Abstract
Objective
Transrectal ultrasound–guided core biopsies of the prostate gland and prostatectomies have become common procedures at many community hospitals in Canada, especially in the era of serum prostate-specific antigen (PSA) screening for prostate cancer. The Gleason grading of prostate cancer in biopsies and prostatectomies is a major determinant used for treatment planning. There is evidence in the literature that suggests important discordance between community hospital pathologists and urological pathologists with respect to the Gleason grading of prostate cancer. Our objective was to determine the diagnostic rates and Gleason scoring patterns for prostate gland biopsies and prostatectomies at our institution compared with the literature.
Methods
We conducted a retrospective review of all prostate gland biopsies and prostatectomies performed at the Grey Bruce Health Services from January 2005 to September 2005. We collected data from 194 biopsies and 44 prostatectomies. We obtained prebiopsy serum PSA levels and digital rectal exam results for all patients from urologists’ office records.
Results
The average age for men having biopsies was 65.8 (standard deviation [SD] 8.6) years, and the average prebiopsy serum PSA level was 8.7 (median 7.1, SD 6.2) μg/L. The rates of diagnosis from prostate gland biopsies of benign (17.6%), high-grade prostatic intraepithelial neoplasia (11.0%), atypical small acinar proliferation suspicious for invasive malignancy (13.2%) and invasive prostatic adenocarcinoma (58.2%) at our institution were significantly different than those reported in the literature (p < 0.001). We observed a significant variation in the rates of these diagnoses among the community hospital pathologists in our study (p = 0.004). There was a strong correlation between the increasing number of positive core biopsy sites and increasing Gleason scores in biopsies (p < 0.001). There was also a strong correlation between increasing pre-biopsy serum PSA levels and increasing Gleason scores in biopsies (p < 0.001). A substantial proportion (21.9%) of the biopsies given the Gleason score of 6 had a Gleason score of 7 in the prostatectomy specimen.
Conclusion
Our results showed a significant difference in prostate gland biopsy categorical diagnoses compared with the literature. There were also significant differences in categorical diagnoses of prostate gland biopsies among the community hospital pathologists in our study. The data identify a strong positive correlation between the increasing number of positive core biopsy sites and increasing Gleason scores in biopsies, as well as a strong positive correlation between increasing prebiopsy serum PSA levels and increasing Gleason scores in biopsies that revealed cancer. We would encourage other community hospital pathologists, in collaboration with their urologists, to review periodically their prostate gland pathology practices in an attempt to improve the uniformity of diagnoses.
Introduction
Transrectal ultrasound–guided core biopsies of the prostate gland and prostatectomies have become common procedures at many community hospitals in Canada, especially in the era of serum prostate-specific antigen (PSA) screening for prostate cancer. The Gleason score1 is a required element in pathology reports that assists urologists and oncologists to decide what treatments they should recommend to patients with invasive prostatic adenocarcinoma. A poor correlation between general pathologists and urological pathologists with respect to Gleason grading has been reported.2–8 Community hospital pathologists have been reported to “undergrade” prostate cancer in biopsy material. It has been suggested that all cancers that community hospital pathologists diagnose based on prostate gland biopsies should be reviewed by a urological pathologist before definitive treatment.9
Our objective was to compare the reporting of prostate gland biopsies and prostatectomies in our community hospital with data in the literature and to assess the homogeneity of prostate gland pathology reporting among the pathologists at our institution.
Methods
The Grey Bruce Health Services is a 240-bed community hospital in rural Ontario with a catchment population of 157 000. There are 3 urologists and 3 pathologists at our institution. All 6 are active participants in the prostate pathology service.
We searched the pathology database at the Grey Bruce Health Services retrospectively from Jan.1, 2005, to Sept. 31, 2005. We retrieved pathology reports for all prostate gland biopsies and prostatectomies for this 9-month period. We also retrieved prostate gland biopsy reports for patients who had a prostatectomy with biopsy before Jan. 1, 2005. The Grey Bruce Health Services Ethics Committee approved our study.
