- Email: [email protected]
Needle Biopsy Associated Tumor Tracking of Adenocarcinoma of the Prostate
0822-534 7/91/1453-1003$03.00/0 VoL 145,
T~E JOURNAL 'Ji< UROLOGY i() 1991 by AMERICAN lJROLOGICAL ASSOCIATION, TNC.
Copyright
NEEDLE BIOPSY ASSOCIATED TUMOR TRACKING OF ADENOCARCINOMA OF THE PROSTATE SHELDON S. BASTACKY, PATRICK C. WALSH AND JONATHAN L EPSTEIN From the Departments of Urology and Pathology, The Johns Hopkins University School of Medicine and James Buchanan Brady Urological Institute, The Johns Hopkins Hospital, Baltimore, Maryland
ABSTRACT
We reviewed 350 previously biopsied completely submitted clinical stage B radical prostatectomy specimens resected between January 1, 1987 and December 31, 1988 in an attempt to identify the incidence of needle biopsy associated tumor tracking into periprostatic soft tissue. We identified 7 cases (2.0%) of needle biopsy associated tumor tracking, 3 in which the only tumor penetration in the gland was limited to the needle track. The maximal soft tissue extension from the biopsy site ranged from 0.1 to 1.2 cm. and approached the nearest soft tissue margin to within 0.5 mm. in 4 cases. In contrast to prior reports showing clinically evident tracking only with transperineal biopsies from high grade tumors, 6 of our 7 cases were of intermediate grade (in the glandular and tracking components) and 6 had transrectal biopsies. Needle biopsy associated tumor tracking occurred with core (14 gauge) and biopsy gun needles (18 gauge). An additional 13 cases (3. 7%) showed some features of needle biopsy associated tumor tracking but they were equivocal. These findings have significant implications in light of recent proposals advocating serial mapping of prostate cancers using the biopsy gun with potential conservative observation of smaller tumors. KEY WORDS:
prostatic neoplasms; adenocarcinoma; biopsy, needle
Perinea! tumor seeding is an uncommon complication following transperineal needle biopsy of advanced stage adenocarcinoma of the prostate. 1 We have sporadically noted tracking of intermediate grade prostate cancer in radical prostatectomy specimens. In light of recent studies suggesting conservative observation for a select group of prostate cancer patients following needle biopsy,2 we decided to investigate the incidence and clinicopathological features of needle biopsy associated tumor tracking. MATERIALS AND METHODS
From January 1, 1987 to December 30, 1988, 364 patients underwent radical retropubic prostatectomy at our hospital for clinical stage B prostate adenocarcinoma in whom a prior transrectal or transperineal needle biopsy was performed. Fourteen cases were excluded because of subtotal submission of the surgical specimen, leaving 350 cases meeting the criteria for inclusion into this study. Following resection the intact prostate was weighed, coated over its entire surface with india ink and fixed in 10% buffered formalin for 18 to 24 hours. Following fixation distal (apical) Accepted for publication September 19, 1990.
and proximal (basal) shave urethral margins were removed for histological examination. The gland was then serially sectioned at 2 to 3 mm. intervals perpendicular to the long (apical-basal) axis of the gland. Each cross section of the prostate was further divided (that is into halves, quarters and so forth) to conform to the size of the cassette. The entire prostate was embedded and designated in a manner permitting localization of each section within the prostate. The tumor was graded using the Gleason grading system. 3 The histological sections were screened for identification of the needle biopsy site. The biopsy sites were characterized histologically by 1 or more of the following features: 1) a stellate parenchymal defect with surrounding fibrosis, 2) a localized area of scarring within the gland disrupting its normal architecture, 3) a capsular indentation with fibrosis or 4) capsular scar with periprostatic hemosiderin deposition. The biopsy track, unless oriented precisely parallel to the plane of tissue section, was often traceable serially over consecutive sections. Needle biopsy associated tumor tracking was characterized histologically the presence of isolated periprostatic neoplastic glands within dense fibrous tissue associated with hemosiderin deposition and tracking away from the glandular biopsy
FIG. LA, tumor contains irregular biopsy site defects (arrows) lined by cuboidal epithelium surrounded by fibrosis. Reduced from XlO. B, several histological sections proximal to tumor seen in part A. There were several malignant glands embedded within fibrous tissue around seminal vesicles containing numerous hemosiderin-laden macrophages, consistent with needle biopsy site tracking. Reduced from x35. 1003
1004
BASTACKY, WALSH AND EPSTEIN
Some prostates were suggestive of needle biopsy tracking yet they were considered equivocal because of either the presence of minimal capsular penetration apparently not related to the biopsy site, yet which also extended into the biopsy site region, or periprostatic soft tissue tumor in the region of the needle biopsy site, which was predominantly limited to perineural spaces. Prostatic specimens demonstrating extensive capsular penetration and concomitant involvement of the biopsy site region were considered nondiagnostic for needle biopsy tracking and were tabulated as negative. In cases diagnostic for needle biopsy tracking the method of biopsy (transrectal or transperineal), gauge of needle and interval between needle biopsy and radical prostatectomy were identified. Additional histological characteristics noted in eluded the assessment of tumor grade in the gland and within the tracking components, and the distance measured by ocular micrometer of the most peripheral tumor within the needle tracks to the inked soft tissue margin. RESULTS
FIG. 2. Single small nest of tumor (arrow) tracking away from biopsy site. Reduced from X30. Inset shows single small nest of tumor within fibrous tissue. Hemosiderin-laden macrophages (not illustrated) were present in surrounding tissue. Reduced from Xl65.
