Hemorrhagic Cystitis in a Dog Receiving Carboplatin

Introduction

Oncology patients are at risk for developing hemorrhagic cystitis for various reasons, including toxic chemical agents, thrombocytopenia, radiation, or infection secondary to myelosuppression.¹ Sterile hemorrhagic cystitis has been reported in dogs receiving cyclophosphamide with either acute or delayed onset of clinical signs.^2–4 Common presenting complaints include hematuria, stranguria, dysuria, and pollakiuria. The clinical signs that result from cyclophosphamide are not a direct toxicity of the compound but rather of its toxic metabolites, in particular acrolein, although the exact mechanism of action is still unknown.¹ Hemorrhagic cystitis following treatment with carboplatin has been reported in the human literature; however, to the authors’ knowledge, has not been reported in the veterinary literature.^5–11 The authors of the case reports in human patients with carboplatin-associated hematuria and/or renal failure propose the potential need for hydration prior to carboplatin administration to avoid hemorrhagic cystitis. By reporting this case, the study authors wish to make other veterinarians aware of the potential for hemorrhagic cystitis to occur. The authors propose that clinicians using carboplatin judiciously monitor for development of hemorrhagic cystitis because recording those events will provide data such that a better idea of the true incidence of hemorrhagic cystitis can be determined.

Case Report

An 8 yr old castrated male Labrador retriever mixed-breed dog was referred to the Veterinary Medical Centre, Western College of Veterinary Medicine, for treatment options for osteosarcoma (OSA) of the left proximal humerus. Following amputation, the dog received three doses of carboplatin^a. The protocol at the Veterinary Medical Centre involved four to six 21 day cycles of carboplatin at a dose of 300 mg/m² IV beginning within 10–14 days following amputation. The premixed aqueous solution of 10 mg/mL carboplatin was administered undiluted over 15 min. Ten days after the third treatment, the dog presented on an emergency basis with a 1 day history of hematuria, stranguria, and pollakiuria. A free-catch urinalysis was submitted. The urine specific gravity was 1.047; pH was 5.0; there was 2+ protein, 1+ bilirubin, 4+ blood; and no glucose and ketones. The sediment contained 0–3 white blood cells/high-power field, 25–40 red blood cells/high-power field, 0–2 epithelial cells with few clumps, scant struvite and bilirubin crystals, 0–2 granular casts/low-power field, and negative bacteria. Urine was submitted for culture and sensitivity, and a complete blood count was normal. Serum biochemical analysis was performed with no abnormalities except mild hypercholesterolemia (cholesterol was 7.57 mmol/L; reference range, 2.70–5.94 mmol/L). As the dog was receiving myelosuppressive therapy, antibiotics were started (amoxicillin^b 22 mg/kg per os q 12 hr) pending urine culture results and were discontinued when the culture was found to be negative.

No improvement in clinical signs was seen after 24 hr of therapy. The dog returned for further diagnostics. Abdominal radiographs were unremarkable. An abdominal ultrasound showed an irregularly thickened urinary bladder wall cranioventrally. No bladder masses were visualized, and no uroliths were present. The urine contained a moderate amount of floating debris. Those findings were considered more consistent with cystitis than with neoplasia.

The following day, cystoscopy was performed. Again, no masses were visualized but the bladder wall appeared thickened and edematous. While anesthetized for cystoscopy, attempts to biopsy bladder mucosa were unsuccessful. Instead, a traumatic catheterization of the trigone area was performed. Cytologic evaluation of the catheterized urine sample revealed many clusters of pleomorphic transitional epithelial cells, characterized by large size and variable nucleus/cytoplasm ratio with abundant, deep basophilic, finely granular cytoplasm and blebbed cytoplasmic margins. Nuclei were round and oval with coarsely stippled chromatin and occasional prominent nucleoli. Moderate to marked anisocytosis and anisokaryosis were noted in the population (Figure 1). Those findings were interpreted as possibly representative of transitional cell carcinoma; however, nonmalignant inflammation of the bladder urothelium could also cause those morphologic changes. The dog was treated with meloxicam^c (0.1 mg/kg per os q 24 hr) for pain, and clinical signs associated with cystitis subsided over a period of 4 wk. Meloxicam was administered daily for 7 days, tapered over the next 3 wk, and then stopped. Clinical signs of cystitis did not recur when meloxicam was discontinued. Meloxicam was subsequently used for joint pain on an as needed basis. Carboplatin was removed from the chemotherapy treatment plan and was substituted with doxorubicin^d (30 mg/m² 21 days after the last carboplatin treatment). The dog received one treatment of doxorubicin with no adverse effects. Further doxorubicin treatments were recommended but declined by the owner.

