Reimbursement Review

Ferric Carboxymaltose (Ferinject)

Sponsor: CSL Vifor

Therapeutic area: Iron deficiency anemia

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Clinical Review

Abbreviations

Executive Summary

An overview of the submission details for the drug under review is provided in Table 1.

Table 1: Background Information of Application Submitted for Review

NOC = Notice of Compliance.

Introduction

Anemia is a medical condition in which the blood has reduced capacity to carry oxygen because of low hemoglobin levels. Although anemia has a range of causes, iron deficiency is the most prevalent.¹ Iron deficiency anemia (IDA) results when body iron levels are insufficient to sustain hemoglobin synthesis.² Alberta clinical practice guidelines describe IDA as “a serious condition whereby low levels of iron are associated with anemia and the presence of microcytic hypochromic red cells.”³ British Columbia clinical guidelines define anemia based on hemoglobin levels relative to the population mean: “[a] low hemoglobin level, most frequently defined as a hemoglobin value over 2 standard deviations below the gender- and age-adjusted mean. A hemoglobin value below the local, lab-specific lower reference interval indicates anemia.”⁴

In a Canadian Health Measures Survey study, approximately 2% of the population in Canada was identified as having IDA, with the prevalence in females approximately 9 times higher than that of males. In a recent study by Cooper et al., IDA prevalence in the total population, corrected for inflammation ranged from 2% to 2.5%. Notably, less than 1% of males exhibited IDA regardless of the correction method, whereas this figure rose to approximately 4% for all females. For females of child-bearing age, the corrected IDA prevalence ranged between 5% and 6.2%.⁵ Iron deficiency and IDA during the prenatal period have been associated with adverse effects on neonatal growth, neurocognitive development, and perinatal morbidity and mortality. A 2022 study assessed iron status during early and late pregnancy among 60 pregnant women receiving 27 mg/day of elemental iron as part of a randomized trial.⁶ High rates of iron deficiency were observed: 28% of participants exhibited probable iron deficiency at baseline (mean gestational age = 16 weeks), with the proportion increasing to 81% by study end (mean gestational age = 32 weeks).⁶ This deterioration in iron status occurred despite widespread supplementation and highlights a potential need for additional iron beyond the recommended dose to support needs during later pregnancy. While the general Canadian population maintains a relatively low prevalence of anemia (approximately 3%), this is not the case among Canada's Indigenous population. In a 2017 study of the Inuit population in Nunavik (aged 16 years and older), the prevalence of anemia in women of child-bearing age was 20%, with 14% of women experiencing IDA and 23% exhibiting iron deficiency without anemia.⁷ Information for other groups of Indigenous Peoples were not available.

Anemia has a range of causes, so a diagnosis of IDA is based on identification of both anemia and iron deficiency. IDA is diagnosed based on blood tests showing low hemoglobin levels coupled with markers for iron deficiency, typically ferritin levels and transferrin saturation (TSAT).

The key goals in the treatment of iron deficiency and IDA are the correction of the hemoglobin deficit and repletion of iron stores (the correction phase), and maintenance of iron levels over time (the maintenance phase).^2,8 Oral iron, often in the form of iron salts (e.g., ferrous gluconate, ferrous sulphate, or ferrous fumarate), is the first-line therapy for most cases of iron deficiency and IDA, and is relatively safe, effective, and inexpensive. Some patients may be unable to absorb ferrous sulphate adequately because of impaired intestinal uptake resulting from gastrointestinal (GI) disease or clinical conditions such as chronic inflammation, which may in turn lead to elevated levels of hepcidin.^9-11 In other instances, the iron deficit becomes so profound that severe symptomatic anemia results and transfusion of red blood cells may be required.^9,12 Oral iron supplements carry the additional burden of GI side effects, which can have a detrimental impact on patient adherence. There is also a need to deliver iron rapidly in certain clinical situations, including, for example, when a patient with iron deficiency requires an urgent surgery with risk of blood loss.¹³ IV iron treatment may also be preferred over oral iron in some situations, such as for chemotherapy-induced anemia^14-16 and in patients with chronic kidney disease (CKD), with or without concomitant erythropoiesis-stimulating treatment.^17-20 In children, IV iron substitution may be necessary when oral iron therapy is not possible, ineffective, or poorly tolerated. According to Mattiello et al.,²¹ a switch to IV iron is proposed for cases with severe anemia (hemoglobin < 70 g/L) to avoid transfusion, for patients with an underlying secondary disease fulfilling a formal indication for IV iron (e.g., IBD, chronic GI or genitourinary bleeding, or celiac disease), and/or for situations of nonadherence to oral treatment and symptomatic refractory IDA with a clinical impact.²¹

The objective of this Clinical Review is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of ferric carboxymaltose 50 mg/mL by IV infusion in the treatment of IDA in adult and pediatric patients aged 1 year and older when oral iron preparations are not tolerated or are ineffective.

Perspectives of Patients, Clinicians, and Drug Programs

The information in this section is a summary of input provided by the patient and clinician groups who responded to a call for input and from a clinical expert consulted by Canada’s Drug Agency (CDA-AMC) for the purpose of this review.

Patient Input

• One patient group, the Gastrointestinal Society, provided input to this submission. The group obtained input through discussions with health care professionals and researchers, and as well as surveys and interviews conducted on digestive and liver diseases, although the precise number of respondents was not reported.

• The GI Society describes dizziness, extreme fatigue, headache, shortness of breath, tiredness, and muscle weakness as common symptoms of IDA. IDA can have significant impact on a patient, physically, mentally, financially, and socially, and they emphasized the extreme fatigue that can lead to social isolation.

• The GI Society emphasized the tolerability issues associated with oral iron products, and the infusion-related side effects associated with parenteral iron as limitations of treatment. Other limitations include the practical issues associated with having to take time off work and obtain transport to attend infusions.

Clinician Input

Input From Clinical Experts Consulted by CDA-AMC

• The clinical expert consulted for this review noted the key limitations of orally administered iron, namely tolerability, adherence, and delayed onset of effect. The expert noted the importance of having additional options for parenteral iron, as some patients have tolerability issues. The expert also noted that, in many cases, ferric carboxymaltose may be seen as simply another option among several formulations of IV iron.

• The patients most suited for ferric carboxymaltose would be adults with iron deficiency or IDA, those unable to tolerate or respond to oral iron, and preoperative patients with IDA according to the clinical expert; those least suitable would be patients with a history of hypophosphatemia. The expert did not believe there were any issues with diagnosing IDA, with ferritin and TSAT being key tests of iron stores.

• The clinical expert noted that the key outcomes for assessing response include a rise in hemoglobin, improvement in symptoms of iron deficiency and IDA, improved function, and reduced transfusion burden. The clinical expert believed that 4 weeks would be the minimum follow-up for assessing the efficacy of IV iron supplementation, and 12 weeks would be ideal.

Clinician Group Input

• A group of 12 clinicians, including those with specialties in cardiology, obstetrics and gynecology, hematology, gastroenterology, and nephrology, as well as a primary care physician, provided input.

• The clinician group was in general agreement with the input provided by the clinical expert.

• The clinician group was in agreement with the clinical expert that important unmet needs with IV iron supplementation include the need to reduce visits to hospitals and infusion clinics and to have a product available that has evidence supporting its use during pregnancy.

• The clinician group did not elaborate on its experience with ferric carboxymaltose, but did note that it has been available in Europe for 20 years.

Drug Program Input

• The Formulary Working Group (FWG) asked whether there are specific lab thresholds that should be considered for initiation of oral iron, and the clinical expert responded that, in patients with non-CKD, non–chronic heart failure (CHF), non-oncology, the definition of IDA is hemoglobin less than 130 g/L (for males) or less than 120 g/L (for females) and ferritin less than 30 mcg/L. Values for hemoglobin also vary in pregnancy (first and third trimester: < 110 g/L; second trimester: < 105 g/L) and in children (< 24 months: < 105 g/L; 24 to 59 months: < 110 g/L; 5 to 11 years: < 115 g/L). The clinical expert also provided specific cutoffs for ferritin and TSAT based on various conditions such as inflammation (TSAT less than 20% and ferritin less than 100 mcg/L),CKD (TSAT less than 30% and ferritin less than 500 mcg/L), CHF(ferritin less than 100 mcg/L or ferritin less than 300 mcg/L with TSAT less than 20%); and dialysis (ferritin less than 200 mcg/L or TSAT less than 20%).

• The FWG asked how an inadequate response to oral iron is defined and the appropriate length of follow-up before making that assessment. The clinical expert stated that failure to normalize iron stores within 3 months or failure to increase hemoglobin by 10 g/L within 4 weeks indicated an inadequate response. In response to whether there are populations in which oral iron is inappropriate, the clinical expert identified those with bariatric surgery, gastrectomy, IBD, CKD, prior small bowel resection, preoperative IDA (4 weeks or less to surgery), or profound IDA (e.g., symptomatic) and patients with a history of heart failure.

• The FWG asked what the benefits of ferric carboxymaltose are over alternatives such as iron sucrose in patients receiving hemodialysis. The clinical expert responded that, other than a higher dose of iron infused at the session during which it is administered, there is no major benefit in patients coming regularly for dialysis. In response to an FWG question about how many patients would require more than 1 infusion, the clinical expert responded that, at any given time, a minority of patients will receive more than 1 g (1,000 mg), although patients frequently require longstanding treatment (i.e., they need IV iron every 3 months) if they have ongoing risk factors for iron deficiency (e.g., heavy menstrual bleeding or malabsorption).

Clinical Evidence

The sponsor conducted a systematic search-and-selection procedure to identify studies relevant to the efficacy and harms of ferric carboxymaltose relative to relevant comparators in the Canadian setting. The sponsor identified 60 reports of 40 clinical trials, including studies among patients with iron deficiency and heart failure. However, heart failure is outside the scope of this report. The current report focuses only on select studies of IDA identified in the sponsor’s systematic review, with detailed summaries and appraisals of these studies. High-level characteristics and results of the supportive studies submitted to Health Canada as well as studies of the use of ferric carboxymaltose during pregnancy are provided in Appendix 1.

A summary of the clinical evidence included by the sponsor in the review of ferric carboxymaltose is presented in 3 sections. To inform the broad indication under review, the first section, Pivotal Studies and Select Randomized Controlled Trials, includes the pivotal studies submitted to Health Canada for the indication under review and other RCTs from the sponsor’s systematic review that enrolled patients with IDA arising from various causes. The second section includes indirect evidence from the sponsor. The third section summarizes studies addressing gaps and includes 1 study of pediatric patients to inform the use of ferric carboxymaltose among children and adolescents, 1 study among adult patients with IDA arising from various causes to provide additional evidence for the broad indication, and 5 studies comparing the efficacy of ferric carboxymaltose and ferric derisomaltose.

Pivotal Studies and Select Randomized Controlled Trials

Description of Studies

The systematic review portion of this report focuses on 4 studies: the 2 pivotal phase III RCTs per the Health Canada review (VIT-IV-CL-015 and FERGIcor) and 2 phase III RCTs (1VIT05006 and VIT-IRON-2011-004) selected from the sponsor’s submitted systematic review. The pivotal trials focused on specific populations: patients with CKD (VIT-IV-CL-015, N = 240) and those with IBD (FERGIcor, N = 304), with each study randomizing patients 1:1 to either ferric carboxymaltose or iron sucrose. Study 1VIT05006 and Study VIT-IRON-2011-004 included a more heterogeneous IDA population (i.e., patients had IDA arising from various causes). Study 1VIT05006 (N = 559) was a placebo-controlled trial with a crossover design and Study VIT-IRON-2011-004 (N = 371) was a noninferiority study that randomized patients 1:1 to either ferric carboxymaltose or iron sucrose. The primary outcome in 3 of the trials was to assess hemoglobin response, as an increase from baseline. The 1VIT05006 trial was not designed to assess efficacy outcomes.

In the FERGIcor study, the mean age was 52.7 years (standard deviation [SD] = 13.8 years) in the ferric carboxymaltose group and 51.0 years (SD = 12.6 years) in the iron sucrose group, while in VIT-IV-CL-015 and the other studies the mean age was approximately 40 years. In Study VIT-IV-CL-015, approximately 42% of patients were female and in the FERGIcor study 59% were female. In the nonpivotal trials enrolling patients with IDA arising from various causes, approximately 90% of the patients were female. The overall mean hemoglobin level was 93.2 g/L in Study VIT-IV-CL-015, 102.2 g/dL in the FERGIcor study, █████ ███ in Study 1VIT05006, and less than 80 g/L in Study VIT-IRON-2011-004.

Efficacy Results

Hospitalizations (All-Cause and Anemia-Related)

This outcome was not assessed in any of the 4 trials that were the focus of this systematic review section.

Hemoglobin

In the VIT-IV-CL-015 study, the primary response rate was defined as the percentage of patients reaching an increase in hemoglobin of 10 g/L or greater at 4 weeks after baseline. The percentage of patients with a response in the ferric carboxymaltose group was 46.4% (45 of 97), and in the iron sucrose group it was 37.2% (32 of 86), with a reported P value for the between-group difference of 0.2101. The mean increase in hemoglobin levels in the ferric carboxymaltose group at week 4 was ███ ███████ ████ ███ ███ and in the iron sucrose group ███ ███████ ████ ███ ███ with a reported P value for the between-group difference of ████████. No between-group difference and confidence interval (CI) was reported for either end point.

In the FERGIcor study, the primary outcome was patients with a response of an increase in hemoglobin of at least 20 g/L at week 12. In the full analysis set (FAS), the percentage patients with IDA who responded to treatment was 65.79% (150 of 240) in the ferric carboxymaltose group and 53.64% (118 of 220) in the iron sucrose group (Table 17). The between-group difference was 12.15% higher (95% CI, 3.07% to 20.97%; P = 0.004) in the ferric carboxymaltose group compared with that in the iron sucrose group. The FAS set was also analyzed using the last observation carried forward (LOCF) and the worst-case method and the results were consistent with the primary analysis.

In the VIT-IRON-2011-004 study, 99.5% of patients (184 of 185) in each of the ferric carboxymaltose and 98.3% (177 of 180) in the iron sucrose group achieved an increase in hemoglobin of at least 20 g/L from baseline to week 8, for a between-group difference of 1.13 (95% CI, −2.02 to 4.68) (Table 17). The criteria for noninferiority was met. From mean baselines of 77.4 g/L (SD = 14.95 g/L) in the ferric carboxymaltose group and 80.5 g/L (SD = 14.45 g/L) in the iron sucrose group, the mean change from baseline to week 8 was ████ ███ ███████ in the ferric carboxymaltose group and ████ ███ ███████ in the iron sucrose group. The between-group difference was ████ ███ ████ ███ █████ ██ ██████ ███████.

Patient-Reported Health-Related Quality of Life

In the FERGIcor study, for the physical component of the SF-36, from a mean baseline of 44.17 (SD = 7.36) in the ferric carboxymaltose group and 44.98 (SD = 7.23) in the iron sucrose group, there were changes from baseline to week 12 of 3.88 (SD = 6.77) and 2.64 (SD = 7.14), respectively, for a reported between-group P value of 0.157. For the mental component of the SF-36, from mean baselines of 40.02 (SD = 11.04) in the ferric carboxymaltose group and 41.30 (SD = 11.70) in the iron sucrose group, there were changes from baseline to week 12 of 5.91 (SD = 10.74) and 5.56 (SD = 10.36), respectively, for a between-group P value of 0.583. For the Inflammatory Bowel Disease Questionnaire (IBDQ) total score, from mean baselines of 150.8 (SD = 35.2) in the ferric carboxymaltose group and 152.7 (SD = 34.4) in the iron sucrose group, there were changes from baseline to week 12 of 21.1 (SD = 32.3) and 19.7 (SD = 28.8), respectively, for a between-group P value of 0.872. Between-group differences with CIs were not reported for any of the health-related quality of life (HRQoL) measures.

HRQoL was not assessed in VIT-IV-CL-015 or VIT-IRON-2011-004 studies.

Serum Ferritin

In the VIT-IV-CL-015 study, the mean baseline for serum ferritin was ████ █████ ███████ ██ ████ in the ferric carboxymaltose group and ████ █████ ███████ ██ ████ in the iron sucrose group. After week 4, there were changes from baseline of █████ █████ ███████ █████ █████ in the ferric carboxymaltose group and █████ █████ ███████ ███ █████ in the iron sucrose group (Table 17), with a reported between-group P value of ██████. The between-group difference and CI were not reported.

In the FERGIcor study, the mean baseline was █████ █████ ████ in the ferric carboxymaltose group and █████ █████ ████ in the iron sucrose group. After week 4, there were changes from baseline of ████ █████ ████ in the ferric carboxymaltose group and █████ █████ ████ in the iron sucrose group (Table 17) for an estimated between-group difference (repeated measures analysis) of █████████ ███ █████ ██ ████████ ███████.

In the VIT-IRON-2011-004 study, the mean baseline was ████ █████ ███████ in the ferric carboxymaltose group and ████ ██████ ███████ in the iron sucrose group. After week 8, there were changes from baseline of ███ █████ █████████ in the ferric carboxymaltose group and ██████ █████ ████████ in the iron sucrose group, for a least square mean between groups of █████ █████ ████ ███ ██████ ███████ ███████ (Table 17).

Transferrin Saturation

In Study VIT-IV-CL-015, there were █████ ████████ ███████ in the ferric carboxymaltose group and █████ ████████ ███████ in the iron sucrose group who had a normal TSAT of between 20% and 50% by week 4, for a reported between-group P value of ████████. The between-group difference and CI were not reported.

In the FERGIcor study, 117 of 222 patients (52.7%) in the ferric carboxymaltose group and 76 of 209 patients (36.4%) in the iron sucrose group had a normal TSAT level (20% to 50%) at week 12. The odds ratio (OR) was 2.05 (95% CI, 1.37 to 3.06; P < 0.001) in favour of ferric carboxymaltose (Table 17). The absolute between-group difference and CI were not reported.

The TSAT response (achieving a normal TSAT) was not assessed in the VIT-IRON-2011-004 study.

Patients Without Anemia

For patients without anemia in the FERGIcor trial, (a hemoglobin level of ≥ 120 g/L for females or ≥ 130 g/L for males), the percentage of patients at week 12 was 72.8% (166 of 228) in the ferric carboxymaltose group and 61.8% (136 of 220) in the iron sucrose group. The OR was 1.65 (95% CI, 1.10 to 2.46; P = 0.015). The absolute between-group difference and CI were not reported.

