- Email: [email protected]
Prostatic Transition Zone Directed Needle Biopsies Uncommonly Sample Clinically Relevant Transition Zone Tumors
Prostatic Transition Zone Directed Needle Biopsies Uncommonly Sample Clinically Relevant Transition Zone Tumors Chadwick F. Haarer, Anuradha Gopalan, Satish K. Tickoo, Peter T. Scardino, James A. Eastham, Victor E. Reuter and Samson W. Fine* From the Departments of Pathology (CFH, AG, SKT, VER, SWF) and Surgery (PTS, JAE), Memorial Sloan-Kettering Cancer Center, New York, New York
Purpose: We compared prostate cancer detected in transition zone directed needle biopsies with those in corresponding radical prostatectomy specimens. Materials and Methods: We reviewed needle biopsy and radical prostatectomy slides from 61 patients in whom cancer was present on transition zone directed needle biopsy. We assessed needle biopsy cancer features as well as transition zone lesions and dominant tumor sites on radical prostatectomy. Results: Prostate cancer was detected in 25 of 61 left (41%), 23 of 61 right (38%) and 13 of 61 bilateral (21%) transition zone directed needle biopsies. On radical prostatectomy 24 of 61 cases (39.5%) had no tumor in the transition zone, 24 of 61 (39.5%) had nondominant transition zone cancer and 13 of 61 (21%) had a dominant transition zone lesion. Of cases with cancer in the left and right transition zone directed needle biopsy 18 of 38 (47%) and 17 of 36 (47%), respectively, had no transition zone tumor or showed tumor in the contralateral transition zone only at radical prostatectomy. In 8 cases the transition zone directed core was the only one with cancer on needle biopsy and 2 of 8 (25%) such cases showed dominant transition zone cancer at radical prostatectomy. Conclusions: Cancer identified in transition zone directed needle biopsy cores was not from the transition zone or did not reflect a dominant transition zone lesion in almost 80% of cases. Cancer identified in a left or right transition zone directed needle biopsy did not predict ipsilateral transition zone cancer in almost 50% of cases. These findings suggest that such biopsies do not adequately characterize transition zone tumors. Thus, care should be taken in their interpretation.
Abbreviations and Acronyms NB ⫽ needle biopsy PSA ⫽ prostate specific antigen RP ⫽ radical prostatectomy TZ ⫽ transition zone Submitted for publication February 6, 2009. Study received institutional review board approval. Supported by the Sidney Kimmel Center for Prostate and Urologic Cancers. * Correspondence: Department of Pathology, Memorial Sloan-Kettering Cancer Center, 1275 York Ave., Room C505, New York, New York 10065 (telephone: 212-639-5066; FAX: 646-4222070; e-mail: [email protected]).
Key Words: prostate; prostatic neoplasms; biopsy, needle; prostatectomy; diagnosis IN recent years we and others have reported a trend toward an increasing number of dominant anterior prostatic tumors.1–3 A significant percent of these cancers are located in the prostatic TZ, typically more difficult to detect by digital rectal examination1 and poorly visualized/distinguished on imaging.4,5 Therefore, the need for an NB technique that can adequately sample this region is crucial. Trans-
rectal NBs directed at the TZ have been proposed to detect these tumors with investigators reporting a 1.8% to 13% increase in cancer detection.6 – 8 While some groups have considered the relationship between cancer detected on these directed NBs and RP findings,9,10 few have investigated how reliably TZ directed NBs reveal TZ localized cancer or compared the location or morphology of tumor on
0022-5347/09/1824-1337/0 THE JOURNAL OF UROLOGY® Copyright © 2009 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 182, 1337-1341, October 2009 Printed in U.S.A. DOI:10.1016/j.juro.2009.06.042
www.jurology.com
1337
1338
PROSTATIC TRANSITION ZONE BIOPSIES RARELY SAMPLE RELEVANT TUMORS
Table 1. Laterality of TZ directed NB cancer vs RP findings No. RP TZ Ca (%) NB TZ Ca
None
Contralat
Ipsilat
Lt or bilat Rt or bilat
14 (36) 13 (36)
4 (11) 4 (11)
20 (53) 19 (53)
these NBs with tumor identified on RP. Thus, we examined matched 14-core NB (including right and left TZ directed NBs) and RP specimens to determine whether 1) presumptive TZ directed NBs in fact sample TZ cancer and 2) features of TZ directed NB cancer could predict dominant TZ cancer on RP.
