Differences in tumor core distribution between palpable and nonpalpable prostate tumors in patients diagnosed using extensive transperineal ultrasound-guided template prostate biopsy

Authors


  • See related editorial on pages 1761–2, this issue.

Abstract

BACKGROUND

The authors performed extensive transperineal ultrasound-guided template prostate biopsies to investigate carcinoma core distribution.

METHODS

Between August 2000 and May 2004, 371 men underwent template biopsies. Three hundred twelve patients had not undergone a previous biopsy (first group) and 59 had undergone previous transrectal sextant biopsies (repeat group). Of the 312 patients in the first group, 236 had normal digital rectal examination (DRE) findings (DRE– first group) and 76 patients had an abnormal DRE (DRE+ first group). A mean of 20.1 biopsy cores (range, 9–38 cores) was taken from the entire prostate. The region > 2.0 cm from the rectal face of the prostate was defined as the anterior region and the remaining area was defined as the posterior region.

RESULTS

In the DRE– first group, the carcinoma core rate (number of tumor cores/number of biopsy cores) in the anterior region (7.2%) did not differ from that of the posterior region (7.3%) (P = 0.9635). However, in the DRE+ first group, the carcinoma core rate in the posterior region (22.0%) was found to be higher than in the anterior region (13.2%) (P < 0.0001). In the repeat group, the carcinoma core rate in the posterior region (3.1%) was significantly (P = 0.0008) lower than that exhibited in the anterior region (7.2%).

CONCLUSIONS

The results of the current study suggest that nonpalpable prostate carcinoma is distributed equally within the entire prostate, although palpable carcinoma is distributed mainly in the posterior region and many of the tumor foci in the anterior region may be missed by a transrectal sextant biopsy. The examination of radical prostatectomy specimens is required to prove these results. Cancer 2005. © 2005 American Cancer Society.

Since Hodge et al. developed the technique of transrectal ultrasound-guided sextant prostate biopsy it has been considered the gold standard for the early detection of prostate carcinoma.1 However, many investigators have recommended transition zone biopsies plus sextant biopsies or more extensive field biopsies for all patients, as well as those individuals at high risk for prostate carcinoma despite having had a previous negative biopsy.2–6 High rates of detection for prostate carcinoma on extended field biopsies have been reported in patients with a previous negative biopsy. Furthermore, Stewart et al.7 developed a saturation needle biopsy method for those patients. They hypothesized that increasing the number of biopsy cores may improve the detection rate in men with persistent indications for a repeat biopsy. A tumor must be supposed to be distributed equally throughout the prostate so that the detection rate may improve by increasing the number of biopsy cores, but to our knowledge this has not been proven to date. Recently, brachytherapy guided by transrectal ultrasound and assisted by a grid template has been performed widely for the treatment of prostate carcinoma. We have adapted the template grid technique for prostate biopsy to detect early-stage prostate carcinoma in men who have not undergone a previous biopsy, as well as those who had previous negative biopsies and remained at high risk for prostate carcinoma.8 In the current study, we analyzed the distribution of carcinoma cores in the prostate with systematic ultrasound-guided transperineal template biopsy to demonstrate whether prostate carcinoma is distributed equally throughout the entire prostate.

MATERIALS AND METHODS

Between August 2000 and May 2004, 371 men ranging in age from 43–87 years (mean age of 67.4 years) underwent systematic ultrasound-guided biopsy using the transperineal template technique. Of these 371 patients, 312 had undergone no previous prostate biopsy (first biopsy group) and 59 had undergone previous transrectal sextant biopsies (repeat biopsy group). Of the 312 men in the first biopsy group, 236 were found to have normal digital rectal examination (DRE) findings (DRE– first biopsy group) and 76 were found to have an abnormal DRE (DRE+ first biopsy group). Table 1 shows patient characteristics of the DRE– and DRE+ first biopsy groups, as well as the repeat biopsy group. The mean ages and PSA levels in the patients in the DRE+ first biopsy group were found to be significantly higher than those in the DRE– first biopsy group. The mean prostate volume and the number of biopsy cores in the DRE– first biopsy group were significantly higher than those in the DRE+ first biopsy group, whereas the mean prostate volume divided by the number of biopsy cores was not found to be significantly different between the two groups. The PSA level was measured using an AxSYM assay system (Abbott Laboratories, North Chicago, IL).

