Dogs and humans with neoplasia have an increased risk of venous thromboembolism (VTE). Human cancer patients have a 4- to 7-fold increased relative risk of VTE compared to healthy controls.1–6 VTE events can be asymptomatic and discovered incidentally during cancer staging,1,7–10 or they can be symptomatic and contribute significantly to patient morbidity and mortality. VTE has been identified as one of the leading causes of mortality in human cancer patients.11,12 Indeed, mortality in human cancer patients with a history of VTE was 22%, compared to 4.4% in those with VTE without cancer (although cancer patients in this study had a higher incidence of comorbid conditions, such as elevated creatinine).12 Another study13 of cancer patients in Denmark showed an overall risk of mortality after VTE of 13.6% at 30 days, and 48.3% at 1 year. In a prospective observational study11 in human cancer patients receiving chemotherapy, thrombosis was a common cause of mortality, second only to progression of cancer. Neoplasia is the most common underlying disease in dogs with documented thrombotic events, both venous and arterial.14–19 The exact mechanisms of the hypercoagulability associated with neoplasia are incompletely understood and are likely multifactorial. While a thorough review of hypercoagulability causes in patients with cancer is beyond the scope of this paper, proposed mechanisms include thrombocytosis, a hyperinflammatory state, and circulating procoagulant extracellular molecules.20–22
Coated platelets are a subpopulation of platelets that are formed ex vivo after dual stimulation with thrombin and a glycoprotein (GP) VI agonist, such as convulxin.23,24 Their surface is coated with procoagulant glycoproteins and alpha granule contents, bridged by serotonin.24 In supporting the assembly of the prothrombinase and tenase complexes, coated platelets have an enhanced ability to generate thrombin, making them prothrombotic.24 In murine models, coated platelets have been shown to promote platelet-driven fibrin formation,25 and decreased coated platelet formation is associated with decreased platelet procoagulant activity in a modified prothrombinase assay.26
The ability of coated platelets to support prothrombinase activity and thrombin generation, both central to coagulation, may enable coated platelets to participate in VTE development.27 Increased coated platelet potential, ie, the percent of platelets converted to coated platelets ex vivo after dual stimulation, is associated with an increased risk of thromboembolic disease in humans.28 Humans with ischemic stroke history produce a greater percentage of coated platelets when compared with healthy controls.29–31 Conversely, decreased coated platelet potential is associated with increased clinical bleeding in humans and in dogs.28 Humans with hemorrhagic stroke history form decreased coated platelets compared with healthy controls.32 Dogs with canine Scott syndrome have a decreased ability to form coated platelets, which may contribute to the increased clinical bleeding observed with this syndrome.23 Additionally, humans receiving serotonin-specific reuptake inhibitors have a tendency to form fewer coated platelets.33 These patients also have a mild increase in bleeding tendency34–36 and a mildly reduced risk of VTE,37 which may be partially due to the role of serotonin in the formation of coated platelets.
Human cancer patients have been shown to have increased subsets of procoagulant platelets expressing phosphatidylserine when compared with healthy controls.38,39 However, the role of coated platelets in the development of VTE in dogs with neoplasia is not known. If coated platelets are found to play a role in VTE in dogs with neoplasia, they could represent a potential therapeutic target.
The primary objective of this study was to evaluate coated platelet potential for the first time in dogs with neoplasia compared with healthy control dogs. It was hypothesized that dogs with neoplasia would produce a greater percentage of coated platelets when compared with healthy control dogs.
Materials and Methods
Dogs
Client-owned dogs presenting to a veterinary teaching hospital with a cytologic or histopathologic diagnosis of either lymphoma or solid tumors (eg, carcinoma, sarcoma, mast cell tumor, etc) were recruited for inclusion in the study. Enrollment occurred over a 3-month period from September to December 2015. Staff- and student-owned dogs that were free of disease based on history and physical examination findings, and not receiving medications other than standard preventative treatments, were recruited as healthy control dogs. Dogs were excluded from enrollment in any group if body weight was less than 5 kg or if they had received any anticoagulant or antiplatelet medications (eg, non-steroidal anti-inflammatory medications) within the 96 hours preceding sampling. For cancer patients, all diagnostic and therapeutic plans were at the discretion of the attending clinician. Dogs were separated into groups based on their clinical remission status for data analysis (either “remission – yes”, or “remission – no”). Dogs in the lymphoma group were classified with regard to clinical stage and substage according to the World Health Organization (WHO) clinical staging system for lymphoma in domestic animals.40 Dogs were categorized by stage and substage according to available clinical information. Stage was determined based on physical exam and diagnostic testing, performed at the discretion of the attending clinician. Substage was determined by the attending clinician. Dogs were included independent of whether they were receiving chemotherapy treatment. Informed owner consent was obtained for all dogs before enrollment in the study, and all procedures were approved by the Institutional Animal Care and Use Committees of Iowa State University (8-14-7840K) and the University of Minnesota (1312-31131A), and the Institutional Biosafety Committee of the University of Minnesota (1509-33048H).
Sample collection procedure
Venous blood (1.8 ml) was collected into a syringe containing 0.2 ml acid-citrate dextrose (Sigma-Aldrich), to a total volume of 2 ml. To generate platelet-rich plasma (PRP) for flow cytometry, the acid-citrate dextrose–anticoagulated blood was combined with an equal volume (2 ml) of buffered saline glucose citrate (129 mM NaCl, 14 mM Na3citrate, 11 mM glucose, and 10 mM NaH2PO4, pH 7.3) and centrifuged at 170 X g for 8 minutes at room temperature (25°C). The supernatant PRP was harvested and coated platelet assay reactions were performed within 4 hours.
