CADTH Reimbursement Review

Ravulizumab (Ultomiris)

Sponsor: Alexion Pharma GmBH

Therapeutic area: Atypical hemolytic uremic syndrome

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Stakeholder Input

Clinical Review

Executive Summary

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

Table 1: Submitted for Review

aHUS = atypical hemolytic uremic syndrome; NOC = Notice of Compliance; TMA = thrombotic microangiopathy.

Introduction

Atypical hemolytic uremic syndrome (aHUS) is a life-threatening, ultra-rare disease in which patients are susceptible to sudden and progressive episodes of complement-mediated thrombotic microangiopathy (TMA) that most commonly damage the kidneys and include extrarenal, multiorgan involvement.^1,2 Patients typically present with signs and symptoms of the triad of thrombocytopenia, hemolysis, and acute kidney injury.³ The disease is primarily caused by an inherited or acquired dysregulation of complement-regulatory proteins, resulting in uncontrolled complement activation.^2-4 Over the past few years, there has been an increasing consensus that, in the majority of patients, aHUS may involve both genetic predisposition (e.g., pathogenic variants, autoantibodies, or at-risk polymorphisms in complement genes) and a triggering condition in order for the clinical event of a TMA to occur.^5-7 aHUS biomarkers include complement component 3 (C3), complement component 5a (C5a), complement component 5b-9 (C5b-9), factor B, complement factors B, H, and I, CH50, AH50, d-dimer, and anticomplement factor H (CFH) antibodies.^8,9 Low levels of C3, CH50, AH50, and CFB—along with increased levels of C5a, C5b-9, Bb, anti-CFH autoantibodies, and d-dimer—are usually found in patients with aHUS.⁸ aHUS can occur at any age, but onset during childhood is more common than in adulthood (60% versus 40%, respectively). Diagnosis is currently based on excluding other causes of TMA.^10,11 Therefore, the risk of misdiagnosis of aHUS may exist in clinical practice.^11,12 Although a positive genetic test can help to confirm a previously clinically diagnosed case of aHUS, it is not required to make the diagnosis of aHUS or to commence treatment. A clinical differential diagnosis remains the primary method of establishing a diagnosis of aHUS.^6,11 According to the clinical experts consulted by CADTH for this review, 30% to 40% of patients with aHUS may have no known genetic predisposition. According to the clinical experts, patients with aHUS who have DGKE mutations are unlikely to benefit from treatment with C5 inhibitors (e.g., eculizumab and ravulizumab).

The incidence and prevalence of aHUS vary widely.^3,13 A 2020 systematic literature review of the global epidemiology of aHUS reported that, for all ages, the annual incidence ranged from 0.23 per million population to 1.9 per million population.¹³ It was also reported that, for all age groups, the annual incidence was 4.9 per million population.¹³ There is limited published prevalence data for aHUS specific to Canada and the US.¹³ A Canadian study published in 2004 reported an incidence of aHUS in children of 2 cases per million over a 4-year period.¹⁴ Most recently, a 2020 analysis of 37 patients in Canada (15 pediatric and 22 adult) enrolled in the aHUS Global Registry (an observational, noninterventional, multicentre study that prospectively and retrospectively collects data on patients who have a clinical diagnosis of aHUS, irrespective of treatment)¹⁵ estimated that there are potentially 74 patients with aHUS in Canada.¹⁶

Prior to the approval of ravulizumab, eculizumab, a terminal complement inhibitor, was considered the standard of care for the treatment of patients with aHUS in most jurisdictions for over a decade.¹⁷ Eculizumab is the only Health Canada–approved drug indicated for the treatment of aHUS.¹⁸ However, it is not reimbursed in all Canadian jurisdictions (refer to Appendix 1). Furthermore, eculizumab imposes a substantial treatment burden on patients due to its shorter half-life and requirement for biweekly doses.¹⁸ The frequent dosing schedule of eculizumab is burdensome for patients, potentially affecting their health-related quality of life (HRQoL). It is also health care resource–intensive, which also drives infusion-related costs with eculizumab.¹⁹ The clinical experts consulted by CADTH for this review indicated that there is an unmet need for alternative effective therapies with acceptable toxicity profiles that can help patients achieve TMA remission and improve HRQoL. The appropriate duration of treatment with anticomplement therapy is unknown.

Ravulizumab (10 mg/mL concentrate for solution for infusion) is a terminal complement inhibitor that specifically binds to C5, inhibiting its cleavage to C5a and C5b, preventing the generation of the terminal complement complex membrane attack complex (MAC) or C5b9. Health Canada has previously issued market authorization for ravulizumab for paroxysmal nocturnal hemoglobinuria.^18,20 The Health Canada–proposed indication of interest for this review is for the treatment of adult patients with aHUS. The Health Canada–recommended dosing regimen consists of a single loading dose followed 2 weeks later by the first maintenance dose, with subsequent maintenance doses administered every 8 weeks for patients weighting greater than or equal to 20 kg or every 4 weeks for those weighing less than 20kg (Table 3). The sponsor’s reimbursement request is identical to the Health Canada–approved indication.

The objective of this clinical report is to review the beneficial and harmful effects of ravulizumab for the treatment of adult and pediatric patients with aHUS by inhibiting complement-mediated TMA.

Stakeholder Perspectives

The information in this section is a summary of input provided by the patient groups who responded to CADTH’s call for patient input and from the clinical experts consulted by CADTH for the purpose of this review.

Patient Input

One patient advocacy group, aHUS Canada, provided input for the treatment of aHUS. This group gathered information from 19 caregivers and 41 patients from inside and outside Canada through an online survey conducted in June 2022. Of these 60 respondents, 19 had experience with the drug under review.

Respondents identified anemia, low platelet count, and acute renal failure as the most difficult primary symptoms to control. Lack of quality of life, helplessness, post-traumatic stress disorder, fatigue, headache, high blood pressure, inability to travel, frequent hospital visits, kidney issues and dialysis are some of the experiences that respondents have while living with aHUS. According to aHUS Canada, aHUS patients with dialysis who need a kidney transplant are not eligible for transplant in Canada unless they receive eculizumab infusions during the transplant. Caregivers to patients with aHUS also face emotional and financial challenges because the process to access eculizumab or alternatives differs from province to province. Respondents described financial struggles, anxiety about access to treatment, the need to protect organs, exhaustion, memory loss, and brain fog as aspects of the disease that are among the hardest to control.

Respondents identified plasma therapy (fresh frozen plasma or plasmapheresis), eculizumab infusions, and long-term dialysis as the currently available treatments for patients with aHUS. Reported side effects included nausea, headache, fatigue, anaphylactic reaction to plasma, vein collapse, infection, anxiety, kidney failure, uncontrolled blood pressure, migraines, exhaustion, memory loss, brain fog, central line issues, muscle crumps, insomnia, abdominal pain, fever and chills, weight gain or loss, and being refractory to plasma therapy, among others.

When discussing their expectations for new drugs, patients reported that access to treatment and freedom of choice were critical components in managing the disease. However, quality of life was the most commonly cited desired outcome, and it was affected by factors like choice in care, frequency of appointments, and drug affordability. The abilities to travel, focus on family, and have more time between appointments were also described as critical to patients’ mental health. Moreover, frequent blood tests and IV therapies or ports were reported to be significant problems for many patients. While 1 caregiver pointed out the importance of maintaining “venous access for continuous access to eculizumab,” other patients shared their ineligibility for ports due to damaged veins from the disease. Patient also expressed the importance of less frequent treatments.

When discussing their experiences with ravulizumab, patients listed benefits that included more energy, less vein damage, fewer treatments, fewer symptom fluctuations, greater freedom of choice, and less anxiety. However, they also reported experiencing headache, nausea, and body aches right after their infusion or during the month after the infusion.

Clinician Input

Input From Clinical Experts Consulted by CADTH

Unmet Needs

The clinical experts consulted by CADTH for this review indicated that administering eculizumab every 2 weeks interferes with a patient’s quality of life by consuming time that could be spent working, travelling, or with friends and family. It can also be an issue when it comes to venous access fatigue and comes with the societal cost of nursing and allied health care support. The biggest limitation to the current treatment is prohibitive cost: most centres will fund an initial treatment or a few treatments, but very few centres have the resources to fund lifelong treatment. Inclusion in provincial formularies is inconsistent across provinces, and private insurance coverage is not common. Patients or their health teams advocate for subsidies or payment in full, but are not always successful. With respect to venous access fatigue, most patients would be candidates for portacaths or central lines, which are normally offered to chemotherapy patients.

The clinical experts indicated that the mechanism of action of ravulizumab is the same as that of eculizumab. Ravulizumab would not be added to other treatments: it would replace eculizumab as the treatment of choice for aHUS. The clinical experts indicated that they believe that ravulizumab would have likely similar or equivalent efficacy to eculizumab, with the potential for a better therapeutic profile and reduced therapeutic burden. These experts believed that ravulizumab would become the first-line treatment of choice because it offers improved patient quality of life and cost-effectiveness compared to eculizumab. The clinical experts mentioned that, theoretically—as CADTH has found with other biologics that use the same target molecule—tachyphylaxis to 1 medication may open up options to treat with the second, so acquired nonresponse may be a consideration to switch therapies. Improvements in patient HRQoL are expected to be significant after switching from eculizumab to ravulizumab

The clinical experts indicated that the patients most suitable for treatment with ravulizumab are those who have been diagnosed with aHUS. The patients least suitable are those with TMA that is clearly due to a secondary cause, such as malignant hypertension, malignancy, or infection. There may be some benefit in using eculizumab in some patients with autoimmune disease with histological evidence of TMA and evidence of complement dysregulation (e.g., some variants of lupus). According to the clinical experts, the patients with aHUS who are most in need of intervention are those with severe TMA with associated end organ damage, such as acute kidney injury or brain ischemia. The clinical experts indicated that patients who qualify for treatment would be identified by physicians with expertise in TMAs, such as nephrologists, hematologists, and internal medicine physicians. These physicians would make the diagnosis based on clinical examination, lab investigations, and genetic testing for complement dysregulation, and by excluding other causes of TMA.

To diagnose aHUS, there needs to be evidence of TMA, such as schistocytes, elevated lactate dehydrogenase, decreased haptoglobin, decreased hemoglobin, and thrombocytopenia. These lab abnormalities should also coincide with 1 or more of the following: neurological symptoms, acute renal failure, or gastrointestinal symptoms, although any organ system can be involved (e.g., pancreas, heart). Diagnosing aHUS can be very challenging because there is no single diagnostic test that can confirm it. In many situations, it is a diagnosis of exclusion. For this reason, misdiagnosis of this condition is a risk. One clinical expert indicated that testing has improved while the difficulties of diagnosis have decreased, suggesting that these diagnostic challenges may have been a greater issue 10 years or 15 years ago, when genetic and biochemical assessments of complement pathways were less accessible; however, these tests are now more available, often on a quick turnaround, even when sent out of province. One clinical expert indicated that haptoglobin is not the most reliable diagnostic indicator; lactodehydrogenase (LDH) level is more reliable.

The clinical experts indicated that etiologies that mimic TMA need be excluded, including infections, medications, malignancy, scleroderma, antiphospholipid antibody syndrome, systemic lupus erythematosus, malignant hypertension, disseminated intravascular coagulation, preeclampsia, and hemolysis, elevated liver enzymes, and low platelets (HELLP) syndrome. Thrombotic thrombocytopenia purpura (TTP) can be distinguished from aHUS by measuring ADAMTS13 level. If ADAMTS13 is higher than 5% and the patient is resistant to plasma exchange, then the diagnosis is more likely to be aHUS than TTP. Screening for complement mutations and antibodies should be performed. More sophisticated testing is available as well, including soluble C5b-9 levels: these levels are elevated during aHUS and reduced with treatment because C5b-9 is generated as a product of complement activation. If it is initially low, most centres will monitor C3 and complement component 4 levels for recovery.

The clinical experts indicated that early initiation of plasmapheresis until the diagnosis is confirmed is critical, given the high mortality risk of untreated TTP. One clinical expert indicated that most centres have access to ADAMTS13 activity testing with a turnaround time of 24 hours to 48 hours. The approach to treatment in adults, particularly older adults, may include plasmapheresis before the result is known. One clinical expert specializing in pediatric nephrology indicated that, if feasible, it is best to wait for the results for pediatric patients, because plasmapheresis is not recommended in this population; however, local resources also dictate its use and whether centres can procure C5 inhibitors quickly. In pediatrics, where TTP is less common, clinicians would likely not initiate plasmapheresis first, but agree it would be prudent to do so for adult patients, particularly older adults. The clinical experts emphasized that once aHUS has been diagnosed, C5 inhibition may be used as first-line therapy.

The clinical experts indicated that the treatment goals for aHUS are resolution of the TMA with normal platelet and LDH counts as well as resolution of acute kidney injury, neurological sequelae and stabilization of end organ damage. The required duration of treatment with C5 inhibition is unknown. Based on available data, if there are no high-risk complement genetic variants, then termination of treatment could be considered after 6 months to 12 months. However, according to the clinical experts, it is possible to discontinue treatment with ravulizumab in patients with aHUS without a genetic mutation in complement 3 months to 6 months after remission is achieved. Lifelong treatment may be considered for patients with high-risk complement genetic variations. The clinical experts mentioned that 30% to 40% of patients may have no known genetic disposition. As noted previously, patients with aHUS who have DGKE mutations are unlikely to benefit from treatment with C5 inhibitors (e.g., eculizumab and ravulizumab). Clinical experts highlighted that patients with DGKE mutations can safely come off C5 inhibitors if no response to treatment has been observed, because it is unlikely to help. The outcomes indicating a favourable response include resolution of TMA (i.e., normalization of LDH and platelet count), stabilization of end organ damage (such as acute kidney injury and brain ischemia), transplant graft survival (in susceptible individuals), and dialysis avoidance (in patients who have not yet developed end-stage kidney disease [ESKD]).

Close monitoring of the patient for 1 year after discontinuing therapy is recommended for monitoring relapse. Treatment discontinuation in patients with a high-risk mutation in complement is associated with a 50% relapse rate; therefore, discontinuing treatment in these patients is more challenging. Treatment discontinuation also needs to be considered in the setting of severe infections. However, 1 clinical expert indicated that this would entail restarting the medication, either with a reduced dose or with prophylactic anti-infectives.

The clinical experts indicated that ravulizumab can be given at home with nursing support or at an infusion centre. A specialist, such as a nephrologist or hematologist with expertise in TMA, needs to monitor the patient.

Clinician Group Input

No clinician group input was received for this review.

Drug Program Input

The drug plan identified the following jurisdictional implementation issues: considerations for initiation of therapy, considerations for continuation or renewal of therapy, considerations for prescribing of therapy, and system and economic issues. The clinical experts consulted by CADTH provided responses to the drug program implementation questions. For details, refer to Table 4: Summary of Drug Plan Input and Clinical Expert Response.

Clinical Evidence

Pivotal and Protocol-Selected Studies

Description of Studies

Two manufacturer-sponsored studies were included in this review: Study 311²¹ and Study 312.²²

Study 311 is an ongoing, phase III, prospective, multicentre, single-arm, open-label trial that includes adult patients with aHUS.²¹ Its key objective is to evaluate the safety and efficacy of ravulizumab (IV infusion) in adult patients (aged 18 years and older) with aHUS who are complement inhibitor treatment–naive. The study consists of a screening period (up to 7 days), a 26-week initial evaluation period, and an extension period until the product is registered or approved (in accordance with country-specific regulations) or for up to 4.5 years. Enrolment started on March 18, 2017, and is ongoing.²¹ The cut-off date for the data reported herein was July 2, 2019. As of the cut-off date, a total of 58 adult patients were included, and 56 patients had received at least 1 dose of ravulizumab. The primary outcome was complete TMA response during the initial 26-week evaluation period, which was defined as normalization of hematologic parameters (platelet count and LDH) and an improvement of at least 25% in serum creatinine from baseline. The secondary outcomes were hematologic normalization (platelet count and LDH), hematologic TMA parameters (platelet count, LDH, and hemoglobin), hemoglobin response (more than 2% increase), dialysis requirement status, estimated glomerular filtration rate (eGFR), chronic kidney disease (CKD) stage, fatigue (measured using the Functional Assessment of Chronic Illness Therapy – Fatigue scale [FACIT-F]), HRQoL (measured using the 3-Level EQ-5D), and safety. Health care resource utilization, patient-reported aHUS symptoms, and extrarenal signs and symptoms of aHUS were reported as exploratory outcomes on a by-patient basis (no summary data provided).

