208 Chapter 15 Table 1. Overview data of 3 rounds: 1st according to the WHO classification, 2nd after tutorial about revised classification and 3rd round repeat revised classification after feedback of the results in the 1st and 2nd rounds including case specific heatmap analysis, underlying histology, biological collapse and recurrence. (continued) Round 1 Round 2 Round 3 Pattern Micropapillary 0.27 0.24 NA Pattern Acinar 0.28 0.27 NA Pattern Solid 0.65 0.62 NA *) Size determined with digital mouse in histologic image. The coefficient of variation of invasive measurement (CVinv) was determined for the largest invasive line (Largest) and for sum of maximum 3 invasive lines (Sum1-3). The CVinv was calculated over all cases including “0” for diagnosis of “No invasion” (incl. “0”) and separately only for the drawn invasive lines (excl. “0”). The coefficient of variation for total tumor size (CVtot) is low compared to CVinv. SD = standard deviation, MIA = minimal invasive adenocarcinoma In the first reading of the 70 cases according to the WHO classification unanimous invasion “yes” was scored in 31 cases (44%). In 5 cases (7%) at least one score was given for invasion “yes” and “not sure”, while in another 6 cases (8%) invasion was scored as “yes” and “no”. In the remaining 26 cases (37%) all options were scored at least once. So, in about half of cases, there was disagreement between pathologists in the diagnosis of invasion. The distribution of pathologists’ WHO pattern scores across 70 cases is shown for round 1 in Figure 2A. The assignment of the dominant (lepidic, papillary, micropapillary, acinar and solid) revealed a near unanimous score (>39 pathologists) in only 6 cases: five cases with solid and 1 with acinar pattern, while in 21 cases 70% of the pathologists scored the same dominant pattern: n= 13 solid, n=7 acinar and n=1 papillary. The distribution of pathologists’ scores for ‘invasion’, ‘no invasion’ or ‘not sure’ is for each case shown in Figure 3A. Of note, in about half of the cases all pathologists agreed on invasion. However, in a about a third of the cases there is variation across samples and individual pathologists. This is not explained by geographical differences (Table 2).
RkJQdWJsaXNoZXIy MTk4NDMw