Harms Results

Adverse Events

In the overall study population in the VIT-IV-CL-015 trial, ██ ███████ patients in the ferric carboxymaltose group and ██ ███████ patients in the iron sucrose group had at least 1 adverse event (AE). In the FERGIcor study, AEs were reported in ████████ patients in the ferric carboxymaltose group and in ███ ███████ patients in the iron sucrose group.

In the overall study population in the 1VIT05006 trial, ███ ███████ patients in the ferric carboxymaltose group and ███ ███████ patients in the placebo group reported at least 1 AE. In the VIT-IRON-2011-004 study, ███ ███████ patients in the ferric carboxymaltose group and ███ ███████ patients in the iron sucrose group had AEs.

Serious Adverse Events

In the VIT-IV-CL-015 study, serious adverse events (SAEs) were reported in | ██████ patients in the ferric carboxymaltose group and ██ ██████ patients in the iron sucrose group. In the FERGIcor study, SAEs were reported in ██ ██████ patients in the ferric carboxymaltose group and ██ ██████ patients in the iron sucrose group.

In the 1VIT05006 study, | ██████ patient in the ferric carboxymaltose group and | ██████ patients in the placebo group had an SAE. In Study VIT-IRON-2011-004 ██ patients ██████ in the ferric carboxymaltose group and | ██████ patients in the iron sucrose group had an SAE.

Withdrawals Due to Adverse Events

In the VIT-IV-CL-015 study, | ██████ patients in the ferric carboxymaltose group and ███████ patients in the iron sucrose group withdrew from study medication due to AEs. In the FERGIcor study, 7 patients (2.9%) in the ferric carboxymaltose group and 2 patients (0.8%) in the iron sucrose group withdrew from the study drug because of an AE.

In Study 1VIT05006, | ██████ patients in the ferric carboxymaltose group and | ██████ patients in the placebo group withdrew from study medication due to an AE. In the VIT-IRON-2011-004 study, | ██████ patients in the ferric carboxymaltose group and ██████ patients in the iron sucrose group withdrew from study medication due to an AE.

Mortality

Across the studies, ███ ███████ in the ferric carboxymaltose group died during Study VIT-IV-CL-015 because of an acute anterior myocardial infarction. ███ ███████ died more than a week after study medication was withdrawn due to a nonserious AE. In Study 1VIT05006, ███ ███████ in the ferric carboxymaltose group died due to pneumonia. There were no deaths reported in the other 2 studies.

Notable Harms

In the FERGIcor study, hypophosphatemia was observed in 6 patients (2.5%) in the ferric carboxymaltose group and none were observed in the iron sucrose group. Hypophosphatemia was not reported in the VIT-IV-CL-015 study.

In the VIT-IRON-2011-004 study, hypophosphatemia was observed in ██ patients in the ferric carboxymaltose group ███████ and in ██ patients ██████ in the iron sucrose group. The number of patients with decreased blood phosphorous or hypophosphatemia was not reported in Study 1VIT05006.

Critical Appraisal

• None of the active-controlled trials were blinded, and this may bias assessment of patient-reported outcomes such as HRQoL as well as the assessment of subjective harms. One of the 3 studies did not plan for any formal comparisons between ferric carboxymaltose and iron sucrose, while the other 2 did not implement a multiple-testing procedure, increasing the risk that statistically significant results (beyond the primary outcome) were false positives. Fewer patients in the ferric carboxymaltose group compared to the iron sucrose group withdrew from Study VIT-IV-CL-015 (████ ██████ █████), and this may also bias assessment of efficacy and harms in this relatively short (4-week) study. Between-group differences with CIs were infrequently reported, limiting the usefulness of judgments about the precision of the effect estimates.

• Few if any patients in the included studies identified as Indigenous, and this is an important gap because there is a disproportionate number of patients with IDA in Canada who are Indigenous.

• One of the pivotal studies, FERGIcor, assessed HRQoL, but the other studies did not, and no studies formally assessed the impact of ferric carboxymaltose on fatigue and other important outcomes in this patient population.

GRADE Summary of Findings and Certainty of the Evidence

In the absence of a complete body of evidence for any patient population or comparison, no Grading of Recommendations Assessment, Development and Evaluation (GRADE) assessment was performed for this review.

Table 2: Results for Ferric Carboxymaltose Versus Iron Sucrose for Patients With Iron Deficiency Anemia

ANCOVA = analysis of covariance; CFB = change from baseline; CKD = chronic kidney disease; CI = confidence interval; FAS = full analysis set; HRQoL = health-related quality of life; IBD = inflammatory bowel disease; IBDQ = Inflammatory Bowel Disease Questionnaire; NR = not reported; OR = odds ratio; PPS = per-protocol set; SF-36 = Short-Form (36) Health Survey; TSAT = transferrin saturation.

^aThe statistical analysis plan for this study did not call for any formal hypothesis testing; therefore, all reported P values should be considered supportive in nature.

^bThese P values have not been adjusted for multiple comparisons.

Source: Sponsor Summary of Clinical Evidence and the CSR for FERGIcor, VIT-IV-CL-015, and VIT-IRON-2011-004

Long-Term Extension Studies

There were no long-term extension studies submitted by the sponsor.

Indirect Comparisons

The sponsor submitted a summary of indirect evidence from a published network meta-analysis (NMA) that compared the effects of ferric carboxymaltose with those of other IV iron therapies in the context of adults with IDA associated with IBD.²²

Description of Studies

The NMA submitted by the sponsor included 5 RCTs. All studies were performed in patients with IBD. The interventions in the studies included ferric carboxymaltose, iron sucrose, iron isomaltoside (also known as ferric derisomaltose), and oral iron.

Efficacy and Harms Results

Point estimates for the ORs comparing ferric carboxymaltose with ferric derisomaltose and iron sucrose favoured ferric carboxymaltose; however, the 95% credible intervals (CrIs) were wide, included effects favouring the comparator interventions, and showed no differences. Harms were not assessed in the NMA. Therefore, the summary of the published NMA submitted by the sponsor was insufficient to determine whether there is a difference in efficacy or harms for ferric carboxymaltose compared to other injectable forms of iron in patients with IDA and inflammatory bowel disease.

Critical Appraisal

The sponsor selected a published NMA that used a Bayesian approach with fixed-effects models for the analyses. Limited information is available regarding the methods used in the NMA model. The NMA was informed by a systematic review of relevant databases, although the search may not have been exhaustive. The date last searched was June 2016, more than 8 years ago at the time of writing this report. Relevant studies available since this time would have been excluded from the analyses. The authors performed an assessment of bias for the included studies but did not explain how the results of this assessment were incorporated into the analyses. It was not clear if the authors incorporated an assessment of clinical heterogeneity into their analyses.

Another limitation of the NMA is that it contained a small amount of data from 5 studies, resulting in a network with only 1 or 2 studies connecting the nodes, along with wide CrIs. Only 1 outcome was assessed in the NMA. There was no analysis of harms or other outcomes that are important for patients.

There was heterogeneity in the time point for end point evaluation, which could have biased the results. Few patient characteristics were reported across the trials, challenging a thorough assessment of the plausibility of the exchangeability assumption. Interpretation of the effect estimates was limited by imprecision. The 95% CrIs for the comparisons of ferric carboxymaltose and ferric derisomaltose and iron sucrose were wide and included the potential for no difference or that either treatment could be favoured.

Another significant limitation is related to the sponsor’s lack of clear, a priori criteria for selecting the NMA by Aksan et al.²² At least 2 other published NMAs have investigated the relative efficacy of ferric carboxymaltose in broader IDA populations. There was no protocol outlining a priori how 1 or more NMAs would be selected for presentation among multiple NMAs reporting on the same population, comparisons, and outcomes. As such, there is risk of bias in the selection of the NMA.

Summary of Indirect Treatment Comparison

A summary of the published NMA submitted by the sponsor was insufficient to determine whether there is a difference in efficacy for ferric carboxymaltose compared to other injectable forms of iron in patients with IDA and IBD. Important efficacy and harms outcomes were not investigated.

Studies Addressing Gaps in the Evidence from Pivotal Studies and Select RCTs

This section presents additional evidence from 6 studies that address gaps in the evidence from pivotal trials and select RCTs: 1 study in pediatric patients, 1 study in patients with IDA resulting from various causes, and 4 studies comparing ferric carboxymaltose with ferric derisomaltose in patients with IDA.

Study 1VIT17044

Study 1VIT17044 is a phase III, multicentre, randomized, active-controlled, open-label clinical trial conducted at 30 sites across 4 countries (the US, Ukraine, Poland, and Canada). The trial enrolled 79 patients who were assigned to receive ferric carboxymaltose (n = 40) or oral iron (n = 39). The trial aimed to investigate the efficacy and safety of ferric carboxymaltose versus oral iron in pediatric patients with IDA and a documented history of inadequate response to oral iron. In the 1VIT17044 trial, the least square mean change in hemoglobin from baseline to day 35 obtained through an analysis of covariance (ANCOVA) model was 2.22 g/dL (██ █ █████) in the ferric carboxymaltose group and 1.92 g/dL (██ █ █████) in the oral iron group. The treatment difference at day 35 was 0.30 g/dL (95% CI, −0.28 t 0.88; P = 0.3108). The results of the mixed model for repeated measures (MMRM) and the subgroup analyses align with the main analysis. The least square mean change in ferritin from baseline to day 35 obtained through the ANCOVA model were ██████ █████ ███ █ ███████ in the ferric carboxymaltose and █████ █████ ███ █ ███████ in the oral iron group. The treatment difference at day 35 was ██████ █████ ████ ███ ██████ ███████ ████████████████. The least square mean change in TSAT from baseline to day 35 obtained through the ANCOVA model was ██████ ███ █ ██████ in the ferric carboxymaltose and █████ ███ █ ██████ in the oral iron group. The treatment difference at day 35 was ██████ ████ ███ █████ ██████ ████████████████. In the 1VIT17044 trial, a larger percentage of participants in the ferric carboxymaltose group than in the oral iron group experienced at least 1 treatment-emergent adverse event (TEAE) ██ ███████ versus ██████████. Numerically more patients in the ferric carboxymaltose group than in the oral iron group experienced metabolism and nutrition disorders ████ ███ ███ and hypophosphatemia ████ ███ ███. Numerically fewer patients in the ferric carboxymaltose group than the oral iron group experienced GI disorders █████ ███ ██████ and constipation ███ ███ ██████. One patient in the ferric carboxymaltose group experienced a TEAE that led to treatment discontinuation. None of the patients experienced any SAE or any TEAE leading to death.

Critical Appraisal

Internal validity: Although the methods for randomization were likely appropriate, due to the small sample size there is an increased risk that prognostic balance was not achieved, as evidenced by imbalances in patients’ baseline disease and demographic characteristics. Notably, the baseline serum ferritin level was higher in the oral iron arm. There were also baseline imbalances by ethnicity and by body mass index (BMI). The effect of these differences on efficacy outcomes is unclear. The trial was open-label; however, the efficacy outcomes are objective, so it is unlikely that bias was introduced to the measurements. There is a risk of bias in the reporting of subjective harms (e.g., GI disorders and headaches) because patients knew which treatment they had been assigned (e.g., it is possible that known harms could be overestimated). The authors used the LOCF method to impute missing outcomes data and conducted sensitivity analyses using an MMRM under the missing-at-random assumption. Although neither method may be appropriate (the LOCF may not be reflective of the true trajectory of the outcome and the MMRM assumes data are missing at random, which is not possible to assess and may not be plausible), the attrition rate was low (5% or less) in each group. As such, it is unlikely that missing data would have introduced bias. The intention-to-treat (ITT) analysis was appropriate for estimating the effect of assignment to the interventions. Because there were no adjustments for multiple comparisons, there is an increased risk of type I error (false positives) for statistically significant results. Although the subgroup analyses were preplanned, these were unlikely sufficiently powered to detect subgroup differences.

External validity: The included patients were aged 1 to 17 years and results are not generalizable to other age groups. Further, given the small sample size, it is unlikely that the results would be broadly generalizable to all pediatric patients with IDA. As the comparator in the trial was oral iron, this study does not inform about the efficacy or harms of ferric carboxymaltose relative to other IV iron formulations in pediatric patients. Indigenous Peoples, who are disproportionately affected by IDA, are not represented in this trial. Although the outcomes measures were appropriate, some outcomes that may be important to patients (e.g., HRQoL) were not reported.

Study 1VIT09031

The 1VIT09031 trial is a phase III, multicentre, randomized, active-controlled, open-label study of the efficacy and safety of IV ferric carboxymaltose in adult patients with IDA who had an unsatisfactory response or intolerance to oral iron. Cohort assignment was based on results from a 14-day run-in period with oral iron. Patients with inadequate response to oral iron (hemoglobin increase < 1 g/dL) were assigned to cohort 1, and patients who were intolerant of oral iron were assigned to cohort 2. Oral iron was the comparator arm in cohort 1, and other IV iron standard of care (SOC) as chosen by the investigator was the comparator in cohort 2. In cohort 1, ███ patients received ferric carboxymaltose and ███ received oral iron. In cohort 2, ███ patients received ferric carboxymaltose and ███ received IV SC. Patients were followed up to day 35 for efficacy assessment, and up to day 120 for safety assessment. For the protocol-specified primary treatment group comparison (cohort 1), the mean increases in hemoglobin from baseline to the highest value between baseline and day 35 or time of intervention (TOI) were 1.57 g/dL ███ █ █████ in group A (ferric carboxymaltose) and 0.80 ███ █ █████ in group B (oral iron) (P = 0.001). No between-group difference and CI were reported. In a post hoc comparison of group C (ferric carboxymaltose) versus group D (IV SOC) (cohort 2), the mean increases in hemoglobin from baseline to the highest value between baseline and day 35 or TOI were 2.90 g/dL ███ █ █████ in group C (ferric carboxymaltose) and 2.16 ███ █ █████ group D (IV SOC). No between-group difference and CI were reported. Subgroup analysis revealed the mean increase in hemoglobin from baseline to the highest value between baseline and day 35 or TOI was greater for the ferric carboxymaltose group than the comparator group regardless of baseline hemoglobin value or etiology of IDA.

In cohort 1, the proportion of patients achieving hemoglobin above 12.0 g/dL was █████ in the ferric carboxymaltose group and █████ in the oral iron group (P < ████). The proportion of patients with a clinically meaningful increase in hemoglobin (as defined by the investigators) was █████ in the ferric carboxymaltose group and ████ in the oral iron group (P < ████). The proportion of patients with a hemoglobin level greater than 12 g/dL and a ferritin increase of 160 ng/mL or greater was █████ in the ferric carboxymaltose group and ████ in the oral iron group (P < ████). The mean change in hemoglobin was ███ g/dL (SD = ███) in the ferric carboxymaltose group and ███ g/dL (SD = ███) in the oral iron group (P < ████). The means of changes in ferritin were █████ ng/mL (SD = █████) in the ferric carboxymaltose group and ████ ng/mL (SD = ████) in the oral iron group (P < ████). The mean change in TSAT was █████ (SD = ████) in the ferric carboxymaltose group and █████ (SD = ████) in the oral iron group (P < ████). Between-group differences and CIs were not reported for any outcome.

In cohort 2, the proportion of patients achieving hemoglobin above 12.0 g/dL was █████ in the ferric carboxymaltose group and █████ in the IV iron group (P < ████). The proportion of patients with a clinically meaningful increase in hemoglobin (as defined by the investigators) was █████ in the ferric carboxymaltose group and █████ in the IV iron group (P < ████). The proportion of patients with a hemoglobin level greater than 12 g/dL and a ferritin increase of 160 ng/mL or greater was █████ in the ferric carboxymaltose group and ████ in the IV iron group (P < ████). The means of changes in hemoglobin were ███ g/dL (SD = ███) in the ferric carboxymaltose group and ███ g/dL (SD = ███) in the IV iron group (P < ████). The mean change in ferritin was █████ ng/mL (SD = █████) in the ferric carboxymaltose group and ████ ng/mL (SD = █████) in the IV iron group (P < ████). The mean change in TSAT was ██████ (SD = ████) in the ferric carboxymaltose group and ████ (SD = ████) in the oral iron group (P < ████). Between-group differences and CIs were not reported for any outcome.

There were numerically more TEAEs in group A ███████ compared with group B ███████. Overall rates of TEAEs were similar between group C (ferric carboxymaltose) and group D (IV SOC) (██ ██ ███ across groups). The TEAS most commonly experienced (≥ 5%) were nausea in group A ██████, hypophosphatemia in group C ██████, and dizziness in group D ██████. No TEAE was experienced in 5% or more of patients in group B. █████ ██████ patients in group A (ferric carboxymaltose), ██ ██████ patients in group B (oral iron), ██ ██████ patients in group C (ferric carboxymaltose), and ████████ patients in group D (IV iron) experienced at least 1 SAE during the treatment phase. █████ ██████ patients in group A (ferric carboxymaltose) and ████ in the group B (oral iron) experienced hypersensitivity reactions. ███ ██████ patients in group C (ferric carboxymaltose) and █████ in group D (IV iron) experienced hypersensitivity reactions. Hypersensitivity reactions in the patients receiving ferric carboxymaltose were either grade 2 or 3, and in the IV SOC group they were grades 1 to 3. ██████ ██████ patients in group A (ferric carboxymaltose) and ██████ in group B (oral iron) experienced skin and subcutaneous tissue disorders. ████████ ██████ patients in group C (ferric carboxymaltose) and ████████ in group D (IV iron) experienced skin and subcutaneous tissue disorders. All TEAEs associated with skin and subcutaneous tissue disorders (erythema, pruritus, rash, maculopapular rash, and urticaria) were either grade 1 or 2. In group A (ferric carboxymaltose), ████ of patients experienced hypophosphatemia and █████ experienced potentially clinically significant (PCS) low-phosphorus levels compared with ██ ███ ████ in group B (oral iron), respectively. In group C (ferric carboxymaltose), ████ of patients experienced hypophosphatemia and █████ experienced PCS low-phosphorus levels compared with ██ ███ ████ in group D (IV SOC), respectively. Hypophosphatemia was mainly grade 1 to 3 in severity, with a single grade 4 event. ████ PCS low-phosphorus events were grade 3, with ██ grade ██ events in patients in the ferric carboxymaltose group. ██ PCS low-phosphorus events were associated with serious or severe AEs. ███ patient had events of somnolence, fatigue, tingling finger, swollen hand, and an elevated white blood cell count on days when phosphorus levels were grade 4. In total, ██ deaths were reported in the study, ██ in group B (oral iron), ██ in group C (ferric carboxymaltose), and ███ in group D (IV iron). ████ were considered related to study drug by investigator.

