Drugs (2015) 75:101–127
`DOI 10.1007/s40265-014-0332-3
`
`A D I S D R U G E V A L U A T I O N
`
`Ferric Carboxymaltose: A Review of Its Use in Iron Deficiency
`
`Gillian M. Keating
`
`Published online: 27 November 2014
`Ó Springer International Publishing Switzerland 2014
`
`Ò
`Ò
`Abstract Ferric carboxymaltose (Ferinject
`)
`, Injectafer
`is an intravenous iron preparation approved in numerous
`countries for the treatment of iron deficiency. A single high
`dose of ferric carboxymaltose (up to 750 mg of iron in the
`US and 1,000 mg of iron in the EU) can be infused in a
`short time frame (15 min). Consequently, fewer doses of
`ferric carboxymaltose may be needed to replenish iron
`stores compared with some other intravenous iron prepa-
`rations (e.g. iron sucrose). Ferric carboxymaltose improved
`self-reported patient global assessment, New York Heart
`Association functional class and exercise capacity in
`patients with chronic heart failure and iron deficiency in
`two randomized, placebo-controlled trials (FAIR-HF and
`CONFIRM-HF). In other randomized controlled trials,
`ferric carboxymaltose replenished iron stores and corrected
`anaemia in various populations with iron-deficiency anae-
`mia,
`including patients with chronic kidney disease,
`inflammatory bowel disease or heavy uterine bleeding,
`postpartum iron-deficiency anaemia and perioperative
`anaemia. Intravenous ferric carboxymaltose was generally
`well tolerated, with a low risk of hypersensitivity reactions.
`
`The manuscript was reviewed by: E. Bisbe, Department of
`Anaesthesia, Hospital Mar-Esperanc¸a, IMIM Hospital del Mar
`Medical Research Institute, Barcelona, Spain; B. Bokemeyer,
`Gastroenterology Practice, Minden, Germany; M. Lainscak, General
`Hospital Celje, Division of Cardiology, Celje, Slovenia; I.C.
`Macdougall, Renal Unit, King’s College Hospital, London, UK.
`
`G. M. Keating (&)
`Springer, Private Bag 65901, Mairangi Bay 0754, Auckland,
`New Zealand
`e-mail: demail@springer.com
`
`It was generally better tolerated than oral ferrous sulfate,
`mainly reflecting a lower incidence of gastrointestinal
`adverse effects. The most common laboratory abnormality
`seen in ferric carboxymaltose recipients was transient,
`asymptomatic hypophosphataemia. The higher acquisition
`cost of ferric carboxymaltose appeared to be offset by
`lower costs for other items, with the potential for cost
`savings.
`In conclusion,
`ferric carboxymaltose is an
`important option for the treatment of iron deficiency.
`
`carboxymaltose
`Ferric
`summary
`
`in iron deficiency:
`
`a
`
`A single high dose of ferric carboxymaltose can be
`infused in a short time frame (15 min), rapidly
`replenishing iron stores
`
`Compared with some other intravenous preparations,
`fewer doses may be needed to replenish iron stores
`
`Showed improvement in self-reported patient global
`assessment, New York Heart Association functional
`class and exercise capacity in chronic heart failure
`and iron deficiency
`
`Replenished iron stores and corrected anaemia in
`various populations with iron-deficiency anaemia
`
`Generally well tolerated, with a low risk of
`hypersensitivity reactions, and better tolerated than
`oral ferrous sulfate
`
`Associated with transient, asymptomatic
`hypophosphataemia
`
`Potential for cost savings
`
`Pharmacosmos, Exh. 1057, p. 1
`
`

`
`102
`
`1 Introduction
`
`Globally, iron deficiency is the most commonly occurring
`nutritional deficiency and the most common cause of
`anaemia [1, 2]. Iron deficiency is a common feature of
`various chronic diseases [e.g. chronic heart failure (CHF),
`chronic kidney disease] and its aetiology is often multi-
`factorial [3]. Iron deficiency may be associated with poor
`nutrition, increased utilization of iron (e.g. during preg-
`nancy), blood loss [e.g. heavy uterine bleeding, blood loss
`associated with gastrointestinal (GI) disorders, surgical
`blood loss], chronic inflammation or
`impaired iron
`absorption [e.g. in inflammatory bowel disease (IBD)] [3–
`5]. Although particularly prevalent in developing countries,
`anaemia remains a significant problem in developed
`countries [2, 5]. Diagnostic criteria for anaemia vary
`between studies [3], although WHO defines anaemia as a
`haemoglobin level of \130 g/L in men, \120 g/L in non-
`pregnant women and \110 g/L in pregnant women [2].