MATERIALS AND METHODS After receiving institutional review board approval the surgical pathology archives at our institution were queried for cases of prostate cancer diagnosed on a TZ directed NB, regardless of whether cancer was present in nonTZ cores, and subsequent RP was performed between 1999 and 2006. During this period the standard prostate NB protocol included 1 core each from the right and left lateral apex, mid and base, right and left medial apex, mid and base, and right and left TZs for a total of 14 cores. All original slides of NB and RP specimens were reviewed by 2 of us (CFH and SWF). For each TZ directed NB we evaluated Gleason score, percent of tissue with carcinoma, tumor site on the core and percent of cancer with TZLOOK morphology, defined as variably sized glands composed of columnar cells with clear to pale pink cytoplasm and basally oriented nuclei.11,12 Tumor site in the core was denoted as the middle, the end or the middle and the end(s) of the core. All RP specimens were inked bilaterally, prosected and whole mounted, as previously described.13 For each RP specimen we assessed the presence of tumor in the TZ as well as the site, Gleason score and percent TZ-LOOK in the dominant lesion. Designating anterior tumors as TZ or PZ origin was based on relationships between glandular zones and the anterior fibromuscular stroma,13,14 and on our recent study highlighting variability in the anterior prostatic anatomy from apex through base.13 In cases of multiple tumor nodules each case had a tumor nodule designated as dominant, that is the lesion with the largest diameter and involving the most sections.
RESULTS From institutional databases and pathology reports we identified 65 cases with available original slides in which a cancer diagnosis was made on TZ directed NB. Four cases were excluded from analysis due to a history of radiation and/or hormonal therapy (3) or inability to confirm a carcinoma diagnosis on biopsy upon review (1), leaving 61 as the study cohort. Cancer was found on the left side only in 25 of 61 cases (41%), on the right side only in 23 of 61 (38%) and on bilateral TZ directed NB in 13 of 61 (21%).
On RP we found no cancer in the TZ in 24 of 61 cases (39.5%), nondominant TZ cancer in 24 of 61 (39.5%) and dominant TZ cancer in 13 of 61 (21%). Of cancers identified on left TZ directed NB (left side only or bilateral) 18 of 38 (47%) showed no tumor in the TZ or right TZ cancer only at RP. Similarly 17 of 36 cases (47%) with cancer on right TZ biopsy (right side only or bilateral) had no tumor in the TZ or left TZ cancer only (table 1). In 8 cases (13%) cancer in the TZ directed core was the only tumor identified in the biopsy specimen, including 2 with a corresponding dominant TZ tumor on RP. We then compared the location and percent of cancer in TZ directed cores and cancer in the TZ on RP. Including cancer detected bilaterally, 39, 13 and 22 cases had tumor at the middle, either end or middle plus end(s) of the biopsy core, respectively. Of these cases 23 of 39 (59%), 9 of 13 (69%) and 15 of 22 (68%) had any TZ tumor in the resection specimen with an increasing trend in the percent containing a dominant TZ tumor, that is 22% mid core (5 of 23), 33% end (3 of 9) and 53% middle plus end(s) (8 of 15). The mean percent involvement of TZ directed NB by cancer was 27% in cases with dominant TZ tumors and in those with nondominant/no TZ tumor at RP. Most RP specimens with TZ tumor, including 32 of 37 (87%) with any TZ tumor and 11 of 13 (85%) with dominant TZ tumor, contained 50% or less involvement on TZ directed NB (table 2). Only 2 of the 5 cases with greater than 50% NB involvement by cancer showed TZ dominant tumors. When we compared Gleason scores on TZ directed NB and RP, we found an overall 59% exact correlation and no significant difference between nonTZ dominant and TZ dominant cases (27 of 48 or 56% vs 9 of 13 or 69%, respectively). Finally, we evaluated TZ-LOOK morphology in cancer bearing TZ directed cores and in RP tumors. TZ-LOOK extent was divided into 4 categories, including none— 0%, focal—1% to 25%, nonfocal—26% to 50% and predominant— greater than 50%. On TZ directed NB 60 of 61 cases (98%) showed no (49 of 61 or 80%) or focal (11 of 61 or 18%) TZ-LOOK histology. Similarly in 12 of 13 cases with dominant TZ cancer on RP corresponding TZ directed cores showed no (6) or focal (6) TZ-LOOK. Table 2. Cancer on TZ directed NB vs TZ cancer in RP specimens RP TZ Tumor (No. Pts) % NB Ca
None
Any
Dominant
25 or Less 26–50 51–75 Greater than 75
11 9 0 4
26 6 3 2
9 2 1 1
PROSTATIC TRANSITION ZONE BIOPSIES RARELY SAMPLE RELEVANT TUMORS
DISCUSSION Given the difficulty in detecting anterior prostatic tumors, including dominant cancer nodules, on clinical examination, imaging and NB,1,4 –5,8 a number of investigators have examined the value of TZ directed NB with conflicting results. Some groups argue for the usefulness of such biopsies in patients with previous negative NB and persistently increased PSA or those with gray zone (4.0 to 9.9 ng/ml) PSA,15–17 while others state that even in these specific settings few additional prostate cancers are detected.18,19 However, most groups have noted that including TZ directed NB in routine protocols has a relatively low yield with only 1.5% to 8% increased cancer detection over that of traditional peripheral zone biopsy alone. Hence, most investigators argue that these NBs might be reserved for specific clinical settings