Critical Appraisal

Internal validity: Although the methods for randomization and allocation concealment appeared to be adequate, imbalances were evident at baseline in some important clinical characteristics. Most notably, serum ferritin was higher in group C (ferric carboxymaltose) compared with group D (IV SOC). It is not certain if this imbalance may have biased the results. The trial was open-label; however, the efficacy outcomes are objective, so it is unlikely that bias was introduced in their measurement. There is a risk of bias in the reporting of subjective harms because patients knew which treatment they had been assigned (e.g., it is possible that known harms could be overestimated). Across groups, up to ███ of patients did not complete the study, and methods for handing missing data are not clear. As such, there is a risk of bias due to missing outcomes data, but the extent and direction of the bias cannot be predicted. The modified intention-to-treat (mITT) analysis was appropriate for estimating the effect of assignment to the interventions. Although the mITT population was a subset of the ITT population, few patients (< 4% across groups) were excluded from the mITT analysis set. Because there were no adjustments for multiple comparisons, there is an increased risk of type I error (false positives) for statistically significant results. Between-group differences and CIs were not reported for any outcome, precluding judgments about the precision of the effects.

External validity: In this study, all patients received 2 doses of 15 mg/kg, up to a maximum single dose of 750 mg and a maximum cumulative dose of 1,500 mg. These values were all lower than the recommended dosing in the product monograph, which recommends a maximum single relative dose of 20 mg/kg, a maximum single absolute dose of 1,000 mg, and a maximum total dose of 2,000 mg. In group D, patients could have been assigned to any of 5 IV iron formulations, only 2 of which were considered relevant comparators for this review. As such, the generalizability of the results may be limited. The comparison of group A (ferric carboxymaltose) to group B (oral iron) does not inform the efficacy and harms of ferric carboxymaltose relative to other IV oral formulations available in Canada, limiting the applicability of these results. The study was conducted at 84 centres in the US only, with no patients from Canada and no representation of Indigenous Peoples, who are disproportionately affected by IDA. As only adults were enrolled in the study, it is uncertain whether the results can be generalized to pediatric patients. For group A (ferric carboxymaltose) and group B (oral iron), the study selected patients based on adherence to oral iron following a run-in phase, and a large proportion of these patients █████ were not enrolled. As such, the enrolled patients may not be representative of patients seen in clinical practice (as patients with lower adherence were not enrolled). Although relevant outcomes were investigated in the study, other outcomes that may be important to patients (e.g., HRQoL) were not investigated.

Zoller (2023)

Zoller (2023) is a multicentre, randomized, double-blind, active-controlled, clinical trial conducted at 20 outpatient hospital clinics in Austria, Denmark, Germany, Sweden, and the UK. The trial compared the incidence of hypophosphatemia after treatment with ferric carboxymaltose (N = 48 patients) versus ferric derisomaltose (N = 49 patients) in 97 patients with IDA and IBD. By day 70 (the end of the trial), the levels of ferritin and TSAT increased in both treatment groups. The hemoglobin increase by day 70 in the ferric derisomaltose group was 24.9 g/L (95% CI, 21.1 g/L to 28.8 g/L) and in the ferric carboxymaltose group was 25.2 g/L (95% CI, 21.3 g/L to 29.1 g/L). The between-group difference and CI were not reported for any efficacy outcome. Both ferric derisomaltose and ferric carboxymaltose resulted in improvement in fatigue symptoms and increased Functional Assessment of Chronic Illness Therapy–Fatigue (FACIT-F) scale scores (possible scores range from 0 to 52), which was statistically significantly greater for patients treated with ferric derisomaltose versus ferric carboxymaltose at days 35 and 49. The between-group difference and CI were not reported at any follow-up time point.

Numerically, more patients experienced hypophosphatemia and vitamin D deficiency in the ferric carboxymaltose group compared with those in the ferric derisomaltose group (hypophosphatemia: 28.6% versus 2.1% and vitamin D deficiency: 34.7% versus 22.9%, respectively). Numerically, fewer patients experienced headache and nausea in the ferric carboxymaltose group compared with those in the ferric derisomaltose group (headache: 10.2% versus 18.8%; and nausea: 2.0% versus 12.5%, respectively). Discontinuation due to AEs occurred among 6.3% of patients in the ferric derisomaltose group and 2.0% of patients in the ferric carboxymaltose group. There were no deaths in the trial. Hypophosphatemia in this trial was defined as a serum phosphate level of less than 2.0 mg/dL. The primary end point was the incidence of hypophosphatemia at any time after the first dose to day 35, which was reported as 8.3% (4 of 48) in the ferric derisomaltose group and 51.0% (25 of 49) in the ferric carboxymaltose group with an adjusted risk difference of −42.8% (95% CI, –57.1% to –24.6%; P < 0.0001) favouring ferric derisomaltose. The majority of patients recovered from hypophosphatemia by day 70. In an analysis by diagnosis of IBD in the ferric carboxymaltose group, the risk differences were 43.1% and 45.5% higher for patients with ulcerative colitis and Crohn disease, respectively (interaction P value = 0.1948). The highest incidence of hypophosphatemia occurred within 2 weeks of treatment in both arms. The secondary safety end point was the incidence of hypophosphatemia at any time from baseline to day 70, which occurred among 12.5% (6 of 48) of patients in the ferric derisomaltose group and 59.2% of patients (29 of 49) in the ferric carboxymaltose group, with an adjusted risk difference of −46.6% (95% CI, −60.9% to –28.1%; P < 0.0001) favouring ferric derisomaltose. Per the investigators, the mean decreases in phosphate concentration from baseline after the first and second doses were significantly greater after ferric carboxymaltose infusion compared to ferric derisomaltose infusion (between-group differences and CIs were not reported). For more than 1 month after the second infusion, 4.7% (2 of 43) of patients in the ferric carboxymaltose group remained hypophosphatemic. On day 70 the mean serum phosphate remained significantly lower in the ferric carboxymaltose group compared to the ferric derisomaltose group (between-group differences and CIs were not reported).

Critical Appraisal

Internal validity: Although the methods for randomization appeared appropriate, there is an increased risk that prognostic balance was not achieved due to the small sample size, as evidenced by imbalances in some baseline disease and demographic characteristics. Notably, there were imbalances at baseline in IBD diagnosis (Crohn disease or ulcerative colitis). The use of some concomitant medications (e.g., interleukin inhibitors and vitamin D supplements) was also imbalanced between groups. The effect of these differences on the efficacy and safety results is uncertain. Although the absolute risk differences for hypophosphatemia among patients with Crohn disease and ulcerative colitis who were treated with ferric carboxymaltose versus ferric derisomaltose were similar (and the interaction P value was not statistically significant), the subgroups were small, and the analysis was likely not sufficiently powered to detect subgroup differences. As the trial was double-blinded and methods to maintain the blinding appeared to be adequate, there is likely a low risk of bias in the measurement of the outcomes. Considering the importance of patient-reported outcomes, the use of the FACIT-F scale, which is a reliable and valid instrument for measuring fatigue in IBD, was appropriate.²³ Because the trial was double-blinded, the risk of bias due to reporting subjective patient-reported outcomes is low. However, judgments about the precision of the effects are precluded because the between-group difference and CIs were not reported at any follow-up time point. For the analysis of hypophosphatemia, 2 patients in the ferric derisomaltose group and 1 patient in the ferric carboxymaltose group did not have a postbaseline observation and were imputed as having hypophosphatemia in the primary analysis. As the proportion of patients with missing data was low for this outcome, there is likely a low risk of bias due to missing outcomes data. Further, a post hoc sensitivity analysis in which these patients were imputed as either being free of hypophosphatemia or were excluded yielded similar results to the primary analysis. There was no imputation of missing values in this trial, except for the change from baseline in patients with no postbaseline measurements, which was set to zero at the first postbaseline visit. Furthermore, the attrition rates were 10% in the ferric derisomaltose group and 12.5% in the ferric carboxymaltose group; as a result, the risk of bias because of missing data is not high. Statistical analyses in this trial were not adjusted for multiple comparisons. As such, there is an increased risk of false-positive conclusions for statistically significant results. Particularly for efficacy outcomes (e.g., change in hemoglobin, ferritin, TSAT levels and fatigue), between-group differences with CIs were not reported, and no conclusions could be made about the magnitude (including clinical importance) of the estimated effects and their precision.

External validity: All patients enrolled in this trial had IBD; however, the results may not be broadly generalizable given the small sample size of the trial. The length of follow-up was relatively short, which limits the usefulness of conclusions about efficacy and safety over a longer period and among patients who require chronic treatment with IV iron. Although the efficacy outcomes measured were appropriate, conclusions about the magnitude of the estimated effects and their precision were limited due to insufficient reporting of between-group differences and CIs. Indigenous Peoples, who are disproportionately affected by IDA, were not represented in this study. In this trial, patients received a single IV infusion of 1,000 mg at baseline (day 0) and, depending on the a priori calculated iron dose, either 500 mg or 1,000 mg at day 35. According to the product monograph for ferric carboxymaltose, a single dose should not exceed 15 mg iron/kg of body weight or 1,000 mg of iron. Based on the product monograph for ferric derisomaltose²⁴ the allowable iron dose per infusion is limited to 20 mg iron/kg of body weight. The dosage administered in this trial therefore aligns with those of the product monographs.

Emrich (2020)

Emrich (2020) is a prospective, single-centre, double-blind study. The study randomized 26 women with IDA to receive ferric carboxymaltose (n = 13) and or ferric derisomaltose (n = 13). This trial aimed to assess hypophosphatemia after high-dose iron repletion with ferric carboxymaltose and ferric derisomaltose. The trial assessed the quality of life using the SF-36, functional impairment by the Sheehan disability scale, and fatigue by the German version of the Multidimensional Fatigue Inventory, but the results have not been published. Changes in the levels of hemoglobin, ferritin, and TSAT were not trial end points and were not reported. The primary outcome was hypophosphatemia at any postinfusion study visit. This was reported for 75% (9 of 12) of the patients in the ferric carboxymaltose group and 8% (1 of 13) of the patients in the ferric derisomaltose group, as measured at study visit 4 (days 5 to 9) (P = 0.001). At study visit 5 (days 33 to 37), 25% (3 of 12) and 8% (1 of 13) of patients in the ferric carboxymaltose and ferric derisomaltose groups, respectively, had hypophosphatemia. The between-group difference with CI was not reported at any follow-up time point.

Critical Appraisal

Internal validity: The methods of randomization appeared to be appropriate; however, there are imbalances in some baseline characteristics that may be due to the small sample size. Furthermore, due to logistical reasons, after the inclusion of 26 patients (instead of the estimated 30 to achieve 80% power) the interim analysis was conducted, and the trial was completed after interim analysis. Therefore, there is the possibility that if more patients had been enrolled, then the effect size may have been differed. Due to lack of information in the publication, the adequacy of blinding is unclear. As only 1 patient was excluded from the analyses, the risk of bias because of missing data is low. There is an increased risk of false-positive conclusions for statistically significant results, as the statistical analyses were not adjusted for multiple comparisons. The between-group differences and CIs were not reported for the primary outcome of hypophosphatemia, precluding conclusions about the magnitude of the estimated effect and its precision.

External validity: Given the small sample size of the trial and considering that all patients enrolled in this trial were female and Caucasians (wording used in the source) the results may not be broadly generalizable. Further, the results also are not generalizable to the patients who were excluded from the study because of advanced CKD, pregnancy, ongoing lactation, untreated hyperparathyroidism, hemochromatosis, active malignancy, bronchial asthma, atopic dermatitis, active alcohol or drug abuse, or a history of a psychological illness or seizures. The length of follow-up was relatively short, which limits conclusions about the safety result (hypophosphatemia) over a longer duration. In this trial, only 1 infusion was performed, which limits detection of consequences of repeated infusions.

Wolf (2020)

Wolf (2020) reports on 2 identically designed, open-label, randomized clinical trials that assessed the effects of ferric carboxymaltose and ferric derisomaltose on hypophosphatemia. In trial A, 123 patients were randomized to receive ferric carboxymaltose (N = 61) or ferric derisomaltose (N = 62). In trial B, 122 patients were randomized to receive ferric carboxymaltose or ferric derisomaltose (N = 61 in each group). In total, 122 patients were randomized to receive ferric carboxymaltose and 123 to receive ferric derisomaltose. The results of hemoglobin, ferritin, and TSAT measurements from trials A and B are presented; however, changes from baseline and between-group differences with CIs were not reported for hemoglobin, ferritin, and TSAT outcomes at any time point.

Overall, in the ferric carboxymaltose versus ferric derisomaltose groups, 27 of 60 patients (45.0%) versus 7 of 63 patients (11.1%) patients in trial A and 28 of 57 (49.1%) versus 14 of 62 (22.6%) in trial B experienced AEs. Serious or severe hypersensitivity reactions occurred in 1 patient (0.8%) in the ferric derisomaltose group (swollen eyelid unilaterally) and in 2 patients (1.7%) in the ferric carboxymaltose group (dyspnea and swelling). Regarding the specific adverse drug reaction, in the ferric carboxymaltose versus ferric derisomaltose groups, 12 (20.0%) versus 0 patients in trial A and 7 (12.3%) versus 0 in trial B experienced decreased blood phosphorus. In the ferric carboxymaltose versus ferric derisomaltose groups, 12 (20.0%) versus 0 patients in trial A and 14 (24.6%) versus 2 (3.2%) in trial B experienced hypophosphatemia. The incidence of hypophosphatemia at any time from baseline to day 35 in the ferric derisomaltose group compared with the ferric carboxymaltose group in trial A was 7.9% versus 75.0% (adjusted rate difference = −67.0%; 95% CI, −77.4% to −51.5%; P < 0.001) favouring ferric derisomaltose, and in trial B it was 8.1% versus 73.7% (adjusted rate difference = −65.8%; 95% CI, −76.6% to −49.8%; P < 0.001) favouring ferric derisomaltose.

Critical Appraisal

Internal validity: Both trials A and B were open-label randomized clinical studies. Although the methods for randomization appeared to be appropriate, there were some imbalances in baseline characteristics, such as sex and race, and minor imbalances in ferritin and TSAT levels. These imbalances may have resulted from the small sample sizes of the included studies, which increases the risk that prognostic balance between groups may not have been achieved. The open-label nature of the study may increase the risk of bias in determining the magnitude of the subjective safety outcomes. The efficacy outcomes are objective and unlikely to be at risk of bias on account of the open-label design. However, the changes from baseline and between-group differences with CIs were not reported for some of the efficacy outcomes (hemoglobin, ferritin, and TSAT) which precludes making judgments about the precision of the effects. There was no adjustment for multiple comparisons, which increases the risk of type I error (false positives) for statistically significant results. There was no imputation of missing values in this trial, except for the change from baseline in patients with no postbaseline measurements, which was set to zero at the first postbaseline visit. The risk of bias due to missing data on outcomes is low, considering the low rate of attrition in both trials.

External validity: The included patients were mostly women with IDA due to gynecological bleeding. In both trials, most patients were white; therefore, results may not be generalizable to a broader population. Although the outcome measures were appropriate, some outcomes that may be important to patients (e.g., HRQoL) were not reported. Patients with conditions such as alcohol or drug abuse, pregnancy or lactation, untreated hyperparathyroidism, kidney transplant, a body weight less than 50 kg, or hemochromatosis or other iron-storage disorder were excluded, and the results are not generalizable to these groups of patients. In this study, all patients in the ferric carboxymaltose groups received 750 mg of ferric carboxymaltose on days 0 and 7, whereas those in the ferric derisomaltose group received 1,000 mg on day 0. According to the product monograph for ferric carboxymaltose, a single dose should not exceed 15 mg iron/kg of body weight or 1,000 mg of iron. Based on the product monograph for ferric derisomaltose, the allowable iron dose per infusion is limited to 20 mg iron/kg of body weight. While the dose administered aligned with the product monographs, patients in the ferric carboxymaltose arm received 500 mg more iron compared with those in the ferric derisomaltose arm. Because of the short follow-up time, this study does not inform the long-term clinical implications of these drugs. Additionally, the clinical outcomes associated with hypophosphatemia have not been reported.

Conclusions

Results from 2 phase III pivotal trials in patients with IDA secondary to IBD and to CKD, and another phase III RCT selected because it included patients with more generic IDA, suggest that ferric carboxymaltose is at least as effective at increasing hemoglobin as iron sucrose, after 4 to 12 weeks of follow-up. None of the included studies were designed to formally assess other efficacy outcomes, such as TSAT and ferritin levels, or important outcomes such as HRQoL; however, there was no indication from the included trials of a notable, clinically significant difference between ferric carboxymaltose and iron sucrose for any of these outcomes. There was no indication of a clear difference in risk of AEs or SAEs associated with ferric carboxymaltose versus iron sucrose, with the exception of hypophosphatemia, which may occur more frequently with ferric carboxymaltose. There is evidence that ferric carboxymaltose may be efficacious and safe for use in pregnancy and in children, and the clinical expert consulted for this review noted that there is a shortage of parenteral iron formulations with evidence in these populations. There is also evidence that the efficacy of ferric carboxymaltose is comparable to that of ferric derisomaltose; however, there is clearly an increased risk of hypophosphatemia with ferric carboxymaltose versus this other iron formulation. The clinical significance of this increased risk of hypophosphatemia is not known.

Introduction

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of ferric carboxymaltose 50 mg/mL IV infusion in the treatment of IDA in adult and pediatric patients aged 1 year and older when oral iron preparations are not tolerated or are ineffective.

Disease Background

Contents within this section have been informed by materials submitted by the sponsor and clinical expert input. The following have been summarized and validated by the CDA-AMC review team.

Overview of the Condition

Iron deficiency is defined heterogeneously, but can be characterized as a condition in which iron availability is insufficient to meet the body’s needs.²⁵ The British Columbia clinical guidelines on the diagnosis and management of iron deficiency define the condition as “insufficient total body iron stores, caused by increased requirements, decreased intake, increased loss, and/or decreased absorption.”⁴ Prolonged iron deficiency can lead to anemia.