`Iron metabolism is a complex process and the 25-amino
`acid peptide hepcidin is the key regulator of iron homeo-
`stasis [6]. Hepcidin is upregulated in chronic inflammation
`(as seen in cancer or autoimmune conditions), which
`contributes to the development of anaemia of chronic dis-
`ease [6]. Other factors affecting the regulation of hepcidin
`include iron availability, erythropoiesis and hypoxia [6].
`Iron deficiency may be absolute or functional [7].
`Absolute iron deficiency occurs when iron stores are so low
`that no iron is available for the production of haemoglobin
`[7]. The negative iron balance compromises erythropoiesis
`resulting in iron-deficiency anaemia [4, 8]. In functional
`iron deficiency, body iron stores are normal or increased,
`but there is a failure of these stores to release iron rapidly
`enough to support the demands of the bone marrow [3, 7].
`Iron supplementation is used to correct anaemia and
`replenish iron stores in patients with iron-deficiency
`Ò
`anaemia [3]. Ferric carboxymaltose (Ferinject
`, Injecta-
`Ò
`)
`is an intravenous iron preparation approved in
`fer
`numerous countries for the treatment of iron deficiency.
`This article reviews the clinical efficacy, tolerability and
`safety of ferric carboxymaltose in iron deficiency, as well
`as summarizing its pharmacological properties and results
`of pharmacoeconomic analyses. Throughout this article,
`the dose of ferric carboxymaltose and other iron prepara-
`tions is expressed in milligrams of iron.
`
`2 Pharmacodynamic Properties
`
`The pharmacodynamic properties of ferric carboxymaltose
`have been reviewed previously [9]; this section provides a
`brief overview.
`
`G. M. Keating
`
`Ferric carboxymaltose is formulated as a colloidal
`solution of polynuclear iron(III) oxyhydroxide stabilized
`by the carbohydrate polymer carboxymaltose [10, 11]. The
`ferric carboxymaltose solution has isotonic osmolarity and
`a pH of 4.5–7.0 and the complex has a molecular mass of
`&150 kDa [10–12]. Ferric carboxymaltose is designed to
`provide controlled delivery of iron to the macrophages of
`the reticuloendothelial system in the liver, spleen and bone
`marrow, with subsequent delivery to the transport protein
`transferrin, without
`large amounts of ionic iron being
`released into the serum [10, 11].
`The stability of the ferric carboxymaltose complex
`permits the administration of high doses of iron [11]. In
`mice, the 50 % lethal dose (LD50) of intravenous ferric
`carboxymaltose was [1,000 mg iron/kg bodyweight [10],
`compared with LD50 values of 230 mg iron/kg for oral iron
`sulfate, [200 mg iron/kg for intravenous iron sucrose and
`[2,500 mg iron/kg for intravenous iron dextran [13].
`In patients with anaemia receiving radiolabelled ferric
`carboxymaltose, maximum red cell uptake of 59Fe of
`61–99 % was seen after 16–24 days [14]. Intravenous
`ferric carboxymaltose transiently increased total serum iron
`concentrations (see Sect. 3) [15]. Results of clinical trials
`demonstrating the beneficial effects of ferric carboxymal-
`tose on haemoglobin levels, serum ferritin levels and
`transferrin saturation are discussed in Sect. 4.
`Ferric carboxymaltose was associated with transient,
`asymptomatic hypophosphataemia in patients with iron
`deficiency (see Sect. 5). The mechanisms underlying this
`hypophosphataemia are unclear, although one possible
`mechanism is an effect on the phosphate-regulating peptide
`hormone fibroblast growth factor 23 (FGF23). In women
`with heavy uterine bleeding and iron-deficiency anaemia,
`serum levels of intact FGF23 increased from baseline to a
`significantly (p \ 0.05) greater extent with intravenous
`ferric carboxymaltose than with intravenous iron dextran,
`with no significant between-group difference in the change
`from baseline in plasma levels of C-terminal FGF23 [16].
`The transient increase in intact FGF23 was accompanied
`by a transient reduction in serum phosphate levels [16].
`Ferric carboxymaltose lowered elevated platelet counts
`and normalized platelet activation in patients with IBD [17,
`18]. For example, in patients with IBD and secondary
`thrombocytosis in the ThromboVIT trial, mean platelet
`counts were significantly (p \ 0.01) lower at week 6 in
`patients receiving ferric carboxymaltose 500–1,500 mg
`than in placebo recipients, although there was no signifi-
`cant difference in the proportion of ferric carboxymaltose
`versus placebo recipients achieving a reduction in platelet
`count of C25 % [18]. Significant (p \ 0.05) reductions
`from baseline in platelet activation were also seen with
`ferric carboxymaltose [18].
