Abstract
Decreased neutrophil function following administration of chemotherapy has been reported in dogs with lymphoma. The first objective of our study was to determine if neutrophil oxidative burst and phagocytic activity are affected by chemotherapy 7 to 10 days following initiation of treatment in dogs with lymphoma and non-lymphoma malignancies. The second objective was to determine if there is a correlation between neutrophil numbers and neutrophil function before or after initiation of chemotherapy. Flow cytometric assessment of neutrophil oxidative burst and phagocytosis following stimulation with Escherichia coli was performed in 9 dogs diagnosed with lymphoma and 17 non-lymphoma tumor-bearing dogs pre- and post-chemotherapy, as well as 14 tumor-free control dogs. Spearman rank correlation was performed to determine if blood neutrophil numbers and neutrophil function were significantly correlated. Lymphoma patients showed significantly reduced percentage neutrophil oxidative burst post-chemotherapy compared to healthy controls as well as compared to pre-chemotherapy values (P = 0.0022 and P = 0.0020, respectively). Lymphoma patients also exhibited significantly reduced neutrophil phagocytosis activity post-chemotherapy compared to controls and pre-chemotherapy values (P = 0.0016 and P = 0.014, respectively). Dogs with non-lymphoma malignancies also showed a significant decrease in both percentage oxidative burst and phagocytosis post-chemotherapy compared to pre-chemotherapy values (P = 0.00040 and P = 0.029, respectively). Neutrophil numbers and function were not significantly correlated. The results of the study suggest that chemotherapeutic treatment decreases neutrophil oxidative burst and phagocytic activity 7 to 10 days post-treatment in dogs with various malignancies. Furthermore, neutrophil numbers cannot be used to predict neutrophil function.
Résumé
Une diminution de la fonction des neutrophiles après l’administration d’une chimiothérapie a été rapportée chez des chiens atteints de lymphome. Le premier objectif de notre étude était de déterminer si la stimulation oxydative des neutrophiles et l’activité phagocytaire sont affectées par la chimiothérapie 7 à 10 jours après le début du traitement chez les chiens atteints de lymphomes et de tumeurs malignes non lymphomateuses. Le deuxième objectif était de déterminer s’il existe une corrélation entre les nombres de neutrophiles et la fonction des neutrophiles avant ou après le début de la chimiothérapie. L’évaluation par cytométrie en flux de la stimulation oxydative des neutrophiles et de la phagocytose après stimulation par Escherichia coli a été réalisée chez neuf chiens diagnostiqués avec un lymphome et 17 chiens avec des tumeurs non lymphomateuses avant et après la chimiothérapie, ainsi que 14 chiens témoins sans tumeur. Une corrélation des rangs de Spearman a été effectuée pour déterminer si les nombres de neutrophiles sanguins et la fonction des neutrophiles étaient significativement corrélés. Les patients atteints de lymphome ont montré un pourcentage significativement réduit de stimulation oxydative des neutrophiles après la chimiothérapie par rapport aux témoins sains ainsi que par rapport aux valeurs pré-chimiothérapie (P = 0,0022 et P = 0,0020, respectivement). Les patients atteints de lymphome ont également présenté une activité de phagocytose par les neutrophiles significativement réduite après la chimiothérapie par rapport aux témoins et aux valeurs pré-chimiothérapie (P = 0,0016 et P = 0,014, respectivement). Les chiens atteints de tumeurs malignes non lymphomateuses ont également montré une diminution significative du pourcentage de stimulation oxydative et de la phagocytose post-chimiothérapie par rapport aux valeurs pré-chimiothérapie (P = 0,00040 et P = 0,029, respectivement). Les nombres et la fonction des neutrophiles n’étaient pas significativement corrélés. Les résultats de l’étude suggèrent que le traitement chimiothérapeutique diminue la poussée oxydative des neutrophiles et l’activité phagocytaire 7 à 10 jours après le traitement chez les chiens atteints de diverses tumeurs malignes. De plus, les nombres de neutrophiles ne peuvent pas être utilisés pour prédire la fonction des neutrophiles.