Anemia is a medical condition in which the blood has reduced capacity to carry oxygen, either due to a reduced number of red blood cells or low hemoglobin levels. Anemia has a range of causes; however, iron deficiency is the most prevalent.¹ IDA results when body iron levels are insufficient to sustain hemoglobin synthesis.² Alberta clinical practice guidelines define IDA as “a serious condition whereby low levels of iron are associated with anemia and the presence of microcytic hypochromic red cells.”³

The British Columbia clinical guidelines define anemia based on hemoglobin levels relative to the population mean, as “[a] low hemoglobin level, most frequently defined as a hemoglobin value over 2 standard deviations below the gender- and age-adjusted mean. A hemoglobin value below the local, lab-specific lower reference interval indicates anemia.”⁴ Patients with IDA report dizziness, extreme fatigue, headache, shortness of breath, and muscle weakness as frequent symptoms of their condition.

Prevalence of IDA in the General Population

In a Canadian Health Measures Survey study, approximately 2% of the population in Canada were identified as having IDA, with females experiencing a prevalence approximately 9 times higher than males. Total population IDA prevalence, corrected for inflammation in a recent study by Cooper et al.,⁵ ranged from 2% to 2.5%. Notably, less than 1% of males exhibited IDA regardless of the correction method, whereas this figure rose to approximately 4% for all females. For females of child-bearing age, the corrected IDA prevalence ranged between 5% and 6.2%.⁵

Prevalence of Iron Deficiency and IDA in Pregnant Females

Maternal ID, anemia, and IDA are most prevalent during the third trimester of pregnancy and may manifest as fatigue, restless leg syndrome, pica, hair loss, and irritability, which are often dismissed as pregnancy-related symptoms. Iron deficiency and IDA during the prenatal period have been associated with adverse effects on neonatal growth, neurocognitive development, and perinatal morbidity and mortality. A 2022 study conducted in British Columbia assessed iron status during early and late pregnancy among 60 pregnant patients receiving 27 mg/day of elemental iron as part of a randomized trial.⁶ High rates of iron deficiency were observed: 28% of participants exhibited probable iron deficiency at baseline (mean gestational age = 16 weeks), with the proportion increasing to 81% by study end (mean gestational age = 32 weeks).⁶ This deterioration in iron status occurred despite widespread supplementation and highlights a potential need for additional iron beyond the recommended dosage to support the needs of later stages of pregnancy.

Prevalence of Iron Deficiency and IDA in Indigenous Adults and Children

While the general Canadian population maintains a relatively low prevalence of anemia (approximately 3%), this is not the case among the Indigenous population. In a 2017 study of the Inuit in Nunavik (16 years and older), women of child-bearing age exhibited a 20% prevalence of anemia, with 14% of them experiencing IDA and 23% demonstrating iron deficiency without anemia, while the remainder had another, unspecified type of anemia.⁷ Data were not available for other Indigenous Peoples.

Anemia has a range of causes, so a diagnosis of IDA is based on identification of both anemia and iron deficiency. IDA is diagnosed based on blood tests showing low hemoglobin levels coupled with markers for ID, typically ferritin levels and TSAT.

Standards of Therapy

Contents within this section have been informed by materials submitted by the sponsor and clinical expert input. The following have been summarized and validated by the review team.

The key goals in the treatment of iron deficiency and IDA are the correction of the hemoglobin deficit and repletion of iron stores (the correction phase) and maintenance of iron levels over time (the maintenance phase).^2,8

Oral iron, often in the form of ferrous sulphate, is the first-line therapy for most cases of iron deficiency, and IDA and is relatively safe, effective, and inexpensive. Some patients may be unable to absorb ferrous sulphate adequately due to impaired intestinal uptake resulting from GI disease or clinical conditions such as chronic inflammation, which may in turn lead to elevated levels of hepcidin.^9-11 In other instances, the rate of absorption of even high-dose oral ferrous sulphate is insufficient to correct the anemia, and blood transfusion may be indicated.^9,12

Oral iron supplements also carry the additional burden of GI side effects, which can have a detrimental impact on patient adherence to treatment.^2,8 A systematic review and meta-analysis of RCTs found a significantly increased risk of GI side effects, particularly constipation, nausea, and diarrhea, for ferrous sulphate compared to placebo (OR = 2.32; 95% CI = 1.74 to 3.08; P < 0.0001).²⁶

There is also a need to deliver iron rapidly in certain clinical situations, including, for example, when a patient with iron deficiency requires an urgent surgery with risk of blood loss.¹³ While repletion of iron stores with oral iron may require administration over several months, the correction of iron stores following IV iron administration occurs within a few weeks.^8,12

IV iron treatment may also be preferred over oral iron in other situations. Oncology guidelines suggest that IV iron is superior to oral iron in combination with erythropoiesis-stimulating agents (ESAs) for chemotherapy-induced anemia.^14-16 There is also strong evidence from clinical trials supporting the preference of IV iron in patients with CKD with or without concomitant ESA treatment.^17-20 In children, IV iron substitution may be necessary in situations where oral iron therapy is not possible, ineffective, or poorly tolerated. According to Mattiello et al.,²¹ a switch to IV iron is proposed for cases with severe anemia (hemoglobin < 70 g/L) to avoid transfusion, for cases with an underlying secondary disease fulfilling a formal indication for IV iron (e.g., IBD, chronic GI or genitourinary bleeding, or celiac disease), and/or for situations of nonadherence to oral treatment and symptomatic refractory IDA with clinical impact.²¹

In cases of severe IDA, blood transfusion may be needed to increase hemoglobin to secure sufficient oxygen transport to the tissues and simultaneously replete iron stores.^27,28 Blood transfusions are the first-line treatment of iron deficiency only in cases of severe and critical anemia. Generally, blood transfusions are not indicated to treat iron deficiency due to the risk of potential disease transmission and transfusion-related complications.^29-32

The earliest IV iron preparations were associated with acute toxicity resulting from the release of bioactive labile iron. Subsequent iron formulations were complexes with carbohydrates, which release iron more slowly. Newer IV iron products, such as ferric carboxymaltose can be given in larger doses over a shorter period.¹²

In addition to ferric carboxymaltose, there are currently 3 other IV iron products approved in Canada: Ferrlecit (sodium ferric gluconate complex in sucrose; 12.5 mg/mL), Monoferric (ferric derisomaltose; 100 mg/mL), Venofer (iron sucrose; 20 mg/mL, pms-Iron Sucrose (iron sucrose injection; 20 mg/mL). None of these products is indicated for pediatric populations. The product monograph for ferric derisomaltose cautions against its use in pregnancy based on data from animal studies, and the product monograph for iron sucrose states that it should only be used if the potential benefit outweighs the potential risk to the fetus. Most of these products must be administered over a long period of time and/or with a series of sessions to attain a cumulative dose of 1,000 mg.

Drug Under Review

Each millilitre of ferric carboxymaltose contains 50 mg of elemental iron, and the dose is determined by the iron need and followed up with assessments after iron repletion. Ferric carboxymaltose is indicated for the treatment of IDA in adult and pediatric patients aged 1 year and older when oral iron preparations are not tolerated or are ineffective and for the treatment of iron deficiency in adult patients with New York Heart Association Class II or III heart failure and to improve exercise capacity. The diagnosis of iron deficiency must be based on laboratory tests.

Ferric carboxymaltose contains iron in a stable ferric state as a non-dextran iron complex consisting of a polynuclear iron-hydroxide core with a carbohydrate ligand. The complex is designed to provide utilizable iron for the iron transport and storage proteins in the body (transferrin and ferritin, respectively).

The sponsor’s reimbursement request is consistent with the indication. However, this review focuses only on the IDA indication while another review examines iron deficiency in adult patients with heart failure. Ferric carboxymaltose was granted a Notice of Compliance on March 11, 2024, under the standard review process at Health Canada.

Key characteristics of ferric carboxymaltose are summarized in Table 3 with other IV treatments available for IDA. According to the clinical expert consulted for this review, Ferrlecit is not broadly used in Canada relative to the other IV iron formulations. It was also not considered a comparator in the sponsor’s systematic review. As such, it was not a relevant comparator for this review.

Table 3: Key Characteristics of Ferric Carboxymaltose, Ferric Derisomaltose, and Iron Sucrose

CKD = chronic kidney disease; IDA = iron deficiency anemia.

^aHealth Canada–approved indication.

Source: Product monographs for ferric carboxymaltose,³³ ferric derisomatlose,²⁴ and iron sucrose.³⁴

Perspectives of Patients, Clinicians, and Drug Programs

Patient Group Input

This section was prepared by the review team based on the input provided by patient groups. The full original patient input(s) received by the team have been included in the Perspectives of Patients, Clinicians, and Drug Programs section of this report.

One patient group, the GI Society, provided input to this submission. The GI Society is a national charity formed in 2008 on the groundwork of its partner organization, the Canadian Society of Intestinal Research, and is committed to improving the lives of people with GI and liver conditions, supporting research, advocating for appropriate patient access to health care, and promoting GI and liver health. The GI Society stated that it obtained information for this submission from meetings and discussions with health care professionals and researchers, as well as surveys and interviews conducted on digestive and liver diseases. The number of survey respondents or interviewees was not mentioned; however, the patient group clarified that none of the patients had experience with ferric carboxymaltose.

According to the patient group input, dizziness, extreme fatigue, headaches, shortness of breath, tiredness, and muscle weakness are common and frequent symptoms of IDA. Other signs include extreme paleness, cold hands and feet, depression, and cravings for things that are not food (pica) or having a poor appetite, particularly in infants and children. The GI Society stated that IDA can have significant effects on a person physically, mentally, financially, and socially. Because IDA often arises from a pre-existing condition, such as celiac disease and/or IBD, many individuals must manage treating both conditions. IDA symptoms, especially extreme fatigue, can make it difficult to take part of daily activities, responsibilities, and social gatherings, leading many individuals to feel isolated.

The GI Society explained that common side effects of oral iron products are constipation, stomach upset, nausea, vomiting, dark stools, or teeth staining from liquid iron products. Some individuals are also unable to absorb or tolerate oral iron well for several reasons, such as time constraints from surgery, pregnancy, or childbirth; malabsorption; and intolerable side effects with oral iron. The GI Society added that, although many iron supplements are available over the counter, several have various marketing claims that may not necessarily be true, some may not have been subject to quality control regulations during manufacturing, and the efficacy claims of others may not have the clinical experience to support their efficacy claims. These products can also incur high out-of-pocket costs for patients. The GI Society believes that infusion-related side effects, such as staying an extra 30 minutes after an infusion to monitor for allergies or adverse reactions; staining of the skin or veins with iron (which can take 6 months or more to fade); Fishbane reaction, which is a pseudoallergy that includes flushing of the skin and can cause temporary (about 15 minutes) muscle contractions; taking time off school or work and arranging transportation to get to the clinic, are some of the disadvantages of IV iron administration. Moreover, some patients do not like the infusion process itself and the need for an IV insertion.

The GI Society stated that having alternative treatment options is crucial and that patients, their families, and health care professionals need access to treatments that have higher doses of iron, are better absorbed, and have fewer side effects.

Clinician Input

Input From the Clinical Expert Consulted by CDA-AMC

All CDA-AMC review teams include at least 1 clinical specialist with expertise regarding the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, and providing guidance on the potential place in therapy). The following input was provided by 1 clinical specialist with expertise in the diagnosis and management of IDA.

Unmet Needs

The clinical expert consulted for this review noted the key limitations of oral iron as tolerability (due to GI side effects) and adherence, as well as slow onset of effect (noteworthy for patients requiring rapid correction of their hemoglobin). The limitations of IV iron are its expense, and lack of coverage for those less than 65 years of age. These limitations are exacerbated when patients require urgent treatment (such as preoperatively or in pregnancy) as they do not have time to apply for additional coverage. The clinical expert noted that, because of these limitations, patients will access emergency departments for their IV iron. The clinical expert also noted that some patients may have a reaction to IV iron. Although they may try other IV formulations, their options are limited if the alternatives fail. The clinical expert noted that high-dose IV iron (1 g formulations instead of 300 mg with iron sucrose) is needed to reduce hospital visits for patients, and there is no approved therapy for IV iron replacement in pregnancy.

Place in Therapy

In the opinion of the clinical expert consulted for this review, ferric carboxymaltose would represent an alternative to ferric derisomaltose. The clinical expert believed that it would be used in the first line in situations where IV iron would be considered the first line, otherwise it would represent second-line therapy after failure to tolerate or respond to oral iron. The clinical expert did not describe ferric carboxymaltose as a drug that would shift the treatment paradigm for patients with IDA. The clinical expert noted that, for patients with iron deficiency or IDA, access to ferric carboxymaltose should be based on failure to tolerate or respond to oral iron after a 4-week trial, except in situations such as end-stage renal disease on dialysis, heart failure with reduced ejection fraction, moderate to severe anemia, and preoperatively in patients with iron deficiency and anemia, as these patients do not have time to respond to oral iron. The clinical expert agreed that patients with IBD should also have access to IV iron, per the European Crohn’s and Colitis Organization’s 2024 guidelines,³⁵ when their disease is clinically active, there is previous intolerance to oral agents, or if ESAs are needed. The clinical expert also agreed that patients with chemotherapy-induced anemia are also among those most suited for treatment with ferric carboxymaltose, given that the National Comprehensive Cancer Network guidelines advocate the use of IV iron over oral iron in this patient population.¹⁶

Patient Population

According to the clinical expert consulted for this review, the patients most suited for treatment with ferric carboxymaltose would be those adults with iron deficiency or IDA (ferritin less than 30 mcg/L or TSAT less than 20%), with an alternate definition for those with CHF or end-stage renal disease, pediatric patients with IDA, those unable to tolerate or respond to oral iron and preoperative patients with IDA. According to the clinical expert, the least suitable patients would be those with a history of clinically significant hypophosphatemia (i.e., those patients who are symptomatic or require treatment for their hypophosphatemia). The clinical expert also noted that those most in need of intervention are those under the age of 65 who lack drug coverage, and pregnant patients.

The clinical expert noted there are no issues with diagnosing IDA, and that ferritin and TSAT are the 2 most informative tests to assess iron stores. The clinical expert went on to note that ferritin is the gold-standard laboratory test, with levels of less than 50 mcg/L representing iron insufficiency and less than 30 mcg/L highly suggestive (and thus considered diagnostic) of iron deficiency. Ferritin can be elevated in cases of inflammation; in those cases, a TSAT less than 20% is informative and used as a diagnosis.

Assessing the Response to Treatment

The most important outcomes for assessing response include a rise in hemoglobin, improvement in symptoms of iron deficiency and IDA, improved function, and reduced transfusion burden. The clinical expert believed that 4 weeks would be the minimum duration of follow-up (with 12 weeks as ideal) to assess the efficacy of iron supplementation. Follow-up in the included trials that reported efficacy data ranged between 4 and 12 weeks.

Discontinuing Treatment

According to the clinical expert, reasons for discontinuation include severe allergic reaction or anaphylaxis (once), and a different formulation should be considered in cases of minor allergic reactions. The clinical expert noted that in the population with non-CHF and non–end-stage renal disease a ferritin level greater than 50 mcg/L and a sustained TSAT of greater than 20% should be warrant discontinuation.

Prescribing Considerations

The clinical expert did not believe that a specialist would be required to order ferric carboxymaltose, adding that it should be administered in community, hospital, and private infusion clinics; emergency departments; and inpatient units by home care nurses.

Clinician Group Input

This section was prepared by the CDA-AMC review team based on the input provided by clinician groups. The full original clinician group input received by CDA-AMC is included in the Perspectives of Patients, Clinicians, and Drug Programs section of this report.

• A group of 12 clinicians, including those with specialties in cardiology, obstetrics and gynecology, hematology, gastroenterology, and nephrology, as well as a primary care physician, were assembled to provide input.

• The clinician group was in general agreement with the input provided by the clinical expert.

• The clinician group agreed with the clinical expert that important unmet needs with IV iron supplementation include the ability to reduce visits to the hospital or infusion clinic and to have a product available that has evidence for use in pregnancy.

• The clinician group did not elaborate on its experience with ferric carboxymaltose; however, it did note that the drug has been available in Europe for 20 years.

Drug Program Input

The drug programs provide input on each drug being reviewed through our Reimbursement Review processes by identifying issues that may affect their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted by the review team are summarized in Table 4.

Table 4: Summary of Drug Plan Input and Clinical Expert Response

CHF = chronic heart failure; CKD = chronic kidney disease; IBD = inflammatory bowel disease; IDA = iron deficiency anemia; pERC = pan-Canadian Oncology Drug Review Expert Review Committee; TSAT = transferrin saturation.

Clinical Evidence

The objective of the Clinical Review report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of ferric carboxymaltose 50 mg/mL IV infusion in the treatment of IDA in adult and pediatric patients aged 1 year and older when oral iron preparations are not tolerated or are ineffective. The focus is on comparing ferric carboxymaltose to relevant comparators and identifying gaps in the current evidence.

The sponsor conducted a systematic search and selection to identify studies relevant to the efficacy and harms of ferric carboxymaltose relative to relevant comparators in the Canadian setting. The sponsor identified 60 reports of 40 clinical trials, including studies among patients with iron deficiency and heart failure. However, heart failure is outside the scope of this report. The current report focuses only on select studies identified in the sponsor’s systematic review, with detailed summaries and appraisals of these studies. Appendix 1 contains additional supportive studies submitted to Health Canada (Table 34, Table 35, Table 36, Table 37 and Table 38), as well as studies of ferric carboxymaltose in pregnancy (Table 39).

A summary of the clinical evidence included by the sponsor in the review of ferric carboxymaltose is presented in 3 sections, with a critical appraisal of the evidence included at the end of each section. To inform the broad indication under review, the first section, Pivotal Studies and Select RCTs, includes the pivotal studies submitted to Health Canada for the indication under review and other RCTs from the sponsor’s systematic review that enrolled patients with IDA arising from various causes. The second section includes indirect evidence from the sponsor. The third section, studies addressing gaps, includes a study among pediatric patients to inform the use of ferric carboxymaltose among children and adolescents that was excluded from the sponsor’s systematic review due to an irrelevant comparator (oral iron); a study among adult patients with IDA arising from various causes to provide additional evidence for the broad indication that was excluded from the sponsor’s systematic review due to an irrelevant comparator (oral iron); and 5 studies comparing ferric carboxymaltose with ferric derisomaltose, as this was identified as a relevant comparator, but these 5 studies were not included in the Pivotal Studies and Select RCTs section.