`
`Pharmacosmos, Exh. 1057, p. 2
`
`

`
`Ferric Carboxymaltose: A Review
`
`103
`
`Prolonged exposure to high levels of non-transferrin
`bound iron (as seen in untreated patients with iron overload
`disorders) may trigger oxidative stress in organs such as the
`liver, heart and pancreas [19]. Thus, there has been concern
`that iron compounds that release large amounts of ionic
`iron into the circulation may potentiate oxidative stress [20,
`21]. However, ferric carboxymaltose appears to be asso-
`ciated with a relatively low risk of oxidative stress,
`reflecting the fact that iron is predominantly deposited in
`the reticuloendothelial system [10, 11]. For example, sig-
`nificant (p \ 0.01) increases in oxidative stress and pro-
`inflammatory markers were seen in the liver, heart and
`kidneys of rats administered low-molecular-weight (LMW)
`or high-molecular-weight (HMW) iron dextran or sodium
`ferric gluconate (ferrous gluconate), compared with rats
`administered ferric carboxymaltose or iron sucrose [22]. In
`addition, ferric carboxymaltose had no effect on plasma
`levels of intercellular adhesion molecule (ICAM) or vas-
`cular adhesion molecule or proinflammatory markers in
`patients with nondialysis-dependent chronic kidney disease
`[23].
`However, reactive oxygen species production, ICAM-1
`expression and apoptosis were significantly (p \ 0.005 vs.
`control) increased from baseline when mononuclear cells
`from haemodialysis patients or healthy volunteers were
`cultured in vitro with ferric carboxymaltose, iron sucrose,
`iron dextran or sodium ferric gluconate, with significantly
`(p \ 0.005) greater effects seen in mononuclear cells from
`haemodialysis patients than in mononuclear cells from
`healthy volunteers [20].
`Baseline levels of erythrocyte glutathione peroxidase
`(eGPx)
`(an antioxidant
`enzyme) were
`significantly
`(p = 0.01) higher among patients with nondialysis-depen-
`dent chronic kidney disease and iron-deficiency anaemia
`who responded to a single dose of ferric carboxymaltose
`1,000 mg than in nonresponders [24]. Multivariate analysis
`revealed that eGPx levels independently predicted response
`to ferric carboxymaltose [24].
`
`3 Pharmacokinetic Properties
`
`Total serum iron concentrations increased rapidly following
`the intravenous administration of single ferric carboxy-
`maltose doses of 100, 500, 800 or 1,000 mg of iron to
`patients with mild iron-deficiency anaemia; the 100 mg
`dose was administered as an intravenous bolus and the 500,
`800 and 1,000 mg doses were administered as 15-min
`intravenous infusions [15]. Following single doses of ferric
`carboxymaltose 100, 500, 800 and 1,000 mg, geometric
`mean maximum serum iron concentrations of 37, 156, 319
`and 331 lg/mL, respectively, were reached in a mean time
`of 0.3, 0.3, 1.0 and 1.2 h, respectively [15]. The geometric
`
`mean area under the serum concentration-time curve (AUC)
`from time zero to 24 h was 333 lg h/mL with ferric carb-
`oxymaltose 100 mg, and the AUC from time zero to 72 h
`was 2,345, 5,171 and 6,277 lg h/mL with ferric carboxy-
`maltose 500, 800 and 1,000 mg, respectively [15]. Repeated
`administration of ferric carboxymaltose 500 or 1,000 mg
`did not result in accumulation of iron in serum [25].
`The iron in ferric carboxymaltose had a volume of
`distribution of &3 L [15, 26]. Ferric carboxymaltose did
`not cross the placenta in an in vitro perfusion model [27].
`The iron in ferric carboxymaltose was rapidly cleared
`from plasma [26], with positron emission tomography
`demonstrating that the major portion of a radiolabelled
`injected iron dose was distributed to the bone marrow, with
`uptake also seen in the liver and spleen [14]. Following
`single doses of
`ferric carboxymaltose 100–1,000 mg,
`geometric mean clearance was 2.6–4.3 mL/min [15].
`In vitro experiments showed that the carbohydrate moiety
`underwent hydrolysis to oligoglucose units (e.g. maltotri-
`ose, maltose and glucose) [11].
`Renal elimination of iron following administration of
`ferric carboxymaltose was negligible [26, 28]. Following
`single doses of ferric carboxymaltose 100–1,000 mg, the
`geometric mean terminal
`elimination half-life was
`7.4–12.1 h and the geometric mean residence time was
`11.2–16.6 h [15].