FIG. 3. Several nests of tumor within fibrosis containing hemosiderin (hemosiderin seen at higher magnification) are seen in periprostatic soft tissue (upper right). These nests of tumor are several histological sections distant from main tumor mass and are only site of capsular penetration in this case. Reduced from X30.
site toward the inked soft tissue margin. Care was taken to avoid over diagnosing needle biopsy tracking by excluding cases in which all of the capsular penetration (extension of tumor into periprostatic adipose tissue) was thought to be intrinsic to the tumor itself rather than resulting from the needle biopsy. Consequently, any case that showed perineural tumor invasion (a feature commonly seen with ordinary capsular penetration) in an area otherwise suggestive of needle biopsy tracking was diagnosed as showing only equivocal needle biopsy site tracking. In some cases with needle biopsy tracking the only site of capsular penetration in the entire prostate was in the region of the needle track. Needle biopsy tracking also was diagnosed in glands with nonbiopsy related capsular penetration as long as the capsular penetration in the region of the needle track was in a different site.
Seven cases (2 %) of needle biopsy tracking were identified, 3 in which the only area of capsular penetration was limited to the biopsy needle track (figs. 1 to 4). In the other 4 cases there was capsular penetration at the needle biopsy tracking site as well as capsular penetration elsewhere in the gland. The histological grade of the tracking component consistently correlated with that of the tumor within the gland. In 6 of these cases the tumor grade was intermediate (Gleason score 5 to 7) and in only 1 case was it high (Gleason score 9) grade. The maximum soft tissue extension of needle biopsy tracking tumor from the glandular biopsy site ranged from 1 to 12 mm. Needle biopsy site tracking tumor extended to within 0.5 mm. of the inked margin of resection in 4 cases. In all cases the tumor within the needle track was microscopic and only detected after careful pathological examination. Needle biopsy associated tumor tracking occurred with the core (14 gauge) and biopsy gun (18 gauge) needles. Six cases were associated with transrectal biopsy and 2 following a transperineal approach (1 patient was biopsied on 2 separate occasions, once transrectally and once transperineally) (see table). The interval between the initial biopsy and subsequent radical prostatectomy averaged 3.5 months with a range of 2 to 7 months. One patient had elevated serum prostate specific antigen (PSA) levels 3 years following radical prostatectomy in whom there was a positive margin unrelated to the needle tracking. None of the other patients has demonstrated either local or distant recurrence following radical prostatectomy, although the postoperative intervals have been short. An additional 13 cases (3. 7%) of equivocal needle biopsy associated tumor tracking were identified, including 12 with intermediate (Gleason score 5 to 7) grade and 1 with high (Gleason score 8 or greater) grade tumor in the glandular and tracking components (fig. 5). In several cases benign elements, such as nonneoplastic glands and corpora amylacea, were seen tracking away from the needle biopsy site (fig. 6). DISCUSSION
Implantation of tumor along a needle biopsy track is a recognized potential complication of this procedure. In 1953 Clarke et al reported the first case of locally seeded prostate cancer resulting in a discrete perinea! nodule. 4 Subsequently, 18 additional cases of prostate biopsy associated soft tissue tumor implantation have been described in the literature, primarily as single case reports. 1 , 4- 15 Of these cases 18 have occurred in the setting of prostatic adenocarcinoma and 1 involved a primary prostatic leiomyosarcoma. 6 All cases have been associated with transperineal needle biopsy and each has presented as a clinically discrete nodule ranging in size from
NEEDLE BIOPSY ASSOCIATED TUMOR TRACKING OF ADENOCARCINOMA OF PROSTATE
1005
FIG. 4. A, extensive fibrosis with hemosiderin was seen in area of biopsy site. Fibrotic biopsy site contained focus of high grade carcinoma (arrow), which was only site of capsular penetration in this case. Reduced from X30. B, higher magnification reveals poorly differentiated adenocarcinoma of prostate within fibrotic needle biopsy site showing hemosiderin deposition (arrow). Reduced from X440. Clinicopathological features of patients with biopsy associated tumor tracking Pt. No.
Route of Biopsy
Biopsy Needle Type
Needle Biopsy Grade
Prostatectomy Specimen Grade
Tumor Tracking Grade
Tumor Tracking Only Site of Capsular Penetration
1
Transrectal Transrectal Transperineal Transrectal Transrectal Transrectal Transperineal Transrectal
Core Core Core Gun Gun Gun Gun Gun
3+3=6 2+3=5 2+3=5 2+3=5 3+3=6 5+4=9 3+3=6 3+3=6
3+2=5 3+4=7 3+2=5
2+3=5 3+3=6 3+2=5
Neg. Neg. Pos.