Table 1. Patient Characteristics
 DRE− first biopsy group (Group A)DRE+ first biopsy group (Group B)Repeat biopsy groupP value (Group A vs. Group B)
  1. DRE: digital rectal examination; DRE-first biopsy group: group with normal digital rectal examination results; DRE+ first biopsy group: group with abnormal digital rectal examination results; SD: standard deviation; PSA: prostate-specific antigen.

No. of patients2367659 
Age (yrs) (mean ± SD)66.9 ± 7.269.8 ± 8.066.4 ± 7.50.0039
PSA (ng/mL) (mean ± SD)7.91 ± 5.1416.36 ± 18.6518.03 ± 62.710.0034
Prostate volume (mL) (mean ± SD)42.5 ± 18.639.3 ± 21.356.8 ± 28.30.0410
No. of biopsy cores (mean ± SD)20.4 ± 4.118.9 ± 4.320.8 ± 4.90.0100
Prostate volume/no. of biopsy cores (mean ± SD)2.08 ± 0.852.02 ± 0.872.69 ± 1.140.4398

Patients had received fluoroquinolone antibiotics for 5 days beginning the day before the procedure. All procedures were performed in the operating room with the patient in the lithotomy position receiving spinal anesthesia. A 14-French Foley catheter was placed to easily detect the urethra on the ultrasound image, and was removed the next morning. After DRE, a Bruel & Kjaer transrectal ultrasound probe 8551 covered with a standard condom containing scanning gel was inserted into the rectum and then was mounted in the Bruel & Kjaer stepping unit UA1084 (Bruel & Kjaer North America, Inc., Norcross, GA). The prostate was scanned transversely from base to apex using the stepping unit. A step size of 5 mm was selected. The matrix pattern was superimposed onto the transverse image 1 cm proximal from the apex. There were scales of A to G with 1-cm intervals on the x axis, and scales of 1–7 with 1-cm intervals on the y axis. The stepping unit was set as the D-line on the urethra and 1-line 0.5-cm upward from the rectal face of the prostate. Because in the majority of cases the urethra was located on the D-2 spot, we defined the area over the 2-line as the anterior region and the remaining area as the posterior region, as described previously.8 Our definition results in the anterior region of the prostate including the majority of the transition zone and the far lateral peripheral zone, whereas the posterior region includes the remainder of the transition zone and the majority of the peripheral zone. An 18-gauge Tru-Cut® biopsy needle (ACN™ Biopsy Needle; Medical Device Technologies, Inc., Gainesville, FL), was inserted through the corresponding guide channel of the template and a mean of 20.1 biopsy cores (range, 9–38 cores) were taken in each biopsy set using a Magnum Biopty gun (C. R. Bard, Inc., Murray Hill, NJ). Cores were labeled according to the template grid locations, which correspond to D-1.0, B-2.5, etc. Guide channels in which the biopsy needle was inserted were determined systematically and distributed equally throughout the entire prostate. Consequently, the biopsy spots on the transverse image of the prostate formed a diamond shape in the majority of cases (Fig. 1). The prostate volume was calculated as follows: volume = height × width × length × 0.5236, with the height and width being measured in the transverse view and the length being measured in the longitudinal view.

Figure 1.

The matrix pattern was superimposed onto the transverse ultrasound image 1 cm proximal from the apex. Biopsy channels were selected like as a diamond shape. The area over the 2-line was defined as the anterior region and the area under the 2-line was defined as the posterior region.

The Mann–Whitney U test was used to evaluate the statistical significance between the two groups. The chi-square test was used for the comparison of frequencies. A value of P < 0.05 was considered significant.