`When intravenous ferric carboxymaltose was adminis-
`tered to women with postpartum iron-deficiency anaemia,
`mean iron concentrations in breast milk increased from
`0.500 mg iron/kg breast milk at baseline to a maximum
`1.447 mg/kg at 24-h postdose [29]. Neither mother nor
`infants had evidence of adverse events associated with the
`administration of ferric carboxymaltose to breast-feeding
`mothers [29].
`
`4 Therapeutic Efficacy in Iron Deficiency
`
`The main focus of this section is the results of randomized
`controlled trials examining the efficacy of ferric carboxy-
`maltose in iron deficiency. Where relevant, results of
`studies in real-world settings are also briefly discussed,
`with a focus on prospective data.
`
`4.1 In Chronic Heart Failure
`
`The efficacy of ferric carboxymaltose in patients with CHF
`and iron deficiency (with or without anaemia) was exam-
`ined in the randomized, double-blind, placebo-controlled,
`multinational FAIR-HF [30] and CONFIRM-HF [31] trials.
`The trials were of 26 [30] or 52 [31] weeks’ duration.
`Inclusion criteria included New York Heart Association
`(NYHA) class II or III CHF [30, 31],
`left ventricular
`
`Pharmacosmos, Exh. 1057, p. 3
`
`

`
`104
`
`G. M. Keating
`
`ejection fraction (LVEF) of B40 % (for NYHA class II
`patients) or B45 % (for NYHA class III patients) [30] or
`B45 % [31], a haemoglobin level of 95–135 g/L [30] or
`\150 g/L [31], iron deficiency (i.e. serum ferritin level of
`\100 ng/mL, or 100–299 ng/mL with a transferrin satu-
`ration of \20 %) [30, 31] and elevated natriuretic peptide
`levels (brain natriuretic peptide level of [100 pg/mL and/
`or N-terminal-pro-brain natriuretic peptide
`level of
`[400 pg/mL) [31].
`In FAIR-HF, patients received intravenous ferric carb-
`oxymaltose 200 mg once weekly until iron repletion was
`achieved (calculated total iron requirement based on the
`Ganzoni formula) and then every 4 weeks during the
`maintenance phase; the maintenance phase started at week
`8 or 12, depending on the iron repletion dose [30]. In
`CONFIRM-HF, patients received ferric carboxymaltose at
`baseline and, if necessary, at week 6 to correct iron defi-
`ciency (cumulative dose of 500–2,000 mg with the dose
`based on bodyweight and haemoglobin levels), followed by
`maintenance doses of 500 mg at weeks 12, 24 and 36 if
`iron deficiency was still present [31].
`Among ferric carboxymaltose and placebo recipients at
`baseline, 17.4 versus 18.7 % in FAIR-HF [30] and 53.3
`versus 60.3 % in CONFIRM-HF [31] were NYHA class
`II, 82.6 versus 81.3 % [30] and 46.7 versus 39.7 % [31]
`were NYHA class III, mean LVEF was 31.9 versus
`33.0 % [30] and 37.1 versus 36.5 % [31], mean haemo-
`globin was 119 versus 119 g/L [30] and 124 versus
`124 g/L [31], mean serum ferritin was 53 versus 60 ng/
`mL [30] and 57 versus 57 ng/mL [31] and mean trans-
`ferrin saturation was 17.7 versus 16.7 % [30] and 20.2
`versus 18.2 % [31].
`The primary endpoints were self-reported patient global
`assessment and NYHA functional class at week 24 [30] and
`the change in 6-min walk test distance from baseline to
`week 24 [31]. Efficacy was assessed in the full analysis set
`[30, 31].
`
`4.1.1 FAIR-HF Trial
`
`Ferric carboxymaltose was beneficial in patients with CHF
`and iron deficiency with or without anaemia [30]. The odds
`ratios (ORs) for improvements in self-reported patient
`global assessment and NYHA functional class both sig-
`nificantly favoured ferric carboxymaltose versus placebo
`recipients at 24 weeks (Table 1). At week 24, 50 % of
`ferric carboxymaltose recipients and 28 % of placebo
`recipients reported much or moderate improvement (based
`on self-reported patient global assessment), and 47 % of
`ferric carboxymaltose recipients and 30 % of placebo
`recipients were NYHA class I or II [30]. ORs for both
`improvement in self-reported patient global assessment and
`NYHA functional class also significantly (p \ 0.001)
`
`favoured ferric carboxymaltose versus placebo recipients at
`weeks 4 and 12 [30].
