Downloaded from http://jcp.bmj.com/ on June 10, 2015 - Published by group.bmj.com

Original article

Downgrading of biopsy based Gleason score in prostatectomy specimens Kilian M Treurniet,1 Dominique Trudel,2,3 Jenna Sykes,4 Andrew J Evans,3 Antonio Finelli,5 Theodorus H Van der Kwast3 1

Leiden University Medical Centre, Leiden, The Netherlands 2 Cancer Research Center, Laval University, Quebec City, Quebec, Canada 3 Department of Pathology, University Health Network, Toronto, Ontario, Canada 4 Department of Biostatistics, University Health Network, Toronto, Ontario, Canada 5 Division of Urology, Department of Surgical Oncology, University Health Network, University of Toronto, Canada Correspondence to Dr T van der Kwast, Department of Pathology, University Health Network, Toronto General Hospital, 200 Elizabeth Street, Toronto, Canada M5G 2C4; [email protected] Received 31 October 2012 Revised 23 August 2013 Accepted 6 October 2013 Published Online First 29 October 2013

ABSTRACT Aims To assess the frequency and possible causes of downgrading from a Gleason score (GS) 7 at biopsy to a GS ≤6 at radical prostatectomy (RP) in a Canadian referral centre. Methods Data were extracted from diagnostic reports of inhouse biopsies and matching prostatectomy specimens from 2008 to 2011 with a GS 7 at biopsy. Biopsies and corresponding prostatectomy specimens of downgraded cases were reviewed. Pathological features were assessed and possible predictors for downgrading were identified. Results Based on pathology reports, 29 (8.9%, 95% CI 5.8% to 11.9%) of the 327 eligible cases were downgraded from biopsy GS 7 to RP GS 6, 72% of them representing a GS ≤6 with tertiary grade 4 at RP. Agreement at review of downgraded RP specimens for Gleason grading was fair and of borderline significance (κ=0.34, 95% CI −0.01 to 0.68, p=0.055) with 65% agreement for tertiary grade. The predominant Gleason grade 4 pattern found in the downgraded biopsies was ill-formed glands. The number of cores with Gleason grade 4 component was found to be the strongest negative predictor of downgrading ( prereview OR=0.56 (95% CI 0.39 to 0.80, p=0.002), postreview OR=0.19 (95% CI 0.07 to 0.52, p=0.001)). Conclusions The frequency of GS 7 in biopsies subsequently downgraded in RP is low and is associated with International Society of Urological Pathology modified Gleason grade 4 patterns. Downgrading could be attributed in most cases to the presence of a tertiary Gleason grade 4 pattern in the RP specimen. Interobserver agreement for the presence of tertiary grade 4 in RP specimens is moderate.

INTRODUCTION

To cite: Treurniet KM, Trudel D, Sykes J, et al. J Clin Pathol 2014;67: 313–318.

The Gleason score (GS) at biopsy is one of the most important prognostic parameters in prostate cancer. In addition to serum prostate specific antigen (PSA) and clinical stage, GS largely determines the treatment options presented to a patient. In 2005, the original Gleason grading system, dating back from 1966, was modified at the International Society of Urogenital Pathology consensus conference on Gleason grading.1 2 Major changes included the addition of nearly all cribriform patterns and ill-formed glands to the Gleason grade 4 category. Single tumour cells were no longer acceptable for Gleason grade 3.3 The reporting of small amounts of high-grade patterns in a biopsy was specified: if a cancer largely consists of a grade 3 pattern with a minute component of unequivocal high-grade cancer, the latter should be incorporated in the GS as the secondary grade. For

Treurniet KM, et al. J Clin Pathol 2014;67:313–318. doi:10.1136/jclinpath-2012-201323

the GS at radical prostatectomy (RP), the original requirement was upheld: a high-grade component is included in the GS only if it represents at least 5% of the cancer volume. Specifically, when a predominantly GS 6 prostate cancer in a prostatectomy has a Gleason grade 4 cancer comprising less than 5% of the cancer volume, such a case would be reported as a GS 6 (3+3) prostate cancer with a tertiary grade 4 and not as a GS 7 (3+4).2 3 Despite the reported improvement in concordance between GS at biopsy and RP since the introduction of the modified GS,4 5 several studies continued to identify discrepancies between biopsy and RP GS.6–10 While an upgraded GS after RP is most common, 7.4% to 30% of the GS 7 (3+4) biopsies were shown to be downgraded to GS 6 (3+3) or lower at RP.6–9 The consequences of downgrading of a biopsy GS 7 are clinically relevant since patients diagnosed with a GS 7 (3+4) cancer no longer meet the eligibility criteria for active surveillance according to most protocols.11 In this study, we assessed the frequency of downgrading from a GS 7 at biopsy to a GS≤6 at RP based on data from pathology reports and after review of the downgraded cases. A systematic slide review was performed to determine the potential histological factors that may have played a role in the reported grade discrepancies. Further, we identified biopsy parameters that predict downgrading in the subsequent RP.

