Canine Intrapelvic Hemangiosarcoma in 12 Dogs (2000–2019): a Retrospective Case Series
ABSTRACT
Intrapelvic hemangiosarcoma (IP HSA) is an uncommon primary location of canine HSA, and its presentation appears to be different than that of other intra-abdominal HSA locations. The objectives of this study were to report on the clinical presentation, diagnostic options, and clinical outcomes of dogs with IP HSA. Medical records from two veterinary teaching hospitals between 2000 and 2019 were reviewed for dogs diagnosed with IP HSA. Inclusion criteria consisted of a cytologic or histopathologic diagnosis of HSA from a mass arising from the muscles of the pelvic canal. Exclusion criteria included masses arising from organs within the pelvic canal, including the urethra, prostate, and urinary bladder. Twelve dogs were ultimately included in the study. Clinical presentation of IP HSA was varied, with clinical signs including hindlimb lameness, urethral obstruction, and evidence of hemoperitoneum. The median progression free survival was 92 days (range: 1–1057 days). The overall survival time was 165 days (range: 1–1170 days). IP HSA is an underreported location of canine HSA with similar biologic behavior and outcome compared with visceral HSA. It serves as an important differential in patients presenting with new-onset lameness and urethral obstruction.
Introduction
Canine hemangiosarcoma (HSA) is an aggressive, locally invasive, and rapidly metastasizing tumor of blood vessel endothelium accounting for 0.3–2% of all canine tumors.1 Visceral HSA accounts for the majority of cases in dogs, up to 70% in some studies, with the spleen being the most common site.2 Additional common primary locations include the liver, right auricle, subcutaneous tissue, and the retroperitoneal space.3–5 Other less commonly reported nonvisceral locations include ocular, cutaneous, lingual, and intranasal. Because HSA is composed of mutated endothelial cells, it can arise essentially wherever there is a vascular supply. HSA is most commonly diagnosed in dogs 9–12 yr of age, and the breeds most commonly affected include golden retrievers, Labrador retrievers, German shepherd dogs, boxers, and schnauzers.6–8
In general, visceral HSA typically exhibits aggressive biological behavior with early distant metastasis. However, there is much variation in biologic behavior among visceral HSA, necessitating the differentiation and categorization of HSA by organ type for accurate prognostication.2 Splenic HSA is reported to have median survival times (MSTs) ranging from 19–86 days with surgery alone to 5–7 mo with the addition of chemotherapy, most often including doxorubicin.9–11 Primary cardiac and hepatic HSA share a dismal prognosis despite treatment. Conversely, of the visceral HSA subtypes, primary renal HSA may carry a slightly better prognosis with reported MSTs upward of 9 mo.12,13
Nonvisceral locations, including the dermis, subcutaneous tissue, tongue, intramuscular, and retroperitoneal spaces, appear to result in better outcomes. Again, as with visceral HSA, there is still much disparity in the survival times between these locations.7 Canine conjunctival and dermal HSAs can be cured with complete surgical excision with MSTs between 780 and 987 days for surgery alone for strictly dermal HSA versus much shorter survival times associated with strictly subcutaneous and intramuscular HSAs, which tend to act more aggressively.14–16 Reports on subcutaneous HSA provide inconsistent results with MSTs for surgery and adjuvant chemotherapy ranging from 9 mo to 3 yr. One explanation for the discrepancy in survival times between dermal and visceral HSA is likely in part because of the difference in inciting etiology, with dermal HSA suspected to be solar-induced and visceral arising from genetic mutations of pluripotent stem cells within the bone marrow.15,17
Primary intrapelvic HSA (IP HSA) is rarely reported in veterinary literature with only two reported cases, one dog and one horse.6,18 In these reports, the true origin of the mass was difficult to ascertain but was believed to be arising from the peripelvic vasculature based on surgical and necropsy findings.6,18 Based on these definitions, IP HSA would be categorized as a nonvisceral location arising from the peripelvic vessels and would therefore be hypothesized to have a better prognosis when compared with visceral HSA. Yet given the vast differences in outcome among and within the categories of visceral and nonvisceral HSA, evaluation of biologic behavior at each anatomic location is necessary. Here, we report the clinical presentation, treatment, and outcome of 12 dogs with primary IP HSA. Recognition of the clinical signs and diagnostic options for IP HSA, along with a better understanding of overall prognosis, will aid in diagnosis, treatment, and preparation of owners for clinical outcomes associated with this unique anatomic location of HSA.