`Ferric carboxymaltose had a consistent treatment effect
`in patients with anaemia (haemoglobin level B120 g/L)
`and in those without anaemia [32]. ORs for improved self-
`reported patient global assessment were 2.48 (95 % CI
`1.49–4.14; p \ 0.001) in patients with anaemia and 2.60
`(95 % CI 1.55–4.35; p \ 0.001) in patients without anae-
`mia, and ORs for improvement by one NYHA functional
`class were 1.90 (95 % CI 1.06–3.40; p = 0.03) in patients
`with anaemia and 3.39 (95 % CI 1.70–6.75; p \ 0.001) in
`patients without anaemia [32]. It should be noted that
`haemoglobin levels
`at week 24 were
`significantly
`(p \ 0.001) higher in ferric carboxymaltose than in pla-
`cebo recipients in the overall population and in the sub-
`group of patients with anaemia at baseline, with no
`significant between-group difference in the subgroup of
`patients without anaemia at baseline [30].
`The mean distance achieved on the 6-min walk test was
`significantly (p \ 0.001) longer with ferric carboxymaltose
`than with placebo at weeks 4 (294 vs. 269 m), 12 (312 vs.
`272 m) and 24 (313 vs. 277 m) (mean baseline values were
`274 vs. 269 m) [30]. At week 24, the increase in 6-min
`walk distance was significantly correlated with a reduction
`in red cell distribution width (r = -0.25; p \ 0.0001),
`according to results of a post hoc analysis [33]. Although
`red cell distribution width initially increased in ferric
`carboxymaltose recipients, it was significantly (p \ 0.05)
`lower with ferric carboxymaltose than with placebo at
`week 24 [33].
`Ferric carboxymaltose improved health-related quality
`of life (HR-QOL) in patients with CHF and iron deficiency
`[34]. The mean changes from baseline in European Quality
`of Life-5 Dimensions (EQ-5D) visual analogue scale
`(VAS) scores significantly (p \ 0.001) favoured ferric
`carboxymaltose versus placebo recipients at weeks 4 (6.0
`vs. 0.8), 12 (7.9 vs. 2.4) and 24 (9.1 vs. 3.4) (mean baseline
`values were 54 in both treatment groups), as did the mean
`changes from baseline in the Kansas City Cardiomyopathy
`Questionnaire (KCCQ) overall summary scores at weeks 4
`(9.4 vs. 3.5), 12 (12.2 vs. 4.6) and 24 (12.8 vs. 6.2) (mean
`baseline values were 52 and 53). Ferric carboxymaltose
`improved HR-QOL in patients with and without anaemia
`[34]. Multivariate analysis showed that intravenous iron
`therapy, lower NYHA class and better 6-min walk test
`results were associated with higher HR-QOL, whereas a
`history of stroke and reduced renal function were associ-
`ated with lower HR-QOL [35].
`Ferric carboxymaltose improved renal function in iron-
`deficient patients with CHF, according to an additional
`analysis of FAIR-HF (available as a poster) [36]. At week
`24, the change from baseline in estimated glomerular fil-
`tration rate significantly favoured ferric carboxymaltose
`
`Pharmacosmos, Exh. 1057, p. 4
`
`

`
`Ferric Carboxymaltose: A Review
`
`105
`
`Table 1 Efficacy of ferric carboxymaltose in patients with chronic heart failure and iron deficiency. Shown are the primary endpoints in the
`randomized, double-blind, multinational FAIR-HF [30] and CONFIRM-HF [31] trials
`
`Study (study name)
`
`Endpoint
`
`Anker et al. [30, 125] (FAIR-HF)
`
`Self-reported patient global assessment at 24 weeksa
`
`Much improved (% of pts)
`
`Moderately improved (% of pts)
`
`A little improved (% of pts)
`
`Unchanged (% of pts)
`
`A little, moderately or much worse (% of pts)
`
`Dead (% of pts)
`
`Odds ratio for improvement (95 % CI)
`NYHA functional class at 24 weeksb
`
`Class I (% of pts)
`
`Class II (% of pts)
`
`Class III (% of pts)
`
`Class IV (% of pts)
`
`Dead (% of pts)
`
`FCM
`
`16
`
`34
`
`26
`
`18
`
`3
`
`2
`
`2.51 (1.75–3.61)**
`
`6
`
`41
`
`50
`
`1
`
`2
`
`PL
`
`10
`
`17
`
`28
`
`35
`
`7
`
`3
`
`1
`
`29
`
`65
`
`3
`
`3
`
`Odds ratio for improvement by one class (95 % CI)
`Ponikowski et al. [31, 126] (CONFIRM-HF) LSM change from baseline to week 24 in 6MWT distancec
`(m) [mean baseline value; m]
`
`2.40 (1.55–3.71)**
`
`?18* [288]
`
`-16 [302]
`
`6MWT 6-min walk test, FCM ferric carboxymaltose, LSM least squares mean, NYHA New York Heart Association, PL placebo, pts patients
`* p \ 0.01, ** p \ 0.001 vs. PL
`a 292 FCM recipients and 149 PL recipients were evaluable for self-reported patient global assessment
`b 294 FCM recipients and 150 PL recipients were evaluable for NYHA functional class
`c 150 FCM recipients and 151 PL recipients were in the full analysis set
`
`(between-group difference
`versus placebo recipients
`4.0 mL/min/1.73 m2; p = 0.017) [36].