(Traduit par Docteur Serge Messier)
Introduction
Cancer is relatively common in dogs, with reported incidence rates ranging from 142.8/100 000 to 897/100 000 dogs, depending on the geographic location (1). While dogs are increasingly being treated with chemotherapy, such treatment is not benign and patients may experience various adverse events, including bone marrow suppression (2). As a result of myelosuppression, peripheral blood cytopenias may occur, with neutropenia typically occurring first due to the short circulating half-life of neutrophils (4 to 8 h) (2). Moreover, neutrophil count nadirs typically occur 5 to 10 d after chemotherapeutic treatment (3).
As neutrophils are an integral component of the innate immune system, developing severe neutropenia can predispose a patient to life-threatening infections (4). Additionally, infections tend to occur most commonly during periods of neutropenia (5). Severe neutropenia may therefore deter a veterinary oncologist from administering chemotherapy on schedule (6). It is routine practice to conduct a complete blood (cell) count (CBC) before administering chemotherapy, mainly to ascertain the patient’s absolute neutrophil count. Neutrophil values of less than 1.5 × 109/L to 2.5 × 109/L generally result in chemotherapy being postponed (6), although these cutoff values are somewhat arbitrary and may vary among oncologists and institutions.
Due to potential immunosuppression, prophylactic antimicrobial treatment may be used in patients receiving chemotherapy (4). Current guidelines for when chemotherapy should be postponed and whether the patient is placed on antibiotics largely rely on absolute neutrophil counts with concurrent risk factors, such as disease status, hematological malignancies, breed, and weight, also considered (4). Neutrophil function as an additional risk factor for developing infection during neutropenic episodes, however, has yet to be explored.
Studies evaluating neutrophil function in dogs undergoing chemotherapy are limited. One study assessing neutrophil function solely in canine lymphoma patients found that oxidative burst activity decreased 7 d after chemotherapy began, when neutrophils were stimulated with phorbol-myristate-acetate (PMA) and Escherichia coli (7). Another study by the same group found that neutrophil oxidative burst was lower in treatment-naïve canine carcinoma and sarcoma patients than in control dogs, while phagocytic activity was lower in treatment-naïve sarcoma dogs than in controls (8).
The first objective of this study was to determine if neutrophil oxidative burst and phagocytic activity are affected by chemotherapy 7 to 10 d after treatment has begun in canines with lymphoma as well as non-lymphoma malignancies. The second objective was to determine if there is a correlation between neutrophil numbers and neutrophil function in tumor-bearing dogs receiving chemotherapy. We hypothesized that chemotherapy would negatively affect neutrophil function and that neutrophil numbers and neutrophil function would be significantly correlated.
Materials and methods
Cancer patients
Client-owned, tumor-bearing dogs admitted to the Veterinary Medical Center at the Western College of Veterinary Medicine for cancer treatment were recruited for this study. Tumor types were characterized via cytopathologic or histopathologic analysis of tumor specimen(s). Enrollees had not received any previous form of chemotherapy or radiotherapy. Patients with evidence of inflammation on their CBC, as defined by neutrophilia greater than 2× the upper reference interval (> 20 × 109 cells/L) with concurrent toxic change and/or presence of left shift, were excluded from the study to prevent inflammation being a confounder to neutrophil function. Owner consent was obtained with a signed consent form for all dogs recruited to the study. The experiment was approved by the University of Saskatchewan Animal Research Ethics Board.
Paired data from tumor-bearing dogs were used to assess neutrophil function in order to control for individual variation among dogs. A 1-mL baseline (before chemotherapy) whole blood sample preceded the second sample, which was collected 7 to 10 d after initial chemotherapeutic treatment. Samples were placed in sodium heparin vacutainer tube (BD, Franklin Lakes, New Jersey, USA) and analyzed within 24 h of collection, as per assay manufacturer recommendations. A CBC was conducted for each sample at Prairie Diagnostic Services in Saskatoon, Saskatchewan, using an ADVIA 2120i Hematology System (Siemens Canada, Oakville, Ontario), along with a microscopic smear review.
Control patients
Fourteen client-owned control dogs were enrolled in the study. Inclusion criteria for control patients included cancer-free status, absence of metabolic disease, inflammation, infection, or pyrexia based on history, physical examination, and a CBC. A single blood sample was taken from each control patient for neutrophil function analysis and a CBC. Sample procedure, handling, and analysis were identical to those previously described for tumor-bearing dogs.