METHODS Patient selection Patients diagnosed at our tertiary care referral centre with a biopsy GS 7 prostate cancer diagnosis between June 2008 and July 2011 were identified using the laboratory information system of the University Health Network (UHN), Toronto, Canada. Only inhouse cases with a corresponding RP performed at UHN were included in the analysis. Histopathological reporting of prostate biopsies and RP specimens was done by dedicated subspecialty urological pathologists. The institution’s ethical board approved this study and institutional authorisation was obtained (Coordinated Approval Process for Clinical Research ID: 12-5109).

Pathological assessment Gleason grades at biopsy and at RP, including presence of a tertiary grade, were recorded. For downgraded cases, we recorded pathological stage at RP and assessed their PSA levels. For the biopsy data-set, the number of cores in the biopsy, the number of cores involved with cancer, percentage 313

Downloaded from http://jcp.bmj.com/ on June 10, 2015 - Published by group.bmj.com

Original article of the tissue involved with cancer and percentage of cancer containing Gleason grade 4 or 5 patterns were extracted from the pathology reports. RP data were also collected, including GS, weight of the prostate, the percentage of prostate involved with cancer and, if present, the percentage of cancer containing Gleason grade 4 patterns. Both biopsy and RP slides from cases that were downgraded to a GS≤6, without regard to tertiary grade, were retrieved for pathology review by an experienced urogenital pathologist (THVdK). During the biopsy slide review, data obtained in the pathological reports were verified. Additionally, the variants of Gleason grade 4 pattern (ie, large or small cribriform, large- or small-fused and ill-formed glands) and possible causes of overgrading of the biopsy like tangentially sectioned glands were recorded. In order to blind the pathologist during review of the RP specimens, 10 random non-downgraded RP were added to the set of downgraded RP, but their findings were not included in the inter-observer statistics analysis. During review, the RP cases were graded and the proportion of high-grade patterns in the cancer was recorded if present.

Statistical analysis The frequency of downgraded cases was calculated and a 95% CI for this frequency was determined assuming an asymptotic binomial distribution. The pathological data of downgraded cases were compared with those of cases with concordant GS at biopsy and matching RP using the student t test. Q–Q plots were used to assess the normality assumption of all variables and non-normal distributions were corrected using ln-transformation. Binary logistic regression modelling was done to predict the probability of downgraded GS for the prereview and postreview data-sets. The variables most strongly associated with GS downgrading were selected via the purposeful selection algorithm.12 In this iterative algorithm, variables that first showed potential for association on univariate analysis at p=0.2 are chosen as the base model, and the remaining variables were then added one-by-one to this base model. If the coefficients of the variables in the base model changed by 20% when a new variable was added, then the model was updated with this new confounder. A weighted κ score with a quadratic weighting scheme was calculated to measure agreement for RP GS after review of the downgraded RP specimens. Two-sided p values of 0.05 were used to assess statistical significance. All analyses were performed using SPSS V.20.0.0.

RESULTS Pathology findings based on pathology reports ( prereview) A total of 327 cases with a GS 7 at biopsy and with corresponding RP were identified. In the prereview data-set, 29 of the 327 (8.9%; 95% CI 5.8% to 11.9%) of the GS 7 (3+4)/(4+3) biopsies were downgraded to a ≤GS 6 after RP (10.5% of the GS 7 (3+4) cases). This includes one case with a tertiary grade 5. In 21 of the 29 (72%) downgraded RP, a tertiary grade 4 was reported in the RP specimen (table 1). If the Epstein eligibility criteria for active surveillance is applied (ie, PSA 50% cancer), 11 of the 29 downgraded cases would have been eligible for active surveillance if their biopsy had been graded as GS 6.13 Furthermore, among the 18 men not eligible for active surveillance based on pretreatment criteria, extra-prostatic extension (two focal, two established) was found at RP in four cases. In all, 24 of 327 biopsy GS 7 cases (7.3%) were upgraded to GS 7 with tertiary grade 5 (n=16) or GS≥8 (n=8). 314

Table 1 Gleason score concordance of biopsy and radical prostatectomy Radical prostatectomy Gleason score (%) ≤6 Biopsy Gleason score Prereview data 3+4 4+3 Total Postreview data 3+4, 4+3 6 Total

No tertiary grade

Tertiary grade

≥7

7 (2.7%) 0 (0%) 7 (2.1%)

20 (7.8%) 2 (2.8%) 22 (6.7%)

229 (89.5%) 256 69 (97.2%) 7 298 (91.2%) 327

4 (1.3%) 1 (20%) 5 (1.6%)

10 (3.2%) 4 (80%) 14 (4.4%)

302 (95.6%) 316 0 (0%) 5 302 (94.1%) 321

Total

Analysis of the biopsy parameters associated with ( prereview) downgrading revealed significantly lower percentages of highgrade cancer at biopsy (respectively p=0.0007) and numbers of biopsy cores with a high-grade component ( p

Downgrading of biopsy based Gleason score in prostatectomy specimens.

To assess the frequency and possible causes of downgrading from a Gleason score (GS) 7 at biopsy to a GS ≤6 at radical prostatectomy (RP) in a Canadia...
500KB Sizes 0 Downloads 0 Views

Recommend Documents