Genomic tumor analysis provides clinical guidance for the management of diagnostically challenging cancers in dogs

Esther Chon Vidium Animal Health, Translational Genomics Research Institute, Scottsdale, AZ

Search for other papers by Esther Chon in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM
,
Guannan Wang Vidium Animal Health, Translational Genomics Research Institute, Scottsdale, AZ

Search for other papers by Guannan Wang in
Current site
Google Scholar
PubMed
Close
 PhD
,
Derick Whitley Vidium Animal Health, Translational Genomics Research Institute, Scottsdale, AZ

Search for other papers by Derick Whitley in
Current site
Google Scholar
PubMed
Close
 DVM, DACVP
,
Sharadha Sakthikumar Vidium Animal Health, Translational Genomics Research Institute, Scottsdale, AZ

Search for other papers by Sharadha Sakthikumar in
Current site
Google Scholar
PubMed
Close
 PhD
,
Manisha Warrier Vidium Animal Health, Translational Genomics Research Institute, Scottsdale, AZ

Search for other papers by Manisha Warrier in
Current site
Google Scholar
PubMed
Close
 MS
,
Shukmei Wong Vidium Animal Health, Translational Genomics Research Institute, Scottsdale, AZ

Search for other papers by Shukmei Wong in
Current site
Google Scholar
PubMed
Close
 MS
,
Natalie Duran Vidium Animal Health, Translational Genomics Research Institute, Scottsdale, AZ

Search for other papers by Natalie Duran in
Current site
Google Scholar
PubMed
Close
 MS
,
Jonathan Adkins Vidium Animal Health, Translational Genomics Research Institute, Scottsdale, AZ

Search for other papers by Jonathan Adkins in
Current site
Google Scholar
PubMed
Close
 BS
,
Martin Boateng Vidium Animal Health, Translational Genomics Research Institute, Scottsdale, AZ

Search for other papers by Martin Boateng in
Current site
Google Scholar
PubMed
Close
 BS
,
Zhanyang Zhu Vidium Animal Health, Translational Genomics Research Institute, Scottsdale, AZ

Search for other papers by Zhanyang Zhu in
Current site
Google Scholar
PubMed
Close
 PhD
,
Salvatore Facista Vidium Animal Health, Translational Genomics Research Institute, Scottsdale, AZ

Search for other papers by Salvatore Facista in
Current site
Google Scholar
PubMed
Close
 BS
,
David Haworth Vidium Animal Health, Translational Genomics Research Institute, Scottsdale, AZ

Search for other papers by David Haworth in
Current site
Google Scholar
PubMed
Close
 DVM, PhD
, and
William Hendricks Vidium Animal Health, Translational Genomics Research Institute, Scottsdale, AZ

Search for other papers by William Hendricks in
Current site
Google Scholar
PubMed
Close
 PhD

Click on author name to view affiliation information

Abstract

OBJECTIVE

To evaluate the diagnostic, prognostic, and therapeutic utility of a cancer genomic diagnostic assay (SearchLight DNA; Vidium Animal Health) for diagnostically ambiguous cancer cases.

ANIMALS

69 privately owned dogs with ambiguous cancer diagnoses and for which the genomic assay was performed.

PROCEDURES

Genomic assay reports generated between September 28, 2020, and July 31, 2022, for dogs with malignancy or suspected malignancy were reviewed to determine the assay’s clinical utility defined as providing diagnostic clarity, prognostic information, and/or therapeutic options.

RESULTS

Genomic analysis provided diagnostic clarity in 37 of 69 cases (54%; group 1) and therapeutic and/or prognostic information in 22 of the remaining 32 cases (69%; group 2) for which the diagnosis remained elusive. Overall, the genomic assay was clinically useful in 86% (59/69) of cases.

CLINICAL RELEVANCE

To our knowledge, this was the first study to evaluate the multifaceted clinical utility of a single cancer genomic test in veterinary medicine. Study findings supported the use of tumor genomic testing for dogs with cancer, particularly those that are diagnostically ambiguous and therefore inherently challenging to manage. This evidence-driven genomic assay provided diagnostic guidance, prognostic support, and therapeutic options for most patients with an unclear cancer diagnosis that would otherwise have an unsubstantiated clinical plan. Furthermore, 38% (26/69) of samples were easily obtained aspirates. Sample factors (sample type, percentage of tumor cells, and number of mutations) did not influence diagnostic yield. Our study demonstrated the value of genomic testing for the management of canine cancer.

Abstract

OBJECTIVE

To evaluate the diagnostic, prognostic, and therapeutic utility of a cancer genomic diagnostic assay (SearchLight DNA; Vidium Animal Health) for diagnostically ambiguous cancer cases.

ANIMALS

69 privately owned dogs with ambiguous cancer diagnoses and for which the genomic assay was performed.

PROCEDURES

Genomic assay reports generated between September 28, 2020, and July 31, 2022, for dogs with malignancy or suspected malignancy were reviewed to determine the assay’s clinical utility defined as providing diagnostic clarity, prognostic information, and/or therapeutic options.

RESULTS

Genomic analysis provided diagnostic clarity in 37 of 69 cases (54%; group 1) and therapeutic and/or prognostic information in 22 of the remaining 32 cases (69%; group 2) for which the diagnosis remained elusive. Overall, the genomic assay was clinically useful in 86% (59/69) of cases.

CLINICAL RELEVANCE

To our knowledge, this was the first study to evaluate the multifaceted clinical utility of a single cancer genomic test in veterinary medicine. Study findings supported the use of tumor genomic testing for dogs with cancer, particularly those that are diagnostically ambiguous and therefore inherently challenging to manage. This evidence-driven genomic assay provided diagnostic guidance, prognostic support, and therapeutic options for most patients with an unclear cancer diagnosis that would otherwise have an unsubstantiated clinical plan. Furthermore, 38% (26/69) of samples were easily obtained aspirates. Sample factors (sample type, percentage of tumor cells, and number of mutations) did not influence diagnostic yield. Our study demonstrated the value of genomic testing for the management of canine cancer.