The urologists at our institution perform trans-rectal ultrasound–guided core biopsies, obtaining 2 cores from each site (left base, left mid, left apex, right base, right mid and right apex). They submit the biopsies from each site in separate specimen containers in 10% formalin. More than 95% of the biopsies submitted contain 12 cores of tissue. We report prostate gland biopsies and prostatectomies using a synoptic-like report adapted from templates developed by the College of American Pathologists. Our reports meet the criteria established by Cancer Care Ontario for the reporting of prostate gland biopsies and prostatectomies.
We collected the following data from each biopsy report: age, the most severe diagnostic abnormality in each set of biopsies (benign, high-grade prostatic intraepithelial neoplasia (HGPIN), atypical small acinar proliferation (ASAP) suspicious for cancer, invasive prostatic adenocarcinoma, urologist (R.D., P.M., T.W.), and pathologist (K.N., B.R., B.S.). For patients who received a diagnosis of invasive adenocarcinoma at biopsy and for all prostatectomies, we collected the following additional data: Gleason score; number of biopsy sites positive for cancer; and the presence or absence of perineural invasion, angiolymphatic invasion and extraprostatic extension. We obtained the prebiopsy serum PSA level and the findings of the digital rectal exam immediately before biopsy from the urologists’ office records. Data were available for all patients who had a biopsy and a prostatectomy.
We compiled all of the data in an Excel (Microsoft Corp.) spreadsheet and performed statistical analyses using SAS9 software (SAS Inc.).
Results
We performed 194 sets of prostate gland biopsies on 182 patients during the study period. Ten patients had 2 sets of biopsies and 1 patient had 3 sets. For the patients who had repeat biopsies, 8 had received a previous diagnosis of HGPIN and 3 had a previous diagnosis of ASAP. We included in our analysis only the first set of prostate gland biopsies for each patient. The mean age of the patients who had biopsies was 65.8 (standard deviation [SD] 8.6, range 41–89) years and the mean prebiopsy serum PSA level was 8.7 (SD 6.2, median 7.1, range 0.3–47.5) μg/L, excluding 1 outlier value of 107 μg/L. Table 1 shows that for each pathologist there was no significant difference in patient age (F test, p = 0.39) or prebiopsy serum PSA (Kruskal–Wallis [Mann–Whitney-Wilcoxon] test, p = 0.24) among the pathologists. The majority of patients (108, 59.3%) had prebiopsy serum PSA levels between 4.0 and 10.0 μg/L. About one-quarter of patients (52, 28.6%), had prebiopsy serum PSA levels greater than 10.0 μg/L. A minority of patients (22, 12.1%) had serum PSA levels lower than 4.0 μg/L; results of the digital rectal examinations of all of these patients were abnormal.
Table 1.
Characteristics of patients who had biopsies and prostatectomies, by participating study pathologist
| Pathologist | No. of patients | Mean (SD) age, yr | Mean (SD) prostate-specific antigen level, μg/L |
|---|---|---|---|
| A | 65 | 66.8 (8.9) | 9.0 (6.6) |
| B | 65 | 66.1 (8.1) | 9.2 (6.7) |
| C | 52 | 64.4 (8.6) | 7.6 (4.9) |
| All biopsies | 182 | 65.8 (8.6) | 8.7 (6.2) |
| All prostatectomies | 44 | 62.7 (5.8) | 7.9 (4.1) |
SD = standard deviation.
Table 2 shows the distribution of diagnoses for each pathologist and the total distribution for our institution compared with rates published in the literature. The rate of diagnosis of invasive prostatic adenocarcinoma at our institution (58.2%) was 1.8 times greater than that reported in the literature (33%). Our rate of precancer diagnoses (HGPIN and ASAP, 24.2%), was 2.0 times greater than that reported in the literature. The rate of benign diagnoses (17.6%) was 3.1 times lower than that reported in the literature. The rates of all diagnoses at our institution were significantly different than those reported in the literature (χ2 test, p < 0.001).
Table 2.
Categorical diagnoses (n = 182) and distribution at biopsy, by participating study pathologist
| Diagnostic category; no. (%) of patients
|
||||
|---|---|---|---|---|
| Pathologist | Benign | HGPIN | ASAP | Cancer |
| A | 19 | 4 | 7 | 35 |
| B | 5 | 11 | 14 | 35 |
| C | 8 | 5 | 3 | 36 |
| Totals | 32 (17.6) | 20 (11.0) | 24 (13.2) | 106 (58.2) |
| Literature | (55.0) | (7.0) | (5.0) | (33.0) |
ASAP = atypical small acinar proliferation; HGPIN = high-grade prostatic intraepithelial neoplasia.
