Introduction

Introduction

The contents of this website are: guidelines of the parameters used by veterinary pathologists to evaluate tumors (mitotic count, margins, lymphovascular invasion, necrosis), including outcome assessments, synoptic reporting, computational pathology, definitions, images of mitotic figures, atypical mitotic figures and mitotic like figures and protocols designed to gather required data elements of specific tumor types. Updates for existing guidelines and protocols will be posted, dated, and older versions archived. These guidelines and protocols are intended to provide a standardized approach to the gross, histological and cytological assessment of tumors and accrue data so that, over time, large data sets with comparable information can be assessed and studies uniformly validated. This will permit the collection of comprehensive data which are comparable between investigators and institutions. Validation of existing study results and grading schemes in veterinary oncology are needed and should be published before being adopted clinically and used to provide prognoses or direct therapy. The philosophy in veterinary oncology seems to be that once a system or method is created, it is put into use and remains in use regardless of whether the system has been validated. It is not known how that system or the methods used will perform when different pathologists apply it to different populations of patients. Consensus statements that support the use of published grading schemes are not validation. Most of our existing studies that used gross and histologic assessments to predict patient outcomes need to be repeated with standardized methods and terminology. 

Evaluating tumors with the same methods is necessary for data to be merged, compared, and applied widely. Pairing this information with thorough historical and clinical information and documented patient outcomes will permit meaningful determination of useful prognostic and or predictive parameters. Parameters that are not correlated with accurate patient outcomes have limited clinical utility. Providing this information in an internationally accessible forum will allow information to be accrued from a multitude of institutions with diverse geographical areas and will cast a wider net for accruing large data sets that are universally applicable. 

Guidelines are intended to be “best practices” for our methods, but they extend beyond “best practices” as they provide brief literature reviews, identify areas of weakness and or conflicting approaches, and suggested fields of investigation for future studies to improve a method. The Guidelines and Protocols are based on review of the literature and the expertise of authors; they are intended to bring consistency and reproducibility to the evaluation of tumors. They are not certified by any standards-related body and they represent the authors’ interpretation and application of the data reviewed. Whether a governing body will participate in further development, implementation and updating these Guidelines and Protocols will partially depend upon the success of the website and how widely it is accepted by the veterinary oncology community. Application of these Guidelines and Protocols will vary for different laboratories and pathologists, depending on the availability of equipment, technical expertise, caseload and goals of the laboratory and personnel. Editors and reviewers of journals can use guidelines and protocols as checklists to ensure that the methods used, and the parameters evaluated are complete and provided in such detail that others can reproduce the methods. Results of oncology studies that fail to replicate have multiple causes, one of which is inadequately detailed methods. 

Guidelines and protocols are organized as checklists and currently, the protocols for tumors are extensive, as, for many tumors, there have not been detailed, thorough, replicated, large-scale examinations of multiple factors involved in prognosis. The protocols will be most useful for investigators gathering complete data to determine which parameters are predictive and may not be practical for routine diagnostic evaluation. Shortened versions of the guidelines and protocols will be posted based upon the authors consensus of the key parameters recommended to be included in diagnostic reports. These condensed versions will be aligned with synoptic reporting. Since all parameters detailed in protocols are unlikely to be included in diagnostic reports, record review studies will be insufficient for determining the prognostic features of tumors. Furthermore, to ensure consistency, tumor tissue or images used in oncology publications should be reviewed by author pathologist(s) as a component of the study methods. Guidelines and Protocols will be modified when studies determine which parameters and methods have the most predictive value.  

We would like others to help us maintain the relevancy and usefulness of this website. We are recruiting volunteers to help develop new guidelines and protocols, contribute images and design teaching aids. We are developing a catalog of images for mitotic figures, atypical mitotic figures, mitotic like figures, fascial planes, percentage necrosis, and lymphovascular invasion. Protocols are needed for additional tumors and additional guidelines are needed. Future guidelines might include cytologic assessments, cellular and nuclear pleomorphism, proliferative indices, molecular profiling, genetic tests, lymph node evaluation, statistical designs for oncology studies and synoptic checklists for surgical pathology reports. Future protocols could address any aggressive tumor type or tumors with an existing grading system(s). Use current protocols as a template and existing literature to identify what needs validation and improvement. Contact the website administrators (listed below) about potential new guidelines, protocols or other content. At the end of each guideline and protocol is a section titled: Future Considerations which is designed to identify deficiencies and possible ways to improve that method or protocol. Each guideline and protocol have the email of the “communication author(s)” for contact about a specific content issue. If you do not receive a timely reply, send an email to Don Meuten or call. The website is interactive and we encourage our colleagues to read, suggest edits, and provide references we missed, and link to our Facebook page.  

Soon we will create a biobank of diagnostic oncological pathology, housing histological and cytological slides, digital images of tumors and associated clinical records. The biobank and an operational protocol are being developed, it will be housed at The Joint Pathology Center, Silver Spring Maryland and a link will appear on this website. 

To compare and merge data between labs, and to apply that information to patient care requires that pathologists use the same methods when evaluating tumors. Using detailed, standardized methods should increase consistency, and likely, reproducibility. However, variability of the measured parameter will persist and there is no guarantee that improving the consistency of how a parameter is assessed will improve its predictability. Any parameter measured, whether it is a quantitative assay such as serum calcium concentration, or a subjectively determined parameter such as mitotic count, has variability. In clinical pathology this variability can be determined statistically and is reported as the coefficient of variation (CV). The CV varies with the concentration and with the methodology used to measure the analyte. Similarly, we expect variable CV for subjective parameters used for tumor assessment, even when standardized guidelines are followed. The problem is we do not know the CV for tumor parameters, nor do we know the significance of this variability relative to tumor prognosis. Is the prediction of the patient’s outcome affected by interpathologist variability? If the mitotic count is within one mitotic figure of the next grade, does that really matter? How much variability is acceptable depends on the element examined, how it correlates with outcome and how that information is applied clinically. But if we use different methods the variability will be unpredictable, results cannot be compared and should not be applied to clinical cases. The first step is to at least use the same methods when assessing tumors. Research should continue to develop new methods and hopefully compare the new method to existing methods so we can compare the utility of both. Using standardized methods, and determining their variability are the first steps but they do not guarantee it will increase the predictability of the method. Presently we cannot assess how useful some parameters are because the data is based on relatively small numbers of cases and the methods used between studies were not consistent. Furthermore, we are only evaluating one half of the cancer team – the tumor. We can assess a tumor microscopically, so we do it. But the host’s defenses or lack of may be equally or more important than the tumor’s biology or the parameters we use to evaluate the tumor. Future tests need to profile the host as well as the tumor. 

To advance our understanding of tumor behavior and enable more effective cancer therapy, we must standardize our approach to tumor diagnosis and grading and pool our resources and collective knowledge. This website will provide a forum to accomplish these goals, particularly with the insights and contributions from all of you. Standardization of tumor assessment and reporting these parameters in a complete and easily readable format will help oncologists provide prognoses and select treatments for pets with aggressive cancers. 

Thank you for contributions, and we look forward to hearing from you. 


Don Meuten (don_meuten@ncsu.edu / (919)605-1691), Frances Moore (fmoore1977@centurytel.net), Taryn Donovan (Taryn.Donovan@amcny.org), Christof Bertram (Christof.Bertram@vetmeduni.ac.at), and Michael Dark (darkmich@ufl.edu)