Included Studies

Clinical evidence from the following sources are included in the CDA-AMC review and appraised in this document:

• two pivotal studies and 2 other RCTs in patients with IDA arising from various causes

• one indirect treatment comparison

• seven additional studies addressing gaps in evidence.

Pivotal Trials and Selected RCTs

Contents within this section have been informed by materials submitted by the sponsor. The following have been summarized and validated by the CDA-AMC review team.

Description of Studies

Characteristics of the pivotal studies are summarized in Table 5. Characteristics of 2 additional studies identified in the sponsor’s systematic review are summarized in Table 6.

Table 5: Details of Pivotal Trials

AUC = area under the curve; CVD = cardiovascular disease; EPO = erythropoietin; FCM = ferric carboxymaltose; IBD = inflammatory bowel disease; IBDQ = Inflammatory Bowel Disease Questionnaire; IDA = iron deficiency anemia; RCT = randomized control trial; SF-36v2 = Short-Form (36) Health Survey Version 2; TSAT = transferrin saturation.

Note: Three additional reports were included (Health Canada reviewers report and Clinical Study Reports for VIT-IV-CL-015 and FERGIcor).

Source: Sponsor’s Summary of Clinical Evidence.³⁷

Table 6: Studies in Patients With IDA Arising From Various Causes

ALT = alanine transaminase; AST = aspartate transaminase; CO = crossover; CVD = cardiovascular disease; FCM = ferric carboxymaltose; HbAg = hepatitis B surface antigen; HBV = hepatitis B virus; HCV = hepatitis C virus; hsCRP = high-sensitivity C-reactive protein; IBDQ = Inflammatory Bowel Disease Questionnaire; IDA = iron deficiency anemia; NDD-CKD = non-dialysis dependent chronic kidney disease; NR = not reported; QoL = quality of life; RCT = randomized control trial; SF-36 = Short-Form (36) Health Survey; TSAT = transferrin saturation.

Note: Three additional reports were included (Health Canada reviewers report and Clinical Study Reports for 1VIT05006 and VIT-IRON-2011-004).

Source: Details included in the table are from the Clinical Study Reports for 1VIT05006³⁹ and VIT-IRON-2011-004.⁴⁰

Study VIT-IV-CL-015 is a pivotal phase III, open-label, active-controlled, multicentre randomized trial that enrolled 240 patients with IDA in hemodialysis or hemodiafiltration with IDA secondary to chronic renal failure. ████████ ████ ████████ ███████ ████ ██ █████ ███ ███ ██ ████ ████ ██ █████ ██ █████ ████████ ██████████ ████████ ████ ████████ ███ ███████████ ████ █ ██████ ███████ ███ ████████ ████ ██████████ ██ ██████ ██████ ██████████████ █ ███████ ██ ██ ████ ███████ █████ ████████ ████████ ████ ████████ ███ ███████ ████ ████████ ████ ███████ ██ █████ ████████ ██ █████ ██ ██ ███ ██ ███ ████ ████ ████████ ██ █████ █ █████ █████ ███ ███ ██ █████ ██████████ ████████ ████ ████████ ████████ ██ ███████ ██████ ██████████████ ██ ██ ████ ███████ ██ █ ███ ██████ ██████████ █████████ ██ ███████ ███ ████████ ██ ██████ ██ █████████ █████ ███ ██ ████ ███ ██ ████ █████ █████████████ ███ ███████ ███ █████ ██ ███████████ █████ ███████████ ███████ ███ ███████ █████████ ██ ████ █████ ███ ██ ███████ ███ ████████ ██ ██████ ██████████████ ██████████ ███ ████ ███████ ██████████ ██ ████████ ██ ████████████ ████ ███ ██████████ ████ ███████ █████ ████████ ███ ███ █████████ █████████ ███ ██ ███████ ███ ██████ ██ ██████ ██████████████ ██████████ ███ ████ ███████ ██████████ ██ ████ ███████ ███████████

Study FER-IBD-07-COR (FERGIcor) is a pivotal phase III, open-label, active-controlled, multicentre randomized trial of 485 patients with IDA and IBD. Patients were enrolled in this trial between October 28, 2008, and December 10, 2009, from 88 centres in 14 European countries. Randomization was based on a 1:1 ratio and was conducted using a predefined, computer-generated randomization list without stratification. After a 14 day-screening period during which study eligibility was assessed, patients were randomly assigned to either ferric carboxymaltose (n = 244) or IV iron sucrose (n = 241). Patients returned for safety and efficacy assessments at weeks 4, 8 and 12. Patients who were not anemic at week 12 were invited to continue to participate in the maintenance study. The primary objective of this study was to evaluate the noninferiority in efficacy of the standardized dosage regimen of ferric carboxymaltose compared to individually calculated dosage regimens of iron sucrose in the correction of IDA in patients with IBD in remission. The secondary objective was to evaluate the safety and tolerability of the standardized correction dose regimen of ferric carboxymaltose.

Study 1VIT05006 was a phase III, randomized, blinded, placebo-controlled crossover study that compared the safety and tolerability of ferric carboxymaltose versus placebo in patients who have IDA. ████████ ████ ████████ ██ ████ █████ ████ ██ █████ ██ ███ ██████ ███████ ██████ ███ ██████████ █████████ ████ ███ █████ ███ ███ ████ ███████ █████████ ██ ████████ ███ █████ █████ █ █████████ ██████ ██ ██ ██ █ █████ █████████████ ██ ███ ███ ████████ ███ ███████ ███ █████ ██ ███████████ █████ ████████ ██████ █████ ██ █ ███ █████ ████ ██████████████ ██ ███████ █████ ████ █ ███████ ██ ███ ██████ ██████ █ ███ ███ █████████ ██ █████ ███████ ██████ ██ ███ ███ █████████ ██ ███ ███ ██████ ███ ███ ███ █████████ ██ █████ ███████████ ████████ ████████ █ ██████ ███████ ████ ██ ██████ ██████ ███████████████ ██ █████ ████████ ████ ███ ██ ███████ █████████████ ████ ██ ███████ ██ ███ ██ ███ ██ ███ █ ████ ███████ ████ ██ ███████ ███ █████ █████████████ ███ ███ █████ ████ ██ █████ █████ ██ ████ ████████ ████ ███ █ ██ ███████████ █████ ██ ███ ██████ ████ ██ █████ ████ ██ ███ ██ ███ ███ ███ ██████ █████ ██ ████ ████████ ████ ███ █ ██████ ████ ██ █████ ████ ██ ███ ███ █████████ ██ ███ ████████ ███ █████ █████████ ██████ ███ █████████ ██ ███ ███████████████ ███████ ██ ███ ████ ██ █████ █████ ████████████ ████ ███ ████████ ██ ██████ ██████████████ ████ ███████ ██ ████████ ██ ████████ ██ ██ █████ ██████████

Study VIT-IRON-2011-004 was a phase III, randomized, open-label active-controlled trial that tested the noninferiority of ferric carboxymaltose versus iron sucrose in Chinese patients with various conditions that result in IDA. A total of 371 patients were randomized at 19 sites in China beginning June 6, 2017, to either ferric carboxymaltose (n = 188) or iron sucrose (n = 183). Randomization was conducted using a predefined, computer-generated randomization list, using interactive response technology and was stratified by hemoglobin level (< 100 g/L versus ≥ 100 g/L). This was a 10-week study, with a screening period of up to 1 week, and a 9-week treatment period, which included follow-up. Ferric carboxymaltose was administered on day 1 and if needed, days 8 and 15, up to a single dose of 1,000 mg iron, while the iron sucrose dose was calculated individually using the Ganzoni formula, up to a maximum single dose of 200 mg iron 3 times a week.

Populations

Inclusion and Exclusion Criteria

The 2 pivotal studies included adult patients with IDA secondary to specific conditions, chronic renal failure (Study VIT-IV-CL-015) and IBD (the FERGIcor study). The other 2 studies were chosen for this section because they included patients with IDA due to a variety of causes and as such could inform the broad indication under review. Study 1VIT05006 there were 3 cohorts, a CKD, 2 with IBD, and cohort 3 included patients with IDA secondary to other conditions. Patients in Study VIT-IRON-2011-004 simply had to have microcytic hypochromic anemia due to any cause. Each study used different laboratory parameter cutoffs for hemoglobin from 70 g/L to 120 g/L (females) or 130 g/L (males) in the FERGIcor study, to 115 g/L or less in Study VIT-IV-CL-015, 120 g/L or less in Study 1VIT05006 or less than 110 g/L (females) or 120 g/L (males) in Study VIT-IRON-2011-004.

Interventions

VIT-IV-CL-015

The administration of ferric carboxymaltose and iron sucrose was conducted during patients’ scheduled hemodialysis sessions by medical personnel during the trial period.

███ ██████ ████ ██ ███ ████████████ ███ ██ █████████████ █████ ████ ██████████ ██ ██ ████ ███ ██ ██ ███████ ████████ ██ ██████ ██████████████ ██ █████ ███ ██████████ ██████ ██████████████ ███ ████████ ██ █ ██ ████████ ██████████ ███ ██ ██ ████ ███ ████████ ██ ███ █████████ ██ ██ █████ █████████ ████████ ████ ███ ████████████ ██████ ████ █ ████ █████ ███ █████ ██ █ ████████ ████████

███ ██████ ████ ██ ███ ██████████ ███ ██ ███ ████ ██████████ ██ ██ ████ ███ ███ ██ ███████ ████████ ██ ████ ███████ ██ █████ ███ ██████████ ████ ███████ ███ ████████ ██ █ ██ ████████ ██████████ ███ ██ ██ ████ ███ ████████ ██ ███ █████████ ██ ██ █████████ ████ ██ ███████ ████████ ████ ███ ████████████ ██████ ████ █ ████ █████ ███ █████ ██ █ ████████ ████████ ███ █████ ████ ███ ████████████ ████ █ ███████ ██ ██ ███ ███ ██ ████████████ ███████ ██ ███ ██ ████ █ ███ ████ █████████ ████ ███ █████████ ███ ██ █████ █████ ██████ ██ ████████ █████ ███ ██████████ ██████████ ██████████ ████ ██████ ██ ███ ███████ ████████ ███ ████████ ██ ████ ██████ ████████ ████████ █ ████ ██ ███ ██ ██ ████ ██ ██████ ██████████████ ██ ████ ████████ ██████ ███ ███ █████ ████ █████ ███ ████ ████ ███ ██ ████ ██ ██████ ██████████████ ██ ████ ███████ █████████ ██ ███ ██████████ ████ ████████ ██████████████ ███████████ ██████ ██████ ████ ███████ ██ ██████████████

Study FERGIcor

In FERGIcor, ferric carboxymaltose and iron sucrose were administered intravenously to patients by study personnel during study follow-up visits.

The dosage form of the intervention was 5% weight by volume iron (50 mg/L), as a sterile solution of ferric carboxymaltose in water for injection. Drip infusion ferric carboxymaltose was diluted in sterile 0.9% sodium chloride. The intervention was packaged in 10 mL vials containing 500 mg of iron per vial. The ferric carboxymaltose dose was selected in accordance with the results of a randomized, controlled study in patients with IBD⁴¹ and was calculated using a simplified dosing matrix based on baseline hemoglobin (below or above 10 g/dL) and patients’ weight (below or above 70 kg). Using this dosing matrix, patients received a total of 1,000 mg, 1,500 mg, or 2,000 mg IV iron, split into up to 3 infusions (on days 1, 8 and 15). The total weekly dose was not to exceed 1,000 mg for any patient, regardless of their body weight.

The form of the iron sucrose was similar to that described previously. The dosage of iron sucrose was selected for each patient based on the individually calculated iron dose following the Ganzoni formula. Patients received iron sucrose infusions twice a week for a maximum of 11 infusions, depending on their calculated iron deficit. A single dose was not to exceed 200 mg of iron.

Study 1VIT05006

██ ███ ██ ████████ ████████ ██ ██████ ██████████████ ████████ █ ███████ ████ ██ █████ ██ ██ ████ ██ ██████ ███ █████ ███ ██████ ██ ██ ███ ████ ██ ███████ ██████████████ █████ ██ ███ ████ ██████ ████ ███ █████████ ██████ █████ ██ ███ ██ █████ ██ ████ ████████████ ██ ███ ██ ██████ ██████ ████ ██ ███████ ███ █████ ██ ███ ██ ███ ██ ████ ████████████ ██ ███ ██ ██████ ██████ ████ ██ ████████ ████████ ████████ ██ ███████ ████ ████ █████ █████ ██ ██████ ████ ██████ ███ ███ ███ ██ ██ ██████ ██████ ███ ███ ██████ ██ ██ ██ █████ ███ ██ ██ ██████ ███████ ████ ██ ███████ ██████████████ ██ ███ ██ ████████ ███████ ████ ██ ███ █████ █████ ███ ████████ ██████ ██████████████ ██ ███████ ██ ██████ ████ ████ ███████ ██ ██████████

Study VIT-IRON-2011-004

In the ferric carboxymaltose group, study drug administration occurred on day 1 and, if needed, on days 8 and 15. Patients randomized to iron sucrose received injections and/or infusions of a maximum of 200 mg of iron 3 times a week until the calculated total iron dose was administered. For both injection and infusion, ferric carboxymaltose was to be administered up to a maximum single dose of 1,000 mg iron (up to a maximum of 20 mg/kg every 2 weeks) over at least 5 minutes for the 500 mg dose and over at least 15 minutes for the 1,000 mg dose. Administration time was not to exceed 30 minutes for either the 500 or the 1,000 mg dose. Iron sucrose was administered by slow IV push injection at a rate of 1 mL of undiluted solution per minute (10 mL of the product in at least 10 minutes; administration time should not exceed 60 minutes) or by drip infusion in a dilution of 1 mL of iron sucrose in maximum 20 mL (0.9% weight by volume) physiological saline (10 mL of the product in maximum 200 mL saline). The drip infusion rate was 200 mL solution in at least 30 minutes and should not exceed 60 minutes.

Other forms of oral or IV iron were prohibited, as were ESAs and blood transfusions. If patients required ESAs or a transfusion for symptomatic anemia, they were withdrawn from the study.

Outcomes

A list of efficacy end points assessed in this Clinical Review is provided in Table 7, followed by descriptions of the outcome measures. Summarized end points are based on outcomes included in the sponsor’s Summary of Clinical Evidence as well as any outcomes identified as important to this review by the clinical expert consulted by CDA-AMC and input from patient and clinician groups and public drug plans. Using the same considerations, the CDA-AMC review team selected end points considered to be most relevant to inform expert committee deliberations and finalized this list of end points in consultation with members of the expert committee. HRQoL was prioritized because of the impact that IDA has on patient quality of life, and all instruments used were reported. Of the various laboratory markers reported, hemoglobin, ferritin, and TSAT were identified as most relevant by the clinical expert. Of these 3 parameters, hemoglobin was considered to be most important by the clinical expert, and it was therefore reported as both a binary outcome (patients who responded) and change from baseline, while ferritin was reported as change from baseline and TSAT was reported as a binary outcome, based on input from the clinical expert. Regarding time points, the clinical expert agreed that it was best to report the latest where available, noting that some laboratory parameters such as ferritin peak immediately after infusion and then decline gradually over time.

In the absence of a complete body of evidence for any patient population or comparison, the certainty of evidence was not assessed by the CDA-AMC review team for any outcome.

Table 7: Outcomes Summarized From Pivotal Studies and Select RCTs

HRQoL = health-related quality of life; IBDQ = Intestinal Bowel Disease Questionnaire; SF-36 = Short-Form (36) Health Survey; TSAT = transferrin saturation.

Sources: Sponsor’s Summary of Clinical Evidence³⁷ and Clinical Study Reports for VIT-IV-CL-015,⁴² FERGIcor,⁴³ 1VIT05006,³⁹ and VIT-IRON-2011-004.⁴⁰

Hemoglobin

Hemoglobin is a protein in red blood cells that delivers oxygen to tissues, and low levels of hemoglobin indicate anemia. In IDA, therapy consists of the of the iron deficit and treatment of the underlying disease; the aim of treatment is to return both hemoglobin and iron stores to target levels. The WHO criteria for anemia are hemoglobin levels of less than 110 g/L in children aged 6 to 59 months and in pregnant patients, less than 115 g/L in children aged 5 to 11 years, less than 105 g/L for patients in the second trimester of a pregnancy, less than 120 g/L in children aged 12 to14 years and in patients who are not pregnant, and less than 130 g/L in men.⁴⁴ Differences in hemoglobin of 20 g/L or greater are considered clinically important in various conditions, although differences as small as 1 g/dL have been reported to have significant clinical impacts in patients with CKD and heart failure.^45-47 In Study VIT-IV-CL-015, the primary end point was an assessment of the percentage of patients to reach the target increase (≥ 10 g/L or ≥ 1 g/dL) in hemoglobin in 4 weeks. In the FERGIcor study, the primary end point was the number of patients responding with a minimum increase of 20 g/L at week 12 compared with baseline. Also in the FERGIcor study, a hemoglobin level greater than 160 g/L was indicative of iron overload. In Study VIT-IRON-2011-004, the primary outcome was the percentage of patients achieving an increase in hemoglobin of at least 20 g/L from baseline.

In the FERGIcor study, the number of patients who were nonanemic at week 12 was also reported as a secondary outcome. Nonanemic was defined as achieving a hemoglobin level of at least 120 g/L in females and at least 130 g/L in males by week 12.

Inflammatory Bowel Disease Questionnaire

The IBDQ was a secondary outcome measure in the FERGIcor study. The IBDQ is a validated and reliable disease-specific tool to measure HRQoL in adult patients with IBD, UC, or Crohn disease.⁴⁸ It consists of 32 questions scored in 4 domains: bowel symptoms (e.g., loose stools, abdominal pain), emotional health (e.g., anger, depression, irritability), systemic function (e.g., fatigue, sleep pattern), and social function (e.g., ability to attend work and social events). The response for each question ranges from 1 to 7, with 1 corresponding to significant impairment and 7 corresponding to no impairment. The overall IBDQ score is the sum of the responses to each of the questions. A total IBDQ score can range from 32 (very poor HRQoL) to 224 (perfect HRQoL). The questionnaire can be interviewer-administered or self-administered and involves a recall period of the last 2 weeks. Table 8 provides more details about the validity, reliability, and responsiveness of this scale.