RESULTS

Prostate carcinoma was detected in 180 of the total 371 patients (49%). It was detected in 159 of the 312 patients in the first biopsy group (51%) and in 21 of the 59 patients in the repeat biopsy group (36%). Prostate carcinoma was detected in 111 of the 236 patients in the DRE– first biopsy group (47%) and in 48 of the 76 patients in the DRE+ first biopsy group (63%). When patients in the DRE– first biopsy group were stratified into PSA categories of 4–5.99 ng/mL, 6–7.99 ng/mL, 8–9.99 ng/mL, and ≥ 10 ng/mL, the carcinoma detection rate was 38% (44 of 115 patients), 56% (28 of 50 patients), 42% (11 of 26 patients), and 64% (28 of 44 patients), respectively. When patients in the DRE+ first biopsy group were stratified into PSA categories of < 4 ng/mL, 4–5.99 ng/mL, 6–7.99 ng/mL, 8–9.99 ng/mL, and ≥ 10 ng/mL, the carcinoma detection rate was 11% (1 of 9 patients), 41% (7 of 17 patients), 75% (9 of 12 patients), 80% (8 of 10 patients), and 82% (23 of 28 patients), respectively.

From 371 patients, 7452 cores were obtained, 678 of which (9.1%) were found to contain evidence of prostate carcinoma. Of the 4806 biopsy cores taken in the DRE- first biopsy group, 349 (7.3%) contained evidence of prostate carcinoma. In the DRE+ first biopsy group, the number of biopsy cores obtained was 1428, 273 of which (19.1%) were found to contain prostate carcinoma. In the repeat biopsy group, the number of biopsy cores was 1224, and 57 (4.7%) cores contained carcinoma. Tables 2–4 show the number of biopsy cores obtained and the cores found to contain carcinoma at each channel on the template in the DRE– first, DRE+ first, and repeat biopsy groups. In the DRE– first biopsy group, the biopsy cores were symmetrically distributed, with the exception of the circumference of the urethra, although the number of biopsy cores appeared to decrease in peripheral areas. Carcinoma cores in the DRE– first biopsy group also were found to be symmetrically distributed within the entire prostate, with the exception of line 4.0, at which the anterior fibromuscular stroma was distributed in most cases not involving a considerably larger prostate, and therefore no carcinoma was observed. In the DRE– first biopsy group, the ratios of the number of carcinoma cores to the number of biopsy cores obtained (referred to as the carcinoma core rate) of Lines 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, and 4.0 were 6.9%, 7.1%, 7.9%, 7.2%, 8.0%, 6.9%, and 0%, respectively (Table 2). The carcinoma core rates of all lines, with the exception of Line 4.0, were similar. Consequently, the carcinoma core rate in the anterior region of the prostate (135 of 1865 biopsy cores obtained, or 7.2%) was not found to be significantly different (P = 0.9635) from that of the posterior region (214 of 2941 biopsy cores obtained or 7.3%). The carcinoma core rates in the DRE+ first biopsy group of Lines 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, and 4.0 were 22.5%, 22.1%, 21.5%, 13.0%, 16.9%, 3.9%, and 0%, respectively (Table 3). The carcinoma core rate in the posterior region (211 of 958 biopsy cores obtained or 22.0%) was found to be significantly (P < 0.0001) higher than in the anterior region (62 of 470 biopsy cores obtained or 13.2%). Of 273 carcinoma-positive cores in this group, 211 (77%) were found in the posterior region. The carcinoma core rates in the repeat biopsy group of Lines 1.0, 1.5, 2.0, 2.5, 3.0 3.5, and 4.0 were 2.5%, 2.5%, 4.3%, 5.4%, 10.1%, 7.1%, and 3.7%, respectively (Table 4). The carcinoma core rate gradually increased as the line rose from Line 1.0 to 3.0, and then decreased. Consequently, the carcinoma core rate in the anterior region (34 of 472 biopsy cores obtained or 7.2%) was significantly (P = 0.0008) higher than in the posterior region (23 of 752 biopsy cores obtained or 3.1%). Of 57 carcinoma-positive cores, 34 (60%) were distributed in the anterior region of the prostate. There was no significant difference in the carcinoma core rates noted between the right lobe and the left lobe of the prostate among all groups.

Table 2. No. of Carcinoma-Positive Cores and Biopsy Cores, and the Carcinoma Core Rate at Each Channel on the Template in the DRE− First Biopsy Group
 AABBBCCCDDDEEEFFFGTotal 
  1. DRE: digital rectal examination; DRE− first biopsy group: group with normal digital rectal examination results.