We observed significant variation among pathologists at our institution with respect to the rendering of categorical diagnoses in prostate gland biopsies (χ2 test, p = 0.004) (Table 2). The majority of cancers (84.9%) were given a Gleason score of 7 (49.1%) or 6 (35.8%). There were no cancers given a Gleason score of less than 6 at biopsy. There was significant variation among the pathologists with respect to the Gleason grading of cancers diagnosed at biopsy (χ2 test, p = 0.001) (Table 3). There was a strong correlation in the biopsies between the increasing number of positive core biopsy sites and increasing Gleason scores (Spearman correlation test, p < 0.001) (Table 4). There was also a strong correlation in the cancerous biopsies between increasing prebiopsy serum PSA levels and increasing Gleason scores (Spearman correlation test, p < 0.001) (Table 5).
Table 3.
Gleason score distribution for cancers diagnosed at biopsy (n = 106), by participating study pathologist
| Gleason score; no. (%) of patients
|
|||
|---|---|---|---|
| Pathologist | 6 | 7 | 8/9 |
| A | 13 | 17 | 5 |
| B | 4 | 22 | 9 |
| C | 21 | 13 | 2 |
| Totals | 38 (35.8) | 52 (49.1) | 16 (15.1) |
| Literature | (49.0) | (41.0) | (10.0) |
Table 4.
Number of positive core biopsy sites per case versus Gleason score for cancers diagnosed at biopsy (n = 106)
| No. of positive core biopsy sites per patient
|
||||||
|---|---|---|---|---|---|---|
| Gleason score | 1 | 2 | 3 | 4 | 5 | 6 |
| 6 | 17 | 11 | 5 | 3 | 2 | 0 |
| 7 | 6 | 13 | 15 | 9 | 2 | 7 |
| 8/9 | 2 | 2 | 7 | 4 | 0 | 1 |
| Totals | 25 | 26 | 27 | 16 | 4 | 8 |
Table 5.
Correlation between diagnostic category and prebiopsy prostate specific antigen level
| Diagnosis | No. of patients | Mean (SD) prostate-specific antigen level, μg/L |
|---|---|---|
| Benign | 32 | 8.9 (7.2) |
| HGPIN | 20 | 6.4 (3.6) |
| ASAP | 24 | 7.1 (4.2) |
| All cancers | 106 | 9.4 (6.6) |
| Gleason score | ||
| 6 | 38 | 6.8 (3.2) |
| 7 | 52 | 9.9 (7.4) |
| 8/9 | 16 | 13.9 (7.3) |
ASAP = atypical small acinar proliferation; HGPIN = high-grade prostatic intraepithelial neoplasia; SD = standard deviation.
We performed immunohistochemistry for the presence of basal cells (cytokeratin 34βE12) on 74/182 biopsies (40.7%). We sent a minority of biopsies (16, 8.8%) for external consultation to the group of urological pathologists at the London Health Science Centre in London, Ont. We forwarded the majority (14 /16) with a diagnosis of “atypical small acinar proliferation suspicious for invasive malignancy” to determine whether there were sufficient features for a definitive diagnosis of invasive prostatic adenocarcinoma. We forwarded the remaining 2 for confirmation of Gleason grading. The consultant urological pathologists diagnosed “minimal” cancer in 8/14 (57%) of the cases in which we suspected cancer.
The comparison of prostate gland biopsies and prostatectomies showed that there was a marked increase in the reporting of perineural invasion in the prostatectomy specimens (Table 6). We never reported angiolymphatic invasion and extraprostatic extension in the biopsy specimens, yet 7/44 (15.9%) tumours showed extraprostatic extension at prostatectomy. We reported seminal vesicle invasion in 3/44 (6.8%) prostatectomies. There were 36 pT2 tumours and 8 pT3 tumours in our study. A large proportion of cancers diagnosed at biopsy and given a Gleason score of 6 (38.9%) received a final Gleason score of 7 at prostatectomy. We sampled lymph nodes in 38/44 (86.3%) prostatectomies. All of the lymph nodes sampled were free of cancer. Surgical margins were positive for cancer in 15/44 (34.1%) prostatectomy specimens.