Assessment of oxidative burst
The PHAGOBURST Kit (341058; Glycotope Biotechnology, Heidelberg, Germany) was used to assess neutrophil intracellular oxidative burst. The assay procedure was as follows: 2 separate 25-μL aliquots of heparinized whole blood were simultaneously incubated in a 37°C water bath for 10 min; 5 μL of opsonized (immunoglobulin and complement) 1 × 109 Escherichia coli/mL (strain: LE392, Mol.Gen.Genet. 150: 53–61.1977) was added to 1 tube, while 5 μL of wash solution was added to the other (negative control). Following incubation, 5 μL of dihydrorhodamine123 (D123) was added to each tube and incubated for another 10 min in a 37°C water bath. Leukocytes were then partially fixed and erythrocytes lysed by incubation in 500 μL of lysing solution for 20 min at room temperature, followed by a washing step. Both samples were then incubated in a light-protected ice bath with 50 μL of DNA stain solution in order to discount aggregation artifact of bacteria and cells during initial flow cytometric assessment. This was followed by adding primary and secondary antibodies as described in the next section.
Assessment of phagocytosis
The PHAGOTEST Kit (341060; Glycotope Biotechnology) was used to assess neutrophil phagocytic activity. To begin, 25 μL of heparinized whole blood was incubated with 5 μL of fluorescein isothiocyanate (FITC)-labeled opsonized 1 × 109 E. coli/mL in a 37°C water bath for 10 min. A second identical control tube was incubated on ice to inhibit phagocytosis. Then, 25 uL of quenching solution was added to quench the fluorescence of surface-bound bacteria, leaving only fluorescence of phagocytized bacteria. Two washing steps were then carried out, followed by cell lysing, another washing step, and incubation with DNA stain solution as previously described. Primary and secondary antibodies were then added.
Addition of antibodies
To ensure that only neutrophils were assessed for levels of oxidative burst and phagocytic activity, primary and secondary antibodies were added to all samples, as described in previously published studies of neutrophil function in canine cancer patients (7,8). Before adding primary antibodies to the tubes, 10 μL from each tube was collected and pooled to make an isotype control sample in order to rule out nonspecific antibody binding. The specificity of anti-canine neutrophil antibody was determined using mouse IgG1 isotype control staining. Primary antibodies (mouse anti-canine neutrophil, clone CAD048A; Kingfisher Biotech, Saint Paul, Minnesota, USA) were added to all samples, excluding the isotype control, followed by ice-incubation for 15 min.
After 2 wash steps, secondary antibodies [AffiniPure F(ab′) fragment donkey anti-mouse IgG-Allophycocyanin (APC)-labeled; Jackson ImmunoResearch, West Grove, Pennsylvania, USA] were added to each sample, including the isotype control sample. The tubes were incubated in a light-protected ice bath for 15 min. After 2 more washing steps, the samples were resuspended in 400 μL of wash solution and immediately analyzed using the flow cytometer.
Flow cytometry and data analysis
Flow cytometry data were acquired using the CytoFLEX Flow Cytometer (Beckman Coulter Canada, Mississauga, Ontario). The same acquisition parameters were used for all flow cytometry experiments. Events from approximately 50 000 peripheral leukocytes (WBCs) were acquired. The data were analyzed using the data software FlowJo (Tree Star, Ashland, Oregon, USA). Initially, total white blood cells (WBCs) were plotted and gated using the forward versus side scatter plot. Neutrophils were then identified via specific antibody labelling: APC versus side scatter plot. Neutrophils were then gated to plot APC versus FITC (oxidative burst or phagocytosis).
To assess the phagocytic and oxidative burst activity of the neutrophils, fluorescence-based gating was set on negative control samples with approximately 1% to 2% of the control cell population included in the gate. The gates were applied to the samples with stimulated neutrophils. Phagocytosis was determined as the percentage of neutrophils engulfing fluorescent-labeled E. coli. Oxidative burst was analyzed both as the percentage of neutrophils undergoing oxidative burst and mean fluorescence intensity (MFI) of neutrophils undergoing oxidative burst, a correlate of the degree of oxidative burst per neutrophil. Fluorescence data were acquired using a log scale.
