Effectiveness and safety of ICL670 in iron-loaded patients with thalassaemia: a randomised, double-blind, placebo-controlled, dose-escalation trial

Summary

Background

Transfusional iron overload is a potentially fatal complication of the treatment of thalassaemia. We aimed to investigate short-term efficacy, pharmacokinetic/pharma-codynamic (PK/PD) relations, and safety of ICL670, a novel, tridentate, orally active iron chelator.

Methods

We enrolled 24 patients and divided them into three cohorts consisting of a minimum of seven individuals. Patients were admitted to a metabolic unit and consumed a diet with a defined content of iron. Two patients in each cohort were randomly allocated placebo. Five or more patients received one daily dose of ICL670 at 10, 20, or 40 mg kg⁻¹ day⁻¹, from day 1 to 12. Net iron excretion (NIE) was measured between days 1 and 12. Primary objectives included assessment of safety and tolerability (measured by adverse events and clinical laboratory monitoring), pharmacokinetics (measured as drug and drug-iron complex), and cumulative net iron excretion (measured by faecal and urine output minus food input). Analysis was for efficacy.

Findings

ICL670 was absorbed promptly and was detectable in the blood for 24 h. Exposure (area under the curve of plasma concentration) to ICL670 at pharmacokinetic steady state was proportional to dose. All three doses resulted in positive NIE. The NIE achieved at 20 mg kg⁻¹day⁻¹ would prevent net iron accumulation in most patients transfused with 12–15 mL packed red-blood-cells kg⁻¹ month⁻¹, equivalent to 0·3–0·5 mg iron kg⁻¹ day⁻¹. A linear relation (PK/PD) was recorded between exposure to ICL670 and total iron excretion, by contrast with placebo (r²=0·54, p<0·0001). Skin rashes were noted in four patients treated at 20 and 40 mg kg⁻¹ day⁻¹, and one patient also developed grade 2 transaminitis.

Interpretation

ICL670 given once daily at 20 mg/kg seems to be an effective orally active iron chelator and is reasonably well tolerated. Long-term studies are now necessary to establish the practical contribution of this drug.

Introduction

Patients with thalassaemia and others who need chronic red-blood-cell transfusions die of complications of iron overload if they do not receive proper chelation. A safe and effective oral iron chelator has been the goal of therapeutic research of these disorders for several decades.1, 2 Deferoxamine—a hexadentate compound with a 1/1 stoichiometry for iron—was the first effective chelator and has few toxic effects. However, it is not orally active and its short biological half-life mandates that it be administered by subcutaneous infusion over 8–12 h a day. Although this regimen prevents iron-induced heart disease³ and extends life,⁴ the method of administration is unacceptable to many patients. At least a third of those who have access to the treatment do not adhere to it.

Many iron-chelating compounds have been synthesised and assessed preclinically to identify potential orally active drugs.⁵ Only one, deferiprone, a hydroxypyridin-4-one,⁶ has been studied in large clinical trials.7, 8, 9, 10 Deferiprone is a bidentate molecule with a 3/1 stoichiometry for iron and hence has very low efficiency for iron chelation.¹¹ Although rapidly absorbed, it is also quickly converted to an inactive form by glucuronidation,¹² and is speedily removed from the circulation.¹³ Hence, deferiprone provides effective chelation for only a brief period after dosing. Published clinical studies of this molecule show that it is ineffective in production of net iron removal in many patients even at doses above which toxic effects would be likely.1, 2, 10, 14, 15, 16 The drug's toxic effects at the presently recommended dose are uncertain10, 17, 18 and a matter of great debate.

We aimed to study the clinical effectiveness of ICL670 (4-[3,5-bis(2-hydroxyphenyl)-1,2,4-triazol-1-yl]-benzoic acid), an orally active tridentate iron chelator of a class of compounds known as 3,5-bis(ortho-hydroxyphenyl)-1,2,4 triazoles. Ryabukhin and colleagues originally described the metal chelating capability of this class of compounds.¹⁹ ICL670 was developed on the basis of computer modelling in a collaboration between synthetic chemists at the Saarland University and medicinal chemists at Novartis Pharma AG.²⁰ Its favourable 2/1 stoichiometry for iron makes it an attractive candidate for further investigation.

ICL670 has undergone extensive preclinical testing in many species, including the iron-loaded marmoset, a standard model for assessment of iron chelators.¹¹ Four major points emerge from these studies: (1) after an oral dose of ICL670, iron is excreted largely in the faeces; less than 10% is excreted in the urine; (2) one oral dose is up to five times more potent than one non-continuous subcutaneous dose of deferoxamine and up to ten times more potent than one oral dose of deferiprone, on an equimolar basis; (3) ICL670 is highly selective for iron and does not induce excretion of zinc or copper above the background (in work done in marmosets); and (4) the drug is promptly absorbed and circulates for several hours. The main toxic effect noted in these preclinical studies was changes in renal tubular epithelial cells; these changes were only seen in non-iron-loaded animals; nephrotoxic effects were not seen in iron-overloaded marmosets and were significantly reduced in iron-overloaded rats.

We aimed to establish the safety, tolerability, pharmacokinetics, and net iron chelation and elimination after repeated daily oral doses of ICL670.