`
`4.1.2 CONFIRM-HF Trial
`
`A significant improvement from baseline to week 24 in the
`6-min walk test distance was seen with ferric carboxy-
`maltose versus placebo in patients with CHF and iron
`deficiency in the CONFIRM-HF trial (Table 1), with a
`least squares mean between-group difference of 33 m
`(p = 0.002) [31]. The improvement in 6-min walk test
`distance was maintained at later time points; the difference
`between ferric carboxymaltose and placebo recipients was
`?42 m at week 36 and ?36 m at week 52 (both
`p \ 0.001). No significant interaction was seen in terms of
`the treatment effect in patients with anaemia (haemoglobin
`\120 g/L) or without anaemia (haemoglobin C120 g/L)
`(p-value for interaction 0.15) [31].
`Ferric carboxymaltose recipients were significantly
`(p \ 0.05) more likely than placebo recipients to have
`improvements in self-reported patient global assessment at
`weeks 12, 24, 36 and 52 and in NYHA class at weeks 24,
`36 and 52 [31]. Significant (p \ 0.05) improvements were
`seen with ferric carboxymaltose versus placebo in terms of
`the least squares mean change in fatigue scores at weeks
`
`12, 24, 36 and 52 and for KCCQ scores at weeks 12, 36 and
`52. A significant (p = 0.002) between-group difference
`favouring ferric carboxymaltose was seen for the least
`squares mean change in EQ-5D VAS score at week 36,
`with no significant between-group difference seen at other
`time points [31].
`significantly (p \ 0.001)
`Between-group differences
`favoured ferric carboxymaltose versus placebo recipients
`in terms of ferritin levels at week 24 (265 ng/mL) and
`week 52 (200 ng/mL), transferrin saturation at week 24
`(8.9 %) and week 52 (5.7 %) and haemoglobin levels at
`week 24 (6 g/L) and week 52 (10 g/L) [31].
`
`4.2 In Chronic Kidney Disease
`
`Several randomized, open-label, multicentre trials exam-
`ined the efficacy of intravenous ferric carboxymaltose in
`patients with chronic kidney disease and iron-deficiency
`anaemia [37–41]. Patients had nondialysis-dependent
`chronic kidney disease [37–39, 41] and/or were undergoing
`haemodialysis [40, 41]. Patients with nondialysis-depen-
`dent chronic kidney disease had haemoglobin levels of
`B110 g/L [39] or B115 g/L [37, 41] or at least one hae-
`moglobin value of 90–110 g/L within 4 weeks of ran-
`domization [38], and a ferritin level of B100 ng/mL [37,
`
`Pharmacosmos, Exh. 1057, p. 5
`
`

`
`106
`
`G. M. Keating
`
`38],\200 ng/mL with transferrin saturation of B20 % [38]
`or B300 ng/mL with transferrin saturation of B25 % [39]
`or B30 % [37, 41]. Patients with chronic kidney disease
`who were undergoing haemodialysis had a haemoglobin
`level of B115 g/L [40] or B125 g/L [41] and a ferritin
`level of \200 ng/mL or transferrin saturation of \20 %
`[40] or a ferritin level of \500 ng/mL and transferrin sat-
`uration of B30 % [41]. Patients were followed for 30 days
`[41], 120 days [37], 8 weeks [39, 40] or 56 weeks [38].
`Ferric carboxymaltose was compared with oral ferrous
`sulfate [38, 39], intravenous iron sucrose [37, 40] or stan-
`dard medical care [41]. Stable treatment with erythropoi-
`esis-stimulating agents was permitted in four trials [37, 39–
`41], with no erythropoiesis-stimulating agent therapy per-
`mitted in a fifth trial [38].
`Primary endpoints included the mean change from
`baseline to the highest observed haemoglobin level at any
`time up to day 56 [37], the proportion of patients achieving
`an increase in haemoglobin of C10 g/L [39, 40] and the
`time to initiation of other anaemia management or occur-
`rence of a haemoglobin trigger [38]; safety was the primary
`endpoint in one study, with efficacy endpoints considered
`exploratory secondary endpoints [41]. One trial is only
`available as an abstract [40].
