Patients with β-thalassemia major require iron-chelation therapy to avoid the complication of iron overload. Until recently, deferoxamine (DFO) was the major iron chelator used in patients requiring chronic hypertransfusion therapy, but DFO required continuous subcutaneous therapy. The availability of deferasirox (Exjade®), an orally active iron chelator, over the past 4 years represented a necessary alternative for patients requiring chelation therapy. However, there have been increasing reports of proximal renal tubular dysfunction and Fanconi Syndrome1–4 associated with deferasirox in the literature. We report a case of hypophosphatemic osteomalacia secondary to deferasirox therapy.
A 28-year-old woman with β-thalassemia major was referred for management of low bone mineral density (BMD). Routine monitoring with dual energy X-ray absorptiometry (DXA) had revealed a progressive decline in the preceding 3 years, with a 16%, 28%, and 34% loss in her total body, lumbar spine, and femoral neck BMD, respectively (Fig. 1). Total-body BMD was 0.928 g/cm2 (Z-score −1.6), L2–L4 BMD was 0.749 g/cm2 (Z-score −2.3), and femoral neck BMD was 0.594 g/cm2 (Z-score −2.3).
She required monthly transfusions but was poorly compliant with DFO therapy leading to serum ferritin levels of >13,000 µg/L. The introduction of deferasirox 3 years ago led to improved compliance and a steady decline in serum ferritin levels. Her past medical history included renal calculi, hepatitis C, and diet-controlled diabetes.
The patient was currently well with no bone or joint pain, muscle weakness, or fatigue. However, she had mild transient right hip pain several months earlier following a fall during exercise, but the pain had not required medical attention. She did not smoke or consume alcohol, had good dietary calcium intake, and participated in regular weight-bearing exercise. Her menses were regular and medications consisted of vitamin D3 1000 IU daily and deferasirox 1 g daily.
Biochemical testing revealed a low serum phosphate (0.39 mmol/L) with normal renal function, serum calcium (2.31 mmol/L, normal range 2.2–2.6 mmol/L), 25OH vitamin D (77 pmol/L, normal range 75–250 pmol/L), and PTH (2.6 pmol/L, normal range 1.5–7 pmol/L). Alkaline phosphatase (ALP) was two to three times the upper limit of normal. Tubular reabsorption of phosphate (TRP) was abnormal at 47.2% (normal >80%) in the setting of low serum phosphate. The FGF-23 concentration was normal at 28 pg/mL (normal <54 pg/mL). The 24-hour urine calcium excretion was normal at 5.25 mmol/day (normal range = 2.0–7.5 mmol/day). Serum ferritin was 475 µg/L, indicative of compliance with deferasirox treatment. Diffuse demineralization was evident on thoracolumbar X-ray, and a bone scan demonstrated healing fractures of the right superior and inferior pubic rami and two older rib fractures. A urine metabolic screen showed generalized aminoaciduria (lysine, glycine, proline, glutamine, glutamate, cysteine all 1+ and citraline 2 + ; normal range undetectable), glycosuria (4 + , normal range: undetectable) and detection of protein (2 + , normal range: undetectable). A urine β2-microglobulin was elevated at 0.6 mg/dL (normal range <0.16 mg/L). Deferasirox therapy was subsequently ceased (late 2009) and oral phosphorous 500 mg twice daily was commenced. The patient declined bone biopsy.
Three days after stopping deferasirox and commencing low-dose phosphate supplementation, the serum phosphate had normalized. Furthermore, normal serum phosphate and TRP was maintained off deferasirox and phosphate. Repeat urine metabolic screen showed no aminoaciduria, protein urine, or glycosuria. Unfortunately, a repeat urine β2-microglobulin was not available. DXA performed 3 months later demonstrated a 12% improvement in the BMD at the lumbar spine and a 19% and 27% improvement at the left and the right femoral neck, respectively (since the preceding DXA 12 months earlier).
The reintroduction of deferasirox at reduced dosage was considered necessary as the patient declined DFO therapy despite increasing levels of serum ferritin (>1400 µg/L). Deferasirox was recommenced 2 to 3 months following its cessation at 250 mg daily. Although low doses of deferasirox were tolerated with no decrease in serum phosphate levels, escalation of deferasirox dose to 750 mg daily led to a reduction in serum phosphate and TRP of 74%. Stabilization of serum phosphate levels occurred with a reduction of deferasirox dose to 500 mg daily (Table 1).