4.0  0/4 (0%)0/1 (0%)0/16 (0%)0/5 (0%) 0/2 (0%)0/16 (0%)0/2 (0%)0/2 (0%) 0/48 (0%) 
3.5  0/3 (0%)2/36 (6%)0/12 (0%)9/98 (9%) 9/100 (9%)1/8 (13%)0/49 (0%)  21/306 (6.9%) 
3.0  1/72 (1%)2/33 (6%)12/194 (6%)3/41 (7%) 6/31 (19%)24/191 (13%)3/35 (9%)2/68 (3%) 53/665 (8.0%) 
2.5 0/4 (0%)1/18 (6%)10/187 (5%)3/33 (9%)15/184 (8%) 13/175 (7%)3/33 (9%)16/195 (8%)0/12 (0%)0/5 (0%)61/846 (7.2%)135/1865 (7.2%)
2.0 0/2 (0%)4/162 (2%)3/46 (7%)23/210 (11%)3/34 (9%)0/6 (%)3/24 (13%)16/205 (8%)3/40 (8%)15/156 (10%) 70/885 (7.9%) 
1.5 0/11 (0%)1/26 (4%)15/198 (8%)8/34 (24%)11/194 (6%)1/61 (2%)8/200 (4%)3/37 (8%)23/205 (11%)1/22 (5%)0/10 (0%)71/998 (7.1%) 
1.01/1 (100%)0/1 (0%)8/159 (5%)3/44 (7%)20/211 (9%)5/39 (13%)7/185 (4%)3/36 (8%)17/209 (8%)2/34 (6%)7/137 (5%)0/2 (0%)73/1058 (6.9%)214/2941 (7.3%)
Total1/1 (100%)0/18 (0%)15/444 (3.4%)35/545 (6.4%)66/710 (9.3%)46/595 (7.7%) 42/568 (7.4%)64/699 (9.2%)47/560 (8.4%)25/397 (6.3%)0/17 (0%)  
 163/2312 (7.0%)8/252 (3.2%)178/2241 (7.9%)349/4806 (7.3%) 
Table 3. No. of Carcinoma-Positive Cores and Biopsy Cores, and the Carcinoma Core Rate at Each Channel on the Template in the DRE+ First Biopsy Group
 AABBBCCCDDDEEEFFFGGTotal 
  1. DRE: digital rectal examination; DRE+ first biopsy group: group with abnormal digital rectal examination results.