Automated platelet counts were obtained on ADVIA 2120 hematology analyzers at both institutions.
Coated platelet assay
All activation experiments were performed in a 100 μl total assay volume containing 1 μl PRP and the following reagents (final concentrations): 2.5 mM CaCl2, 1.25 mM MgCl2, 1 μg/ml biotin-fibrinogen (Sigma-Aldrich), 2.5 μg/ml rattlesnake toxin (Miami Serpentarium), 0.5 U/ml bovine thrombin (Sigma-Aldrich), 0.1 mM gly-pro-arg-pro-NH2, 150 mM NaCl, and 10 mM HEPES, pH 7.5. The reaction tubes were incubated at 37°C for 5 minutes. Following incubation, labeling to detect platelet membrane GP IIbIIIa (CD41/CD61; 0.8 μg fluorescein isothiocyanate-abciximab [FITC-abx]) and biotinylated fibrinogen (0.5 μg phycoerythrin-streptavidin (PE-SA); Streptavidin, R-Phycoerythrin Conjugate; Molecular Probes) was performed to identify platelets and to label surface bound biotinylated fibrinogen characteristic of coated platelets, respectively, as previously described.23,41–44 After incubating an additional 5 minutes at 37°C, the reaction was stopped by fixation with 200 μl of 1.5% formalin in HEPES-saline (10 mM HEPES and 150 mM NaCl, pH 7.5). Samples were incubated at room temperature (25°C) for 15 minutes, while protected from light, then diluted with 200 μl of HEPES-saline (10 mM HEPES and 150 mM NaCl, pH 7.5).
Samples were collected with a FACSCanto (ISU; FACSCanto, BD Biosciences) or an LSR II (UMN; LSR II/Fortessa; BD Biosciences) cytometer and analyzed with FlowJo software (FlowJo, v10, Tree Star Inc). After initial optimization and compensation on each cytometer, the same settings were used throughout the study. Samples were gated to collect 10,000 platelet positive events, based on FITC-abx positivity as a function of forward scatter. Platelets were first identified by forward (FSC) and side scatter (SSC) properties, and platelet gates were fine-tuned based on FITC-abx positivity (Figure 1). Within the platelet population, coated platelets were identified and gated based on their PE-SA positivity. Results are reported as the percentage of platelets that were converted to coated platelets (PE-SA positive).
Flow cytometric analyses of platelet surface fibrinogen expression. Pseudocolor plots show platelet rich plasma dual stimulated with thrombin (0.5 U/ml) and rattlesnake toxin (2.5 μg/ml), and double labeled to detect constitutive platelet membrane GP IIbIIIa (fluorescein isothiocyanate-abciximab) and biotinylated fibrinogen (phycoerythrin-streptavidin). A gate is constructed around the characteristic forward and side scatter of platelets (A). The platelet population (plt1) is then further delineated based on fluorescein isothiocyanate-abciximab labeling (B). The platelet population (plt2) is then further separated based on phycoerythrin-streptavidin labeling (C). In this example, 30.5% of platelets are expressing biotinylated fibrinogen on their surface following dual stimulation with thrombin and rattlesnake toxin (coats).
Citation: American Journal of Veterinary Research 83, 10; 10.2460/ajvr.22.03.0042
Flow cytometric analyses of platelet surface fibrinogen expression. Pseudocolor plots show platelet rich plasma dual stimulated with thrombin (0.5 U/ml) and rattlesnake toxin (2.5 μg/ml), and double labeled to detect constitutive platelet membrane GP IIbIIIa (fluorescein isothiocyanate-abciximab) and biotinylated fibrinogen (phycoerythrin-streptavidin). A gate is constructed around the characteristic forward and side scatter of platelets (A). The platelet population (plt1) is then further delineated based on fluorescein isothiocyanate-abciximab labeling (B). The platelet population (plt2) is then further separated based on phycoerythrin-streptavidin labeling (C). In this example, 30.5% of platelets are expressing biotinylated fibrinogen on their surface following dual stimulation with thrombin and rattlesnake toxin (coats).
Citation: American Journal of Veterinary Research 83, 10; 10.2460/ajvr.22.03.0042
Flow cytometric analyses of platelet surface fibrinogen expression. Pseudocolor plots show platelet rich plasma dual stimulated with thrombin (0.5 U/ml) and rattlesnake toxin (2.5 μg/ml), and double labeled to detect constitutive platelet membrane GP IIbIIIa (fluorescein isothiocyanate-abciximab) and biotinylated fibrinogen (phycoerythrin-streptavidin). A gate is constructed around the characteristic forward and side scatter of platelets (A). The platelet population (plt1) is then further delineated based on fluorescein isothiocyanate-abciximab labeling (B). The platelet population (plt2) is then further separated based on phycoerythrin-streptavidin labeling (C). In this example, 30.5% of platelets are expressing biotinylated fibrinogen on their surface following dual stimulation with thrombin and rattlesnake toxin (coats).
Citation: American Journal of Veterinary Research 83, 10; 10.2460/ajvr.22.03.0042
Statistical analysis
Sample sizes were calculated before the study initiation to detect a 10% difference between cases and controls, a difference approximate to that previously detected between healthy people and those with ischemic stroke,29–31 with 80% power and an alpha of 0.05. Calculations were performed using the standard deviation of a published coated platelet assay.45 Based on these calculations, it was determined that each of the three groups (healthy control dogs, dogs with lymphoma, and dogs with solid tumors) needed to contain 13 dogs to detect a significant difference.