Materials and Methods
Data Collection
The medical records of Texas A&M University Veterinary Medical Teaching Hospital and Colorado State University Veterinary Teaching Hospital from 2000 to 2019 were interrogated for cases that fit the following inclusion criteria: an IP sarcoma as the primary lesion with cytologic or histopathologic confirmation of HSA, necropsy or case description information, and outcome. A total of 12 dogs fit the criteria to be evaluated for this retrospective case series.
Medical record information collected included signalment, pertinent history, laboratory data, diagnostic imaging, cytology, histopathology, treatment performed (if any), progression free survival (PFS), and overall survival. Because of the small number of cases, dogs undergoing definitive therapy, regardless of whether microscopic disease was obtained with clean surgical margins, were grouped together for outcome information. For this group, progression free survival was defined as time from definitive initial treatment until disease progression in all cases.
Clinical staging was performed based on the modified World Health Organization staging criteria for canine HSA.16 Staging diagnostics consisted of complete blood count, serum biochemistry panel, abdominal ultrasound, thoracic radiographs, and/or abdominal and/or thoracic computed tomography (CT).
Statistical Methods
All data were recorded using Microsoft Excel softwarea (version 16.27). GraphPad Prismb was used to generate medians and Kaplan-Meier Survival curves. Three patients (Dogs 7, 9, and 12) were censored at the time of their last medical visit because of the lack of follow-up and outcome data on these cases. Statistical comparisons between treatment protocol were not performed because of the low numbers of dogs in any group.
Results
Patient Population
The population consisted of 12 dogs (Table 1), 6 of whom were females and 6 were males, resulting in a 1.0 female-to-male ratio. The entire population was altered. Represented breeds included golden retriever (n = 2), mixed breed (n = 3), boxer (n = 2), and one of each of the following breeds: schnauzer, Shetland sheepdog, Labradoodle, greyhound, and Bedlington terrier. The median age at the time of diagnosis was 11.5 yr. The median body weight was 22 kg, ranging from 7.4 to 39.6 kg. The weights of mixed-breed dogs ranged from 16 to 39.6 kg.
Pertinent signalment, presenting complaint, and physical exam findings are summarized in Table 1. Presenting complaints included stranguria (n = 3), diarrhea/dyschezia (n = 3), pelvic limb pain/weakness/discomfort (n = 3), hematuria (n = 1), hematochezia (n = 1), and shortness of breath (n = 1). The mass was palpated by the owner or on routine physical exam in four dogs (Dogs 2, 5, 6, and 10), none of whom had any related clinical signs. In Dog 1, the mass was discovered on CT scan for staging 28 days after splenectomy for two small splenic masses, both of which were histologically confirmed to be HSA. In Dogs 5 and 6, a caudal abdominal mass was incidentally palpated on routine physical exam before the annual dental cleaning. Dog 7 presented to the referring veterinarian for pelvic limb weakness. Abdominal radiographs revealed a hepatic mass. The IP mass was not discovered until surgery for the hepatic mass and was not noted on staging CT scan of the abdomen 2 days before. Dog 2 presented for additional diagnostics after an incidental pulmonary mass was found on thoracic radiographs for a workup of shortness of breath, with the IP mass being discovered incidentally on abdominal CT scan. Based on the appearance of the pulmonary nodule on thoracic CT scan and the radiographic interpretation, it was suspected to be a metastatic lesion from the primary IP HSA.
