Pegylated Liposomal Doxorubicin Causes Kidney-limited Thrombotic Microangiopathy

Abstract

A definite causal link between pegylated liposomal doxorubicin (PLD) and renal-limited thrombotic microangiopathy (TMA) remains unestablished. Here, we report two cases of PLD-induced renal-limited TMA, one in a patient with myxofibrosarcoma and the other in a patient with liposarcoma. The two patients received a high cumulative dose of PLD, and both presented with a rise in serum creatinine and proteinuria. Kidney biopsy revealed TMA with chronic mesangiolysis and capillary wall double contouring. Neither patient had concomitant exposure to TMA-causing drugs, such as gemcitabine, anti-vascular endothelial growth factor agents, or mammalian target of rapamycin (mTOR) inhibitors. The workup for secondary causes of TMA was negative in both patients. The cessation of PLD therapy led to improvement or stabilization in serum creatinine and proteinuria in both patients. These two cases provide a clear causal link between PLD and renal-limited TMA. The high cumulative dose of PLD increases the risk of renal TMA. Early recognition of PLD-induced renal TMA can lead to timely cessation of PLD therapy and potentially preserve kidney function.

Introduction

Doxorubicin is an antineoplastic agent used to treat several malignancies.¹ Pegylated liposomal doxorubicin (PLD) is a liposomal formulation of doxorubicin, in which doxorubicin is coated with polyethylene glycol and contained in a liposome. The polyethylene glycol coating prolongs the plasma half-life of PLD by preventing its degradation by the reticuloendothelial system.² The alteration in the structure and pharmacokinetic profile of PLD attenuates the risk of cardiotoxicity and myelosuppression observed with the parent compound doxorubicin.^3,4 Furthermore, the prolonged half-life of PLD in the vascular compartment enhances its delivery and accumulation in tumor tissue.⁵ PLD’s improved efficacy and side effect profile allows for its prolonged use, resulting in high cumulative drug exposure.

Thrombotic microangiopathy (TMA) is a group of syndromes characterized by microvascular thrombosis, microangiopathic hemolytic anemia, thrombocytopenia, and end-organ damage, including kidney injury. Renal-limited TMA can occur due to underlying malignancy or anticancer therapies.⁶ Renal-limited TMA has been reported with PLD use, but in these cases, patients had concomitant exposure to TMA-causing drugs, such as gemcitabine, anti-vascular endothelial growth factor (VEGF) agents, or mammalian target of rapamycin (mTOR) inhibitors.^7–11 Here, we describe two cases of renal-limited TMA in patients with sarcoma following prolonged PLD therapy and without concomitant exposure to TMA-causing drugs.

Case Reports

Case 1

A 72-year-old woman was diagnosed with high-grade myxofibrosarcoma in the left thigh. She underwent tumor resection, but two years later, developed tumor recurrence and underwent a second resection. Two years following the second tumor resection, she developed metastatic disease in the lungs. She underwent lung wedge resection and initiated chemotherapy. She was treated with several anti-cancer agents, including gemcitabine, docetaxel, olaratumab, pazopanib and pembrolizumab. She developed metastatic disease in the left kidney and underwent a left radical nephrectomy four years ago. The left nephrectomy tissue showed normal glomeruli with minimal tubular atrophy and interstitial fibrosis. Following the left nephrectomy, she received six cycles of pembrolizumab. A year after undergoing nephrectomy, she developed a lesion in the right kidney and a peri-hilar mass consistent with metastatic disease. She received six cycles of doxorubicin for a cumulative dose of 450 mg/m2, followed by 11 cycles of PLD for a cumulative dose of 440 mg/m2. At the end of 11 cycles of PLD, her radiologic scans showed improvement with complete resolution of the metastatic lesion in the right kidney, so PLD was stopped. One year after stopping PLD, she was noted to have new lung metastatic lesions, so PLD was reinitiated. She received eight cycles of PLD for a cumulative dose of 320 mg/m2. Her total PLD exposure over 2.5 years was 760 mg/m2.