Three of the reaction tubes, containing identical reagents, were considered technical replicates and were averaged. Results were discarded if one of the technical replicates was considered an outlier compared to the other two (defined as a difference of > 10% coated platelets) or if assay failure occurred (defined as either a lack of stimulation above resting levels, or maximal stimulation, > 95%, above resting levels). The overall intra-assay coefficient of variation (CV) was calculated by averaging the CVs from the triplicate measurements of each sample analyzed.
Normality was confirmed using qq-plots and boxplots. To compare the percentage of coated platelets between the three groups, an ANOVA was performed followed by pairwise 95% confidence intervals adjusted for multiple comparisons using Tukey’s method.
To see how the relationship between percent coated and group depends on total platelet count, an analysis of covariance (ANCOVA) model was fit with total platelet count as covariate. This model was performed only on the dogs with cancer. No visual evidence of an interaction was seen so the interaction was not included in the model. Total platelet count was missing for one dog in the solid group, so for comparison with the ANCOVA, the initial ANOVA model was refit on this subset of the data. The mean difference between groups, with a 95% confidence interval (CI) still adjusted for three comparisons, is reported for both the refit ANOVA model and the ANCOVA model. The coefficient for total platelet count, with 95% CI, is also reported from the ANCOVA model.
Results
Dogs
In total, 55 dogs were screened for enrollment in the study and 47 were enrolled. Clinical characteristics of enrolled dogs are summarized (Supplementary Tables S1–S3). Dogs were excluded due to assay failure early in the course of the study (n = 4, 2 from the healthy control group and 2 from the lymphoma group), emergence of significant underlying disease within 2 months of sampling, making them ineligible to be included as healthy controls (2; 1 case was diagnosed with prostatic adenocarcinoma and 1 case was diagnosed with a severe submandibular abscess), and lack of a definitive diagnosis (n = 2). Healthy dogs were followed for a period of 6 months to ensure they remained clinically healthy. The 47 dogs included in the study consisted of healthy controls (n = 14), dogs with a diagnosis of lymphoma (19), and dogs with solid tumors (14).
Clinical characteristics of dogs in the healthy control group
The healthy control dogs had a mean age of 5.0 years (range, 1 to 9 years) and consisted of spayed females (7/14; 50%) and castrated males (7/14; 50%). Breeds represented in the healthy control group were mixed breed (3/14; 21.4%), Australian Shepherd (2/14; 14.3%), American Staffordshire (Pit Bull) Terrier (2/14; 14.3%), Labrador Retriever (2/14; 14.3%), and 1 each (1/14; 7.1%) of German Wirehaired Pointer, Rough-Coated Collie, Golden Retriever, Greyhound, and Border Collie.
Clinical characteristics of dogs in the lymphoma group
Of the dogs with a diagnosis of lymphoma, 12/19 (63.2%) were considered to have clinical disease at the time of enrollment and 7/19 (36.8%) were considered to be in clinical remission. Dogs in the lymphoma group had a mean age of 9.2 years (range, 3 to 15 years) and consisted of spayed females (8/19; 42.1%), castrated males (9/19; 47.4%), 1 intact female (5.3%), and 1 intact male (5.3%). Breeds represented in the lymphoma group included Beagle (2/19; 10.5%), Golden Retriever (2/19; 10.5%), and 1 each (1/19; 5.3%) of Vizsla, Bernese Mountain Dog, mixed breed, Doberman, Jack Russell Terrier, Border Collie, Boxer, Newfoundland, Standard Poodle, Cairn Terrier, Chow Chow, Cocker Spaniel, German Shepherd Dog, Basset Hound, and Irish Setter. Diagnosis of lymphoma was based on a combination of cytology and clonality testing (12/19; 63.2%), histopathology (5/19; 26.3%), or cytology alone (2/19; 10.5%). The most common WHO Stage at the time of diagnosis of lymphoma was Stage III (9/19; 47.4%), followed by Stage V (5/19; 26.3%), Stage II (3/19; 15.8%), and Stage IV (2/19; 10.5%). At the time of diagnosis, 10 dogs were considered to be substage b (10/19; 52.6%), and 9 dogs were substage a (9/19; 47.4%). Immunophenotyping was consistent with B-cell neoplasia in 10 cases and T-cell neoplasia in 5 cases. Four dogs with lymphoma did not have immunophenotyping performed. Twelve of the dogs with lymphoma were receiving chemotherapy at the time of enrollment, including 5 of the 12 dogs that were not in clinical remission and all 7 of the dogs in clinical remission. The most common chemotherapy protocol was the CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) protocol (7/12; 58.3%). The following chemotherapy protocols were used in one dog each: modified MOPP (mechlorethamine, vincristine, prednisone, procarbazine); chlorambucil, vincristine, and cyclophosphamide; doxorubicin and lomustine; and doxorubicin alone. Ten of the dogs were treated with systemic glucocorticoids (prednisone [9/10; 90%] and prednisolone [1/10; 10%]) as part of the CHOP protocol (7/10; 70%) and in 1 dog each (1/10; 10%) as part of the MOPP protocol, concurrently with doxorubicin, and as the only treatment, at the time of enrollment. Comorbidities are summarized (Supplementary Table S2). None of the comorbid conditions present are known to be associated with coated platelet formation based on extrapolation from the human literature, and none are known to be associated with hypercoagulability in dogs.