Short-Form (36) Health Survey

Version 2 of the SF-36 was used as a secondary outcome measure in the FERGIcor study. This generic quality-of-life instrument is considered a standard instrument for assessing patient-based health care outcomes and was used because of the positive relationship reported between hemoglobin and all but 2 of the survey’s domains.⁴⁹ The SF-36 was developed to measure generic health concepts relevant across age, disease, and treatment groups.⁵⁰ The SF-36 measures 8 multi-item dimensions of health: physical functioning (10 items), social functioning (2 items), role limitations due to physical problems (4 items), role limitations due to emotional problems (3 items), mental health (5 items), energy and/or vitality (4 items), pain (2 items) and general health perception (5 items). There is a further unscaled single item asking respondents about health changes over the past year. The score of the SF-36 ranges from 0 to 100, with higher scores indicating better health. Table 8 provides for more details about the validity, reliability, and responsiveness of this scale.

Serum Ferritin

Ferritin is an iron-storage protein whereas serum ferritin is an indicator of iron status.⁵¹ In the presence of inflammation (i.e., elevated C-reactive protein), serum ferritin levels of less than 100 mcg/L are indicative of IDA⁵² and, according to the clinical expert, less than 30 mcg/L in the noninflamed population. Target levels of serum ferritin in the VIT-IV-CL-015 study were 200 mcg/L to 800 mcg/L. In both the VIT-IV-CL-015 and FERGIcor studies, serum ferritin levels greater than 800 mcg/L were also indicative of iron overload. No minimal important difference for serum ferritin was identified in the literature. The change in serum ferritin from baseline at weeks 2, 4, 6, and 8 was assessed as a secondary outcome in the VIT-IRON-2011-004 study, and change from baseline at weeks 1, 2, and 4 was assessed as a secondary outcome in the VIT-IRON-2011-004 study. The change from baseline in ferritin was also assessed as a secondary outcome in the FERGIcor study, adjusted for baseline measurement at subsequent visits (week 1 to 12).

Transferrin Saturation

TSAT is the ratio of serum iron concentration to the total iron-binding capacity expressed as a percentage, and is an indicator of iron deficiency. TSAT levels of less than 20% are a diagnostic criterion for IDA.⁵² In the VIT-IV-CL-015 and FERGIcor studies, TSAT outcome target levels were between 20% and 50%, assessed as a secondary outcome. In the VIT-IV-CL-015 and FERGIcor studies, TSAT values greater than 50% were indicative of iron overload. In the VIT-IRON-2011-004 study, only the change from baseline in TSAT, not patients with a TSAT responds, was assessed. No minimal importance difference for TSAT was identified in the literature.

Table 8: Summary of Outcome Measures and Their Measurement Properties

CDAI = Crohn Disease Activity Index; CKD = chronic kidney disease; FACT-An = Functional Assessment of Cancer Therapy–Anemia; IBD = inflammatory bowel disease; IBDQ = Inflammatory Bowel Disease Questionnaire; SF-36 = Short-Form (36) Health Survey; UC = ulcerative colitis.

Statistical Analysis

Sample Size and Power Calculation

Study VIT-IV-CL-015

The sample size was chosen by the sponsor based on feasibility rather than statistical arguments. The trial was not aimed to prove a specific statistical hypothesis. Results were analyzed primarily to provide estimates and descriptive statistics. The investigators decided that a total of 240 patients would be randomized (120 patients in the ferric carboxymaltose group and 120 patients in the iron sucrose group).

Study FERGIcor

Sample sizes of 183 in both the ferric carboxymaltose and iron sucrose groups provided the basis for achieving 90% power (beta = 0.10) at a 2.5% significance level by using a 1-sided equivalence test of proportions when the proportion for the ferric carboxymaltose group with standardized correction dosage regimen was 0.82 and the proportion in the iron sucrose group being tested for equivalence was 0.75. The maximum allowable difference between these proportions that would indicate they were equivalent (the range of equivalence) was 0.07. No justification was provided for this noninferiority margin. Assuming a dropout rate of 15%, a total of 420 patients were needed to be randomized to ensure 366 patients completed the study.

Study 1VIT5006

No formal sample size calculations were performed. The sponsor noted that “the statistical precision of estimated parameters (as measured by CI width) will be adequate with the planned sample size”; however, no further justification was provided.

Study VIT-IRON-2011-004

The planned sample size per study protocol was determined to be a total of 368 patients (184 per treatment group), to ensure that at least 147 patients are available for the primary analysis. The sample size was based on showing noninferiority in the difference in the proportion of patients achieving an increase in hemoglobin of 2 g/dL or more at any time up to week 8 between ferric carboxymaltose and iron sucrose. Using a 1-sided alpha level of 2.5%, and a −15% noninferiority margin, it was estimated that 147 patients per group would have an 80% power to detect whether ferric carboxymaltose is noninferior to iron sucrose with an expected proportion of patients in the iron sucrose group with a response of 70%. No justification was provided for the noninferiority margin. The total planned sample size was 368, which allowed for an estimated dropout rate of 20% of the randomized patients.

Statistical Testing and Multiple-Testing Procedure

The methods for statistical testing of relevant efficacy end points, including methods for imputing missing data and for sensitivity analyses in the included trials are listed in Table 9. There was no mention of multiple testing procedures in any of the trials.

Subgroup Analyses

In the FERGIcor study, a subgroup analysis was performed on patients with a hemoglobin response (patients with an increase from baseline of at least 20 g/L), looking at those who had a baseline hemoglobin of at least 70 g/L and less than 100 g/L, or patients with a baseline hemoglobin of at least 100 g/L and less than 120 g/L. This was a descriptive analysis with no formal comparisons planned.

In the 1VIT05006 study, no formal efficacy analyses were planned; however, for safety analyses, subgroups were summarized for any subgroup that included at least 30 patients, including different types of CKD (hemodialysis-dependent, peritoneal dialysis–dependent, and non–dialysis-dependent) and postpartum anemia and CKD.

No subgroup analyses were performed in Study VIV-IV-CL-015 study or in Study VIT-IRON-2011-004.

Secondary Outcomes of the Studies

Table 9 describes the methods used in the statistical analysis of the efficacy end points in the VIT-IV-CL-015, FERGIcor, and VIT-IRON-2011-004 studies.

Table 9: Statistical Analysis of Efficacy End Points

ANCOVA = analysis of covariance; HRQoL = health-related quality of life; IBDQ = Inflammatory Bowel Disease Questionnaire; LOCF = last observation carried forward; NA = not applicable; NR = not reported; TSAT = transferrin saturation.

Source: Sponsor’s Summary of Clinical Evidence.³⁷

Analysis Populations

Relevant analysis populations for the included trials are described in Table 10.

Table 10: Analysis Populations of VIT-IV-CL-015, FERGIcor, 1VIT5006, and VIT-IRON-2011-004 Studies

FAS = full analysis set; PPS = per-protocol set; TSAT = transferrin saturation.

Sources: Sponsor’s Summary of Clinical Evidence³⁷ and Clinical Study Reports for 1VIT05006³⁹ and VIT-IRON-2011-004.⁴⁰

Results

Patient Disposition

A summary of patient disposition from the pivotal trials is in Table 11. A summary of patient disposition in the 2 RCTs in patients with IDA arising from various causes is provided in Table 12.

Table 11: Summary of Patient Disposition From Pivotal Studies

FAS = full analysis set; FCM = ferric carboxymaltose; NR = not reported; PP = per-protocol set.

Source: Sponsor’s Summary of Clinical Evidence.³⁷

Table 12: Disposition for Trials In Patients With IDA Arising From Various Causes

FAS = full analysis set; FCM = ferric carboxymaltose; IDA = iron deficiency anemia; PPS = per-protocol set.

^aPatients may have had more than 1 reason for screening failure.

Sources: Clinical Study Report for 1VIT05006³⁹ and VIT-IRON-2011-004.⁴⁰

Baseline Characteristics

The baseline characteristics of patients enrolled in the VIT-IV-CL-015 and FERGIcor pivotal trials are provided in Table 13, and those from the trials among patients with IDA arising from various causes are in Table 14. The mean ages in the VIT-IV-CL-015 study were ████ ██████ years in the ferric carboxymaltose group and ████ ██████ in the iron sucrose group. In the other studies the mean age was around 40 years. Nearly all patients in the pivotal trials (at least 97% across treatment groups) were white. In the VIT-IV-CL-015 study approximately ███ of patients were female and in the FERGIcor study 59% were female. In the trials involving patients with IDA arising from various causes, almost all (approximately 90%) of the patients were female. In the 1VIT05006 study, ███ were Caucasian (as worded in the source), and in the VIT-IRON-2011 to 04 study all the patients were Chinese. Overall mean hemoglobin levels were ████ ███ in the VIT-IV-CL-015 study, █████ ████ in the FERGIcor study, █████ ███ in Study 1VIT05006, and less than ██ ███ in Study VIT-IRON-2011-004. There were differences in serum ferritin levels between studies, with a baseline of ████ █████ in Study VIV-IV-CL-015 and ████ █████ in the FERGIcor study, and also differences in the studies with patients with IDA arising from other conditions: ████ █████ in Study 1VIT05006 and ███ █████ in Study VIT-IRON-2011-004. Similarly, there were differences between studies in TSAT in the pivotal trials, with a mean baseline of █████ in Study VIT-IV-CL-015, and ████ in the FERGIcor study, while the mean baselines were ████ in Study 1VIT05006 and ████ in Study VIT-IRON-2011-004.

The baseline characteristics outlined in Table 13 and Table 14 are limited to those most relevant to this review or those expected to affect the outcomes or interpretation of the study results.

Table 13: Summary of Baseline Characteristics of Patients From Pivotal Studies

FCM = ferric carboxymaltose; IBD = inflammatory bowel disease; SD = standard deviation; TSAT = transferrin saturation.

Source: Sponsor’s Summary of Clinical Evidence.³⁷

Table 14: Baseline Characteristics of Patients With IDA From Various Causes in 1VIT05006 and VIT-IRON-2011-004

FCM = ferric carboxymaltose; IDA = iron deficiency anemia; NR = not reported; SD = standard deviation; TSAT = transferrin saturation.

^aThe cause of IDA is unknown.

Sources: Clinical Study Reports for 1VIT05006³⁹ and VIT-IRON-2011-004.⁴⁰

Exposure to Study Treatments

Exposure to study treatments is summarized in Table 15 for the pivotal studies and in Table 16 for the studies of patients with IDA arising from various causes.

Table 15: Summary of Patient Exposure From Pivotal Studies

FCM = ferric carboxymaltose; NR = not reported; SD = standard deviation.

Source: Sponsor’s Summary of Clinical Evidence.³⁷

Table 16: Exposure of Patients With IDA From Various Causes in 1VIT05006 and VIT-IRON-2011-004

FCM = ferric carboxymaltose; IDA = iron deficiency anemia; NR = not reported; SD = standard deviation.

Sources: Clinical Study Reports for 1VIT05006³⁹ and VIT-IRON-2011-004.⁴⁰

Efficacy

Key results of the pivotal trials and other RCTs among patients with IDA arising from various causes are summarized in Table 17.

Table 17: Summary of Efficacy Results (Pivotal Trials and Select RCTs)

ANCOVA = analysis of covariance; CFB = change from baseline; CI = confidence interval; CKD = chronic kidney disease; FAS = full analysis set; FCM = ferric carboxymaltose; HRQoL = health-related quality of life; IBD = inflammatory bowel disease; IBDQ = Inflammatory Bowel Disease Questionnaire; IDA = iron-deficient anemia; NR = not reported; OR = odds ratio; PPS = per-protocol set; RCT = randomized controlled trial; SD = standard deviation; SF-36 = Short-Form (36) Health Survey; TSAT = transferrin saturation.

^aThe Statistical Analysis Plan for this study did not call for any formal hypothesis testing, therefore, all reported P values should be considered supportive in nature

^bP values have not been adjusted for multiple comparisons.

Sources: Sponsor’s Summary of Clinical Evidence³⁷ and Clinical Study Reports for 1VIT05006³⁹ and VIT-IRON-2011-004.⁴⁰

Hospitalizations (All-Cause and Anemia-Related)

This outcome was not assessed in any of the 4 trials that were the focus of the systematic review section.

Hemoglobin

In Study VIT-IV-CL-015, the primary response rate was defined as the percentage of patients reaching an increase in hemoglobin of ≥ 10 g/L at 4 weeks after baseline. The percentages of patients with a response (per-protocol set) were 45 of 97 (46.4%) in the ferric carboxymaltose group and 32 of 86 (37.2%) in the iron sucrose group, with a reported P value for the between-group difference of 0.2101. Similar responses were reported for the ITT population: ██████ ███████ in the ferric carboxymaltose group and ██████ ███████ in the iron sucrose group. The mean (range) increase of hemoglobin in the ferric carboxymaltose group at week 4 was ███ █████ ███ ███ and in the iron sucrose group ███ █████ ███ ███ with a reported P value for the between-group difference of ██████. No between-group difference and CI were reported for either end point.

In the FERGIcor study, the primary outcome was patients who exhibited a response, defined as an increase in hemoglobin of at least 20 g/L at week 12. In the FAS, the percentage of patients with a response was 65.79% (150 of 240) in the ferric carboxymaltose group and 53.64% (118 of 220) in the iron sucrose group (Table 17). The between-group difference was 12.15% higher (95% CI, 3.07 to 20.97; P = 0.004) in the ferric carboxymaltose group compared with the iron sucrose group. The FAS set was also analyzed using the LOCF and the worst-case methods; the results were consistent with the primary analysis.

In Study VIT-IRON-2011-004, 99.5% of patients in the ferric carboxymaltose group and 98.3% of patients in the iron sucrose group achieved an increase in hemoglobin of at least 20 g/L from baseline to week 8, for a between-group difference of 1.13 (95% CI, −2.02 to 4.68) (Table 17). The criteria for noninferiority were met. The results for the FAS population were identical to those of the PPS. From a mean baseline of ████ ███ ███████ in the ferric carboxymaltose group and ████ ███ ███████ in the iron sucrose group, the mean change from baseline to week 8 was ████ ███ ███████ in the ferric carboxymaltose group and ████ ███ ███████ in the iron sucrose group. The between-group difference was ████ ███ ████ ███ █████ ██ ██████ ███████.

Patient-Reported HRQoL

In the FERGIcor study, for the physical component of the SF-36, from mean baselines of 44.17 (SD = 7.36) in the ferric carboxymaltose group and 44.98 (SD = 7.23) in the iron sucrose group, there were changes from baseline to week 12 of 3.88 (SD = 6.77) and 2.64 (SD = 7.14) (Table 17) for a reported between-group P value of 0.157. For the mental component of the SF-36, from a mean baseline of 40.02 (SD = 11.04) in the ferric carboxymaltose group and 41.30 (SD = 11.70) in the iron sucrose group, there was a change from baseline to week 12 of ████ ███████ ███ ████ ███████, for a between-group P value of ███████. For the IBDQ total score, from mean baselines of 150.8 (SD = 35.2) in the ferric carboxymaltose group and 152.7 (SD = 34.4) in the iron sucrose group, there was a change from baseline to week 12 of ████ ██████ ███ ████ ██████, for a between-group P value of ███████. Between-group differences and CIs were not reported for any of the HRQoL measures.

HRQoL was not assessed in either Study VIT-IV-CL-015 or Study VIT-IRON-2011-004.

Serum Ferritin

In Study VIT-IV-CL-015, from mean baselines of ████ mcg/L (range = ██ ███) in the ferric carboxymaltose group and ████ mcg/L (range = ██ ████ in the iron sucrose group, after week 4 there were changes from baseline of █████ mcg/L (range = █████ ████) in the ferric carboxymaltose group and █████ █████ (range = ███ ████) in the iron sucrose group (Table 17), with a reported between-group P value of less than ████. The between-group difference and CI were not reported.

In the FERGIcor study, from mean baselines of █████ mcg/L (95% CI, not reported [NR]) in the ferric carboxymaltose group and █████ mcg/L (95% CI, NR) in the iron sucrose group, after week 4 there were changes from baseline of ████ mcg/L (95% CI, NR) in the ferric carboxymaltose group and █████ mcg/L (95% CI, NR) in the iron sucrose group (Table 17). for an estimated difference between groups (repeated measures analysis) of █████ (95% CI, █████ ██ ██████; P < █████).

In Study VIT-IRON-2011-004, from mean baselines of ████ mcg/L (█████) in the ferric carboxymaltose group and ████ mcg/L (█████) in the iron sucrose group, after week 8 there were changes from baseline of 200 mcg/L (███████) in the ferric carboxymaltose group and ██████ mcg/L (██████) in the iron sucrose group, for a least square mean between groups of 52.65 mcg/L (95% CI, ██████ █████), P < █████ (Table 17).

Transferrin Saturation

In VIT-IV-CL-015, there were █████ patients (█████) in the ferric carboxymaltose group and █████ patients (██████ in the iron sucrose group who had a normal TSAT of between 20% and 50% by week 4, for a reported between-group P value ██████. The between-group difference and CI were not reported.

In the FERGIcor trial, there were 117 or 222 patients (52.7%) in the ferric carboxymaltose group and 76 of 209 patients (36.4%) in the iron sucrose group with a normal TSAT level (20% to 50%) at week 12. The OR was ████ ████ ██ █████ ██████ ███████ in favour of ferric carboxymaltose (Table 17). The absolute between-group difference and CI were not reported.

The TSAT response (achieving a normal TSAT) was not assessed in VIT-IRON-2011-004.|

Patients Without Anemia

In the FERGIcor trial, for the proportion of patients without anemia (a hemoglobin level of ≥ 120 g/L for females or ≥ 130 g/L for males) at week 12, the percentages of patients with a response were 72.8% (166 of 228) in the ferric carboxymaltose group and 61.8% (136 of 220) in the iron sucrose group. The OR was ████ ████ ██ █████ █████ ███████. The absolute between-group difference and CI were not reported.

Harms

Refer to Table 18 and Table 19 for harms data.

Adverse Events

In the overall population in Study VIT-IV-CL-015, ██████ ███████ patients in the ferric carboxymaltose group and ██████ ███████ patients in the iron sucrose group had at least 1 AE. In the FERGIcor study, there were AEs in ███████ █████ patients in the ferric carboxymaltose group and in ███████ ███████ patients in the iron sucrose group. The most common AEs, defined as occurring in at least 2% of patients in either treatment group of the overall population, are presented in Table 18.

In the overall population in Study 1VIT05006, ███████ ███████ patients in the ferric carboxymaltose group and ███████ ███████ patients in the placebo group reported at least 1 AE. In Study VIT-IRON-2011-004, ███████ ███████ patients in the ferric carboxymaltose group and ███████ ███████ patients in the iron sucrose group had an AE. The most common AEs are reported in Table 19.