Abdominal palpation on physical exam revealed a palpable mass in six patients, abdominal discomfort but no palpable mass in two patients, and the absence of abdominal discomfort or a palpable mass in three patients. A mass was palpable on rectal palpation in five dogs, not palpable in six dogs, and not reported in one dog. Hindlimb weakness or lameness was noted in two dogs. Additional abnormal physical exam findings included one each of the following: heart murmur, inability to urinate, a single enlarged prescapular lymph node, and cranial organomegaly on abdominal palpation. Dog 12 displayed no abnormal physical exam findings on presentation.
Clinicopathologic abnormalities at diagnosis are presented in Table 2 and were categorized as mild, moderate, and severe. These categories were based on similar criteria for grade 1, grade 2, and grade 3 toxicities, respectively, as outlined by the Veterinary Cooperative Oncology Group.
Although the intent of the Veterinary Cooperative Oncology Group-Common Terminology Criteria for Adverse Events grading scheme is to categorize the severity of adverse events to treatment, the authors employed similar parameters in this study as a means of standardization of parameter deviation because the study population was derived from two separate institutions with different reference ranges. Of the 10 patients for whom packed cell volume (PCV) was evaluated, 3 had a mild, nonregenerative anemia. Complete blood counts were performed in nine patients; one had a mild mature neutrophilia, one had a moderate mature neutrophilia, and two had mild thrombocytopenia.
Serum biochemistry panel results were available for 10 patients. Four dogs had mild to moderate elevations in liver enzymes, two dogs had mildly elevated lactic acid levels, one dog had a mildly elevated globulin, and two dogs had mildly elevated albumin. The remaining values were not clinically significant. None of the hematologic abnormalities were consistent among patients or prognostically significant. Coagulation panels were not consistently performed on each patient.
Staging diagnostics for each patient are outlined in Table 2. Briefly, of the six dogs for whom thoracic radiographs were performed, five were within normal limits. The only dog with thoracic radiograph abnormalities was Dog 2, who presented for shortness of breath to the primary veterinarian, at which time a pulmonary mass was noted on thoracic radiographs. The caudal abdominal mass was incidentally palpated by the owner before presentation to the referral institution.
Six patients had abdominal ultrasounds performed at the tertiary institution. In five dogs, a caudal abdominal mass was noted with sizes ranging from 0.45 cm in diameter to 14.0 × 10.0 × 9.9 cm. The tissue of origin was unidentified in four of the dogs. In Dog 11, the mass was suspected to be arising from the prostate, and in Dog 12, the mass was suspected to be a pericolic lymph node. In Dog 5, the caudal abdominal mass was palpated on physical exam; however, only a 7.0 × 4.9 cm splenic mass was noted in the original ultrasound report. The caudal abdominal mass was later discovered on abdominal CT scan before splenectomy and excised surgically. Dog 2 had evidence of a peritoneal effusion with a PCV of 27% with a peripheral PCV of 39%, suggestive of a hemoabdomen. This was the only dog with a confirmed hemoabdomen. Aside from ill-defined splenic nodules in two dogs, the remainder of the studies were unremarkable.
Four dogs had both thoracic and abdominal CT scans performed, whereas one dog (Dog 2) had only a thoracic CT scan, and four dogs had only an abdominal CT scan. In seven dogs, a heterogeneous, soft-tissue-to-fluid-attenuating, rim contrast enhancing caudal abdominal mass was noted, ranging from the smallest dimension of 2.6 cm to the largest dimension of 15.0 cm, resulting in displacement of the local viscera (Figure 1). In Dog 7, an enlarged, mineralized pericolic lymph node was noted but was later confirmed to be HSA after surgical removal and histopathology. Three dogs had splenic nodules, two of which were suspected metastasis from the caudal abdominal mass because of the number and size of nodules present in the spleen compared with the IP mass. Two dogs had enlargement of a single intra-abdominal lymph node, with only one having measurements available. Both were suspected to be metastatic based on size and appearance when compared with the IP mass. Dog 6 had evidence of an intramuscular metastatic nodule within the left epaxial muscle (Figures 1C, D). Three dogs had hepatic nodules with benign differentials prioritized based on CT appearance. Dog 5 displayed a region of geographic lysis of the L1 vertebral body, with the differentials being an age-related change or an aggressive bony lesion, including possible metastasis. The cranioventral peritoneal mass noted in Dog 9 was suspected to be an omental cyst, nodular fat necrosis, lymphadenopathy, or metastatic neoplasia and was later confirmed to be metastatic HSA from the primary IP mass upon surgical excision and histopathology. The soft-tissue pleural nodules on thoracic CT of Dog 10 were suspected to be pleural thickening, mesothelial hypertrophy, or neoplasia (metastatic, mesothelioma). Lastly, the hypoattenuating lesion of L2 in Dog 12 was prioritized to be a benign lesion (hemangioma, fatty replacement) or—less likely—metastasis. Thoracic CT revealed evidence of pulmonary metastatic disease in three of the five dogs, and a large heterogeneous mass within the right caudal lung lobe of Dog 2 in addition to the smaller pulmonary nodules suspected to be metastasis.
