Hemangiosarcoma is a common mesenchymal malignancy often affecting larger breed dogs. This disease is characterized by a primary cavitated lesion in the viscera (spleen, heart, liver), muscular or subcutaneous tissue with a high propensity to spread early and often to other locations in the body such as the lungs, lymph nodes and other viscera(20, 21). Because this is a tumor of blood vessels, a sudden onset of hypovolemic shock due to acute spontaneous bleeding maybe the first clinical sign observed by the pet owner(24). Non-invasive diagnostics may be highly suggestive of hemangiosarcoma; however, it is very difficult to collect cytologic or biopsy samples of these tumors without a surgical approach in many cases due to the fragility of the tumor and the extraordinary amount of blood flow the tumors possess(21, 25). For this reason, new tests that can help direct client decision making without increasing the risk of acute hemorrhage are needed.
Though nucleosomes have DNA wrapped around them, they stand apart from other forms of circulating DNA through their pathophysiology. Nucleosomes play an active role in sepsis as part of neutrophil extracellular traps (NETs) which serve to remove extracellular pathogens from the blood(26). In cancer, they are suspected to play a role in metastasis and injections of nucleosomes into mice with xenografts led to increased metastatic lesions and new tumor manifestations(27). They have also been shown to enhance endothelial cell expression of interleukin 8 leading to neoangiogenesis in tumor tissue and ultimately tumor progression(28). Finally, they have also been shown to play a role in tumor cell immune evasion by inhibiting natural killer cell-mediated tumor lysis(29).
Though not specific for hemangiosarcoma, elevated nucleosome levels in the plasma of patients with hemangiosarcoma may provide additional information that can be useful to the client and veterinarian alike. In this cohort of 77 dogs, 80.5% of them (62 of 77) had mean plasma nucleosome concentrations above the 67.5 ng/mL designated cut off which is higher than the nucleosome levels of all 134 healthy dogs included in this study. A previously published manuscript evaluating the nucleosome compartment of healthy dogs determined that factors such as fasting, processing time, sample type and storage conditions can alter nucleosome concentrations in the plasma samples of healthy dogs [16]. For this reason, special care was taken in the selection and handling of these samples to ensure that all samples were handled the same way. When properly prepared, the plasma samples collected from normal dogs showed very little variability, similarly to what is seen in humans7.
The cohorts of both healthy dogs and those diagnosed with hemangiosarcoma were fairly diverse representing a variety of ages, breeds, body sizes and genders. However, nucleosome levels were unaffected by these factors (see supplement data). Only a diagnosis of hemangiosarcoma was able to differentiate between the two groups. Furthermore, the mean and median plasma nucleosome concentrations in the healthy dogs were surprisingly consistent across a variety of ages, body sizes and genders (supplemental data) with SEMs ≤ 5 ng/mL in all categories evaluated except body weight where dogs weighing more than 45 kg had an SEM of 14.7 ng/mL. These data demonstrate that the nucleosome compartment is consistently low in healthy dogs across the board.
In this study nucleosome concentrations varied by location; however, these results should be viewed with caution for all locations except splenic HSA due to the low numbers of cases in the other groups. Interestingly, 100% (5/5) of the primary bone HSA cases were elevated. A study evaluating 48 cases of osteosarcoma (OSA) by this group demonstrated elevated nucleosome concentrations in only 17/48 cases (35%) with the highest levels in those cases with metastasis (data not shown). More cases of primary osseous HSA are needed; however, this assay may be able to differentiate HSA from OSA when it is localized to the bone.
Nucleosome concentrations are predictive of stage in humans in gastrointestinal cancer, while there is no correlation to stage for other cancer types(30). In humans with lung and breast cancer, even the early stages of disease were found to have elevated serum nucleosome concentrations, similar to what was found here in dogs with hemangiosarcoma(31). In the present study, elevated nucleosome concentrations were detected in 67% of dogs with stage I disease, 76% of dogs with stage II disease and 90% of dogs with stage III disease. Stage I HSA is characterized by a solitary lesion that is < 5cm in diameter, no evidence of hemorrhage and no evidence of metastasis (locoregional or distant)(32). Stage II HSA is characterized by a lesion that is either 5 cm or greater or one that has ruptured with or without locoregional metastasis and no distant metastasis(32). These tumors may also be invading into the subcutaneous tissues. Stage III HSA is characterized by highly invasive tumors with or without distant metastasis or tumors of any size with distant metastasis (32). This ability to detect early stage disease can be particularly useful as a screening test for older dogs by possibly detecting hemangiosarcoma before there is an acute bleeding episode.
Like most types of cfDNA, nucleosomes are cleared from the plasma quickly, usually within an hour(3, 4). In healthy individuals’ nucleosomes are often cleared within minutes through endonuclease degradation, immunocomplexing with anti-nucleosome antibodies, phagocytosis and lysosomal degradation, hepatic metabolism or direct excretion through the kidneys [4]. However, in diseased patients, nucleosome elimination may be delayed when they bind to acute phase proteins (4). Though cancer is not the only cause of elevated nucleosome concentrations in humans or dogs, the short half-life of this type of cfDNA can provide a real time window into the amount of apoptosis or cell turn over in that patient at any given time. Elevations in plasma nucleosomes have been used in humans for monitoring response to therapy or recovery in a variety of diseases such as autoimmune disease, cancer, sepsis, viral infections and trauma (3, 30, 33–35). For this reason, plasma nucleosome concentrations may be a reliable and effective way to screen otherwise healthy canine patients at risk for hemangiosarcoma. Caution should be used; however, when evaluating plasma nucleosome concentrations of clinically ill dogs as nucleosomes are not specific for hemangiosarcoma or other common cancers.
Nucleosome concentrations have been used to monitor treatment response in rodents with HeLa xenografts as well as humans with cervical cancer and were found to reliably predict responders in both groups(36). Circulating nucleosome concentrations went down significantly after one cycle of platinum-based chemotherapy in the women with cervical cancer who had measurable responses. In patients with acute myeloid leukemia undergoing induction chemotherapy, nucleosome concentrations were an early predictor of treatment responders(37). Plasma nucleosome concentrations also correlated with response to therapy in patients with non-small cell lung cancer(38, 39). Here, we present a single case as a proof of concept that plasma nucleosome concentrations may serve as a surrogate for therapeutic response in patients with hemangiosarcoma. Though many more data points are needed before the utility of this test can be applied as a treatment monitoring tool for dogs with hemangiosarcoma, it is promising that nucleosome concentrations in dogs may mirror that which is seen in humans with a variety of cancers.
Other proteins, such as C-reactive protein (CRP) and Thymidine kinase have also been used to diagnose and monitor disease status in dogs with hemangiosarcoma. CRP is an acute phase protein that has been previously reported as a useful tool for evaluating remission status in canines with lymphoma(40). Thymidine kinase is an enzyme involved in pyrimidine synthesis and increases in extracellular TK activity could indicate the overall degree of DNA synthesis and dying cells. TK has also been reported as a useful marker of remission status in canine lymphoma patients(40). We evaluated both CRP and TK along with nucleosome concentrations in this patient as a proof of concept and found that CRP and TK values did not significantly change during the course of therapy and were not correlated with the clinical response in this one patient. Additional cases are needed to determine how effective nucleosomes concentrations are for disease monitoring when compared to CRP and TK.