%) who underwent saturation biopsy and in 45 of 87 patients (51.7%) who underwent 10-core biopsy (p ⬎0.9). Breakdown by PSA level failed to show benefit to the saturation technique for any degree PSA increase. In PSA range 2.5 to 3.9 ng/dl, 6 of 21 (28.6%) men had cancer during saturation as did 7 of 18 (38.9%) during 10-core biopsy. In PSA range 4.0 to 9.9 ng/dl cancer was found in 47 of 101 patients who underwent saturation biopsy (46.5%) and 19 of 40 men who underwent 10-core biopsy (47.5%). Therefore, men with PSA 2.5 to 9.9 ng/dl were found to have cancer in 53 of 122 (43.4%) saturation biopsies and 26 of 58 (44.8%) 10-core biopsies. Although the differences were not
Template showing location of cores obtained during saturation biopsy. Men in group 2 (10-core biopsies) had 1 biopsy core obtained from each of same sectors.
SATURATION TECHNIQUE NOT ADVANTAGEOUS FOR INITIAL PROSTATE BIOPSY TABLE 2. Results No./Total No. (%) PSA (ng/dl)
Group 1 (24-core saturation biopsy)
Group 2 (10-core biopsy)
2.5–3.9 4.0–9.9 2.5–9.9 Greater than 10.0 All
6/21 (28.6) 47/101 (46.5) 53/122 (43.4) 10/17 (59.9) 62/139 (44.6)
7/18 (38.9) 19/40 (47.5) 26/58 (44.8) 19/29 (65.5) 45/87 (51.7)
No statistically significant difference for any level PSA.
evidence of Gleason score 6 adenocarcinoma but final pathological examination of the resected specimen failed to identify the cancer. DISCUSSION Sextant biopsy has been clearly shown to have an unacceptably high false-negative rate.4 Multiple authors have shown increased cancer detection in the range of approximately 40% for men with PSA between 4 and 10 ng/dl using 10 to 12-core biopsy schemes. One recent report of 107 patients showed a 49% detection rate in those with PSA between 4.0 and 10.0 ng/dl,13 consistent with our detection rate of 44.8% in patients who underwent 10-core biopsy. We recently demonstrated that office based saturation prostate biopsy increases the detection rate of prostate cancer in patients with suspicious clinical findings following negative standard biopsy. Based on our findings that even such extended initial biopsy schemes miss almost a third of cancers, we attempted to improve the detection rate using a saturation technique for initial biopsy. In contrast to our theory that this practice would increase cancer detection, we found that increasing the number of cores to 24 does not yield a higher detection rate. In fact, our detection rate was less than our own experience using 10 cores, although this difference was not statistically significant. One explanation could be that this population was tested during a later time when screening might have decreased the total number of cancers in the study population. This seems unlikely as it immediately followed the comparison group. Based on our recent data for patients with 10core biopsy, we chose to compare results in these patients to a sequential cohort in lieu of randomization. These data cannot justify randomization in further study. We did not investigate use of more than 24 cores, but the trend observed makes it unlikely that doing so would change our findings. Regardless, there is clearly no increased detection rate in our patient population using 24 cores. The falsenegative rate remains undefined. We believe a cancer small enough to avoid detection by a 24-core technique would be unlikely to be clinically significant. Moreover, we focused on the lateral and anteroapical areas known to be the most likely site of malignancy. These patients will be followed to determine how many are eventually found to have cancer. The rates of clinically insignificant cancer detection from both groups are consistent with the literature using fewer cores,14 suggesting that most cancers detected as a result of increased PSA are significant regardless of biopsy scheme used. There were more clinically insignificant cancers in the saturation group, but the difference was not statistically significant.
