Limited sampling of radical prostatectomy specimens with excellent preservation of prognostic parameters of prostate cancer.
Abstract: * Context.--The widespread use of the serum prostate-specific antigen test has increased the early detection of prostate cancer and consequently reduced grossly definable prostate cancers.

Objective.--To find the most efficient gross sampling method for radical prostatectomy specimens not only preserving important prognostic factors but also being cost effective.

Design.--We initially analyzed clinicopathologic features of the entire prostate sections from 148 radical prostatectomy specimens, which then were used to examine the impact of 5 partial sampling methods on tumor stage, Gleason score, extraprostatic extension, resection margin status, and paraffin block numbers. The methods included submission of (1) alternative slices, (2) alternative slices plus biopsy-positive posterior quarters, (3) every posterior half, (4) every posterior half plus one midanterior half, and (5) alternative slices plus peripheral 3-mm rim of the remaining prostate.

Results.--Prostate cancers and their extraprostatic extension and resection margin involvement were commonly located in the right posterior portion of the prostate. Method 5 was most efficient, detecting all cases with extraprostatic extension and resection margin involvement and reducing 25% of paraffin blocks compared with the entire sampling of the prostate. The Gleason scores were retained in most of cases, except reversal of the primary and secondary Gleason grade component in only 2 cases (1%). Only 4 cases (3%) were downstaged within the same T2 stage.

Conclusions.--These results demonstrate that sampling of alternative slices plus peripheral rim of the remaining prostate is the most efficient partial sampling method for radical prostatectomy specimens.

(Arch Pathol Lab Med. 2009;133:1278-1284)
Article Type: Report
Subject: Prostate cancer (Diagnosis)
Prostate cancer (Care and treatment)
Diagnostic specimens (Management)
Diagnosis, Laboratory (Methods)
Diagnosis, Laboratory (Equipment and supplies)
Prostatectomy (Methods)
Prostatectomy (Equipment and supplies)
Authors: Kim, Kyungeun
Pak, Pil June
Ro, Jae Y.
Shin, Dongik
Huh, Soo-Jin
Cho, Yong Mee
Pub Date: 08/01/2009
Publication: Name: Archives of Pathology & Laboratory Medicine Publisher: College of American Pathologists Audience: Academic; Professional Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2009 College of American Pathologists ISSN: 1543-2165
Issue: Date: August, 2009 Source Volume: 133 Source Issue: 8
Topic: Event Code: 200 Management dynamics; 440 Facilities & equipment Computer Subject: Company business management
Geographic: Geographic Scope: United States Geographic Code: 1USA United States
Accession Number: 230247118
Full Text: The incidence of prostate cancer is increasing worldwide, and prostate cancer is the most common cancer in men in the United States. (1) Widespread use of serum prostate-specific antigen (PSA) screening test promotes early detection of nonpalpable small prostate cancers. Consequently, it has been difficult for pathologists to identify prostate cancers during gross examination and to sample tissues for microscopic examination of important prognostic information, such as the Gleason score, stage, extraprostatic extension, and resection margin involvement. (2)

Entire sampling of the prostate most accurately provides the important prognostic parameters; however, it demands labor, time, and expense, because 26 to 42 paraffin blocks are required when the entire sampling method is used. (3-5) According to a survey by the American Society of Clinical Pathologists, only 12% of pathologists use the entire sampling method, and most pathologists adopt one of various partial sampling methods. (6) There have been previous studies to propose the most efficient partial sampling method. (4,5,7-9) These partial sampling methods were able to reduce the number of paraffin blocks; however, they were at the expense of losing the important prognostic information, especially T stage, extraprostatic extension, and resection margin involvement status.

To find the most efficient gross sampling method that is not only cost effective but also preserves the pathologic prognostic information, we first analyzed the clinicopathologic features of 148 prostate cancers that were processed by the entire sampling method. We then simulated 5 different partial sampling methods using tumor maps to compare their impact on the stage, Gleason score, extraprostatic extension, resection margin involvement, and paraffin block numbers.

