New Quality Framework Brings Human Cancer Registry Standards to Veterinary Medicine

When a dog is diagnosed with cancer, the tumor that grows inside it tells a story that reaches far beyond the clinic. Companion animals share our homes, our environments, and increasingly our disease risks, which is why veterinary oncologists have spent decades trying to build cancer registries for pets modeled on the systems that track human malignancies. Yet these animal cancer registries have always faced an uncomfortable question: how good is the data they produce? A new study from Italian veterinary researchers now offers the first structured answer, adapting the rigorous quality standards developed for human cancer registration into a practical framework that any animal registry can apply.

The research, published in Frontiers in Veterinary Science, was conducted by a team at the Istituto Zooprofilattico Sperimentale dell’Abruzzo e del Molise (IZSAM) in Teramo, together with colleagues from the Istituto Zooprofilattico Sperimentale del Piemonte, Liguria e Valle d’Aosta in Torino. Led by Marta Cresci, the group set out to solve a problem that has quietly hampered veterinary oncology for years. Roughly 25 to 30 animal cancer registry initiatives have been described internationally over recent decades, ranging from small hospital-based databases to regional population-based networks, but only a subset remains actively operational or routinely updated. With approximately 34 percent of canine deaths attributable to cancer, and with companion animals serving as valuable models for comparative oncology, the absence of a standardized way to assess registry quality has left veterinary public health without a reliable benchmark.

The framework’s architecture borrows directly from the gold standards of human cancer epidemiology. Since the International Association of Cancer Registries was formed in 1966, and later through the European Network of Cancer Registries and Italy’s AIRTum, human registry quality has been assessed along four dimensions: comparability, validity, completeness, and timeliness. Comparability measures how closely a registry’s coding and classification procedures adhere to international guidelines. Validity refers to the proportion of cases with a given characteristic that truly possess it. Completeness ensures that all cancer cases in a defined population are captured, and timeliness measures how rapidly a registry collects, processes, and reports complete data. The Italian team adapted these four dimensions and added two veterinary-specific domains: registry organization and cancer data sources, reflecting the reality that veterinary registries frequently depend on voluntary participation by veterinary practices and may differ substantially in governance, funding, and geographical coverage.

To build the framework, the researchers conducted a systematized literature review following the typology of Grant and Booth, searching PubMed and Scopus from database inception through December 2025, and consulting grey literature from the International Agency for Research on Cancer, the National Cancer Institute, the European Network of Cancer Registries, and AIRTum. The search yielded 566 records on human registry quality assessment, of which four studies and seven grey literature reports survived eligibility screening, plus eight studies describing Italian animal cancer registries. From this evidence base, the team identified and retained 16 quality indicators, each evaluated against three predefined selection criteria: documented use in human registry quality assessment, applicability to the organizational and epidemiological characteristics of veterinary cancer registries, and availability of objective, reproducible, routinely collectable data. Indicators requiring information not routinely available in veterinary surveillance systems or relying on unverifiable assumptions were excluded.

As a proof of concept, the framework was applied to the Abruzzo Animal Cancer Registry, a pathology-based registry established in 2014 and managed by IZSAM. The registry currently includes neoplastic diagnoses generated by the IZSAM Histopathology Unit during routine diagnostic activities, relying exclusively on passive data collection, meaning that cases are recorded only when veterinary practitioners voluntarily submit biological samples for diagnostic purposes. Because reporting animal cancer cases is not mandatory in Italy, participation is promoted through a regional funding scheme that covers the costs of histopathological examinations for samples submitted within the registry’s catchment area, a mechanism that recognizes the registry as a One Health surveillance tool for monitoring cancer occurrence in companion animals and supporting environmental and public health surveillance.

The results, drawn from 2,265 canine tumors registered between January 2020 and December 2024, reveal both strengths and weaknesses with unusual transparency. All registered cases originated from passive case reporting, with no active case-finding procedures implemented. The number of veterinary facilities contributing to the registry ranged from 34 to 39 annually, corresponding to reporting coverage of just 19.48 to 23.38 percent of all veterinary facilities operating in the Abruzzo Region. Standardized coding systems were applied throughout the study period: every tumor received an ICD-9 topography code, morphological coding was performed using ICD-O-3 between 2020 and 2023, and in 2024, 88 tumors were coded according to ICD-O-3 while 437 were coded using the Vet-ICD-O-Canine-1 morphology classification, a veterinary adaptation of the human oncology coding system that represents an important step toward harmonization.

