Ovarian carcinomatosis in a dog managed with surgery and intraperitoneal, systemic, and intrapleural chemotherapy utilizing indwelling pleural access ports

Abstract

A 3-year-old Weimaraner dog was presented with bilateral papillary ovarian carcinoma and abdominal carcinomatosis. Treatment included ovariectomy, intraperitoneal cisplatin, and systemic carboplatin. Pleural carcinomatosis 473 days following surgery was treated with intrapleural cisplatin through indwelling pleural access ports. Euthanasia occurred 1154 days following surgery due to malignant pleural effusion without peritoneal effusion.

Ovarian neoplasia is rare in modern companion canine populations due to the high incidence of ovariectomy/ovariohysterectomy (1). In populations of entire female dogs it has been estimated that the prevalence of ovarian neoplasia is 6.25% (2). Of these cases approximately 50% are epithelial in origin and 64% of those that are epithelial in origin are malignant (3). In 1 study, 10 of 18 dogs with ovarian adenocarcinomas had metastatic disease, with local carcinomatosis being the most common metastatic manifestation (3).

There is no established standard treatment for canine ovarian carcinoma with regional metastasis (carcinomatosis) (1) and in many cases dogs are euthanized due to a perceived poor prognosis (4,5). In humans ovarian neoplasia is the second most common gynecological cancer and most patients have a high clinical stage of disease at presentation and a poor prognosis for long-term survival (6). The standard of care for human ovarian adenocarcinoma is cytoreduction of gross disease to < 1 cm thickness followed by the use of combination chemotherapy with a platinum compound and paclitaxel (6–8).

Case description

A 3-year, 3-month-old female intact Weimaraner dog was presented with a 5- to 6-day history of abdominal distension without change in demeanor or appetite. On examination, there was marked abdominal distension with a fluid wave and the rest of the examination was unremarkable. Abdominal sonography revealed a large volume effusion and bilateral ovarian masses. The left ovary was partially cavitated and measured 5.3 × 2.6 cm (Figure 1) while the right was homogenous and smaller, measuring 3.9 × 2.0 cm (Figure 2). Cytology of both the abdominal fluid and the ovarian masses was consistent with carcinoma. Thoracic radiographs were unremarkable as was an echocardiogram. A comprehensive biochemistry panel and a complete blood cell count were within normal limits.

Figure 1.

Sonographic image of the partially cavitated left ovarian mass.

Figure 2.

Sonographic image of the right ovarian mass. Note the surrounding echogenic free fluid.

On exploratory laparotomy 5 L of serosanguinous fluid were drained from the abdomen. The peritoneum was grossly thickened with widespread erythematous and nodular changes. The uterus was adhered to the colon and there was marked mesenteric lymphadenopathy, with the largest nodes estimated to have a maximum cross-sectional thickness of 10 to 20 mm. The bilaterally enlarged ovaries were removed via ovariectomy but rigorous resection of the generalized regional metastatic disease was not anatomically feasible. Histopathology confirmed bilateral ovarian papillary carcinoma with extensive vascular and lymphatic infiltration. The diagnosis of carcinomatosis was based on the clinical presentation, the gross appearance, the presence of carcinoma cells within the effusion, and extensive vascular infiltration on histology.

Three weeks following surgery urine culture was negative and serum creatinine was normal; further urinalysis was not performed. A blood smear was evaluated at this time with no abnormalities noted in abundance or morphology of any cell line. Abdominal sonography showed small amounts of generalized free fluid and a mildly irregular renal outline while the residual disease seen at surgery could not be identified. For 4 h prior to intraperitoneal (IP) treatment saline diuresis was performed with 18.3 mL/kg body weight (BW) per hour of intravenous saline (0.9% NaCl) solution. Pretreatment antiemetics were administered with maropitant (Cerenia; Zoetis, Silverwater, NSW, Australia), 1 mg/kg BW, SC 4 h prior to chemotherapy and 2 doses of butorphanol (Torbugesic; Zoetis) 0.4 mg/kg BW, IM, 4 h and 10 min prior to chemotherapy. An 18 G rigid IV catheter was placed through the surgically prepared ventral midline with the dog in lateral recumbency. A solution of cisplatin (DBL Cisplatin; Hospira Australia, VIC, Australia) was prepared in 1 L of room temperature 0.9% NaCl, using a closed system transfer device (PhaSeal, Becton, Dickinson and Company, New Jersey, USA) to a total dose of 70 mg/m². No fluid was removed before instillation of this solution into the peritoneal space over 5 to 10 min. The dog was rolled back and forth repeatedly, to distribute the fluid, then was briskly walked before a further 2 h of saline diuresis. On all occasions the intracavitary chemotherapy was tolerated without any apparent discomfort or nausea.