MATERIALS AND METHODS

Patients and Clinical Data

This study was approved by the Asan Medical Center Institutional Review Board. This retrospective study initially included 190 consecutive prostate cancer patients who received radical prostatectomy at Asan Medical Center (Seoul, South Korea) from January 1 to December 31, 2006. Among these patients, 42 were excluded--28 because of insufficient glass slides or tumor maps and 14 because of preoperative hormonal therapy. Clinical information, including age, preoperative serum PSA, and the results of digital rectal examination and radiologic studies, was obtained from the patients' electronic medical records. The tumor stage was assigned according to the 2002 American Joint Committee on Cancer TNM staging system. (10) Briefly, T1 was defined as clinically inapparent tumor involving one or both lobes but not palpable or visible by imaging. T1 included incidentally detected tumors involving 5% or less (T1a) or more than 5% (T1b) of transurethrally resected prostate tissue specimens. T1c included tumors identified by needle biopsy because of elevated PSA levels. T2 included tumors confined within the prostate involving one half of one lobe or less (T2a), more than one half of one lobe (T2b), and both lobes (T2c). The tumors with extraprostatic extension and seminal vesicle invasion belonged to T3a and T3b, respectively.

Pathologic Examination of Radical Prostatectomy Specimens

Each radical prostatectomy specimen was measured, weighed, and fixed in 10% buffered neutral formalin. All prostates were processed using the entire sampling method. In brief, after application of 4 different color inks, red, yellow, blue, and green, on the surface of the right, left, anterior, and posterior prostate, respectively, both seminal vesicles and vasa deferentia were removed. The basal portion of the prostate was shaved using a transverse cut in 144 cases (97.3%). The apical portion was amputated at 5 mm from the apical margin in all cases, thus generating a cone-shaped specimen which was then serially sectioned parasagitally at approximately 2-mm intervals. The remaining prostate was sliced serially at 3- to 5-mm intervals perpendicular to the prostatic urethra. The slices were arranged orderly, and then the outline and direction of the slides, prostatic urethra, and cassette number were recorded on an overlain overhead projector film to generate a tumor map. Each slice was embedded in 1 cassette or divided into 2 to 4 cassettes according to its size. Both shaved basal and amputated apical prostates were submitted separately.

On microscopic examination, the glass slides of all cases were reviewed for pathologic evaluation of T stage, Gleason grade and score, tumor size, extraprostatic extension, resection margin involvement, seminal vesicle invasion, lymphatic invasion, and lymph node and other organ involvement by 2 pathologists. Tumor boundaries and the location of extraprostatic extension and resection margin involvement, if any, were recorded on the tumor maps. The tumor grade was assigned according to the Gleason scoring system. (11) The extent of extraprostatic extension and resection margin involvement were defined as focal when a few glands were involved in no more than 2 separate sections and nonfocal when there was any involvement greater than focal. (12,13)

Analysis of the Spatial Distribution Pattern of Prostate Cancer

All 148 tumor maps showing boundaries of the prostate and tumor were scanned and then converted into digital images to apply to a 3-dimensional reconstruction computer program developed in the Department of Biomedical Engineering, Asan Medical Center. The computer program was designed to demonstrate intraprostatic tumor location and to measure tumor volume, which was obtained by multiplying total tumor area ([cm.sup.2]) by the mean height of slices (0.4 cm).

To analyze the spatial distribution pattern of multiple cases as a whole, the size of the prostate images of each case was standardized and then overlapped to generate composite images. All digital tumor maps were shown in a grayscale according to the frequency of tumors. Dark gray represented high frequency, and brighter gray represented low frequency. Because the basal and apical prostates were not recorded on the tumor maps, the digital images did not include the basal and apical prostates.

Comparison of Partial Sampling Methods

The comparison of partial sampling methods was performed by manual simulation of 5 partial sampling methods using the tumor maps generated during gross examination. The partial sampling methods consisted of imaginary sampling of alternative slices (method 1), alternative slices plus biopsy-positive posterior quarters (method 2), every posterior half (method 3), every posterior half plus one midanterior half (method 4), and alternative slices plus 3 mm of peripheral rim of the remaining prostate (method 5). We compared the tumor extent and stage, extraprostatic extension, resection margin involvement, and the number of paraffin blocks of each partial sampling method to those of the entire sampling method. In addition, the Gleason grade and score status were evaluated only on the partial sampling method 5. In methods 1, 2, and 5, which examined alternative slices, the slice selection was started from the most apical one. In method 5, the outer 3-mm rims from each unselected slice were designed to submit in one cassette.