Citation: Journal of the American Animal Hospital Association 58, 1; 10.5326/JAAHA-MS-7142
Two dogs had echocardiograms, which revealed American College of Veterinary Internal Medicine stage B1 mitral valve disease in Dog 1 and was unremarkable in Dog 7. Nine dogs had fine-needle aspirates of the caudal abdominal mass performed, none of which experienced obvious immediate or long-term complications from the procedure. Cytologic samples were diagnostic for a sarcoma, suspect HSA, in five dogs, and nondiagnostic or hemodiluted in four dogs. Of the diagnostic samples, the cells were oval to polygonal to spindloid with scant to moderate amounts of basophilic cytoplasm and an eccentrically located nucleus. There were frequent multinucleated cells and nucleoli ranged from 1 to 4.
IP HSA was confirmed in 10 of the 12 cases via histopathology, 1 via biopsy of the caudal abdominal mass, 8 via histopathology of the surgically excised mass, and 1 at necropsy. Tumors were encapsulated and unencapsulated, densely cellular with polygonal to spindloid cells with indistinct cell borders, moderate homogeneous eosinophilic cytoplasm, and marked anisocytosis and anisokaryosis. Mitotic index ranged from 2 to 38, with a mean of 12 per 10 high-powered fields. Dog 1 was initially diagnosed with splenic HSA when two small (<5 cm) cavitated masses were seen via ultrasound at the referring veterinarian’sclinic. The spleen was removed and sent off for histopathologic diagnosis. The initial presenting complaint for this dog was difficultly climbing stairs, hindlimb weakness, and difficulty sleeping characterized by whining and restlessness at night. This did not resolve after surgery, and the dog was referred to a specialist 4 wk after the surgery once the biopsy results were available. At the time of the visit to the specialist, a rectal exam revealed a large and painful pelvic mass. A CT scan performed at that time revealed the large cavitated IP mass, a heterogeneous contrast-enhancing hepatic mass, multifocal intra-abdominal lymphadenopathy, and pulmonary nodules, all of which were presumed metastasis from the primary IP mass. This mass was determined to be the cause of the initial presenting complaints of difficulty climbing stairs and restlessness secondary to pain. A second case without histopathologic confirmation (Dog 3) was diagnosed via cytology alone. In this case, primary consideration was given to HSA by the clinical pathologist based on the presence of a cavitated mass within the IP canal resulting in bony lysis and the cytologic evidence of marked hemodilution, cytologic appearance of the cells, and evidence of previous hemorrhage. Cytologically, the majority of nucleated cells were spindled to stellate mesenchymal cells with a moderate amount of cytoplasm, variable N:C ratio, few mitotic figures, and rare binucleation.
Dog 3 was also euthanized shortly after presentation to the emergency service for the inability to urinate and ribbon-like stools. She had also been dragging her right hind limb over the last several weeks. On physical exam, there was a palpable firm mass within the caudal abdomen, and she demonstrated mild right hindlimb lameness. A thoracic and abdominal CT (Figures 1A, B) scan revealed a large cavitated mass within the right pelvic canal as well as suspect pulmonary metastasis. Cytology of the mass was consistent with sarcoma, most likely HSA, and humane euthanasia was elected.