Clinical characteristics of dogs in the solid tumor group
Of the dogs with solid tumors, 12/14 (85.7%) were considered to have residual disease at the time of enrollment and 2/14 (14.3%) were considered to be in clinical remission. Dogs in the solid tumor group had a mean age of 9.4 years (range, 5 to 13 years) and consisted of castrated males (9/14; 64.3%), spayed females (4/14; 28.6%), and 1 intact male (7.1%). Breeds represented in the solid tumor group included mixed breed (4/14; 28.6%), English Springer Spaniel (2/14; 14.3%), Golden Retriever (2/14; 14.3%), and 1 each (7.1%) of Boxer, Boston Terrier, English Setter, German Shepherd Dog, Labrador Retriever, and Australian Shepherd. Diagnosis of neoplasia in dogs with solid tumors was based on histopathology (8/14; 57.1%), cytology (3/14; 21.4%), or both (2/14; 14.3%). One dog was enrolled that had a presumptive diagnosis of an insulinoma based on clinical signs, documented hypoglycemia with concurrent elevated blood insulin levels, and evidence of a pancreatic mass on computed tomography.
Dogs with solid tumors had a variety of cytologic or histopathologic tumor types, including anal gland adenocarcinoma (2/14; 14.3%), insulinoma (2/14; 14.3%; one presumptive and one confirmed based on histopathology), osteosarcoma (2/14; 14.3%), chondrosarcoma (2/14; 14.3%), and 1 each (1/14; 7.1%) of mast cell tumor, nasal adenocarcinoma, hemangiosarcoma, oligodendroglial ganglioglioma, and clear cell adnexal carcinoma. One dog had an oral melanoma that had previously been surgically excised with incomplete margins, and an anal gland adenocarcinoma diagnosed on the day of study enrollment that had therefore not been treated. Two dogs (2/14; 14.3%) were receiving chemotherapeutic treatment at the time of enrollment, including one that had previously received a full course of fractionated radiation therapy that was receiving carboplatin for an anal gland adenocarcinoma, and one dog with hemangiosarcoma that was receiving doxorubicin and a bispecific ligand-targeted angiotoxin as part of a clinical trial. The dog that was receiving carboplatin was also receiving metoclopramide and metronidazole for chemotherapy-related side effects. Three dogs (3/14; 21.4%) were receiving systemic glucocorticoids. Comorbidities are listed in Supplementary Table S3. None of the comorbid conditions present are known to be associated with coated platelet formation based on extrapolation from the human literature, and none are known to be associated with hypercoagulability in dogs.
Coated platelet technical replicates
Results of one of the technical replicate reactions were discarded in 6 instances (6/141; 4.3%). This occurred due to consideration as an outlier (defined as a difference of > 10% coated platelets compared to the other 2 technical replicate tubes) in 5/141 reactions (3.5%) and due to assay failure in 1/141 reaction (0.7%).
Comparison of coated platelet percentage between healthy dogs and dogs with solid tumors or lymphoma
When all dogs were included regardless of remission status, the mean percentage (± standard deviation) of coated platelets produced by healthy control dogs was 30.8% (±11.8%; range, 15.1 to 49.5%), the mean percentage of coated platelets produced by dogs with solid tumors was 41.7% (±9%; range, 26.3 to 56.5%), and the mean percentage of coated platelets produced by dogs with lymphoma was 30.5% (±15.9%; range, 4.5 to 63%). There was overlap of the distribution of data between the three groups (Figure 2).
Coated platelet percentages in healthy dogs compared to dogs with lymphoma and solid tumors. Cases with active disease are shown in red, whereas cases considered to be in clinical remission are shown in blue. The horizontal lines represent the median coated platelet percentage for each group when all cases are included (regardless of remission status), and the boxes represent the interquartile range when all cases are included.
Citation: American Journal of Veterinary Research 83, 10; 10.2460/ajvr.22.03.0042
Coated platelet percentages in healthy dogs compared to dogs with lymphoma and solid tumors. Cases with active disease are shown in red, whereas cases considered to be in clinical remission are shown in blue. The horizontal lines represent the median coated platelet percentage for each group when all cases are included (regardless of remission status), and the boxes represent the interquartile range when all cases are included.
Citation: American Journal of Veterinary Research 83, 10; 10.2460/ajvr.22.03.0042
Coated platelet percentages in healthy dogs compared to dogs with lymphoma and solid tumors. Cases with active disease are shown in red, whereas cases considered to be in clinical remission are shown in blue. The horizontal lines represent the median coated platelet percentage for each group when all cases are included (regardless of remission status), and the boxes represent the interquartile range when all cases are included.
Citation: American Journal of Veterinary Research 83, 10; 10.2460/ajvr.22.03.0042
We observed a greater mean percentage of coated platelets in dogs with solid tumors, compared with healthy control dogs, by 10.9 percentage points (95% CI: −1.0, 22.8; P = .07; Figure 2). We observed a mean percentage of coated platelets in dogs with lymphoma that was less than healthy control dogs by 0.3 percentage points (95% CI: −11.4, 10.8; P = .99).
When dogs in clinical remission were excluded, the mean percentage of coated platelets produced by dogs with solid tumors was 41.4% (range, 26.3 to 56.5%) and the mean percentage of coated platelets produced by dogs with lymphoma was 26.7% (range, 4.5 to 63%). We similarly observed a greater mean percentage of coated platelets in dogs with solid tumors, compared with healthy control dogs, by 10.6 percentage points (95% CI: −2.2, 23.4; P = 0.1). We observed a mean percentage of coated platelets in dogs with lymphoma that was less than healthy control dogs by 4.1 percentage points (95% CI: −16.9, 8.6; P = .7; Figure 2).
The average intra-assay CV for coated platelets in this study was 6.9%. No CV was greater than 20%, which is the threshold often used as the limit of acceptable precision for assays.
