U J - Spring 2012.pdf 486 | Urological Oncology Inter/Intra-Observer Reproducibility of Gleason Scoring in Prostate Adenocarci- noma in Iranian Pathologists Alireza Abdollahi,1 Alipasha Meysamie,2 Sara Sheikhbahaei,2 Ali Ahmadi,1 Hedieh Moradi- Tabriz,1 Mohammadreza Bakhshandeh,1 Hassan Hosseinzadeh1 Purpose: To measure the level of inter/intra-observer reproducibility among pa- thologists as far as Gleason scoring of adenocarcinoma of the prostate is con- cerned. Materials and Methods: A total of 101 prostate biopsy slides, diagnosed with ad- were exposed to Gleason scoring. Two months later, the slides were re-examined by three of the same pathologists. Thereafter, the kappa was calculated for the data pathologists. Results: Inter-observer reproducibility was inappropriate, but intra-observer di- agnostic reproducibility was almost perfect with a corresponding percentage of Conclusion: The inter-observer reproducibility was poor. Keywords: prostatic neoplasms, neoplasm grading, methods, humans Corresponding Author: Alireza Abdollahi, MD Department of Pathology, Imam Khomeini Teaching Hospital, Tehran University of Medical Sciences, Teh- ran, Iran Tel: +98 912 122 0588 Fax: +98 21 8826 9844 E-mail: dr_p_abdollahi@ yahoo.com Received December 2010 Accepted April 2011 1Department of Pathol- ogy, Tehran University of Medical Sciences, Tehran, Iran 2Department of Com- munity Medicine, Tehran University of Medical Sciences, Tehran, Iran Urological Oncology 487Vol. 9 | No. 2 | Spring 2012 |U R O LO G Y J O U R N A L Reproducibility of Gleason Grading | Abdollahi et al INTRODUCTION P -cer, is the most prevalent type of cancer in men in the United States. It is also the sec- ond leading cause of cancer-related deaths in men, just following lung cancer. The overall prostate cancer detection rate in our community is 3.5%. The gold standard in the diagnosis of the PCa is biopsy and making a histological diagnosis of carcinoma. When the tissue sample indicates presence of carcinoma, its Gleason scoring is one of the most important elements in reporting. In this method, tumors are graded, based on their pattern of growth and the level of differentiation, from 1 to 5; grade 1 has the lowest and grade 5 the highest level of differentiation. One of the contributing factors to this observed up- grading in Gleason scoring is the level of patholo- gist experience. Since Gleason score is one of the most important prognostic factors for the out- come of treatment in PCa and even determines the treatment of choice for the tumor, a high de- gree of precision in its reporting and the agreement among the reports of the different pathologists for the same sample are crucial issues. Despite the fact that Gleason scoring is simple, there is an inter- observer variability of the scores. Gleason once said that “If I re-score my previously scored sam- ples, in 50% of cases, I report the same scores and scores”. - cates the concordance rate of the report varied be- tween 0.16 and 0.836. For example, in a study conducted by Rodriguez-Urrego and colleagues, the inter-observer agreement was excellent with k good with k = 0.65 and even more. In this study, regarding the crucial importance of the reproducible and concordant reporting of the samples among different pathologists, we want to obtain approximations for these two variables among the pathologists working in Iran. MATERIALS AND METHODS In this cross-sectional study, 101 tissue samples of the prostate adenocarcinoma obtained through needle biopsy were re-examined to be Gleason- scored. thin microscopic sections, and after being stained Table 1. Gleason’s microscopic grading system of the prostate carcinoma. Stage Description 1 Single, separate, uniform glands in closely packed masses with a definite, usually rounded edge limiting area of tumor. 2 Single, separate, slightly less uniform glands, loosely packed (separated by small amounts of stroma), with less sharp edge. 3a Single, separate, much more variable glands, may be closely packed, but usually irregularly separated, with ragged, poorly defined edge. 3b Like 3a, but very small glands or tiny cell clusters. 