3+3=6 5+4=9 3+2=5 3+3=6
3+3=6 5+4=9 3+3=6 3+3=6
Pos. Pos. Neg. Neg.
2 3 4 5 6 7
0.3 to 5 cm., with an average latency following biopsy of 16 months. 1 Only 4 previous studies have addressed the incidence of needle biopsy associated tumor seeding in prostate cancer, all of which were limited to the detection of a macroscopic perinea! nodule. 1•7•9 • 15 Burkholder and Kaufman,7 and Blackard et al9 reported the lower incidences of 0.15% and 0.2% among 1,351 and 522 biopsy positive patients, respectively. Moul et al demonstrated the highest incidence of clinical perinea! seeding (1 %) in a series of 502 biopsy positive patients. 1 Emtage and PerezMarrero reported an incidence of 0.16% among 575 biopsy positive and negative patients, underestimating the incidence because of the incorporation of negative biopsies into the denominator. 15 All cases except 1 were high (C or D) clinical stage. Our 2% incidence of needle biopsy associated tumor tracking was surprisingly high, especially given that our patient population was restricted to those with clinical stage B lesions. The incidence of needle biopsy associated tumor tracking seen in our study cannot be directly compared to prior studies of clinical macroscopic tumor seeding. All of our cases were microscopic and they were only detected following careful pathological examination of the entirely embedded radical prostatectomy specimen. Another reason for the discrepancy between the incidence of needle biopsy associated tumor tracking in our study and prior studies may be that the incidence of transperineal seedings in prior studies was reduced due to the sterilization of small tumor seedings by either radiotherapy or adjunctive hormonal therapy that was often administered to patients with high stage disease. Furthermore, our detection rate of needle biopsy tracking was improved as a result of our ability to distinguish microscopically those cases of needle biopsy tracking from those in which the confounding presence
FIG. 5. Several neoplastic glands showing capsular penetration (upper right) are seen in region of biopsy site defect (bottom center). In addition, there is single neoplastic gland (arrow) suggestive of needle tracking. Inset shows higher magnification of gland suggestive of needle biopsy tracking since it is embedded within fibrotic needle biopsy track containing numerous hemosiderin-laden macrophages. Because gland also shows perineural invasion and was adjacent to other glands showing capsular penetration away from needle biopsy site, case was labeled as equivocal for needle biopsy associated tumor tracking. Reduced from
xno.
of conventional tumor capsular penetration was present. It is for this reason that tumor tracking due to transrectal biopsy has probably never been reported, since clinically one cannot distinguish needle biopsy tracking following transrectal biopsy
1006
BASTACKY, WALSH AND EPSTEIN
FIG. 6. A, needle biopsy site in area distant from tumor demonstrates tracking of benign glands within fibrotic needle track (arrow). Reduced from x9. B, higher magnification shows several benign glands within fibrotic needle track. Reduced from X30.
from nonbiopsy related direct extension of prostate carcinoma posteriorly into the perirectal soft tissue. I6 -I 8 A recent study has evaluated a variety of proposed risk factors that have been retrospectively correlated with needle biopsy related tumor seeding. I Moul et al found clinical stage to be the most highly correlated parameter, since 17 of 18 cases (94%) in the literature with adenocarcinoma of the prostate and needle biopsy associated tumor tracking had high (C or D) clinical stage disease at biopsy. Only 1 clinical stage B case treated by total perineal prostatectomy demonstrated perineal recurrence 7 years later. 11 All of our patients had clinical stage B disease precluding an assessment of this variable as a risk factor; however, our data demonstrate that it is not rare to find microscopic seeding within the needle biopsy track in patients with lower clinical stage disease as well. Histological grade also has been regarded by several authors as an important risk factor. However, Moul et al, by summarizing data in the literature, found that only 11 of 19 patients (58%) had high grade tumor. I In our study 6 of 7 patients (85%) with needle biopsy tracking and 11 of 13 (88%) with equivocal needle biopsy tracking had intermediate grade histology, in part reflecting the distribution of histological grades in our patient population. In our patients as well as in the majority of patients reported on in the literature the histological grade of the tumor implant was similar to that of the glandular component. I We also identified rare cases of minimal implantation of benign glands and corpora amylacea into the periprostatic soft tissue. These findings suggest that seeding is neither limited to high grade tumor nor necessarily to neoplastic cells; it may represent a passive and relatively nonselective ectopic implantation and growth of biological material deposited by needle in the soft tissue. Moul et al noted the development of distant metastases synchronously or within 16 months of presentation of the macroscopic perineal nodules in all 6 of their patients, concluding that this phenomenon reflects aggressive tumor behavior and positively predicts disease progression.I Followup clinical data were available for 6 of our 7 patients with needle biopsy tracking. The post-biopsy interval ranged from 2 months to 3 years (5 patients followed for more than 1 year) with 5 of the 6 showing no evidence of local or distant recurrence and undetectable postoperative PSA levels. One patient with a positive margin unrelated to the needle biopsy tracking area had elevated serum PSA levels 3 years postoperatively and underwent radiation treatment. However, all of these patients had the tumor removed by radical surgery. Our data suggest that subclinical seeding is a more prevalent process than was formerly believed. In particular, tumor seeding within the needle track can occur following transrectal biopsy, a phenomenon that has never been previously recognized. Furthermore, our study also demonstrates the novel finding of seeding following the thin needle biopsy gun technique. It is noteworthy that 3 of our 7 patients with needle biopsy tracking