One month after receiving the last dose of PLD, she was seen in the nephrology clinic for evaluation of elevated creatinine and proteinuria. Her creatinine was 1.8 mg/dL, and her urine protein-to-creatinine ratio (UPCR) was 2400 mg /g. The patient had a history of hypertension and chronic kidney disease (CKD). Her baseline creatinine was 0.9–1.1 mg/dL, and baseline UPCR was 600 mg/g. The cause of CKD was attributed to hypertension and reduced renal mass due to a previous unilateral nephrectomy. She was not on renin-angiotensin system (RAS) inhibitors due to history of a severe allergic reaction to ace inhibitors. Home medications included amlodipine, alendronate, and omeprazole. The rise in serum creatinine, from 1.1 to 1.8 mg/dL, occurred over the preceding 12 weeks. Physical examination revealed lower extremity edema and high blood pressure (BP) of 153/72 for which furosemide was prescribed. Laboratory investigations are summarized in Table 1. Kidney biopsy showed chronic TMA with extensive mesangiolysis and capillary wall double contour formation. No acute thrombi were seen within the glomeruli or arteries. The glomeruli were mildly enlarged but no proliferative or sclerotic glomerular features were observed. Electron microscopy (EM) showed glomerular basement membrane duplication with cellular interposition, but no deposits were observed (Figure 1, panels A, B). PLD was stopped and three months following the cessation of PLD, serum creatinine remained stable at 1.8 mg/dL, UPCR declined from 2400 mg/g to 750 mg/g, and BP improved to 127/67. At the time of the last follow-up, patient was awaiting radiologic scans to assess the status of underlying malignancy.

Table 1.

Laboratory Investigations

	Patient 1	Patient 2	Reference Range
Hemoglobin, g/dL	10.4	7.9	11.2–15.4
WBC count, ×109/L	5.4	6.3	4.0–11.0
Platelet count, ×109/L	113	233	160–400
Reticulocyte count, ×109/L	66	37	36–111
Haptoglobin, mg/dL	159	151	40–240
Lactate dehydrogenase, U/L	195	317	130–250
Sodium, mEq/L	135	139	133–143
Potassium, mEq/L	4.5	3.9	3.3–4.9
Bicarbonate, mEq/L	26	22	18–29
Creatinine, mg/dL	1.8	2.0	0.6–1.1
eGFR, ml/min/1.73m2	30	24	≥60
SUN, mg/dL	38	39	6–20
Calcium, mg/dL	9.6	9.4	8.5–10.5
Albumin, g/dL	3.8	3.8	3.8–5.0
AST, U/L	20	17	≤37
ALT, U/L	12	<10	≤55
ALP, U/L	74	71	≤130
Total bilirubin, mg/dL	0.3	0.3	≤1.2
C3, mg/dL	103	149	76–161
C4, mg/dL	27	34	14–40
ANA, Immunofluorescence	<1:80	<1:80	<1:80
Anticardiolipin, lupus anticoagulant, anti-β2 microglobulin antibodies	Negative	Negative	Negative
Anti-centromere, Scleroderma (Scl-70)	Negative	-	Negative
Anti-PR3 antibodies, units	<0.2	-	<0.2
Anti-MPO antibodies, units	<0.2	-	<0.2
ANCA Comprehensive	Negative	Negative	Negative
PLA2R antibody, ELISA, RU/mL	<2	n/a	<14
INR	0.97	0.98	0.8–1.2
PTT, seconds	27.4	31.7	23–36
Hepatitis B surface antigen	Non-reactive	Non-reactive
Hepatitis B core antibody	Non-reactive	Non-reactive
Hepatitis C antibody	Non-reactive	Non-reactive
SPEP/IF	No monoclonal protein	No monoclonal protein
Serum free kappa, mg/dL	3.2	3.6	0.33–1.94
Serum free lambda, mg/dL	2.1	1.69	0.57–2.63
Kappa: Lambda ratio	1.54	2.13	0.26–1.65
Factor H complement antigen (mg/dL)	33	38.4	18.5–40.8
Factor B complement antigen (mg/dL)	34	50.5	15.2–42.3
C4d complement (mcg/mL)	1.4	2.7	<9.9
CBb complement (mcg/mL)	0.9	1.8	<1.7
SC5b-9 complement	151	260	<251
Alternate complement pathway test interpretation	No serologic evidence of complement abnormalities	No serologic evidence of complement abnormalities
Urine sediment	Bland	Trace Blood 5 RBC/HPF
Urine protein-to-creatinine ratio, mg/g	2400	4089	<30
UPEP/IF	No monoclonal protein	No monoclonal protein
Echocardiogram	Normal left ventricular systolic and diastolic function	Normal left ventricular systolic and diastolic function
Renal imaging	No hydronephrosis	No hydronephrosis