Introduction

A substantial challenge facing veterinarians who manage cancer patients is an unclear diagnosis. Without a definitive diagnosis, a clear therapeutic path and prognosis are difficult to determine. A definitive cancer diagnosis generally starts with histopathology and/or cytopathology review of the suspected tumor or mass, and an accurate diagnosis is necessary to determine the appropriate treatment plan and prognosis. However, sometimes a pathologic diagnosis may not be possible due to a variety of factors such as inadequate tissue sampling, poor differentiation of neoplastic cells, or lack of distinguishing features in the sample. In such situations, additional diagnostic tests are often sought to obtain a clearer diagnosis, such as additional pathology reviews, special stains/immunohistochemistry/immunocytochemistry, PCR for antigen receptor rearrangement (PARR), flow cytometry, or additional tissue biopsy.112 Often, even with these additional tests, a diagnosis remains elusive, making case management challenging for the clinician by hindering development of a definitive treatment course and evading a proper prognosis. Therefore, additional methods are needed to overcome this diagnostic challenge.

One method of diagnosis utilizes genomics, with the inherent understanding that cancer is a genetic disease resulting from congenital or acquired mutations in oncogenes and tumor suppressor genes.13,14 These mutations are fundamental to the cancer hallmarks that lead to their clinical behavior.15 Our understanding of the role of genomic alterations in human cancer has exploded over the past few decades, enabling the inception of precision medicine, a paradigm shift to therapy directed at mutations and pathway dysregulation unique to cancers in individual patients, and to the development of genomic biomarkers to inform cancer diagnoses and treatment decisions.1618 Genomic diagnostics and molecular pathology are now commonly incorporated into the clinical care of human cancer patients. Numerous mutation-based biomarkers serving as pathognomonic mutations are even included in professional guidelines for diagnostic work-up along with many others that help provide diagnostic clarity. For example, the Association of Molecular Pathology (AMP), American Society of Clinical Oncology (ASCO), and College of American Pathologists (CAP) consensus guidelines19 on interpretation and reporting of mutations in human cancer testing highlight diagnostic mutations commonly utilized in cancer testing. Additionally, many such mutations are cataloged across numerous human clinical cancer databases such as, for example, Memorial Sloan Kettering Cancer Center’s OncoKB,20 which documents 22 genes with many mutations that are required for testing to aid in diagnosis along with 53 genes that support diagnosis, etc. Similarly, the body of knowledge surrounding the role of genomics in canine oncology is also quickly expanding, empowering veterinarians to use the same advanced technology to uncover a canine tumor’s mutations to elucidate the tumor’s behavior and potential sensitivity to drugs targeting those mutations.21 Additionally, the homology between the canine and human genomes22,23 as well as the shared mutation profiles between canine and human cancers2429 can be leveraged to optimize our insight into the tumor’s identity, helping to overcome the challenge we can face in diagnostically elusive cases.

Tumor profiling with genomic analysis therefore has the unique potential to provide diagnostic clarity while also guiding therapy and providing prognostic information. Tumor genomic profiling can be performed by several methods, such as via whole-genome sequencing, whole-exome sequencing, or targeted sequencing.25,3034 Targeting specific genes for sequencing allows for the utilization of small, cost-effective panels that are focused on mutations with high biomarker evidence levels. Targeted sequencing panels can also be performed with a relatively quick turnaround time, making them optimal for the fast-paced veterinary clinical environment. These targeted panels are particularly valuable for veterinary oncology since they favorably balance low cost with high clinical actionability, particularly when this genomic testing uses a reporting framework that incorporates structured, comprehensive, and peer-reviewed genomic biomarker data such as that based on the AMP/ASCO/CAP consensus guidelines19 closely followed and regulated in human cancer genomic diagnostics. A diagnostic genomic biomarker comprises a pathogenic mutation known to be enriched in (and sometimes pathognomonic for) a specific tumor type in dogs and/or humans. A prognostic genomic biomarker is a mutation associated with specific clinical outcomes in canine and/or human cancer. A therapeutic genomic biomarker is a mutation associated with response or resistance to a specific therapy in canine and/or human cancer. Each tumor’s mutations can be described in a concise, descriptive report for the veterinarian along with evidence statements to indicate their role as a biomarker while also presenting a qualitative review of genomic findings in the context of the patient’s pathology findings and clinical history.

Given the frustrating nature of ambiguous cancer diagnoses and potential for genomic testing to provide multiple avenues of assistance to the practicing veterinarian, the objective of this study was to evaluate the diagnostic, prognostic, and therapeutic utility of a clinical genomic diagnostic assay (SearchLight DNA; Vidium Animal Health) for diagnostically ambiguous cases. We hypothesized that this assay would bring clinically valuable information to the majority of cases within our cohort of diagnostically unclear tumors.

Materials and Methods

Samples

For this evaluation study, the Vidium Animal Health patient database was searched for records of client-owned dogs with clinical genomic reports from testing performed between September 28, 2020, and July 31, 2022, for malignancy or suspected malignancy. Dogs were included if their diagnosis was equivocal and their record contained a pathology report or written medical history that included any of the following words: poorly differentiated, anaplastic, probable, possible, suspect, suggestive, malignant neoplasm, round-cell tumor, and atypical. A diagnosis was considered equivocal if the diagnosis did not include a specific histotype (eg, lymphoma, histiocytic sarcoma, or squamous cell carcinoma), and/or there was at least 1 other differential diagnosis provided by the pathologist. Dogs with malignancy highly suspected (eg, dogs with pulmonary lesions with history of multiple different tumor histologies) were included, regardless of whether a formal pathology review was performed. Dogs were excluded if their diagnosis was unequivocal (ie, the diagnosis was a specific histotype).

Clinical cancer genomic diagnostic assay

A pan-cancer, tumor genomic sequencing panel was used as previously described (Figure 1).35 Briefly, fine-needle aspirate (FNA) samples or biopsy specimens (fresh frozen, formalin-fixed or formalin-fixed paraffin-embedded [FFPE] blocks, scrolls, or slides) of suspected tumors were submitted by veterinarians directly or by the veterinarians’ reference laboratories (at –80°C overnight for fresh frozen samples and at room temperature with overnight or 2-day shipping for all other samples) to Vidium Animal Health. Once received, samples were grossly processed and embedded if not already performed (for formalin-fixed or fresh frozen tissues), tumor content was determined, and DNA was extracted as previously described within 2 to 3 days of receipt. Samples were then verified as having passed quality metrics and then underwent sequencing of targeted genomic regions enriched using a proprietary custom panel of hybridization-based capture probes to evaluate the presence of multiple mutation types, including single-nucleotide variants (SNVs), copy number variants (CNVs), and internal tandem duplications (ITDs) in the carefully selected 120 cancer genes (1,358 exonic regions and 429 exon-proximal regions of the genome across 11,554 probes targeting 482.3 kbp of sequence space) selected on the basis of our prior studies,36,37 content of human cancer gene panels, and curation of canine and human cancer genomic biomarker literature (Appendix).