4.0  0/1 (0%) 0/2 (0%)   0/2 (0%)0/1 (0%)   0/6 (0%) 
3.5   0/7 (0%)0/4 (0%)1/14 (7%) 0/14 (0%)0/4 (0%)1/6 (17%)0/1 (0%)0/1 (0%) 2/51 (3.9%) 
3.0 0/1 (0%)2/14 (14%)1/6 (17%)9/49 (18%)2/13 (15%)1/1 (100%)3/11 (27%)7/53 (13%)2/7 (29%)1/11 (9%)  28/166 (16.9%) 
2.5 0/1 (0%)14 (25%)4/49 (8%)2/21 (10%)8/51 (16%) 4/45 (9%)3/18 (17%)9/50 (18%)0/7 (0%)1/1 (100%) 32/247 (13.0%)62/470 (13.2%)
2.0 0/1 (0%)7/54 (13%)6/18 (33%)14/64 (22%)3/13 (23%)1/3 (33%)3/18 (17%)17/63 (27%)5/15 (33%)7/44 (16%)  63/293 (21.5%) 
1.5 0/2 (0%)3/10 (30%)7/57 (12%)13/29 (45%)12/59 (20%)5/23 (22%)10/55 (18%)7/23 (30%)14/60 (23%)1/8 (13%)1/5 (20%) 73/331 (22.1%) 
1.00/1 (0%) 7/45 (16%)8/20 (40%)10/66 (15%)6/14 (43%)12/49 (24%)6/19 (32%)13/63 (21%)4/17 (24%)9/38 (24%) 0/2 (0%)75/334 (22.5%)211/958 (22.0%)
Total0/1 (0%)0/5 (0%)20/128 (15.6%)26/157 (16.6%)48/235 (20.4%)32/164 (19.5%) 26/162 (16.0%)47/226 (20.8%)35/156 (22.4%)18/109 (16.5%)2/7 (28.6%)0/2 (0%)  
 126/690 (18.3%)19/76 (25.0%)128/662 (19.3%)273/1428 (19.1%) 
Table 4. No. of Carcinoma-Positive Cores and Biopsy Cores, and the Carcinoma Core Rate at Each Channel on the Template in the Repeat Biopsy Group
 AABBBCCCDDDEEEFFFGGTotal 
4.0  0/1 (0%)0/1 (0%)0/8 (0%)0/5 (0%) 0/3 (0%)0/7 (0%)0/1 (0%)1/1 (100%)  1/27 (3.7%) 
3.5  0/2 (0%)0/2 (0%)0/6 (0%)1/21 (5%) 1/19 (5%)0/7 (0%)2/15 (13%) 0/1 (0%) 6/84 (7.1%) 
3.0 0/1 (0%)2/16 (13%)1/13 (8%)5/43 (12%)2/12 (17%) 0/5 (0%)5/44 (11%)0/12 (0%)1/14 (7%)  16/159 (10.1%) 
2.5 0/3 (0%)1/9 (11%)1/37 (3%)1/18 (6%)3/38 (8%) 3/34 (9%)0/13 (0%)2/39 (5%)0/9 (0%)0/2 (0%) 11/202 (5.4%)34/472 (7.2%)
2.0 0/1 (0%)1/34 (3%)0/21 (0%)3/46 (7%)0/14 (0%)0/1 (0%)0/12 (0%)3/49 (6%)0/19 (0%)3/36 (8%)  10/233 (4.3%) 
1.5 0/5 (0%)0/11 (0%)0/41 (0%)0/20 (0%)1/41 (2%)0/13 (0%)1/39 (3%)0/16 (0%)3/41 (7%)1/11 (9%)0/4 (0%) 6/242 (2.5%) 
1.00/1 (0%)0/1 (0%)2/36 (6%)1/22 (5%)2/45 (4%)0/21 (0%)0/43 (0%)1/18 (6%)1/46 (2%)0/15 (0%)0/27 (0%)0/1 (0%)0/1 (0%)7/277 (2.5%)23/752 (3.1%)
Total0/1 (0%)0/11 (0%)6/109 (5.5%)5/147 (3.4%)11/186 (5.9%)7/152 (4.6%) 6/130 (4.6%)9/182 (4.9%)7/142 (4.9%)6/98 (6.1%)0/8 (0%)0/1 (0%)  
 29/606 (4.8%)0/57 (0%)28/561 (5.0%)57/1224 (4.7%) 

Complications were reported to be minor and rare, and occurred in 8 of the 371 patients (2.2%). No patient reported a high fever, and all instances of hematuria were improved conservatively. Complications included hematuria and urinary retention that required overnight catheterization in 6 patients, continuous (lasting > 1 month) hematospermia in 1 patient, and continuous hematuria and anal pain in one patient.

DISCUSSION

Many studies of tumor distribution within zones of the prostate have been reported to date.9–12 Epstein et al.11 reported that in 85% of nonpalpable prostate carcinoma cases diagnosed using a transrectal biopsy (classified as T1c), the major tumor mass was located in the peripheral zone. Chen et al. used computer modeling and reported that 74% of tumor foci were located in the peripheral zone.12 In their study. 87% of the specimens (156 of 180 specimens) were palpable tumors, with the data being based on radical prostatectomy specimens taken from the patients who were diagnosed using transrectal sextant biopsy. In contrast, Igel et al. used the transperineal template technique and reported that prostate carcinoma was identified in the transition zone area in 76% of patients (29 of 38 patients) who had previously undergone transrectal prostate biopsies and had negative results.13 These results appear to be inconsistent, but we believe we are able to offer an explanation for the apparent discrepancy between them. In the DRE+ first biopsy group, the carcinoma core rate in the posterior region of the prostate was significantly higher than in the anterior region. Of 273 carcinoma-positive biopsies, 211 (77%) were in the posterior region, suggesting that palpable carcinomas are distributed more often in the posterior region compared with the anterior region, a finding that is consistent with the result reported by Chen et al.12 However, in the repeat biopsy group, the carcinoma core rate in the posterior region was found to be significantly lower than in the anterior region, and 60% of the carcinoma-positive biopsies (34 of 57 biopsies) were taken from the anterior region, a finding that is largely consistent with the results reported by Igel et al.13 These results suggested that a segment of tumors occurring in the anterior region (mostly the transition zone) had been missed by previous transrectal sextant biopsies, although many of the tumors occurring in the posterior region (mostly the peripheral zone) had been detected. We demonstrated that carcinoma-positive biopsy cores are distributed equally throughout the entire prostate in patients with normal DRE findings and no previous prostate biopsy, which is quite different from the result reported by Epstein et al.11 Emiliozzi et al. reported that transperineal biopsy missed fewer tumors than transrectal biopsy, and that it was superior to transrectal biopsy for the detection of early prostate carcinoma.14 It is probable that transrectal biopsies missed more tumors occurring in the transition zone than in the peripheral zone.