Conversion factors for units: serum creatinine in mg/dL to μmol/L, ×88; urea nitrogen in mg/dL to mmol/L, ×0.357. eGFR, estimated glomerular filtration rate; ANA, antinuclear antibody; anti-MPO, anti–myeloperoxidase antibodies; anti-PR3, anti-proteinase 3 antibodies; ALT, alanine aminotransferase; AST, aspartate aminotransferase; ALK, alkaline phosphatase; ANCA, antineutrophilic cytoplasmic antibody, SUN, serum urea nitrogen; PLA2R antibody, phospholipase A2 receptor antibody, SPEP/IF, serum protein electrophoresis and immunofixation; UPEP/IF, urine protein electrophoresis and immunofixation; HPF, high power field, RBC, red blood cells, WBC, white blood cell.

Figure 1.

(A) Chronic thrombotic microangiopathy with mesangiolysis and capillary wall double contour formation (Jones stain; original magnification x600). (B) Endothelial injury with subendothelial widening and electron lucent material (electron microscopy; original magnification x 12000). (C) Chronic thrombotic microangiopathy with capillary wall multilayering and mesangiolysis (periodic acid–Schiff stain; original magnification, x400). (D) Endothelial injury with subendothelial widening, basement membrane duplication, and subendothelial ‘hyaline’ accumulation (electron microscopy; original magnification, x12000).

Case 2

A 72-year-old female with a history of well-controlled hypertension and retroperitoneal liposarcoma was referred to nephrology for evaluation of elevated serum creatinine. At the time of the initial cancer diagnosis, she underwent tumor resection and left nephrectomy at an outside institution. Patient received several anti-cancer agents including abemaciclib, pembrolizumab and epacadostat. Due to the progression of disease, she was initiated on PLD. In eighteen months before presentation to the nephrology clinic, she had received PLD at 40 mg/m² per cycle for a cumulative dose of 1240 mg/m². Her baseline serum creatinine was 1.0 mg/dL, but it had slowly risen over seven months. Home medications included alendronate, amlodipine, and levothyroxine. On initial assessment, her BP was 156/90, and she had lower extremity edema, so hydrochlorothiazide was prescribed. Laboratory data showed a serum creatinine of 2 mg/dL and UPCR of 4089 mg/g. Laboratory investigations are summarized in Table 1. The kappa light chains and kappa/lambda ratio was elevated but this was in the setting of decreased kidney function. Patient underwent kidney biopsy and light microscopy showed chronic endothelial injury suggestive of chronic TMA involving the glomeruli with glomerular capillaries basement membrane exhibiting double contour formation. EM showed endothelial injury with subendothelial widening, basement membrane duplication, and subendothelial ‘hyaline’ accumulation (Figure 1, panels C, D). PLD was stopped. The serum creatinine peaked at 5.4 mg/dl but improved to 3.4 mg/dL four months later. The peripheral edema resolved, UPCR declined to 1040 mg/g and BP improved to 123/69.