Statistical Analysis

The statistical analysis was performed using SPSS 12.0K software (SPSS, Chicago, Illinois). It included a paired t test and a one-way ANOVA test to compare regional tumor frequency and tumor volume. The relationship between extraprostatic extension or resection margin involvement and PSA level or Gleason score was evaluated using the [chi square] test. The differences were regarded as statistically significant at P < .05.

RESULTS

Clinicopathologic Features of 148 Prostate Cancers

The resected prostates weighed 38.1 g on average (range, 19-90), and were sectioned into 8 slices on average (range, 4-13). All cases were diagnosed as acinar adenocarcinoma, except for 5 cases, which were accompanied by focal ductal adenocarcinoma component. The clinicopathologic data of the patients are summarized in Table 1. The mean age was 64.4 years (range, 42-78 years). The mean preoperative PSA level was 8.7 ng/mL (range, 0.5-41.6 ng/mL), with 59.6% of the patients belonging to the borderline range between 4 and 10 ng/mL. The clinical stage was assigned in 136 patients, except for 12 patients whose medical records of both digital rectal examination and radiologic studies were not available. Digital rectal examination was performed in 125 patients, and a palpable nodule(s) was detected in 65 patients (52%). Transrectal ultrasonography and/or magnetic resonance imaging were performed in 122 patients, and a suspicious or distinct tumor(s) was noted in 66 patients (54.1%). The tumor was detected by both digital rectal examination and radiologic examination in 39 patients. About two-thirds of the tumors (92 cases; 62.2%) were clinically detected preoperatively, and the remaining 44 cases (29.7%) were detected by the elevated serum PSA without clinically palpable or visible abnormalities (T1c). Twelve patients (8.1%) were TX cases.

The tumors were located mostly in both lobes (124 cases; 83.8%) and at multiple sites (115 cases; 77.7%). Lymph node dissection was performed in 144 cases (97.3%), and lymph node metastasis was noted in 5 cases (3.4%). Twelve cases (8.1%) involved one or both seminal vesicles. However, none of the cases had invasion of other adjacent structure or distant organ metastasis. The Gleason score was 7 or less in most cases (127 cases; 85.8%). The mean tumor volume was 3.43 mL (range, <0.1-23.0 mL), and the median tumor volume was 2.25 mL. The tumor volume was increased with higher preoperative PSA levels and higher Gleason scores (P = .001 and P < .001, respectively). When cases were divided into 3 groups according to the preoperative PSA levels, less than 4 ng/ mL, 4 to 10 ng/mL, and greater than 10 ng/mL, the mean tumor volumes were 2.4 mL, 2.8 mL, and 5.4 mL, respectively. When they were divided into 3 groups according to the Gleason scores 2 to 6, 7, and 8 to 10, the mean tumor volumes were 2.2 mL, 3.2 mL, and 6.8 mL, respectively.

Intraprostatic Distribution Pattern of Prostate Cancers

The composite digital images of all 148 prostate cancers showed a distinct spatial distribution pattern (Figure 1). The tumor distribution was analyzed on transverse and vertical views. In the transverse view, the tumors were located more frequently in the posterior portion than in the anterior portion (P < .001). Interestingly, tumors were noted more frequently in the right lobe than in the left lobe (P = .04). When the prostate was divided into 4 quarters in the transverse view, the tumors were most frequent in the right posterior portion (P = .001). On the vertical view, the prostate cancers were located predominantly at the lower and mid levels than at the upper level (P < .001), as previously described. (14-17) The tumors also were found in the basal and apical sections in 29 cases (19.6%) and 90 cases (60.8%), respectively.

The prostate cancers also showed distinct spatial distribution pattern according to the preoperative PSA level and the Gleason score, although there was a tendency to be distributed more diffusely when the PSA values were high (Figure 2). When the preoperative PSA levels were below 4 ng/mL, the tumors were located frequently in the posterior portion of both lobes. When preoperative PSA levels were in the borderline range, as in 88 cases (59.5%), the prostate cancers were located more frequently in the right posterior portion (P = .003) and at the mid and lower levels (P < .001). When Gleason scores were 7 or below, as in 127 cases (85.8%), the prostate cancer was also located more frequently in the right posterior portion (P < .001) and at the lower level (P < .001). On the other hand, when the Gleason scores were higher than 7, the tumor was distributed more diffusely, with a tendency to be located more often in the left lobe than in the right lobe, however, this was not statistically significant (P = .11).