had tumor penetration limited to the needle biopsy track, upstaging the disease from pathological stage B to pathological stage C. In addition, the needle biopsy track component closely approached the soft tissue margin of resection in several cases. These findings have potentially significant implications in light of recent proposals advocating serially mapping of prostate cancer using the small caliber gun needle with potentially conservative observation of smaller tumors. 2 Subclinical microscopic seeding, as seen in our study, does not necessarily indicate obligative progression to a clinical macroscopic tumor nodule. It may be that for progression from microscopic to macroscopic tumor to occur in this situation, the tumor may have to be inherently more aggressive as reflected, for example, by high histological grade or advanced clinical stage. However, subclinical seeding must invariably precede the development of a macroscopic implant nodule. Given the relatively short-term intervals between the needle biopsies and the detection of tumor seeding in the radical prostatectomy specimens seen in the current study, our findings raise concern over the potential long-term risk of tumor tracking in patients who might be followed for many years with conservative therapy following needle biopsy of carcinoma of the prostate. Furthermore, there would be no way to monitor the incidence of post-transrectal biopsy tumor tracking following conservative treatment, since clinically one could not distinguish biopsy versus nonbiopsy related capsular penetration in the region of the transrectal biopsy site. Finally, we believe that our data should not be construed as evidence supporting the use of fine needle aspiration over the use of either larger gauge or biopsy gun needles for the diagnosis of adenocarcinoma of the prostate. The incidence of this phenomenon with needle biopsy techniques was low and, more importantly, there is no evidence to suggest unfavorable consequences of needle biopsy tracking if definitive therapy is instigated shortly after the needle biopsy diagnosis. We also do not know if needle biopsy tracking occurs following fine needle aspiration. Furthermore, fine needle aspiration of the prostate is a technique not without its own problems, especially in terms of the diagnostic difficulties that many pathologists have with interpreting this material. REFERENCES 1. Moul, J. W., Miles, B. J., Skoog, S. J. and McLeod, D. G.: Risk
factors for perinea! seeding of prostate cancer after needle biopsy. J. Urol., 142: 86, 1989. 2. 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. 3. Mellinger, G. T., Gleason, D. and Bailar, J., III: The histology and prognosis of prostatic cancer. J. Urol., 97: 331, 1967. 4. Clarke, B. G., Leadbetter, W. F. and Campbell, J. S.: Implantation of cancer of the prostate in site of perinea! needle biopsy: report of a case. J. Urol., 70: 937, 1953.
NEEDLE BIOPSY ASSOCIATED TUMOR TRACKING OF ADENOCARCINOMA OF PROSTATE
5. Goldman, E. J. and Samellas, W.: Local extension of carcinoma of the prostate following needle biopsy. J. Urol., 84: 575, 1960. 6. Fortunoff, S.: Needle biopsy of prostate: a review of 346 biopsies. J. Urol., 87: 159, 1962. 7. Burkholder, G. V. and Kaufman, J. J.: Local implantation of carcinoma of the prostate with percutaneous needle biopsy. J. Urol., 95: 801, 1966. 8. Labardini, M. M. and Nesbit, R. M.: Perinea! extension of adenocarcinoma of the prostate gland after punch biopsy. J. Urol., 97: 891, 1967. 9. Blackard, C. F., Soucheray, J. A. and Gleason, D. F.: Prostate needle biopsy with perinea! extension of adenocarcinoma. J. Urol., 106: 401, 1971. 10. Puigvert, A., Elizalde, C. and Matz, J. A.: Perinea! implantation of carcinoma of the prostate following needle biopsy: a case report. J. Urol., 107: 821, 1972. 11. Desai, S. G. and Woodruff, L. M.: Carcinoma of prostate. Local extension following perinea! needle biopsy. Urology, 3: 87, 1974.
1007
12. Addonizio, J. C. and Kapoor, S. N.: Perinea! seeding of prostatic carcinoma after needle biopsy. Urology, 8: 513, 1976. 13. Warden, S.S., Schellhammer, P. F. and el-Mahdi, A.: Well differentiated carcinoma of the prostate seeding a perinea! needle biopsy tract. Brit. J. Urol., 56: 436, 1984. 14. Brausi, M., Latini, A. and Palladini, P. D.: Local seeding of anaplastic carcinoma of prostate after needle biopsy. Urology, 27: 63, 1986. 15. Emtage, J. B. and Perez-Marrero, R.: Extension of carcinoma of prostate along perinea! needle biopsy tract. Urology, 27: 548, 1986. 16. Fry, D. E., Amin, M. and Harbrecht, P. J.: Rectal obstruction secondary to carcinoma of the prostate. Ann. Surg., 189: 488, 1979. 17. Goldfarb, S. and Leiter, E.: Invasion of the rectum by carcinoma of the prostate. Arch. Surg., 115: 1117, 1980. 18. Winter, C. C.: The problem of rectal involvement by prostatic cancer. Surg., Gynec. & Obst., 105: 136, 1957.