487
The cases in which no cancer was identified on final pathological examination of radical retropubic prostatectomy specimens were concerning, although this occurred in a single case using both techniques. The original biopsy slides and the resected specimen were reexamined by at least 2 board certified pathologists, and the diagnosis was agreed upon in each circumstance. Goldstein et al described “vanishing cancer syndrome” for this phenomenon.15 This has recently been reported to occur in less than 1% of radical prostatectomy specimens.16 We believe the most likely explanation to be that a small cancer can be overlooked even by a team of experienced urological pathologists. Its prognostic significance is reportedly similar to any clinically insignificant cancer. Our urological pathology team reviewed these 2 cases and took our standard approach that the effort required for an exhaustive search to find such a small tumor was an unwise use of resources, and would not ultimately change disease management. We agree with this assessment. A concern regarding biopsy schemes obtaining such widespread sampling is the possibility of overdiagnosis. The question is whether cancers detected in such schemes are clinically significant or potentially harmful. If not, it might be preferable that the patients never learn of their existence to avoid the costs and risks of unnecessary treatment. The concern of overdetection must be weighed against the risk of clinically significant malignancy being missed by inadequate biopsy strategies that have considerable false-negative rates. We believe the best approach to balance these risks at present involves the concept espoused by Carroll, “unlinking detection and treatment, as they are separate processes.”17 This series demonstrates our commitment to this concept, as more than a fourth of the men with cancer in each group are currently being treated with watchful waiting/active surveillance (table 4). Many cases of low volume, low grade disease may be managed expectantly, and treatment can be safely delayed until clinical indicators such as sequential PSA, repeat biopsy or physical examination indicates disease progression and the need for deferred intervention.18 If this approach is accepted for men whose biopsies suggest clinically insignificant disease, careful observation will usually allow detection of progression while still curable. This report shows that the complication rate with 24 cores is consistent with published rates for biopsy schemes using fewer cores, although there was a trend (not statistically significant) for increased bleeding compared to what we experienced in our 10-core biopsy population. There appeared to be no greater risk of infection. However, even if there is no increase in complications, the use of saturation biopsy cannot be justified as an initial biopsy technique based on these data. If is feasible that a larger series might reveal a small advantage to saturation biopsy as an initial biopsy strategy, but our findings in this pilot study make that appear unlikely and preclude us from further exploration of the concept. TABLE 3. Gleason score (%) No./Total No. (%)
Gleason Gleason Gleason Gleason
6 7 8 9
Group 1 (24-core saturation biopsy)
Group 2 (10-core biopsy)
39/62 (62.9) 18/62 (29.0) 3/62 (4.8) 2/62 (3.2)
27/45 (60.0) 13/45 (28.9) 1/45 (2.2) 4/45 (8.8)
488
SATURATION TECHNIQUE NOT ADVANTAGEOUS FOR INITIAL PROSTATE BIOPSY TABLE 4. Treatments chosen
Radical prostatectomy External beam radiation Brachytherapy Hormonal ablation Watchful waiting or no decision made yet
Group 1 (24-core saturation biopsy)
Group 2 (10-core biopsy)
19 7 17 1 18
15 2 10 4 14
Based on these findings combined with those of our previous report on the use of saturation biopsy as a repeat biopsy strategy, we believe the literature supports use of a 10 to 12-core laterally based biopsy technique for initial biopsy and an office based saturation biopsy for subsequent biopsies. Our saturation technique obtains 20 laterally based cores with a focus on the apex and anterior horn based on our experience that we have never identified cancer on a medial biopsy in the absence of a positive lateral core (and no longer separate the cores based on this).19 Due to our observation that the false-negative rate approaches 30% even for extended biopsy schemes involving 10 to 12 cores, we recommend repeat biopsy within the following year for men who have persistence of suspicion for prostate cancer either due to persistently increased PSA or palpable abnormality. The timing during that year is individualized based on the suspicion of the physician and the input of the patient. The presence of high grade prostatic intraepithelial neoplasia on initial biopsy does not lead to the automatic recommendation of immediate repeat biopsy based on reports showing that the likelihood of finding cancer based on this indication alone is low, but we recommend that men with high grade prostatic intraepithelial neoplasia undergo repeat (office based saturation) biopsy 3 years after the initial biopsy regardless of PSA as recommended by Lefkowitz et al.20 Our experience has been that a positive biopsy following a negative repeat saturation biopsy is rare and has occurred in less than 5% of men who underwent subsequent repeat saturation biopsy due to rapidly increasing PSA values. However, until we gather enough experience with this population, we continue to biopsy on a judgment basis if PSA increases more than 20% from a previous saturation biopsy level. CONCLUSIONS Although office based saturation prostate biopsy improves cancer detection in men with suspicion of cancer following a negative biopsy, it does not appear to offer benefit as an initial biopsy technique. These findings suggest that further efforts at extended biopsy strategies beyond 10 to 12 cores are not appropriate as an initial biopsy strategy. We believe the literature supports our recommendation for a 10 to 12core laterally based biopsy technique for initial biopsy and office based saturation biopsy for subsequent biopsies.