Distribution Pattern of Extraprostatic Extension and Resection Margin Involvement

Among the 148 prostate cancer specimens, extraprostatic extension was noted in 50 cases (34%), resection margin involvement in 48 cases (32%), and both in 22 cases (15%). Most of the cases demonstrated nonfocal involvement of extraprostatic extension (46 cases; 92.0%) and resection margin (41 cases; 89.1%). The mean lengths of extraprostatic extension and resection margin involvement were 12.3 mm (range, 2-32 mm) and 8.3 mm (range, 2-20 mm), respectively. Extraprostatic extension and resection margin involvement were commonly identified as a single focus (72% and 90% of positive cases, respectively). The location of extraprostatic extension and resection margin involvement was similar to that of the overall tumor distribution pattern, with the most frequent site being the right posterior quarter. On the vertical view, extraprostatic extension was most frequently noted at the mid level, whereas resection margin involvement was at the lower level. Details are tabulated in Table 2.

Extraprostatic extension and resection margin involvement were noted more frequently in patients with high preoperative PSA levels (P = .04 and .01, respectively) and Gleason scores (P < .001 and .001, respectively). When the prostate cancers were divided into 3 groups according to the preoperative PSA levels less than 4 ng/mL, 4 to 10 ng/mL, and greater than 10 ng/mL, extraprostatic extension was noted in 7 (35%), 22 (25%), and 21 (52.5%) cases, and resection margin involvement in 5 (25%), 21 (23.9%), and 20 (50%) cases, respectively. When the prostate cancers were divided into 3 groups according to the Gleason scores 2 to 6, 7, and 8 to 10, extraprostatic extension was noted in 5 (11.4%), 29 (34.9%), and 16 (76.2%) cases, and resection margin involvement in 8 (18.2%), 25 (30.1%), and 13 (61.9%) cases, respectively.

Comparison of Partial Sampling Methods to the Entire Sampling Method

We simulated the impact of partial sampling methods on paraffin block numbers and important pathologic prognostic parameters, that is, T stage, extraprostatic extension, and resection margin involvement in 136 cases for which stage information was available (Table 2). Because the tumor was located more frequently at the lower level than the upper level, the slice selection was started from the most apical slice when a partial sampling method examined alternative slices. We included partial sampling method 5 because both extraprostatic extension and resection margin involvement were commonly present at the periphery of the prostate.

[FIGURE 1 OMITTED]

Partial sampling method 1, which examined alternative slices, was the most economic and reduced the total paraffin block number to 16.4 from 26.6 blocks (61.7%) of entire sampling, but at the greatest expense for missing extraprostatic extension (8 cases; 16%) and resection margin involvement (8 cases; 17%). Method 2, which added needle biopsy-positive posterior quarters to the alterative slices, slightly improved the loss of extraprostatic extension (6 cases; 12%) and resection margin involvement (6 cases; 13%). Partial sampling methods 3 and 4 examining every posterior half only or with one midanterior half, respectively, were similar to or slightly better than partial sampling method 2, with only slight improvement in detecting extraprostatic extension and resection margin involvement. However, the paraffin block numbers were the highest in the partial sampling method 4. As expected, the partial sampling method 5, which examined alternative slices plus the peripheral 3-mm rim of the remaining prostate, was identical to the entire sampling method regarding detection of the extraprostatic extension and resection margin involvement (Table 3). Among the 136 cases, only 4 T2c cases (3%) were downstaged. One case was downstaged to T2b, and 3 cases to T2a. However, they were still within the same T2 stage (Figure 3). The partial sampling method 5 reduced the paraffin block numbers to 20.0 (75.2% of the entire sampling). Because the above prognostic factors were best preserved by the partial sampling method 5, we also examined its impact on the Gleason scoring. The Gleason scores of all cases were well retained, although primary and secondary Gleason grades were reversed in 2 cases: from 3 + 4to 4 + 3 in one case, and from 4 + 3to 3 + 4 in the other case.

