Search

Newer data suggest that double expression of MYC and BCL2 proteins (DE) evaluated by quantitative immunohistochemistry (qIHC) may be a powerful marker of worse prognosis in diffuse large B cell lymphoma (DLBCL). Testing for DE status, defined as >40% MYC+ and >50% BCL2+ tumor cells, is recommended in the WHO 2016 classification and clinical trials are using DE scoring to assign therapy arms. However, other data suggest that significant variability in manual DE scoring diminishes the predictive value. Error sources include high interobserver variability (IOV) associated with field choice, discrimination of tumor immunoreactivity from adjacent non-neoplastic cells, cell-to-cell variability in staining intensity, crush artifacts and necrosis. Thus, there is a need for standardized, reproducible approaches for DE scoring by qIHC. To address this need, we have begun developing a novel machine-learning approach to analyze IHC digital pathology whole-slide images, focusing initially on MYC IHC. Digital whole-slide images (400x) of 22 DLBCL cases were uploaded to a web-based annotation platform. Using all cases, one annotator created 138 regions of interest (ROIs) containing approximately 200 nucleated cells and representing a variety of tissue types. Eight pathologists were assigned the same 10 ROIs in which to annotate all nuclei from which ground-truth seed nucleus labels (location, classification) were created for a validation set. Nuclei were classified as “tumor-positive”, “tumor-negative”, “non-tumor-positive”, “non-tumor-negative”, or “unknown”. This generated a set of 15,792 annotations with 1974 +/- 272 (Avg+/-STD) labels/annotator. Agglomerative hierarchical clustering afforded the creation of 2299 ground-truth seed locations. A maximum diameter of 3 mm/cluster was set by visual inspection of annotations. Of these seed locations, 1041 (45%) were detected by 8/8 annotators and, on average, 6/8 agreed on class. 302 +/- 72 (Avg+/-STD) “tumor positive” labels per annotator generated 382 seeds locations, 178 (47%) of which were detected by 8/8 annotators, with an average of 7.5/8 agreeing on class. 286 +/- 168 (Avg+/-STD) “tumor-negative” labels per annotator generated 336 seeds, 195 (58%) of which were detected by 8/8 annotators, with an average of 5/8 agreeing on class. Among all classes, the “tumor-positive” label displayed best overall label agreement whereas the “tumor-negative“ label yielded similar localization rate, but lower class agreement. These promising early findings provide a novel basis for quantifying IOV and utilizing multi-observer agreement to create a ground-truth validation set for a supervised machine learning approach to qIHC. Future efforts will make use of these data to optimize the validation set by rationally determining the number of additional annotations required, optimizing the number of annotators per ROI required, devising an adaptive approach to nuclear clustering based on nuclear density, and utilizing the additional 31,422 annotations in hand from all annotators as a robust algorithm training set.

Objectives: A long-standing problem in lung cancer pathology has been to determine whether two anatomically distinct foci of lung carcinoma are independent primaries or intrapulmonary metastases. While several proposals exist with respect to this problem, it is not known how pathologists use these proposals in actual practice. Methods: We performed a voluntary survey of the Pulmonary Pathology Society, a self-selected group of pathologists with a specialty interest in pulmonary pathology, to determine their opinion on staging various specific scenarios of multiple foci of lung cancer. Results: We found that there was a great deal of disagreement and uncertainty in the approach to these scenarios unless there were also unambiguous molecular markers or a dramatic difference in histology. Conclusions: Pathologists have a high degree of uncertainty and disagreement on staging multiple lung carcinomas. An improved pathologic method is required to distinguish separate primary tumors from intrapulmonary metastases. Until that is available, improved nomenclature is needed to convey the intrinsic uncertainty of the situation.

This Internet-based quiz (http://kathrin.unibas.ch/polyp/) tested the diagnostic variability of 168 pathologists in the diagnosis of 20 colorectal polyps on 3 representative images, including hyperplastic polyps (HPs), traditional serrated adenomas (TSAs), sessile serrated adenomas (SSAs), and tubulovillous adenomas (TVAs). Interobserver variability for each of the 20 lesions was significant and was most pronounced for SSAs. Correct answers were independent of the participant's experience with TVAs, HPs, and TSAs. Participants with gastrointestinal subspecialty training and those who had read a reference article on serrated polyps gave a significantly higher percentage of correct answers for SSAs. The nomenclature used for serrated polyps was generally inconsistent. Our results suggest significant shortcomings in the routine H&E diagnosis of serrated colorectal polyps. A diagnostically unifying concept for lesions of the serrated neoplasia pathway, standardization of nomenclature, training of pathologists, and possibly development of ancillary techniques are of paramount importance for accurate patient management.