Dog 4 was euthanized shortly after presentation with no additional diagnostics. This patient presented through the emergency service of the referral hospital for the inability to urinate. She had evidence of stranguria and hematuria for 4 mo before presentation. Upon physical exam, the dog’s abdomen was turgid, and a mass was palpated ventrally on rectal examination. An abdominal focused assessment with sonography for trauma revealed a large cavitated mass, and the owners elected humane euthanasia. Necropsy confirmed the IP mass to be HSA without an obvious organ of origin. This was the only patient with a necropsy performed.
Clinical Stage
Staging was based on the modified scheme of the World Health Organization clinical staging system for canine HSA, yielding one patient with stage I, six with stage II, and five with stage III disease. Corresponding stages are displayed in Table 2. Three patients (Dogs 5, 7, and 12) had evidence of locoregional metastasis (N1) without distant metastasis (M0). Metastasis was confirmed in Dogs 5 and 7 via histopathology of locoregional lymph nodes extirpated during surgery, and metastasis was suspected in Dog 12 because of the enlargement of a splenic lymph node (1.4 cm thickness). Dog 1 had both locoregional nodal and distant splenic, hepatic, and pulmonary metastases; all except the pulmonary metastases were confirmed via histopathology of the respective locations at the time of surgery. Four dogs had no evidence of locoregional metastasis but either confirmed or suspected distant metastases. Dogs 2 and 3 had evidence of pulmonary nodules on thoracic CT, Dog 6 had evidence of a suspected intramuscular metastasis on abdominal CT, and Dog 9 had hepatic and omental metastasis confirmed via a wedge biopsy and surgical extirpation, respectively. The splenic mass noted on abdominal CT for Dog 5 was consistent with a splenic hematoma, with no evidence of neoplasia.
Treatment
Treatments for the 12 dogs are outlined in Table 1. Ten of the 12 dogs received some form of treatment, whereas the remaining 2 (Dogs 3 and 4) were euthanized at the time of diagnosis. Diagnosis was confirmed at necropsy for Dog 4, whereas cytology consistent with a sarcoma and presence of a cavitated mass in the IP space led to the presumptive diagnosis of IP HSA in Dog 3. Eight of the 10 dogs treated underwent surgery via a midventral laparotomy with 6 achieving complete surgical excision. The pelvis was undisturbed in all cases, and no perioperative complications were reported. None of the patients undergoing surgical excision required a blood transfusion intra- or postoperatively. In Dog 11, it was discovered intraoperatively that the suspected prostatic mass was not arising from the prostate, which appeared normal at surgery. Six of the 10 dogs treated received chemotherapy, one of whom was a dog that had complete surgical excision, as follows: 4 of the 6 received single-agent doxorubicin, 1 received single-agent carboplatin, and 1 received metronomic therapy consisting of cyclophosphamide and piroxicam. Two dogs also received palliative radiation therapy for pain associated with bony invasion. Dog 1 had a palliative course of 4 by 8 Gray to a total of 32 Gray delivered in weekly fractions. Dog 11 received a similar palliative protocol in conjunction with metronomic cyclophosphamide. Additional therapies included propranolol (n = 2) and zoledronic acid (n = 1).
Survival Time
All 12 dogs were included in the overall survival analyses, with three dogs being censored at the date of the last hospital visit. Only dogs receiving some form of treatment (n = 10) were included in the PFS, again with three dogs being censored. Two untreated dogs were euthanized at the time of diagnosis. The median PFS was 92 days (range: 1–1057 days) (Figure 2A). The overall survival time (OST) was 165 days (range: 1–1170 days) (Figure 2B).
Citation: Journal of the American Animal Hospital Association 58, 1; 10.5326/JAAHA-MS-7142
Discussion
IP HSA is a rare form of HSA in dogs. Although sparsely reported in veterinary literature, with only two case reports existing, this variant arises from the peripelvic vasculature of the pelvic canal and not from the organs within the pelvic canal, such as the rectum, bladder, or prostate.6,18 Thus far, only a single case report of canine IP HSA exists in veterinary literature, and to the authors’ knowledge, there are no reports of human IP angiosarcoma.6 This case series reports the presenting complaints, treatment modalities, and outcome of 12 dogs diagnosed with IP HSA from two academic teaching institutions over a period of 19 yr.
