The international system for serous fluid cytopathology: Interobserver agreement

Interobserver variability in cytopathology: How much do we agree?

Interobserver variability remains a major challenge for cytopathologists despite the development of standardized reporting and classification systems. Indeed, whereas moderate-to-good interobserver agreement is generally achievable when the differential diagnosis between benign and malignant entities is straightforward, high levels of variability make the diagnostic interpretation of atypical and suspicious samples not consistent. This review explores the landscape of interobserver agreement in cytopathology across different anatomical sites.

Comprehensive evaluation of benign and malignant etiologies of different serous effusions with the International System for Reporting Serous Fluid Cytopathology: A multi-institutional study in Taiwan

BACKGROUND

The International System for Reporting Serous Fluid Cytopathology (ISRSFC) was introduced globally in 2019 in response to the absence of a standardized reporting system for serous fluid cytology. This study presents experiences implementing this system across three distinct hospitals in Taiwan.

METHODS

A total of 6177 serous fluid specimens in three hospitals in Taiwan between 2018 and 2020 were retrospectively reviewed and reclassified according to the ISRSFC. Cytohistological correlation and chart review were further performed to investigate etiologies and risks of malignancy (ROMs).

RESULTS

Reclassification showed that 34 (0.7%) of 4838 pleural effusions were nondiagnostic (ND), 4086 (84.5%) were negative for malignancy (NFM), 201 (4.2%) were atypia of undetermined significance (AUS), 92 (1.9%) were suspicious for malignancy (SFM), and 425 (8.8%) were malignant (MAL). The 1231 ascites cases contained 13 (1.1%) ND, 1004 (81.6%) NFM, 53 (4.3%) AUS, 31 (2.5%) SFM, and 130 (10.6%) MAL specimens. In pleural effusions, the ROM was 2.9% for ND, 14.0% for NFM, 52.2% for AUS, 85.9% for SFM, and 95.1% for MAL. In ascites, it was 15.4% for ND, 19.1% for NFM, 52.8% for AUS, 83.9% for SFM, and 92.3% for MAL. In pericardial effusions, it was 0.0% for ND, 11.6% for NFM, 30.8% for AUS, 100.0% for SFM, and 95.2% for MAL. Different effusions' most common benign and malignant etiologies were also disclosed.

CONCLUSIONS

These multi-institutional data have determined the diagnostic usefulness of the ISRSFC, which provides pathologists and physicians with invaluable assistance in correctly classifying effusions for further management.

Review of the impact of the International System for Serous Fluid Cytopathology

The International System for Reporting Serous Fluid Cytology (TIS) has been proposed by an expert working team composed of the International Academy of Cytology and the American Society of Cytopathology, following an international survey. Since its introduction, the TIS has gained worldwide acceptance, and this review aims to assess its global impact. A literature search revealed 25 studies which have presented data on the impact of the TIS. Most of them provide data, including risk of malignancy (ROM) for each diagnostic category, separately for pleural, peritoneal and pericardial effusions, while a few do not separate them. A few studies focus on specific diagnoses like mesothelioma on specific types of fluids or more specific issues like the optimal fluid volume for cytology or interobserver variability. A synopsis of the data from the literature search is presented in four tables. The ROM assessment is discussed, as well as interobserver variability and the use of ancillary diagnostic immunochemistry. In conclusion, our review of the published data suggests that the TIS is a valid classification scheme that has been widely accepted by pathologists globally, is highly reproducible and makes a valuable contribution to clinical therapeutic management.

Digital Examination of LYmph node CYtopathology Using the Sydney system (DELYCYUS): An international, multi-institutional study

BACKGROUND

After a series of standardized reporting systems in cytopathology, the Sydney system was recently introduced to address the need for reproducibility and standardization in lymph node cytopathology. Since then, the risk of malignancy for the categories of the Sydney system has been explored by several studies, but no studies have yet examined the interobserver reproducibility of the Sydney system.

METHODS

The authors assessed interobserver reproducibility of the Sydney system on 85 lymph node fine-needle aspiration cytology cases reviewed by 15 cytopathologists from 12 institutions in eight different countries, resulting in 1275 diagnoses. In total, 186 slides stained with Diff-Quik, Papanicolaou, and immunocytochemistry were scanned. A subset of the cases included clinical data and results from ultrasound examinations, flow cytometry immunophenotyping, and fluorescence in situ hybridization analysis. The study participants assessed the cases digitally using whole-slide images.