Statistical analysis
Statistical analysis was carried out using GraphPad Prism 6 (GraphPad Software, San Diego, California, USA). A 1-way analysis of variance (ANOVA)-Kruskal-Wallis test was used to compare oxidative burst and phagocytosis between control and tumor-bearing patients before and after chemotherapy. A Wilcoxon matched pairs signed rank test was used to compare neutrophil function before and after chemotherapy. The Wilcoxon rank-sum test was used to compare the median age between control and cancer dogs. Spearman rank correlation was conducted to determine if blood neutrophil numbers and neutrophil function were significantly correlated. P-values of < 0.05 were considered significant.
Results
Study population
A total of 26 tumor-bearing dogs was recruited to the study and 9 of the 26 patients were diagnosed with lymphoma. Non-lymphoma tumor-bearing dogs were divided into sarcoma patients (9/26), carcinoma patients (4/26), mast cell tumor patients (3/26), and leukemia patients (1/26). Signalment, tumor type, and chemotherapy for the recruited cancer patients are summarized in Table I.
Table I.
Signalment, tumor type, and chemotherapy of canine cancer patients.
| Tumor category | Age (years) | Sex | Breed | Tumor type | Chemotherapy received |
|---|---|---|---|---|---|
| Lymphoma | 13 | FS | Border collie/blue heeler mix | Multicentric lymphoma | Mitoxantrone |
| 7 | FS | Pitbull | Multicentric lymphoma | Cyclophosphamide | |
| 10 | FS | Miniature poodle | Multicentric lymphoma | Vincristine | |
| 10 | MN | Border collie mix | Multicentric lymphoma | Vincristine | |
| 9 | FS | Labrador retriever mix | Multicentric lymphoma | Vincristine | |
| 9 | MN | Shetland sheepdog | Multicentric lymphoma | Vincristine | |
| 8 | FS | Border collie | Multicentric lymphoma | Vincristine | |
| 12 | MN | German shepherd | Multicentric lymphoma | Vincristine | |
| 2 | FS | Husky mix | Intestinal lymphoma | Vincristine | |
| Sarcoma | 8 | FS | Labrador retriever | Osteosarcoma | Carboplatin |
| 11 | FS | German shepherd | Osteosarcoma | Carboplatin | |
| 3 | MN | Komondor | Osteosarcoma | Carboplatin | |
| 5 | MN | Golden retriever | Osteosarcoma | Carboplatin | |
| 10 | MN | Golden retriever | Osteosarcoma | Carboplatin | |
| 12 | FS | Labrador retriever | Metastatic hemangiosarcoma | Doxorubicin | |
| 10 | MN | Border collie | Hemangiosarcoma | Doxorubicin | |
| 6 | MN | Pitbull | Hemangiosarcoma | Doxorubicin | |
| 10 | MN | Coonhound | Histiocytic Sarcoma | Doxorubicin | |
| Carcinoma | 11 | FS | Golden retriever mix | Anal sac adenocarcinoma | Mitoxantrone |
| 11 | MN | Shetland sheepdog | Urothelial carcinoma | Mitoxantrone | |
| 10 | MN | Portuguese water dog | Hemangiosarcoma | Mitoxantrone | |
| 7 | FS | Pomeranian | Mammary carcinoma | Doxorubicin | |
| Mast cell tumor (MCT) | 9 | MN | Golden retriever | High-grade MCT | Vinblastine |
| 3 | MN | Labrador retriever | High-grade MCT | Vinblastine | |
| 3 | MN | French bulldog | Grade-II MCT | Vinblastine | |
| Leukemia | 8 | MN | Rough collie | Acute leukemia (suspect lymphoblastic) | Cyclophosphamide |
FS — Female spayed; MN — Male neutered.
At time of sampling, 4 patients had non-life-threatening concurrent illnesses that included mitral valve insufficiency without clinical signs of heart failure, osteoarthritis, adrenal mass without evidence of clinical hyperadrenocorticism, chronic kidney disease, and uncomplicated urinary tract infection (UTI). The median age of tumorbearing dogs (9 y, range: 2 to 13 y) and control dogs (8.6 y, range: 0.66 to 15 y) was not statistically different (P = 0.7109). Concurrent medications received by the tumor patients included prednisone (9/26), maropitant (23/26), nonsteroidal anti-inflammatory drugs (NSAIDs) (5/26), gabapentin (4/26), diphenhydramine (1/26), famotidine (1/26), trazodone (3/26), telmisartan (1/26), antibiotics (3/26), and omeprazole (1/26). No patients had evidence of fever, inflammation, or infection at the time of blood collection.