Figure 1
Figure 1

Schematic of steps taken for the use of a genomic assay (SearchLight DNA; Vidium Animal Health) in identifying and annotating mutations occurring within a tumor. This genomic assay is a tumor-only, next-generation sequencing canine gene panel covering 120 genes associated with canine or human cancer. Mutation types identified include single-nucleotide variants (SNV), copy number variants (CNV), and internal tandem duplications (ITD). Once identified, these mutations then undergo bioinformatic analysis and annotation as a biomarker of diagnosis, prognosis, and/or therapy based on published literature curated and stored within a proprietary knowledge base (Vidium Insight; Vidium Animal Health). Annotation of these mutations provides meaning to each mutation for clinical application by the veterinarian.

Citation: Journal of the American Veterinary Medical Association 261, 5; 10.2460/javma.22.11.0489

Briefly, the laboratory workflow includes genomic DNA extraction, measurement of DNA quantity and quality, sample pooling and construction of sequencing libraries utilizing the hybrid-capture probes described above, assessment of library quantity and quality, and paired-end sequencing on Illumina sequencing instruments in a dedicated veterinary genomics reference laboratory operating in a Good Laboratory Practices–like environment modeled on that of human cancer genomic diagnostic laboratories. Bioinformatic analysis, filtering, and annotation (Supplementary Figure S1) includes the use of DNA sequencing variant identification algorithms (variant callers) with high reported performance in human data and that we also benchmarked in canine genomic data. Raw sequence data were demultiplexed, FASTQs generated and quality controlled, and then aligned to the canine reference genome, CanFam3.147. Consensus SNV calls from the variant callers Mutect2 (GATK version 4.1.4.0; Broad Institute) and Pisces (version 5.2.5.20; Illumina Inc) were identified for further functional annotation. We used Manta (version 1.6; Illumina Inc) and CNVkit (version 0.9.6) for ITD and CNV variant calls, respectively. A pool of normals comprising 14 confirmed non–cancer-bearing tissues from diverse anatomic sites and breeds was utilized for tumor-only copy number calling with CNVkit. The primary analysis pipeline was automated, tested, and run on the DNAnexus cloud-based computing platform (DNAnexus Inc). Candidate pathogenic variants occurring at variant allele frequencies of ≥ 5% were annotated according to predicted impact and filtered to remove common, likely benign single-nucleotide polymorphisms from the European Variant Archive and internal data with population allele frequencies ≥ 1% from assessment of ≥ 10 dogs. CNV thresholds were determined empirically utilizing samples with known CNVs identified by orthogonal genomic techniques. For ITD calling with Manta, only ITDs identified in KIT and FLT3 were called. These highly filtered mutations were then used to query a comprehensive, proprietary genomic biomarker database (Vidium Insight; Vidium Animal Health) comprising canine and human genomic biomarker associations from clinical guidelines and peer-reviewed literature. A diagnostic biomarker was defined as a mutation found to be enriched in a specific tumor type, predicted to be pathogenic, and suspected to contribute to cancer development (ie, functional data support their role in cancer), alone or in combination. A prognostic biomarker was defined as a mutation associated with outcome prediction, and a therapeutic biomarker was defined as a mutation associated with response to a specific therapy. The evidence level for each biomarker designation was stated in each report and was based on published guidelines19 that are robustly used in human cancer genomic diagnostic assessments. Mutations with biomarker associations populated clinical genomic reports for each case along with supporting evidence statements, while those without biomarker associations populated reports as Variants of Uncertain Significance.

This laboratory test and bioinformatic analysis workflow have been analytically validated as previously described,35 utilizing principles for validation of clinical genomic diagnostic assays and analysis methods.38,39 Although no canine reference standards exist, we utilized dilution studies, replicate analyses, and orthogonal genomic characterization techniques with control samples to define analytical sensitivity, specificity, and reproducibility. Performance metrics included 95.2% and 100% sensitivity for detection of SNVs and ITDs, respectively, at variant allele fractions as low as 7.5%; 99.9% and 100% specificity for detection of SNVs and ITDs, respectively; 98.8% and 88% concordance across sequencing runs, instruments, and operators for SNV/ITD and CNV detection, respectively; and high performance across canine tumor and tissue types (with high performance being defined as having a median of > 250X sequence coverage with > 90% of target bases at > 100X coverage in 6 FNAs and 15 FFPEs in a validation cohort and with 88% concordance in variants identified between an FFPE tumor and a cell line derived from the same tumor).

For each dog in our study, the clinical genomic report included all mutations tabulated with accompanying evidence statements to indicate their role as a biomarker of diagnosis, prognosis, or therapy, alone or in combination. A qualitative analysis of the genomic findings in the context of the patient’s pathology findings and clinical history was also performed by genomic scientists (WH, GW, and SS) and included as an overview and interpretation of the mutations identified. Independent of diagnostic insight, prognostic biomarkers as well as therapeutic biomarkers with matching targeted therapies were also described in the same report.

Data collection and evaluation

Sample-specific information (sample type, tumor content, and number of mutations within each sample), patient demographics (age, gender, and breed), and previously performed diagnostics were gathered from each report. To evaluate diagnostic utility, each sample’s qualitative analysis was reviewed to determine assignment of dogs to group 1 (diagnostic clarity achieved with genomic analysis) versus group 2 (no diagnostic clarity achieved with genomic analysis). Group 1 included dogs for which genomic analysis was able to provide diagnostic clarity. Diagnostic clarity was achieved if either a potential underlying histology (eg, a malignant round-cell neoplasia being clarified as a lymphoma) or histologic class (eg, a malignant neoplasia being clarified as a sarcoma) was provided or if a tumor could be classified as either likely benign versus malignant (eg, histiocytic neoplasm being clarified as malignant, not inflammatory). The process by which a mutation (or a group of mutations) provided diagnostic clarity involved the mutation triggering a strong diagnostic biomarker association from the comprehensive genomic biomarker database (Vidium Insight; Vidium Animal Health) with a specific tumor type (ie, the mutation was predicted to be pathogenic and has been previously identified in a specific human and/or canine cancer); the bioinformatics pipeline capturing this evidence; and the genomic scientists (WH, GW, and SS) reporting it in the qualitative analysis. Group 2 included dogs for which genomic analysis could not provide diagnostic clarity, either because it could not clarify a potential underlying histologic characteristic or it could not further subclassify the tumor histologic characteristics beyond that already broadly diagnosed as a carcinoma, sarcoma, or round-cell tumor.