and should not be a component of the initial biopsy session,6,7,16,20 especially since they have the potential to cause hematuria and discomfort during urination. While those groups examined the efficacy of TZ directed NB to detect cancer, few have looked at whether these NBs truly sample the TZ and/or at the clinical relevance of detected tumors. Ex vivo studies by Fink et al19 and the study by Pelzer et al of 9 cases in which TZ directed NB tumor was the only cancer detected on NB10 show that almost no cases with positive TZ directed cores had TZ cancer only at RP. In another ex vivo study McNeal and Noldus determined that these biopsies are useful to identify tumors less than 5.0 cc but could miss more common anterior tumors in the 2.0 to 5.0 cc range.9 Our findings clearly show that TZ directed NBs uncommonly sample clinically relevant TZ tumors. In this series approximately 40% of patients with positive TZ directed cores had no TZ cancer and almost 80% had nondominant TZ cancer even when TZ biopsy contained the only cancer in the NB. We
1339
also noted significant discrepancies in detected tumor laterality with approximately 50% of positive left or right TZ directed NBs showing no or contralateral TZ tumor only. These unexpected findings may be partially explained by technical, topographic and mechanical considerations. TZ directed NB involves advancing the biopsy needle approximately 1 cm into the mid medial prostate before firing.9 Individual variability in peripheral zone tissue volume, and the presence and extent of benign prostatic hypertrophy may impact the ability to effectively sample anteromedial TZ tissue. Also, to avoid the urethra the biopsy needle may not be placed medial enough to properly sample the TZ in a given case. Finally, there is evidence to suggest that TZ tissue is relatively resistant to needle penetration. In an ex vivo study using dyes to track needle paths McNeal and Noldus observed that TZ NB selectively sampled tissue between nodules of benign prostatic hyperplasia, tending to spare hyperplastic nodules.9 They postulated that this phenomenon may be secondary to the mobility of hyperplastic nodules, an argument that might be extended to some TZ localized prostate cancer nodules. Most morphological features in TZ directed NB cancer could not distinguish dominant TZ tumors. The mean percent involvement of NB was not different in cases with or without dominant TZ cancer, as highlighted by 4 with 75% or greater NB involvement but no TZ tumor. Moreover, while prostate cancer heterogeneity is well accepted and the limited material in a NB often does not reflect the overall Gleason score of RP tumors, our results show that exact NB-RP Gleason score correlation was not significantly different between patients with vs those without dominant TZ tumors at RP. Similarly we recently noted that the clear cell or TZ-LOOK histology described by McNeal et al11 may be found in a considerable number of peripheral zone tu-
Posterior peripheral zone cancer (A) with TZ-LOOK morphology (B)
1340
PROSTATIC TRANSITION ZONE BIOPSIES RARELY SAMPLE RELEVANT TUMORS
mors,12 implying that this morphology should not be used to determine whether one has actually sampled the TZ on NB. The current study reinforces this notion with only 1 case showing extensive TZ-LOOK on NB and a dominant TZ lesion (see figure). Conversely most cases showed no or focal TZ-LOOK on cancer bearing TZ biopsies. Hence, the extent, Gleason score and morphology of tumors detected on TZ directed NB did not predict dominant TZ cancer. The only significant association in this series was between the cancer site on the TZ directed core and a dominant TZ tumor on RP. The proportion of dominant TZ cancers gradually increased in cases with tumor at the middle, end and middle plus end(s) of the TZ biopsy, respectively. To our knowledge this phenomenon has not been previously documented and it is somewhat difficult to explain, especially in light of the similar tumor volumes on NB in the documented cases with vs without dominant TZ tu-
mors. Although we did not formally attempt to distinguish NB core ends (closest to the rectum vs most intraprostatic) and have found this difficult to determine in practice, it is possible that at least in some cases the end of the core corresponds to the part of the needle closest to the TZ. Alternatively using the logic of McNeal and Noldus9 we suggest that high volume, dominant TZ cancer renders TZ tissue less resistant to penetration, facilitating more accurate biopsy of these lesions.
CONCLUSIONS We report the inconsistency with which TZ directed NBs sample cancer in the TZ. As such, we urge caution in interpreting cancer bearing TZ directed biopsies since they do not appear to be particularly useful to accurately characterize TZ/anterior prostatic tumors.
REFERENCES 1. Bott SRJ, Young MPA, Kellett MJ et al: Anterior prostate cancer: is it more difficult to diagnose? BJU Int 2002; 89: 886. 2. Koppie TM, Bianco FJ Jr, Kuroiwa K et al: The clinical features of anterior prostate cancers. BJU Int 2006; 98: 1167. 3. Al-Ahmadie HA, Tickoo SK, Olgac S et al: Anterior-predominant prostatic tumors: zone of origin and pathologic outcomes at radical prostatectomy. Am J Surg Pathol 2008; 32: 229. 4. Terris MK, Freiha FS, McNeal JE et al: Efficacy of transrectal ultrasound for identification of clinically undetected prostate cancer. J Urol 1991; 146: 78.
prostate biopsies are not indicated. J Urol 1997; 157: 204.
