Masitinib-associated minimal change disease with acute tubular necrosis resulting in acute kidney injury in a dog.

A 10-year-old spayed female Labrador Retriever was brought to the Heart of Texas Veterinary Specialty Center for additional treatment of metastatic mast cell tumor after failing vinblastine,11 APP Pharmaceutical, LLC, Schaumburg, IL lomustine,22 CeeNu, Bristol-Myers Squibb, Princeton, NJ chlorambucil,33 Leukeran, GlaxoSmithKline, Research Triangle Park, NC and toceranib phosphate44 Palladia, Pfizer Animal Health, New York, NY treatment. The original mast cell tumor was grade I, but recurrent lesions at the site of previous resection were reported to be grade III. On examination, the dog was bright, alert, and responsive. Four cutaneous mast cell tumors were present near the site of the incision of the original tumor resection (right lateral flank), and the right iliac lymph node was enlarged. The only abnormality noted on a CBC, biochemical profile, and urinalysis was an increased serum alkaline phosphatase activity (366 U/L; reference range, 23–212 U/L). Serum creatinine concentration was 1.4 mg/dL (reference range, 0.5–1.8 mg/dL). Urine specific gravity was 1.020, and a urine dipstick protein test was negative. Masitinib,55 Masitinib mesylate, AB1010, AB Science, Paris, France a tyrosine kinase inhibitor (TKI), was started at the recommended dosage (12.5 mg/kg PO q24h). Monitoring was performed as recommended by the manufacturer (every 2 weeks for hypoalbuminemia and proteinuria; every 4 weeks for azotemia, anemia, neutropenia, increased aspartate aminotransferase or alanine aminotransferase activity, and hyperbilirubinemia). At 2 weeks, serum albumin concentration and urine protein:creatinine ratio (UPC) were within reference intervals. After 4 weeks of treatment, serum albumin concentration was normal (3.0 g/dL; reference range, 2.2–3.9 g/dL) and minimal proteinuria was noted (UPC, 0.6; reference range, <0.5). All other variables remained within their reference intervals. Two weeks later, testing identified marked proteinuria (UPC, 20.8) and severe hypoalbuminemia (1.2 g/dL). Masitinib was immediately discontinued. Within 24 hours, the patient began vomiting and showed signs of cranial abdominal discomfort. Results of a CBC and biochemical profile performed at the referring veterinary hospital showed increased urea nitrogen concentration (52 mg/dL; reference range, 7–27 mg/dL), but similar serum creatinine concentration (1.5 mg/dL) as observed previously. These findings were thought to be consistent with gastrointestinal bleeding. In addition, hyperphosphatemia (7.1 mg/dL; reference range, 2.5–6.8 mg/dL) and hypoalbuminemia (1.0 g/dL) were present. The dog was transferred to the Heart of Texas Veterinary Specialty Center where she was found to be lethargic and severely edematous. Abnormalities on abdominal ultrasound examination included heterogeneous echogenicity of the pancreas and a small amount of free peritoneal fluid. The dog was treated with lactated Ringers solution66 Lactated Ringer's Solution, Baxter Healthcare Corporation, Deerfield, IL (30–40 mL/kg/d), hydroxyethyl starch77 Hydroxyethyl starch, Braun Medical, Allentown, PA (20 mL/kg/d), aspirin88 Aspirin USP, Professional Compounding Center of America, Houston, TX (0.5 mg/kg PO q24h), benazepril99 Benazepril Hydrochloride, Amneal Pharmaceuticals of New York, Hauppauge, NY (0.5 mg/kg PO q24h), ranitidine1010 Ranitidine, Bedford Laboratories, Bedford, OH (1 mg/kg IV q12h), and sucralfate1111 Sucralfate, Nostrum Laboratories, Kansas City, MO (1 g PO q8h). Despite treatment, she developed acute oliguric renal failure and became anuric despite administration of diltiazem1212 Diltiazem, Baxter Healthcare Corporation, Deerfield, IL (3–5 μg/kg/min IV, constant rate infusion), mannitol1313 Mannitol, Hospira, Incorporated, Lake Forest, IL (0.56 g/kg IV q4h for 3 treatments), and furosemide1414 Furosemide, Intervet International GmBH, Germany (0.5 mg/kg/h IV, constant rate infusion). The dog continued to deteriorate, and a biochemical profile obtained 2 days after transfer disclosed azotemia (urea nitrogen concentration 110 mg/dL; serum creatinine concentration 5.4 mg/dL), hyperphosphatemia (>16.1 mg/dL), hypoalbuminemia (1.1 g/dL), and increased serum alkaline