Complete blood cell counts showed a normal leukogram on day 7, and uncomplicated Veterinary Co-operative Oncology Group — Common Terminology Criteria for Adverse Events (VCOG-CTCAE) grade 3 (9) neutropenia [0.84 × 10⁹/L; reference interval (RI): 2.0 to 12.0 × 10⁹/L] on day 14. Three further doses of cisplatin were given at 3-weekly intervals following the same protocol but with a reduced dose of 52.5 mg/m². These were tolerated well, and VCOG-CTCAE grade 1 neutropenia (1.77 × 10⁹/L) was noted at day 20 following the fourth dose. Prior to each dose serum creatinine levels were evaluated, and remained within the reference interval. No gastrointestinal toxicities were observed.

Three doses of IV carboplatin (DBL Carboplatin; Hospira Australia) were given at 3-weekly intervals starting 26 d after the last intracavitary chemotherapy. Complete blood cell counts and serum creatinine levels were monitored before each dose and at 1 and 2 wk after treatment. Following 300 mg/m² carboplatin an uncomplicated VCOG-CTCAE grade 4 neutropenia (0.45 × 10⁹/L) was identified on day 16. Further doses were reduced to 225 mg/m² without further neutropenia. Uncomplicated VCOG-CTCAE grade 2 thrombocytopenia was noted at 16 d following the first treatment (75 × 10¹²/L; RI: 175 to 500 × 10¹²/L) and 14 d following the second treatment (92 × 10¹²/L). No platelet count was done after the third dose.

Brief abdominal sonograms at the time of the third dose of peritoneal cisplatin as well as the second and third doses of systemic carboplatin showed no evidence of free abdominal fluid. Monthly comprehensive abdominal sonography was performed for 3 mo following completion of the course of carboplatin and showed a very small volume of free abdominal fluid (on both occasions a single pocket of fluid measuring less than 1 × 1 cm) initially, which was undetectable by 3 mo following chemotherapy. In total, 15 abdominal sonograms were performed following completion of the initial course of chemotherapy, with the maximum period between imaging being 207 d. No sonograms identified peritoneal-free fluid or mesenteric lymphadenopathy for the remainder of the patient’s life, with the last abdominal imaging performed 1112 d after surgery.

At 473 d post-surgery the dog was presented with a recent history of expiratory wheeze. While an abdominal sonogram was normal at this time a bilateral pleural effusion was identified sonographically and was confirmed radiographically. No mass effect or nodular pattern was identified with either imaging modality. Thoracocentesis was performed and 560 mL of serosanguinous fluid was obtained. Fluid submitted for cytology showed malignant epithelial cells similar in morphology to cells in the previous carcinomatosis and consistent with a high probability of progressive disease.

Thirty-two days later intrapleural cisplatin therapy was initiated. Serum creatinine was within normal limits and urine sediment was inactive with a negative culture and a specific gravity of 1.026. Treatment with saline diuresis and maropitant was performed, as previously, while butorphanol was not administered. Bilateral indwelling pleural access ports (Canine PleuralPort; Norfolk Vet Products, Illinois, USA) were placed under general anesthesia. Effusion was drained through the ports; the volume was not recorded but was sonographically similar to the initial pleural effusion. Intrapleural chemotherapy was then performed under the same general anesthesia with a total dose of 52.5 mg/m² cisplatin divided into 2 volumes of 125 mL/m² of 0.9% NaCl each instilled into 1 hemithorax, using Huber needles (PosiGrip 22G Huber needle; Norfolk Vet Products, Skokie, Illinois, USA) to access the indwelling pleural ports. The patient was rocked following drug instillation and, following recovery from anesthesia, was taken outside for a brisk walk to aid drug distribution within the thoracic cavity.