We compared the interobserver reproducibility of estimating the adequacy of the squamous component of conventional Papanicolaou (Pap) smears using traditional and newly proposed criteria. Forty conventional Pap smears with varying degrees of squamous cellularity were reviewed by 13 observers who evaluated adequacy (satisfactory vs unsatisfactory) based on the traditional criterion of estimating 10% slide coverage. After being introduced to the new criterion and the reference images, the observers reevaluated adequacy on the same set of smears, using the new criterion and the reference images. With the original criterion of 10% slide coverage, 15 smears had a unanimous designation; the overall kappa value was 0.49 ( P < .001). With the newly proposed adequacy criterion and reference images, 17 smears had a unanimous designation; the overall kappa value was 0.60 ( P < .001). The difference in the kappa correlation coefficients was statistically significant ( P = .007). While traditional and newly proposed criterion resulted in fair interobserver agreement, it seemed that the newly proposed criterion, along with the use of reference images, for evaluating adequacy of the squamous component of conventional Pap smears result in better interobserver reproducibility.

The effect of using a 30% cutoff for the proportion of HER2+ cells on the interobserver variability in the interpretation of HER2 immunohistochemical results was evaluated. Immunostained sections from 96 cases of breast carcinoma were reviewed by 10 pathologists and scored as positive (3+) when uniform strong membranous staining was identified in at least 10% of tumor cells; the actual percentage of cells with such staining was also estimated. The agreement rates and the kappa values using a 30% cutoff were compared with those using a 10% cutoff. These proved to be higher in 62% and 66% of measurements, respectively, with average interobserver rates and kappa values of 72% and 0.54 using the 30% cutoff and 70% and 0.49 using the 10% cutoff (P=.001 for all comparisons). Using a 30% cutoff for the percentage of HER2+ cells by immunohistochemical analysis modestly decreased interobserver variability in the interpretation of HER2 immunohistochemical results.

The density and distribution of lymphocytes infiltrating the vertical growth phase of primary cutaneous melanomas has been suggested by several studies to be of prognostic significance. However, few pathologists comment on tumor-infiltrating lymphocytes (TILs), and there is the perception that the assessment of TILs is subject to great interobserver variability. We studied interobserver agreement on the categorization of TILs; 20 cases of primary cutaneous malignant melanoma with a vertical growth phase component were circulated among 3 pathologists and 3 dermatologists. For each case, TILs were classified as brisk, nonbrisk, or absent according to Clark. Only 1 pathologist (a dermatopathologist) was familiar with the classification of TILs. Observers were given written guidelines and a brief tutorial before their examination of the slides. Our results show that with little instruction, overall agreement among observers was good (kappa values, 0.6 or more), especially among pathologists (kappa values, > 0.7). Three observers had excellent agreement among each other (kappa values, > 0.75). These findings suggest that the categorization of TILs can be easily taught and can be applied with an acceptable level of reproducibility in routine diagnostic practice.

We examined diminutive colonic polyps to identify relationships between thermal electrocoagulation or resection trauma cytologic artifacts, type of thermal electrocoagulation, polyp size, and the interobserver variation among 3 pathologists. The 3 pathologists independently evaluated 119 colonic polyps 5 mm or less in maximum dimension for diagnosis and degree of thermal electrocoagulation or resection trauma cytologic artifacts. The maximum dimension of the polyps and type of thermal electrocoagulation were recorded. The average percentage of polyps in which a definitive diagnosis could not be made because of cytologic artifacts was 16.5% (range, 11.8%-19.3%). Decreasing polyp size was associated linearly with the inability to make a definitive diagnosis owing to cytologic artifacts. Polyps smaller than 2 mm significantly more often could not be definitively diagnosed by at least 1 pathologist owing to cytologic artifacts, including some polyps that were excised without thermal electrocautery. Interobserver variation increased with decreasing polyp dimension. Two millimeters seems to represent a cut point, below which the likelihood that a definitive diagnosis can be made can be increased if thermal electrocoagulation is used. This small size seems to make them especially susceptible to cytologically injurious forces.

The Recombigen®-HIV-l LA Test (Cambridge BioScience Corporation, Worcester, MA) uses recombinant peptides derived from the env gene product in a latex agglutination assay for the detection of antibodies to human immunodeficiency virus type 1 (HIV-1) and was recently approved by the Food and Drug Administration for marketing in the United States. It is intended for use as a screening test in physicians' offices, emergency rooms, and other settings where enzyme immunoassays are not practical or available. Concern has been raised over the sensitivity, specificity, and difficulty in interpretation of the agglutination pattern. The authors report on the sensitivity and interobserver variability of the assay as performed in a blinded fashion in a hospital laboratory by technologists experienced with other latex agglutination assays. In the first study, sera from 50 patients positive by enzyme immunoassay (EIA) (Abbott® HIV EIA) and western blot (WB), performed with EPITOPE® HIV western blot strips were assayed by one technologist using the latex agglutination technique. Forty-six samples were positive and four were negative, yielding a sensitivity of 92%. In the second study, 30 samples consisting of 10 negative by EIA and WB, 10 borderline by EIA and/or indeterminate by WB, and 10 positive by EIA and WB were evaluated by three technologists with the latex agglutination technique. There was agreement among all three technologists in 24 of 30 samples (80%). There was disagreement over one sample from the negative group (one technologist obtained a single false positive result), three from the borderline/indeterminate group, and two from the positive group (three technologists obtained false negative results on two samples). In summary, the authors report interobserver variation in interpreting 20% of tests, reflecting difficulty in assessing weak agglutination. Sensitivity of 92% is below that achievable with the EIA or WB techniques and limits the usefulness of the latex agglutination assay as a screening test.