Effect of chemotherapy on neutrophil oxidative burst in lymphoma patients
While both control and lymphoma patients showed similar oxidative burst activity before treatment, the mean percentage of neutrophils undergoing oxidative burst was significantly lower after chemotherapy [56.73 ± standard deviation (SD): 21.67] in lymphoma patients than in controls (81.83 ± 10.60) (Figure 1A; P = 0.0022). Oxidative burst as measured by MFI was also significantly lower after chemotherapy (3.13 × 104 ± 1.51 × 104) compared to controls (4.72 × 104 ± 1.57 × 104) (Figure 1B; P = 0.029). Compared to values before chemotherapy (mean percentage = 81.20 ± 9.67; MFI = 5.090 × 104 ± 2.34 × 104), both mean percentage and MFI were significantly decreased after chemotherapy (Figure 1C; P = 0.0020 and Figure 1D; P = 0.020, respectively).
Figure 1.
A, B — Comparison of neutrophil oxidative burst in healthy control dogs and dogs with lymphoma before and 7 to 10 d after initial chemotherapeutic treatment as measured by percentage oxidative burst and oxidative burst mean fluorescence intensity (MFI), respectively. Phagoburst % indicates percentage of neutrophils undergoing oxidative burst. Phagoburst MFI indicates the mean oxidative burst activity on a single cell basis. C, D — Comparison of neutrophil oxidative burst in dogs diagnosed with lymphoma before chemotherapy and 7 to 10 d after initial chemotherapeutic treatment, as measured by percentage oxidative burst and oxidative burst MFI, respectively. P indicates P-value when comparing data sets.
Effect of chemotherapy on neutrophil phagocytosis in lymphoma patients
Both control and lymphoma patients showed similar phagocytic activity before treatment; however, the mean percentage of neutrophils with phagocytic activity (75.53 ± 15.32) was significantly lower after chemotherapy than in controls (92.59 ± 4.41) (Figure 2A; P = 0.0016). Furthermore, compared to values before chemotherapy (88.98 ± 7.31), the mean percentage of neutrophils with phagocytic activity was significantly lower after chemotherapy (Figure 2B; P = 0.014).
Figure 2.
A — Comparison of neutrophil phagocytosis in healthy control dogs and dogs with lymphoma before and after initial chemotherapy. Phagocytosis % indicates percentage of neutrophils undergoing phagocytosis. B — Comparison of neutrophil phagocytosis in dogs diagnosed with lymphoma before chemotherapy and 7 to 10 d after initial chemotherapeutic treatment. P indicates P-value when comparing data sets.
Effect of chemotherapy on oxidative burst in non-lymphoma cancer patients
As was seen with the lymphoma patients, compared to values before chemotherapy (mean percentage = 82.31 ± 14.04; MFI = 6.06 × 104 ± SD 2.46 × 104), both mean percentage oxidative burst and oxidative burst as measured by MFI were significantly decreased after chemotherapy (mean percentage = 71.80 ± 20.92; MFI = 5.22 × 104 ± 2.30 × 104) (Figure 3A; P = 0.00040 and Figure 3B; P = 0.020, respectively).
Figure 3.
A, B — Comparison of neutrophil oxidative burst in non-lymphoma tumor-bearing dogs before and 7 to 10 d after initial chemotherapeutic treatment as measured by percentage oxidative burst and oxidative burst mean fluorescence intensity (MFI), respectively. Phagoburst % indicates percentage of neutrophils undergoing oxidative burst. Phagoburst MFI indicates the mean oxidative burst activity on a single cell basis. C — Comparison of neutrophil phagocytic activity in non-lymphoma tumor-bearing dogs before and after initial chemotherapeutic treatment. Phagocytosis % indicates percentage of neutrophils undergoing phagocytosis. P indicates P-value when comparing data sets.
Effect of chemotherapy on phagocytic activity in non-lymphoma patients
As with the lymphoma patients, compared to pre-chemotherapy values (88.04 ± 8.66), mean percentage phagocytosis was significantly decreased after chemotherapy (84.15 ± 8.27) (Figure 3C; P = 0.029).
Correlation between neutrophil numbers and neutrophil function (oxidative burst and phagocytosis)
No significant correlation was found between neutrophil numbers and function in cancer patients before or after chemotherapy (Table II).