Study 312 is an ongoing, phase III, prospective, multicentre, single-arm, open-label trial conducted in pediatric patients (younger than 18 years) with aHUS.²² The study includes 2 cohorts (i.e., cohort 1 and cohort 2). Cohort 1 includes 21 children with aHUS who are complement inhibitor–naive. The key objective for cohort 1 is to evaluate the safety and efficacy of ravulizumab (IV infusion) in this group. Cohort 2 includes 10 children with aHUS treated with eculizumab. The key objective for cohort 2 is to evaluate the safety and efficacy of ravulizumab (IV infusion) in children with aHUS with stable TMA parameters before a switch to ravulizumab. The study consists of a screening period (up to 7 days), a 26-week initial evaluation period, and an extension period that runs until the product is registered or approved (in accordance with country-specific regulations) or for up to 4.5 years. Enrolment for this study started on September 1, 2017, and is ongoing.²² The cut-off date for the data reported herein was December 3, 2019. As of the cut-off date, a total of 21 pediatric patients were included in cohort 1, and 18 patients had received at least 1 dose of ravulizumab. In cohort 2, a total of 10 pediatric patients were included, and all 10 patients received at least 1 dose of ravulizumab. The primary outcome was complete TMA response during the initial 26-week evaluation period, which was defined as normalization of hematologic parameters (platelet count and LDH) and at least a 25% improvement in serum creatinine from baseline (for cohort 1 only). The secondary outcomes were hematologic normalization (platelet count and LDH), hematologic TMA parameters (platelet count, LDH, and hemoglobin, cohort 1 only), hemoglobin response (great than 2% increase, cohort 1 only), dialysis requirement status, eGFR, CKD stage, fatigue (measured using FACIT-F), and safety. Health care resource utilization, patient-reported aHUS symptoms, and extrarenal signs and symptoms of aHUS were reported as exploratory outcomes on a by-patient basis (no summary data provided).

Efficacy Results

Complete TMA Response

At week 26 of Study 311, complete TMA response was observed in 30 patients of the 56 patients in the full analysis set (FAS) (53.6%; 95% confidence interval [CI], 39.6% to 67.5%). At the data cut-off date (median follow-up time = 75.57 weeks), complete TMA response was observed in 34 patients of the 56 patients in the FAS (60.7%; 95% CI, 47.0% to 74.4%). In Study 312, cohort 1, at week 26, complete TMA response was observed in 14 patients of the 18 patients in the FAS (77.8%; 95% CI, 52.4% to 93.6%). At the data cut-off date (median follow-up time: 82.43 weeks), complete TMA response was observed in 17 patients of the 18 patients in the FAS (94.4%; 95% CI, 72.7% to 99.9%).

Hematologic Normalization

In Study 311, hematologic normalization was defined as normalization of platelets and LDH. At week 26, hematologic normalization was observed in 41 patients of 56 patients in the FAS (73.2%; 95% CI, 60.7% to 85.7%). As of the data cut-off date, hematologic normalization was observed in 45 patients of the 56 patients in the FAS (80.4%; 95% CI, 69.1% to 91.7%). In Study 312, cohort 1, at week 26, hematologic normalization was observed in 16 patients of the 18 patients (88.9%; 95% CI, 65.3% to 98.6%). As of the data cut-off date, hematologic normalization was observed in 17 patients of the 18 patients in the FAS (94.4%; 95% CI, 72.7% to 99.9%).

Individual Hematologic Parameters

In Study 311, the mean (standard deviation [SD]) platelet count improved to a normal value after the initiation of ravulizumab treatment and remained stable during the extension period at the data cut-off date. Similarly, the mean LDH value decreased from baseline to within a normal range at week 26 and was sustained during the extension period at the data cut-off date. The mean hemoglobin value increased more gradually over time. The mean hemoglobin value was 120.27 g/L (normal value = 130 g/L to 175 g/L) at week 26 and remained above 120 g/L during the extension period at the data cut-off date. At week 26, 40 patients of the 56 patients (71.4%; 95% CI, 58.7% to 84.2%) in the FAS achieved a hemoglobin response. As of the data cut-off date, 45 patients of the 56 patients (80.4%; 95% CI, 69.1% to 91.7%) in the FAS achieved a hemoglobin response. In Study 312, cohort 1, similar improvements were observed in platelet count, LDH, and hemoglobin at week 26 and at the data cut-off date. In Study 312, cohort 2, hematologic parameters (platelet count, LDH, and hemoglobin) remained stable during the initial 26 weeks as well as through the data cut-off date.

Time to Complete TMA Response

In Study 311, as of the data cut-off date, complete TMA response was achieved at a median time of 86 (range, 7 days to 401 days). In Study 312, for pediatric patients, the median time to complete TMA response was 30 days (range, 15 days to 351 days).

Fatigue (FACIT-F)

In Study 311, an improvement of at least 3 points in FACIT-F score, which is considered a clinically meaningful improvement,²³ was observed in 37 patients of 44 patients (84.1%) with available data at week 26. During the extension period, 33 patients of 40 patients (82.5%) with available data had at least a 3-point improvement from baseline at the day 351 visit. In cohort 1 of Study 312, 3 patients of 9 patients (33.3%) had at least a 3-point improvement in the FACIT-F total score from baseline at week 26. All 9 patients had at least a 3-point improvement from baseline at day 351. In Study 312, cohort 2, there was no notable improvement or worsening compared to baseline in the pediatric FACIT-F scores for all 8 patients during the initial 26 weeks through day 351 of the extension period.

HRQoL (Measured Using the 3-Level EQ-5D)

In Study 311, patients in the FAS showed improved 3-Level EQ-5D scores at week 26, and this improvement continued to day 351 of the extension period.

Renal Function (eGFR, CKD Stage Shifting, Dialysis Status)

Estimated Glomerular Filtration Rate

In Study 311, the mean eGFR gradually improved during the initial 26 weeks. During the extension period, the mean eGFR remained stable above 50 mL/min/1.73 m² for the 43 patients who reached the day 407 visit. Overall, the mean eGFR value at baseline was 15.86 mL/min/1.73 m². The mean eGFRs were 51.83 mL/min/1.73 m² at week 26 and 50.30 mL/min/1.73 m² at day 407. In Study 312, cohort 1, the mean eGFR value at baseline was 26.4 mL/min/1.73m² (SD = 21.17 mL/min/1.73m²). The eGFR was 108.5 mL/min/1.73 m² (SD = 56.87 mL/min/1.73 m²) at week 26 and remained above 100 mL/min/1.73 m² for the 14 patients who reached the day 407 visit. In Study 312, cohort 2, the eGFR remained generally stable for all 10 patients from week 26 through the data cut-off date.

CKD Stage

In Study 311, in patients with available baseline and week 26 data, 32 of 47 patients (68.1%) in the FAS had improvement in CKD stage compared to baseline; 2 patients experienced a worsening of their CKD stage. During the extension period, for the 42 patients with available baseline data and day 407 data, 29 patients (69.0%) had improvement in CKD stage compared to baseline; the 2 patients who experienced worsening CKD stage at week 26 remained at stage 5 at the last available visit during the extension period. In Study 312, all but 2 patients in cohort 1 had an improved CKD stage at week 26; the shift was substantial, with 14 patients improving by 2 or more stages. None of the patients worsened in CKD stage at week 26 or during the extension period. In Study 312, 8 patients of 10 patients in cohort 2 began at CKD stage 1 and remained stable; 2 patients worsened during the initial 26 weeks. During the extension period, all 10 patients had no change in CKD stage by day 351 compared to baseline (refer to Table 2).

Dialysis Requirement Status

In Study 311, at baseline or within 5 days before the first dose of the study drug, 29 patients (51.8%) in the FAS had received renal dialysis (Table 2, Table 23). During the initial 26 weeks, 17 patients of these 29 patients (58.6%) discontinued dialysis. As of the data cut-off date, 18 patients of 29 patients (62.1%) had discontinued dialysis. Of the 27 patients who were not on dialysis at baseline, 7 patients (25.9%) initiated dialysis during the initial 26 weeks. As of the data cut-off date, 4 patients (14.8%) remained or started on dialysis. In Study 312, cohort 1, of the 6 patients in the FAS who were receiving kidney dialysis at baseline, 4 patients discontinued dialysis within the first 36 days of exposure to ravulizumab (Table 2, Table 23). All 6 patients had discontinued dialysis by day 193. Among patients who were not on dialysis at baseline, 0 patients initiated dialysis after starting treatment with ravulizumab. In Study 312, as of the data cut-off date, 0 of the 10 patients in cohort 2 initiated dialysis after starting treatment with the study drug.

Plasma Therapy–Free Status

Plasma therapy was prohibited during the trials; therefore, it was not an outcome assessed in the pivotal studies. However, plasma therapy was reported in the section on the concomitant therapy. In Study 311, 3 patients (5.2%) received plasma therapy; this was considered a protocol violation. No patients received plasma therapy in either cohort of Study 312.

Other Outcomes

Mortality, bleeding, packed red blood cell (RBC) transfusions, and soluble MAC levels were not assessed as efficacy outcomes in the 2 pivotal studies (Study 311 and Study 312). Symptoms (aside from fatigue) and hospitalization were reported on a by-patient basis in the 2 Clinical Study Reports (CSRs) submitted by the sponsor; there were no summary data submitted. Therefore, symptom reduction and hospitalization are not reported herein.

Harms Results

The key harm findings of Study 311 and Study 312 are shown in Table 2. In both studies, as of the data cut-off date, all patients experienced at least 1 treatment-emergent adverse event (TEAE). In Study 311, the most common adverse events (occurring in at least 30% of patients) were headache (n = 22; 37.9%), diarrhea (n = 19; 32.8%), and vomiting (n = 18; 31.0%). In Study 312, cohort 1, the most common adverse events (occurring in at least 30% of patients) were pyrexia (n = 10, 47.6%) and headache, diarrhea, vomiting, and nasopharyngitis (each occurring in 7 patients [33.3%]). In Study 312, cohort 2, the most common adverse event (occurring in at least 30% of patients) was oropharyngeal pain (n = 3; 30%). In Study 311, a total of 33 patients (56.9%) experienced a serious adverse event (SAE). Each SAE was reported in 1 patient. Exceptions were pneumonia and hypertension, each of which occurred in 3 patients (5.2%), and septic shock, urinary tract infection, aHUS, and malignant hypertension, each of which occurred in 2 patients (3.4%). In Study 312, cohort 1, the SAEs that occurred in greater than or equal to 2 patients were gastroenteritis viral infection and abdominal pain; each occurred in 2 patients (9.5%). In Study 312, cohort 2, no SAE was reported in more than 1 patient. In Study 311, a total of 3 patients (5.2%) experienced adverse events leading to discontinuation of the study drug. In Study 312, cohort 1, a total of 1 patient (4.8%) experienced adverse events leading to discontinuation of the study drug. In Study 312, cohort 2, 0 patients experienced adverse events leading to discontinuation of the study drug.

In Study 311, 4 patients died during the initial 26-week evaluation period. One of these 4 patients died due a pretreatment SAE (cerebral arterial thrombosis); 3 patients (5.2%) died due to treatment-emergent SAEs that were not considered to be related to the study drug. In Study 312, cohort 1 and cohort 2, no patients had died due to adverse events as of the data cut-off date. Regarding notable harms, as identified in the review protocol, no meningococcal disease was reported in either study. In Study 311, sepsis, hypersensitivity to the drug, and antidrug antibodies were each reported in 1 patient (1.7%). Infusion-related reactions were not reported. In Study 312, cohort 1, 1 patient (5.6%) reported hypersensitivity; no other notable harms were reported. In Study 312, cohort 2, no notable harms were reported.

Table 2: Summary of Key Results From the Pivotal and Protocol-Selected Studies

AE = adverse event; CKD = chronic kidney disease; CI = confidence interval; DC = discontinuation; FACIT-F = Functional Assessment of Chronic Illness Therapy – Fatigue; LDH = lactate dehydrogenase; NA = not assessed; NR = not reported; SAE = serious adverse event; SD = standard deviation; TMA = thrombotic microangiopathy.

^a95% CIs for the proportion were based on the asymptotic Gaussian approximation method with a continuity correction.

^b95% CIs for the proportion were based on exact confidence limits using the Clopper-Pearson method.

Source: Clinical Study Reports.^21,22

Critical Appraisal

The main limitation of the 2 included pivotal studies (Study 311 and Study 312) is the single-arm design, which does not include a comparator arm. Due to the rare and severe nature of aHUS, a randomized control group was not likely to be feasible. Such a design, in addition to a lack of consideration of confounding variables, precludes causal inferences (i.e., the outcomes cannot be directly attributed to ravulizumab). Without an active comparator, standard of care, or statistical hypothesis testing, it is not possible to confirm the relative therapeutic benefit or safety of ravulizumab versus other available treatments (such as eculizumab in this population) or standard of care. In addition, both Study 311 and 312 were open-label trials; the study investigators and patients were aware of their treatment status, which increases the risk of detection and performance biases that have the potential to influence outcome reporting. However, the primary and most secondary outcomes (aside from fatigue and HRQoL) are objective end points for which the risk of bias due to open-label design is low. The potential for bias is a greater concern for the subjective end points, such as safety, fatigue (measured using the FACIT-F), and HRQoL (measured using the 3-Level EQ-5D). The direction of anticipated bias related to these outcomes is unclear. It is possible that known harms and anticipated benefits would be overreported.

For the longer-term subjective end points (HRQoL and fatigue), there is a potential risk of bias because complete measures were lacking for a large number of patients (especially for the extension period), leading to substantial missing data on certain outcomes. There may have been differential recall bias, and/or those patients remaining in the study may have differed in some systematic way compared to those who remained in the study and provided responses. Overall, the magnitude and direction of the impact of these missing data and of recall bias on the patient-reported and HRQoL outcomes is unknown. No minimally important difference (MID) was identified for HRQoL measures in the aHUS population. The overall the findings for HRQoL should be viewed as supportive evidence only.

One more potential limitation was that the efficacy assessment was not based on the intention-to-treat population (for Study 311 and Study 312, cohort 1); instead, it included patients who received at least 1 dose of the study drug. A total of 2 patients (3.4%) in Study 311 and 3 patients (14.3%) in Study 312, cohort 1 were excluded from the primary FAS analysis. In addition, it is also noted that 43 patients (76.79%) in Study 311 and 14 patients (66.7%) in Study 312, cohort 1 experienced major protocol violations (N = 25, 43.1% in Study 311 and N = 9, 42.9% in Study 312), the majority of which were related to the eligibility criteria. Although a per-protocol (PP) analysis was performed (N = 44, 75.9% for Study 311 and N = 18, 85.7% for Study 312, cohort 1) and showed results that were consistent with the FAS analysis, not all patients with the major protocol violation, especially those related to eligibility criteria, were excluded from the analysis. Therefore, there is a potential impact on the results (although the direction of the impact is not clear). The main limitation of Study 312, cohort 2 (pediatric patients with aHUS who switched from eculizumab to ravulizumab) was that the sample size (N = 10) was small, which meant the overall dataset was more sensitive to outliers and skewed distribution.