The patient characteristics within this retrospective study population had similarities and differences when compared with previous reports of visceral HSA. Unlike other studies suggesting a possible male predisposition, there appeared to be no apparent sex predilection in this group.7,19 Because of the limited sample size, conclusions on breed predisposition for IP HSA were unable to be drawn, yet the population did include some breeds known to be predisposed to HSA, including two golden retrievers, two boxers, and one schnauzer. The median age of diagnosis of 11.5 yr is also comparable to that of visceral HS, A being between 9 and 12 yr of age, and the median weight of 22 kg suggests that this disease also occurs in large-breed dogs.6–8 None of the mixed-breed dogs in this study were toy or miniatures, with weights of the mixed-breed dogs ranging from 16.0 to 39.6 kg.
Presenting complaints of the patients in this study were distinctly different from those typical of visceral HSA, particularly splenic HSA. The most common presenting clinical signs for visceral HSA resulting in hemoabdomen are weakness and lethargy, followed by pallor, anorexia, abdominal enlargement, and depression.20,21 These clinical signs are often episodic and assumed to be associated with intratumor or intracavitary bleeding events.21,22 In contrast, the most common presenting complaint for clinical patients in this case series was stranguria, followed by dyschezia and pelvic limb pain/weakness, making the argument for consideration of IP HSA as a differential diagnosis in patients presenting with such clinical signs. CT imaging confirmed urethral obstruction in two patients and invasion of the pelvic bones in two patients. Similarly, the one existing case report of canine IP HSA described a 12 yr old female spayed schnauzer with a 1 wk history of constipation.6
In four patients, the IP mass was discovered incidentally by the owner or on routine physical examination with abdominal palpation, advocating the importance of a thorough physical exam. The mass was readily palpable during rectal examination of five dogs. Dog 1 presented to its primary veterinarian for hindlimb weakness and was diagnosed with two small (<0.5 cm) splenic nodules via abdominal ultrasound and was ultimately taken to surgery for splenectomy 1 mo before diagnosis of an IP mass. Both abdominal ultrasound and surgery failed to identify the IP HSA that was later discovered when the patient underwent an abdominal CT for continued clinical signs. Given the presence of two small splenic nodules compared with the large size of the IP mass and the fact that the original presenting complaints of hindlimb weakness, reluctance to use the stairs, and restlessness at night were due to the IP mass, it is highly likely that the IP mass was the primary lesion in this patient.
In a study of 39 cases of canine nontraumatic hemoabdomen, Pintar et al. determined neoplasia to be the most frequent cause (80%), with 70% of these cases being due to ruptured visceral HSA.20 The present case series contained only one patient (Dog 2) with the presence of an active hemoabdomen at the time of diagnosis characterized by a small amount of fluid noted on the CT scan. This patient presented to the primary veterinarian for shortness of breath and was referred to the tertiary institution for staging for a pulmonary mass discovered on thoracic radiographs performed by the referring veterinarian. The IP mass and hemoabdomen were incidentally discovered on abdominal CT scan performed for staging of the pulmonary mass. The PCV of the abdominal fluid was 27%, with a peripheral PCV of 39%.
Another distinction between this patient population and that with splenic HSA is the presenting clinicopathologic abnormalities. Approximately 50% of dogs with visceral HSA have evidence of disseminated intravascular coagulopathy, with mild thrombocytopenia being the most common laboratory finding.11 Furthermore, both thrombocytopenia and neutrophilia have been documented as negative prognostic factors for dogs with splenic HSA.23 In the present study, only one dog was found to have a mild thrombocytopenia combined with mild anemia (PFS = 84 days; OST not available), whereas one dog demonstrated mild neutrophilia (time to progression was not available; OST = 165 days), and two dogs had a moderate neutrophilia (time to progression = 84 and 92 days; OST of latter dog = 94 days).
