Mitosis Counting in Breast Cancer: Object-Level Interobserver Agreement and Comparison to an Automatic Method.

BACKGROUND: Tumor proliferation speed, most commonly assessed by counting of mitotic figures in histological slide preparations, is an important biomarker for breast cancer. Although mitosis counting is routinely performed by pathologists, it is a tedious and subjective task with poor reproducibility, particularly among non-experts. Inter- and intraobserver reproducibility of mitosis counting can be improved when a strict protocol is defined and followed. Previous studies have examined only the agreement in terms of the mitotic count or the mitotic activity score. Studies of the observer agreement at the level of individual objects, which can provide more insight into the procedure, have not been performed thus far. METHODS: The development of automatic mitosis detection methods has received large interest in recent years. Automatic image analysis is viewed as a solution for the problem of subjectivity of mitosis counting by pathologists. In this paper we describe the results from an interobserver agreement study between three human observers and an automatic method, and make two unique contributions. For the first time, we present an analysis of the object-level interobserver agreement on mitosis counting. Furthermore, we train an automatic mitosis detection method that is robust with respect to staining appearance variability and compare it with the performance of expert observers on an "external" dataset, i.e. on histopathology images that originate from pathology labs other than the pathology lab that provided the training data for the automatic method. RESULTS: The object-level interobserver study revealed that pathologists often do not agree on individual objects, even if this is not reflected in the mitotic count. The disagreement is larger for objects from smaller size, which suggests that adding a size constraint in the mitosis counting protocol can improve reproducibility. The automatic mitosis detection method can perform mitosis counting in an unbiased way, with substantial agreement with human experts.

Validity of whole slide images for scoring HER2 chromogenic in situ hybridisation in breast cancer.

AIM: Whole slide images (WSIs) have stimulated a paradigm shift from conventional to digital pathology in several applications within pathology. Due to the fact that WSIs have not yet been approved for primary diagnostics, validating their use for different diagnostic purposes is still mandatory. The aim of this study was to test the validity of WSI in assessing human epidermal growth factor receptor 2 (HER2) status in breast cancer specimens using chromogenic in situ hybridisation (CISH). MATERIALS AND METHODS: Ninety-six HER2 CISH slides were scored by two observers on a light microscope (400× viewing magnification) and on WSI (40× scanning magnification, one focus plane) with a minimum of 6 weeks washout period. The concordance between digital and microscopic HER2 scores was assessed. RESULTS: Digitally, 93/96 cases could be assessed (96.8%). Microscopic and digital evaluation of HER2 amplification status were concordant in 68/93 cases ((73.1%, 95% CI: 0.639 -0.823), κ 0.588). CISH underscoring was most noticeable in the amplified and equivocal categories while the highest level concordance was seen in cases with a normal copy number. Additionally there was a noticeable tendency to underestimate the average HER2 scores on WSI: lower in 59 and higher in 11 cases. There was no major difference in time spent for microscopic scoring (86.9 s) and digital scoring (81.7 s). CONCLUSIONS: There was a reasonable concordance between microscopic scoring and WSI-based scoring of HER2 copy number of CISH slides. Nevertheless, WSIs scanned on a single focal plane are insufficient to assess HER2 gene amplification status by scoring CISH due to the noticeable tendency towards digitally underestimating the number of HER2 spots. Scanning at multiple focus planes may offer better resolution for improved digital CISH spot counting.

Improved quality of patient care through routine second review of histopathology specimens prior to multidisciplinary meetings.