VCOG-CTCAE grade 1 inappetence occurred during the first week after intrapleural cisplatin. Complete blood cell counts were monitored at weekly intervals for 3 wk with no myelosuppression noted. Two further doses of intrapleural cisplatin were given at 3-week intervals, without anesthesia but otherwise following the same protocol. Thoracocentesis was also performed on these occasions with 310 mL and 410 mL removed from the left hemithorax and no effusion noted on the right. While no adverse effects were noted, a self-limiting episode of presumed unrelated infectious tracheitis was diagnosed at an emergency clinic 10 d after the second dose.

A fourth dose of intrapleural cisplatin was intended but VCOG-CTCAE grade 1 azotemia developed (creatinine 172 mmol/L RI: < 159 mmol/L), with concurrent urine specific gravity of 1.027, after a total cumulative cisplatin dose of 385 mg/m². As thoracic and abdominal sonograms demonstrated no evidence of disease at this time, cisplatin was discontinued. Consideration was given to using further cycles of IV carboplatin as had been done previously. The dog tolerated IV carboplatin well at that time and renal function was normal; however, as carboplatin is renally excreted, it was considered that risk of side effects was now higher. Therefore considering the remission status, the potential for risk was considered greater than the potential for benefit and no further chemotherapy was administered at this time.

Monitoring by thoracic and abdominal sonography and aspiration via the ports showed no evidence of disease at rechecks 231 and 438 d following the first dose of intrapleural cisplatin.

At 607 d following the first pleural cisplatin the patient was presented with a 2-week history of lethargy, increased respiratory effort, reduced appetite, and coughing. Abdominal sonography showed no evidence of malignancy but thoracic radiographs showed left-sided effusion. Aspiration yielded 450 mL of fluid from the left hemithorax which cytologically revealed large numbers of neoplastic epithelial cells consistent with metastatic ovarian carcinoma. Repeated aspiration was required at 2- to 7-day intervals with similar volumes returned.

Twelve days following recurrence, creatinine was 164 mmol/L and 225 mg/m² carboplatin was administered into the left hemithorax in 250 mL of 0.9% NaCl through the indwelling pleural access port. Following a short improvement in clinical signs thoracocentesis was required at 15 and 18 d with 520 mL and 548 mL removed from the left hemithorax, respectively. Uncomplicated VCOG-CTCAE grade 3 neutropenia (1.07 × 10⁹/L) and uncomplicated VCOG-CTCAE grade 4 thrombocytopenia (17 × 10⁹/L) were identified on day 15.

After 3 wk, a second dose of carboplatin was given in the left hemithorax at a reduced dose of 180 mg/m². The effusion recurred after 1 wk and the patient was euthanized 1154 d after the date of the ovariectomy.

Discussion

Previously reported treatments for canine ovarian carcinoma are mostly limited to surgical resection of gross tumor tissue (1,3,4,10,11). Treatment of high-stage ovarian carcinoma in dogs with IP chemotherapy has been reported very rarely (10,12), with reports of only 3 cases of carcinomatosis treated in this way, only 1 of which was confirmed ovarian adenocarcinoma (10).

In 1 case a multi-agent systemic chemotherapy protocol containing cyclophosphamide, chlorambucil, and nitrosurea, resulted in at least 10 mo remission (13). In the single case of IP cisplatin as a treatment of canine ovarian carcinoma a second look celiotomy soon after 5 treatments of IP cisplatin suggested a complete remission, but the dog died acutely of hemoabdomen 2 mo later (10). Another study retrospectively reviewed the use of intracavitary carboplatin and/or mitoxantrone in dogs with carcinomatosis, sarcomatosis, or mesothelioma; however, there were no cases of ovarian carcinoma or peritoneal effusion (14).