If a nonpalpable prostate carcinoma grows equally throughout the entire prostate, biopsy samples should be taken equally from the entire prostate in patients with normal DRE results. It was believed previously that the majority of prostate carcinomas occurred in the peripheral zone. McNeal and Bostwick reported that 70% of prostate carcinomas occurred in the peripheral zone and 25% occurred in the transition zone. However, it should be noted that their reported transition zone occupies only 5% of the prostate, although the peripheral zone occupies 70% of the prostate.15 This suggests that the occurrence rate of prostate carcinoma per unit volume is higher in the transition zone than in the peripheral zone. Transition zone volume increases in the hypertrophied prostate, but to our knowledge it has not been demonstrated whether the tumor occurrence rate per unit volume increases, decreases, or remains unchanged as the volume of the transition zone increases. Although the carcinoma occurrence rate per unit volume decreases as the transition zone volume increases, it is not probable that the carcinoma occurrence rate per unit volume in the transition zone becomes significantly lower than that in the peripheral zone. If a nonpalpable tumor is distributed equally throughout the entire prostate, the carcinoma detection rate should improve as the number of biopsy cores obtained increases and biopsy samples are taken equally from the entire prostate. The transperineal template biopsy technique is perfectly suited to taking samples equally from throughout the entire prostate. The carcinoma detection rate in patients with no previous prostate biopsy, normal DRE results, and a PSA level of 4.0–9.99 ng/mL was 43.5% (83 of 191 patients) in the current investigation, which is significantly higher than that based on diagnosis using the transrectal sextant biopsy method. However, we do not know whether that is sufficient and how many tumor foci we miss using the extensive transperineal ultrasound-guided template prostate biopsy method. Our policy is that tumor foci with a dimension of > 10 mm, which is identical to a tumor volume of 0.5 cc if the tumor focus is a sphere, should be detected. To do this, we made the biopsy spots on the transverse image of the prostate form a diamond shape with a diagonal measuring 10-mm in length (Fig. 1). In our preliminary study of a comparison of pathologic findings from step-sectioning of radical prostatectomy specimens with that of biopsy specimens, we estimated detection rates of tumor foci with dimensions of < 5 mm, 5–10 mm, 10–15 mm, 15–20 mm, and ≥ 20 mm to be 25%, 67%, 90%, 92%, and 100%, respectively (data not shown). However, repeat transperineal template biopsies are necessary to assess the sensitivity and specificity of the biopsy method.

All patients in the current study were Japanese. Shiraishi et al. reported that latent carcinoma growing in the transition zone is more prevalent in Japanese individuals than in black and white Americans.16 They reported that 28.9% of latent tumors diagnosed in Japanese patients were distributed within the transition zone, whereas the rate of detection of latent tumors occurring within the transition zone was 8.8% for black Americans and 19.5% for white Americans. It is possible that racial differences influencing the distribution of prostate carcinoma may have influenced our results. Furthermore, an investigation of radical prostatectomy specimens from patients who were diagnosed using extensive transperineal ultrasound-guided template biopsy would be required to demonstrate that nonpalpable tumors are distributed equally within the entire prostate and that palpable tumors are distributed mainly within the peripheral zone.

Ancillary