View Figure

• Download as Powerpoint
• Download Figure

A follow-up abdominal ultrasound was performed 5 mo after the initial diagnosis. The urinary bladder, as well as all other abdominal organs, appeared grossly normal. At the time of the last follow-up examination (48 mo postdiagnosis of OSA), the dog was alive with no evidence of metastatic disease and no signs of cystitis.

Discussion

Platinum analogs are commonly used in the treatment of canine OSA either alone or in combination with doxorubicin.^12–19 Because of the toxicities of renal damage and severe nausea and vomiting seen with cisplatin, other platinum compounds were developed.²⁰ Carboplatin has similar activity in canine OSA and appears to provide equivalent clinical effectiveness with survival times ranging from 262 days to 540 days.^12–19 Carboplatin is not emetogenic and is significantly less nephrotoxic than cisplatin, thus negating the need for diuresis and antiemetic therapy.²⁰ Nevertheless, carboplatin is associated with other toxicities, for which diligent monitoring is still indicated.

Acute renal failure due to carboplatin is rare in humans and, to the authors’ knowledge, has not been reported in the veterinary literature.^5,6 Hemorrhagic cystitis is considered to be even rarer in humans. Indeed, a literature review reveals only a single case of presumed recurrent hemorrhagic cystitis in a patient receiving high-dose carboplatin.⁷ The patient described in this report was thrombocytopenic during the first episode, but on the second occasion the platelet count was normal. The patient did not have cystoscopy performed so the diagnosis could not be confirmed. Nevertheless, the complication was reported as hemorrhagic cystitis.

Two other case reports are very similar in that patients with ovarian carcinoma and no history of renal disease developed gross hematuria after receiving carboplatin.^8,9 Both patients had normal platelet counts, and diagnostic tests revealed bilateral ureteral obstruction secondary to blood clots that caused renal failure.

A woman with locally advanced breast cancer received one cycle of chemotherapy with carboplatin, docetaxel, and trastuzumab.¹⁰ She presented 2 days after completion of chemotherapy with a history of gross hematuria. Blood work and urine cultures were normal. Results of computed tomography and ultrasound of the abdomen and pelvis were normal. Cystoscopy was also normal with no evidence of hemorrhagic cystitis. A retrograde pyelogram did not show either any tumor or filling defects. Neither trastuzumab nor docetaxel are known to cause hematuria. Thus, carboplatin was removed from the protocol and the patient’s hematuria resolved.

In contrast to those reports, a phase 2 study of single-agent carboplatin for testicular seminoma reported one patient suffered from hematuria, but the authors of that study did not believe there was a likely relationship to either the tumor or carboplatin.¹¹ However, as the aim of that study was efficacy rather than toxicity, no further investigation was performed to identify a causal relationship between carboplatin and hematuria; therefore, a possible toxicity could not be ruled out.

Authors believe that platinum analogs in general and carboplatin in particular can be toxic to transitional epithelium, causing sloughing and bleeding.⁸ Although the pathogenesis of such an effect is unclear, it may be similar to that for the metabolites of cyclophosphamide, ifosfamide, and other related compounds that are also well known to be toxic to the urothelium.

Conclusion

In the patient discussed in this report, a presumptive diagnosis of hemorrhagic cystitis was made based on clinical signs, urinalysis, and cytologic analysis of a traumatic catheterization sample. As there was no immediate indication for biopsy, it was not performed on the patient to further confirm cystitis histologically. Even though hematuria is not a recognized complication of carboplatin therapy, there are several case reports in the human literature. In some of those cases, other chemotherapy drugs were used concomitantly that may have had an idiosyncratic interaction leading to that particular toxicity. However, carboplatin was used as a single agent in some of those cases, so it is more likely that it was the underlying cause. The clinical history, imaging results, and urinalysis findings in the canine patient described in this report are similar to those seen in dogs with either cyclophosphamide-induced cystitis or an occult urinary tract infection. However, the history and work-up ruled out both processes as contributors to the hematuria in this patient, leaving carboplatin administration as the sole cause of hemorrhagic cystitis in this case, especially because the signs resolved following discontinuation of carboplatin therapy and treatment with nonsteroidal anti-inflammatory drugs. Authors of case reports involving carboplatin-associated hematuria and renal failure in humans propose the potential necessity for aggressive hydration, as is typically recommended for cisplatin and cyclophosphamide, as a means of avoiding hemorrhagic cystitis. Further monitoring for that type of toxicity in canine patients is warranted to help determine whether diuresis is also indicated in canine patients receiving carboplatin.