[FIGURE 2 OMITTED]

COMMENT

Partial sampling methods of radical prostatectomy specimens have been a controversial issue. Entire sampling of the prostate has been emphasized in order to detect all resection margin-positive cases. Patients with negative resection margin on the entire sampling method tended to have a lower risk of PSA recurrence and postoperative adjuvant treatment than those diagnosed by partial sampling methods. (18) The PSA recurrence also depends on extraprostatic extension. For example, a previous study showed that the 5-year PSA recurrence-free rate was highest (96.7%) in prostate-confined margin-negative tumors and followed by focal extraprostatic extension and margin-negative disease (89.7%). Tumors with both resection margin involvement and extensive extraprostatic extension demonstrated the lowest (58.5%) 5-year freedom of PSA recurrence. (19) These studies emphasize a careful pathologic examination of radical prostatectomy specimens to detect extraprostatic extension and resection margin involvement for the prediction of outcome and proper postoperative patient management.

Several studies have been conducted to determine the most efficient partial sampling method to not only preserve important prognostic parameters but also reduce the number of paraffin blocks. (3,5,7,18) Sampling of alternative slices of clinically palpable tumors detected 75% of capsular penetration and 87.5% of resection margin involvement. (7) The detection rate was improved by the sampling of gross carcinoma plus the apical section, which detected 91% of extraprostatic extension and 96% of resection margin involvement, with a reduction of the mean paraffin block number to 13 (31%) compared with 42 for the entire sampling. (3) These partial sampling methods are more cost effective than the entire sampling method, but extracapsular extension and resection margin involvement could be missed in a few cases. In our study, partial sampling method 5, examining alternative slices plus peripheral 3-mm rim of the remaining prostate slices, was the most comparable of the 5 partial sampling methods to the entire sampling method. Method 5 detected all extraprostatic extension and resection margin involvement. However, it is still possible to overlook small foci of extraprostatic extension and resection margin involvement because only one 4-[micro]m tissue section was prepared from the 3- to 5-mm prostatic tissue embedded in the paraffin block. However, this issue is not specific for just partial sampling methods, but also applied to the entire sampling method.

[FIGURE 4 OMITTED]

Because nonpalpable prostate cancers have been increasing recently, accounting for more than 50% of radical prostatectomy cases in the United States, Sehdev et al (5) compared 9 partial sampling methods to the entire sampling method, using nonpalpable, needle biopsy-proven (stage T1c) prostate cancers. They showed sampling of every posterior section plus one midanterior section, and if sizable tumor was seen, the remaining ipsilateral anterior sections were submitted. This method detected 96% of extraprostatic extension and 100% of resection margin involvement, with reduction of the mean block numbers from 34 to 27 (79%). (5) Our partial sampling method 4 examined every posterior half and one midanterior section, which detected 90% of extraprostatic extension and 91% of resection margin involvement with the reduction of mean block numbers to 21.2 (79.7%). Because our initial study did not consider additional submission even if any sizable tumor was present, we added remaining ipsilateral anterior section if any sizable tumor was present to match the previous study. (5) With this partial sampling method, we could detect all extraprostatic extension and resection margin involvement cases like the previous report; however, the partial sampling method required more blocks (26.9 blocks) than that of the entire sampling method (26.6 blocks). The discrepancy of the block number was derived partly from the difference in study design. The previous study included only nonpalpable, needle biopsy-proven prostate cancers (stage T1c), whereas our study included both nonpalpable and palpable tumors, which increased reexamination rate (96 cases; 64.9%) for the remaining ipsilateral anterior sections. (5)

It has been demonstrated that 68% to 80% of clinically diagnosed cancers are located in the peripheral zone. (15,17) In fact, sampling of the peripheral 3-mm rim of the remaining prostate slices in addition to the alternative slices in method 5 correctly staged 6 cases among the 10 downstaged cases by alternative slice sampling only (method 1). Although the other 4 cases were downstaged using method 5, from T2c to T2a in 3 cases and to T2b in 1 case, the downstage in these cases was clinically insignificant because the tumor stages remained T2 (Figure 3).

It is well known that the majority of prostate cancers are localized in the peripheral zone and in the apex; however, the side predominance has not been well established. (20) One previous study analyzed 81 cases and showed a high frequency of prostate cancer in the left lobe of the prostate. (16) In contrast, our study analyzed 148 cases and showed right lobe predominance, especially in cases with borderline increase of preoperative PSA level (4-10 ng/mL). Because the recognition of the intraprostatic distribution pattern of prostate cancer could improve tumor detection rate of the preoperative needle biopsy procedure, further studies should be undertaken to assess the right or left lobe predominance.