AIM: Double reading may be a valuable tool for improving quality of patient care by identifying diagnostic errors before final sign-out, but standard double reading would significantly increase costs of pathology. We assessed the added value of intradepartmental routine double reading of histopathology specimens prior to multidisciplinary meetings. METHODS: Diagnoses, treatment plans and prognoses of patients are often discussed at multidisciplinary meetings. As part of the daily routine, all pathology specimens to be discussed at upcoming multidisciplinary meetings undergo prior intradepartmental double reading. We identified all histopathology specimens from 2013 that underwent such double reading and determined major and minor discordance rates based on clinical relevance between the initial and consensus sign-out diagnoses. RESULTS: We included 6796 histopathology specimens that underwent double reading, representing approximately 8% of all histopathology cases at our institution in 2013. Double reading diagnoses were concordant in 6566 specimens (96.6%). Major and minor discordances were observed in 60 (0.9%) and 170 (2.5%) specimens, respectively. Urology specimens had significantly more discordances than other tissues of origin, Gleason grading of prostate cancer biopsies being the most frequent diagnostic problem. Furthermore, premalignant and malignant cases showed significantly higher discordance rates than the rest. The vast majority (90%) of discordances represented changes within the same diagnostic category (eg, malignant to malignant). CONCLUSIONS: Routine double reading of histopathology specimens prior to multidisciplinary meetings prevents diagnostic errors. It resulted in about 1% discordant diagnoses of potential clinical significance, indicating that second review is worthwhile in terms of patient safety and quality of patient care.

Interlaboratory Variability in the Histologic Grading of Colorectal Adenocarcinomas in a Nationwide Cohort.

Differentiation grade of colorectal adenocarcinoma (CRC) is a prognostic factor and important for therapy selection. In patients with stage II colon cancer, poor differentiation is an indication for adjuvant chemotherapy. The variability in daily practice in the grading of CRC was assessed in a nationwide cohort. Using the Dutch Pathology Registry (PALGA), all synoptically reported CRC resections from 2010 to 2013 were identified. Proportions of poorly differentiated (PD) adenocarcinomas were determined and compared between 35 laboratories by univariable and multivariable logistic regression analyses. In total, 11,719 resections of 11,681 patients were included, of which 1427 (12.2%) were PD (range between 35 laboratories: 5.0% to 33.2%). After adjustment for case mix, 4 (11%) laboratories still reported a significantly lower (n=2) or higher (n=2) proportion of PD adenocarcinoma compared with the reference laboratory. Seven of 8 investigated laboratories showed considerable intralaboratory variation between pathologists as well. In a subgroup of 2812 patients (2813 tumors) who could have been eligible for adjuvant chemotherapy solely on the basis of the differentiation grade (stage II colon cancer patients without other high-risk factors [ie, T4, <10 lymph nodes evaluated, perforation, ileus, or angioinvasion]), 258 (9.2%) were PD (range between laboratories: 0% to 22.7%). In this subgroup, 4 laboratories still diagnosed significantly more PD adenocarcinomas after multivariable logistic regression analysis, increasing the number of colon cancer patients eligible for adjuvant therapy. In conclusion, this large nationwide cohort demonstrates considerable interlaboratory and intralaboratory variation in differentiation grading of CRC. Better standardization of grading criteria is needed for optimal determination of prognosis and treatment selection.

Interlaboratory variability in the grading of dysplasia in a nationwide cohort of colorectal adenomas.

AIMS: Although high-grade dysplasia (HGD) is a risk factor for malignant transformation and the future development of adenomas/carcinomas, grade is not incorporated in the Dutch guidelines for colonoscopy surveillance, partly because of presumed interobserver variability. The aim of this study was to analyse, in a nationwide cohort of colorectal adenomas, the interlaboratory variability in the grading of dysplasia in daily practice. METHODS AND RESULTS: From the Dutch Pathology Registry, all synoptically reported classic adenomas in The Netherlands in 2013 were identified. The proportion of adenomas with HGD was determined for biopsies and polypectomies, and compared between 37 laboratories by the use of multivariable logistic regression analyses. In total, 21 030 colonoscopies of 20 270 patients were included. HGD was reported in 530 (3.6%) of 14 866 adenomas diagnosed on biopsies (range between laboratories: 0-13.6%) and in 983 (11.8%) of 8346 adenomas diagnosed on polypectomies (range: 3.1-42.9%). After adjustment for case mix, 13 (35%) laboratories reported a significantly lower or higher frequency of HGD than average. CONCLUSIONS: We observed considerable interlaboratory variation in the grading of dysplasia in colorectal adenomas, which could be only partly explained by differences in case mix. Therefore, better standardization of grading criteria is needed before grade of dysplasia can usefully be incorporated in colonoscopy surveillance guidelines.

Improved cytodiagnostics and quality of patient care through double reading of selected cases by an expert cytopathologist.