Serious Adverse Events

In Study VIT-IV-CL-015, SAEs were reported in █████ ██████ patients in the ferric carboxymaltose group and █████ ██████ patients in the iron sucrose group. In the FERGIcor study SAEs were reported in ██████ ██████ patients in the ferric carboxymaltose group and ██████ ██████ patients in the iron sucrose group. The most frequently occurring SAEs are reported in Table 18.

In Study 1VIT05006, █████ ██████ patients in the ferric carboxymaltose group and █████ ██████ patients in the placebo group had an SAE. In VIT-IRON-2011-004 ██████ ██████ patients in the ferric carboxymaltose group and █████ ██████ patients in the iron sucrose group had an SAE (Table 19).

Withdrawals Due to Adverse Events

In Study VIT-IV-CL-015, █████ ██████ patients in the ferric carboxymaltose group and █████ ██████ patients in the iron sucrose group withdrew from study medication due to AEs. In the FERGIcor study, 7 of 224 patients (2.9%) in the ferric carboxymaltose group and 2 of 239 patients (0.8%) in the iron sucrose group withdrew from the study drug due to an AE. Reasons for withdrawal from study drug are reported in Table 18.

In Study 1VIT05006, █████ ██████ patients in the ferric carboxymaltose group and █████ ██████ patients in the placebo group withdrew from study medication due to an AE. In Study VIT-IRON-2011-004 █████ ██████ patients in the ferric carboxymaltose group and █████ ██████ patients in the iron sucrose group withdrew from study medication due to an AE (Table 19).

Mortality

███ patient in the ferric carboxymaltose group died during Study VIT-IV-CL-015 because of acute anterior myocardial infarction, which was serious, severe in intensity and unlikely to be related to study medication (Table 18). ███ ███████ died more than a week after study medication was withdrawn due to another nonserious AE. In 1VIT05006, ███ ███████ in the ferric carboxymaltose group died due to pneumonia. These were the only deaths across the 4 studies (Table 19).

Notable Harms

In the FERGIcor study, hypophosphatemia was observed in 6 patients (2.5%) in the ferric carboxymaltose group, and none were observed in the iron sucrose group. Hypophosphatemia was not reported in Study VIT-IV-CL-015.

In Study VIT-IRON-2011, hypophosphatemia was observed in ██ patients receiving ferric carboxymaltose ███████ and in ██ patients ██████ in the iron sucrose group. The number of patients with decreased blood phosphorous or hypophosphatemia was not reported in 1VIT05006.

Table 18: Summary of Harms Results From Pivotal Studies

AE = adverse event; FCM = ferric carboxymaltose; NA = not applicable; NR = not reported; TEAE = treatment-emergent adverse event.

Source: Sponsor’s Summary of Clinical Evidence.³⁷

Table 19: Summary of Harms Results for Trials in Patients With IDA From Various Causes

AE = adverse event; FCM = ferric carboxymaltose; IDA = iron deficiency anemia; NA = not applicable; NR = not reported; SAE = serious adverse events; TEAE = treatment-emergent adverse event.

Sources: Clinical Study Reports for 1VIT05006³⁹ and VIT-IRON-2011-004.⁴⁰

Critical Appraisal

Internal Validity

The pivotal trials and the other trials with IDA caused by various conditions appeared to be reasonably well-designed with measures taken to maintain allocation concealment during the randomization process and to achieve reasonable similarity in baseline characteristics between groups. All of the active-controlled trials lacked blinding. It is unlikely that laboratory values such as hemoglobin, ferritin, and TSAT would be influenced by patient knowledge of their treatment assignment. However, it is possible that patient-reported outcomes such HRQoL and subjective harms could be influenced by patients knowing which group they were assigned to. This bias may be mitigated by the fact that both groups received active treatment. However, there still may be a bias if patients assume that a newer therapy is more effective, or the reverse may be true if patients assume that a more established therapy is more effective. As Study 1VIT05006 was double-blinded, there is a low risk of bias in the measurement of the outcomes. As there were no substantial differences in reported harms in this trial, it is unlikely that patients would have become unblinded due to known effects of the intervention or comparator drugs.

There were fewer patients who withdrew from Study VIT-IV-CL-015 in the ferric carboxymaltose group than from the iron sucrose group █████ ██████ ██████, with the largest single reason being an AE █████ ██████ █████. There was not only a numerical difference in withdrawals between the 2 groups, but the percentage of withdrawals in the iron sucrose group was also high, at greater than 10%. This difference in withdrawals may bias assessments of efficacy outcomes if patients are withdrawing due to a lack of efficacy, and, particularly in this case, where there is a notable difference in withdrawals due to AE, assessment of harms. The sponsor did not plan for any formal statistical analyses in this study, no sensitivity analyses were performed, and no attempt was made to replace missing values. The direction of the potential bias is uncertain. There were no clear differences in withdrawals in the FERGIcor, VIT-IRON-2011 to 04, and 1VIT05006 studies.

The VIT-IV-CL-015 study was not planned to perform formal comparisons between ferric carboxymaltose and iron sucrose, and the other 2 studies that assessed efficacy outcomes did not appear to have a multiple-testing procedure in place. This limits the strength of conclusions that can be drawn about the efficacy of ferric carboxymaltose versus iron sucrose, outside of the primary outcome of patients with a hemoglobin response, as any statistically significant results reported after the primary outcome are at risk of type I error (a false-positive result).

The 2 active-controlled studies, FERGIcor and VIT-IRON-2011-004, that planned for formal between-group comparisons both used noninferiority designs; however, neither provided a justification for the noninferiority margin they chose. For example, in Study VIT-IRON-2011-004, the investigators used a noninferiority margin of 15% for the primary outcome and assumed a proportion patients who responded of 70% in the iron sucrose group. However, it is unclear where either of these estimates were obtained. The impact of this omission is likely minimal, given that there was a higher proportion of patients who responded in the ferric carboxymaltose group compared with the iron sucrose group. Both of these noninferiority trials planned to conduct their primary analysis of efficacy in the per-protocol population rather than in the FAS or ITT population. Because the results for the per-protocol population and FAS or ITT population were consistent with each other in each trial, there are no concerns with this approach.

There was a crossover design used in 1VIT05006, which was the only blinded study and the only study with a placebo control. Patients crossed over only 7 days after receiving their initial treatment assignment, and it is unclear if that was adequate for washout to prevent additive effects from first-line treatment.

The scales used to assess HRQoL, and the laboratory markers used to assess efficacy are all well-established and well validated instruments or assays. However, it is unclear how useful the IBDQ instrument would be for assessing the specific impact of IDA in patients with IBD. A generic instrument, the SF-36, was also used to assess HRQoL in this study, and it is unclear whether this instrument can separate changes in HRQoL due to the underlying condition (IBD) from those due to improvements in IDA.

Between-group differences, with a 95% CI, were not always reported in the included studies because formal comparisons were not planned. However, where provided, there appeared to be a reasonable degree of precision around the mean difference for most outcomes. Where absolute between-group differences with CIs were not reported (e.g., HRQoL, change from baseline hemoglobin, and patients with a TSAT response in the FERGIcor study, and patients with a hemoglobin response, change from baseline hemoglobin, change from baseline ferritin, and patients with a TSAT response in Study VIT-IV-CL-015), the precision of the effect estimates (i.e., whether the 95% CIs included values that would suggest a conclusion different from that suggested by the point estimate) could not be judged. There tended to be greater variability in within-group changes in outcomes such as ferritin; for example, in VIT-IV-CL-015, the change from baseline in ferritin to week 4 ranged from −205 mcg/L to 1,786 mcg/L. A formal GRADE assessment was not performed in this review given that a complete body of evidence was not included for any patient population or comparison.

External Validity

The only study conducted among patients with IDA arising from various causes that reported efficacy results was conducted solely in China, and all participants were Chinese. The clinical expert did not believe this to be a significant issue with respect to generalizability, as hemoglobinopathies were excluded from the trial (conditions such as thalassemia that are seen more commonly in patients who are Chinese). There were few patients who identified as Indigenous in any of the included studies, even though there are a disproportionate number of Indigenous people who have IDA in Canada. The clinical expert consulted on this review believed that the pivotal trial that included patients with IBD would likely be generalizable to any patient with IDA, whereas the pivotal trial that included patients with CKD would be less generalizable, given the nature of CKD, in which the pathophysiology of anemia is much more complicated. The relatively short follow-up in the pivotal trials and in Study 1-VIT-IRON-2011-004 (4 to 12 weeks) is likely not adequate for assessing the effects of ferric carboxymaltose on hard clinical outcomes, such as hospitalizations for anemia; however, the clinical expert believed that a minimum of 4 weeks would likely be adequate, but not ideal, for assessing laboratory response to an iron infusion in Canada.

The clinical expert noted that IV iron was administered more frequently in the trial among patients with CKD (2 to 3 times weekly) than it would be in Canadian practice, where it would be administered weekly (at most). The clinical expert also noted that dosing of IV iron replenishment in Canada is based on the Ganzoni formula, and this differs slightly from the approach taken in the clinical trials. The clinical expert believed iron sucrose as an appropriate comparator, one that is commonly used in Canada. However, there was no pivotal trial or trial among patients with IDA arising due to various conditions with a comparison to ferric derisomaltose.

The clinical expert found that, across the trials, the inclusion criteria for some lab parameters were different than would be seen in clinical practice, noting that, in Canada, a ferritin level of less than 50 mcg/L would warrant treatment, while the clinical trials used a ferritin level of less than 100 mcg/L or less than 200 mcg/L for inclusion. It is not clear what the implications are with respect to generalizability.

The clinical expert consulted for this review believed that hemoglobin is likely the most important laboratory value when assessing response to ferric carboxymaltose or other iron replacement products, and agreed with positioning hemoglobin as a primary outcome. The clinical expert also believed that the cutoffs used to define a patient who responded in the pivotal trials and other select trials as appropriate. However, the clinical expert also agreed that clinical outcomes such as fatigue and HRQoL are important measures of treatment success, and these outcomes were absent from 3 of the included trials. Although HRQoL was measured in the FERGIcor trial, it is uncertain whether the measurement tools used were appropriate to estimate the HRQoL impacts of IV iron therapy.

GRADE Summary of Findings and Certainty of the Evidence

Results

In the absence of a complete body of evidence for any patient population or comparison, a formal GRADE assessment was not performed for this review.

Indirect Evidence

Contents within this section were informed by materials submitted by the sponsor. The following have been summarized and validated by the CDA-AMC review team.

Objectives for the Summary of Indirect Evidence

The objective of this section is to summarize and critically appraise the methods and findings of the indirect evidence submitted by the sponsor. The sponsor submitted a summary of indirect evidence from a published NMA that compared the effects of ferric carboxymaltose with those of other IV iron therapies in the context of adults with IDA associated with IBD.²²

The sponsor noted that at least 2 other published NMAs compare the effects of ferric carboxymaltose with those of other iron therapies.^63,64 These NMAs, by Pollock and Muduma (2019) and Han et al. (2023), compared the effects of ferric carboxymaltose and iron derisomaltose among patients with IDA secondary to various conditions (excluding pregnancy), and during pregnancy, respectively. The reason given by the sponsor for selecting the NMA by Aksan et al. was the relatively low levels of statistical, clinical, and methodological heterogeneity compared to the other 2 NMAs. The methods for identifying the NMA by Aksan et al. and other candidate NMAs were not reported. There was no protocol outlining a priori how 1 or more NMAs would be selected for presentation among multiple NMAs reporting on the same population, comparisons, and outcomes.

Description of Indirect Comparison

Aksan et al. published a systematic literature review and NMA in 2017 that used a literature search updated to June 2016. The authors screened 322 records that were potentially of interest, from which 15 studies were selected for inclusion in the systematic review. Of these, 5 studies were included in the NMA based on the systematic review methods and selection criteria described in Table 20.

Table 20: Study Selection Criteria and Methods for Systematic Review Contributing to the NMA Submitted by the Sponsor

NMA = network meta-analysis; RCT = randomized controlled trial.

Source: Sponsor’s Summary of Indirect Comparisons.²²

Network Meta-Analysis

Objectives

The objective of the NMA was to compare the efficacy and tolerability of different IV iron formulations used to treat IDA in IBD patients in a systematic review and Bayesian network meta-analysis.

Study Selection Methods

A literature search was performed in PubMed, Scopus, the Web of Science, and the Cochrane Library to identify original studies analyzing the efficacy and safety of IV iron as monotherapy for IDA in patients with IBD. The last search was completed in June 2016. Only publications in the English language were included.

Network Meta-Analysis Methods

The NMA was performed within a Bayesian framework using Markov chain Monte Carlo methods. The authors did not provide any information on the prior distributions used for the modelling parameters. Similarly, no information was provided on the convergence diagnostic, the burn-in period, the number of iterations, or the number of chains used.

A fixed-effect model and a random-effect model were fitted on the data. The deviance information criterion (DIC) was used to assess the model fit and to select the model for the main analysis.

Nodes in the network were built considering different types of IV iron compounds. However, different doses of the same drug were pooled in the same nodes.

Statistical heterogeneity was tested using I² statistics. Clinical and methodological heterogeneity was informally assessed by reviewing the included studies. However, more formal analyses and methods, such as meta-regression analyses, were missing. Network consistency was checked using the node-split approach. No subgroup analyses were performed.

The ITC results were presented as ORs related to response rate. Only 1 outcome was assessed in the NMA. Harms were not analyzed.

Results of the Network Meta-Analysis

Summary of Included Studies

The authors included 5 RCTs in the NMA. All studies were performed in patients with inflammatory bowel disease. The interventions in the studies included ferric carboxymaltose, iron sucrose, iron isomaltoside (also known as ferric derisomaltose), and oral iron. The evidence network is depicted in Figure 1. A summary of the characteristics of the included studies is presented in the following section. Sources of clinical and methodological heterogeneity across the included studies is summarized in Table 21.

Summary of Design and Population Characteristics of the RCTs in the NMA (Aksan et al.)²²

• Evstatiev et al. (2011), efficacy outcomes measured at 12 weeks

  • N = 483 (ferric carboxymaltose subgroup: n = 244; iron sucrose subgroup: n = 239)

  • Median age: 39.5 years (ferric carboxymaltose group); 38 years (iron sucrose group)

  • Interventions:

    • IV ferric carboxymaltose, 500 or 1,000 mg; up to 3 infusions (mean total dose = 1,377 mg ± 381 mg)

    • IV iron sucrose, 200 mg; up to 11 infusions, twice weekly (mean total dose = 1,160 mg ± 316 mg)

• Kulnigg et al. (2008), efficacy outcomes measured at 12 weeks

  • N = 196 (ferric carboxymaltose group: n = 136, oral iron group: n = 60)

  • Median age: 40 years (ferric carboxymaltose group); 45 years (ferrous sulphate group)

  • Interventions:

    • IV ferric carboxymaltose: iron requirements calculated according to Ganzoni formula; maximum permitted dose of 1,000 mg, or 15 mg/kg body weight in patients weighing less than 66 kg; a maximum of 3 infusions was permitted (per treatment cycle)

    • Oral iron sulphate: 100 mg ferrous iron, administered twice daily for 12 weeks

• Lindgren et al. (2009), efficacy outcomes measured at 20 weeks

  • N = 91 (iron sucrose group: n = 45; oral iron group: n = 46)

  • Median age: 42 years

  • Interventions:

    • IV iron sucrose: total dose individually determined according to SPC, based on the Ganzoni formula; drug given either in a single weekly dose of 200 mg, or every second week until cumulative dose reached; 1,000 mg was given to replenish iron stores; mean dose = 1,708 mg ± 331 mg

    • Oral iron sulphate: 100 mg, 2 tablets twice daily; mean dose = 38,387 mg ± 19,955 mg

• Reinisch et al. (2013), efficacy outcomes measured at 8 weeks

  • N = 327 (219 in the iron isomaltoside group and 108 in the oral iron sulphate group)

  • Median age: 36 years

  • Interventions

    • IV iron isomaltoside (also called ferric derisomaltose): Ganzoni formula used; patients randomized to either a single once-weekly infusion of up to 1,000 mg of iron isomaltoside over 15 minutes until reaching cumulative dosage or to a single once-weekly 500 mg bolus injection over 2 minutes until reaching cumulative dosage Mean dose: 885 mg ± 238 mg (infusion) and 883 mg ± 296 mg (bolus)

    • Oral iron sulphate: 200 mg oral iron daily for 8 weeks; total dose: 11,200 mg elemental iron

• Schröder et al. (2005), efficacy outcomes measured at 6 weeks

  • N = 46 (iron sucrose group: n = 22; oral iron group: n = 24)

  • Median age): 35 years (males); 33 years (females)

  • Interventions:

    • Iron sucrose group received a loading dose of iron sucrose of the maximal recommended concentration of 7 mg/kg body weight administered as a drip infusion over 3.5 hours, at visit 1; thereafter, 200 mg of iron sucrose was infused over 30 minutes once or twice weekly over 5 weeks in a fixed schedule; median total dose = 1,418 mg (1,375 mg to 1,452 mg); total mean dose = 1,418 mg)

    • Oral iron sulphate; 100 mg to 200 mg per day for 6 weeks; median total dose = 4,200 mg; total mean dose = 5,600 mg.

Figure 1: Network of Evidence

Source: Sponsor’s Summary of Indirect Comparisons.²²

Risk-of-Bias Assessment of Included RCTs

All trials were considered at low risk of selection bias by the NMA authors (i.e., randomization and allocation concealment were considered adequate). Participants and personnel were not blinded to the interventions in 4 of 5 studies; these were considered at unclear risk of performance bias by the NMA authors. There was incomplete outcome data for 2 studies that lacked clear numerical presentations of hematological results and associated statistics. It is unclear from the manuscripts of 3 studies whether the authors had full access to collected data. Both of the latter cases were considered at unclear risk of selective reporting bias by the NMA authors.²²

Table 21: Assessment of Clinical and Methodological Heterogeneity of Trials in the NMA

FCM = ferric carboxymaltose; FD = ferric derisomaltose; IDA = iron deficiency anemia; NMA = network meta-analysis; RCT = randomized controlled trial.