14. McNeal JE: The zonal anatomy of the prostate. Prostate 1981; 2: 35.
8. Chang JJ, Shinohara K, Hovey RM et al: Prospective evaluation of systemic sextant transition zone biopsies in large prostates for cancer detection. Urology 1998; 52: 89.
15. Maeda H, Ishitoya S, Aoki Y et al: Value of systematic transition zone biopsy in the detection of prostate cancer. Int J Urol 1997; 4: 567.
9. McNeal J and Noldus J: Limitations of transition zone needle biopsy findings in the prediction of transition zone cancer and tissue composition of benign nodular hyperplasia. Urology 1996; 48: 751. 10. Pelzer AE, Bektic J, Berger AP et al: Are transition zone biopsies still necessary to improve prostate cancer detection? Results from the Tyrol screening project. Eur Urol 2005; 48: 916.
16. Liu IJ, Macy M, Lai YH et al: Critical evaluation of the current indications for transition zone biopsies. Urology 2001; 57: 1117. 17. Ishizuka O, Mimura Y, Oguchi T et al: Importance of transition zone biopsies in patients with grayzone PSA levels undergoing the ultrasoundguided systematic ten-biopsy regimen for the first time. Urol Int 2005; 74: 23. 18. Keetch DW and Catalona WJ: Prostatic transition zone biopsies in men with previous negative biopsies and persistently elevated serum prostate specific antigen values. J Urol 1995; 154: 1795.
5. Zakian KL, Eberhardt S, Hricak H et al: Transition zone prostate cancer: metabolic characteristics at 1H MR spectroscopic imaging: initial results. Radiology 2003; 229: 241.
11. McNeal JE, Redwine EA, Freiha FS et al: Zonal distribution of prostatic adenocarcinomas: correlation with histologic pattern and direction of spread. Am J Surg Pathol 1988; 12: 897.
6. Bazinet M, Karakiewicz PI, Aprikian AG et al: Value of systematic transition zone biopsies in the early detection of prostate cancer. J Urol 1996; 155: 605.
12. Garcia JJ, Al-Ahmadie HA, Gopalan A et al: Do prostatic transition zone tumors have a distinct morphology? Am J Surg Pathol 2008; 32: 1709.
19. Fink KG, Hutarew G, Esterbauer B et al: Evaluation of transition zone and lateral sextant biopsies for prostate cancer detection after initial sextant biopsy. Urology 2003; 61: 748.
13. Fine SW, Al-Ahmadie HA, Gopalan A et al: Anatomy of the anterior prostate and extraprostatic space: a contemporary surgical pathology analysis. Adv Anat Pathol 2007; 14: 401.
20. Morote J, Lopez M, Encabo G et al: Value of routine transition zone biopsies in patients undergoing ultrasound-guided sextant biopsies for the first time. Eur Urol 1998; 35: 294.
7. Terris MK, Pham TQ, Issa MM et al: Routine transition zone and seminal vesicle biopsies in all patients undergoing transrectal ultrasound guided
EDITORIAL COMMENT This is a cohort study of patients with RP diagnosed by TZ directed NB. Although the authors conclude that the diagnostic yield of TZ directed biopsies is low, this population is a highly select, retrospectively constructed cohort. As such, the external validity and applicability to clinical practice of these findings remain unclear.
During the study period patients underwent routine 14-core biopsy with only 2 cores directed at the TZ, presumably because anterior tumors were not suspected. A more germane study population would be those in whom an anterior tumor is suspected but the diagnosis remains elusive, that is patients with
PROSTATIC TRANSITION ZONE BIOPSIES RARELY SAMPLE RELEVANT TUMORS
rapidly increasing PSA, several prior negative biopsies and unremarkable digital rectal examination. In these patients multiple TZ cores, perhaps in conjunction with magnetic resonance or other advanced imaging, may potentially yield valuable information. Another issue is whether these data would alter clinical decision making. While it may be true that TZ directed biopsies do not accurately characterize
1341
anterior tumors, histopathological distinctions between transitional and peripheral tumors on biopsy do not routinely enter into standard treatment algorithms for prostate cancer. J. Kellogg Parsons Moores Comprehensive Cancer Center University of California-San Diego La Jolla, California
REPLY BY AUTHORS Bilateral TZ directed NBs are routinely performed in our practice and not only reserved for specific clinical scenarios. We investigated the subset of our patients in whom cancer was detected in these biopsies. We acknowledge that the findings in TZ directed NBs performed in specific patients with multiple previous negative biopsies, gray zone PSA or imaging suspicious for an anterior tumor may yield somewhat different findings. However, to our knowledge no large series has compared cancer detected in routinely performed TZ directed biopsies to RP findings. We believe that the nearly 80% of cases with no
or nondominant tumors in the TZ at RP is enough to urge caution in interpretation of such biopsies, regardless of clinical circumstance. Furthermore, while we are aware that histopathological distinctions between transition and peripheral zone tumors on NB may not enter standard treatment algorithms, we demonstrate that these pathological features in TZ NBs do not predict the features of the tumor seen at prostatectomy. Additionally, TZ-LOOK histology has been widely described and used as a feature of TZ tumors, and we show that it does not predict TZ dominant lesions on biopsy.