While the evidence in veterinary medicine regarding intra-cavitary chemotherapy is sparse, this treatment remains part of the standard of care for high stage ovarian carcinoma in human medicine and accordingly there has been extensive research into maximizing its efficacy. Experimental animal research (15,16) and human trials (6,7,17,18) demonstrate the benefits of IP cisplatin therapy over IV chemotherapy in abdominal carcinomatosis. Cisplatin has been demonstrated to penetrate into the tissue to a greater depth than carboplatin when given as an IP dose (19) and for this reason cisplatin was selected for intracavitary chemotherapy in this case. Despite this, the penetration of cisplatin is only a few millimeters (15) limiting its use as a sole therapy. Several techniques have been developed to combat this problem including cytoreductive surgery prior to chemotherapy (6,8,20), the use of IV chemotherapy following IP treatment including cisplatin (21), and measures to increase the penetration of the cisplatin including hypotonic carrier fluids (22), instillation of adrenaline in the carrier fluids (23), regional hyperthermia (23–25), use of IV sodium thiosulfate to reduce systemic side effects and so allow the use of a higher IP dose (26), and high pressure abdominal lavage (27). To date none of these have been investigated in dogs. Reducing disease to < 1 cm rather than < 2 cm would intuitively be expected to be more efficacious; however, while this is suggested in some studies, a conclusive benefit has not been demonstrated (8).

In our case cytoreduction was performed but residual disease exceeded the recommended maximum thickness in human medicine of 1 cm, although it was less than a previously used cut-off of 2 cm for optimal cytoreduction (7,8). In 1 randomized study in humans there was a significant benefit to the use of IV cisplatin with cytoreduction to a maximum diameter of 2 cm (18). For this reason a combination of IP cisplatin followed by carboplatin IV was used to address the need for deeper tissue penetration resulting from suboptimal cytoreduction. It was not clear whether the addition of IV therapy was necessary and it is possible that the IP cisplatin reduced the tumor burden to the extent that complete penetration became possible during later treatments. Previous studies in humans have suggested that 5 or 6 cycles of platinum chemotherapy may be as effective as 10 or 12 cycles (28,29). In this case the initial protocol of 7 cycles allowed a short course of each IP and IV chemotherapies and kept the total number of treatments similar to these studies (28,29). Indeed the presumed recurrence of disease within the pleural space was controlled for 607 d with just 3 doses of intracavitary cisplatin, although it is not known whether this would have been the best possible outcome. The protocols used were tolerated extremely well clinically, though dose reductions were required for both drugs due to asymptomatic neutropenia with cisplatin and asymptomatic neutropenia and thrombocytopenia with carboplatin. This is of particular interest because a perceived high morbidity of intracavitary cisplatin in humans has reduced its uptake by oncologists (7).

This case also highlights potential use of pleural access devices for chemotherapeutic drug administration as well as symptomatic/palliative thoracocentesis with the ports remaining fully functional and without adverse effects for 649 d. The use of pleural access ports has previously been described in the veterinary literature (30) and these ports are reported by the manufacturer to be indicated for administration of chemotherapy agents (31). To our knowledge functional ports have not been reported for this duration in dogs and their use for intra-pleural chemotherapy has not been described in peer-reviewed veterinary literature prior to this article.

There is no well-established survival time for ovarian carcinoma in dogs, though the prognosis is considered poor in cases with metastatic disease (1). The outcome in this case compares favorably with average survival times for high-stage ovarian carcinoma in humans (6,17,18,20). This case demonstrates long-term resolution of a peritoneal malignant effusion with surgery and intracavitary and systemic chemotherapy and a long-term (607 d) remission from a malignant pleural effusion with intracavitary chemotherapy alone. It can only be speculated as to whether or not longer remission of the thoracic component of the disease could have been achieved if renal compromise had not curtailed the course of intrapleural cisplatin.

This case demonstrates the potential for long-term responses to intracavitary chemotherapy and may provide encouragement to clinicians in the face of otherwise sparse information in the literature regarding the management of ovarian carcinomatosis. Canine ovarian carcinoma with carcinomatosis should be considered a disease for which there may be reasonable palliative options. Further study is needed to define case selection and treatment protocols to optimize the role of intracavitary chemotherapy in canine carcinomatosis. This case report also highlights the potential for indwelling pleural access ports to be used as a convenient and safe method to administer chemotherapeutic agents. CVJ