Table II.
Correlation between neutrophil numbers and neutrophil function.
| Neutrophil function and neutrophil numbers | Spearman’s correlation coefficient | P-value |
|---|---|---|
| Oxidative burst before chemo versus neutrophil count before chemo | Percentage: −0.50 | 0.12 |
| MFI: −0.07 | 0.74 | |
| Phagocytosis before chemo versus neutrophil count before chemo | Percentage: −0.30 | 0.17 |
| MFI: −0.1522 | 0.488 | |
| Oxidative burst 7 to 10 d after chemo versus neutrophil count after chemo | Percentage: −0.02 | 0.93 |
| MFI: 0.13 | 0.54 | |
| Phagocytosis 7 to 10 d after chemo versus neutrophil count after chemo | Percentage: 0.25 | 0.24 |
| MFI: 0.22 | 0.31 |
MFI — mean fluorescence intensity.
Discussion
This study builds on previous work describing neutrophil function in canine lymphoma patients (7). We expanded on that study to include patients diagnosed with other malignancies. Furthermore, unlike previous studies of canine cancer neutrophil function (7,8), primary and secondary antibodies were added to every blood sample analyzed, which allowed for a more exact gating strategy by ensuring that only neutrophils were gated during analysis of neutrophil function. To the authors’ knowledge, this is the first veterinary study to investigate the correlation between absolute neutrophil numbers and neutrophil function in canine cancer patients receiving chemotherapy.
Lymphoma patients showed significantly reduced neutrophil oxidative burst activity after chemotherapy compared to healthy controls and to values before chemotherapy. Our findings further support the initial data of LeBlanc et al (7), which showed decreased neutrophil oxidative burst 7 d after chemotherapy in canine lymphoma patients compared to values before chemotherapy. Although bactericidal activity of neutrophils was not directly measured in our study, decreased microbicidal activity for Staphylococcus aureus and E. coli has been partially associated with reduced neutrophil oxidative burst (5).
Similar to neutrophil oxidative burst, our study revealed reduced neutrophil phagocytic activity 7 to 10 d after chemotherapy in lymphoma patients compared to both healthy controls and to values before chemotherapy. This is in contrast with Leblanc et al (7), who reported no significant change in phagocytic activity from baseline values 7 d after chemotherapy when neutrophils were stimulated with E. coli. More precise gating strategies due to anti-canine neutrophil antibodies being added to all analyzed samples may have allowed for this significant difference to be identified in the present study. Studies assessing the effect of chemotherapy on neutrophil phagocytic activity in human cancer patients have had conflicting results. In some studies, decreased phagocytic activity was reported after chemotherapy (9–11), while others have reported no significant change in neutrophil phagocytosis (5,12). Different experimental designs may explain the disparity in the results.
Dogs with non-lymphoma malignancies also exhibited a significant decrease in neutrophil oxidative burst and phagocytic activity after versus before chemotherapy. This novel finding suggests that chemotherapy not only negatively affects neutrophil oxidative burst in lymphoma canine cancer patients, but also in dogs affected with other malignancies. Decreased neutrophil oxidative burst and phagocytosis have been described in a number of published studies in humans with different forms of malignancies (5,9–11,13). Larger-scale studies are needed to compare neutrophil function by specific tumor type and chemotherapy protocol.
No significant difference was noted in neutrophil function between control dogs and canine cancer patients before chemotherapy was started. LeBlanc et al (7) likewise found no such difference between control dogs and dogs with lymphoma before chemotherapy. In an earlier study, however, LeBlanc’s group had previously found that dogs with untreated carcinomas and sarcomas had lower percentages of neutrophils capable of oxidative burst compared to healthy dogs when neutrophils were stimulated with phorbol-myristate-acetate (PMA), while no significant difference was noted in oxidative burst and phagocytosis when neutrophils were stimulated with E. coli (8). As our findings replicated this study, it could be concluded that one reason for a significant difference not being noted in the control group and cancer group before treatment could be the method by which the neutrophils were stimulated in vitro.