Comparison of platelet counts in dogs with solid tumors and dogs with lymphoma
The mean platelet count of the neoplasia cases included in this study was 333 X 103 platelets/μl (range, 124 to 667 X 103/μl). Dogs with lymphoma had a mean (±standard deviation) platelet count of 321 X 103/μl (±160 X 103/μl; range, 124 to 667 X 103/μl). Of the lymphoma dogs, 7/19 (36.8%) had platelet counts outside of the reference range (200 to 500 X 103/μl); 4/7 were below the reference range (124 to 192 X 103/μl) and 3/7 were above the reference range (506 to 667 X 103/μl). Dogs with solid tumors had a mean platelet count of 349 X 103/μl (±102 X 103/μl; range, 124 to 667 X 103/μl). Of the solid tumor dogs, 2/13 (15.4%) had platelet counts outside of the reference range (200-500 X 103/μl); 1 had a platelet count below the reference range (167 X 103/μl) and 1 had a platelet count above the reference range (545 X 103/μl). One dog in the solid tumor group did not have a platelet count performed on the day of enrollment but historically had platelet counts within the normal reference range.
The healthy control dogs had a mean platelet count of 226 X 103/μl (±46 X 103/μl; range, 153 to 294 X 103/μl). Four of the 14 healthy control dogs had platelet counts below the reference range (range, 153 to 183 X 103/μl), and no healthy control dog had a platelet count above the reference range.
Using the model with only data for which total platelet count was present, the mean percentage of coated platelets was higher in dogs with solid tumors than dogs with lymphoma by 11.4 percentage points (95% CI: −1.1, 23.8). When total platelet count was included as a covariate, the estimate was similar and the precision increased only slightly, with an adjusted mean difference of 10.3 percentage points (95% CI: −1.5, 22.1). In the ANCOVA model, the total platelet count was associated with percent coated platelets, with an increase of 100 X 103/μl in total platelets associated with a 3.8 percentage point increase in percent coated platelets (95% CI: 0.34, 7.2; Figure 3).
Percent coated platelets by total platelet count (X 103/μl) in dogs with lymphoma and solid tumors, with points colored by group. Lines and bands represent linear fits and 95% confidence intervals for each group.
Citation: American Journal of Veterinary Research 83, 10; 10.2460/ajvr.22.03.0042
Percent coated platelets by total platelet count (X 103/μl) in dogs with lymphoma and solid tumors, with points colored by group. Lines and bands represent linear fits and 95% confidence intervals for each group.
Citation: American Journal of Veterinary Research 83, 10; 10.2460/ajvr.22.03.0042
Percent coated platelets by total platelet count (X 103/μl) in dogs with lymphoma and solid tumors, with points colored by group. Lines and bands represent linear fits and 95% confidence intervals for each group.
Citation: American Journal of Veterinary Research 83, 10; 10.2460/ajvr.22.03.0042
Thrombotic and hemorrhagic events
None of the dogs included in this study had evidence of a thrombotic event at the time of enrollment, historically, or for a period of 2 years after enrollment. With the exception of the dog with hematuria, which was presumed secondary to either a urinary tract infection or sterile hemorrhagic cystitis secondary to cyclophosphamide, no dog included in this study had evidence of clinical bleeding.
Discussion
The primary objective of this study was to evaluate coated platelet potential in dogs with neoplasia compared to healthy controls. Given the large range (15.1 to 49.5%) of coated platelet formation in our healthy dog population, there was overlap between coated platelet percentage between solid tumor dogs, dogs with lymphoma, and healthy control dogs. However, our data are compatible with dogs with solid tumors having a higher mean coated platelet percentage than the healthy control dogs.
Hypercoagulability has been well documented in both humans and dogs with cancer. Neoplasia was found to be the most common underlying disease found in dogs with documented pulmonary thromboembolism,15,16 portal vein thrombosis,17,19 and splenic vein thrombosis.46 Although the data from this study did not document a definitive role for coated platelets in canine neoplasia, it was the first time this subset of procoagulant platelets has been evaluated in dogs with cancer.
There is still much to be learned regarding the underlying causes of the hypercoagulability that is seen in cancer patients. Several studies have documented a tendency toward hypercoagulability in canine cancer patients. Andreasen et al47 showed that 47/71 (66.2%) of canine cancer patients had prothrombotic thromboelastography (TEG) profiles. Carcinoma patients, with the exception of mammary carcinomas, had increased clot strength on TEG and increased fibrinogen. Dogs with distant metastases had significantly elevated fibrinogen and d-dimer levels compared to those with only local disease.47 Another study48 evaluating TEG in dogs with neoplasia found similar results, with dogs with malignant disease displaying prothrombotic TEG profiles, including decreased time to clot formation and increased clot strength. Adenocarcinomas were the most common neoplasia associated with hypercoagulability in that study. Kol et al49 showed that 17/27 (63%) of dogs with multicentric lymphoma also had prothrombotic TEG parameters, and this tendency persisted throughout treatment and for up to a month after clinical remission was attained. In these lymphoma patients, accelerated clot formation on TEG at the end of the chemotherapy regimen was associated with decreased survival.49 Limited information is known about the correlation between coated platelet formation ex vivo and TEG parameters; however, Scott syndrome dogs (that have a decreased ability to form coated platelets) had prolonged mean clotting times (K) and prolonged time to maximal rate of thrombus generation on nontissue factor–activated TEG compared with same day healthy control samples.50 Additionally, TEG parameters and percentage of coated platelets have been shown to change over time in stored platelet concentrate samples from healthy human donors.51
Coated platelets have not previously been evaluated in dogs with cancer and were investigated here as a potential mechanism for their documented hypercoagulability. In the current study, we observed a greater mean percentage of coated platelets in dogs with solid tumors compared to healthy controls and dogs with lymphoma. However, a major limitation of this study is that the variability in our patient population was notably larger than that in the data set used to estimate the variability for our power calculation,45 resulting in this study potentially being underpowered and having less precision than planned. Additionally, the sample sizes were calculated to detect a difference of 10% in coated platelet production. This percentage is considered clinically significant in humans.30,31 However, it is unknown if the difference in coated platelet percentage required for clinical significance in humans can be translated to dogs. Another limitation is that inter-assay variability for this assay for coated platelets in dogs has not been established. Our population of healthy control dogs displayed a large range (15.1 to 49.5%) of coated platelets upon ex vivo stimulation. We did confirm that the assay utilized in this study had very low intra-assay variability. However, samples in this study were analyzed on two different cytometers at two different institutions. Attempts were made to minimize other variables when possible. The assays were performed by the same investigators, using the same reagents and antibodies, using the same analysis software. Use of the same reagents and analysis software was shown to improve consistency of results in a flow cytometric study of canine lymphoproliferative disorders.52 This same study also identified a strong agreement across 9 institutions using various cytometers but the same antibodies. However, given these limitations, this study should be considered a hypothesis-generating study that will better inform future targeted studies.