Discussion

Doxorubicin, an anthracycline anticancer drug, inhibits DNA and RNA synthesis in tumor cells. In the human body, doxorubicin can lead to the generation of free radicals, and these free radicals attack the myocardial mitochondrial membrane causing cardiotoxicity. The dose-dependent doxorubicin cardiotoxicity limits its clinical use. Compared to doxorubicin, PLD attenuates the risk of cardiotoxicity and myelosuppression but increases the risk of hypersensitivity reactions and mucocutaneous adverse effects. The kidney toxicity profile differs between non-liposomal anthracyclines and PLD. Non-liposomal anthracyclines, such as doxorubicin, and daunorubicin have been reported to cause proteinuria due to minimal change disease, focal segmental glomerulosclerosis, and collapsing glomerulopathy, although such cases are rare.^12–14 In these reports of doxorubicin or daunorubicin-associated proteinuria, none of the patients had TMA on the kidney biopsy.

Here we describe two cases of renal-limited TMA in patients with sarcoma following prolonged exposure to PLD without concomitant or recent exposure to TMA-causing drugs. Both patients had a single kidney due to a prior nephrectomy. Patient #1 had received gemcitabine and pazopanib (tyrosine kinase inhibitor) before undergoing nephrectomy. Although tyrosine kinase inhibitors can cause TMA, findings of TMA were absent in the nephrectomy sample. In both cases, PLD was administered following nephrectomy. Both patients presented with hypertension, lower extremity edema, rise in serum creatinine and proteinuria. The kidney biopsy in both patients was consistent with chronic TMA. In both cases, PLD cessation improved or stabilized serum creatinine and proteinuria. Neither of these two patients required dialysis.

There are a few case reports of PLD-associated renal TMA. However, simultaneous exposure to other TMA-causing drugs obscured the causal link between PLD and renal TMA in these cases. In a case series of patients with ovarian cancer, three patients with PLD exposure had biopsy-proven renal TMA.⁷ However, two of the three patients had also received bevacizumab and gemcitabine, drugs known to cause TMA. In another case report, a patient with metastatic breast cancer developed acute kidney injury (AKI) two months following the cessation of PLD. Although the kidney biopsy showed acute and chronic TMA, the biopsy was performed one year after PLD cessation.⁹ In a recently published report, a patient with a kidney transplant and Kaposi sarcoma developed renal-limited TMA with PLD therapy; however, the patient had recent exposure to gemcitabine and sirolimus (mTOR inhibitor), drugs known to cause TMA.¹⁰ The two patients presented in this report had renal-TMA and neither patient had recent or concomitant exposure to TMA-causing drugs. Furthermore, in patient #1, TMA was absent in the nephrectomy tissue prior to the PLD initiation.

Renal-limited TMA due to PLD is a drug-induced, dose-related TMA associated with high cumulative PLD exposure. In previously published case reports of presumed PLD-associated renal TMA, the cumulative dose of PLD was 880 mg/m2 to 1445 mg/m².^7,9–11 In the two patients described here, the cumulative PLD exposure was 760 mg/m² and 1240mg/m², respectively. The exact mechanism of PLD-induced renal TMA is not well-elucidated. PLD-induced TMA likely involves reactive oxygen species (ROS)-mediated cellular injury causing endothelial dysfunction and platelet aggregation. It is plausible that the presence of underlying CKD or reduced renal mass due to prior nephrectomy increases the risk of PLD-induced TMA and such patients must be monitored closely.

In conclusion, we present two cases of renal-limited TMA caused by PLD. The lack of concurrent exposure to other TMA-causing drugs provides a definitive causal link between PLD and renal-limited TMA. Patients on prolonged PLD therapy should be monitored closely for signs of kidney injury, such as rise in creatinine, proteinuria, and hypertension. Clinicians should be aware of PLD-induced renal-limited TMA as early recognition of this entity can lead to timely cessation of PLD therapy, limiting kidney injury and preventing the progression of kidney disease.