Table 6.
Characteristics of cancers diagnosed at biopsy v. at prostatectomy (n = 44)
| Gleason score
|
Pathologic feature
|
|||||
|---|---|---|---|---|---|---|
| Procedure | 6 | 7 | 8/9 | PN | AL | EPE |
| Biopsies | 18 | 24 | 2 | 13 | 0 | 0 |
| Prostatectomies | 11 | 32 | 1 | 38 | 1 | 7 |
PN = perineural invasion; AL = angiolymphatic invasion; EPE = extraprostatic extension.
Discussion
The rate of diagnosis of invasive prostatic adenocarcinoma at our institution (58.2%) was substantially higher than the average rate (33%) reported in the literature.10,11 We compared the rates of categorical diagnoses in our study with those reported in 2 large-scale studies involving 78 290 patients.10,11 Although our population may not have been directly comparable to these large-scale predominantly American studies, in the absence of similar large-scale Canadian data, we used the rates of categorical diagnoses in these studies for comparison. The demographics and pre-biopsy serum PSA levels of our patient population compare favourably with those reported in the literature.11 Our results raise the possibility that there are important differences between the approaches to prostate gland biopsies performed by American and Canadian urologists. We speculate that Canadian urologists, who work in a single-payer environment with limited resources, may have more stringent criteria for deciding to what patients they will offer a prostate gland biopsy. Alternatively, the widespread use of serum PSA screening for prostate cancer may have been delayed in our geographic region. This would suggest that the increased rate of prostate cancer that we observed in the biopsies may be secondary to case-finding in the initial wave of more widespread serum PSA screening. Regardless of the definitive explanation for our findings, we question whether Canadian urologists should be performing greater numbers of prostate gland biopsies.
Our results showed an increased rate of pre-cancer diagnoses (HGPIN and ASAP, 24.2%) compared with the rate reported in the literature (13%).10–13 The literature indicates that these pre-cancer diagnoses show poor interobserver reproducibility among community hospital and urological pathologists.12–15 Several recent reviews outlined the diagnostic criteria and the clinical significance of HGPIN and ASAP in prostate gland biopsies.12–15 It is possible that the more widespread publication of diagnostic criteria for these pre-cancer diagnoses will lead to greater degrees of diagnostic homogeneity among community hospital pathologists. Regardless, the increased rates of precancer diagnoses at our institution are an important and possibly worrisome trend, in that HGPIN and ASAP have been indications for early repeat biopsies by the urologists at our institution. We would again speculate that the increased rates of precancer diagnoses may be related to a selection bias as described above.
The data show that there are significant differences among the pathologists at our institution in the Gleason grading of prostate cancers at biopsy. Significant variation in the grading of prostate cancer has been reported previously among general and urological pathologists.2–7,16 Without review of all of these prostate gland biopsies by an expert urological pathologist, it would not be possible to determine if an individual pathologist was “under-” or “overgrading.” Previous studies have indicated that community pathologists have a tendency to undergrade prostate cancer biopsies compared with urological patholo-gists.2,4–6,8 There was no evidence of this phenomenon at our institution. Recent studies have shown that a teaching program by an expert urological pathologist and/or the completion of a Web-based tutorial can substantially improve the grading of prostate cancer by community hospital pathologists.6,8,17 The results of our study suggest that such a program may be beneficial in our department to improve the homogeneity of Gleason grading in biopsy specimens. Alternatively, a formal inter-and intraobserver variability study at our institution may be beneficial.
We observed significant positive correlations between the increasing number of positive core biopsy sites and Gleason scores in biopsies, as well as between increasing prebiopsy serum PSA levels and increasing Gleason scores in biopsies. These data agree well with previous studies that have associated increasing cancer volume with increasing Gleason score.18 The data also agree with studies suggesting that prebiopsy serum PSA levels are predictive of biopsy Gleason scores.19 The results from our study also show that the pre-biopsy serum PSA levels are useful markers for the stratification of risk in patients in whom cancer was diagnosed at biopsy.