The cases enrolled were very heterogeneous, but all eligible cases were included since coated platelets had not yet been evaluated in dogs with cancer. Cases differed with regards to treatment status, with some dogs having had surgical treatments of their neoplasia, and with some dogs receiving chemotherapy, glucocorticoids, and/or other medications. The effect of each of these treatments on coated platelets is not currently known; however, it must be acknowledged that these treatments may have influenced our results. It was decided to include dogs that were receiving chemotherapy and glucocorticoids in this initial hypothesis-generating study because both the underlying neoplasia and the treatment effects have potential implications for thrombotic outcomes. Endogenous and exogenous glucocorticoids have a documented association with hypercoagulability in both dogs and humans.53–56 Human cancer patients receiving chemotherapy have an increased risk of thrombotic disease compared to human cancer patients that are not receiving chemotherapy.3,11,57,58 Similarly, thrombosis is also a known potential complication after cancer surgery in humans.59 To improve the precision of these findings, coated platelets should be evaluated in a larger or more uniform population of dogs with neoplasia. Coated platelets could also be evaluated in dogs with cancer at the time of diagnosis, before starting therapy, to reduce the potential for confounding variables associated with treatment. The data from our current study can be employed to determine the appropriate sample size for future studies.
In summary, our data are compatible with dogs with solid tumors having a greater mean percentage of coated platelets when compared to healthy control dogs. Further studies investigating coated platelet production in more specific subsets of neoplasia and investigating additional mechanisms of hypercoagulability in dogs with neoplasia are warranted. It may also be warranted to investigate potential therapies to mitigate the formation of coated platelets in dogs at risk of thrombosis, such as serotonin-specific reuptake inhibitors.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org
Acknowledgments
This study was supported by the Iowa State University Veterinary Clinical Sciences Research Incentive fund.
The authors thank Dr. George Dale for the generous provision of coated platelet reagents (biotinylated fibrinogen, rattlesnake toxin, and fluorescein isothiocyanate abciximab) and for guidance with the coated platelet assay.
The authors declare that there were no conflicts of interest.
The authors acknowledge the Flow Cytometry Facility of Iowa State University, especially Shawn Rigby, and the University Flow Cytometry Resource of the University of Minnesota. The authors thank Austin Viall, DVM, MS, DACVP, for guidance in data analysis. This work would not have been possible without the talents and time of Unity Jeffery, VetMB, PhD.
References
- 1.↑
Connolly GC, Menapace L, Safadjou S, Francis CW, Khorana AA. Prevalence and clinical significance of incidental and clinically suspected venous thromboembolism in lung cancer patients. Clin Lung Cancer. 2013;14(6):713–718.
- 2.
Cronin-Fenton DP, Søndergaard F, Pedersen LA, et al. Hospitalisation for venous thromboembolism in cancer patients and the general population: a population-based cohort study in Denmark, 1997-2006. Br J Cancer. 2010;103(7):947–953.
- 3.↑
Falanga A. The incidence and risk of venous thromboembolism associated with cancer and nonsurgical cancer treatment. Cancer Invest. 2009;27(1):105–115.
- 4.
Falanga A, Russo L. Epidemiology, risk and outcomes of venous thromboembolism in cancer. Hamostaseologie. 2012;32(2):115–125.
- 5.
Khorana AA, Connolly GC. Assessing risk of venous thromboembolism in the patient with cancer. J Clin Oncol. 2009;27(29):4839–4847.
- 6.↑
Sud R, Khorana AA. Cancer-associated thrombosis: risk factors, candidate biomarkers and a risk model. Thromb Res. 2009;123(Suppl 4):S18–S21.
- 7.↑
Di Nisio M, Lee AYY, Carrier M, Liebman HA, Khorana AA, Subcommittee on Haemostasis and Malignancy. Diagnosis and treatment of incidental venous thromboembolism in cancer patients: guidance from the SSC of the ISTH. J Thromb Haemost. 2015;13(5):880–883.
- 8.
Douma RA, Kok MGM, Verberne LM, Kamphuisen PW, Büller HR. Incidental venous thromboembolism in cancer patients: prevalence and consequence. Thromb Res. 2010;125(6):e306–309.
- 9.
Khorana AA, O’Connell C, Agnelli G, Liebman HA, Lee AYY, Subcommittee on Hemostasis and Malignancy of the SSC of the ISTH. Incidental venous thromboembolism in oncology patients. J Thromb Haemost. 2012;10(12):2602–2604.