Double reading may be a valuable tool for improving the quality of patient care by restoring diagnostic errors before final sign-out, but standard double reading would significantly increase costs of pathology. The aim of this study was to assess the added value of routine double reading of defined categories of clinical cytology specimens by specialized cytopathologists. Specialized cytopathologists routinely re-diagnosed blinded defined categories of clinical cytology specimens that had been signed out by routine pathologists from January 2012 up to December 2013. Major and minor discordance rates between initial and expert diagnoses were determined, and both diagnoses were validated by comparison with same-site histological follow-up. Initial and expert diagnoses were concordant in 131/218 specimens (60.1 %). Major and minor discordances were present in 28 (12.8 %) and 59 (27.1 %) specimens, respectively. Pleural fluid, thyroid and urine specimens showed the highest major discordance rates (19.4, 19.2 and 16.7 %, respectively). Histological follow-up (where possible) supported the expert diagnosis in 95.5 % of specimens. Our implemented double reading strategy of defined categories of cytology specimens showed major discordance in 12.8 % of specimens. The expert diagnosis was supported in 95.5 % of discordant cases where histological follow-up was available. This indicates that this double reading strategy is worthwhile and contributes to better cytodiagnostics and quality of patient care, especially for suspicious pleural fluid, thyroid and urine specimens. Our results emphasize that cytopathology is a subspecialization of pathology and requires specialized cytopathologists.

The value of autopsies in the era of high-tech medicine: discrepant findings persist.

AIMS: Although the autopsy is still the gold standard for quality assessment of clinical diagnoses, autopsy rates have been declining over the last decades to <10%. The aim of this study was to investigate the value of autopsies in the high-tech medicine era by determining the frequency of discrepancies between clinical and autopsy diagnoses. METHODS: We classified all adult autopsy cases (n=460), performed at Symbiant, Pathology Expert Centre, in 2007 and 2012/2013, as having major, or minor discrepancy or total concordance. The roles of possible contributory factors were analysed. Finally, we assessed the role of microscopic examination in identifying cause of death. RESULTS: Major and minor discrepancies were found in 23.5% and 32.6% of the classifiable autopsies, respectively. Most commonly observed major discrepancies were myocardial infarction, pulmonary embolism and pneumonia. Improper imaging and discontinuation of active treatment were significantly associated with a higher and a lower frequency of major discrepancies, respectively. Comparing 2007 and 2012/2013, the frequency of minor discrepancies significantly increased from 26.8% to 39.3%. Final admission length of >2 days was significantly associated with a lower frequency of class III minor discrepancies. Microscopic examination contributed to establishing cause of death in 19.6% of the cases. CONCLUSIONS: Discrepant findings persist at autopsy, even in the era of high-tech medicine. Therefore, autopsies still should serve as a very important part of quality control in clinical diagnosis and treatment. Learning from individual and system-related diagnostic errors can aid in improving patient safety.

Added value of HER-2 amplification testing by multiplex ligation-dependent probe amplification in invasive breast cancer.

BACKGROUND: HER-2 is a prognostic and predictive marker, but as yet no technique is perfectly able to identify patients likely to benefit from HER-2 targeted therapies. We aimed to prospectively assess the added value of first-line co-testing by IHC, and multiplex ligation-dependent probe amplification (MLPA) and chromogenic in situ hybridization (CISH). METHODS: As local validation, HER-2 MLPA and CISH were compared in 99 breast cancers. Next, we reviewed 937 invasive breast cancers, from 4 Dutch pathology laboratories, that were prospectively assessed for HER-2 by IHC and MLPA (and CISH in selected cases). RESULTS: The validation study demonstrated 100% concordance between CISH and MLPA, if both methods were assessable and conclusive (81.8% of cases). Significant variation regarding percentages IHC 0/1+ and 2+ cases was observed between the laboratories (p<0.0001). Overall concordance between IHC and MLPA/CISH was 98.1% (575/586) (Kappa = 0.94). Of the IHC 3+ cases, 6.7% failed to reveal gene amplification, whereas 0.8% of the IHC 0/1+ cases demonstrated gene amplification. Results remained discordant after retrospective review in 3/11 discordant cases. In the remaining 8 cases the original IHC score was incorrect or adapted after repeated IHC staining. CONCLUSIONS: MLPA is a low-cost and quantitative high-throughput technique with near perfect concordance with CISH. The use of MLPA in routinely co-testing all breast cancers may reduce HER-2 testing variation between laboratories, may serve as quality control for IHC, will reveal IHC 0/1+ patients with gene amplification, likely responsive to trastuzumab, and identify IHC 3+ cases without gene amplification that may respond less well.