3c Sharply and smoothly circumscribed rounded masses of papillary or loose cribriform tumor (papillary intra- ductal tumor). 4a Raggedly outlined, raggedly infiltrating, fused glandular tumor. 4b Like 4a, with large pale cells (hypernephroid). 5a Sharply circumscribed, rounded masses of almost solid cribriform tumor, usually with central necrosis (comedocarcinoma). 5b Ragged masses of anaplastic carcinoma with only enough gland formation or vacuoles to identify it as adenocarcinoma. 488 | randomly selected pathologists to be scored using was based on the degree of glandular differentia- tion and the growth pattern of the tumor compared Sections that cannot be scored, those extracted from patients previously treated with anti-andro- genic drugs or radiotherapy, and samples contain- ing less than 5 malignant acini were excluded from the study. After selection of the samples, a code was given to each of them. Thereafter, the scores given by each sheet. Two months later, the same samples with altered code were sent back to three of the patholo- gists to be re-scored. Finally, the concordance rate This research was carried out according to the principles of the Declaration of Helsinki. The local Ethics Medical Committee of Tehran University of Medical Sciences approved the study protocol. Our statistical analysis included calculation of second data report and comparison of kappa be- tween pathologists. Kappa varied between 0 and 1; the greater the kappa, the higher the concord- Table 2. Percentages of agreement and Kappa values of all possible pair combination of 5 pathologists’ grading scores.*† O1T1 O2T1 O3T1 O4T1 O5T1 O3T2 O4T2 O5T2 O1T1 40.00% (30.40% to 49.60%) 36.50% (27.06% to 45.94%) 60.00% (50.40% to 69.60%) 35.70% (26.31% to 45.09%) 39.10% (29.54% to 59.10% (49.46% to Agreement, % 0.24 0.25 0.19 0.19 0.21 0.47 Kappa O2T1 46.60% 34.50% 50.00% (40.20% to 41.40% (31.75% to 51.05%) 31.00% (21.94% to 40.06%) 46.60% Agreement, % 0.34 0.19 0.15 0.35 Kappa O3T1 37.10% (27.63% to 46.57%) 52.60% 62.39%) 94.29%) 31.00% (21.94% to 40.06%) Agreement, % 0.19 0.4 0.12 0.34 Kappa O4T1 41.00% (31.36% to 50.64%) 37.60% 47.09%) (70.56% to 40.20% (30.59% to Agreement, % 0.25 0.2 0.72 0.24 Kappa O5T1 52.10% (42.31% to 95.06%) Agreement, % 0.39 0.21 Kappa O3T2 (21.75% to Agreement, % 0.11 0.35 Kappa O4T2 41.00% (31.36% to 50.64%) Agreement, % 0.24 Kappa * P < .001. † Urological Oncology 489Vol. 9 | No. 2 | Spring 2012 |U R O LO G Y J O U R N A L 0.60 moderate agreement, 0.61 to 0.80 substan- perfect agreement. Eventually, the data were analyzed both descriptively and analytically using SPSS (the Statistical Package for the Social Sci- ences, Version 15.0, SPSS Inc, Chicago, Illinois, RESULTS Percentages of agreement and Kappa values of all - Overall kappa values in different Gleason scores Using weighted kappa values, there was no sig- between poorly differentiated and moderately dif- Intra-observer diagnostic reproducibility was al- most perfect with a corresponding percentage of DISCUSSION Today, the Gleason system (prostate adenocarci- tumor grading, crucial for both patients and doc- tors. We found an extremely low reproducibility be- - - bility has a good level when the slides were reported - lacks an integrated and regular education system in pathology. In addition, there are obvious limitations in the ac- curacy of grading based on the small amount of tis- sue available from needle biopsies of the prostate. On the other hand, one should recognize a patho- logical misinterpretation. Differences of opinion are related to different interpretations of tumor grading, which is a qualitative indicator. Naturally, this qualitative factor may be interpreted differ- ently by pathologists. Therefore, the difference in interpretation should not be construed as an error. In this study, the reproducibility was meaningfully proportionate with Gleason score of the samples. It seems that the reproducibility would rise when the Ozdamar and colleagues reported an acceptable in- ter-observer