Xavier Escoda Llorens, Ignacio Amarillo-Gómez, Isaac Carrasco-Rivero, Ignacio Rodríguez-Pizà, Antonio Meléndez-Lazo
{"title":"Comparative Evaluation of Cytological and Histopathological Diagnostic Performance in Detecting Nodal Metastasis of Canine Mast Cell Tumours.","authors":"Xavier Escoda Llorens, Ignacio Amarillo-Gómez, Isaac Carrasco-Rivero, Ignacio Rodríguez-Pizà, Antonio Meléndez-Lazo","doi":"10.1111/vco.70008","DOIUrl":null,"url":null,"abstract":"<p><p>Cytological evaluation of regional lymph nodes (LNs) is routinely used for staging canine mast cell tumours (MCTs), but its diagnostic accuracy requires further validation. This multicentre retrospective study compared Krick's cytological criteria with Weishaar's histopathological classification for detecting LN metastasis in canine MCTs, and evaluated whether cytology with modified Wright-Giemsa stain (WGS) could outperform routine haematoxylin-eosin (H&E) histology in identifying early metastatic LNs. Paired cytological and histopathological evaluations were performed on 65 LNs from 52 dogs undergoing lymphadenectomy, with toluidine blue staining (TBS) applied when metastasis was not evident on H&E. Cytology samples were classified using Krick's criteria (normal/reactive, possible, probable, certain metastasis). Histopathology was graded according to Weishaar's system (HN0-HN3). Diagnostic performance was assessed for three cytologic positivity groups (G): G1 (certain), G2 (certain/probable), and G3 (certain/probable/possible). Cytology detected 87.5% of early (HN2) and 95% of overt (HN3) metastases, outperforming H&E (25% and 85%, respectively). G1 showed moderate sensitivity (75%) but high specificity (93.1%), positive predictive value (93.1%), and positive likelihood ratio (10.5). G3 yielded the highest sensitivity (91.7%) but the lowest specificity (62.1%), with a negative predictive value of 85.7% and a negative likelihood ratio of 0.13. ROC analysis revealed notable diagnostic performance with Kappa values indicating moderate to substantial agreement between cytology and histopathology. Our results validate cytology with WGS as a reliable tool for LN staging in MCTs, outperforming routine H&E histology in the detection of early metastatic LNs. However, the 'possible' and 'probable' categories should be interpreted cautiously. TBS is essential for accurate nodal histopathological assessment. Definitive staging should integrate histopathology and advanced nodal mapping techniques to optimise sentinel LN identification.</p>","PeriodicalId":23693,"journal":{"name":"Veterinary and comparative oncology","volume":" ","pages":""},"PeriodicalIF":1.9000,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Veterinary and comparative oncology","FirstCategoryId":"97","ListUrlMain":"https://doi.org/10.1111/vco.70008","RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"VETERINARY SCIENCES","Score":null,"Total":0}
引用次数: 0
Abstract
Cytological evaluation of regional lymph nodes (LNs) is routinely used for staging canine mast cell tumours (MCTs), but its diagnostic accuracy requires further validation. This multicentre retrospective study compared Krick's cytological criteria with Weishaar's histopathological classification for detecting LN metastasis in canine MCTs, and evaluated whether cytology with modified Wright-Giemsa stain (WGS) could outperform routine haematoxylin-eosin (H&E) histology in identifying early metastatic LNs. Paired cytological and histopathological evaluations were performed on 65 LNs from 52 dogs undergoing lymphadenectomy, with toluidine blue staining (TBS) applied when metastasis was not evident on H&E. Cytology samples were classified using Krick's criteria (normal/reactive, possible, probable, certain metastasis). Histopathology was graded according to Weishaar's system (HN0-HN3). Diagnostic performance was assessed for three cytologic positivity groups (G): G1 (certain), G2 (certain/probable), and G3 (certain/probable/possible). Cytology detected 87.5% of early (HN2) and 95% of overt (HN3) metastases, outperforming H&E (25% and 85%, respectively). G1 showed moderate sensitivity (75%) but high specificity (93.1%), positive predictive value (93.1%), and positive likelihood ratio (10.5). G3 yielded the highest sensitivity (91.7%) but the lowest specificity (62.1%), with a negative predictive value of 85.7% and a negative likelihood ratio of 0.13. ROC analysis revealed notable diagnostic performance with Kappa values indicating moderate to substantial agreement between cytology and histopathology. Our results validate cytology with WGS as a reliable tool for LN staging in MCTs, outperforming routine H&E histology in the detection of early metastatic LNs. However, the 'possible' and 'probable' categories should be interpreted cautiously. TBS is essential for accurate nodal histopathological assessment. Definitive staging should integrate histopathology and advanced nodal mapping techniques to optimise sentinel LN identification.
期刊介绍:
Veterinary and Comparative Oncology (VCO) is an international, peer-reviewed journal integrating clinical and scientific information from a variety of related disciplines and from worldwide sources for all veterinary oncologists and cancer researchers concerned with aetiology, diagnosis and clinical course of cancer in domestic animals and its prevention. With the ultimate aim of diminishing suffering from cancer, the journal supports the transfer of knowledge in all aspects of veterinary oncology, from the application of new laboratory technology to cancer prevention, early detection, diagnosis and therapy. In addition to original articles, the journal publishes solicited editorials, review articles, commentary, correspondence and abstracts from the published literature. Accordingly, studies describing laboratory work performed exclusively in purpose-bred domestic animals (e.g. dogs, cats, horses) will not be considered.