Alongside the qualitative diagnostic analysis, prognostic and therapeutic biomarkers were also evaluated to determine overall clinical utility. For this specific evaluation, the genomic assay was considered to have diagnostic utility if it provided diagnostic clarity based on qualitative analysis, prognostic utility if it provided canine-specific prognostic information, and therapeutic utility if it identified a mutation’s sensitivity to a targeted therapeutic that was available in at least 1 major US compounding pharmacy (ie, available to veterinarians).

Statistical analysis

Statistical analyses were performed to evaluate whether sample-specific factors (sample type, tumor content, and number of mutated genes) could affect diagnostic yield (and therefore placement into group 1 vs group 2). For continuous variables (tumor content and number of mutated genes), the Mann-Whitney test was performed. For sample type (categorical variables), the χ2 test for trend was performed. Where applicable, 95% CIs of the percentages were calculated using an online calculator (Sample Size Calculators for designing clinical research).40 For differences between groups, a value of P < .05 was considered significant. Statistical analysis was performed with standard software (Prism version 9.4.1; GraphPad Software).41

Results

Of the 358 reports reviewed, 69 met the inclusion criteria and were evaluated for this study. A total of 91 mutations in 60 genes were identified within this cohort (Supplementary Table S1). Each of the 69 reports was for 1 dog. Of these 69 dogs, 37 (54%) were assigned to group 1 and 32 (46%) were assigned to group 2 (Table 1). Mean age was 9.1 years (range, 1 to 15 years) for all dogs, 9 years (range, 1 to 15 years) for group 1, and 9.2 years (range, 2 to 14 years) for group 2. Of the 69 cases, 26 (38%) had samples submitted as aspirates and 32 (46%) had prior diagnostic tests performed (Table 2). Tumor content (U = 286; P = .3421), number of mutations (U = 508.5; P = .316), and sample type (χ2 [1, n = 69] = 0.1492; P = .6993) were not significantly different between the groups.

Table 1

Sex and breeds of 69 dogs for which genomic analysis (SearchLight DNA; Vidium Animal Health) performed on fine-needle aspirate or biopsy samples of suspected malignancies between September 28, 2020, and July 31, 2022, provided diagnostic clarity (group 1; n = 37) versus those for which diagnosis remained elusive (group 2; 32).

Variable All Group 1 Group 2
No. of dogs No. of dogs No. of dogs
Sex
 Female spayed 38 23 15
 Female intact 1 1 0
 Male castrated 28 12 16
 Male intact 2 1 1
Breed
 Mixed breed 22 13 9
 Golden Retriever 7 6 1
 Labrador Retriever 4 0 4
 Beagle 3 3 0
 Pembroke Welsh Corgi 3 1 2
 Unknown breed 3 3 0
 Basset Hound 1 1 0
 Catahoula Leopard Dog 2 0 2
 German Shepherd Dog 2 0 2
 Miniature Schnauzer 2 1 1
 Siberian Husky 2 2 0
 Vizsla 2 2 0
 Australian Cattle Dog 1 1 0
 Belgian Malinois 1 0 1
 Boxer 1 0 1
 Cardigan Welsh Corgi 1 1 0
 Cocker Spaniel 1 0 1
 Dachshund 1 0 1
 Doberman Pinscher 1 0 1
 English Setter 1 0 1
 German Shorthaired Pointer 1 1 0
 Greyhound 1 0 1
 Maltese 1 1 0
 Poodle (Miniature) 1 0 1
 Poodle (Standard) 1 0 1
 Samoyed 1 1 0
 Shih Tzu 1 0 1
 Yorkshire Terrier 1 0 1
Table 2

Characteristics of samples submitted for the genomic analysis described in Table 1 for all dogs and stratified by dogs in group 1 versus group 2.

Variable All Group 1 Group 2
Submitted sample type
 Total aspirates 26 (38) 13 (19) 13 (19)
 Total biopsies 43 (62) 24 (35) 19 (27)
  Blocks 1 (1.5) 1 (1.5) 0 (0)
  Blocks (MD) 1 (1.5) 1 (1.5) 0 (0)
  Formalin-fixed (MD) 2 (3) 1 (1.5) 1 (1.5)
  Fresh frozen 2 (3) 2 (3) 0 (0)
  Scrolls 22 (32) 15 (22) 7 (10)
  Slides 8 (12) 2 (3) 6 (8.5)
  Slides (MD) 7 (10) 2 (3) 5 (7)
Mutations per sample
 Median (range) 4 (0–28) 5 (0–26) 3.5 (0–28)
 Mean 5.9 6.6 5.1
Tumor content (as percent of all cells) per sample
 Median (range) 80 (20–98) 77.5 (20–95) 60 (20–98)
 Mean 75.3 74.6 75.9
Prior diagnostic tests performed 32 (46) 20 (29) 12 (17)
 Pathology additional opinion only 7 (10) 4 (6) 3 (4)
 IHC or ICC only 12 (17) 7 (10) 5 (7)
 Pathology additional opinion and IHC 10 (14.5) 6 (8.5) 4 (6)
 Pathology additional opinion, IHC, and PARR 2 (3) 2 (3) 0 (0)
 PARR, FC, and IHC 1 (1.5) 1 (1.5) 0 (0)

Data are reported as number and percentage, unless otherwise indicated.

FC = Flow cytometry. ICC = Immunocytochemistry. IHC = Immunohistochemistry. MD = Macrodissection. PARR = PCR for antigen receptor rearrangement.