phosphatase activity (363 U/L). Urinalysis on urine collected by catheterization showed a dark yellow to orange color with urine specific gravity 1.024, 4+ proteinuria, and 3+ blood. On sediment examination, >100 red cells/high power field and 5–10 white cells/high power field were observed. The UPC was 12.9. The dog also was hypertensive (systolic blood pressure, 170 mmHg). Euthanasia was performed because of the dog's deteriorating clinical condition and unsatisfactory response to treatment. The owner declined full necropsy; but, kidneys were sampled for pathologic evaluations immediately after death. No gross lesions were observed during a brief evaluation of the abdominal organs. Samples of kidney and urine were submitted to the Texas Veterinary Renal Pathology Service for analysis by light microscopy, immunofluorescence, and transmission electron microscopy (TEM), and urinary protein electrophoresis was performed. Tissue for light microscopy was sectioned at 3-μm thickness and stained routinely with hematoxylin and eosin, periodic acid Schiff, Masson's trichrome, and Jones methenamine silver stains. The unfixed tissue sample for immunofluorescence evaluation was frozen and embedded in Optimal Cutting Temperature compound.1515 Tissue-Tek OCT Compound, Sakura Finetek USA, Torrance, CA After cryosectioning, sections were routinely labeled with an appropriate dilution of fluorescein isothiocyanate-labeled anti-IgG,1616 Bethyl Labs, Montgomery, TX anti-IgM,1616 Bethyl Labs, Montgomery, TX anti-IgA,1616 Bethyl Labs, Montgomery, TX anti-C1q,1717 Dako North America, Carpinteria, CA anti-C31616 Bethyl Labs, Montgomery, TX , anti-Lambda light chains,1717 Dako North America, Carpinteria, CA and anti-Kappa light chains1717 Dako North America, Carpinteria, CA antibodies. All slides were examined with an epifluorescence microscope1818 Olympus, Center Valley, PA using appropriate filters. Tissue for electron microscopy was sectioned and evaluated by the authors (CB and FC) at the University of Georgia and Texas A&M University, respectively. Urine supernatant was frozen at −80°C until analysis using nonreducing sodium dodecyl sulfate polyacrylamide gel electrophoresis. The urine was normalized for urine specific gravity using the formula 0.065/(urine specific gravity-1), and the sample and commercial standards1919 Mark 12™ Protein Standard, Life Technologies, Grand Island, NY were loaded onto a gel.2020 NuPAGE® 4–12% Bis-Tris gel, Life Technologies, Grand Island, NY Gels were stained2121 Imperial Protein Stain, Thermo Fisher Scientific Inc, Rockford, IL and imaged.2222 Gel Doc™, Bio-Rad, Hercules, CA Images were analyzed using BioNumerics,2323 BioNumerics, Applied Maths, Austin, TX with automated band detection using densitometric curves and manual verification using the dried gel. Protein bands within the sample then were assigned a molecular weight based on the standards. Light microscopic examination indicated that the glomerular tufts were normal, but there was mild-to-moderate dilatation of Bowman's capsules and the parietal epithelium was hypertrophied and vacuolated. These parietal epithelial cell abnormalities were segmental in nature, and the cells nearest the urinary pole were more severely affected than those nearer the hilus (Fig 1A). Furthermore, there was diffuse proximal tubular epithelial vacuolar degeneration with multifocal acute tubular injury consisting of multiple singly necrotic epithelial cells in tubular cross-sections (Fig 1B). A few tubules were more severely affected; these tubules were dilated, lined by attenuated epithelial cells, and contained sloughed cellular debris, hyaline protein casts, or both. Arcuate caliber arteries had evidence of mild acute intramural hemorrhage associated with apoptosis of smooth muscle cells of the tunica media (Fig 1C). Multiple small (non-occlusive) intravenous thrombi were present in veins (Fig 1D). Patchy acute interstitial inflammation was present in the cortex and corticomedullary junction, often associated with necrotic tubules or intravenous thrombi. Immunostaining evaluation was negative, showing no evidence of immune-complex deposition or complement activation. On TEM, marked global podocyte activation and injury consisting of severe cellular swelling, global foot process effacement, and