In the present study, 50% of patients had evidence of metastasis at the time of diagnosis. Four of the 12 dogs (33.3%) presented with locoregional metastasis and 3/12 dogs (25.0%) with distant metastases were identified via a combination of thoracic radiographs, thoracic CT, abdominal ultrasound, abdominal CT, and whole-body CT. Only one of these dogs had evidence of both locoregional and distant metastasis at the time of diagnosis. Evidence of enlarged locoregional lymph nodes and pulmonary nodules not evaluated via histopathology were considered to be metastatic lesions in this study as to not underestimate metastatic potential of IP HSA. This finding is within the range of metastatic rates displayed by splenic HSA ranging from 29 to 90% in previous studies and in contrast to other locations that have a lower propensity for metastasis, such as renal, cutaneous, and conjunctival HSA.1,13,14,24 In addition, one dog (Dog 3) had intramuscular metastases within the epaxial muscles on abdominal CT scan. HSA is the most common sarcoma subtype to result in skeletal muscle metastases (SMMs).25 In a retrospective study of 61 dogs, SMMs were identified in 24.6% of dogs with HSA, all of which had additional metastatic lesions within the liver, spleen, and omentum. The only clinical sign reported in 60% of these patients with SMMs was lameness not attributable to orthopedic or neurologic disease, whereas the remaining 40% were asymptomatic.19 The patient in the present study did not display evidence of lameness, likely because of the location of the SMM being the epaxial muscles. Conversely, the four dogs that displayed hindlimb lameness, weakness, or discomfort did not have overt evidence of SMMs; however, the entirety of the limbs of these patients were not included in the imaging studies and therefore may have been missed. Additionally, the hindlimb lameness, weakness, or discomfort in these cases can be potentially explained by the IP HSA causing bony lysis (Figures 1A, B). Given the high propensity for SMM in cases of HSA along with the presenting complaints of lameness, there is argument for whole-body CT or PET-CT scans in the staging of all patients with HSA, including those with IP HSA.
Because of the retrospective nature of this study and the lack of advanced imaging diagnostics, such as a whole-body CT scan, for all patients, it is possible that the rate of metastasis of IP HSA is underestimated. As demonstrated by the previously mentioned patients, abdominal ultrasound alone may not be sufficient for diagnosis of an IP mass based on its actual location in the pelvic canal. Abdominal ultrasound would not appropriately stage a patient in the case of intramuscular metastases, thereby further highlighting the utility and importance of CT as a standard component of staging in cases of IP masses.
In addition to the similar metastatic potential, IP HSA shares similarities in the PFS and OST when compared with splenic HSA. The PFS for splenic HSA corresponds to that of the overall PFS for dogs in this study. In this study, the overall PFS was 92 days regardless of treatment modality (Figure 2A). In a study comparing low-dose chemotherapy with a standard single-agent doxorubicin protocol, Lana et al. documented a PFS of 126 days for patients with splenic HSA treated with doxorubicin alone.26 Similarly, intramuscular HSA and a subset of subcutaneous HSA patients have been shown to have an overall PFS of 116 days.27 Based upon the similarities in biologic behavior and survival time with splenic HSA, it is presumed that IP HSA would respond similarly in a large population; however, future studies should focus on doxorubicin treatment in a larger sample size of patients with IP HSA. Fifty percent of dogs in this case series received some form of chemotherapy, including single-agent doxorubicin (n = 4) and metronomic cyclophosphamide (n = 2). Dog 11 received one dose of carboplatin that was not well tolerated because of gastrointestinal toxicity and was switched to metronomic cyclophosphamide. Because of the small sample size, statistical analysis was unable to determine the significance in PFS with surgery alone versus surgery with adjuvant chemotherapy and/or palliative radiation therapy.