The genomic assay was diagnostically supportive of either a broad histologic class or specific histology in 45% (31/69) of patients, and it was able to differentiate malignant versus likely benign neoplasia in 9% (6/69) of patients (group 1; Table 3; Supplementary Table S2). In every case for which there were identified therapeutic biomarker associations, treatment options were provided on the basis of peer-reviewed publications with high evidence levels (predominantly Tier 1)19 supporting matches between the specific mutations and the targeted therapeutics. Likewise, every identified prognostic biomarker association between a specific mutation and important clinical event (such as disease progression or patient survival) was also based on published literature with supporting evidence levels.19 For group 2 cases in which diagnostic clarity could not be provided, the genomic assay was able to provide treatment options alone for 2 patients, prognostic information alone for 12 patients, and treatment with prognostic information in 8 patients, totaling 22 patients (69% [22/32]) for which therapeutic and/or prognostic information was provided, even in the face of no further diagnostic clarity.

Table 3

Distribution (number, percentage, and 95% CI of the percentage) of diagnoses for the suspected malignancies sampled for the 37 dogs in group 1 (dogs for which genomic analysis provided diagnostic clarity) described in Table 1.

Diagnosis from genomic data No. of dogs Percentage (95% CI)
Malignant vs likely benign 6 8.70 (3.26–17.97)
 Likely benign 2 2.90 (0.35–10.08)
 Malignant 4 5.80 (1.60–14.18)
Histologic diagnosis 31 44.93 (32.92–57.38)
 Adrenocortical carcinoma 1 1.45 (0.04–7.81)
 Cholangiocarcinoma 1 1.45 (0.04–7.81)
 Pulmonary adenocarcinoma 1 1.45 (0.04–7.81)
 Squamous cell carcinoma 1 1.45 (0.04–7.81)
 Carcinoma 2 2.90 (0.35–10.08)
 Hemangiosarcoma 1 1.45 (0.04–7.81)
 Osteosarcoma 1 1.45 (0.04–7.81)
 Histiocytic sarcoma 2 2.90 (0.35–10.08)
 Langerhans cell histiocytosis 1 1.45 (0.04–7.81)
 Melanoma 5 7.25 (2.39–16.11)
 Sarcoma 5 7.25 (2.39–16.11)
 Mesothelioma 1 1.45 (0.04–7.81)
 Lymphoma 8 11.59 (5.14–21.57)
 Hematopoietic cancer 1 1.45 (0.04–7.81)
Total 37 53.62 (41.20–65.72)

Two cases did not have prior formal pathology performed although malignancy was highly suspected by the clinician. One patient had progressively enlarging pulmonary nodules after a history of splenic hemangiosarcoma, a heart-based mass, and oral melanoma; for this patient, the genomic assay was able to narrow the diagnosis of pulmonary lesions to melanoma. The second patient had multiple progressively enlarging liver nodules after a prior history of adrenal carcinoma. For this patient, the genomic assay was able to narrow the diagnosis of the liver lesions to adrenocortical carcinoma.

The genomic assay was able to provide clinical utility in 86% (59/69) of all patients, either diagnostically, prognostically, therapeutically, or any combination thereof (Figure 2). The greatest number of mutated genes were identified as diagnostic biomarkers, followed by therapeutic and prognostic biomarkers, respectively (Figure 3). A gene’s biomarker categorization was not mutually exclusive, meaning that 1 gene could be designated as a biomarker of 1, 2, or all 3 categories.

Figure 2
Figure 2

Histograms of the clinical utility (provision of information for the diagnosis [Dx], prognosis [Px], or therapeutic options [Tx], alone or in combination) of the genomic assay described in Figure 1 when used to evaluate fine-needle aspirate or biopsy samples of suspected malignancies in 69 dogs for which assay reports were generated between September 28, 2020, and July 31, 2022, showing all dogs grouped together (total; dark blue; n = 69) and on the basis of whether assay results provided the Dx (group 1; light gray; 37) versus did not provide the Dx (group 2; light blue; 23). The number above each bar represents the number of dogs; the percentages shown are of the 69 dogs for which the assay results had the respective clinical utility. Categories are mutually exclusive, with results for each dog appearing only once (not repeated between categories). The combined clinical utility of all categories totals 86% (59/69).

Citation: Journal of the American Veterinary Medical Association 261, 5; 10.2460/javma.22.11.0489

Figure 3
Figure 3

Mutated canine genes identified by the genomic assay described in Figure 1, grouped according to clinical utility in the study described in Figure 2, with the number of mutated genes circled above the respective bar. Genes could bear anywhere between 1 and 18 different mutations (eg, TP53 bore 18 different mutations, such as copy number loss, p.Arg226His, p.Tyr214fs, splice site mutation, etc) but were only counted once for this tally, regardless of the number of mutations identified for any 1 gene.

Citation: Journal of the American Veterinary Medical Association 261, 5; 10.2460/javma.22.11.0489

Discussion

This study provided quantitative evidence of the high utility of a genomic diagnostic assay that brings digestible, clinically applicable, and heretofore unavailable information to the veterinary health team. Patients were included in this study because of an unclear cancer diagnosis, with nearly half (46% [32/69]) of these patients already having had at least 1 other prior diagnostic test performed with no further clarity on diagnosis. The lack of a diagnosis forces the clinician to treat the patient empirically and limits the ability to estimate prognosis. In this study, genomic analysis was able to provide diagnostic clarity in over half of the patients (54% [37/69]; group 1). Even for patients in which the diagnosis remained elusive (group 2), genomic analysis provided prognostic and/or therapeutic information in the majority (69% [22/32]) of those patients. Overall, tumor genomic analysis provided diagnostic clarity, prognostic information, and/or available targeted therapeutic option(s) for most (86% [59/69]) patients in this cohort of diagnostically challenging cases.

Genomic diagnostics is an emerging field in veterinary medicine. To the best of the authors’ knowledge, this genomic assay is the first of its kind in veterinary medicine to provide diagnostic, prognostic, and therapeutic information in a single report based on structured biomarker data. The assay is fundamentally modeled after human FDA-approved tissue-based companion diagnostics, such as FoundationOne CDx (Foundation Medicine, Inc), that are clinically and analytically validated for all solid tumors. The biomarker associations for mutations identified in each sample are based on guidelines and standards for the interpretation and reporting of sequence variants in cancer from a joint consensus recommendation of the AMP, ASCO, and CAP.19 This veterinary genomic assay is therefore holding itself to a high standard to provide veterinarians with clinically useful information based on an individual patient’s cancer genomic profile.