Table 1. Change in Serum Phosphate With Deferasirox Dose
Improvements in BMD had been maintained 6 months following the reintroduction of low-dose deferasirox and X-ray confirmed healing of pelvic fractures. However, the ALP (although improved) had not normalized to levels prior to deferasirox initiation, and serum procollagen 1 N-terminal propeptide (P1NP) levels remained elevated. Cessation of deferasirox and reinitiation of DFO therapy was recommended to ascertain whether normalization of ALP levels would occur, but the patient declined due to prior difficulties with DFO treatment. Consideration is currently being given to oral therapy with deferiprone and her serum ferritin levels are being closely monitored on low-dose deferasirox therapy.
Thalassemia major is an inherited disorder of impaired haemoglobin synthesis and in its severe form requires life-long transfusion and concomitant iron chelation therapy.5 DFO has been the standard iron chelating agent used for the past 4 decades but requires parenteral administration. Deferiprone and deferasirox are orally active iron chelators and are being increasingly used in those with transfusion dependent thalassemia major.
Renal dysfunction is an increasingly recognized problem in β-thalassemia major but has been poorly characterized.6–8 The underlying cause for renal abnormalities is controversial but has been associated with chronic hypoxia related to anemia,8, 9 oxidative stress9, 10 and DFO.6, 10 Furthermore, serum creatinine is notoriously unreliable as a marker of kidney function, and previous studies have reported normal creatinine levels in β-thalassemia major patients with early tubular and glomerular dysfunction.8, 9, 11, 12 Our patient had evidence of proximal tubular dysfunction on deferasirox with generalized aminoaciduria, glycosuria, detection of urine protein, and elevated urine β2-microglobulin levels. Furthermore, cessation of deferasirox was followed by normalization of urinary amino acid, protein, and glucose excretion.
In a phase III trial of deferasirox in thalassemia major, a mild dose-dependent and nonprogressive increase in creatinine was noted.13 However, more serious renal dysfunction had been associated with deferasirox use including biopsy proven acute interstitial nephritis.14 More recently, reports of acquired Fanconi's syndrome have been reported with the use of deferasirox1–4 with recurrence after cessation and rechallenge.2
The administration of parenteral iron has been shown to suppress renal phosphate reabsorption resulting in hypophosphatemia through increased levels of FGF-23.15 The presence of normal FGF-23 levels, hypophosphatemia, and reduced tubular phosphate reabsorption in this case, suggests that FGF-23 is not the major cause of renal phosphate wasting.
Fanconi's syndrome leads to reduced absorption of glucose, amino acids, phosphate, and bicarbonate by the proximal renal tubule. This can lead to impaired bone mineralization and osteomalacia.16–18 Unfortunately, a bone biopsy was declined, which would have provided a definitive diagnosis of osteomalacia. However, the loss and subsequent increase in BMD associated with deferasirox cessation and dose reduction strongly suggests a causal relationship. Although the ALP levels had not normalized 6 months following dose reduction of deferasirox, elevated levels of ALP have been reported to persist for 9 months despite treatment in individuals with osteomalacia.19 The patient denied any current bone pain or muscle weakness but did report mild, transient right hip pain several months earlier, corresponding to the pubic rami fractures noted on bone scan. The patient denied any history of rib pain. The absence of symptoms requiring medical attention despite radiological evidence of fracture associated with osteomalacia is concerning and highlights the need for medical screening for potential renal tubular dysfunction and its complications.
To our knowledge, this is the first reported case of deferasirox induced renal tubulopathy leading to hypophosphataemic osteomalacia. The effect on phosphate wasting was dose dependent, with rapid normalization of serum phosphate and TRP following the drug's cessation. Renal tubular dysfunction including generalized aminoaciduria, protein urine, and glycosuria also reversed after deferasirox was stopped.2 In this case, reintroduction of low-dose deferasirox was possible without further compromising BMD. This case illustrates the dramatic effect of chronic hypophosphataemia secondary to renal phosphate loss on the skeleton and demonstrates the potential for renal tubular toxicity of deferasirox despite normal serum creatinine levels. We believe this warrants careful monitoring of renal tubular function, serum phosphate levels, and bone mineral density in those receiving treatment with deferasirox.
All authors state that they have no conflicts of interest.
This work was supported through an unrestricted educational grants from Servier, Merck Sharp and Dohme, and Sanofi-Aventis to FM and PW, and a National Health and Medical Research Council (Australia) Senior Principal Research Fellowship to PJF. Prince Henry's Institute is supported by the Victorian Government's Operational Infrastructure Support program. We thank Dr Roderick Clifton-Bligh from The Kollings Institute, Sydney, Australia, for measuring the FGF-23 level.
Authors' roles: Drafting manuscript: FM and PW. Revising manuscript content: PJF, LJ, PGK, JCGD, BJS, and DKB. Approving final version of manuscript: FM, PW, PJF, LJ, PGK, JCGD, BJS, and DKB.