T~E JOURNAL 'Ji< UROLOGY i() 1991 by AMERICAN lJROLOGICAL ASSOCIATION, TNC.
Copyright
NEEDLE BIOPSY ASSOCIATED TUMOR TRACKING OF ADENOCARCINOMA OF THE PROSTATE SHELDON S. BASTACKY, PATRICK C. WALSH AND JONATHAN L EPSTEIN From the Departments of Urology and Pathology, The Johns Hopkins University School of Medicine and James Buchanan Brady Urological Institute, The Johns Hopkins Hospital, Baltimore, Maryland
ABSTRACT
We reviewed 350 previously biopsied completely submitted clinical stage B radical prostatectomy specimens resected between January 1, 1987 and December 31, 1988 in an attempt to identify the incidence of needle biopsy associated tumor tracking into periprostatic soft tissue. We identified 7 cases (2.0%) of needle biopsy associated tumor tracking, 3 in which the only tumor penetration in the gland was limited to the needle track. The maximal soft tissue extension from the biopsy site ranged from 0.1 to 1.2 cm. and approached the nearest soft tissue margin to within 0.5 mm. in 4 cases. In contrast to prior reports showing clinically evident tracking only with transperineal biopsies from high grade tumors, 6 of our 7 cases were of intermediate grade (in the glandular and tracking components) and 6 had transrectal biopsies. Needle biopsy associated tumor tracking occurred with core (14 gauge) and biopsy gun needles (18 gauge). An additional 13 cases (3. 7%) showed some features of needle biopsy associated tumor tracking but they were equivocal. These findings have significant implications in light of recent proposals advocating serial mapping of prostate cancers using the biopsy gun with potential conservative observation of smaller tumors. KEY WORDS:
prostatic neoplasms; adenocarcinoma; biopsy, needle
Perinea! tumor seeding is an uncommon complication following transperineal needle biopsy of advanced stage adenocarcinoma of the prostate. 1 We have sporadically noted tracking of intermediate grade prostate cancer in radical prostatectomy specimens. In light of recent studies suggesting conservative observation for a select group of prostate cancer patients following needle biopsy,2 we decided to investigate the incidence and clinicopathological features of needle biopsy associated tumor tracking. MATERIALS AND METHODS
From January 1, 1987 to December 30, 1988, 364 patients underwent radical retropubic prostatectomy at our hospital for clinical stage B prostate adenocarcinoma in whom a prior transrectal or transperineal needle biopsy was performed. Fourteen cases were excluded because of subtotal submission of the surgical specimen, leaving 350 cases meeting the criteria for inclusion into this study. Following resection the intact prostate was weighed, coated over its entire surface with india ink and fixed in 10% buffered formalin for 18 to 24 hours. Following fixation distal (apical) Accepted for publication September 19, 1990.
and proximal (basal) shave urethral margins were removed for histological examination. The gland was then serially sectioned at 2 to 3 mm. intervals perpendicular to the long (apical-basal) axis of the gland. Each cross section of the prostate was further divided (that is into halves, quarters and so forth) to conform to the size of the cassette. The entire prostate was embedded and designated in a manner permitting localization of each section within the prostate. The tumor was graded using the Gleason grading system. 3 The histological sections were screened for identification of the needle biopsy site. The biopsy sites were characterized histologically by 1 or more of the following features: 1) a stellate parenchymal defect with surrounding fibrosis, 2) a localized area of scarring within the gland disrupting its normal architecture, 3) a capsular indentation with fibrosis or 4) capsular scar with periprostatic hemosiderin deposition. The biopsy track, unless oriented precisely parallel to the plane of tissue section, was often traceable serially over consecutive sections. Needle biopsy associated tumor tracking was characterized histologically the presence of isolated periprostatic neoplastic glands within dense fibrous tissue associated with hemosiderin deposition and tracking away from the glandular biopsy
FIG. LA, tumor contains irregular biopsy site defects (arrows) lined by cuboidal epithelium surrounded by fibrosis. Reduced from XlO. B, several histological sections proximal to tumor seen in part A. There were several malignant glands embedded within fibrous tissue around seminal vesicles containing numerous hemosiderin-laden macrophages, consistent with needle biopsy site tracking. Reduced from x35. 1003
1004
BASTACKY, WALSH AND EPSTEIN
Some prostates were suggestive of needle biopsy tracking yet they were considered equivocal because of either the presence of minimal capsular penetration apparently not related to the biopsy site, yet which also extended into the biopsy site region, or periprostatic soft tissue tumor in the region of the needle biopsy site, which was predominantly limited to perineural spaces. Prostatic specimens demonstrating extensive capsular penetration and concomitant involvement of the biopsy site region were considered nondiagnostic for needle biopsy tracking and were tabulated as negative. In cases diagnostic for needle biopsy tracking the method of biopsy (transrectal or transperineal), gauge of needle and interval between needle biopsy and radical prostatectomy were identified. Additional histological characteristics noted in eluded the assessment of tumor grade in the gland and within the tracking components, and the distance measured by ocular micrometer of the most peripheral tumor within the needle tracks to the inked soft tissue margin. RESULTS
FIG. 2. Single small nest of tumor (arrow) tracking away from biopsy site. Reduced from X30. Inset shows single small nest of tumor within fibrous tissue. Hemosiderin-laden macrophages (not illustrated) were present in surrounding tissue. Reduced from Xl65.