Since a list of genomic mutations, even with biomarker annotations, can be challenging to translate into clinical actionability, a qualitative analysis is routinely performed and communicated to the veterinarian for each sample. This process is similar to what is done in pathology reports, in which a trained expert integrates the microscopic observations with the clinical history. This analysis includes a succinct overview and interpretation of the mutations for clinical application. Particularly for this cohort of patients with ambiguous diagnoses, a qualitative analysis is relevant due to the differing degrees of diagnostic specificity among mutations, necessitating an assessment of an aggregate of mutations (ie, the presence, absence, and combination of various mutations) within the sample’s mutational milieu in light of the tumor location, pathology reports, and clinical history to support a diagnosis of cancer and/or a specific cancer type. Accordingly, the qualitative analysis for each case was specifically reviewed in this study for the determination of whether diagnostic clarity was achieved. This analysis, regardless of cancer diagnosis, is especially valuable for the veterinarian as genomics becomes increasingly more utilized within veterinary medicine, particularly for these diagnostically challenging cases.

Importantly, mutations identified in each case can be supportive of, but are rarely pathognomonic for, a specific cancer type, although some pathognomonic mutations do exist. The basis of such support is directly derived from peer-reviewed publications, with attention to the evidence levels underpinning their association with the specific cancer type (diagnosis) as well as with prognosis and/or therapy. All of these biomarker associations curated from peer-reviewed literature are stored within a routinely updated proprietary database. When mutations are identified in a tumor sample, any diagnostic, prognostic, and therapeutic associations with that mutation are automatically extracted and reported. After annotation, a qualitative analysis of reported biomarkers is performed by our genomic scientists (WH, GW, and SS) in consultation with a board-certified oncologist (EC) and board-certified pathologist (DW) to contextualize the genomic findings in the framework of the key clinical questions. This qualitative analysis occurs for all cases, including those that are considered likely benign, which are diagnosed as such on the basis of the absence of mutations commonly found in malignant lesions (eg, inactivating mutations in tumor suppressor genes such as TP53 or CDKN2B) in conjunction with pathology data and any other additional diagnostic tests (eg, immunohistochemistry). As new genomic information is published and curated into the genomic database, these genomic analyses will continue to evolve alongside newly discovered mutations and novel biomarker associations.

Sample-specific factors (sample type, tumor content, and number of mutations identified within the sample) were not significantly different between the group with diagnostic clarity (group 1) and the group with no diagnostic clarity (group 2). These analyses were performed to evaluate whether sample-related factors could influence diagnostic yield. While the lack of a significant difference between groups could have been due to an overall small sample size, the finding that preanalytical sample factors (tumor content and sample type) did not differ between groups could be a reflection of the rigor of this genomic assay’s sequencing criteria, including the requirement of all samples to have met a validated minimum threshold of tumor content as well as DNA quantity and quality metrics before they were sequenced. Furthermore, since sample type did not appear to make a difference in diagnostic outcome, an argument can be made to more strongly encourage submission of FNA samples for genomic analysis, due to the ease, lower cost, and lower morbidity associated with FNAs, compared to tissue biopsies, which are contrastingly more expensive, require some form of anesthesia to perform, and require tissue healing after the biopsy procedure. While histopathology of a biopsy is generally the next diagnostic step if cytopathology from an FNA proves inconclusive, some patients may not be suitable candidates for a biopsy procedure due to patient comorbidities and/or owners’ constraints or morbidities associated with biopsy procedures. Furthermore, most cases (62% [43/69]) within this study cohort had histopathology performed as their initial diagnostic test and still did not have a clear diagnosis. Therefore, it is reasonable to consider FNAs for genomic analysis, whether the clinical aim is for diagnostic support, prognostic guidance, or targeted therapeutic options. Clinicians can get the same high-quality genomic information from a relatively easy and inexpensive FNA, which can be considered for all tumor samples that can be aspirated.

The clinical utility of this genomic assay for this cohort was high at 86% (59/69), even with the stringent definitions of utility unique to this study. As with the criteria used to define diagnostic clarity, the criteria to evaluate prognostic and therapeutic utility were narrow for this calculation. Only canine-specific prognostic information was included (compared to prognostic information for both dogs and humans, which is included in the report), and only therapeutic associations of sensitivity (not resistance) to targeted therapies currently available to the veterinarian were considered (compared to all targeted therapies that meet the stringent therapeutic biomarker criteria for the report but that may not be available yet to veterinarians). These strict definitions were used to reflect information that could be more practically applied to canine patients by the veterinarian. Therefore, the 86% (59/69) clinical utility calculated in this study was an underestimation of the larger capacity of this genomic assay’s clinical utility that uses broader definitions to include therapeutics that are expected to be available to veterinarians in the future, and that uses human-specific prognostic information to stand in the gap where there may not yet be such associations in the canine literature.

While this genomic assay was able to guide the diagnosis in over half (54% [37/69]; group 1) of diagnostically equivocal cancer cases, we do not recommend that genomics wholly replace pathology for the diagnosis of cancer. In fact, all cases submitted for genomic analysis during the evaluation time frame had prior pathology (histology or cytology) performed, and most (81% [289/358]) had an unequivocal, definitive diagnosis from pathology, with or without additional diagnostics such as immunohistochemistry or PARR. However, 19% (69/358), of these cases (ie, all the cases reviewed for the present study) had an elusive diagnosis, even with conventional diagnostics. In this specific cohort, nearly half of the cases had additional diagnostics performed with no conclusive diagnosis, and genomic analysis was pursued for additional information for all cases. Genomic analysis is a reasonable next step in these situations. This genomic assay is undergirded by published data to support an enrichment of certain mutations in certain cancers and is therefore able to provide diagnostic guidance if those mutations are present in the sample in question. Furthermore, in more than 90% of cases with a definitive pathology diagnosis (manuscript under preparation), this genomic assay corroborated the diagnosis based on published mutations identified in the specific canine cancers or parallel human cancers, lending support to the reliability of this assay in providing diagnostic guidance when other conventional diagnostics have failed to give conclusive results.

Our goal with the present study was to determine whether and to what extent this genomic assay was able to provide diagnostic clarity, therapeutic options, and prognostic information in all patients with ambiguous diagnoses. However, clinical outcomes data could provide further support for the prognostic and therapeutic information provided on the genomic report. While collection of such data was outside the scope of this study, we understand the value of this information for the clinician and we are actively collecting this data for separate evaluation. Another limitation of this study was its retrospective nature and any associated biases that such studies can bring. Finally, while mutation status at the genomic level is crucial to uncover, especially vis-à-vis therapeutics that target such mutations, gene expression correlates could have added further support to this study’s findings.