There have been previous studies examining the impact of partial sampling methods on Gleason scoring. (5,7) Sampling of alternative slices of clinically palpable prostate tumors resulted in Gleason score change in 13 of 52 total cases. Seven cases had only one Gleason score difference and were considered intraobserver variations. (7) Therefore, Gleason scoring in 48 cases (90%) was maintained by the partial sampling method. (7) Another study examined 104 cases of clinically palpable and nonpalpable tumors. They reported that various partial sampling methods, including whole-mount alternative sampling, showed a Gleason score within plus or minus one of that assigned by entire histologic examination in 98% of cases. (3) These previous studies demonstrated that the Gleason score was relatively well preserved even when partial sampling methods were employed. Our study including both palpable and nonpalpable tumors showed better preservation of Gleason score than the previous reports. The Gleason scores of all cases were preserved, although 2 cases showed a reversal of the most predominant Gleason pattern and the second Gleason pattern. This might be attributed to the sampling of the peripheral prostate, which was the common site of prostate cancers.

The method examining peripheral prostatic capsule was reported previously to harvest fresh prostate cancer tissue (4,5,7-9); however, to the best of our knowledge, our study is the first to apply the method to the routine sampling of formalin-fixed radical prostatectomy specimens. As we simulated the partial sampling methods using tumor maps generated during the entire sampling method, we tested the practical usefulness of the partial sampling method 5. The prostate was processed as alternative slices, the slice selection was started from the most apical slice, and the peripheral 3-mm rims of unselected slices were then shaved using a standard scalpel blade (Figure 4). The 4 peripheral rims from quadrant sections of each slice could then be submitted in one cassette. This procedure can be performed easily.

Although the partial sampling method 5 detected tumors of all cases in our study, it is possible that a tumor may not be detected if there is a small unifocal tumor in the central prostate of unselected slices. Therefore, the remaining central prostate of the unselected slices should be saved in a well-oriented status until the final diagnosis is rendered. Because this is a retrospective study performed at a single institution, a prospective multicenter study with more cases might help to verify our results.

In conclusion, our study demonstrated that the sampling of alternative slices plus peripheral rim of the remaining prostate is the most efficient partial sampling method for radical prostatectomy specimens, not only reducing 25% of paraffin blocks but also preserving well important pathologic prognostic factors, such as tumor stage, Gleason score, extraprostatic extension, and resection margin involvement.

References

(1.) Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer J Clin. 2005;55:10-30.

(2.) Bostwick DG, Grignon DJ, Hammond ME, et al. Prognostic factors in prostate cancer. College of American Pathologists Consensus Statement 1999. Arch Pathol Lab Med. 2000;124:995-1000.

(3.) Hall GS, Kramer CE, Epstein JI. Evaluation of radical prostatectomy specimens. A comparative analysis of sampling methods. Am J Surg Pathol. 1992;16: 315-324.

(4.) Mazzucchelli R, Santinelli A, Lopez-Beltran A, Scarpelli M, Montironi R. Evaluation of prognostic factors in radical prostatectomy specimens with cancer. Urol Int. 2002;68:209-215.

(5.) Sehdev AE, Pan CC, Epstein JI. Comparative analysis of sampling methods for grossing radical prostatectomy specimens performed for nonpalpable (stage T1c) prostatic adenocarcinoma. Hum Pathol. 2001;32:494-499.

(6.) True LD. Surgical pathology examination of the prostate gland. Practice survey by American society of clinical pathologists. Am J Clin Pathol. 1994;102: 572-579.

(7.) Cohen MB, Soloway MS, Murphy WM. Sampling of radical prostatectomy specimens. How much is adequate? Am J Clin Pathol. 1994;101:250-252.

(8.) Hollenbeck BK, Bassily N, Wei JT, et al. Whole mounted radical prostatectomy specimens do not increase detection of adverse pathological features. J Urol. 2000;164:1 583-1586.

(9.) Bova GS, Fox WM, Epstein JI. Methods of radical prostatectomy specimen processing: a novel technique for harvesting fresh prostate cancer tissue and review of processing techniques. Mod Pathol. 1993;6:201-207.

(10.) Greene FL, Page DL, Fleming ID, et al. American Joint Committee on Cancer (AJCC) Cancer Staging Manual. 6th ed. New York, NY: Springer; 2002.