Overall, according to the clinical experts consulted by CADTH, the inclusion and exclusion criteria of the 2 pivotal studies (Study 311 and Study 312) were reasonable and the baseline patient characteristics, concomitant medications, and prohibited medications were reflective of patients observed in clinical practice for the indication under review. Finally, it is not clear whether the magnitude of the treatment effect estimates observed in the relatively small study sample will be replicable in a larger study sample or generalizable to the target population in real-world clinical practice.

Indirect Treatment Comparisons

Direct comparisons between ravulizumab and eculizumab are likely to be infeasible due to the rare and severe nature of aHUS. Therefore, for this submission, a systematic literature review was conducted to identify any sources of indirect treatment comparisons (ITCs) between ravulizumab and eculizumab, or between ravulizumab and best supportive care. No ITCs were identified in the CADTH search.

Description of Studies

Overall, 1 study, a sponsor-submitted ITC, was available to assess the relative efficacy of ravulizumab versus eculizumab using a patient-level, propensity-based primary analysis.

Efficacy Results

Among adult patients with aHUS who had not had a kidney transplant, the sponsor did not note any statistically significant differences between ravulizumab and eculizumab with respect to mortality, complete TMA response, LDH, platelets, EQ-5D visual analogue scale (VAS), FACIT subscales, renal function, or dialysis status at 6 months when using a stabilized weights model. Sensitivity analyses exploring pediatric patients without kidney transplant, adult patients with kidney transplant, and adult patients without kidney transplant using propensity matching were broadly concordant with the primary analysis.

No data were available with respect to the presence of severe bleeding, hemoglobin concentration change over time, plasma therapy–free status, packed RBC transfusion, hospitalizations, or soluble MAC.

Harms Results

No evidence for relative safety or harms was presented for review.

Critical Appraisal

Overall, the submitted ITC was subject to several limitations that add uncertainty to the conclusions presented. Principally, it is unclear whether all clinically meaningful covariates were accounted for within the sponsor’s ITC; residual confounding may occur from these characteristics not being accounted for in the primary analysis. Similarly, there remain potentially important unmeasured confounding characteristics, such as a 10-year gap between the studies of eculizumab and ravulizumab. During this period, there may have been changes to standard of care, increased awareness or capacity to diagnose disease, and changes in health care system capacity. These are all confounding factors that cannot be excluded from the current analysis. Finally, a few reporting characteristics were absent, such as rationale of exclusion for studies, specification of the estimands used in the analysis, units of outcomes, and baseline covariates of interest.

Other Relevant Evidence

No other relevant evidence was identified.

Conclusions

The evidence for the clinical benefits and harms of ravulizumab in the treatment of aHUS was based on the 2 sponsor-submitted, pivotal, multinational, single-arm, open-label, prospective phase III trials (Study 311 for adults with aHUS and Study 312 for pediatric patients with aHUS). The majority of pediatric and adult patients who were complement inhibitor treatment–naive experienced hematological normalization, improved renal function, and improved HRQoL with ravulizumab treatment. Despite uncertainty around the magnitude of the clinical benefit attributable to ravulizumab (given the limitations inherent in the single-arm trial design), the lack of formal hypothesis testing, and the relatively small sample size, the clinical experts indicated that the benefits observed in the 2 trials appeared clinically meaningful, considering that aHUS is an extremely rare and life-threatening disease. For adult patients who were complement inhibitor–experienced, no evidence was identified to inform the switch from eculizumab to ravulizumab. The expected benefit of switching lies in the reduced number of infusions required (because of the longer half-life of ravulizumab versus eculizumab). Although the 10 patients who switched from eculizumab to ravulizumab in Study 312 appeared to have a maintained TMA response, due to the small sample size, it remains unclear whether these findings are reflective of what would be observed in the larger population of patients with aHUS. The sponsor also submitted a propensity score–weighted analysis comparing ravulizumab with eculizumab; however, due to several methodological limitations, no robust conclusion could be drawn on the comparative efficacy and safety of ravulizumab versus eculizumab. The safety profile of ravulizumab observed in the 2 trials appeared consistent with the known safety profile of ravulizumab, and no additional safety signals were identified.

Introduction

Disease Background

aHUS is a life-threatening, ultra-rare disease in which patients are susceptible to sudden and progressive episodes of complement-mediated TMA. These episodes most commonly affect the kidneys, but can also include extrarenal, multiorgan involvement.^1,2 An acute aHUS episode requires emergency care; however, patients with aHUS are also at ongoing risk of systemic, life-threatening, and multisystem complications over the long-term. Extrarenal manifestations are common, especially in newly diagnosed patients (i.e., ≤ 6 months from diagnosis).²⁴ aHUS is primarily caused by inherited or acquired dysregulation of complement-regulatory proteins, resulting in uncontrolled complement activation.^2-4 Historically, kidney failure and death were common outcomes; however, improved understanding of the condition has led to the discovery of novel therapies (i.e., complement inhibitors, including eculizumab and ravulizumab) that may reduce the risk of these complications.²⁵ The uncontrolled complement activation of aHUS causes inflammation, endothelial activation and damage, and a prothrombotic and/or procoagulant state, resulting in systemic TMA.^2,26-28 Over the past few years, there has been an increasing consensus that, in the majority of patients, for the clinical event of a TMA to occur, aHUS may involve both genetic predisposition (e.g., pathogenic variants, autoantibodies, or at-risk polymorphisms in complement genes) and a triggering condition.^5-7 Atypical HUS can occur at any age, but childhood onset is more common than adult onset (60% versus 40%, respectively). When onset occurs in childhood, the disease affects males and females equally, whereas in adulthood, the disease affects women more frequently. Most children affected by aHUS (70%) will have the disease before the age of 2 years. aHUS biomarkers include C3, C5a, C5b-9, factor B, complement factors B, H, and I, CH50, AH50, d-dimer, as well as anti-CFH antibodies.^8,9 Low levels of C3, CH50, AH50, and CFB—along with increased levels of C5a, C5b-9, Bb, anti-CFH autoantibodies, and d-dimer—are usually noted in patients with aHUS.⁸

Diagnosis of aHUS is currently based on exclusion of other causes of TMA.^10,11 Therefore, the potential risk of misdiagnosis of aHUS may exist in clinical practice.¹¹ Although a positive genetic test can help to confirm a clinically diagnosed case of aHUS, complement gene mutations are identified in only 50% to 60% of patients with aHUS^6,29,30 and are not required to make the diagnosis or commence treatment. A clinical differential diagnosis remains the primary method of establishing a diagnosis of aHUS.^6,11 The clinical experts consulted by CADTH for this review indicated that 30% to 40% patients with aHUS may have no known genetic disposition.

The incidence and prevalence of aHUS vary widely.^3,13 This is attributed to the heterogeneity of patients with aHUS, ambiguity surrounding its clinical presentation, and difficulties with diagnosing aHUS.¹³ A 2020 systematic literature review of the global epidemiology of aHUS reported that the annual incidence of aHUS ranged from 0.26 per million population to 0.75 per million population among people aged 20 years and younger. For all ages, the annual incidence ranged from 0.23 per million population to 1.9 per million population.¹³ These estimates are in line with the estimate reported in the US of 1 case per million to 2 cases per million in the general population.¹⁴ The 2020 systematic literature review also reported that the prevalence of aHUS ranged from 2.2 per million population to 9.4 per million population in people aged 20 years and younger, whereas the prevalence in all age groups (based on only 1 study) was 4.9 per million population.¹³ Most studies providing these data were from Europe and Oceania, given that there are limited published prevalence estimates for aHUS from countries such as Canada and the US.¹³ A Canadian study published in 2004 reported an incidence of aHUS in children of 2 cases per million over a 4-year period.¹⁴ It has also been reported that aHUS occurs in approximately 1 in 1 million births and affects 60 patients to 90 patients in Canada.^31,32 Most recently, a 2020 analysis of 37 patients in Canada (15 pediatric and 22 adult) enrolled in the aHUS Global Registry (an observational, noninterventional, multicentre study that prospectively and retrospectively collects data on patients with a clinical diagnosis of aHUS, irrespective of treatment)¹⁵ estimated that there are potentially 74 patients with aHUS in Canada.¹⁶

Standards of Therapy

Prior to the approval of ravulizumab, the terminal complement inhibitor eculizumab was considered the standard of care in most jurisdictions for the treatment of patients with aHUS for more than a decade.¹⁷ Eculizumab is the only Health Canada–approved drug indicated for the treatment of aHUS.¹⁸ However, eculizumab is not reimbursed across all Canadian jurisdictions (refer to Appendix 1). Furthermore, eculizumab imposes a substantial treatment burden on patients, due to its shorter half-life (compared to ravulizumab) and requirement for biweekly dosing.¹⁸ This results in missed days of work or school to accommodate visits to an infusion centre and requires careful scheduling of travel and other life events to accommodate biweekly treatment. Frequent infusions also make venous access ports necessary for some patients, especially children, which puts them at risk of port-related complications (e.g., infection and thrombosis).¹⁹ The frequent dosing schedule of eculizumab is health care resource–intensive, which also drives infusion-related costs with eculizumab.¹⁹ The clinical experts consulted by CADTH for this review indicated that there is an unmet need for alternative effective therapies with acceptable toxicity profiles that help patients with aHUS achieve TMA remission and improved HRQoL. The clinical experts anticipated that ravulizumab would have similar or equivalent efficacy as eculizumab, with the potential of a better therapeutic profile and/or reduced therapeutic burden. The clinical experts indicated that patients with aHUS who have DGKE mutations are unlikely to benefit from treatment with C5 inhibitors (e.g., eculizumab and ravulizumab).

The appropriate duration of treatment with anticomplement therapy is unknown. Both eculizumab and ravulizumab are C5 inhibitors, the major difference between these drugs is duration of action; ravulizumab has a longer duration. Compared with eculizumab, ravulizumab may have a similar clinical benefit, but be less burdensome for patients and the health care system. The choice between them is individualized.⁹

Drug

Ravulizumab (10 mg/mL concentrate for solution for infusion) is a terminal complement inhibitor that specifically binds to C5, inhibiting its cleavage to C5a and C5b and preventing the generation of MAC or C5b9. Health Canada has previously issued market authorization for ravulizumab for paroxysmal nocturnal hemoglobinuria.¹⁸

The Health Canada–proposed indication of interest for this review is the treatment of adult patients with aHUS. The Health Canada–recommended dosing regimen consists of a single weight-based IV loading dose followed 2 weeks later by the first IV maintenance dose; IV maintenance doses are then administered every 8 weeks (or every 4 weeks for children weighing ˂ 20 kg) (Table 3). The sponsor’s reimbursement request is identical to the Health Canada–approved indication.

Table 3: Key Characteristics of Ravulizumab and Eculizumab

aHUS = atypical hemolytic uremic syndrome; C5 = complement component 5; q.4.w. = once every 4 weeks; q.8.w. = once every 8 weeks; TMA = thrombotic microangiopathy.

^aHealth Canada–approved indication.

Source: Product monographs for Ultomiris and Soliris.^17,18,20

Stakeholder Perspectives

Patient Group Input

This section was prepared by CADTH staff based on the input provided by patient groups.

One patient advocacy group, aHUS Canada, provided input on the treatment of aHUS. This group gathered information from 19 caregivers and 41 patients from inside and outside Canada through an online survey conducted in June 2022. Of these 60 respondents, 19 had experience with the drug under review.

Respondents identified anemia, low platelet count, and acute renal failure as the most difficult primary symptoms to control. Diminished quality of life, helplessness, post-traumatic stress disorder, fatigue and/or exhaustion, headache, high blood pressure, inability to travel, frequent hospital visits, and kidney issues and/or dialysis are some of the experiences that respondents have while living with aHUS. According to aHUS Canada, aHUS patients who require dialysis and need a kidney transplant are not eligible for a transplant in Canada unless they receive eculizumab infusions at the time of the transplant.

Caregivers to patients with aHUS also face emotional and financial challenges because the process to access eculizumab and alternatives differs from province to province. Respondents described financial struggles, anxiety about access to treatment, the need to protect organs, exhaustion, memory loss and/or brain fog as aspects of the disease that are hardest to control.

Respondents identified plasma therapy (fresh frozen plasma or plasmapheresis), eculizumab infusions, and long-term dialysis as the currently available treatments for patients with aHUS. Side effects reported by the respondents included nausea, headache, fatigue, anaphylactic reaction to plasma, vein collapse, infection, anxiety, kidney failure, uncontrolled blood pressure, migraines, exhaustion, memory loss and/or brain fog, central line issues, muscle crumps, insomnia, abdominal pain, fever and chills, weight gain or loss, and being refractory to plasma therapy, among others.

While discussing their expectations for new drugs, patients reported that access to treatment and freedom of choice were critical components in managing the disease. However, quality of life was the most commonly cited desired outcome, and it was affected by factors like choice in care, frequency of appointments, and drug affordability. The abilities to travel, focus on family, and have more time between appointments were also described as critical to patients’ mental health. Moreover, frequent blood tests and IV therapies or ports were reported to be significant problems for many patients. While 1 caregiver pointed out the importance of maintaining “venous access for continuous access to eculizumab,” other patients shared their ineligibility for ports due to damaged veins from the disease. Patients also expressed the importance of requiring less frequent treatments.

While discussing their experiences with ravulizumab, patients listed benefits that included more energy, less vein damage, fewer treatments, fewer symptom fluctuations, greater freedom of choice, and less anxiety. However, they also reported experiencing headache, nausea, and body aches right after their infusion or during the month after the infusion.

Clinician Input

Input From Clinical Experts Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise in 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 2 clinical specialists with expertise in the diagnosis and management of adult and pediatric patients with aHUS.

Unmet Needs

The clinical experts consulted by CADTH for this review indicated that administration of eculizumab every 2 weeks interferes with a patient’s quality of life by consuming time that could be spent working, travelling, or with friends and family. Administration of eculizumab every 2 weeks can also be an issue when it comes to venous access fatigue and comes with the societal cost of nursing and allied health care support. The biggest limitation to the current treatment is prohibitive cost: most centres will fund an initial treatment or a few treatments, but very few have the resources to fund lifelong treatment. Inclusion in provincial formularies is inconsistent across provinces. and private insurance coverage is not common. Often, patients or their health teams advocate for subsidies or payment in full, but they are not always successful. With respect to venous access fatigue, most patients would be candidates for portacaths or central lines, which are normally offered to chemotherapy patients.

Place in Therapy

The mechanism of action of ravulizumab is the same as that of eculizumab, which is the only other approved treatment for aHUS. Ravulizumab would not be added to other treatments. Instead, the clinical experts believed it would replace eculizumab as the treatment of choice for aHUS. The clinical experts anticipated that ravulizumab would have similar or equivalent efficacy as eculizumab, with the potential of a better therapeutic profile and/or reduced therapeutic burden. The reasons clinical experts believed ravulizumab could become the first-line treatment of choice included the potential for improved patient quality of life and better cost-effectiveness because of the fewer infusions required.

Theoretically, as with other biologics that use the same target molecule, tachyphylaxis to 1 medication may open up options to treat with the second; therefore, acquired nonresponse may be a consideration to switch therapies.

Patient Population

The patients most suitable for treatment with ravulizumab are those diagnosed with aHUS. The patients most in need of intervention are those with severe TMA with associated end organ damage, such as acute kidney injury or brain ischemia.

The patients who are least suitable for treatment with ravulizumab are those with TMA that is clearly due to a secondary cause, such as malignant hypertension, malignancy, or infection. There may be some benefit to using eculizumab in patients with certain autoimmune diseases in which there is histological evidence of TMA as well as evidence of complement dysregulation (e.g., some variants of lupus).

Patients who qualify for treatment would be identified by physicians with expertise in TMAs. These include nephrologists, hematologists, and internal medicine physicians, who would make the diagnosis based on clinical examination, lab investigations, and genetic testing for complement dysregulation, and by excluding other causes of TMA.