In this retrospective study, the OST was 165 days (range: 1–1170 days) (Figure 2B). This is comparable to the OST of 179 days in dogs with visceral HSA treated with either surgery alone or surgery with adjuvant single-agent doxorubicin.2 One explanation for the short survival time may be related to the fewer number of cases of IP HSA; however, it should be noted that rare cases in this study population achieved long-term disease control. This, along with the fact that IP HSA may not be as readily recognized by clinical signs or on routine diagnostics and procedures, including abdominal ultrasound or surgery, is likely to lead to a delay in diagnosis. Because of the retrospective nature of this case report, sufficient information regarding duration of clinical signs before presentation was lacking. The small sample size of the present study precluded OST calculation for surgical removal alone compared with surgical excision with adjuvant chemotherapy. However, it is well known that adjuvant chemotherapy, often including doxorubicin, increases MST to 5–9 mo in dogs with visceral HSA.2,23 Additional studies with a larger number of patients in each treatment category (no treatment versus surgery versus surgery with adjunctive chemotherapy) are needed to better determine the effect of such treatment on survival time in IP HSA cases.
Mitotic index and nuclear pleomorphism have previously been shown to be of prognostic value for HSA from various anatomic locations.7,10,23 Of the reported mitotic indices in this study, there was a range from 2 to 38, with a mean of 12 per 10 high-powered fields. One long-term survivor (Dog 11) had a small mass that was found incidentally and was removed completely as a stage I tumor with a mitotic index of 3, suggesting that early identification and intervention may result in prolonged survival times. This patient later died of osteosarcoma. Lack of sufficient data and sample size prohibited prognostication calculations based on mitotic index.
Limitations
Limitations of the current study are due mostly to the retrospective nature of the cases and include a heterogeneous patient population with variations in tumor size, stage, and treatment modality, as well as completeness of medical records in some patients. The small sample size of the population was also a contributing factor and precluded certain statistical analyses. Although the present study population consists of relatively few cases over a period of 2 decades, it provides an in-depth description of a rarely reported, yet important, disease process. The hope of the authors is that this case series will serve as an impetus for future work on this unique location of HSA and will contribute to a growing body of literature regarding clinical presentation and prognostic factors.
Conclusion
This case series described the presenting complaints, diagnostic findings, treatment, and outcome of 12 dogs with primary IP HSA. Given the findings, it highlights the importance of a thorough physical exam and the benefits of a full-body CT for patients presenting with disruption of normal urination and/or defecation and those with hindlimb signs, such as weakness or pain. Overall, the meta-static rate of this variant of HSA was similar to that previously described for splenic HSA and appears to have a similar OST. Prospective studies involving IP HSA are required for further characterization of prognostic factors, optimal treatment modalities, and outcomes for dogs with this rare but clinically important anatomic variant of HSA.
(A, B) Dog 3: A 9 yr old female spayed greyhound presenting for a 1 day history of stranguria and dyschezia and ribbon-like stool. (A) Transverse section of pelvis demonstrating intrapelvic mass resulting in severe compression of the rectum and lysis of the right medial ilial cortex. (B) Sagittal section exhibiting the intrapelvic mass. (C, D) Dog 6: A 13 yr old male castrated schnauzer presenting for evaluation of a caudal abdominal mass palpated on physical examination. Retrospectively, the owners noted the patient was reluctant to jump on furniture. (C) Transverse section of left epaxial muscle metastasis. (D) Sagittal section of intrapelvic mass found incidentally on abdominal palpation during routine physical examination.
(A) Kaplan-Meier PFS curve for 10 dogs with intrapelvic HSA because 2 dogs were euthanized at the time of diagnosis. Survival times were calculated from the day of diagnosis. Median time to progression was 92 days, range 1–1057. (B) Kaplan-Meier OST curve for 12 dogs with intrapelvic HSA. Three patients (Dogs 7, 9, and 12) were censored at the time of their last medical visit because of a lack of follow-up and outcome data on these cases. Median survival time was 165 days, range 1–1170. HSA, hemangiosarcoma; OST, overall survival time; PFS, progression-free survival.
Contributor Notes
From the Department of Small Animal Clinical Sciences (K. L. B., C. R. D., H.W.-R.), College of Veterinary Medicine & Biomedical Sciences (K. N. H.), Texas A&M University, College Station, Texas; and Surgical Oncology, Flint Animal Cancer Center, Colorado State University, Fort Collins, Colorado (B. W.-J.)