(11.) Epstein JI, Allsbrook WC Jr, Amin MB, Egevad LL. The 2005 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma. Am J Surg Pathol. 2005;29:1228-1242.

(12.) Epstein JI, Amin M, Boccon-Gibod L,etal. Prognostic factors and reporting of prostate carcinoma in radical prostatectomy and pelvic lymphadenectomy specimens. Scand J Urol Nephrol Suppl. 2005;(216):34-63.

(13.) Wheeler TM, Dillioglugil O, Kattan MW, et al. Clinical and pathological significance of the level and extent of capsular invasion in clinical stage T1-2 prostate cancer. Hum Pathol. 1998;29:856-862.

(14.) Chen ME, Johnston DA, Tang K, Babaian RJ, Troncoso P. Detailed mapping of prostate carcinoma foci: biopsy strategy implications. Cancer. 2000;89:1800 1809.

(15.) Epstein JI, Walsh PC, Carmichael M, Brendler CB. Pathologic and clinical findings to predict tumor extent of nonpalpable (stage T1c) prostate cancer. JAMA. 1994;271:368-374.

(16.) Frimmel H, Egevad L, Bengtsson E, Busch C. Modeling prostate cancer distributions. Urology. 1 999;54:102 8-1034.

(17.) McNeal JE, Redwine EA, Freiha FS, Stamey TA. Zonal distribution of prostatic adenocarcinoma. Correlation with histologic pattern and direction of spread. Am J Surg Pathol. 1988;12:897-906.

(18.) Grossfeld GD, Chang JJ, Broering JM, et al. Does the completeness of prostate sampling predict outcome for patients undergoing radical prostatectomy?: data from the CAPSURE database. Urology. 2000;56:430-435.

(19.) Chuang AY, Chang SJ, Horng CF, Tsou MH. Study of prostate cancer pathologic features in Chinese populations. Urology. 2007;69:915-920.

(20.) McNeal JE. Origin and development of carcinoma in the prostate. Cancer 1969;23:24-34.

Kyungeun Kim, MD; Pil June Pak, MS; Jae Y. Ro, MD, PhD; Dongik Shin, PhD; Soo-Jin Huh, PhD; Yong Mee Cho, MD, PhD

Accepted for publication October 16, 2008.

From the Department of Pathology (Drs Kim and Cho) and the Biomedical Engineering Department (Mr Pak and Drs Shin and Huh), Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea; and Department of Pathology, Cornell University Methodist Hospital, Houston, Texas (Dr Ro).

The authors have no relevant financial interest in the products or companies described in this article.

The results presented in this paper have not been previously published either in whole or in part, except in abstract form at the 59th Annual Fall Meeting of the Korean Society of Pathologists, Seoul, South Korea, 2007.

Reprints: Yong Mee Cho, MD, PhD, Department of Pathology, Asan Medical Center, University of Ulsan College of Medicine, 388-1 Pungnap-dong, Songpa-gu, Seoul, Seoul 138-736, Republic of Korea (e-mail: yongcho@amc.seoul.kr).
Table 1. Clinicopathologic Features of 148
Prostate Cancers

Variables                          No.   Cases %

Age, y
  <60                              34      23.0
  60-69                            80      54.0
  [greater than or equal to] 70    34      23.0
Serum PSA, ng/mL
  <4                               20      13.5
  4-10                             88      59.5
  10-20                            32      21.6
  [greater than or equal to] 20     8       5.4
No. of tumor
  Unifocal                         33      22.3
  Multifocal                       115     77.7
Extent
  <1/2 of one lobe                 23      15.5
  >1/2 of one lobe                  1       0.7
  Both lobes                       124     83.8
T stage
  1                                44      29.7
  2                                53      35.8
  3                                39      26.4
  Unknown                          12       8.1
N stage
  NX                                4       2.7
  N0                               139     93.9
  N1                                5       3.4
Gleason score
  2-6                              44      29.7
  7                                83      56.1
  8-10                             21      14.2
Extraprostatic extension
  Absent                           98      66.2
  Focal                             4       2.7
  Nonfocal                         46      31.1
Resection margin involvement
  Absent                           102     68.9
  Focal                             5       3.4
  Nonfocal                         41      27.7
Seminal vesicle involvement
  Absent                           136     91.9
  Present                          12       8.1

Abbreviation: PSA, prostate-specific antigen.