Diagnosing aHUS can be very challenging because no single diagnostic test can confirm the disease. In many situations, it is a diagnosis of exclusion. For this reason, misdiagnosis is a risk. One clinical expert indicated that testing has improved and the difficulties of diagnosis have decreased, suggesting that these diagnostic challenges may have been a greater issue 10 or 15 years ago, when genetic and biochemical assessments of complement pathways were less accessible; however, these tests are now more available, often on a quick turnaround, even when sent out of province. The diagnosis of aHUS requires evidence of TMA, such as schistocytes, elevated lactate dehydrogenase, decreased haptoglobin, decreased hemoglobin, and thrombocytopenia. These lab abnormalities should also coincide with 1 or more of the following: neurological symptoms, acute renal failure, or gastrointestinal symptoms, although any organ system can be involved (e.g., pancreas, heart). One clinical expert indicated that the haptoglobin is not the most reliable diagnostic indicator, and that LDH level is a better test. In patients with aHUS, early initiation of plasmapheresis until the diagnosis is confirmed is critical, given the increased mortality of untreated TTP. Most centres have access to ADAMTS13 activity testing with a turnaround of 24 hours to 48 hours. Adult patients may be offered treatment with plasmapheresis before the results are known; however, for pediatric patients, physicians would prefer to wait for the results, if feasible.

Etiologies that mimic TMA need be excluded during diagnosis, including infections, medications, malignancy, scleroderma, antiphospholipid antibody syndrome, systemic lupus erythematosus, malignant hypertension, disseminated intravascular coagulation, preeclampsia, and HELLP syndrome. TTP can be distinguished from aHUS by measuring ADAMTS13 level. If ADAMTS13 is greater than 5% and the patient is resistant to plasma exchange, then the diagnosis is more likely to be aHUS than TTP. Screening for complement mutations and antibodies should be performed. In pediatric populations, where TTP is less common, clinicians would likely not initiate plasmapheresis first, but agree this would be prudent to do for older patients, particularly older adults. Once aHUS has been diagnosed, C5 inhibition may be used as first-line therapy.

Treatment goals include the resolution of TMA, with normal platelet and LDH counts, as well as the resolution of acute kidney injury and/or neurological sequelae and the stabilization of end organ damage.

More sophisticated testing is available as well, including levels of soluble C5b-9. Levels of C5b-9 are elevated during aHUS and subside with treatment because C5b-9 is generated as a product of complement activation. If levels are initially low, most centres may follow C3 and complement component 4 levels to monitor for recovery. Duration of treatment with C5 inhibition is unknown. Based on the available data, if there are no high-risk complement genetic variants, then termination of treatment could be considered after 6 months to 12 months. Lifelong treatment may be considered for patients with high-risk complement genetic variations; however, 30% to 40% of patients may have no known genetic disposition. As noted previously, patients with aHUS who have DGKE mutations are unlikely to benefit from treatment with C5 inhibitors (e.g., eculizumab and ravulizumab). Clinical experts highlighted that patients with DGKE mutations can safely come off of C5 inhibitors because these are unlikely to help if no response to treatment has been observed.

Assessing Response to Treatment

The outcomes indicating a favourable response include the resolution of TMA (i.e., normalization of LDH and platelet count), the stabilization of end organ damage (such as acute kidney injury and brain ischemia), transplant graft survival (in susceptible individuals), and dialysis avoidance (in patients who are pre-ESKD).

Discontinuing Treatment

It is possible to discontinue treatment with ravulizumab in patients with aHUS who do not have a genetic complement mutation 3 months to 6 months after they achieve remission. Close monitoring of the patient for 1 year after they discontinue therapy is recommended to monitor for relapse. One clinical expert indicated that 30% to 40% of patients do not have a genetic diagnosis. Treatment discontinuation in patients with a high-risk mutation in complement is associated with a 50% relapse rate; therefore, discontinuing treatment in these patients is more challenging.

Treatment discontinuation also needs to be considered in the setting of severe infections. However, 1 clinical expert indicated that this would entail restarting the medication with reduced dose or prophylactic anti-infectives.

Prescribing Conditions

Ravulizumab can be given at home with nursing support or at an infusion centre. A specialist, such as a nephrologist or hematologist with expertise in TMA, needs to monitor the patient.

Additional Information

The clinical experts expressed that despite the cost savings associated with ravulizumab’s less frequent administration, the drug’s cost still needs to be reasonable. The experts highlighted that if the cost is much higher than that of eculizumab, most health systems would constrain use to eculizumab.

Clinician Group Input

No clinician group input was received for this review.

Drug Program Input

Drug programs provide input on each drug being reviewed through CADTH’s 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 CADTH are summarized in Table 4.

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

C5 = complement component 5; CDEC = CADTH Canadian Drug Expert Committee; PNH = paroxysmal nocturnal hemoglobinuria; TMA = thrombotic microangiopathy.

Clinical Evidence

The clinical evidence included in the review of ravulizumab is presented in 3 sections. The Systematic Review section includes pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those studies that were selected according to an a priori protocol. The second section includes indirect evidence from the sponsor and indirect evidence selected from the literature that met the selection criteria specified in the review (when available). The third section includes sponsor-submitted long-term extension studies and additional relevant studies that were considered to address important gaps in the evidence included in the systematic review (if submitted).

Systematic Review (Pivotal and Protocol-Selected Studies)

Objectives

To perform a systematic review of the beneficial and harmful effects of ravulizumab for the treatment of adult and pediatric patients with aHUS to inhibit complement-mediated TMA.

Methods

Studies selected for inclusion in the systematic review included pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those meeting the selection criteria presented in Table 5. Outcomes included in the CADTH review protocol reflect outcomes considered to be important to patients, clinicians, and drug plans.

Table 5: Inclusion Criteria for the Systematic Review

AE = adverse event; aHUS = atypical hemolytic uremic syndrome; CKD = chronic kidney disease; eGFR = estimated glomerular filtration rate; ESRD = end-stage renal disease; FACIT-F = Functional Assessment of Chronic Illness Therapy – Fatigue; HRQoL = health-related quality of life; LDH = lactate dehydrogenase; SAE = serious adverse event; TMA = thrombotic microangiopathy; WDAE = withdrawal due to adverse event.

^aThese outcomes were identified as being of particular importance to patients in the input received by CADTH from patient groups.

The literature search for clinical studies was performed by an information specialist using a peer-reviewed search strategy according to the PRESS Peer Review of Electronic Search Strategies checklist.³³

Published literature was identified by searching the following bibliographic databases: MEDLINE All (1946–) through Ovid and Embase (1974–) through Ovid. All Ovid searches were run simultaneously as a multifile search. Duplicates were removed using Ovid deduplication for multifile searches, followed by manual deduplication in Endnote. The search strategy comprised both controlled vocabulary, such as the National Library of Medicine’s MeSH (Medical Subject Headings), and keywords. The main search concept was Ultomiris (ravulizumab). Clinical trials registries were searched: the US National Institutes of Health’s clinicaltrials.gov, WHO’s International Clinical Trials Registry Platform search portal, Health Canada’s Clinical Trials Database, and the European Union Clinical Trials Register.

No filters were applied to limit the retrieval by study type. Retrieval was not limited by publication date or by language. Conference abstracts were excluded from the search results. Refer to Appendix 1 for the detailed search strategies.

The initial search was completed on July 07, 2022. Regular alerts updated the search until the meeting of the CADTH Canadian Drug Expert Committee on October 26, 2022.

Grey literature (literature that is not commercially published) was identified by searching relevant websites from the CADTH Grey Matters: A Practical Tool For Searching Health-Related Grey Literature checklist.)³⁴ Included in this search were the websites of regulatory agencies (the US FDA and European Medicines Agency). Google was used to search for additional internet-based materials. Refer to Appendix 1 for more information on the grey literature search strategy.

These searches were supplemented by reviewing bibliographies of key papers and contacting appropriate experts. In addition, the sponsor was contacted for information regarding unpublished studies.

Two CADTH clinical reviewers independently selected studies for inclusion in the review based on titles and abstracts, according to the predetermined protocol. Full-text articles of all citations considered potentially relevant by at least 1 reviewer were acquired. Reviewers independently made the final selection of studies to be included in the review, and differences were resolved through discussion.

Findings From the Literature

A total of 2 studies^21,22 were identified from the literature for inclusion in the systematic review (Figure 1). The 2 included studies are presented in 7 documents^{12,21,22,35-38} and summarized in Table 6. A list of excluded studies is presented in Appendix 3.

Figure 1: Flow Diagram for Inclusion and Exclusion of Studies

Table 6: Details of Study 311 (Adults)

AMR = antibody-mediated rejection; CKD = chronic kidney disease; CNI = calcineurin inhibitor; eGFR = estimated glomerular filtration rate; ESKD = end-stage kidney disease; FACIT-F = Functional Assessment of Chronic Illness Therapy – Fatigue; Hb = hemoglobin; HRQoL = health-related quality of life; HUS = hemolytic uremic syndrome; LDH = lactate dehydrogenase; LLN = lower limit of normal; mTORi = mammalian target of rapamycin inhibitor; OL = open label; q.8.w. = every 8 weeks; TMA = thrombotic microangiopathy; ULN = upper limit of normal.

Source: Study 311 Clinical Study Report.²¹

Table 7: Details of Study 312 (Children)

aHUS = atypical hemolytic uremic syndrome; AMR = antibody-mediated rejection; CKD = chronic kidney disease; CNI = calcineurin inhibitor; eGFR = estimated glomerular filtration rate; ESKD = end-stage kidney disease; FACIT-F = Functional Assessment of Chronic Illness Therapy – Fatigue; HUS = hemolytic uremic syndrome; LDH = lactate dehydrogenase, LLN = lower limit of normal; mTORi = mammalian target of rapamycin inhibitor; NA = not applicable; OL = open label; q.4.w. = every 4 weeks; q.8.w. = every 8 weeks; TMA = thrombotic microangiopathy; ULN = upper limit of normal.

Source: Study 312 Clinical Study Report.²²

Description of Studies

Two manufacturer-sponsored studies were included in this review: Study 311²¹ and Study 312.²²

Study 311

Study 311 is an ongoing, phase III, prospective, multicentre, single-arm, open-label trial that includes adult patients with aHUS.²¹ The key objective of the Study 311 is to evaluate the safety and efficacy of ravulizumab (IV infusion) in complement inhibitor treatment–naive adult patients (aged ≥ 18 years) with aHUS. The trial was conducted at 41 sites in 14 countries (including Canada, the US, Australia, and 11 countries in Europe and Asia). The key characteristics of the study design are summarized in Table 6. The study consists of a screening period (≤ 7 days), a 26-week initial evaluation period, and an extension period until the product is registered or approved (in accordance with country-specific regulations) or for up to 4.5 years, whichever occurs first, as illustrated in Figure 2.

Enrolment started on March 18, 2017, and is ongoing.²¹ The cut-off date for the data presented herein was July 2, 2019. As of the cut-off date, a total of 58 patients were included and 56 patients had received at least 1 weight-based dose of IV ravulizumab. The primary outcome was complete TMA response during the initial 26-week evaluation period, which was defined as the normalization of hematologic parameters (platelet count and LDH) and an improvement of at least 25% in serum creatinine from baseline. The secondary outcomes were hematologic normalization (platelet count and LDH), hematologic TMA parameters (platelet count, LDH, and hemoglobin), hemoglobin response (> 2% increase), dialysis requirement status, eGFR, CKD stage, fatigue, HRQoL, and safety outcomes. Health care resource utilization, patient-reported aHUS symptoms, and extrarenal signs and symptoms of aHUS were reported as exploratory outcomes.

Major protocol deviations were reported in 43 patients (74.1%) in Study 311. Among those with protocol violations, 25 patients (43.1%) were in the eligibility and entry criteria and 15 patients (25.9%) were in the category of SAE reporting criteria (Table 39).

Figure 2: Study Design Schematic for Study 311

Source: Study 311 Clinical Study Report.²¹

Study 312

Study 312 is an ongoing, phase III, prospective, multicentre, single-arm, open-label trial conducted in pediatric patients (aged < 18 years) with aHUS.²² Study 312 included 2 cohorts (cohort 1 and cohort 2). Cohort 1 included 21 complement inhibitor–naive children (aged < 18 years) with aHUS. The key objective for cohort 1 was to evaluate the safety and efficacy of ravulizumab (IV infusion) in complement inhibitor–naive children (aged < 18 years) with aHUS. Cohort 2 included 10 children (aged < 18 years) with aHUS who had been treated with eculizumab. The key objective for cohort 2 was to evaluate the safety and efficacy of ravulizumab (IV infusion) in children (aged < 18 years) with aHUS with stable TMA parameters following a switch from eculizumab to ravulizumab treatment. Study 312 was conducted at 20 sites in 8 countries (including the US and 7 countries in Europe and Asia; there were no sites in Canada). The key characteristics of the study design are summarized in Table 7 The study consists of a screening period (≤ 7 days), a 26-week initial evaluation period, and an extension period lasting until the product is registered or approved (in accordance with country-specific regulations) or for up to 4.5 years, whichever occurs first, as illustrated in Figure 3.

The enrolment for this study started on September 1, 2017, and is ongoing.²² The cut-off date for the data presented herein was December 3, 2019. As of the cut-off date, a total of 21 patients were included in Study 312, cohort 1, and 18 patients had received at least 1 weight-based dose of IV ravulizumab. A total of 10 patients were included in cohort 2, and all received at least 1 dose of ravulizumab. The primary outcome was complete TMA response during the initial 26-week evaluation period among patients in cohort 1; response was defined as the normalization of hematologic parameters (platelet count and LDH) and an improvement of greater than or equal to 25% in serum creatinine from baseline. The secondary outcomes were hematologic normalization (platelet count and LDH), hematologic TMA parameters (platelet count, LDH, and hemoglobin for cohort 1 only), hemoglobin response (> 2% increase; cohort 1 only), dialysis requirement status, eGFR, CKD stage, fatigue, and safety outcomes. Health care resource utilization, patient-reported aHUS symptoms, and extrarenal signs and symptoms of aHUS were reported as exploratory outcomes.

Major protocol deviations were reported in 14 patients (66.7%) in cohort 1. Among those with protocol violations, 9 patients (42.9%) were in the eligibility and entry criteria category and 7 patients (33.3%) were in the category of SAE reporting criteria (Table 39). No major protocol violations were reported in cohort 2.

Figure 3: Study Design Schematic for Study 312 (Cohort 1 and Cohort 2)

q.4.w. = every 4 weeks; q.8.w. = every 8 weeks.

Source: Study 312 Clinical Study Report.²²

Populations

Inclusion and Exclusion Criteria

Study 311

Eligible patients were adults (aged 18 years and older) weighing at least 40 kg with evidence of active TMA during the screening period or within 28 days before the start of the screening period, defined as: platelet count less than 150,000/μL; lactate dehydrogenase greater than or equal to 1.5 times the upper limit of normal (ULN) and hemoglobin less than or equal to the lower limit of normal for age and gender; and serum creatinine level greater than or equal to the ULN during the screening period. (Patients who required dialysis for acute kidney injury were also eligible.) Patients with renal transplant were permitted, but must have had either a prior history of aHUS or persistent evidence of TMA in the 4 days after modifying the dose of calcineurin inhibitors or mammalian target of rapamycin inhibitors. Postpartum patients were permitted, but must have had either a prior history of aHUS or persistent evidence of TMA for more than 3 days after childbirth. Patients must have received meningococcal vaccination at the time of starting ravulizumab and were required to receive treatment doses of antibiotics from the time of the first dose of ravulizumab until at least 2 weeks after vaccination.