As veterinarians seek to continuously optimize their efforts in clinical practice, it becomes increasingly more challenging to digest complex information, such as that provided by genomic diagnostics. Having a succinct, digestible report is therefore imperative, especially for cases that are already inherently difficult to manage, such as those with ambiguous diagnoses. The diagnostic test described in this study is unique in providing such a report of genomic findings to the clinician that incorporates up-to-date published literature to support clinically relevant biomarker associations to answer pertinent questions for the care of cancer patients. As shown in this study, genomic testing could also fulfill a need in veterinary oncology by providing useful diagnostic, prognostic, and therapeutic information where there would otherwise be none for diagnostically ambiguous cases.

Supplementary Materials

Supplementary materials are posted online at the journal website: avmajournals.avma.org

Acknowledgments

No third-party funding or support was received in connection with this study or the writing or publication of the manuscript.

The authors are employees of Vidium Animal Health, a subsidiary of the Translational Genomics Research Institute. SearchLight DNA is a product developed and provided by Vidium Animal Health.

References

  • 1.

    Fernandez NJ, West KH, Jackson ML, Kidney BA. Immunohistochemical and histochemical stains for differentiating canine cutaneous round cell tumors. Vet Pathol. 2005;42(4):437-445. doi:10.1354/vp.42-4-437

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2.

    Millanta F, Impellizeri J, McSherry L, Rocchigiani G, Aurisicchio L, Lubas G. Overexpression of HER-2 via immunohistochemistry in canine urinary bladder transitional cell carcinoma - a marker of malignancy and possible therapeutic target. Vet Comp Oncol. 2018;16(2):297-300. doi:10.1111/vco.12345

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3.

    Smedley RC, Lamoureux J, Sledge DG, Kiupel M. Immunohistochemical diagnosis of canine oral amelanotic melanocytic neoplasms. Vet Pathol. 2011;48(1):32-40. doi:10.1177/0300985810387447

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4.

    Barger A, Graca R, Bailey K, et al. Use of alkaline phosphatase staining to differentiate canine osteosarcoma from other vimentin-positive tumors. Vet Pathol. 2005;42(2):161-165. doi:10.1354/vp.42-2-161

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5.

    Priest HL, Hume KR, Killick D, et al. The use, publication and future directions of immunocytochemistry in veterinary medicine: a consensus of the Oncology-Pathology Working Group. Vet Comp Oncol. 2017;15(3):868-880. doi:10.1111/vco.12228

    • Search Google Scholar
    • Export Citation
  • 6.

    Masserdotti C. Proportion of mast cells in normal canine hepatic cytologic specimens: comparison of 2 staining methods. Vet Clin Pathol. 2013;42(4):522-525. doi:10.1111/vcp.12086

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7.

    Riondato F, Comazzi S. Flow cytometry in the diagnosis of canine B-cell lymphoma. Front Vet Sci. 2021;8:600986. doi:10.3389/fvets.2021.600986

  • 8.

    Comazzi S, Riondato F. Flow cytometry in the diagnosis of canine T-cell lymphoma. Front Vet Sci. 2021;8:600963. doi:10.3389/fvets.2021.600963

  • 9.

    Waugh EM, Gallagher A, Haining H, et al. Optimisation and validation of a PCR for antigen receptor rearrangement (PARR) assay to detect clonality in canine lymphoid malignancies. Vet Immunol Immunopathol. 2016;182:115-124. doi:10.1016/j.vetimm.2016.10.008

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10.

    Ehrhart EJ, Wong S, Richter K, et al. Polymerase chain reaction for antigen receptor rearrangement: benchmarking performance of a lymphoid clonality assay in diverse canine sample types. J Vet Intern Med. 2019;33(3):1392-1402. doi:10.1111/jvim.15485

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11.

    Laliberté SM, Poirier VJ, Pinard CJ, Hocker SE, Foster RA. A retrospective comparison of first and second opinion histopathology with patient outcomes in veterinary oncology cases (2011-2019). Vet Comp Oncol. 2022;20(1):198-206. doi:10.1111/vco.12762

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12.

    Regan RC, Rassnick KM, Malone EK, McDonough SP. A prospective evaluation of the impact of second-opinion histopathology on diagnostic testing, cost and treatment in dogs and cats with cancer. Vet Comp Oncol. 2015;13(2):106-116. doi:10.1111/vco.12023

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13.

    Stratton MR, Campbell PJ, Futreal PA. The cancer genome. Nature. 2009;458(7239):719-724. doi:10.1038/nature07943

  • 14.

    Garraway LA, Lander ES. Lessons from the cancer genome. Cell. 2013;153(1):17-37. doi:10.1016/j.cell.2013.03.002

  • 15.

    Hanahan D. Hallmarks of cancer: new dimensions. Cancer Discov. 2022;12(1):31-46. doi:10.1158/2159-8290.CD-21-1059

  • 16.

    Brittain HK, Scott R, Thomas E. The rise of the genome and personalised medicine. Clin Med (Lond). 2017;17(6):545-551. doi:10.7861/clinmedicine.17-6-545

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17.

    Berger MF, Mardis ER. The emerging clinical relevance of genomics in cancer medicine. Nat Rev Clin Oncol. 2018;15(6):353-365. doi:10.1038/s41571-018-0002-6

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 18.

    Tsimberidou AM, Fountzilas E, Nikanjam M, Kurzrock R. Review of precision cancer medicine: evolution of the treatment paradigm. Cancer Treat Rev. 2020;86:102019. doi:10.1016/j.ctrv.2020.102019

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19.

    Li MM, Datto M, Duncavage EJ, et al. Standards and guidelines for the interpretation and reporting of sequence variants in cancer: a joint consensus recommendation of the Association for Molecular Pathology, American Society of Clinical Oncology, and College of American Pathologists. J Mol Diagn. 2017;19(1):4-23. doi:10.1016/j.jmoldx.2016.10.002

    • Search Google Scholar
    • Export Citation
  • 20.

    Chakravarty D, Gao J, Phillips SM, et al. OncoKB: a precision oncology knowledge base. JCO Precis Oncol. 2017;2017:PO.17.00011. doi:10.1200/PO.17.00011

  • 21.

