NEWS & PRESS

Interobserver Agreement Grade

The AJCC Cancer Staging Manual 8th Edition contained tumor grades at the pathological prognostic stage for breast cancer. Because of the known subjectivity of the tumor classification, we wanted to assess Interobserver`s level of compliance with invasive cancer in pathologists and determine its effect on the prognostic pathological stage. 100 successive cases of invasive stage II carcinomas were graduated twice independently, with an intermediate flushing time of 4 weeks, 6 breast pathologists using established nottingham evaluation criteria. The variability between and within the observer was determined for the general class and for each of the three components of the evaluation. The variability of the interobserver was good to very good (range – 0.582-0.850) with greater intra-observer variability (average – 0.766). Tubule Score was the most reproducible element (n- 0.588). The full agreement was reached in 54 cases and 58 times in rounds 1 and 2. In the first round, in only 25 ambiguous cases, this resulted in different pathological prognostic cases, 18 of which were LEVEL AI compared to the IB. In conclusion, the classification agreement between pathologists was good to very good, and ambiguous degrees led to small changes in the pathological stage of prognostic. Category-specific rates for Nottingham`s 35 invasive breast cancers ranged from 0.5 to 0.7%, with a corresponding percentage of 61.4 to 87.8%. Nottingham ≥7 or <7 values were slightly better (up 0.7). The intraobserver agreement for Nottingham grades (87-100%) score (94.7-98%) much better than the Interobserver agreement (although the value of this intraobserver agreement is limited to the extent that only one pathologist was used for this calculation, it gives a sense of reproducibility for the reference standard used to assess the accuracy of the classification in this study).

In general, disagreement over classification was attributed to differences in the classification of Class II carcinomas. Although there was a relatively wide margin in the Interobserver agreement for Nottingham grades, the accuracy for the rating classification was quite high, ranging from 75 to 100% (average, 83.3%) For Class I; 50 to 83.3% (average, 64.6%) 79-100% (average, 92.3%) For Class III tumours. Harvey JM, from Klerk NH, Sterrett GF . Histological classification in breast cancer: interobserver agreement and relationship with other prognostic factors, including poidia. Pathology 1992;24:63-68. This work is supported in part by the National Cancer Institute, National Institutes of Health under RFA #CA-95-011. The content of this document does not necessarily reflect the opinions or guidelines of the National Cancer Institute or any of the CFR`s cooperating centres, nor does it reflect the approval of the U.S. government or the CFR.

Other pathologists who participated in the evaluation of the Interobserver slides are Deon J Venter, MD, and Jane Arms MD, University of Melbourne, Melbourne, Australia. We pay tribute to Thelma Santa Maria and Caroline Tudor at Stanford University and Marie Maguire, Mount Sinai Hospital, Toronto, for their support in the preparation of paper and figurines. Results: overall gross compliance with the Atypia classification was 60%. The agreement adjusted for the overall odds was fair (up 0.28). Based on histological results, cases were stratified in group A (HG dysplasia or worse) and in group B (LG or medium class [IG] dysplasia after follow-up).


Please direct requests and inquiries to Guarachi Wine Partners, 22837 Ventura Blvd, 3rd Floor, Woodland Hills, CA 91364 or call 818-225-5100