microvillus transformation were observed (Fig 1E). Consistent with the immunofluorescence results, no immune-complex deposits were identified by TEM. In addition, vacuolar degeneration and necrosis of proximal tubular epithelial cells was confirmed by TEM (Fig 1F). Taken together, the global effacement of podocyte foot processes with normal-appearing glomerular tufts were diagnostic of minimal change disease (MCD), as reported for humans and veterinary species.1, 2 Urinary protein electrophoresis identified the presence of many low molecular weight and high molecular weight bands, consistent with both tubular (low molecular weight) and glomerular (high molecular weight) proteinuria3 (Fig 2). The number and intensity of bands on the gel were consistent with moderate-to-severe glomerular and tubular damage, although mild alterations of the banding pattern attributable to the presence of hematuria could not be excluded. Notably, the observation of a prominent 14 kilodalton (kDa) band was consistent with the presence of hemoglobin, as seen in grossly hematuric samples (MBN, personal observations). Mild to moderate hematuria will not result in presence of these bands. Masitinib is a TKI that primarily targets the tyrosine kinase receptor c-Kit, and it may target platelet-derived growth factor receptor.4 In a randomized, double-masked, placebo-controlled phase III trial of masitinib in >200 dogs with nonresectable grade II or III mast cell tumors, overall response rate was not significantly different between placebo and masitinib-treated dogs (15% versus 16%).5 However, there was a significant difference in time to progression between the 2 groups (75 versus 118 days). This study suggests that masitinib does have biologic activity against mast cell tumors. Masitinib's most common adverse effects include anorexia, vomiting, diarrhea, and gastrointestinal bleeding. Mild problems are treated with supportive care; temporary cessation of drug administration is instituted for substantial toxicity. A less common, but potentially serious, adverse effect of masitinib is induction of a proteinuric nephropathy.6, 7,2424 Masitinib mesylate [package insert]: AB Science The mechanism of renal damage has not been fully elucidated. Impairment of renal function may be related to a direct effect of masitinib on renal tubules, which have been shown to express c-Kit,8 or on glomerular cells that express platelet-derived growth factor receptors.9 Masitinib rarely was associated with life-threatening or fatal hypoalbuminemia in field studies and in the 39-week safety study.2424 Masitinib mesylate [package insert]: AB Science Neither proteinuria nor hypoalbuminemia was associated with kidney lesions on light microscopic evaluation of histologic sections. The package insert2424 Masitinib mesylate [package insert]: AB Science states, “The studies provide evidence that severe adverse reactions may be prevented if dogs are monitored for hypoalbuminemia every 2 weeks.” The insert suggests that the clinician should monitor biweekly for hypoalbuminemia and proteinuria, and every 4 weeks for azotemia, anemia, neutropenia, increased aspartate aminotransferase or alanine aminotransferase activity, and hyperbilirubinemia. Cessation of treatment is recommended if either hypoalbuminemia or proteinuria (UPC >1) occurs. However, based on the present case, it may be more prudent to recommend a baseline UPC and increased monitoring if the UPC increases substantially above baseline, even if <1. Two cases of well-documented drug-induced MCD in veterinary species (1 dog and 1 cat) have been reported in the literature, and both animals were being treated with a TKI.2, 10 Specifically, the dog was receiving masitinib mesylate for allergic skin disease and the cat was being treated with imatinib mesylate for hypereosinophilic syndrome. Although the dog had an increased serum creatinine concentration when it was first evaluated for drug-induced nephropathy, previous testing had documented mild azotemia before initiation of treatment. Furthermore, masitinib was not associated with a further increase in serum creatinine concentration for up to 6 months after treatment. The cat was only mildly and briefly (on 1 occasion) azotemic during or after imatinib administration. Therefore, ours