    Ostrander EA, Dreger DL, Evans JM. Canine cancer genomics: lessons for canine and human health. Annu Rev Anim Biosci. 2019;7(1):449-472. doi:10.1146/annurev-animal-030117-014523

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22.

    Shearin AL, Ostrander EA. Leading the way: canine models of genomics and disease. Dis Model Mech. 2010;3(1-2):27-34. doi:10.1242/dmm.004358

  • 23.

    Ostrander EA. Franklin H. Epstein lecture. Both ends of the leash-the human links to good dogs with bad genes. N Engl J Med. 2012;367(7):636-646. doi:10.1056/NEJMra1204453

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24.

    Simpson RM, Bastian BC, Michael HT, et al. Sporadic naturally occurring melanoma in dogs as a preclinical model for human melanoma. Pigment Cell Melanoma Res. 2014;27(1):37-47. doi:10.1111/pcmr.12185

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 25.

    Elvers I, Turner-Maier J, Swofford R, et al. Exome sequencing of lymphomas from three dog breeds reveals somatic mutation patterns reflecting genetic background. Genome Res. 2015;25(11):1634-1645. doi:10.1101/gr.194449.115

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26.

    Abdelmegeed SM, Mohammed S. Canine mammary tumors as a model for human disease. Oncol Lett. 2018;15(6):8195-8205. doi:10.3892/ol.2018.8411

    • Search Google Scholar
    • Export Citation
  • 27.

    Megquier K, Turner-Maier J, Swofford R, et al. Comparative genomics reveals shared mutational landscape in canine hemangiosarcoma and human angiosarcoma. Mol Cancer Res. 2019;17(12):2410-2421. doi:10.1158/1541-7786.MCR-19-0221

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28.

    Wang G, Wu M, Durham AC, et al. Molecular subtypes in canine hemangiosarcoma reveal similarities with human angiosarcoma. PLoS One. 2020;15(3):e0229728. doi:10.1371/journal.pone.0229728

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29.

    Schiffman JD, Breen M. Comparative oncology: what dogs and other species can teach us about humans with cancer. Philos Trans R Soc Lond B Biol Sci. 2015;370(1673):20140231. doi:10.1098/rstb.2014.0231

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 30.

    Kim TM, Yang IS, Seung BJ, et al. Cross-species oncogenic signatures of breast cancer in canine mammary tumors. Nat Commun. 2020;11(1):3616. doi:10.1038/s41467-020-17458-0

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 31.

    Sparks A, Woods JP, Bienzle D, Wood GA, Coomber BL. Whole genome sequencing analysis of high confidence variants of B-cell lymphoma in Canis familiaris. PLoS One. 2020;15(8):e0238183. doi:10.1371/journal.pone.0238183

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 32.

    Das S, Idate R, Cronise KE, Gustafson DL, Duval DL. Identifying candidate druggable targets in canine cancer cell lines using whole-exome sequencing. Mol Cancer Ther. 2019;18(8):1460-1471. doi:10.1158/1535-7163.MCT-18-1346

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 33.

    Truvé K, Dickinson P, Xiong A, et al. Utilizing the dog genome in the search for novel candidate genes involved in glioma development-genome wide association mapping followed by targeted massive parallel sequencing identifies a strongly associated locus. PLoS Genet. 2016;12(5):e1006000. doi:10.1371/journal.pgen.1006000

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 34.

    Wang G, Wu M, Durham AC, Mason NJ, Roth DB. Canine Oncopanel: a capture-based, NGS platform for evaluating the mutational landscape and detecting putative driver mutations in canine cancers. Vet Comp Oncol. 2022;20(1):91-101. doi:10.1111/vco.12746

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 35.

    Wong S, Warrier M, Byron S, et al. Design, analytical validation, and diagnostic yield of a novel canine cancer gene sequencing panel. Abstract in: Proceedings of the 2020 VCS Virtual Conference. Veterinary Cancer Society; 2020:122.

    • Search Google Scholar
    • Export Citation
  • 36.

    Lorch G, Sivaprakasam K, Zismann V, et al. Identification of recurrent activating HER2 mutations in primary canine pulmonary adenocarcinoma. Clin Cancer Res. 2019;25(19):5866-5877. doi:10.1158/1078-0432.CCR-19-1145

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 37.

    Wong S, Ehrhart EJ, Stewart S, et al. Genomic landscapes of canine splenic angiosarcoma (hemangiosarcoma) contain extensive heterogeneity within and between patients. PLoS One. 2022;17(7):e0264986. doi:10.1371/journal.pone.0264986

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 38.

    Jennings LJ, Arcila ME, Corless C, et al. Guidelines for validation of next-generation sequencing-based oncology panels: a joint consensus recommendation of the Association for Molecular Pathology and College of American Pathologists. J Mol Diagn. 2017;19(3):341-365. doi:10.1016/j.jmoldx.2017.01.011

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 39.

    Roy S, Coldren C, Karunamurthy A, et al. Standards and guidelines for validating next-generation sequencing bioinformatics pipelines: a joint recommendation of the Association for Molecular Pathology and the College of American Pathologists. J Mol Diagn. 2018;20(1):4-27. doi:10.1016/j.jmoldx.2017.11.003

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 40.

    Kohn MA, Senyak J. Sample size calculators. UCSF Clinical and Translational Science Institute. Accessed February 3, 2023. https://www.sample-size.net/

    • Search Google Scholar
    • Export Citation
  • 41.

    GraphPad Prism. Version 9.4.1 for Mac OS. GraphPad Software; 2022. Accessed November 1, 2022. https://www.graphpad.com

Appendix

Genes and mutations detected by a genomic assay (SearchLight DNA; Vidium Animal Health) used in identifying and annotating mutations in fine-needle aspirate or biopsy samples of suspected malignancies in 69 dogs for which assay reports were generated between September 28, 2020, and July 31, 2022. The 120 genes evaluated by this assay are labeled in the diagonal columns, and the mutation types evaluated by this assay are listed in the left column. Solid circles are placed in the appropriate box(es) to indicate what mutation types are evaluated for the respective gene. CNV = Copy number variant. ITD = Internal tandem duplication. SNV = Single-nucleotide variant.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 4561 1882 96
PDF Downloads 3250 1378 60
Advertisement