Key exclusion criteria were: a deficiency of ADAMTS13 (activity < 5%, suggestive of TTP); known Shiga toxin–related HUS; and other HUSs, such as drug exposure–related HUS with a positive direct Coombs test. Patients receiving immunosuppressive therapies were excluded unless they were part of an established post-transplant antirejection regimen, had confirmed anticomplement antibodies, or were using steroids for a different condition. Patients receiving plasma exchange and/or plasma infusion for a period of 28 days or longer before screening, or who were on chronic dialysis at screening, were excluded. Among patients without a kidney transplant, history of kidney disease suggesting an underlying disease other than aHUS were excluded.

Study 312

Study 312 included pediatric patients (aged < 18 years with > 5 kg body weight). For cohort 1, the key inclusion and exclusion criteria were the same as those for Study 311. Cohort 2 included children with a documented diagnosis of aHUS who were treated with eculizumab for at least 90 days before screening and showed clinical evidence of response, indicated by stable TMA parameters at screening, including LDH lower than 1.5 times ULN; platelet count 150,000/μL or higher; and eGFR greater than 30 mL/min/1.73 m² using the Schwartz formula. The key exclusion criteria for cohort 2 were the same as those for cohort 1 (Table 7).

Baseline Characteristics

Study 311

The main baseline demographics and disease characteristics of the 56 adult patients (for the FAS population) in the trial are summarized in Table 8 and Table 9. In the FAS, the median age (years) at time of first aHUS symptoms was 40.1 years (range, 9.3 years to 76.6 years). The median age at the time of first infusion was 40.1 years (range, 19.5 years to 76.6 years). Thirty-seven patients (66.1%) were female. A total of 51.8% of patients were white and 26.8% were Asian. At baseline, 30 patients (53.6%) met the protocol-specified TMA criteria at day 1, based on central laboratory results. Genetic mutations were present in 2 patients (3.6%), while 52 patients (92.9%) presented without gene mutations (2 patients had unknown status). The median time from the first aHUS symptom to the first dose of ravulizumab was 0.28 months (range, 0 months to 215.0 months).

The baseline median platelet level was 95.25 × 10⁹/L. The baseline median serum LDH level was 508.00 U/L. The baseline median eGFR was 10.00 mL/min/1.73 m². At baseline, 39 patients (72.2%) presented with CKD stage 5, and 9 patients (16.7%) presented with CKD stage 4. At baseline, 29 patients (51.8%) were on dialysis.

Eight patients (14.3%) had received a kidney transplant before entering the study, but none of these transplants were related to aHUS. Prior to the study, the numbers (proportions) of patients who had received plasma exchange and/or plasma infusion (related to the current TMA), packed RBC transfusions, and platelet transfusions were 48 patients (82.8%), 17 patients (29.3%), and 6 patients (10.3%), respectively. Three patients (5.2%) received selective immunosuppressants (not eculizumab) before the study.

The majority of patients (92.9%) had pretreatment extrarenal signs or symptoms of aHUS. At the time of the first dose of the study drug, 48 of 56 patients (85.7%) were hospitalized due to aHUS (refer to Table 36). Fifty-three patients (94.6%) in the FAS had previous hospitalizations and/or emergency room visits due to aHUS. Twenty-seven patients (50.9%) had received intensive care, with a mean duration of stay in the intensive care unit of 10.1 days (SD = 10.0 days).

Study 312, Cohort 1

The main baseline demographics and disease characteristics of the 18 pediatric patients in cohort 1 (for the FAS population) are summarized in Table 8 and Table 9. The median age (years) at the time of first aHUS symptoms was 4.75 years (range, 0.8 years to 14.7 years). The median age at the time of first infusion was 5.2 years (range, 0.9 years to 17.3 years). Ten patients (55.6%) were female. A total of 50% of patients were white, and 27.8% were Asian. All 18 patients (100%) presented without gene mutation. The median time from the first aHUS symptom to the first dose of ravulizumab was not reported.

The baseline median platelet level was 51.25 × 10⁹/L. The baseline median serum LDH level was 1963.00 U/L. The baseline median eGFR was 22.0 mL/min/1.73 m². At baseline, a total of 6 patients (33.3%) presented with CKD stage 5 and 8 patients (44.4%) presented with CKD stage 4. At baseline, patients 6 (33.3%) were on dialysis.

One patient had received a kidney transplant (related to aHUS) before entering the study. Prior to the study, no patients received plasma exchange and/or plasma infusion related to the current TMA. The numbers (proportions) of patients who received packed RBC transfusions and platelet transfusions were 12 patients (57.1%) and 4 patients (19%), respectively. One patient (4.8%) received selective immunosuppressants (not eculizumab) before the study.

Thirteen patients (72.2%) had pretreatment extrarenal signs or symptoms of aHUS at baseline (refer to Table 36). All 18 patients had experienced a hospitalization and/or emergency room visit due to aHUS before the start of screening. Prior to screening, 7 patients (38.9%) had received intensive care during their hospitalizations due to aHUS, with a mean duration of stay in the intensive care unit of 9.0 days (SD = 17.68 days). At the time of the first dose of study drug, 17 patients (94.4%) were hospitalized due to aHUS.

Study 312, Cohort 2

For cohort 2, the main baseline demographics and disease characteristics of the 10 pediatric patients with stable TMA who were eculizumab-treated (for the FAS population) in the trial are summarized in Table 8 and Table 9. The median age (years) at time of first aHUS symptoms was 4.70 years (range, 0.4 years to 8.3 years). The median age at the time of first infusion was 12.5 years (range, 12 years to 15.5 years). Nine patients (90%) were male and 1 patient (10%) was female. A total of 50% of patients were white, and 40% were Asian. The median time from the first aHUS symptom to the first dose of ravulizumab was not reported.

The baseline median platelet level was 281.75 × 10⁹/L (range = 207 × 10⁹/L to 415.5 × 10⁹/L). The baseline median serum LDH level was 206.50 U/L (range, 138.5 U/L to 356 U/L). The baseline median eGFR was 99.75 mL/min/1.73 m² (range, 54 mL/min/1.73 m² to 136.5 mL/min/1.73 m² [normal range, ≥ 60 mL/min/1.73 m²]). No patients presented with CKD stage 5 or stage 4. No patient was on dialysis.

One patient had received a kidney transplant (related to aHUS) before entering the study. Prior to the study, no patients had received plasma exchange and/or plasma infusion related to the current TMA. All 10 patients received and responded to the eculizumab treatment.

One of the 10 patients had pretreatment extrarenal signs or symptoms of aHUS at baseline. None of the 10 patients had experienced a hospitalization and/or emergency room visit due to aHUS before start of screening.

Table 8: Baseline Demographic Characteristics (FAS)

FAS = full analysis set; LDH = lactate dehydrogenase; LLN = lower limit of normal; NR = not reported; TMA = thrombotic microangiopathy; ULN = upper limit of normal.

^aPatients can have multiple races selected.

^bPlatelet count less than 150,000/μL; LDH greater than or equal to 1.5 times the ULN; hemoglobin less than or equal to the LLN; serum creatinine level greater than or equal to the ULN.

Sources: Study 311 Clinical Study Report²¹ and Study 312 Clinical Study Report.²²

Table 9: Disease Characteristics (FAS)

aHUS = atypical hemolytic uremic syndrome; CKD = chronic kidney disease; eGFR = estimated glomerular filtration rate; LDH = lactate dehydrogenase; NR = not reported; SD = standard deviation; TMA = thrombotic microangiopathy.

^aDialysis at baseline was recorded as “yes” for patients who had received dialysis within 5 days before study drug initiation.

^bThe percentage was based on the total number of patients.

^cThe percentage meant 1 of 1 patient (100%) who had a kidney transplant.

^dThe normal value was provided in the Study 311 Clinical Study Report only for adult patients. The normal value was not provided in the Study 312 Clinical Study Report.

^eBaseline CKD stage was available for 54 patients.

Sources: Study 311 Clinical Study Report²¹ and Study 312 Clinical Study Report.^12,22

Interventions

Study 311

A total 56 of 58 enrolled patients received the IV ravulizumab treatment. An interactive voice and/or web response system was used to assign vials containing ravulizumab to each patient. During the initial 26-week initial evaluation period, patients received a weight-based loading dose (refer to Figure 2) of ravulizumab IV on day 1, then every 8 weeks (all patients were ≥ 40 kg and received body weight-based maintenance doses on days 15, 71, and 127). After the 26-week initial evaluation period, all patients rolled into an extension period during which they received ravulizumab every 8 weeks (refer to Figure 2). A patient who discontinued and restarted ravulizumab on a scheduled study visit received a loading dose, a supplemental maintenance dose 2 weeks later, and a maintenance dose 6 weeks later, resuming an every-8-weeks regimen thereafter. If the decision to re-treat with ravulizumab occurred between scheduled study visits, the dosing regimen was to be determined by the Alexion medical monitor and the investigator. In the extension period, patients receive ravulizumab until product registration or approval (in accordance with country-specific regulations) or for up to 4.5 years, whichever occurs first (refer to Figure 2).

Study 312

In cohort 1, a total of 18 patients of 21 patients received the intended ravulizumab treatment. The dosing regimen was the same as that described for Study 311, except that the frequency of the body weight–based dosing regimen was every 8 weeks for patients weighing greater than or equal 20 kg and every 4 weeks for patients weighing less than 20 kg (refer to Figure 3).

In cohort 2, all 10 patients received the intended ravulizumab treatment. Day 1 of the study treatment occurred 14 days from the patient’s last dose of eculizumab. Changes in dosing regimen (i.e., dose amount [mg] or dose frequency [every 4 weeks or every 8 weeks]) were based on the same weight-based regimen (refer to Figure 3). Patients changing from every 4 weeks to every 8 weeks (i.e., those weighing 20 kg or more) or from every 8 weeks to every 4 weeks (i.e., those weighing less than 20 kg) were administered their first dose of the new scheduled dosing regimen (i.e., every 4 weeks or every 8 weeks) on the first ravulizumab administration day.

If the investigator and Alexion medical monitor mutually agreed that a patient would potentially benefit from a supplemental dose of ravulizumab, this supplemental dose may have been administered and the decision was documented. If the investigator and Alexion medical monitor mutually agreed that the infusion volume (120 mL) of the loading dose for patients weighing 5 kg to 9.9 kg (i.e., 600 mg) was too high for an individual patient, this dose may have been administered as 2 separate infusions no more than approximately 24 hours apart. This decision was also documented.

Concomitant Therapy

In both Study 311 and Study 312, concomitant medications were considered to be those the patient received from the first infusion of ravulizumab through 56 days after the patient’s last dose of the study drug. Any concomitant medication deemed necessary for the patient’s standard of care during the study, or for the treatment of any adverse event, was given at the discretion of the investigator.

All patients in the study (even those who had discontinued Study 311 and Study 312 in the extension period, but remained in the study) were prohibited from receiving any of the following medications and procedures from the first dose of the study drug until the completion of the study or early termination of the patient from the study:

• eculizumab or other complement inhibitors

• other investigational drugs or devices as part of a clinical trial

• IV immunoglobulin (unless for an unrelated medical need, e.g., hypogammaglobinemia)

• rituximab

• plasma exchange and/or plasma infusion

• new dialysis within the first 48-hour period following the first dose given in Study 311 and Study 312, unless there was a compelling medical need, such as in cases of hypervolemia unresponsive to diuretics, refractory electrolyte imbalance, or new-onset uremic encephalopathy (exceptions had to be approved on a case-by-case basis by Alexion before dialysis).

Outcomes

A list of efficacy end points identified in the CADTH review protocol that were assessed in the clinical trials included in this review is provided in Table 10. These end points are subsequently summarized. A detailed discussion and critical appraisal of the outcome measures is provided in Appendix 5.

The primary outcome for both Study 311 and Study 312, cohort 1 was complete TMA response (i.e., normalization of platelets and LDH and a 25% serum creatinine improvement). Patients must have met all the criteria for complete TMA response at 2 separate assessments obtained at least 4 weeks (28 days) apart, and at any measurement in between.

The secondary outcomes were hematological parameters (i.e., platelets, LDH, hemoglobin), time to complete TMA response, HRQoL, fatigue, renal function (e.g., eGFR, change in CKD stage, progression to end-stage renal disease), and dialysis-free status. The exploratory outcomes included symptom reduction and hospitalization.

The outcomes were measured at each visit.

Adverse events were assessed throughout both studies.^21,22 A TEAE was defined as any adverse event that started during or after the first infusion of the study drug. Adverse events that started 56 days or more after the last dose of the study drug were not considered to be TEAEs. Adverse events were coded using MedDRA Version 21.0. The severity of adverse events was graded using version 4.03 of the Common Terminology Criteria for Adverse Events.^21,22

Table 10: Summary of Outcomes of Interest Identified in the CADTH Review Protocol

AE = adverse event; CKD = chronic kidney disease; eGFR = estimated glomerular filtration rate; FACIT-F = Functional Assessment of Chronic Illness Therapy – Fatigue; Hb = hemoglobin; LDH = lactate dehydrogenase; NR = not reported; TMA = thrombotic microangiopathy.

Sources: Study 311 Clinical Study Report²¹ and Study 312 Clinical Study Report.²²

Statistical Analysis

The statistical analyses for Study 311 and Study 312 are described herein.

No formal hypothesis testing was planned in the 2 pivotal trials. An interim analysis was planned at the end of the initial 26-week evaluation period after all patients had completed or withdrawn from this phase of the study. This analysis allowed for the evaluation of the primary end point (i.e., complete TMA response). Additionally, a second analysis to summarize long-term efficacy and overall safety was performed at the data-cut-off dates of July 2, 2019, for Study 311 and December 3, 2019, for Study 312.

Power Calculation

Sample size and power calculations were not performed. However, the plan was to enrol approximately 55 patients in Study 311 and 23 to 28 patients in Study 312.

Primary Analysis

Statistical analysis methods for the efficacy outcomes are shown in Table 11. No multiplicity control was performed because there was no formal hypothesis testing. Efficacy analyses were performed using the FAS, the primary efficacy population.

The primary efficacy outcome (complete TMA response at week 26) was assessed by calculating the point estimate and a 95% CI for the proportion of complete TMA responders in patients treated with ravulizumab. The 95% CI was based on the asymptotic Gaussian approximation method with a continuity correction.

Secondary Outcomes

Hematologic Normalization

The number and proportion of patients who achieved hematologic normalization, defined as the normalization of both platelet count and LDH, was summarized over time with a 2-sided 95% CI for each postbaseline time point.

Hematologic TMA Parameters

Hematologic TMA parameters (platelets, LDH, and hemoglobin) were summarized at baseline and at each postbaseline time point using descriptive statistics for continuous variables for the observed value as well as the change from baseline. A mixed model for repeated measures (MMRM) with the fixed, categorical effect of visit and fixed, continuous effect of the specific test’s baseline value as covariates was performed to test whether changes differed from 0 at each time point.

Hemoglobin Response

The number and proportion of patients with an increase in hemoglobin of 20 g/L or more from baseline observed at 2 separate assessments at least 4 weeks apart (and at any measurement in between) was summarized over time by presenting the number and proportion of responders along with a 2-sided 95% CI for each postbaseline time point.

Time to Complete TMA Response

For the secondary efficacy end point of time to complete TMA response, Kaplan-Meier cumulative distribution curves were generated along with 2-sided 95% CIs. Patients who did not have a response were censored at the date of last visit or study discontinuation when the analysis was performed.

Health-Related Quality of Life

Quality of life was evaluated using the 3-Level EQ-5D. These data were summarized at baseline and at each postbaseline time point using descriptive statistics for continuous variables for the observed value as well as the change from baseline. An MMRM with the fixed, categorical effect of visit and fixed, continuous effect of the specific test’s baseline value as covariates was performed to test whether changes differed from 0 at each time point.

Symptoms Reduction

Fatigue was assessed using the FACIT-F version 4. The FACIT-F data were summarized at baseline and at each postbaseline time point using descriptive statistics for continuous variables for the observed value as well as the change from baseline. An MMRM with the fixed, categorical effect of visit and fixed, continuous effect of the specific test’s baseline value as covariates was performed to test whether changes differed from 0 at each time point.