Source: Sponsor’s Summary of Indirect Comparisons.²²

Results

Efficacy

Point estimates for the ORs for hemoglobin response comparing ferric carboxymaltose with ferric derisomaltose and iron sucrose favoured ferric carboxymaltose. However, the 95% CrIs were wide and included effects favouring the comparator interventions, suggesting uncertainty with and imprecision indifferences in the ORs between ferric carboxymaltose versus iron sucrose (OR = 0.70; 95% CrI = 0.48 to 1.0) and ferric carboxymaltose versus iron isomaltoside (OR = 0.69; 95% CrI = 0.34 to 1.4).

Results of the NMA found that treatment with ferric carboxymaltose had a superior response compared to treatment with oral iron (OR = 1.9; 95% CrI, 1.1 to 3.2).

The node-splitting approach showed consistency in the analysis, with all P values without statistical significance. In line with the assessment of consistency, the low I² statistics (10%) showed limited statistical heterogeneity. The fixed-effect model offered a better fit (DIC = 18.215) compared with the random-effect model (DIC = 15.585).

Harms

Harms were not assessed in the NMA.

Critical Appraisal of the Network Meta-Analysis

The sponsor selected a published NMA that used a Bayesian approach with fixed-effects models for the analyses. Limited information is available regarding the methods used in the NMA model. The NMA was informed by a systematic review of relevant databases. However, there was no search of trial registries, reference lists of included studies, or grey literature, and no evidence of consultation with content experts, limiting the comprehensiveness of the search.⁶⁵ The date last searched was June 2016, more than 8 years before this report. Relevant studies available since this time would have been excluded from the analyses. The authors assessed the bias of the included studies but did not explain how the results of this assessment were incorporated into the analyses. All included trials were at unclear overall risk of bias, according to the authors. It was not reported how many reviewers were involved in the risk-of-bias appraisals and whether they worked independently. As such, the risk of errors and/or biases in the appraisals themselves is uncertain. It was not clear if the authors incorporated an assessment of clinical heterogeneity into their analyses. For example, the included study by Kulnigg et al.⁴¹ allowed for a second treatment cycle in case of relapse.

There were several limitations to the NMA. First, it contained a very small amount of data from 5 studies, resulting in a network with only 1 or 2 studies connecting the nodes and wide CrIs. Only 1 outcome was assessed in the NMA. There was no analysis of harms or other outcomes important to patients (e.g., symptoms or HRQoL). Furthermore, because studies of children and adolescents with IBD were excluded, it is uncertain whether the results can be generalized to this population. Generalizability is also limited because the NMA included patients taking oral iron, and therefore patients analyzed in the included trials may differ from those eligible for ferric carboxymaltose according to the Health Canada indication, which specifies that it is for the treatment of IDA “when oral iron preparations are not tolerated or are ineffective.” The inclusion criteria for the NMA did not place restrictions on the treatment dose, and doses could have been used that do not reflect those of clinical practice in Canada.

There was heterogeneity in the time point for end point evaluation, which could have biased the results in favour of 1 of the included treatment groups. Few patient characteristics were reported across the trials, challenging a thorough assessment of the plausibility of the exchangeability assumption. Interpretation of the effect estimates was limited by imprecision. The 95% CrIs for the comparisons of ferric carboxymaltose and ferric derisomaltose and iron sucrose were wide and included the potential for no difference, or that either treatment could be favoured.

Another significant limitation is related to the sponsor’s lack of clear, a priori criteria for selecting the NMA by Aksan et al. At least 2 other published NMAs investigate the relative efficacy of ferric carboxymaltose (in broader IDA populations). The methods for identifying the NMA by Aksan et al. and other candidate NMAs were not reported. There was no protocol outlining a priori how 1 or more NMAs would be selected for presentation among multiple NMAs reporting on the same population, comparisons, and outcomes. As such, there is a risk of bias in the selection of the NMA, as it is possible that it was selected based on its results.

Summary

The summary of a published NMA submitted by the sponsor was insufficient to determine whether there is a difference in efficacy for ferric carboxymaltose compared with other injectable forms of iron in patients with IDA and IBD. Important efficacy and harms outcomes were not investigated.

Studies Addressing Gaps in the Evidence From Pivotal Trials and Select RCTs

Contents within this section were informed by materials submitted by the sponsor. The following summary was validated by the review team.

This section presents additional evidence from 6 studies that address gaps in the evidence from pivotal trials and select RCTs: 1 study in pediatric patients, 1 study in patients with IDA resulting from various causes, and 4 studies comparing ferric carboxymaltose with ferric derisomaltose in patients with IDA.

The sponsor’s systematic review identified 2 trials, both in pediatric patients, that were excluded due to the comparators. Study 1VIT17044 compared ferric carboxymaltose with oral iron, while Study 1VIT18045 was a single-arm trial with no comparator. As ferric carboxymaltose is indicated in the pediatric population, Study 1VIT17044 is summarized here to provide evidence for pediatric patients. The 1VIT18045 trial was not summarized, as it lacks a randomized comparator, limiting causal conclusions.

Additionally, Study 1VIT09031, which investigated the efficacy and harms of ferric carboxymaltose compared with oral iron among adult patients with IDA secondary to various causes with inadequate response or intolerance to oral iron, was excluded from the sponsor’s systematic review because an oral iron formulation is not considered an appropriate comparator. It is summarized here to provide additional information to inform the broad indication for ferric carboxymaltose.

Moreover, an additional 4 studies are summarized here as they provide a comparison of ferric carboxymaltose and ferric derisomaltose, which is considered an appropriate comparator but not evaluated in the sponsor’s pivotal trials and select RCTs. The 4 studies were published as Zoller (2023), Emrich (2020), and Wolf (2020), the latter involving 2 trials.

Description of Studies

Characteristics of the 6 studies summarized in this section are detailed in Table 22.

The 1VIT17044 trial⁶⁶ is a phase III, multicentre, randomized, active-controlled, open-label, clinical trial conducted in 30 sites across 4 countries (the US, Ukraine, Poland, and Canada). The trial enrolled 79 patients who were assigned to receive ferric carboxymaltose (n = 40) or oral iron (n = 39). The trial aimed to investigate the efficacy and safety of ferric carboxymaltose versus oral iron in pediatric patients with IDA and a documented history of inadequate response to oral iron.

The 1VIT09031 trial⁶⁷ is a phase III, multicentre, randomized, active-controlled, open-label study to investigate the efficacy and safety of IV ferric carboxymaltose in adult patients with IDA who had an unsatisfactory response or intolerance to oral iron. Cohort assignment was based on results from a 14-day run-in period with oral iron. Patients with an inadequate response to oral iron (hemoglobin increase < 1 g/dL) were assigned to cohort 1, and patients who were intolerant of oral iron were assigned to cohort 2. Oral iron was the comparator arm in cohort 1, and other IV iron SOC per investigator’s choice was the comparator in cohort 2. In cohort 1, a total of 246 patients received ferric carboxymaltose and 253 received oral iron. In cohort 2, a total of 253 patients received ferric carboxymaltose and 245 received IV SOC. Patients were followed up to day 35 for efficacy assessment, and up to day 120 for safety assessment.

Zoller (2023)⁶⁸ is a multicentre, randomized, double-blind, active-controlled, clinical trial conducted at 20 outpatient hospital clinics in Austria, Denmark, Germany, Sweden, and the UK. The trial compared the incidence of hypophosphatemia after treatment with ferric carboxymaltose (N = 48 patients) versus ferric derisomaltose (N = 49 patients) in 97 patients with IDA and IBD.

Emrich (2020)⁶⁹ is a prospective, single-centre, double-blind study. The study randomized 26 women with IDA to receive ferric carboxymaltose (N = 13) or ferric derisomaltose (n = 13). This trial aimed to assess hypophosphatemia after high-dose iron repletion with ferric carboxymaltose and ferric derisomaltose.

Wolf (2020)⁷⁰ reports on 2 identically designed, open-label, randomized clinical trials that aimed to assess the effects of ferric carboxymaltose and ferric derisomaltose on hypophosphatemia. In trial A, 123 patients were randomized to receive ferric carboxymaltose (N = 61 patients) or ferric derisomaltose (N = 62 patients). In trial B, 122 patients were randomized to receive ferric carboxymaltose or ferric derisomaltose (N = 61 patients in each group). In total, 122 patients were randomized to receive ferric carboxymaltose and 123 to receive ferric derisomaltose.

Table 22: Details of Studies Addressing Gaps in the Evidence From Pivotal Trials and Select RCTs

ALT = alanine transaminase AST = aspartate transaminase; CKD = chronic kidney disease; FACIT-F = Functional Assessment of Chronic Illness Therapy–Fatigue; FCM = ferric carboxymaltose; FD = ferric derisomaltose; HUB = heavy uterine bleeding; IBD = inflammatory bowel disease, IDA = iron deficiency anemia; SOC = standard of care; TEAE = treatment-emergent adverse event; TOI = time of intervention; TSAT = transferrin saturation.

Sources: Clinical Study Reports for 1VIT17044⁶⁶ and 1VIT09031,⁶⁷ Onken (2014),⁷¹ Zoller (2023),⁶⁸ Emrich (2020),⁶⁹ Wolf (2020),⁷⁰ and Sponsor’s Summary of Clinical Evidence.³⁷

Study 1VIT17044

Statistical Analysis

All efficacy analyses were based on the ITT population, which included all patients who received at least 1 dose of the study drug, had a baseline hemoglobin measurement, and at least 1 postbaseline measurement. Patients were analyzed in the group to which they were randomized. The treatment group difference for change in hemoglobin, change in ferritin, and change in TSAT was assessed with an ANCOVA, using treatment and randomization strata (hemoglobin and age categories) as fixed factors and the baseline value of hemoglobin, ferritin, and TSAT as a covariate.

Missing values were imputed using the LOCF. An MMRM was conducted as a sensitivity analysis, assuming data were missing at random. There was a single primary comparison. Supportive and/or secondary analyses were not adjusted for multiple comparisons.

For the primary outcome (change in hemoglobin from baseline), subgroup analyses were conducted for the following subgroups:

• baseline hemoglobin (less than 10 g/dL; 10 g/dL or greater)

• age (1 year to less than 12 years; 12 years or greater to 17 years).

Results

Patient Disposition

In the 1VIT17044 trial, 39 patients (98%) in the ferric carboxymaltose group and 37 (95%) in the oral iron group completed the study. ████ ███ ████ in the ferric carboxymaltose group and ████ ███ ████ in the oral iron group discontinued the study.

Baseline Characteristics

The baseline characteristics of 1VIT17044 trial are summarized in Table 23. Notable imbalances were evident in patient characteristics at baseline. There were fewer Hispanic or Latino patients in the ferric carboxymaltose group compared with the oral iron group ████ ███ ████. There were also fewer patients with an obese BMI (██ ███ ███) and more patients in the normal-weight BMI category ████ ███ ████ in the ferric carboxymaltose group compared with the oral iron group. Mean baseline ferritin was lower in the ferric carboxymaltose group █████ ng/mL) compared with the oral iron group (█████ ng/mL).

Table 23: Summary of Baseline Characteristics of Patients in the 1VIT17044 Trial

BMI = body mass index; FCM = ferric carboxymaltose; SD = standard deviation; TSAT = transferrin saturation.

Note: Data are presented as mean (SD) or n (%)

Sources: Clinical Study Report for 1VIT17044⁶⁶ and sponsor’s Summary of Clinical Evidence.³⁷

Exposure to Study Treatments

In the 1VIT17044 trial, ██ █████ patients were exposed to at least 1 treatment dose. The mean amount of iron administered in the ferric carboxymaltose group was ██████ ██ ███████, with █████ of the patients receiving 2 doses. In the oral iron group, the mean amount of iron administered was ██████ ██ ██████). No additional iron preparations were allowed in the 1VIT17044 study. However, patients were allowed to receive ESAs, provided they were already on stable ESA therapy for at least 8 weeks before screening and that no ESA dosing or product changes were anticipated for the length of the trial. All the patients ██████ in the ferric carboxymaltose group and █████ in the oral iron group reported prior use of medications, most commonly ferrous sulphate (█████ in the ferric carboxymaltose group and █████ in the oral iron group). █████ of patients in the ferric carboxymaltose group and █████ in the oral iron group reported concomitant medications. Salbutamol sulphate was the most common concomitant medication and was used in ████ of patients in the ferric carboxymaltose group and ████ in the oral iron group.

Efficacy

In the 1VIT17044 trial, the least square mean change in hemoglobin from baseline to day 35 obtained through the ANCOVA model were ████ g/dL (standard error [SE] = █████) in the ferric carboxymaltose and ████ g/dL (SE = █████) in the oral iron group. The treatment difference at day 35 was ████ g/dL (95% CI, ██████ ████; P = ██████). The results of the MMRM model and the subgroup analyses align with those of the main analysis. The least square mean change in ferritin from baseline to day 35 obtained through the ANCOVA model were ██████ ng/mL (SE = ██████) in the ferric carboxymaltose and █████ ng/mL (SE = ██████) in the oral iron group. The treatment difference at day 35 was ██████ ng/mL (95% CI, ██████ ██████; P < ██████). The least square mean change in TSAT from baseline to day 35 obtained through the ANCOVA model were ██████ (SE = █████) in the ferric carboxymaltose and █████ (SE = █████) in the oral iron group. The treatment difference at day 35 was ██████ (95% CI, █████ █████; P < ██████).

Harms

In the 1VIT17044 trial, a larger percentage of participants in the ferric carboxymaltose group than in the oral iron group experienced at least 1 TEAE ██ ███████ versus ██ ██████, respectively). Numerically more patients in the ferric carboxymaltose group than the oral iron group experienced metabolism and nutrition disorders ████ ███ ███ and hypophosphatemia ████ ███ ████ ███████████ █████ ████████ ██ the ferric carboxymaltose group than the oral iron group experienced gastrointestinal disorders █████ ███ ██████ and constipation ███ ███ ██████. ███ ███████ in the ferric carboxymaltose group experienced a TEAE that led to treatment discontinuation. ████ ██ ███ ████████ experienced any SAE or any TEAE leading to death. Details are listed in Table 24.

Table 24: Summary of AEs in 1VIT17044 Occurring in 5% or More of Patients in Any Treatment Group (Safety Population)

AE = adverse event; FCM = ferric carboxymaltose, TEAE = treatment-emergent adverse event.

Sources: Clinical Study Report for 1VIT17044⁶⁶ and sponsor’s Summary of Clinical Evidence.³⁷

Critical Appraisal

Internal Validity

Although the methods for randomization were likely appropriate, there is an increased risk that prognostic balance was not achieved due to the small sample size, as evidenced by imbalances in patients’ baseline disease and demographic characteristics. Notably, the baseline serum ferritin level was higher in oral iron arm. There were also baseline imbalances by ethnicity and by BMI. The effect of these differences on efficacy outcomes is unclear. Although the trial was open-label, the efficacy outcomes are objective, making it unlikely that their measurements are biased. There is a risk of bias in the reporting of subjective harms (e.g., gastrointestinal disorders and headache) because patients knew which treatment they had been assigned (e.g., it is possible that known harms could be overestimated). The authors used the LOCF method to impute missing outcomes data and conducted sensitivity analyses using MMRM under the missing-at-random assumption. Although neither method may be appropriate (LOCF may not be reflective of the true trajectory of the outcome and MMRM assumes data are missing at random, which is not possible to assess and may not be plausible), the attrition rate was low (5% or less) in each group. As such, it is unlikely that missing data would have introduced bias. The ITT analysis was appropriate for estimating the effect of assignment on the interventions. Because there were no adjustments for multiple comparisons, there is an increased risk of type I error (false positives) for statistically significant results. Although the subgroup analyses were preplanned, these were unlikely to be sufficiently powered to detect subgroup differences.

External Validity

The included patients are from age 1 to 17 years, and results are not generalizable to other age groups. Given the small sample size, it is also unlikely that the results would be broadly generalizable to all pediatric patients with IDA. As the comparator in the trial was oral iron, this study does not provide information about the efficacy or harms of ferric carboxymaltose relative to other IV iron formulations in pediatric patients. Indigenous Peoples, who are disproportionately affected by IDA, are not represented in this trial. Although the outcomes measures were appropriate, some outcomes that may be important to patients (e.g., HRQoL) were not reported.

Study 1VIT09031

Populations

A 14-day run-in period of 325-mg oral ferrous sulphate tablets (65 mg elemental iron) 3 times a day was required before cohort assignment. Patients who had an inadequate response to oral iron (hemoglobin increase < 1 g/dL) were assigned to cohort 1. Patients who were intolerant of oral iron (e.g., severe gastrointestinal AEs) or deemed unsuitable by investigator for the oral iron run-in (e.g., low hemoglobin requiring rapid correction) were assigned to cohort 2. Patients who tolerated oral iron and had a hemoglobin increase ≥ 1 g/dL, or who were inadequately (< 67%) adherent, were not randomized.

Interventions

In cohort 1, patients were stratified by IDA etiology (heavy uterine bleeding [HUB], GI disorders, and other), baseline hemoglobin, and baseline cardiovascular risk (Framingham model). Patients were randomized 1:1 to receive the following:

• group A — ferric carboxymaltose 15 mg/kg to a maximum of 750 mg per dose on day 0 and 7 (maximum total dose of 1,500 mg), as an undiluted IV push at 100 mg/min

• group B — oral ferrous sulphate 325 mg 3 times a day for an additional 14 days.

In cohort 2, patients were randomized to the following:

• group C — ferric carboxymaltose 15 mg/kg to a maximum of 750 mg per dose on day 0 and 7 (maximum total dose 1,500 mg), as an undiluted IV push at 100 mg/min

• group D — IV standard care (other IV iron: Venofer, Ferrlecit, Feraheme, INFeD, or Dexferrum).

Outcomes

The primary efficacy end point was the mean change in cohort 1 from baseline to the highest observed hemoglobin observed any time between baseline and day 35 or TOI. Supportive efficacy end points in cohort 1 included:

• the proportion of patients achieving hemoglobin > 12 g/dL any time between baseline and day 35 or TOI

• the mean change from baseline to highest observed ferritin any time between baseline and day 35 or TOI

• the proportion of patients achieving hemoglobin > 12 g/dL and ferritin increase ≥ 160 ng/mL any time between baseline and day 35 or TOI

• the proportion of patients achieving a clinically meaningful increase in hemoglobin between baseline and day 35 or TOI (i.e., ≥ 1 g/dL for CKD, ≥ 2 g/dL for HUB or GI disorders, ≥ 3 g/dL for postpartum, and ≥ 2 g/dL for others).