RESULTS

Overall, the authors observed an almost perfect agreement of cytopathologists with the ground truth (median weighted Cohen κ = 0.887; interquartile range, κ = 0.210) and moderate overall interobserver concordance (Fleiss κ = 0.476). There was substantial agreement for the inadequate and malignant categories (κ = 0.794 and κ = 0.729, respectively), moderate agreement for the benign category (κ = 0.490), and very slight agreement for the suspicious (κ = 0.104) and atypical (κ = 0.075) categories.

CONCLUSIONS

The Sydney system for reporting lymph node cytopathology shows adequate interobserver concordance. Digital microscopy is an adequate means to assess lymph node cytopathology specimens.

Deep Learning-Based Computational Cytopathologic Diagnosis of Metastatic Breast Carcinoma in Pleural Fluid

A Pleural effusion cytology is vital for treating metastatic breast cancer; however, concerns have arisen regarding the low accuracy and inter-observer variability in cytologic diagnosis. Although artificial intelligence-based image analysis has shown promise in cytopathology research, its application in diagnosing breast cancer in pleural fluid remains unexplored. To overcome these limitations, we evaluate the diagnostic accuracy of an artificial intelligence-based model using a large collection of cytopathological slides, to detect the malignant pleural effusion cytology associated with breast cancer. This study includes a total of 569 cytological slides of malignant pleural effusion of metastatic breast cancer from various institutions. We extracted 34,221 augmented image patches from whole-slide images and trained and validated a deep convolutional neural network model (DCNN) (Inception-ResNet-V2) with the images. Using this model, we classified 845 randomly selected patches, which were reviewed by three pathologists to compare their accuracy. The DCNN model outperforms the pathologists by demonstrating higher accuracy, sensitivity, and specificity compared to the pathologists (81.1% vs. 68.7%, 95.0% vs. 72.5%, and 98.6% vs. 88.9%, respectively). The pathologists reviewed the discordant cases of DCNN. After re-examination, the average accuracy, sensitivity, and specificity of the pathologists improved to 87.9, 80.2, and 95.7%, respectively. This study shows that DCNN can accurately diagnose malignant pleural effusion cytology in breast cancer and has the potential to support pathologists.

Risk of malignancy assessment of the International System for Reporting Serous Fluid Cytopathology: Experience in a community hospital setting and comparison with other studies

BACKGROUND

The International System for Reporting Serous Fluid Cytopathology (ISRSFC) was published recently to provide standard reporting terminology for serous fluid. To date, several ISRSFC reclassification studies have reported a wide range of diagnostic category frequency and the associated risk of malignancy (ROM). Herein, the authors applied the ISRSFC to report pleural and peritoneal effusions retrospectively in a community hospital setting.

METHODS

With Internal Review Board approval, 446 peritoneal effusion specimens and 299 pleural fluid specimens from 576 patients in three community hospitals over a 12-month period were reviewed and reclassified according to the ISRSFC.

RESULTS

After reclassification, in pleural effusions, 18 (5.0%) were nondiagnostic (ND), 273 (76.0%) were negative for malignancy (NFM), 18 (5.0%) were atypia of undetermined significance (AUS), 6 (1.7%) were suspicious for malignancy (SFM), and 44 (12.3%) were malignant (MAL). In peritoneal effusions, after reclassification, 11 (5.5%) were ND, 168 (77.1%) were NFM, 9 (4.1%) were AUS, 2 (0.9%) were SFM, and 27 (12.4%) were MAL. The calculated ROM was 0.0% for ND, 1.8% for NFM, 37.5% for AUS, 83.3% for SFM, and 100.0% for MAL in peritoneal effusions; and the ROM was 8.3% for ND, 1.2% for NFM, 44.4% for AUS, and 100.0% for both SFM and MAL in pleural effusions. Further analysis demonstrated notable heterogeneity among published ISRSFC reclassification studies, although the overall ROMs did not differ significantly from the ISRSFC-determined ROMs (all p values were > .05 for mean ROM comparisons).

CONCLUSIONS

The findings suggested the necessity for each laboratory to perform its own ROM analysis based on its statistics for ISRSFC-tiered classification terminology.