FIG. 3. Several nests of tumor within fibrosis containing hemosiderin (hemosiderin seen at higher magnification) are seen in periprostatic soft tissue (upper right). These nests of tumor are several histological sections distant from main tumor mass and are only site of capsular penetration in this case. Reduced from X30.
site toward the inked soft tissue margin. Care was taken to avoid over diagnosing needle biopsy tracking by excluding cases in which all of the capsular penetration (extension of tumor into periprostatic adipose tissue) was thought to be intrinsic to the tumor itself rather than resulting from the needle biopsy. Consequently, any case that showed perineural tumor invasion (a feature commonly seen with ordinary capsular penetration) in an area otherwise suggestive of needle biopsy tracking was diagnosed as showing only equivocal needle biopsy site tracking. In some cases with needle biopsy tracking the only site of capsular penetration in the entire prostate was in the region of the needle track. Needle biopsy tracking also was diagnosed in glands with nonbiopsy related capsular penetration as long as the capsular penetration in the region of the needle track was in a different site.
Seven cases (2 %) of needle biopsy tracking were identified, 3 in which the only area of capsular penetration was limited to the biopsy needle track (figs. 1 to 4). In the other 4 cases there was capsular penetration at the needle biopsy tracking site as well as capsular penetration elsewhere in the gland. The histological grade of the tracking component consistently correlated with that of the tumor within the gland. In 6 of these cases the tumor grade was intermediate (Gleason score 5 to 7) and in only 1 case was it high (Gleason score 9) grade. The maximum soft tissue extension of needle biopsy tracking tumor from the glandular biopsy site ranged from 1 to 12 mm. Needle biopsy site tracking tumor extended to within 0.5 mm. of the inked margin of resection in 4 cases. In all cases the tumor within the needle track was microscopic and only detected after careful pathological examination. Needle biopsy associated tumor tracking occurred with the core (14 gauge) and biopsy gun (18 gauge) needles. Six cases were associated with transrectal biopsy and 2 following a transperineal approach (1 patient was biopsied on 2 separate occasions, once transrectally and once transperineally) (see table). The interval between the initial biopsy and subsequent radical prostatectomy averaged 3.5 months with a range of 2 to 7 months. One patient had elevated serum prostate specific antigen (PSA) levels 3 years following radical prostatectomy in whom there was a positive margin unrelated to the needle tracking. None of the other patients has demonstrated either local or distant recurrence following radical prostatectomy, although the postoperative intervals have been short. An additional 13 cases (3. 7%) of equivocal needle biopsy associated tumor tracking were identified, including 12 with intermediate (Gleason score 5 to 7) grade and 1 with high (Gleason score 8 or greater) grade tumor in the glandular and tracking components (fig. 5). In several cases benign elements, such as nonneoplastic glands and corpora amylacea, were seen tracking away from the needle biopsy site (fig. 6). DISCUSSION
Implantation of tumor along a needle biopsy track is a recognized potential complication of this procedure. In 1953 Clarke et al reported the first case of locally seeded prostate cancer resulting in a discrete perinea! nodule. 4 Subsequently, 18 additional cases of prostate biopsy associated soft tissue tumor implantation have been described in the literature, primarily as single case reports. 1 , 4- 15 Of these cases 18 have occurred in the setting of prostatic adenocarcinoma and 1 involved a primary prostatic leiomyosarcoma. 6 All cases have been associated with transperineal needle biopsy and each has presented as a clinically discrete nodule ranging in size from
NEEDLE BIOPSY ASSOCIATED TUMOR TRACKING OF ADENOCARCINOMA OF PROSTATE
1005
FIG. 4. A, extensive fibrosis with hemosiderin was seen in area of biopsy site. Fibrotic biopsy site contained focus of high grade carcinoma (arrow), which was only site of capsular penetration in this case. Reduced from X30. B, higher magnification reveals poorly differentiated adenocarcinoma of prostate within fibrotic needle biopsy site showing hemosiderin deposition (arrow). Reduced from X440. Clinicopathological features of patients with biopsy associated tumor tracking Pt. No.
Route of Biopsy
Biopsy Needle Type
Needle Biopsy Grade
Prostatectomy Specimen Grade
Tumor Tracking Grade
Tumor Tracking Only Site of Capsular Penetration
1
Transrectal Transrectal Transperineal Transrectal Transrectal Transrectal Transperineal Transrectal
Core Core Core Gun Gun Gun Gun Gun
3+3=6 2+3=5 2+3=5 2+3=5 3+3=6 5+4=9 3+3=6 3+3=6
3+2=5 3+4=7 3+2=5
2+3=5 3+3=6 3+2=5
Neg. Neg. Pos.
3+3=6 5+4=9 3+2=5 3+3=6
3+3=6 5+4=9 3+3=6 3+3=6
Pos. Pos. Neg. Neg.