`Ferric carboxymaltose was more effective than oral
`ferrous sulfate in patients with nondialysis-dependent
`chronic kidney disease and iron-deficiency anaemia [38,
`39]. In the FIND-CKD trial, patients whose ferric carb-
`oxymaltose dose was adjusted to maintain a higher ferritin
`level had a significantly (p = 0.026) lower risk of initiating
`other anaemia management or experiencing a haemoglobin
`trigger than ferrous sulfate recipients [hazard ratio (HR)
`0.65; 95 % CI 0.44–0.95], with no significant difference
`seen between patients whose ferric carboxymaltose dose
`was adjusted to maintain a higher versus a lower ferritin
`level (Table 2) [38]. High-ferritin ferric carboxymaltose
`recipients were significantly more likely than low-ferritin
`ferric carboxymaltose or
`ferrous sulfate recipients to
`achieve an increase in haemoglobin level of C10 g/L
`(Table 2). A significantly greater
`least squares mean
`increase from baseline to month 12 in haemoglobin level
`was seen in high-ferritin ferric carboxymaltose recipients
`than in ferrous sulfate recipients, with no significant dif-
`ference seen between low-ferritin ferric carboxymaltose
`recipients and ferrous sulfate recipients (Table 2) [38]. The
`least squares mean change from baseline to month 12 in
`serum ferritin levels was significantly greater with high-
`ferritin ferric carboxymaltose than with ferrous sulfate
`(?451 vs. ?137 ng/mL; p \ 0.001), and was significantly
`smaller with low-ferritin ferric carboxymaltose than with
`ferrous sulfate (?81 vs. ?137 ng/mL; p \ 0.001). The
`least squares mean change from baseline to month 12 in
`transferrin saturation did not significantly differ between
`
`high-ferritin ferric carboxymaltose and ferrous sulfate
`recipients (?15.8 vs. ?13.8 %), although it was signifi-
`cantly smaller with low-ferritin ferric carboxymaltose than
`with ferrous sulfate (?8.5 vs. ?13.8 %; p = 0.001). The
`mean number of ferric carboxymaltose doses needed to
`achieve and maintain ferritin levels in the high- and low-
`ferritin groups was 4.0 and 4.8, respectively [38].
`In a second trial, significantly more ferric carboxymal-
`tose than ferrous sulfate recipients achieved an increase in
`haemoglobin level of C10 g/L and the mean increase from
`baseline to the highest observed haemoglobin level was
`significantly greater with ferric carboxymaltose than with
`ferrous sulfate (Table 2) [39]. The mean changes from
`baseline to study end in serum ferritin levels (?358.8 vs.
`?178.4 ng/mL) and transferrin saturation (?12.1 vs.
`?7.0 %) were significantly (p \ 0.001) greater with ferric
`carboxymaltose than with ferrous sulfate. Only one dose of
`ferric carboxymaltose was administered to 58 % of patients
`[39].
`In the REPAIR-IDA trial, ferric carboxymaltose was
`noninferior to iron sucrose in terms of the mean change
`from baseline to the highest observed haemoglobin level up
`to day 56 in patients with nondialysis-dependent chronic
`kidney disease (Table 2)
`[37]. The 95 % CI
`for
`the
`between-group difference was entirely above zero, sug-
`gesting superiority of ferric carboxymaltose versus iron
`sucrose, although assessment of superiority was not pre-
`specified. Ferric carboxymaltose was also noninferior to
`iron sucrose in terms of the proportion of patients achiev-
`ing an increase in haemoglobin level of C10 g/L up to day
`56; the 95 % CI for the between-group difference was
`entirely above zero (Table 2) [37]. The mean changes from
`baseline to the highest observed value in serum ferritin and
`transferrin saturation were significantly greater with ferric
`carboxymaltose than with iron sucrose (p-values not sta-
`ted). Two ferric carboxymaltose infusions were adminis-
`tered to 96.8 % of patients, and five iron sucrose infusions
`were administered to 91.6 % of patients [37].
`In patients with iron-deficiency anaemia undergoing
`haemodialysis, an increase in haemoglobin level of C10 g/
`L had occurred in 46.4 % of ferric carboxymaltose recip-
`ients and in 37.2 % of iron sucrose recipients at week 4
`(Table 2) [40]. The mean change from baseline in hae-
`moglobin levels is shown in Table 2. At study end, mean
`serum ferritin levels were 465.3 ng/mL in ferric carboxy-
`maltose recipients and 397.7 ng/mL in iron sucrose recip-
`ients (baseline levels of 90.4 and 93.1 ng/mL in the
`corresponding treatment groups) [40].