Dialysis Requirement Status

For patients requiring dialysis within 5 days before ravulizumab treatment initiation, the proportion of patients no longer requiring dialysis was summarized over time using proportions. A 2-sided 95% CI for the proportion receiving dialysis was provided.

eGFR Value and Change From Baseline

Kidney function evaluated by eGFR was summarized at baseline and at each postbaseline time point using descriptive statistics for continuous variables for the observed value as well as the change from baseline. A value of 10 mL/min/1.73 m² for eGFR was imputed for patients requiring dialysis for acute kidney injury. This summary was repeated by kidney transplant status at enrolment. An MMRM with the fixed, categorical effect of visit and fixed, continuous effect of the baseline value as covariates was performed to test whether changes differed from 0 at each time point.

CKD Stage

CKD stage (refer to Table 38) was summarized over time by presenting the number and proportion of patients who improved (excluding those with stage 1 at baseline, given that they cannot improve), worsened (excluding those with stage 5 at baseline, given that they cannot worsen), and stayed the same compared to their CKD stage at baseline. Stage 5 was considered the worst category, while stage 1 was considered the best category. A 2-sided 95% CI for the proportion was provided for each category.

Other Outcomes

Mortality, presence of bleeding, packed RBC transfusions, and soluble MAC level were not assessed as efficacy outcomes in the 2 pivotal studies (Study 311 and Study 312). Symptoms (aside from fatigue) and hospitalization were reported on a by-patient basis in the 2 pivotal studies (CSRs) submitted by the sponsor; there were no summary data submitted. Therefore, symptom reduction and hospitalization have not been reported herein.

Subgroup and Sensitivity Analyses

Subgroup analyses were conducted only for the primary outcome, the complete TMA response. The primary efficacy analysis was repeated separately by the following main relevant subgroups: sex (male, female), age at enrolment (age 12 to 17 years, ≥ 18 years), kidney transplant history (yes, no), gene mutation status (ever positive, always negative), dialysis within 5 days before treatment initiation (yes, no), and whether or not they met all laboratory criteria for TMA, as determined by the central laboratory at day 1. Given that the number of patients in these subgroups may have been small, the CIs were based on exact confidence limits using the Clopper-Pearson method. No subgroup analyses were conducted for any secondary outcomes.

Sensitivity analyses were conducted only for the primary outcome (the complete TMA response). A sensitivity analysis was prespecified to evaluate a slightly modified version of complete TMA response. This modified complete TMA response applied only to the patients who were on dialysis at baseline (i.e., patients requiring dialysis within 5 days before the first dose of ravulizumab [N = 29 patients in Study 311 and N = 6 patients in Study 312, cohort 1]). For these patients, the criterion requiring an improvement from baseline of 25% or more in serum creatinine was replaced by a postbaseline change in dialysis status (from requiring dialysis at baseline to no longer requiring dialysis) that was maintained for at least 4 weeks. The definition of complete TMA response remained the same for all other patients (refer to Table 11). Primary and secondary end points were analyzed on the per protocol set as a sensitivity analysis to observe whether any substantial differences existed in the outcomes for this population compared to the FAS.

Handling of Dropouts and Missing Data

For the evaluation of complete TMA response during the 26-week initial evaluation period (i.e., the primary outcome), patients who missed an efficacy assessment that was part of the definition of complete TMA response while still in the study had their last observation carried forward. For patients who withdrew from the study before week 26, their data up to the time of withdrawal were used to assess complete TMA response. If a day 1 pretreatment assessment was missing, the screening assessment was used as the baseline assessment.

Safety Outcomes

Only descriptive statistics of safety were presented, with evidence summarized based on frequencies and proportion of total patients. Separate summaries were provided for all adverse events, SAEs, and adverse events leading to discontinuation and dose modification. Deaths and their primary causes were summarized.

Table 11: Statistical Analysis of Efficacy End Points (Study 311 and/or Study 312, Where Applicable)

CI = confidence interval; CKD = chronic kidney disease; eGFR = estimated glomerular filtration rate; FACIT-F = Functional Assessment of Chronic Illness Therapy – Fatigue; HRQoL = health-related quality of life; LOCF = last observation carried forward; MMRM = mixed model for repeated measures; TMA = thrombotic microangiopathy.

Sources: Study 311 Clinical Study Report²¹ and Study 312 Clinical Study Report.²²

Analysis Populations

FAS: The FAS was based on a modified intention-to-treat approach in which confirmation of eligibility may have occurred after patients received the study drug. This applied specifically to the inclusion criterion of increased serum creatinine confirmed by a central laboratory and to the following 2 exclusion criteria: known familial or acquired ADAMTS13 confirmed by a central or local laboratory and known Shiga toxin–related HUS confirmed by a central or local laboratory. Accordingly, the FAS included all patients who were determined to have met the previously described criteria, had received at least 1 dose of ravulizumab, and had at least 1 postbaseline efficacy assessment.

PP set: The PP set included all patients in the FAS who received 100% of the planned number of infusions during the 26-week initial evaluation period; did not take any prohibited medications or undergo any prohibited procedures; met the inclusion criteria related to evidence of TMA (i.e., including thrombocytopenia, evidence of hemolysis, and kidney injury, based on laboratory findings, as detailed in Table 6 and Table 7); were willing and able to give written informed consent and comply with the study visit schedule; and did not meet the following exclusion criteria (refer to Table 6 and Table 7):

• positive direct Coombs test

• presence or suspicion of active and untreated systemic bacterial infection that, in the investigator’s opinion, could confound an accurate diagnosis of aHUS or impede the ability to manage the disease

• acute kidney dysfunction within 4 weeks of transplant consistent with the diagnosis of acute AMR according to Banff 2013 criteria,³⁹ among patients with a kidney transplant

• among patients without a kidney transplant, history of kidney disease other than aHUS (such as known kidney biopsy finding suggestive of underlying disease other than aHUS), or known kidney ultrasound finding consistent with an alternative diagnosis to aHUS, or known family history and/or genetic diagnosis of noncomplement-mediated genetic renal disease

• identified drug exposure–related HUS

• plasma exchange and/or plasma infusion for 28 days or longer before the start of screening for the current TMA

• bone marrow transplant and/or hematopoietic stem cell transplant within the past 6 months

• hemolytic uremic syndrome related to known genetic defects of cobalamin C metabolism

• known systemic sclerosis (scleroderma), systemic lupus erythematosus, or antiphospholipid antibody positivity or syndrome

• chronic dialysis (defined as dialysis on a regular basis as renal replacement therapy for ESKD)

• participation in another interventional treatment study or use of any experimental therapy within 30 days before initiation of the study drug on day 1 in this study or within 5 half-lives of that investigational product, whichever was greater

• prior use of eculizumab or other complement inhibitors

• use of tranexamic acid within 7 days before screening.

Safety set: The safety set included all patients who received at least 1 dose of the study drug.

Results

Patient Disposition

Data related to patient disposition in Study 311 and Study 312 are presented in Table 12.

In Study 311, a total of 74 patients were screened, and 58 patients were included. Two treated patients were withdrawn from the study after receiving the first dose of ravulizumab; both were deemed ineligible because they tested positive for Shiga toxin based on stool tests. Their results became known when the local laboratory results were made available following the first dose of the study drug. Among the 56 patients who received at least 1 dose of ravulizumab, 49 patients (84.5%) completed the initial 26-week period; 7 patients (12.5%) discontinued from the treatment; and 9 patients (15.5%) discontinued from the study during the initial 26-week period. At the time of the data cut-off date (July 2, 2019), no patient had completed the treatment, and 17 patients (29.3%) had discontinued the study. The main reasons for discontinuation were withdrawal by patient (n = 5, 8.6%), adverse event (n = 3, 5.2%), physician decision (n = 3, 5.2%), death (n = 2, 3.4%), being deemed ineligible posttreatment (n = 2, 3.4%), and protocol violation (n = 2, 3.4%). A total of 41 patients (70.1%) were continuing in the extension period; 38 of these patients (65.5%) were continuing to receive the study drug.

In Study 312, cohort 1, a total of 21 patients were screened and included. All 21 patients received at least 1 dose of ravulizumab. Seventeen patients (81.0%) completed the initial 26-week period and 4 patients (19.0%) discontinued the study during the initial 26-week period. At the time of the data cut-off date (December 3, 2019), no patient had completed the treatment, and 5 patients (23.8%) had discontinued from both the treatment and the study. The main reasons for discontinuation of treatment were being deemed ineligible posttreatment (n = 2, 9.5%), adverse event (n = 1, 4.8%), physician decision (n = 1, 4.8%), and protocol violation (n = 1, 4.8%). A total of 16 patients (76.2%) were continuing in the extension period.

In Study 312, cohort 2, at the time of the data cut-off date, all 10 included patients had completed the initial 26-week evaluation period and were continuing in the extension period.

Table 12: Disposition of Patients (All Enrolled Patients)

FAS = full analysis set; NR = not reported; PP = per protocol.

^aTwo patients were discontinued from the study because they were deemed ineligible after receiving positive Shiga toxin test results (based on stool tests) after the first dose. Two patients discontinued the study due to death. One patient withdrew consent.

Sources: Study 311 Clinical Study Report²¹ and Study 312 Clinical Study Report.²²

Exposure to Study Treatments

As of the data cut-off date, in Study 311, the median treatment duration in the safety set was 74.07 weeks (range, 0.57 weeks to 118.3 weeks) (refer to Table 40).

For Study 312, cohort 1, the median treatment duration in the safety set was 82.43 weeks (range, 1 week to 110.6 weeks) (refer to Table 40). For Study 312, cohort 2, the median treatment duration in the safety set was 52.29 weeks (range, 49.4 weeks to 58.7 weeks) (refer to Table 40).

Concomitant Treatments

In Study 311, a total of 17 patients (29.3% for the safety set) received packed RBC transfusions, and 3 patients (5.2%) received platelet transfusions; 3 patients (5.2%) received plasma exchange and/or plasma infusion during the study, which was prohibited per the protocol. These patients were discontinued from the study PP; 2 patients discontinued during the initial evaluation period and 1 patient discontinued during the extension period. Nine patients (15.5%) received concomitant blood substitutes and 9 patients (15.5%) received plasma protein fraction selective immunosuppressants. The main potential relevant concomitant medications used during the study are summarized in Table 37.

In Study 312, cohort 1, a total 4 patients (19%) received platelet transfusions. None of the patients received plasma exchange and/or plasma infusion during the study. A total of 2 patients (9.5%) received concomitant blood substitutes, and 6 patients (28.6%) received plasma protein fraction selective immunosuppressants (refer to Table 37).

In Study 312, cohort 2, 0 patients received plasma exchange and/or plasma infusion (refer to Table 37).

Efficacy

Only those efficacy outcomes and analyses of subgroups identified in the review protocol are reported here. Refer to Appendix 4 for detailed efficacy data.

Mortality

Mortality was not assessed as an efficacy outcome in the 2 pivotal studies (Study 311 and Study 312). The information on mortality was reported in the safety outcomes (refer to the Harms section and Table 27).^21,22

Complete TMA Response

Study 311

At week 26, complete TMA response was observed in 30 patients of the 56 patients in the FAS (53.6%; 95% CI, 39.6% to 67.5%) (Table 13 and Figure 7). At the data cut-off date, complete TMA response was observed in 34 patients of the 56 patients in the FAS (60.7%; 95% CI, 47.0% to 74.4%).

Table 13: Complete TMA Response and Components of Complete TMA Response Analysis (Study 311 for Adults, FAS)

CI = confidence interval; FAS = full analysis set; LDH = lactate dehydrogenase; TMA = thrombotic microangiopathy.

^a95% CIs for the proportion were based on the asymptotic Gaussian approximation method with a continuity correction.

^bHematologic normalization includes normalization of platelet count (≥ 150 × 10⁹/L) and normalization of LDH (≤ 246 U/L).

Source: Study 311 Clinical Study Report.²¹

In the PP analysis set, the proportion of responders with complete TMA response was consistent with that in the primary analysis (FAS) at week 26 and at the data cut-off date (Table 42).

Study 311 Subgroup Analysis for Complete TMA Response

Prespecified subgroup analyses for complete TMA response at week 26 are presented in Table 14. At week 26, the complete TMA response rate was generally consistent across subgroups compared with the overall population (53.6%) (refer to Table 14 and Figure 9).

No subgroup analyses were conducted based on baseline platelet count, LDH level, serum creatinine, hemoglobin, duration of plasma therapy, or duration of dialysis before the study. Subgroup analysis was not conducted for the extension phase.

Table 14: Complete TMA Responder Analyses by Subgroups at Week 26 (FAS)

CI = confidence interval; CKD = chronic kidney disease; FAS = full analysis set; NA = not applicable; NR = not reported; TMA = thrombotic microangiopathy.

Notes: The proportion of complete TMA response is based on the responders among treated patients. The numerator is the number of patients achieving complete TMA response during the 26-week initial evaluation period, and the denominator is the number of patients in the FAS. Complete TMA response was not an outcome assessed for Study 312, cohort 2.

^a95% CIs for the proportion are based on exact confidence limits using the Clopper-Pearson method.

^bAlso reported as sensitivity analysis.

Sources: Study 311 Clinical Study Report²¹ and Study 312 Clinical Study Report.²²

Study 311 Sensitivity Analyses for Complete TMA Response

A separate analysis was performed using a modified version of complete TMA response. This modified complete TMA response applied only to the patients who were on dialysis at baseline. For these patients, the criterion requiring an improvement from baseline of greater than or equal to 25% in serum creatinine was replaced by a postbaseline change in dialysis status (from requiring dialysis at baseline to no longer requiring dialysis after treatment) that was maintained for at least 4 weeks. The definition of complete TMA response remained the same for all other patients. At week 26, the modified complete TMA was observed in 32 patients of the 56 patients in the FAS (57.1%; 95% CI, 43.3% to 71.0%). As of the data cut-off date, modified complete TMA response was observed in 36 patients of the 56 patients in the FAS (64.3%; 95% CI, 50.8% to 77.7%).

Study 312, Cohort 1

In Study 312, cohort 1, at week 26, complete TMA response was observed in 14 patients of the 18 patients in the FAS (77.8%; 95% CI, 52.4% to 93.6%) (Table 15 and Figure 8). At the data cut-off date, complete TMA response was observed in 17 patients of the 18 patients in the FAS (94.4%; 95% CI, 72.7% to 99.9%).

Table 15: Complete TMA Response and Components of Complete TMA Response Analysis (Study 312, Cohort 1 FAS)

CI = confidence interval; FAS = full analysis set; LDH = lactate dehydrogenase; TMA = thrombotic microangiopathy.

^a95% CIs for the proportion were based on exact confidence limits using the Clopper-Pearson method.

Source: Study 312 Clinical Study Report.²²

The FAS number and PP set number were the same in Study 312. Therefore, no PP analysis was done.

Study 312, Cohort 2

TMA response was not relevant for this population.

Study 312, Cohort 1 Subgroup Analysis for Complete TMA Response

Prespecified subgroup analyses for complete TMA response at week 26 are presented in Table 14. At week 26, the complete TMA response rate was generally consistent across subgroups compared with the overall population (refer to Table 14 and Figure 10). No subgroup analyses were conducted based on baseline platelet count, LDH level, serum creatine, hemoglobin, duration of plasma therapy, or duration of dialysis before the study. Subgroup analysis was not conducted for the data cut-off date.

Study 312, Cohort 1 Sensitivity Analyses for Complete TMA Response

At week 26, the modified complete TMA was observed in 14 patients of the 18 patients in the FAS (77.8%; 95% CI, 52.4 to 93.6%). As of the data cut-off date, modified complete TMA response was observed in 17 patients of the 18 patients in the FAS (94.4%; 95% CI, 72.7% to 99.9%).