2 3 4 5 6 7
0.3 to 5 cm., with an average latency following biopsy of 16 months. 1 Only 4 previous studies have addressed the incidence of needle biopsy associated tumor seeding in prostate cancer, all of which were limited to the detection of a macroscopic perinea! nodule. 1•7•9 • 15 Burkholder and Kaufman,7 and Blackard et al9 reported the lower incidences of 0.15% and 0.2% among 1,351 and 522 biopsy positive patients, respectively. Moul et al demonstrated the highest incidence of clinical perinea! seeding (1 %) in a series of 502 biopsy positive patients. 1 Emtage and PerezMarrero reported an incidence of 0.16% among 575 biopsy positive and negative patients, underestimating the incidence because of the incorporation of negative biopsies into the denominator. 15 All cases except 1 were high (C or D) clinical stage. Our 2% incidence of needle biopsy associated tumor tracking was surprisingly high, especially given that our patient population was restricted to those with clinical stage B lesions. The incidence of needle biopsy associated tumor tracking seen in our study cannot be directly compared to prior studies of clinical macroscopic tumor seeding. All of our cases were microscopic and they were only detected following careful pathological examination of the entirely embedded radical prostatectomy specimen. Another reason for the discrepancy between the incidence of needle biopsy associated tumor tracking in our study and prior studies may be that the incidence of transperineal seedings in prior studies was reduced due to the sterilization of small tumor seedings by either radiotherapy or adjunctive hormonal therapy that was often administered to patients with high stage disease. Furthermore, our detection rate of needle biopsy tracking was improved as a result of our ability to distinguish microscopically those cases of needle biopsy tracking from those in which the confounding presence
FIG. 5. Several neoplastic glands showing capsular penetration (upper right) are seen in region of biopsy site defect (bottom center). In addition, there is single neoplastic gland (arrow) suggestive of needle tracking. Inset shows higher magnification of gland suggestive of needle biopsy tracking since it is embedded within fibrotic needle biopsy track containing numerous hemosiderin-laden macrophages. Because gland also shows perineural invasion and was adjacent to other glands showing capsular penetration away from needle biopsy site, case was labeled as equivocal for needle biopsy associated tumor tracking. Reduced from
xno.
of conventional tumor capsular penetration was present. It is for this reason that tumor tracking due to transrectal biopsy has probably never been reported, since clinically one cannot distinguish needle biopsy tracking following transrectal biopsy
1006
BASTACKY, WALSH AND EPSTEIN
FIG. 6. A, needle biopsy site in area distant from tumor demonstrates tracking of benign glands within fibrotic needle track (arrow). Reduced from x9. B, higher magnification shows several benign glands within fibrotic needle track. Reduced from X30.
from nonbiopsy related direct extension of prostate carcinoma posteriorly into the perirectal soft tissue. I6 -I 8 A recent study has evaluated a variety of proposed risk factors that have been retrospectively correlated with needle biopsy related tumor seeding. I Moul et al found clinical stage to be the most highly correlated parameter, since 17 of 18 cases (94%) in the literature with adenocarcinoma of the prostate and needle biopsy associated tumor tracking had high (C or D) clinical stage disease at biopsy. Only 1 clinical stage B case treated by total perineal prostatectomy demonstrated perineal recurrence 7 years later. 11 All of our patients had clinical stage B disease precluding an assessment of this variable as a risk factor; however, our data demonstrate that it is not rare to find microscopic seeding within the needle biopsy track in patients with lower clinical stage disease as well. Histological grade also has been regarded by several authors as an important risk factor. However, Moul et al, by summarizing data in the literature, found that only 11 of 19 patients (58%) had high grade tumor. I In our study 6 of 7 patients (85%) with needle biopsy tracking and 11 of 13 (88%) with equivocal needle biopsy tracking had intermediate grade histology, in part reflecting the distribution of histological grades in our patient population. In our patients as well as in the majority of patients reported on in the literature the histological grade of the tumor implant was similar to that of the glandular component. I We also identified rare cases of minimal implantation of benign glands and corpora amylacea into the periprostatic soft tissue. These findings suggest that seeding is neither limited to high grade tumor nor necessarily to neoplastic cells; it may represent a passive and relatively nonselective ectopic implantation and growth of biological material deposited by needle in the soft tissue. Moul et al noted the development of distant metastases synchronously or within 16 months of presentation of the macroscopic perineal nodules in all 6 of their patients, concluding that this phenomenon reflects aggressive tumor behavior and positively predicts disease progression.I Followup clinical data were available for 6 of our 7 patients with needle biopsy tracking. The post-biopsy interval ranged from 2 months to 3 years (5 patients followed for more than 1 year) with 5 of the 6 showing no evidence of local or distant recurrence and undetectable postoperative PSA levels. One patient with a positive margin unrelated to the needle biopsy tracking area had elevated serum PSA levels 3 years postoperatively and underwent radiation treatment. However, all of these patients had the tumor removed by radical surgery. Our data suggest that subclinical seeding is a more prevalent process than was formerly believed. In particular, tumor seeding within the needle track can occur following transrectal biopsy, a phenomenon that has never been previously recognized. Furthermore, our study also demonstrates the novel finding of seeding following the thin needle biopsy gun technique. It is noteworthy that 3 of our 7 patients with needle biopsy tracking