Purpose: We compared prostate cancer detected in transition zone directed needle biopsies with those in corresponding radical prostatectomy specimens. Materials and Methods: We reviewed needle biopsy and radical prostatectomy slides from 61 patients in whom cancer was present on transition zone directed needle biopsy. We assessed needle biopsy cancer features as well as transition zone lesions and dominant tumor sites on radical prostatectomy. Results: Prostate cancer was detected in 25 of 61 left (41%), 23 of 61 right (38%) and 13 of 61 bilateral (21%) transition zone directed needle biopsies. On radical prostatectomy 24 of 61 cases (39.5%) had no tumor in the transition zone, 24 of 61 (39.5%) had nondominant transition zone cancer and 13 of 61 (21%) had a dominant transition zone lesion. Of cases with cancer in the left and right transition zone directed needle biopsy 18 of 38 (47%) and 17 of 36 (47%), respectively, had no transition zone tumor or showed tumor in the contralateral transition zone only at radical prostatectomy. In 8 cases the transition zone directed core was the only one with cancer on needle biopsy and 2 of 8 (25%) such cases showed dominant transition zone cancer at radical prostatectomy. Conclusions: Cancer identified in transition zone directed needle biopsy cores was not from the transition zone or did not reflect a dominant transition zone lesion in almost 80% of cases. Cancer identified in a left or right transition zone directed needle biopsy did not predict ipsilateral transition zone cancer in almost 50% of cases. These findings suggest that such biopsies do not adequately characterize transition zone tumors. Thus, care should be taken in their interpretation.
Abbreviations and Acronyms NB ⫽ needle biopsy PSA ⫽ prostate specific antigen RP ⫽ radical prostatectomy TZ ⫽ transition zone Submitted for publication February 6, 2009. Study received institutional review board approval. Supported by the Sidney Kimmel Center for Prostate and Urologic Cancers. * Correspondence: Department of Pathology, Memorial Sloan-Kettering Cancer Center, 1275 York Ave., Room C505, New York, New York 10065 (telephone: 212-639-5066; FAX: 646-4222070; e-mail: [email protected]).
Key Words: prostate; prostatic neoplasms; biopsy, needle; prostatectomy; diagnosis IN recent years we and others have reported a trend toward an increasing number of dominant anterior prostatic tumors.1–3 A significant percent of these cancers are located in the prostatic TZ, typically more difficult to detect by digital rectal examination1 and poorly visualized/distinguished on imaging.4,5 Therefore, the need for an NB technique that can adequately sample this region is crucial. Trans-
rectal NBs directed at the TZ have been proposed to detect these tumors with investigators reporting a 1.8% to 13% increase in cancer detection.6 – 8 While some groups have considered the relationship between cancer detected on these directed NBs and RP findings,9,10 few have investigated how reliably TZ directed NBs reveal TZ localized cancer or compared the location or morphology of tumor on
0022-5347/09/1824-1337/0 THE JOURNAL OF UROLOGY® Copyright © 2009 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 182, 1337-1341, October 2009 Printed in U.S.A. DOI:10.1016/j.juro.2009.06.042
www.jurology.com
1337
1338
PROSTATIC TRANSITION ZONE BIOPSIES RARELY SAMPLE RELEVANT TUMORS
Table 1. Laterality of TZ directed NB cancer vs RP findings No. RP TZ Ca (%) NB TZ Ca
None
Contralat
Ipsilat
Lt or bilat Rt or bilat
14 (36) 13 (36)
4 (11) 4 (11)
20 (53) 19 (53)
these NBs with tumor identified on RP. Thus, we examined matched 14-core NB (including right and left TZ directed NBs) and RP specimens to determine whether 1) presumptive TZ directed NBs in fact sample TZ cancer and 2) features of TZ directed NB cancer could predict dominant TZ cancer on RP.
MATERIALS AND METHODS After receiving institutional review board approval the surgical pathology archives at our institution were queried for cases of prostate cancer diagnosed on a TZ directed NB, regardless of whether cancer was present in nonTZ cores, and subsequent RP was performed between 1999 and 2006. During this period the standard prostate NB protocol included 1 core each from the right and left lateral apex, mid and base, right and left medial apex, mid and base, and right and left TZs for a total of 14 cores. All original slides of NB and RP specimens were reviewed by 2 of us (CFH and SWF). For each TZ directed NB we evaluated Gleason score, percent of tissue with carcinoma, tumor site on the core and percent of cancer with TZLOOK morphology, defined as variably sized glands composed of columnar cells with clear to pale pink cytoplasm and basally oriented nuclei.11,12 Tumor site in the core was denoted as the middle, the end or the middle and the end(s) of the core. All RP specimens were inked bilaterally, prosected and whole mounted, as previously described.13 For each RP specimen we assessed the presence of tumor in the TZ as well as the site, Gleason score and percent TZ-LOOK in the dominant lesion. Designating anterior tumors as TZ or PZ origin was based on relationships between glandular zones and the anterior fibromuscular stroma,13,14 and on our recent study highlighting variability in the anterior prostatic anatomy from apex through base.13 In cases of multiple tumor nodules each case had a tumor nodule designated as dominant, that is the lesion with the largest diameter and involving the most sections.