is the 1st case report of TKI-associated MCD and acute tubular necrosis (ATN) resulting in proteinuria and acute kidney injury (AKI) with substantial azotemia in a dog or cat. This case illustrates the need for EM and immunofluorescence evaluation in patients with proteinuria, AKI, and histologically normal glomeruli, because the diagnosis of MCD would otherwise not be possible. Because ultrastructural evaluation of glomeruli of proteinuric dogs was not performed in previous safety studies, the true incidence of TKI-associated MCD is unknown. Urinary protein electrophoresis also can provide evidence for glomerular and tubular damage, and in this case, it supported the theory that both glomerular and tubular injury were clinically relevant because the banding pattern corresponded well with the glomerular damage on TEM and the severity of tubular changes observed histologically. Grossly, the dark yellow to orange color of the urine supported mild discoloration caused by the hematuria observed microscopically. Although this degree of blood contamination could have influenced the UPC and electrophoresis results, the banding pattern observed on the gel was consistent with that typically seen in uncontaminated samples from dogs with proteinuric renal disease, with the exception of the approximately 14 kDa band that previously has been identified as hemoglobin in our laboratory and has only been observed by the authors in samples containing substantial hematuria. Interestingly, AKI caused by ATN in the setting of MCD is not uncommon in humans. In fact, up to 25% of cases of adult-onset MCD are concurrently diagnosed with ATN-induced AKI.11 Predisposing factors for human patients with concurrent AKI and MCD are older age, higher systolic blood pressure, and more severe proteinuria.11, 12 Renal biopsy results indicate ATN characterized by tubular epithelial simplification along with more severe arteriosclerosis in human AKI/MCD patients compared with patients having MCD alone.12 As expected, the diagnosis of AKI in the setting of MCD warrants a worse prognosis for long-term renal function compared with MCD alone. Human patients with both syndromes had higher serum creatinine concentrations after resolution of proteinuria,11 and some patients required hemodialysis.12 Furthermore, the previous 2 cases of TKI-associated MCD without AKI in animals had resolution of all clinical signs after cessation of TKI use. The dog of the present report, however, had rapid progression of renal disease. The pathogenesis of MCD with ATN in humans is unknown, but unsampled arterial thromboemboli, acute onset of severe hypoalbuminemia, and proteinuria-induced tubular injury all have been postulated as causes.11, 12 The clinical course of the current case (ie, abrupt onset of severe proteinuria before the onset of AKI) supports the latter theory. Although thrombi were present in the dog of this report, their venous location and small size suggest that they were secondary to the renal disease as opposed to being its cause. Direct nephrotoxic effects of masitinib on the proximal tubular epithelial cells also may have played a role. The arterial mural hemorrhage and apoptosis of smooth muscle cells are rare lesions in the kidneys of small animals, and their pathogenesis and clinical relevance is currently unknown. If they represent the histologic manifestation of arterial vasospasm, they might have played a role in the pathogenesis of the ATN by induction of transient ischemia. Alternatively, they may merely be incidental lesions in this patient with severe AKI. This case report is the 1st detailed description of masitinib-induced MCD and AKI in a dog or cat. It disputes the widely held assumption that renal lesions caused by TKIs are reversible and raises questions about what type and frequency of monitoring is sufficient to promptly detect and minimize adverse effects when masitinib is administered. It also demonstrates a need to find a way to prevent or minimize the risk of this potentially fatal adverse effect. The study was not supported by a grant or presented at a meeting. The authors acknowledge the help of Ralph Nicholes at the Texas Heart Institute for transmission electron microscopy support. Conflict of Interest: Authors disclose no conflict of interest.