variation for Gleason-style grading. Furthermore, Allsbrook and associates examined were involved. The reproducibility stood at an ac- ceptable level. The reason behind different con- clusions from these studies might be related to the pathology education system. Holding meetings for exchange of views, conferences, journals, and group studies may bring views and interpretations together. The lack of such programs in our country must explain the differences. with cancer produced an inter-observer reproduc- training course for pathologists contributing to - lar training to pathologists. Table 3. Inter-observer reproducibility of Gleason’s grad- ing system for prostatic carcinoma.* Gleason Score Kappa Prob > Z 4 - 0.0070 0.5939 5 0.0417 6 0.4033 < 0.001 7 < 0.001 < 0.001 9 0.3402 < 0.001 10 0.3964 < 0.001 Combined < 0.001 Scores 4 and 5 were not significant. Reproducibility of Gleason Grading | Abdollahi et al 490 | This study faced restrictions, such as undermanned samples. Therefore, studies with more pathologists and samples are recommended for the future. CONCLUSION carcinoma, regular and effective training courses are strongly recommended for pathologists in or- der to raise intra/inter-observer reproducibility. ACKNOWLEDGEMENTS The authors are sincerely grateful to the Vice Chan- cellor for Research Affairs at Tehran University of this study. CONFLICT OF INTEREST None declared. REFERENCES 1. Gleason DF, Mellinger GT. Prediction of prognosis for pro- static adenocarcinoma by combined histological grading 2. cancer by measuring free and total serum prostate-specific 3. Hosseini SY, Moharramzadeh M, Ghadian AR, Hooshyar H, prostate cancer by measuring total serum prostate-specific 4. Bostwick DG. Gleason grading of prostatic needle biopsies. Correlation with grade in 316 matched prostatectomies. 5. Vira MA, Tomaszewski JE, Hwang WT, et al. Impact of the percentage of positive biopsy cores on the further stratification of primary grade 3 and grade 4 Gleason score 7 tumors in radical prostatectomy patients. Urology. 6. D'Amico AV, Renshaw AA, Arsenault L, Schultz D, Richie JP. Clinical predictors of upgrading to Gleason grade 4 or 5 disease at radical prostatectomy: potential implications for patient selection for radiation and androgen suppression 7. Kulkarni GS, Lockwood G, Evans A, et al. Clinical predictors of Gleason score upgrading: implications for patients con- sidering watchful waiting, active surveillance, or brachy- Patel AA, Chen MH, Renshaw AA, D'Amico AV. PSA failure following definitive treatment of prostate cancer hav- ing biopsy Gleason score 7 with tertiary grade 5. JAMA. 9. Whittemore DE, Hick EJ, Carter MR, Moul JW, Miranda-Sou- sa AJ, Sexton WJ. Significance of tertiary Gleason pattern 5 in Gleason score 7 radical prostatectomy specimens. J Urol. 10. Albertsen PC, Hanley JA, Gleason DF, Barry MJ. Competing risk analysis of men aged 55 to 74 years at diagnosis man- aged conservatively for clinically localized prostate cancer. 11. Gleason DF. Histologic grading of prostate cancer: a per- 12. Allsbrook WC, Jr., Mangold KA, Johnson MH, Lane RB, Lane CG, Epstein JI. Interobserver reproducibility of Gleason grading of prostatic carcinoma: general pathologist. Hum 13. Melia J, Moseley R, Ball RY, et al. A UK-based investigation of inter- and intra-observer reproducibility of Gleason grad- 14. Lotan TL, Epstein JI. Clinical implications of changing defi- nitions within the Gleason grading system. Nat Rev Urol. 15. Brennan P, Silman A. Statistical methods for assessing ob- 4. 16. Rodriguez-Urrego PA, Cronin AM, Al-Ahmadie HA, et al. Interobserver and intraobserver reproducibility in digital and routine microscopic assessment of prostate needle 17. Ozdamar SO, Sarikaya S, Yildiz L, Atilla MK, Kandemir B, Yildiz S. Intraobserver and interobserver reproducibility of WHO and Gleason histologic grading systems in prostatic Mulay K, Swain M, Jaiman S, Gowrishankar S. Gleason scor- ing of prostatic carcinoma: impact of a web-based tutorial on inter- and intra-observer variability. Indian J Pathol Urological Oncology