Complete TMA Response Status Over Time

Rates of complete TMA response status over time for Studies 311 and 312 are presented in Table 44 and Table 45, respectively. In Study 311, from the median time to complete TMA response (86 days), the proportion of responders was stable. After achieving complete TMA response, some patients had transient periods during which not all components of response continued to be met. In Study 312, for the 14 patients who achieved complete TMA response status during the initial evaluation period, these responses were sustained through the end of the 26-week initial evaluation period. After achieving complete TMA response, some patients had transient periods during which not all components of response continued to be met.

The complete TMA response components status over time for Study 311 and Study 312 are presented in Figure 11 to Figure 12 and Figure 13 to Figure 14, respectively (Appendix 4).

Hematologic Normalization

In Study 311, in the FAS, hematologic normalization was defined as the normalization of platelets and LDH. At week 26, hematologic normalization was observed in 41 patients of 56 patients in the FAS (73.2%; 95% CI, 60.7% to 85.7%) (Table 13 and Figure 11). As of the data cut-off date, hematologic normalization was observed in 45 patients of the 56 patients in the FAS (80.4%; 95% CI, 69.1% to 91.7%) (Table 13 and Figure 12). In the PP set, hematologic normalization was consistent with the primary analysis (FAS, Table 42).

In Study 312, cohort 1, at week 26, in the FAS, hematologic normalization was observed in 16 patients of the 18 patients (88.9%; 95% CI, 65.3% to 98.6%) (Table 15 and Figure 13). As of the data cut-off date, hematologic normalization was observed in 17 patients of the 18 patients in the FAS and PP sets (94.4%; 95% CI, 72.7% to 99.9%) (Table 15 and Figure 14).

Individual Hematologic Parameters

Study 311

Platelet Count: The mean platelet count improved after the initiation of ravulizumab treatment, increasing from 118.52 × 10⁹/L (SD = 86.440 × 10⁹/L) at baseline to 243.54 × 10⁹/L (SD = 160.500 × 10⁹/L) at day 8. The mean platelet count remained above 227 × 10⁹/L at all subsequent visits in the 26-week period. The mean platelet count was 237.96 × 10⁹/L (SD = 73.528 × 10⁹/L) at day 183 (n = 48) and remained stable, at 205 × 10⁹/L or higher, at all visits during the extension period. The mean platelet count was 241.56 × 10⁹/L (67.523 × 10⁹/L) at day 407 (n = 43).

Lactate Dehydrogenase: The mean LDH value decreased from baseline, with the majority of the decrease occurring during the first month of ravulizumab treatment; this mean reduction in LDH was sustained over a 26-week period. The mean LDH value decreased from 702.38 U/L (SD = 557.959) at baseline to 554.31 U/L (SD = 603.954 U/L) at day 8 and further to 293.27 U/L (SD = 156.999 U/L) at day 29. The mean LDH value remained below 250 U/L at all subsequent visits in the 26-week period. The mean LDH value was 194.46 U/L (SD = 58.099 U/L) at day 183 (n = 48) and remained below 215 U/L at all visits during the extension period. The mean LDH value was 192.86 U/L (SD = 67.536 U/L) at day 407 (n = 42).

Hemoglobin Change From Baseline: The mean hemoglobin value increased more gradually over time during the 26-week period. The mean hemoglobin value increased from 86.26 g/L (SD = 14.866 g/L) at baseline to 91.24 g/L (SD = 15.397 g/L) at day 15 and to 113.82 g/L (SD = 17.086 g/L) at day 57, with mean values remaining above this level at subsequent visits in the initial evaluation period. The mean hemoglobin value was 120.27 g/L (SD = 12.946 g/L) at day 183 (n = 48) and remained above 120 g/L at all visits during the extension period. The mean hemoglobin value was 125.21 g/L (SD = 15.557 g/L) at day 407 (n = 43).

Hemoglobin Response (> 20 g/L): During the initial evaluation period, 40 patients of the 56 patients in the FAS (71.4%; 95% CI, 58.7% to 84.2%) achieved a hemoglobin response (i.e., an increase in hemoglobin of ≥ 20 g/L compared to baseline with a confirmatory result) (Table 16). As of the data cut-off date, an additional 5 patients in the FAS achieved a hemoglobin response, bringing the total to 45 patients of 56 patients achieving a hemoglobin response (80.4%; 95% CI, 69.1% to 91.7%).

Table 16: Hemoglobin Response With a Confirmatory Result as of the Data Cut-Off Date (FAS, Study 311)

CI = confidence interval; EOS = end of study; FAS = full analysis set; NR = not reported.

^a95% CIs for the proportion are based on the asymptotic Gaussian approximation method with a continuity correction.

^bFor hemoglobin increases of greater than or equal to 20 g/L at each visit, 95% CIs for the proportions were based on exact confidence limits using the Clopper-Pearson method.

Sources: Study 311 Clinical Study Report²¹ and Study 312 Clinical Study Report.²²

Study 312, Cohort 1

Overall, patients in the FAS showed improvement in all hematologic TMA parameters (platelets, LDH, and hemoglobin) during the initial evaluation period.

Platelet Count: The mean platelet count increased from baseline early in treatment, and this mean improvement was sustained over the duration of the initial 26-week period. The mean platelet count improved after the initiation of ravulizumab treatment, increasing from 60.39 × 10⁹/L (SD = 32.613 × 10⁹/L) at baseline to 285.40 × 10⁹/L (SD = 147.860 × 10⁹/L) at day 8 and to 273.76 × 10⁹/L (SD = 101.404 × 10⁹/L) at day 22. The mean platelet count remained above 304 × 10⁹/L at all subsequent visits in the initial 26 weeks. The mean platelet count was 304.94 × 10⁹/L (SD = 75.711 × 10⁹/L) at day 183 (n = 17) and remained greater than or equal to 218 × 10⁹/L at all visits through the data cut-off date. The mean platelet count was 289.90 × 10⁹/L (SD = 59.795 × 10⁹/L) at day 575 (n = 10).

Lactate Dehydrogenase: The mean LDH value decreased from baseline, with the majority of the decrease occurring during the first month of ravulizumab treatment; this mean reduction in LDH was sustained over the duration of the initial 26 weeks. The mean LDH value decreased from 2,223.47 U/L (SD = 1,321.118 U/L) at baseline to 1,064.83 U/L (SD = 597.947 U/L) at day 8 and further to 326.94 U/L (SD = 121.606 U/L) at day 29. The mean LDH value remained at 311 U/L or lower at all subsequent visits in the initial 26 weeks. The mean LDH value was 262.41 U/L (SD = 59.995 U/L) at day 183 (n = 17) and remained at 262 U/L or lower at all visits through the data cut-off date. The mean LDH value was 248.18 U/L (SD = 53.822 U/L) at day 575 (n = 11).

Hemoglobin Change From Baseline: The mean hemoglobin value increased gradually over time during the initial 26 weeks. It increased from 74.42 g/L (SD = 17.387 g/L) at baseline to 86.93 g/L (SD = 16.589 g/L) at day 15 and 113.06 g/L (SD = 16.634 g/L) at day 57, with mean values remaining above this level at subsequent visits in the initial 26 weeks. The mean hemoglobin value was 120.06 g/L (SD = 8.011 g/L) at day 183 (n = 17) and remained above 114 g/L at all visits through the data cut-off date. The mean (SD) hemoglobin was 120.30 g/L (SD = 9.787 g/L) at day 575 (n = 10).

Hemoglobin Response (Increase ≥ 20 g/L): During the initial evaluation period, 16 patients of the 18 patients in the FAS (88.9%; 95% CI, 65.3% to 98.6%) (Table 16) had an increase in hemoglobin of greater than or equal to 20 g/L compared to baseline with a confirmatory result (i.e., a hemoglobin response). Of the 17 patients who completed the 26 weeks of ravulizumab treatment, 16 patients had a hemoglobin response as of day 99 (Table 16). As of the data cut-off date, 1 additional patient had achieved a hemoglobin response. At the day 575 visit, 10 patients of 11 patients (90.9%; 95% CI, 58.7% to 99.8%) had achieved a hemoglobin response. Of the 2 patients with data through day 743, each has maintained their hemoglobin response.

Study 312, Cohort 2

Hematologic TMA parameters (platelet count, LDH, and hemoglobin) remained stable for patients in cohort 2 during the initial 26 weeks as well as through the data cut-off date (refer to Figure 14, Figure 15, and Figure 16).

Time to Complete TMA Response

Study 311

As of the data cut-off date, complete TMA response was achieved at a median time of 86 days and occurred as early as 7 days following the first dose of ravulizumab (Figure 4). Four patients achieved a complete TMA response during the extension period. The latest response was observed at 401 days.

Figure 4: Time to Complete TMA Response — Kaplan-Meier Cumulative Distribution Curves as of Data Cut-Off Date (Study 311, FAS)

BL = baseline; CI = confidence interval; FAS = full analysis set; NO. = number; TMA = thrombotic microangiopathy.

Source: Study 311 Clinical Study Report.²¹

Study 312, Cohort 1

The median time to complete TMA response was 30 days and occurred as early as 15 days following the first dose of ravulizumab (Figure 5). Three patients achieved a complete TMA response during the extension period. The latest response was observed at 351 days.

Figure 5: Time to Complete TMA Response — Kaplan-Meier Cumulative Distribution Curves (Study 312, Cohort 1, FAS)

Note: Data as of December 3, 2019. Patients must have met all complete TMA response criteria at 2 separate assessments obtained at least 4 weeks (28 days) apart and at any measurement in between. The time of the event of a confirmed complete TMA response was considered as the first time point at which all the criteria for complete TMA response were met. Patients who did not have a response were censored at the date of the last visit or study discontinuation at the time when the analysis was performed.

Source: Study 312 Clinical Study Report.²²

Presence of Severe Bleeding

Severe bleeding was not reported as an efficacy outcome in either Study 311 or Study 312.

HRQoL (Measured Using the 3-Level EQ-5D)

Study 311: At baseline, for the 53 patients in the FAS for whom data were available, the mean 3-Level EQ-5D score was 0.48 (SD = 0.271). Overall, patients in the FAS showed improvement in their 3-Level EQ-5D scores over time during the initial evaluation period, and this improvement continued into the extension period (Figure 24). At day 183, the 46 patients with available data had a mean change from baseline of 0.22 (SD = 0.247) (Table 17). At day 351, 42 patients with available data had a mean change from baseline of 0.25 (SD = 0.256). Observed and model-based values of changes in 3-Level EQ-5D scores over time (time trade-off value set for the US) for the initial evaluation period and during the extension period through the data cut-off date are presented in Figure 25. No reported MID was found for patients with aHUS.

Study 312: In Study 312, cohort 1 and cohort 2, HRQoL was not assessed.

Table 17: Outcomes in 3-Level EQ-5D (FAS)

NA = not assessed; FAS = full analysis set; SD = standard deviation; TTO = time trade-off.

Sources: Study 311 Clinical Study Report²¹ and Study 312 Clinical Study Report.²²

Symptom Reduction

A by-patient listing of the patient-reported aHUS symptoms and extrarenal signs and symptoms of aHUS is provided in the CSRs for both Study 311 and Study 312. Atypical HUS symptoms were examined using a symptoms questionnaire, and results were reported for the FAS. However, the study level result was not summarized by the sponsor in the CSRs. Therefore, symptom reduction has not been reported herein. Fatigue was assessed using FACIT-F score as an HRQoL outcome (refer to the FACIT-F score assessment section).

Fatigue

Study 311

At baseline, the mean FACIT-F score for the 51 patients in the FAS for whom data were available was 24.03 (SD = 15.279). Overall, the patients in the FAS showed improvement in their FACIT-F scores over time during the initial 26 weeks, and this improvement continued into the extension period (Figure 18). At day 183, the 44 patients for whom data were available had a mean improvement from baseline in FACIT-F score of 19.15 (SD = 16.212) (Table 18). During the extension period, patients with available data had mean improvements from baseline in FACIT-F score of 19.29 at the day 351 visit (n = 40); 16.75 at the day 575 visit (n = 22); and 8.81 at the day 743 visit (n = 8) (Table 18).

An improvement of greater than or equal to 3 points in FACIT-F score, considered to be a clinically meaningful improvement,²³ was observed in 37 patients of the 44 patients (84.1%) with available data at week 26. All had a 3-point improvement from baseline by day 29. During the extension period, 33 patients of the 40 patients (82.5%) with available data had a 3-point improvement from baseline at the day 351 visit (Table 18). Five patients of the 8 patients (62.5%; 95% CI, 24.5% to 91.5%) had an improvement from baseline of 3 points on day 743. The observed values of the changes in FACIT-F scores over time and of FACIT-F scores over time for the FAS are presented in Figure 19.

Study 312, Cohort 1

Among the 13 treated patients who were aged 5 years or older in this study, 9 patients had fatigue assessed using the pediatric FACIT-F scale. At the end of the initial 26 weeks (day 183), these 9 patients had a mean improvement in the pediatric FACIT-F score of 16.78 (SD = 14.704) compared to baseline (Table 18). During the extension period, patients with available data had a mean improvement from baseline in FACIT-F score of 16.67 (SD = 15.297) at the day 351 visit (n = 9) and of 17.40 (SD = 17.184) at the day 575 visit (n = 5). Observed values over time in pediatric FACIT-F scores are shown in Figure 20, with change from baseline presented in Figure 21.

Three patients of 9 patients (33.3%) had at least a 3-point improvement in FACIT-F total score from baseline at day 8; 7 patients (77.8%) had at least a 3-point improvement from baseline at day 29; and all 9 patients had at least a 3-point improvement from baseline by day 71 to day 351. At day 575, 5 of 5 patients had a 3-point improvement from baseline (Table 18).

Study 312, Cohort 2

For the 8 treated patients in cohort 2 who were aged 5 years or older, quality of life was assessed using the pediatric FACIT-F scale. There was no notable improvement or worsening compared to baseline in the pediatric FACIT-F scores for all 8 patients during the initial 26 weeks or through day 351 of the extension period (Table 18, Figure 22). During the extension period, the 8 patients had a mean improvement from baseline in FACIT-F score of –1.25 (SD = 2.712); this was observed at the day 351 visit. Observed values and change from baseline in the pediatric FACIT-F score over time for the FAS are presented in Figure 22 and Figure 23, respectively.

Table 18: Outcomes in Pediatric Quality of Life (FACIT-F, FAS)

FACIT-F = Functional Assessment of Chronic Illness Therapy – Fatigue; NR = not reported; SD = standard deviation.

Sources: Study 311 Clinical Study Report²¹ and Study 312 Clinical Study Report.²²

Renal Function

Estimated Glomerular Filtration Rate

Study 311: The mean eGFR gradually improved during the initial 26 weeks (Figure 26, Figure 27). Overall, the mean eGFR value at baseline was 15.86 mL/min/1.73 m² (Table 19) and increased to 51.83 mL/min/1.73 m² by the end of the initial week 26 (day 183, Table 19). During the extension period, the mean eGFR remained stable above 50.30 mL/min/1.73 m² for the 43 patients who reached the day 407 visit.

Study 312, Cohort 1: The mean eGFR improved gradually during the initial 26 weeks (Figure 28, Figure 29). The mean change in eGFR over time for the FAS is presented Table 19 and in Figure 28 and Figure 29. Overall, the mean eGFR value at baseline was 26.4 mL/min/1.73m² (SD = 21.17 mL/min/1.73m²) (Table 21). The eGFR was 108.5 mL/min/1.73m² (SD = 56.87 mL/min/1.73m²) at the end of the initial 26 weeks (i.e., by day 183). During the extension period, the mean eGFR remained above 100 mL/min/1.73 m² for the 14 patients who reached the day 407 visit.

In Study 312, cohort 2, the eGFR remained generally stable for all 10 patients during the initial 26-week period and through the data cut-off date (Figure 30 and Figure 31).