Validity indicators painted an encouraging picture of progressive data refinement. Missing microchip identification decreased from 34 percent in 2020 to 20 percent in 2024, missing breed information fell from 3 to 0 percent, missing sex data dropped from 7 percent to less than 1 percent, and missing date of birth ranged from 41 percent down to 16 percent, with a median of 6 percent across the study period. Owner and pet addresses were available for all registered cases. No missing values were present for cancer-related variables, as the date of diagnosis and method of diagnosis were available for every tumor throughout the study period. No tumors with an unknown primary site were recorded between 2020 and 2024 according to either the ICD-9 or the Vet-ICD-O-Canine-1 topography coding systems. The proportion of tumors classified as primary ranged from 95.97 to 98.39 percent based on the referring practitioner, and from 93.35 to 97.71 percent using the registry-generated unique tumor identifier, while necropsy-derived diagnoses represented between 0.73 and 2.04 percent of registered tumors.

Timeliness and completeness assessments added further nuance. Registration timeliness for passively collected cases, measured as the interval between the acceptance date and the registration date, showed a median of 8.61 months, ranging from 2.4 to 23.7 months across the study period. Completeness, which cannot be directly estimated in veterinary medicine due to the lack of independent and standardized data sources, incomplete mortality information, and the absence of reliable population denominators, was assessed through temporal trends using the Average Annual Percent Change calculated with Joinpoint Regression Software. The annual number of registered tumors increased from 412 cases in 2020 to 525 cases in 2024, yielding an AAPC of 6.84 percent with a 95 percent confidence interval of 2.36 to 11.70 percent. The researchers interpret this cautiously: biologically, an annual increase of almost 7 percent over a five-year period is far too rapid to represent a true shift in underlying population tumor incidence, which tends to change slowly. Instead, the trend is far more consistent with surveillance network consolidation, reflecting progressive stabilization of reporting practices and engagement of participating facilities rather than a corresponding biological increase in tumor occurrence.

The framework’s authors acknowledge important limitations that temper any temptation to declare victory. The literature review was conducted by a single researcher, introducing potential selection bias, and no formal risk-of-bias assessment of the included studies was performed, consistent with the systematized approach adopted. The application to the Abruzzo registry should be considered a proof of concept demonstrating feasibility rather than a formal validation. Some indicators, particularly those related to active case ascertainment, could not be assessed because the registry relies exclusively on passive reporting. The relative importance of the proposed indicators has not yet been formally weighted through expert consensus. The researchers suggest that consensus-based approaches such as Delphi panels, involving veterinary oncologists, epidemiologists, and health data managers, could confirm the relevance of the selected indicators and define weighted scoring systems capable of generating an overall registry quality score, facilitating benchmarking among registries and contributing to harmonized quality standards at national and international levels.

Nevertheless, the study represents a significant first step toward standardizing quality assessment procedures for animal cancer registries. By providing a structured methodology for evaluating registry organization, data quality, and surveillance performance, the framework may support future harmonization efforts and promote more reliable and comparable veterinary oncology data within the One Health framework. As animal cancer registries become more widely established and data collection becomes increasingly standardized and integrated, complementary approaches could be progressively incorporated to provide a more comprehensive assessment of completeness. For now, the Italian team has given veterinary medicine something it has never had before: a common methodological basis for assessing and comparing animal cancer registries operating in different settings, and a practical foundation for the standardized quality assessment, benchmarking, and harmonization that the field has long awaited.

Subject of Research: A framework for assessing quality indicators in animal cancer registries: a case study

Article Title: A framework for assessing quality indicators in animal cancer registries: a case study

Article References: Cresci, M., Cito, F., Petrini, A., Di Teodoro, G., Baffoni, M., Ru, G., & Crescio, M. I. (2026). A framework for assessing quality indicators in animal cancer registries: a case study. Frontiers in Veterinary Science, 13, Article 1947787. https://doi.org/10.3389/fvets.2026.1947787

Image Credits: AI Generated

DOI: 10.3389/fvets.2026.1947787

Keywords: framework, assessing, quality, indicators, animal, cancer, registries, case, scientific research