`In patients with chronic kidney disease (either nondial-
`ysis-dependent or undergoing haemodialysis), no signifi-
`cant difference was seen between patients receiving ferric
`carboxymaltose and those receiving standard medical care
`in terms of
`the change from baseline to day 30 in
`
`Pharmacosmos, Exh. 1057, p. 6
`
`

`
`Ferric Carboxymaltose: A Review
`
`107
`
`Table 2 Efficacy of ferric carboxymaltose in patients with chronic kidney disease and iron-deficiency anaemia. Results of randomized, open-
`label, multicentre trials; patients were followed for 30 days [41], 120 days [37], 8 weeks [39, 40] or 56 weeks [38]
`
`Study
`
`Treatment
`
`No. of
`ptsa
`
`Hb
`
`Meanc
`baseline
`value (g/L)
`
`Meanc change
`from baselined
`(g/L)
`
`Increase of C10 g/Le
`(% of pts)
`
`Pts (%) starting other
`anaemia management
`or having an Hb triggerb
`
`Pts with nondialysis-dependent chronic kidney disease
`IV FCMf
`IV ISCi
`IV high-ferritin
`FCMj
`IV low-ferritin
`FCMj
`Oral FSk
`IV FCMl
`Oral FSk
`Pts with chronic kidney disease undergoing haemodialysis
`46.4g
`Schaefer et al. [40]m
`FCMn
`37.2g
`ISCn
`Pts with chronic kidney disease who were nondialysis dependent or undergoing haemodialysis
`FCMo
`SMCp
`
`Onken et al. [37, 127]
`(REPAIR-IDA)
`
`Macdougall et al. [38, 128]
`(FIND-CKD)
`
`Qunibi et al. [39]
`
`Charytan et al. [41]
`
`1,276
`
`1,285
`
`103.1
`
`103.2
`
`153
`
`101
`
`152
`
`102
`
`308
`
`147
`
`103
`
`97
`
`86
`
`249
`
`249
`
`102
`
`101
`
`100
`
`93
`
`93.4
`
`105.9
`
`104.5
`
`?11.3g,h
`?9.2g
`?14*
`
`?9
`
`?10
`
`?131**
`
`?0.83
`
`?12.7
`
`?9.6
`
`?4.9
`
`?3.3
`
`48.6h
`41.0
`56.9** 
`
`34.2
`
`32.1
`60.4**g
`34.7g
`
`25.7
`
`22.1
`
`23.5*g
`
`32.2g
`
`31.8g
`
`FCM ferric carboxymaltose, FS ferrous sulfate, Hb haemoglobin, ISC iron sucrose, IV intravenous, pts patients, SMC standard medical care
`* p \ 0.05, ** p B 0.001 vs. FS;  p \ 0.001 vs. low-ferritin FCM
`a No. of pts in the modified intent-to-treat [37–39, 41] or per-protocol [40] populations
`b Other anaemia management included erythropoiesis-stimulating agents, blood transfusion or use of an alternative iron therapy, and occurrence
`of an Hb trigger was defined as two consecutive Hb values \100 g/L on or after week 8, without an increase of C5 g/L between the two
`measurements
`c Mean [37, 39–41] or least squares mean values [38]
`d Change from baseline to highest observed Hb at any time up to day 56 [37, 39], or change from baseline to day 30 [41], week 8 [40] or month
`12 [38]
`e Increase of C10 g/L at any time in the study period [37, 39, 41], at week 4 [40] or prior to starting other anaemia management [38]
`f FCM was administered on days 0 and 7 (each dose a maximum of 750 mg)
`g Primary endpoint
`h Noninferiority shown for FCM vs. ICS
`i Five doses of ISC 200 mg were administered between days 0 and 14
`j High-ferritin FCM comprised FCM 1,000 mg followed every 4 weeks by FCM 500 mg if ferritin was 200–400 ng/mL or FCM 1,000 mg if
`ferritin was \200 ng/mL. Low-ferritin FCM comprised FCM 200 mg if ferritin was \100 ng/mL followed every 4 weeks by FCM 200 mg if
`ferritin was \100 ng/mL
`k FS 100 mg twice daily to week 52 [38] or 65 mg three times daily to day 56 [39]
`l First maximum dose of 1,000 mg; second and third maximum doses of 500 mg administered on approximately day 17 and/or day 31 if the
`transferrin saturation remained \30 % and ferritin remained \500 ng/mL
`m Available as an abstract
`n FCM or ISC 200 mg two or three times weekly (administered into the venous line of the dialyser) until the total cumulative dose for each pt
`was reached
`o Nondialysis-dependent pts received a maximum IV dose of FCM 1,000 mg and pts undergoing haemodialysis received FCM 200 mg directly
`into the venous line of the dialyser
`p SMC comprised IV iron (63 % of pts), oral iron (30 %) or no iron treatment (8 %)
`
`haemoglobin levels (