A large proportion (38.9%) of cancers diagnosed at biopsy and given a Gleason score of 6 were given a final Gleason score of 7 at prostatectomy.2,20,21 This upgrading of prostate cancers and its clinical importance has been reported recently in a study by urologists and urological pathologists at the University Health Network in Toronto.21 Our experience agrees well with that reported in previous studies. It has been suggested that the upgrading of prostate cancers at prostatectomy is most likely related to sampling error at the time of biopsy.2 The increase in non–organ confined cancers (pT3 tumours) at prostatectomy compared with biopsy is also most likely related to a sampling error at the time of biopsy. The data from our study indicate that the observed increase in Gleason score at prostatectomy is not substantially different than that observed at a nearby tertiary care centre.
Conclusion
The rates of cancerous, precancerous (HGPIN and ASAP) and benign diagnoses in prostate gland biopsies at our institution were significantly different than those reported in the literature. There was also significant variation in the rates of categorical diagnoses and Gleason grading in biopsies among the community pathologists in our study. Based on the results presented, we would encourage other community hospital pathologists, in collaboration with their urologists, to review periodically their prostate gland pathology practices in an attempt to improve the uniformity of diagnoses and Gleason grading. Since this is the first Canadian publication to report rates of diagnoses from a community hospital, we hope that the data can serve as a reference for other Canadian community hospital pathologists who undertake a similar review.
Acknowledgements
The support of Ms. E. Pyke, GBHS librarian, in obtaining references is appreciated. We thank Ms. L. Borland, Ms. S. Ellacott, Ms. T. Lackey and Ms. D. Bartley for secretarial assistance.
Footnotes
Competing interests: None declared.
See related article on page 524
This article has been peer reviewed.
References
- 1.Epstein JI, Allsbrook WC, Amin MB, et al. The 2005 International Society of Urological Pathology (ISUP) consensus conference on Gleason grading of prostatic carcinoma. Am J Surg Pathol. 2005;29:1228–42. doi: 10.1097/01.pas.0000173646.99337.b1. [DOI] [PubMed] [Google Scholar]
- 2.Steinberg DM, Sauvagoeot J, Piantadosi S, et al. Correlation of prostate needle biopsy and radical prostatectomy Gleason grade in academic and community settings. Am J Surg Pathol. 1997;21:566–76. doi: 10.1097/00000478-199705000-00010. [DOI] [PubMed] [Google Scholar]
- 3.Allsbrook WC, Mangold KA, Johnson MH, et al. Interobserver reproducibility of Gleason grading of prostatic carcinoma: urologic pathologists. Hum Pathol. 2001;32:74–80. doi: 10.1053/hupa.2001.21134. [DOI] [PubMed] [Google Scholar]
- 4.Allsbrook WC, Mangold KA, Johnson MH, et al. Interobserver reproducibility of Gleason grading of prostatic carcinoma: general pathologists. Hum Pathol. 2001;32:81–8. doi: 10.1053/hupa.2001.21135. [DOI] [PubMed] [Google Scholar]
- 5.Renshaw AA, Schultz D, Cote K, et al. Accurate Gleason grading of prostatic adenocarcinoma in prostate needle biopsies by general pathologists. Arch Pathol Lab Med. 2003;127:1007–8. doi: 10.5858/2003-127-1007-AGGOPA. [DOI] [PubMed] [Google Scholar]
- 6.Oyama T, Allsbrook WC, Kurokawa K, et al. A comparison of interobserver reproducibility of Gleason grading of prostatic carcinoma in Japan and the United States. Arch Pathol Lab Med. 2005;129:1004–10. doi: 10.5858/2005-129-1004-ACOIRO. [DOI] [PubMed] [Google Scholar]
- 7.Melia J, Moseley R, Ball RY, et al. A UK-based investigation of inter- and intra-observer reproducibility of Gleason grading of prostatic biopsies. Histopathology. 2006;48:644–54. doi: 10.1111/j.1365-2559.2006.02393.x. [DOI] [PubMed] [Google Scholar]