- 10.↑
Menapace LA, Peterson DR, Berry A, Sousou T, Khorana AA. Symptomatic and incidental thromboembolism are both associated with mortality in pancreatic cancer. Thromb Haemost. 2011;106(2):371–378.
- 11.↑
Khorana AA, Francis CW, Culakova E, Kuderer NM, Lyman GH. Thromboembolism is a leading cause of death in cancer patients receiving outpatient chemotherapy. J Thromb Haemost. 2007;5(3):632–634.
- 12.↑
Monreal M, Falgá C, Valdés M, et al. Fatal pulmonary embolism and fatal bleeding in cancer patients with venous thromboembolism: findings from the RIETE registry. J Thromb Haemost. 2006;4(9):1950–1956.
- 13.↑
Ording AG, Skjøth F, Søgaard M, et al. Increasing incidence and declining mortality after cancer-associated venous thromboembolism: a nationwide cohort study. Am J Med. 2021;134(7):868–876.e5.
- 14.↑
Boswood A, Lamb CR, White RN. Aortic and iliac thrombosis in six dogs. J Small Anim Pract. 2000;41(3):109–114.
- 15.↑
LaRue MJ, Murtaugh RJ. Pulmonary thromboembolism in dogs: 47 cases (1986-1987). J Am Vet Med Assoc. 1990;197(10):1368–1372.
- 16.↑
Johnson LR, Lappin MR, Baker DC. Pulmonary thromboembolism in 29 dogs: 1985-1995. J Vet Intern Med. 1999;13(4):338–345.
- 17.↑
Respess M, O’Toole TE, Taeymans O, Rogers CL, Johnston A, Webster CRL. Portal vein thrombosis in 33 dogs: 1998-2011. J Vet Intern Med. 2012;26(2):230–237.
- 18.
Palmer KG, King LG, Van Winkle TJ. Clinical manifestations and associated disease syndromes in dogs with cranial vena cava thrombosis: 17 cases (1989-1996). J Am Vet Med Assoc. 1998;213(2):220–224.
- 19.↑
Van Winkle TJ, Bruce E. Thrombosis of the portal vein in eleven dogs. Vet Pathol. 1993;30(1):28–35.
- 20.↑
Kim AS, Khorana AA, McCrae KR. Mechanisms and biomarkers of cancer-associated thrombosis. Transl Res. 2020;225:33–53.
- 21.
Hisada Y, Mackman N. Cancer-associated pathways and biomarkers of venous thrombosis. Blood. 2017;130(13):1499–1506.
- 22.↑
Sharma BK, Flick MJ, Palumbo JS. Cancer-associated thrombosis: a two-way street. Semin Thromb Hemost. 2019;45(6):559–568.
- 23.↑
Brooks MB, Catalfamo JL, Friese P, Dale GL. Scott syndrome dogs have impaired coated-platelet formation and calcein-release but normal mitochondrial depolarization. J Thromb Haemost. 2007;5(9):1972–1974.
- 24.↑
Dale GL. Coated-platelets: an emerging component of the procoagulant response. J Thromb Haemost. 2005;3(10):2185–2192.
- 25.↑
Mattheij NJA, Swieringa F, Mastenbroek TG, et al. Coated platelets function in platelet-dependent fibrin formation via integrin αIIbβ3 and transglutaminase factor XIII. Haematologica. 2016;101(4):427–436.
- 26.↑
Jobe SM, Leo L, Eastvold JS, et al. Role of FcRγ and factor XIIIA in coated platelet formation. Blood. 2005;106(13):4146–4151.
- 27.↑
Prodan CI, Stoner JA, Cowan LD, Dale GL. Higher coated-platelet levels are associated with stroke recurrence following nonlacunar brain infarction. J Cereb Blood Flow Metab. 2013;33(2):287–292.
- 28.↑
Podoplelova NA, Nechipurenko DY, Ignatova AA, Sveshnikova AN, Panteleev MA. Procoagulant platelets: mechanisms of generation and action. Hamostaseologie. 2021;41(2):146–153.
- 29.↑
Prodan CI, Dale GL. Coated-platelets in ischemic stroke - potential insight into the etiology of stroke subtypes. Int J Stroke. 2008;3(4):249–250.
- 30.↑
Prodan CI, Joseph PM, Vincent AS, Dale GL. Coated-platelets in ischemic stroke: differences between lacunar and cortical stroke. J Thromb Haemost. 2008;6(4):609–614.
- 31.↑
Prodan CI, Vincent AS, Dale GL. Coated-platelet levels are elevated in patients with transient ischemic attack. Transl Res. 2011;158(1):71–75.
- 32.↑
Prodan CI, Stoner JA, Cowan LD, Dale GL. Lower coated-platelet levels are associated with early hemorrhagic transformation in patients with non-lacunar brain infarction. J Thromb Haemost. 2010;8(6):1185–1190.
- 33.↑
Prodan CI, Joseph PM, Vincent AS, Dale GL. Coated-platelet levels are influenced by smoking, aspirin, and selective serotonin reuptake inhibitors. J Thromb Haemost. 2007;5(10):2149–2151.
- 34.↑
Serebruany VL. Selective serotonin reuptake inhibitors and increased bleeding risk: are we missing something? Am J Med. 2006;119(2):113–116.
- 35.
Aarts N, Akoudad S, Noordam R, et al. Inhibition of serotonin reuptake by antidepressants and cerebral microbleeds in the general population. Stroke. 2014;45(7):1951–1957.