Abbreviations and Acronyms PSA ⫽ prostate specific antigen REFERENCES 1. 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
2. Keetch, D. W., Catalona, W. J. and Smith, D. S.: Serial prostate biopsies in men with persistently elevated serum prostate specific antigen values. J Urol, 151: 1571, 1994 3. Epstein, J. I., Walsh, P. C. and Carter, H. B.: Importance of posterolateral needle biopsies in the detection of prostate cancer. Urology, 57: 1112, 2001 4. Applewhite, J. C., Matagla, B. R. and McCullough, D. L.: Results of the 5 region prostate biopsy method: the repeat biopsy population. J Urol, 168: 500, 2002 5. Borboroglu, P. G., Comer, S. W., Riffenburgh, R. H. and Amling, C. L.: Extensive repeat transrectal ultrasound guided prostate biopsy in patients with previous benign sextant biopsies. J Urol, 163: 158, 2000 6. Nash, P. A., Bruce, J. E., Indudhara, R. and Shinohara, K.: Transrectal ultrasound guided prostatic nerve blockade eases systematic needle biopsy of the prostate. J Urol, 155: 607, 1996 7. Soloway, M. S. and Obek, C.: Periprostatic local anesthesia before ultrasound guided prostate biopsy. J Urol, 163: 172, 2000 8. Jones, J. S., Oder, M. and Zippe, C. D.: Saturation biopsy with periprostatic block can be performed in the office. J Urol, 168: 2108, 2002 9. Rabets, J. C., Jones, J. S., Patel, A. and Zippe, C. D.: Prostate cancer detection with office based saturation biopsy in a repeat biopsy population. J Urol, 172: 94, 2004 10. Gore, J. L., Shariat, S. F., Miles, B. J., Kadmon, D., Jiang, N., Wheeler, T. M. et al: Optimal combination of systematic sextant and laterally directed biopsies for the detection of prostate cancer. J Urol, 165: 1554, 2001 11. Jones, J. S., Ulchaker, J. C., Nelson, D., Kursh, E. D., Kitay, R., Angie, S. et al: Periprostatic local anesthesia eliminates pain of office-based transrectal prostate biopsy. Prostate Cancer Prostatic Dis, 6: 53, 2003 12. Jack, G. S., Cookson, M. S., Coffey, C. S., Vader, V., Roberts, R. L., Chang, S. S. et al: Pathological parameters of radical prostatectomy for clinical stages T1c versus T2 prostate adenocarcinoma decreased pathological stage and increased detection of transition zone tumors. J Urol, 168: 519, 2002 13. Emiliozzi, P., Scarpone, P., DePaula, F., Pizzo, M., Federico, G., Pansadoro, A. et al: The incidence of prostate cancer in men with prostate specific antigen greater than 4.0 ng/ml: a randomized study of 6 versus 12 core transperineal prostate biopsy. J Urol, 171: 197, 2004 14. Chan, T. Y., Chan, D. Y., Stutzman, K. L. R. E. and Epstein, J. I.: Does increased needle biopsy sampling of the prostate detect a higher number of potentially insignificant tumors? J Urol, 166: 2181, 2001 15. Goldstein, N. S., Begin, L. R., Grody, W. W., Novak, J. M., Qian, J. and Bostwick, D. G.: Minimal or no cancer in radical prostatectomy specimens. Report of 13 cases of the “vanishing cancer phenomenon”. Am J Surg Pathol, 19: 1002, 1995 16. Herkommer, K., Kuefer, R., Gschwend, J. E., Hautmann, R. E. and Volkmer, B. G.: Pathological T0 prostate cancer without neoadjuvant therapy clinical presentation and follow-up. Eur Urol, 45: 36, 2004 17. Carroll, P. R.: Early stage prostate cancer— do we have a problem with over-detection, overtreatment or both? J Urol, 173: 1061, 2005 18. Patel, M. I., DeConcini, D. T., Lopez-Corona, E., Ohori, M., Wheeler, T. and Scardino, P. T.: An analysis of men with clinically localized prostate cancer who deferred definitive therapy. J Urol, 171: 1520, 2004 19. Patel, A. R., Jones, J. S., Rabets, J., DeOreo, G. and Zippe, C. D.: Parasagittal biopsies add minimal information in repeat saturation prostate biopsy. Urology, 63: 87, 2004 20. Lefkowitz, G. K., Sidhu, G. S., Torre, P., Lepor, H. and Taneja, S. S.: Is repeat prostate biopsy for high-grade prostatic intraepithelial neoplasia necessary after routine 12-core sampling? Urology, 58: 999, 2001