RESULTS From institutional databases and pathology reports we identified 65 cases with available original slides in which a cancer diagnosis was made on TZ directed NB. Four cases were excluded from analysis due to a history of radiation and/or hormonal therapy (3) or inability to confirm a carcinoma diagnosis on biopsy upon review (1), leaving 61 as the study cohort. Cancer was found on the left side only in 25 of 61 cases (41%), on the right side only in 23 of 61 (38%) and on bilateral TZ directed NB in 13 of 61 (21%).
On RP we found no cancer in the TZ in 24 of 61 cases (39.5%), nondominant TZ cancer in 24 of 61 (39.5%) and dominant TZ cancer in 13 of 61 (21%). Of cancers identified on left TZ directed NB (left side only or bilateral) 18 of 38 (47%) showed no tumor in the TZ or right TZ cancer only at RP. Similarly 17 of 36 cases (47%) with cancer on right TZ biopsy (right side only or bilateral) had no tumor in the TZ or left TZ cancer only (table 1). In 8 cases (13%) cancer in the TZ directed core was the only tumor identified in the biopsy specimen, including 2 with a corresponding dominant TZ tumor on RP. We then compared the location and percent of cancer in TZ directed cores and cancer in the TZ on RP. Including cancer detected bilaterally, 39, 13 and 22 cases had tumor at the middle, either end or middle plus end(s) of the biopsy core, respectively. Of these cases 23 of 39 (59%), 9 of 13 (69%) and 15 of 22 (68%) had any TZ tumor in the resection specimen with an increasing trend in the percent containing a dominant TZ tumor, that is 22% mid core (5 of 23), 33% end (3 of 9) and 53% middle plus end(s) (8 of 15). The mean percent involvement of TZ directed NB by cancer was 27% in cases with dominant TZ tumors and in those with nondominant/no TZ tumor at RP. Most RP specimens with TZ tumor, including 32 of 37 (87%) with any TZ tumor and 11 of 13 (85%) with dominant TZ tumor, contained 50% or less involvement on TZ directed NB (table 2). Only 2 of the 5 cases with greater than 50% NB involvement by cancer showed TZ dominant tumors. When we compared Gleason scores on TZ directed NB and RP, we found an overall 59% exact correlation and no significant difference between nonTZ dominant and TZ dominant cases (27 of 48 or 56% vs 9 of 13 or 69%, respectively). Finally, we evaluated TZ-LOOK morphology in cancer bearing TZ directed cores and in RP tumors. TZ-LOOK extent was divided into 4 categories, including none— 0%, focal—1% to 25%, nonfocal—26% to 50% and predominant— greater than 50%. On TZ directed NB 60 of 61 cases (98%) showed no (49 of 61 or 80%) or focal (11 of 61 or 18%) TZ-LOOK histology. Similarly in 12 of 13 cases with dominant TZ cancer on RP corresponding TZ directed cores showed no (6) or focal (6) TZ-LOOK. Table 2. Cancer on TZ directed NB vs TZ cancer in RP specimens RP TZ Tumor (No. Pts) % NB Ca
None
Any
Dominant
25 or Less 26–50 51–75 Greater than 75
11 9 0 4
26 6 3 2
9 2 1 1
PROSTATIC TRANSITION ZONE BIOPSIES RARELY SAMPLE RELEVANT TUMORS
DISCUSSION Given the difficulty in detecting anterior prostatic tumors, including dominant cancer nodules, on clinical examination, imaging and NB,1,4 –5,8 a number of investigators have examined the value of TZ directed NB with conflicting results. Some groups argue for the usefulness of such biopsies in patients with previous negative NB and persistently increased PSA or those with gray zone (4.0 to 9.9 ng/ml) PSA,15–17 while others state that even in these specific settings few additional prostate cancers are detected.18,19 However, most groups have noted that including TZ directed NB in routine protocols has a relatively low yield with only 1.5% to 8% increased cancer detection over that of traditional peripheral zone biopsy alone. Hence, most investigators argue that these NBs might be reserved for specific clinical settings and should not be a component of the initial biopsy session,6,7,16,20 especially since they have the potential to cause hematuria and discomfort during urination. While those groups examined the efficacy of TZ directed NB to detect cancer, few have looked at whether these NBs truly sample the TZ and/or at the clinical relevance of detected tumors. Ex vivo studies by Fink et al19 and the study by Pelzer et al of 9 cases in which TZ directed NB tumor was the only cancer detected on NB10 show that almost no cases with positive TZ directed cores had TZ cancer only at RP. In another ex vivo study McNeal and Noldus determined that these biopsies are useful to identify tumors less than 5.0 cc but could miss more common anterior tumors in the 2.0 to 5.0 cc range.9 Our findings clearly show that TZ directed NBs uncommonly sample clinically relevant TZ tumors. In this series approximately 40% of patients with positive TZ directed cores had no TZ cancer and almost 80% had nondominant TZ cancer even when TZ biopsy contained the only cancer in the NB. We