- 36.↑
Cochran KA, Cavallari LH, Shapiro NL, Bishop JR. Bleeding incidence with concomitant use of antidepressants and warfarin. Ther Drug Monit. 2011;33(4):433–438.
- 37.↑
Kimmel SE, Schelleman H, Berlin JA, et al. The effect of selective serotonin re-uptake inhibitors on the risk of myocardial infarction in a cohort of patients with depression. Br J Clin Pharmacol. 2011;72(3):514–517.
- 38.↑
Zhao L, Bi Y, Kou J, Shi J, Piao D. Phosphatidylserine exposing-platelets and microparticles promote procoagulant activity in colon cancer patients. J Exp Clin Cancer Res. 2016;35:54.
- 39.↑
Yang C, Ma R, Jiang T, et al. Contributions of phosphatidylserine-positive platelets and leukocytes and microparticles to hypercoagulable state in gastric cancer patients. Tumour Biol. 2016;37(6):7881–7891.
- 40.↑
Valli VE, San Myint M, Barthel A, et al. Classification of canine malignant lymphomas according to the World Health Organization criteria. Vet Pathol. 2011;48(1):198–211.
- 41.↑
Alberio L, Safa O, Clemetson KJ, Esmon CT, Dale GL. Surface expression and functional characterization of alpha-granule factor V in human platelets: effects of ionophore A23187, thrombin, collagen, and convulxin. Blood. 2000;95(5):1694–1702.
- 42.
Brooks MB, Catalfamo JL, Brown HA, Ivanova P, Lovaglio J. A hereditary bleeding disorder of dogs caused by a lack of platelet procoagulant activity. Blood. 2002;99(7):2434–2441.
- 43.
Dale GL, Friese P, Batar P, et al. Stimulated platelets use serotonin to enhance their retention of procoagulant proteins on the cell surface. Nature. 2002;415(6868):175–179.
- 44.↑
Remenyi G, Szasz R, Friese P, Dale GL. Role of mitochondrial permeability transition pore in coated-platelet formation. Arterioscler Thromb Vasc Biol. 2005;25(2):467–471.
- 45.↑
Knudsen T, Kjalke M, Tranholm M, Nichols TC, Jensen AL, Kristensen AT. Development of a flow cytometric assay for detection of coated platelets in dogs and evaluation of binding of coated platelets to recombinant human coagulation factor VIIa. Am J Vet Res. 2011;72(8):1007–1014.
- 46.↑
Laurenson MP, Hopper K, Herrera MA, Johnson EG. Concurrent diseases and conditions in dogs with splenic vein thrombosis. J Vet Intern Med. 2010;24(6):1298–1304.
- 47.↑
Andreasen EB, Tranholm M, Wiinberg B, Markussen B, Kristensen AT. Haemostatic alterations in a group of canine cancer patients are associated with cancer type and disease progression. Acta Vet Scand. 2012;54:3.
- 48.↑
Kristensen AT, Wiinberg B, Jessen LR, Andreasen E, Jensen AL. Evaluation of human recombinant tissue factor-activated thromboelastography in 49 dogs with neoplasia. J Vet Intern Med. 2008;22(1):140–147.
- 49.↑
Kol A, Marks SL, Skorupski KA, et al. Serial haemostatic monitoring of dogs with multicentric lymphoma. Vet Comp Oncol. 2015;13(3):255–266.
- 50.↑
Brooks MB, Randolph J, Warner K, Center S. Evaluation of platelet function screening tests to detect platelet procoagulant deficiency in dogs with Scott syndrome. Vet Clin Pathol. 2009;38(3):306–315.
- 51.↑
Svendsen MS, Kristensen AT, Bochsen L, Salado-Jimena JA, Johansson PI. The in vitro effect of recombinant factor VIIa on coated platelet formation and clot dynamics of stored platelet concentrates. Transfusion. 2009;49(10):2186–2194.
- 52.↑
Meichner K, Stokol T, Tarigo J, et al. Multicenter flow cytometry proficiency testing of canine blood and lymph node samples. Vet Clin Pathol. 2020;49(2):249–257.
- 53.↑
Van Zaane B, Nur E, Squizzato A, et al. Hypercoagulable state in Cushing’s syndrome: a systematic review. J Clin Endocrinol Metab. 2009;94(8):2743–2750.
- 54.
Klose TC, Creevy KE, Brainard BM. Evaluation of coagulation status in dogs with naturally occurring canine hyperadrenocorticism. J Vet Emerg Crit Care (San Antonio). 2011;21(6):625–632.
- 55.
Kol A, Nelson RW, Gosselin RC, Borjesson DL. Characterization of thrombelastography over time in dogs with hyperadrenocorticism. Vet J. 2013;197(3):675–681.
- 56.↑
Park FM, Blois SL, Abrams-Ogg ACG, et al. Hypercoagulability and ACTH-dependent hyperadrenocorticism in dogs. J Vet Intern Med. 2013;27(5):1136–1142.
- 57.↑
Khorana AA, Kuderer NM, Culakova E, Lyman GH, Francis CW. Development and validation of a predictive model for chemotherapy-associated thrombosis. Blood. 2008;111(10):4902–4907.
- 58.↑
Scappaticci FA, Skillings JR, Holden SN, et al. Arterial thromboembolic events in patients with metastatic carcinoma treated with chemotherapy and bevacizumab. J Natl Cancer Inst. 2007;99(16):1232–1239.
- 59.↑
Trinh VQ, Karakiewicz PI, Sammon J, et al. Venous thromboembolism after major cancer surgery: temporal trends and patterns of care. JAMA Surg. 2014;149(1):43–49.