The mechanism by which chemotherapy affects neutrophil function is not well-understood. Previous studies in human medicine suggest that immature neutrophils may have decreased oxidative burst more than mature forms (14). As chemotherapy may result in increased proportions of immature neutrophils in circulation, this could be a contributing factor to decreased oxidative burst after chemotherapy. As a significant left shift was not present in the cancer patients in our study, however, this is unlikely to contribute to the noted difference. Another proposed mechanism for decreased oxidative burst after chemotherapy is alteration of one or more components of the nicotinamide adenine dinucleotide phosphate (NADPH) oxidase complex or related regulatory proteins, which are collectively responsible for generating reactive oxygen species in neutrophils (5,15). Further studies are needed to determine the mechanism by which chemotherapy impedes oxidative burst in cancer patients.
Age was not a significant confounder in the present study, as there was no statistically significant difference between the median age of control dogs and tumor-bearing dogs. The effect of breed on neutrophil function was not explored. Another potential confounding factor was the use of concurrent medication. Decreased oxidative burst following steroid therapy has been reported in rats and humans (16,17). In our study, as steroids were part of the chemotherapeutic protocol in lymphoma patients, their use should be considered as part of the overall effect of chemotherapy on neutrophil function, rather than as a sole confounding factor. Since most cancer patients received concurrent medication to mitigate anxiety and side effects of chemotherapy, it would be unrealistic to exclude such patients in a clinical study. Nevertheless, the effect of concurrent medications on neutrophil function cannot be entirely excluded, although there is limited information on the effects of drugs on neutrophil function in dogs in general.
While most of the canine cancer patients in our study (22/26) did not have any other concurrent illness, the effect of other concurrent diseases such as cardiovascular and kidney disease on neutrophil function cannot be fully discounted, although these effects have yet to be studied. While no evidence of inflammation was found on CBC for any of the recruited patients, measuring acute phase proteins would have been helpful to further rule out a developing systemic inflammatory response.
No correlation was found between neutrophil numbers and function in our study, which suggests that there is no relationship between the two. In particular, there was marked variability in the drop in oxidative burst compared to the drop in neutrophil counts. This may partly explain why some canine cancer patients with marked neutropenia do not develop clinically significant infections, as their remaining neutrophils may function adequately. On the other hand, some patients with mild or marginal neutropenia may succumb to infectious disease, possibly due to inadequate functioning of remaining neutrophils. Prospectively, the clinical significance of decreased neutrophil function could be investigated on an individual patient basis. This information may be helpful in recognizing patients at higher risk of developing infection, at which time, consideration of prophylactic antibiotics could be warranted. A previous study in human patients with acute non-lymphocytic leukemia undergoing chemotherapy reported that patients with decreased neutrophil function before chemotherapy began were at higher risk of developing serious infection after chemotherapeutic treatment (18).
In conclusion, our results suggest that chemotherapy negatively affects neutrophil oxidative burst and phagocytosis in lymphoma and non-lymphoma canine cancer patients 7 to 10 d after chemotherapy. Results also suggest that neutrophil numbers cannot be used to predict neutrophil function. The clinical significance of the decrease in neutrophil oxidative burst and phagocytosis after chemotherapy has yet to be determined.
Acknowledgments
The authors thank Ben Elwood, Igor Moshynskyy, and Betty Chow-Lockerbie for technical support and the Companion Animal Health Fund of Western College of Veterinary Medicine for funding this project.