Table 2. Distribution Pattern of Extraprostatic
Extension and Resection Margin Involvement

                                    Prognostic Parameters

                                                     Resection
                                 Extraprostatic       Margin
                                   Extension,      Involvement,
                                    No. (%)           No. (%)
    Intraprostatic Region           (n = 50)      (n = 31/46) (a)
Transverse view
  Right anterior                   2 (4.0)          6 (19.4)
  Right posterior                 17 (34.0)        10 (32.3)
  Left anterior                    3 (6.0)          2 (6.4)
  Left posterior                  14 (28.0)        10 (32.3)
  Right anterior and posterior     5 (10.0)         1 (3.2)
  Left anterior and posterior      1 (2.0)          0 (0.0)
  Right and left anterior          3 (6.0)          0 (0.0)
  Right and left posterior         3 (6.0)          0 (0.0)
  Others                           2 (4.0) (b)      2 (6.4) (c)
  Total                           50 (100)         31 (100)
Vertical view (d)
  Upper level                      5 (10.0)         9 (19.5)
  Midlevel                        15 (30.0)         4 (8.7)
  Lower level                      4 (8.0)         23 (50)
  Upper and midlevel               7 (14.0)         3 (6.5)
  Lower and midlevel               9 (18.0)         5 (10.9)
  Upper and lower level            0 (0.0)          1 (2.2)
  Upper, mid, and lower
    level                         10 (20.0)         1 (2.2)
  Total                           50 (100)         46 (100)

(a) The location of resection margin involvement of 15 cases
could not be evaluated in the transverse view because it was
present in the apex and base resection margins, which were not
recorded on the tumor maps.

(b) Right posterior and left anterior and posterior, 1 case;
right anterior and posterior and left posterior, 1 case.

(c) Right posterior and left anterior, 1 case; right anterior
and posterior and left posterior, 1 case.

(d) The apex and base involvements were included in the upper
level and lower level, respectively.

Table 3. Comparison of Partial Sampling Methods to Complete
Sampling Method

                                                 Parameters

                            Stage        Extraprostatic Extension,
Methods                   Agreement         None/Focal/Nonfocal

Complete sampling, No.       136                  98/4/46
Partial
samplings, No. (%)

1. Alternative            126 (92.6)  106 (108.2)/2 (50.0)/40 (87.0)
   slices

2. Alternative slices
   plus biopsy- positive  129 (94.9)  104 (106.1)/2 (50.0)/42 (91.3)
   posterior quarters

3. Every posterior        125 (91.9)  106 (108.2)/3 (75.0)/39 (74.8)
   half

4. Every posterior
   half plus one          131 (96.3)  103 (105.1)/3 (75.0)/42 (91.3)
   midanterior half

5. Alternative
   slices plus periph-    132 (97.1)  98 (100.0)/4 (100.0)/46 (100.0)
   eral rim of
   remaining prostate

                                     Parameters

                           Resection Margin Involvement,    Mean Block
Methods                         None/Focal/Extensive            No.

Complete sampling, No.                102/5/41                 26.6
Partial
samplings, No. (%)

1. Alternative            110 (107.8)/4 (80.0)/34 (82.9)    16.4 (61.7)
   slices

2. Alternative slices
   plus biopsy- positive  108 (105.9)/5 (100.0)/35 (85.4)   18.1 (68.0)
   posterior quarters

3. Every posterior        107 (104.9)/5 (100.0)/36 (87.8)   19.8 (74.4)
   half

4. Every posterior
   half plus one          106 (103.9)/5 (100.0)/37 (90.2)   21.2 (79.7)
   midanterior half

5. Alternative
   slices plus periph-    102 (100.0)/5 (100.0)/41 (100.0)  20.0 (75.2)
   eral rim of
   remaining prostate

Figure 3. The impact of partial sampling method 5 on the tumor stage.
Although 4 T2c cases are downstaged (T2b in 1 case and to T2a in 3
cases) compared with the entire sampling method, they still remain in
the same stage T2.

   Method 5     TO     T1c   T2a   T2b   T2c   T3a   T3b
Complete       n = 0   44    10     1    42    29    10

TO    n = 0      0
T1c    44              44
T2a     7                     7
T2b     0                           0
T2c    46                     3     1    42
T3a    29                                      29
T3b    10                                            10
Gale Copyright: Copyright 2009 Gale, Cengage Learning. All rights reserved.