1339
also noted significant discrepancies in detected tumor laterality with approximately 50% of positive left or right TZ directed NBs showing no or contralateral TZ tumor only. These unexpected findings may be partially explained by technical, topographic and mechanical considerations. TZ directed NB involves advancing the biopsy needle approximately 1 cm into the mid medial prostate before firing.9 Individual variability in peripheral zone tissue volume, and the presence and extent of benign prostatic hypertrophy may impact the ability to effectively sample anteromedial TZ tissue. Also, to avoid the urethra the biopsy needle may not be placed medial enough to properly sample the TZ in a given case. Finally, there is evidence to suggest that TZ tissue is relatively resistant to needle penetration. In an ex vivo study using dyes to track needle paths McNeal and Noldus observed that TZ NB selectively sampled tissue between nodules of benign prostatic hyperplasia, tending to spare hyperplastic nodules.9 They postulated that this phenomenon may be secondary to the mobility of hyperplastic nodules, an argument that might be extended to some TZ localized prostate cancer nodules. Most morphological features in TZ directed NB cancer could not distinguish dominant TZ tumors. The mean percent involvement of NB was not different in cases with or without dominant TZ cancer, as highlighted by 4 with 75% or greater NB involvement but no TZ tumor. Moreover, while prostate cancer heterogeneity is well accepted and the limited material in a NB often does not reflect the overall Gleason score of RP tumors, our results show that exact NB-RP Gleason score correlation was not significantly different between patients with vs those without dominant TZ tumors at RP. Similarly we recently noted that the clear cell or TZ-LOOK histology described by McNeal et al11 may be found in a considerable number of peripheral zone tu-
Posterior peripheral zone cancer (A) with TZ-LOOK morphology (B)
1340
PROSTATIC TRANSITION ZONE BIOPSIES RARELY SAMPLE RELEVANT TUMORS
mors,12 implying that this morphology should not be used to determine whether one has actually sampled the TZ on NB. The current study reinforces this notion with only 1 case showing extensive TZ-LOOK on NB and a dominant TZ lesion (see figure). Conversely most cases showed no or focal TZ-LOOK on cancer bearing TZ biopsies. Hence, the extent, Gleason score and morphology of tumors detected on TZ directed NB did not predict dominant TZ cancer. The only significant association in this series was between the cancer site on the TZ directed core and a dominant TZ tumor on RP. The proportion of dominant TZ cancers gradually increased in cases with tumor at the middle, end and middle plus end(s) of the TZ biopsy, respectively. To our knowledge this phenomenon has not been previously documented and it is somewhat difficult to explain, especially in light of the similar tumor volumes on NB in the documented cases with vs without dominant TZ tu-
mors. Although we did not formally attempt to distinguish NB core ends (closest to the rectum vs most intraprostatic) and have found this difficult to determine in practice, it is possible that at least in some cases the end of the core corresponds to the part of the needle closest to the TZ. Alternatively using the logic of McNeal and Noldus9 we suggest that high volume, dominant TZ cancer renders TZ tissue less resistant to penetration, facilitating more accurate biopsy of these lesions.
CONCLUSIONS We report the inconsistency with which TZ directed NBs sample cancer in the TZ. As such, we urge caution in interpreting cancer bearing TZ directed biopsies since they do not appear to be particularly useful to accurately characterize TZ/anterior prostatic tumors.
REFERENCES 1. Bott SRJ, Young MPA, Kellett MJ et al: Anterior prostate cancer: is it more difficult to diagnose? BJU Int 2002; 89: 886. 2. Koppie TM, Bianco FJ Jr, Kuroiwa K et al: The clinical features of anterior prostate cancers. BJU Int 2006; 98: 1167. 3. Al-Ahmadie HA, Tickoo SK, Olgac S et al: Anterior-predominant prostatic tumors: zone of origin and pathologic outcomes at radical prostatectomy. Am J Surg Pathol 2008; 32: 229. 4. Terris MK, Freiha FS, McNeal JE et al: Efficacy of transrectal ultrasound for identification of clinically undetected prostate cancer. J Urol 1991; 146: 78.
prostate biopsies are not indicated. J Urol 1997; 157: 204.
14. McNeal JE: The zonal anatomy of the prostate. Prostate 1981; 2: 35.
8. Chang JJ, Shinohara K, Hovey RM et al: Prospective evaluation of systemic sextant transition zone biopsies in large prostates for cancer detection. Urology 1998; 52: 89.