- 8.Griffiths DFR, Melia J, McWilliams LJ, et al. A study of Gleason score interpretation in different groups of UK pathologists: techniques for improving reproducibility. Histopathology. 2006;48:655–62. doi: 10.1111/j.1365-2559.2006.02394.x. [DOI] [PubMed] [Google Scholar]
- 9.Epstein JI, Walsh PC, Sanfilippo F. Clinical and cost impact of second-opinion pathology: review of prostate needle biopsies prior to radical prostatectomy. Am J Surg Pathol. 1996;20:851–7. doi: 10.1097/00000478-199607000-00008. [DOI] [PubMed] [Google Scholar]
- 10.Orozco R, O’Dowd G, Kunnel B, et al. Observations on pathology trends in 62,537 prostate biopsies obtained from urology private practices in the United States. Urology. 1998;51:186–95. doi: 10.1016/s0090-4295(97)00620-1. [DOI] [PubMed] [Google Scholar]
- 11.Novis DA, Zarbo RJ, Valenstein PA. Diagnostic uncertainty expressed in prostate needle biopsies. Arch Pathol Lab Med. 1999;123:687–92. doi: 10.5858/1999-123-0687-DUEIPN. [DOI] [PubMed] [Google Scholar]
- 12.Epstein JI, Herawi M. Prostate needle biopsies containing prostatic intraepithelial neoplasia or atypical foci suspicious for carcinoma: implications for patient care. J Urol. 2006;175:820–34. doi: 10.1016/S0022-5347(05)00337-X. [DOI] [PubMed] [Google Scholar]
- 13.Iczkowski KA. Current prostate biopsy interpretation. Criteria for cancer, atypical small acinar proliferation, high-grade prostatic intraepithelial neoplasia, and use of immuno-stains. Arch Pathol Lab Med. 2006;130:835–43. doi: 10.5858/2006-130-835-CPBICF. [DOI] [PubMed] [Google Scholar]
- 14.Egevad L, Allsbrokk WC, Epstein JI. Current practice of diagnosis and reporting of prostatic intraepithelial neoplasia and glandular atypia among genitourinary pathologists. Mod Pathol. 2006;19:180–5. doi: 10.1038/modpathol.3800522. [DOI] [PubMed] [Google Scholar]
- 15.Bostwick DG, Meiers I. Atypical small acinar proliferation in the prostate: clinical significance in 2006. Arch Pathol Lab Med. 2006;130:952–7. doi: 10.5858/2006-130-952-ASAPIT. [DOI] [PubMed] [Google Scholar]
- 16.McLean M, Srigley J, Banerjee D, et al. Interobserver variation in prostate cancer Gleason scoring: Are there implications for the design of clinical trials and treatment strategies? Clin Oncol. 1997;9:222–5. doi: 10.1016/s0936-6555(97)80005-2. [DOI] [PubMed] [Google Scholar]
- 17.Kronz JD, Silberman MA, Allsbrook WC, et al. A web-based tutorial improves practicing pathologists’ Gleason grading of images of prostate carcinoma specimens obtained by needle biopsy: validation of a new medical education paradigm. Cancer. 2000;89:1818–23. [PubMed] [Google Scholar]
- 18.Sebo TJ, Bock BJ, Cheville JC, et al. The percent of cores positive for cancer in prostate needle biopsy specimens is strongly predictive of tumor stage and volume at radical prostatectomy. Urology. 2000;163:174–8. [PubMed] [Google Scholar]
- 19.Ochiai A, Troncoso P, Babaian RJ. The relationship between serum prostate specific antigen and tumour volume persists in the current era. J Urol. 2007;177:903–6. doi: 10.1016/j.juro.2006.10.072. [DOI] [PubMed] [Google Scholar]
- 20.Poulos CK, Daggy JK, Cheng L. Preoperative prediction of Gleason grade in radical prostatectomy specimens: the influence of difference Gleason grades from multiple positive biopsy sites. Mod Pathol. 2005;18:228–34. doi: 10.1038/modpathol.3800302. [DOI] [PubMed] [Google Scholar]
- 21.Pinthus JH, Witkos M, Fleshner NE, et al. Prostate cancers scored as Gleason 6 on prostate biopsy are frequently Gleason 7 tumours at radical prostatectomy: implications on outcome. J Urol. 2006;176:979–84. doi: 10.1016/j.juro.2006.04.102. [DOI] [PubMed] [Google Scholar]