References
- 1.Baioni E, Scanziani E, Vincenti MC, et al. Estimating canine cancer incidence: Findings from a population-based tumour registry in northwestern Italy. BMC Vet Res. 2017;13:203. doi: 10.1186/s12917-017-1126-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.MacDonald V. Chemotherapy: Managing side effects and safe handling. Can Vet J. 2009;50:665–668. [PMC free article] [PubMed] [Google Scholar]
- 3.Gustafson DL, Page RL. Cancer chemotherapy. In: Withrow SJ, Vail DM, Page RL, editors. Withrow & MacEwen’s Small Animal Clinical Oncology. 5th ed. St. Louis, Missouri: Elsevier Saunders; 2013. pp. 157–179. [Google Scholar]
- 4.Bisson JL, Argyle DJ, Argyle SA. Antibiotic prophylaxis in veterinary cancer chemotherapy: A review and recommendations. Vet Comp Oncol. 2018;16:301–310. doi: 10.1111/vco.12406. [DOI] [PubMed] [Google Scholar]
- 5.Lejeune M, Sariban E, Cantinieaux B, Ferster A, Devalck C, Fondu P. Granulocyte functions in children with cancer are differentially sensitive to the toxic effect of chemotherapy. Pediatr Res. 1996;39:835–842. doi: 10.1203/00006450-199605000-00016. [DOI] [PubMed] [Google Scholar]
- 6.Fournier Q, Serra JC, Handel I, Lawrence J. Impact of pretreatment neutrophil count on chemotherapy administration and toxicity in dogs with lymphoma treated with CHOP chemotherapy. J Vet Intern Med. 2018;32:384–393. doi: 10.1111/jvim.14895. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.LeBlanc AK, LeBlanc CJ, Rohrbach BW, Kania SA. Serial evaluation of neutrophil function in tumour-bearing dogs undergoing chemotherapy. Vet Comp Oncol. 2015;13:20–27. doi: 10.1111/vco.12015. [DOI] [PubMed] [Google Scholar]
- 8.LeBlanc CJ, LeBlanc AK, Jones MM, Bartges JW, Kania SA. Evaluation of peripheral blood neutrophil function in tumor-bearing dogs. Vet Clin Pathol. 2010;39:157–163. doi: 10.1111/j.1939-165X.2009.00200.x. [DOI] [PubMed] [Google Scholar]
- 9.Bonatto SJ, Oliveira HH, Nunes EA, et al. Fish oil supplementation improves neutrophil function during cancer chemotherapy. Lipids. 2012;47:383–389. doi: 10.1007/s11745-011-3643-0. [DOI] [PubMed] [Google Scholar]
- 10.Gandossini M, Souhami RL, Babbage J, Addison IE, Johnson AL, Berenbaum MC. Neutrophil function during chemotherapy for Hodgkin’s disease. Br J Cancer. 1981;44:863–871. doi: 10.1038/bjc.1981.285. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Davies JE, Whittaker JA, Khurshid M. The effect of cytotoxic drugs on neutrophil phagocytosis in vitro and in patients with acute myelogenous leukaemia. Br J Haematol. 1976;32:21–27. doi: 10.1111/j.1365-2141.1976.tb01871.x. [DOI] [PubMed] [Google Scholar]
- 12.Lehrer RI, Cline MJ. Leukocyte candidacidal activity and resistance to systemic candidiasis in patients with cancer. Cancer. 1971;27:1211–1217. doi: 10.1002/1097-0142(197105)27:5<1211::aid-cncr2820270528>3.0.co;2-3. [DOI] [PubMed] [Google Scholar]
- 13.Hartmann P, Herholz K, Salzberger B, Petereit HF. Unusual and severe symptomatic impairment of neutrophil function after one cycle of temozolomide in patients with malignant glioma. Ann Hematol. 2004;83:212–217. doi: 10.1007/s00277-003-0802-2. [DOI] [PubMed] [Google Scholar]
- 14.Hara N, Ichinose Y, Asoh H, Yano T, Kawasaki M, Ohta M. Superoxide anion-generating activity of polymorphonuclear leukocytes and monocytes in patients with lung cancer. Cancer. 1992;69:1682–1687. doi: 10.1002/1097-0142(19920401)69:7<1682::aid-cncr2820690707>3.0.co;2-n. [DOI] [PubMed] [Google Scholar]
- 15.Nguyen GT, Green ER, Mecsas J. Neutrophils to the ROScue: Mechanisms of NADPH oxidase activation and bacterial resistance. Front Cell Infect Microbiol. 2017;7:373. doi: 10.3389/fcimb.2017.00373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Tsuji C, Shioya S. In vivo effect of methylprednisolone on lipopolysaccharide-induced superoxide production by pulmonary and circulating blood neutrophils in rats. Circ Shock. 1994;42:128–134. [PubMed] [Google Scholar]
- 17.Fukushima K, Ando M, Ito K, Suga M, Araki S. Stimulus- and cumulative dose-dependent inhibition of O2- production by polymorphonuclear leukocytes of patients receiving corticosteroids. J Clin Lab Immunol. 1990;33:117–123. [PubMed] [Google Scholar]
- 18.Hübel K, Hegener K, Schnell R, et al. Suppressed neutrophil function as a risk factor for severe infection after cytotoxic chemotherapy in patients with acute nonlymphocytic leukemia. Ann Hematol. 1999;78:73–77. doi: 10.1007/s002770050475. [DOI] [PubMed] [Google Scholar]