15. Maeda H, Ishitoya S, Aoki Y et al: Value of systematic transition zone biopsy in the detection of prostate cancer. Int J Urol 1997; 4: 567.
9. McNeal J and Noldus J: Limitations of transition zone needle biopsy findings in the prediction of transition zone cancer and tissue composition of benign nodular hyperplasia. Urology 1996; 48: 751. 10. Pelzer AE, Bektic J, Berger AP et al: Are transition zone biopsies still necessary to improve prostate cancer detection? Results from the Tyrol screening project. Eur Urol 2005; 48: 916.
16. Liu IJ, Macy M, Lai YH et al: Critical evaluation of the current indications for transition zone biopsies. Urology 2001; 57: 1117. 17. Ishizuka O, Mimura Y, Oguchi T et al: Importance of transition zone biopsies in patients with grayzone PSA levels undergoing the ultrasoundguided systematic ten-biopsy regimen for the first time. Urol Int 2005; 74: 23. 18. Keetch DW and Catalona WJ: Prostatic transition zone biopsies in men with previous negative biopsies and persistently elevated serum prostate specific antigen values. J Urol 1995; 154: 1795.
5. Zakian KL, Eberhardt S, Hricak H et al: Transition zone prostate cancer: metabolic characteristics at 1H MR spectroscopic imaging: initial results. Radiology 2003; 229: 241.
11. McNeal JE, Redwine EA, Freiha FS et al: Zonal distribution of prostatic adenocarcinomas: correlation with histologic pattern and direction of spread. Am J Surg Pathol 1988; 12: 897.
6. Bazinet M, Karakiewicz PI, Aprikian AG et al: Value of systematic transition zone biopsies in the early detection of prostate cancer. J Urol 1996; 155: 605.
12. Garcia JJ, Al-Ahmadie HA, Gopalan A et al: Do prostatic transition zone tumors have a distinct morphology? Am J Surg Pathol 2008; 32: 1709.
19. Fink KG, Hutarew G, Esterbauer B et al: Evaluation of transition zone and lateral sextant biopsies for prostate cancer detection after initial sextant biopsy. Urology 2003; 61: 748.
13. Fine SW, Al-Ahmadie HA, Gopalan A et al: Anatomy of the anterior prostate and extraprostatic space: a contemporary surgical pathology analysis. Adv Anat Pathol 2007; 14: 401.
20. Morote J, Lopez M, Encabo G et al: Value of routine transition zone biopsies in patients undergoing ultrasound-guided sextant biopsies for the first time. Eur Urol 1998; 35: 294.
7. Terris MK, Pham TQ, Issa MM et al: Routine transition zone and seminal vesicle biopsies in all patients undergoing transrectal ultrasound guided
EDITORIAL COMMENT This is a cohort study of patients with RP diagnosed by TZ directed NB. Although the authors conclude that the diagnostic yield of TZ directed biopsies is low, this population is a highly select, retrospectively constructed cohort. As such, the external validity and applicability to clinical practice of these findings remain unclear.
During the study period patients underwent routine 14-core biopsy with only 2 cores directed at the TZ, presumably because anterior tumors were not suspected. A more germane study population would be those in whom an anterior tumor is suspected but the diagnosis remains elusive, that is patients with
PROSTATIC TRANSITION ZONE BIOPSIES RARELY SAMPLE RELEVANT TUMORS
rapidly increasing PSA, several prior negative biopsies and unremarkable digital rectal examination. In these patients multiple TZ cores, perhaps in conjunction with magnetic resonance or other advanced imaging, may potentially yield valuable information. Another issue is whether these data would alter clinical decision making. While it may be true that TZ directed biopsies do not accurately characterize
1341
anterior tumors, histopathological distinctions between transitional and peripheral tumors on biopsy do not routinely enter into standard treatment algorithms for prostate cancer. J. Kellogg Parsons Moores Comprehensive Cancer Center University of California-San Diego La Jolla, California
REPLY BY AUTHORS Bilateral TZ directed NBs are routinely performed in our practice and not only reserved for specific clinical scenarios. We investigated the subset of our patients in whom cancer was detected in these biopsies. We acknowledge that the findings in TZ directed NBs performed in specific patients with multiple previous negative biopsies, gray zone PSA or imaging suspicious for an anterior tumor may yield somewhat different findings. However, to our knowledge no large series has compared cancer detected in routinely performed TZ directed biopsies to RP findings. We believe that the nearly 80% of cases with no
or nondominant tumors in the TZ at RP is enough to urge caution in interpretation of such biopsies, regardless of clinical circumstance. Furthermore, while we are aware that histopathological distinctions between transition and peripheral zone tumors on NB may not enter standard treatment algorithms, we demonstrate that these pathological features in TZ NBs do not predict the features of the tumor seen at prostatectomy. Additionally, TZ-LOOK histology has been widely described and used as a feature of TZ tumors, and we show that it does not predict TZ dominant lesions on biopsy.