Evaluation of mitotic activity index in breast cancer using whole slide digital images.

INTRODUCTION: Mitotic Activity Index (MAI) is an important independent prognostic factor and an integral part of the breast cancer grading system. Thus, correct estimation of this prognostically relevant feature is essential for guiding treatment decision and assessing patient prognosis. The aim of this study was to validate the use of high resolution Whole Slide Images (WSI) in estimating MAI in breast cancer specimens. METHODS: MAI was evaluated in 100 consecutive breast cancer specimens by three observers on two occasions, microscopically and on WSI with a wash out period of 4 months. MAI was also translated to mitotic scores as in grading. Inter- and intra-observer agreement between microscopic and digital MAI counts and scores was measured. RESULTS: Almost perfect inter-observer agreements were obtained from counting MAI using a conventional microscope (intra-class correlation coefficient (ICCC) 0.879) as well as on WSI (ICCC 0.924). K coefficients reflected good inter-observer agreements among observers' microscopic mitotic scores (average kappa 0.642). Comparable results were also observed among digital mitotic scores (average kappa 0.635). There was strong to perfect intra-observer agreements between MAI counts and mitotic scores for the two diagnostic modalities (ICCC 0.716-0.863, kappa 0.506-0.617). There were no significant differences in mitotic scores using both diagnostic modalities. CONCLUSION: Scoring mitoses using WSI in breast cancer seems to be just as reliable and reproducible as when using a microscope. Further development of software and image quality will definitely encourage the use of WSI in routine pathology practice.

Intra-operative rapid diagnostic method based on CK19 mRNA expression for the detection of lymph node metastases in breast cancer.

Staging by sentinel node (SN) biopsy is the standard procedure for clinically node-negative breast cancer patients. Intra-operative analysis of the SN allows immediate axillary lymph node (ALN) dissection in SN positive patients, but a quick, reliable and reproducible method is lacking. We tested the suitability of a quantitative cytokeratin 19 (CK19) mRNA one step nucleic acid amplification (OSNA#) technique (OSNA-CK19) for intra-operative SN analysis. OSNA-CK19 involves a short manual sample preparation step and subsequent fully automated amplification of CK19 mRNA based on reverse transcription loop-mediated isothermal amplification, with results available within 30-40 min. OSNA-CK19 was compared to histological staining (Hematoxylin&Eosin and CAM5.2 and CK19 immunostaining) of 346 frozen ALNs from 32 breast cancer patients, using half of the lymph node for each method. 267 samples were negative and 61 positive by both methods. Three samples were histology positive and OSNA-CK19 negative. Fifteen samples were histology negative and OSNA-CK19 positive, 11 of which had copy numbers close to the cut-off level of OSNA-CK19. Seven of these 15 samples were RT-PCR positive for epithelial markers and/or showed CK19 protein expression by Western blot suggesting the presence of tumor deposits in the lymph node part investigated by OSNA-CK19. Concordance with histology was 94.8%, and 96.8% after exclusion of the latter 7 discordant cases. Sensitivity was 95.3% and specificity was 94.7% before and 97.1% after discordant case investigation. Our results indicate that OSNA-CK19 can potentially be useful in an intra-operative clinical setting to detect SN tumor involvement in breast cancer patients.

A patient with endobronchial BALT lymphoma successfully treated with radiotherapy.

Low-grade bronchus-associated lymphoid tissue (BALT) lymphoma is a rare tumour originating from the marginal zone of lymphoid tissue. It is a subgroup of B-cell extranodal non-Hodgkin's lymphoma with an indolent course. We report a case of this tumour with characteristic histologic feature. The patient had non-specific respiratory complaints. The tumour occluded the right bronchus intermedius. He received radiation therapy alone, resulting in complete remission of the tumour and disappearance of symptoms.