had tumor penetration limited to the needle biopsy track, upstaging the disease from pathological stage B to pathological stage C. In addition, the needle biopsy track component closely approached the soft tissue margin of resection in several cases. These findings have potentially significant implications in light of recent proposals advocating serially mapping of prostate cancer using the small caliber gun needle with potentially conservative observation of smaller tumors. 2 Subclinical microscopic seeding, as seen in our study, does not necessarily indicate obligative progression to a clinical macroscopic tumor nodule. It may be that for progression from microscopic to macroscopic tumor to occur in this situation, the tumor may have to be inherently more aggressive as reflected, for example, by high histological grade or advanced clinical stage. However, subclinical seeding must invariably precede the development of a macroscopic implant nodule. Given the relatively short-term intervals between the needle biopsies and the detection of tumor seeding in the radical prostatectomy specimens seen in the current study, our findings raise concern over the potential long-term risk of tumor tracking in patients who might be followed for many years with conservative therapy following needle biopsy of carcinoma of the prostate. Furthermore, there would be no way to monitor the incidence of post-transrectal biopsy tumor tracking following conservative treatment, since clinically one could not distinguish biopsy versus nonbiopsy related capsular penetration in the region of the transrectal biopsy site. Finally, we believe that our data should not be construed as evidence supporting the use of fine needle aspiration over the use of either larger gauge or biopsy gun needles for the diagnosis of adenocarcinoma of the prostate. The incidence of this phenomenon with needle biopsy techniques was low and, more importantly, there is no evidence to suggest unfavorable consequences of needle biopsy tracking if definitive therapy is instigated shortly after the needle biopsy diagnosis. We also do not know if needle biopsy tracking occurs following fine needle aspiration. Furthermore, fine needle aspiration of the prostate is a technique not without its own problems, especially in terms of the diagnostic difficulties that many pathologists have with interpreting this material. REFERENCES 1. Moul, J. W., Miles, B. J., Skoog, S. J. and McLeod, D. G.: Risk
factors for perinea! seeding of prostate cancer after needle biopsy. J. Urol., 142: 86, 1989. 2. 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. 3. Mellinger, G. T., Gleason, D. and Bailar, J., III: The histology and prognosis of prostatic cancer. J. Urol., 97: 331, 1967. 4. Clarke, B. G., Leadbetter, W. F. and Campbell, J. S.: Implantation of cancer of the prostate in site of perinea! needle biopsy: report of a case. J. Urol., 70: 937, 1953.
NEEDLE BIOPSY ASSOCIATED TUMOR TRACKING OF ADENOCARCINOMA OF PROSTATE
5. Goldman, E. J. and Samellas, W.: Local extension of carcinoma of the prostate following needle biopsy. J. Urol., 84: 575, 1960. 6. Fortunoff, S.: Needle biopsy of prostate: a review of 346 biopsies. J. Urol., 87: 159, 1962. 7. Burkholder, G. V. and Kaufman, J. J.: Local implantation of carcinoma of the prostate with percutaneous needle biopsy. J. Urol., 95: 801, 1966. 8. Labardini, M. M. and Nesbit, R. M.: Perinea! extension of adenocarcinoma of the prostate gland after punch biopsy. J. Urol., 97: 891, 1967. 9. Blackard, C. F., Soucheray, J. A. and Gleason, D. F.: Prostate needle biopsy with perinea! extension of adenocarcinoma. J. Urol., 106: 401, 1971. 10. Puigvert, A., Elizalde, C. and Matz, J. A.: Perinea! implantation of carcinoma of the prostate following needle biopsy: a case report. J. Urol., 107: 821, 1972. 11. Desai, S. G. and Woodruff, L. M.: Carcinoma of prostate. Local extension following perinea! needle biopsy. Urology, 3: 87, 1974.
1007
12. Addonizio, J. C. and Kapoor, S. N.: Perinea! seeding of prostatic carcinoma after needle biopsy. Urology, 8: 513, 1976. 13. Warden, S.S., Schellhammer, P. F. and el-Mahdi, A.: Well differentiated carcinoma of the prostate seeding a perinea! needle biopsy tract. Brit. J. Urol., 56: 436, 1984. 14. Brausi, M., Latini, A. and Palladini, P. D.: Local seeding of anaplastic carcinoma of prostate after needle biopsy. Urology, 27: 63, 1986. 15. Emtage, J. B. and Perez-Marrero, R.: Extension of carcinoma of prostate along perinea! needle biopsy tract. Urology, 27: 548, 1986. 16. Fry, D. E., Amin, M. and Harbrecht, P. J.: Rectal obstruction secondary to carcinoma of the prostate. Ann. Surg., 189: 488, 1979. 17. Goldfarb, S. and Leiter, E.: Invasion of the rectum by carcinoma of the prostate. Arch. Surg., 115: 1117, 1980. 18. Winter, C. C.: The problem of rectal involvement by prostatic cancer. Surg., Gynec. & Obst., 105: 136, 1957.