Table 19: Results for eGFR (FAS)

eGFR = estimated glomerular filtration rate; FAS = full analysis set; NR = not reported; SD = standard deviation.

Sources: Study 311 Clinical Study Report²¹ and Study 312 Clinical Study Report.²²

CKD Stage

Study 311

Most patients were in CKD stage 4 or stage 5 at baseline (Table 20). For the 47 patients with available baseline and week 26 (day 183) data, 32 patients of 47 patients in the FAS (68.1%) had improvement in their CKD stage compared to baseline: 6 patients improved by 5 stages (i.e., from ESKD to normal renal function); 7 patients improved by 4 stages; 5 patients improved by 3 stages; 4 patients improved by 2 stages; and 10 patients improved by 1 stage (Table 20). Two patients experienced worsening of their CKD stages. One of these patients worsened from stage 4 at baseline to stage 5 at day 8, received dialysis on day 16, and remained at stage 5 for the duration of the initial evaluation period. The other patient worsened from stage 4 at baseline to stage 5 at day 8 and remained at stage 5 for the duration of the initial evaluation period (except for 1 assessment of stage 4 at day 15). Nineteen of the 30 patients who had complete TMA responses continued to have improved renal function during the initial evaluation period after achieving complete TMA response, as assessed by an improvement in CKD stage from the time of complete TMA response to day 183 (Table 20). In the extension period, among the 42 patients with available baseline and day 407 data, 29 patients (69.0%) had improvement in CKD stage compared to baseline: 4 patients improved by 5 stages (i.e., from ESKD to normal renal function); 6 patients improved by 4 stages; 8 patients improved by 3 stages; 2 patients improved by 2 stages; and 9 patients improved by 1 stage (Table 21). The 2 patients who experienced worsening of their CKD stage during the initial evaluation period remained at stage 5 from day 183 through the last available visit during the extension period.

Table 20: CKD Stage Shift From Baseline to End of Initial Evaluation Period (Study 311, at Week 26)

CKD = chronic kidney disease.

Note: Baseline data were derived based on the last available eGFR before starting treatment. Patients with both baseline data and at least 1 value at postbaseline visits were included in the summary.

Source: Study 311 Clinical Study Report.²¹

Table 21: CKD Stage Shift From Baseline to End of Initial Evaluation Period (Study 311, on Day 407)

CKD = chronic kidney disease.

Note: Baseline data were derived based on the last available eGFR before starting treatment. Patients with both baseline data and at least 1 value at postbaseline visits were included in the summary.

Source: Study 311 Clinical Study Report.²¹

Study 312, Cohort 1

Of the patients with CKD stage data at both baseline and week 26, the majority (14 patients of 17 patients) evaluated at baseline were at CKD stage 4 or stage 5; 6 patients (35.3%) were at CKD stage 5 (Table 22). With the exception of 2 patients, all of these patients improved their CKD stage (i.e., shifted to a lower stage) from baseline through the end of the initial evaluation period (day 183); the shift was substantial, given that 14 patients improved by 2 or more stages. For the 17 patients with available data at the end of the initial evaluation period, 15 patients (88.2%) had improvement in their CKD stage compared to baseline (Table 22). Three of these patients improved by 5 stages; 7 patients improved by 4 stages; 2 patients improved by 3 stages; 2 patients improved by 2 stages; and 1 patient improved by 1 stage. Of the 2 patients who had no improvement in CKD stage during the initial evaluation period, 1 patient had a history of kidney transplant before the study. None of the patients worsened in CKD stage during the initial evaluation period. All 14 patients with available baseline and day 407 data had improvements in their CKD stage compared to baseline: 3 patients improved by 5 stages (i.e., from ESKD to normal renal function); 5 patients improved by 4 stages; 1 patient improved by 3 stages; 2 patients improved by 2 stages; and 3 patients improved by 1 stage (Table 23).

Table 22: CKD Stage Shift From Baseline to End of Initial Evaluation Period (Study 312, Cohort 1, at Week 26)

CKD = chronic kidney disease.

Note: Baseline data were derived based on the last available eGFR before starting treatment. Patients with both baseline data and at least 1 value at postbaseline visits were included in the summary.

^aThe percentages for the postbaseline CKD stage at week 26 were based on the 17 patients with available data.

Source: Study 312 Clinical Study Report.²²

Table 23: CKD Stage Shift From Baseline to End of Initial Evaluation Period (Study 312, Cohort 1, on Day 407)

CKD = chronic kidney disease.

Note: Baseline data were derived based on the last available eGFR before starting treatment. Patients with both baseline data and at least 1 value at postbaseline visits were included in the summary.

^aThe percentages for the postbaseline CKD stage at week 26 were based on the 17 patients with available data.

Source: Study 312 Clinical Study Report.²²

Study 312, Cohort 2

The majority of patients in cohort 2 (8 patients of 10 patients) were in CKD stage 1 at baseline; 1 patient was in stage 2, and 1 patient was in stage 3a (Table 24). By week 26, 7 of the patients had no change in their CKD stage; 2 patients had worsened by 1 stage; and 1 patient had worsened by 3 stages. During the extension period, all 10 patients had no change in CKD stage by day 351 compared to baseline (Table 25). The CKD stage remained unchanged for the 3 patients with data available at day 407.

Table 24: CKD Stage Shift From Baseline to End of Initial Evaluation Period (Study 312, Cohort 2, at Week 26, FAS)

CKD = chronic kidney disease.

Source: Study 312 Clinical Study Report.²²

Table 25: CKD Stage Shift From Baseline to End of Initial Evaluation Period (Study 312, Cohort 2, on Day 407, FAS)

CKD = chronic kidney disease.

Source: Study 312 Clinical Study Report.²²

Dialysis Requirement Status

Study 311

At baseline, or within 5 days before the first dose of the study drug, 29 patients (51.8%) in the FAS had received renal dialysis (Table 26). During the initial 26 weeks, 17 patients of these 29 patients (58.6%) discontinued dialysis. As of the data cut-off date, 18 patients of these 29 patients (62.1%) had discontinued dialysis (Table 26). One of these patients discontinued within the first week of study treatment, then received dialysis at 3 time points (day 136, day 138, and day 141) during the initial evaluation period. Of the 27 patients who were not on dialysis at baseline, 7 patients (25.9%) initiated dialysis during the initial 26 weeks; 6 of these 7 patients remained on dialysis at day 183. As of the data cut-off date, 20 patients (74.0%) remained off dialysis.

Seven patients (25.9%) initiated dialysis after starting treatment: 3 patients started and stopped dialysis during the initial evaluation period; 3 patients started receiving dialysis during the initial evaluation period; and 1 patient started receiving dialysis during the extension period.

Study 312, Cohort 1

Of the 6 patients in the FAS who were receiving dialysis at baseline (within 5 days before the first dose of the study drug), 4 patients discontinued dialysis within the first 36 days of exposure to ravulizumab (Table 26). All 6 patients had discontinued dialysis by day 193. No patients initiated dialysis after starting treatment with study drug.

Study 312, Cohort 2

As of the data cut-off date, none of the 10 patients had initiated dialysis after starting treatment with the study drug (Table 23).

Table 26: Dialysis Requirement Status (Data Cut-Off Date)

NA = not assessed.

Sources: Study 311 Clinical Study Report²¹ and Study 312 Clinical Study Report.²²

Plasma Therapy–Free Status

Plasma therapy was prohibited during the trial and considered a protocol violation; thus, it was not an outcome assessed in the pivotal studies. However, plasma therapy was reported as a concomitant therapy. In Study 311, 3 patients (5.2%) received plasma therapy. No patient received plasma therapy in Study 312 (cohort 1 or cohort 2) (refer to Table 37).

Packed RBC Transfusions

RBC transfusions were reported as a concomitant treatment (refer to Table 37). In Study 311, 17 patients (29.3%) received packed RBC transfusions during the study. In Study 312 (cohort 1 and cohort 2), no information on packed RBC transfusions was reported (Table 37).

Hospitalizations

Hospitalizations were reported in the assessments of health resource utilization in both Study 311 and Study 312. However, there were no summaries or analyses of hospitalizations at the study level. Therefore, the data are not extractable and not reported in this review report.

Presence of Soluble MAC

Soluble MAC was not assessed in either Study 311 or Study 312 as an outcome.

Harms

Only those harms identified in the review protocol are reported. Refer to Table 27 for detailed harms data.

Adverse Events

In Study 311, as of the data cut-off date, all patients (N = 58, 100%) had experienced at least 1 TEAE. The most common adverse events (occurring in at least 30% of patients) were headache (n = 22; 37.9%), diarrhea (n = 19; 32.8%), and vomiting (n = 18; 31.0%) (refer to Table 27).

In Study 312, cohort 1, as of the data cut-off date, all patients (N = 21, 100%) had experienced at least 1TEAE. The most common adverse events (occurring in at least 30% of patients) were pyrexia (n = 10; 47.6%), headache (n = 7; 33.3%), diarrhea (n = 7; 33.3%), vomiting (n = 7; 33%), and nasopharyngitis (n = 7; 33.3%) (refer to Table 27).

In Study 312, cohort 2, as of the data cut-off date, all 10 patients (100.0%) had experienced at least 1 TEAE. The most common adverse event (occurring in at least 30% of patients) was oropharyngeal pain (n = 3; 30%) (Table 27).

Serious Adverse Events

In Study 311, as of the data cut-off date, a total of 33 patients (56.9%) had experienced an SAE (Table 27). Each SAE was reported in 1 patient, except for pneumonia and hypertension — both of which occurred in 3 patients (5.2%) — and septic shock, urinary tract infection, and malignant hypertension, each of which occurred in 2 patients (3.4%) (Table 27).

In Study 312, cohort 1, as of the data cut-off date, the SAEs occurring in 2 or more patients were gastroenteritis, viral infection and abdominal pain, each of which occurred in 2 patients (9.5%).

In Study 312, cohort 2, as of the data cut-off date, no SAE had been reported in more than 1 patient.

Withdrawals Due to Adverse Events

In Study 311, as of the data cut-off date, a total of 3 patients (5.2%) had experienced adverse events leading to study drug discontinuation (Table 27).

In Study 312, cohort 1, as of the data cut-off date, 1 patient (4.8%) had experienced adverse events leading to study drug discontinuation (Table 27).

In Study 312, cohort 2, as of the data cut-off date, no patient had experienced adverse events leading to study drug discontinuation (Table 27).

Mortality

In Study 311, 4 patients died during the initial 26-week evaluation period. One of these 4 patients died due a pretreatment SAE (cerebral arterial thrombosis); 3 patients (5.2%) died due to treatment-emergent SAEs that were considered not related to the study drug (2 due to septic shock and 1 due to intracranial hemorrhage)²¹ (refer to Table 27).

In cohort 1 and cohort 2 of Study 312, no patients had died due to adverse events as of the data cut-off date (Table 27).

Notable Harms

In Study 311, as of the data cut-off date, no meningococcal infection had been reported. Sepsis was reported in 1 patient (1.7%). Infusion-related reaction was not reported. Hypersensitivity was reported in 1 patient (1.7%) Antidrug antibodies were reported in 1 patient (1.7%) (Table 27).

In Study 312, cohort 1, as of the data cut-off date, 1 patient (4.8%) had reported hypersensitivity (Table 27). In cohort 2, as of the data cut-off date, no notable harms had been reported (Table 27).

Table 27: Summary of Harms (Safety Analysis, Data Cut-Off Date)

AE = adverse event; NR = not reported; SAE = serious adverse event; WDAE = withdrawal due to adverse event.

Note: The organisms involved in the 2 septic shock cases were Corynebacterium and Candida lusitaniae (n = 1) and Pseudomonas (n = 1). The organisms for pneumonia were not reported in the Clinical Study Report.

Sources: Study 311 Clinical Study Report²¹ and Study 312 Clinical Study Report.²²

Critical Appraisal

Internal Validity

The main limitation of the 2 included pivotal studies (Study 311 and Study 312) is their single-arm design, which does not include a comparator arm. Due to the rare and severe nature of aHUS, a randomized control group was not likely to be feasible. Nonetheless, such a design, in addition to the lack of consideration for confounding variables, precludes causal inferences (i.e., the outcomes cannot be directly attributed to ravulizumab). Without an active comparator, standard of care, or statistical hypothesis testing, it is not possible to confirm the relative therapeutic benefit or safety of ravulizumab against other available treatments (such as eculizumab, in this population) or against standard care.

The clinical diagnosis of aHUS is challenging and relies on the exclusion of other conditions.¹¹ Therefore, it is possible that some patients with other conditions that present similarly to aHUS were enrolled.¹² If so, that would mean that not all of the included patients had a confirmed diagnose of aHUS. For example, in Study 311, 8 patients (14.3%) had a kidney transplant not related to aHUS before entering the study. The clinical experts consulted by CADTH for this review indicated that these patients (i.e., those who had a kidney transplant without a prior aHUS diagnosis) could potentially have aHUS, but that the diagnosis of aHUS in these patients could not be absolutely established. This is important because the clinical experts indicated that patients with a confirmed diagnosis of aHUS would be expected to respond better to ravulizumab than those patients without a confirmed diagnosis of aHUS. As a result, any bias associated with an uncertain diagnosis would be against ravulizumab. In other words, it would make ravulizumab appear less effective in terms of improving TMA parameters.

Furthermore, only 30 patients of 56 patients met all the TMA criteria (i.e., active criteria based on platelet, LDH, and serum creatinine levels) on day 1 of the trial. (All had met these criteria during the screening phase.) However, the subgroup analysis based on the TMA criteria on day 1 (yes or no) showed similar results for patients who met all of the TMA criteria on day 1 and those who did not, which minimizes any potential concern for bias being introduced.

In addition, both Study 311 and 312 were open-label trials, so the study investigators and patients were aware of their treatment status. This increases the risk of detection and performance biases that have the potential to influence outcome reporting. However, the primary end point (TMA response) and most of the secondary end points are considered to be objective response measurements for which the potential for bias due to the open-label design is low. The potential for bias is more of a concern for subjective end points, such as safety, symptoms (e.g., measured using the FACIT-F), and HRQoL (measured using the 3-Level EQ-5D). The direction of anticipated bias related to these outcomes is unclear. It is possible that known harms and anticipated benefits would be overreported.

The clinical experts consulted by CADTH for this review indicated that complete TMA response is usually used in clinical research, but not commonly used in clinical practice. An improvement in serum creatinine of greater than or equal to 25% is usually accepted as a component of the complete TMA response. However, for patients on dialysis, discontinuation of dialysis is more meaningful clinically. Within the pivotal trials, sensitivity analysis replacing the improvement in serum creatinine of greater than or equal to 25% with the discontinuation of dialysis showed a consistent and complete TMA response in the primary analysis.

For HRQoL (i.e., the 3-Level EQ-5D) and symptom scales (i.e., the FACIT-F), there is a potential risk of bias because a large number of patients did not have complete measures, especially during the extension period. There may have been differential recall bias, and/or those patients remaining in the study may have differed in some systematic way from those who remained in the study. Overall, the magnitude and direction of the impact of these missing data and recall bias on patient-reported outcomes, 3-Level EQ-5D, and FACIT-F is unknown. No MID was identified for the 3-Level EQ-5D in the aHUS population; therefore, the clinical importance of potential HRQoL improvements is unknown.

One additional potential limitation was that the efficacy assessment was not based on the intention-to-treat population (for Study 311 or Study 312, cohort 1). Instead, it included patients who received at least 1 dose of the study drug and at least 1 postbaseline efficacy assessment. A total of 2 patients (3.4%) in Study 311 and 3 patients (14.29%) in Study 312, cohort 1 were excluded from the primary FAS analysis. It is also noted that 43 patients (76.79%) in Study 311 and 14 patients (66.7%) in Study 312, cohort 1 experienced a major protocol violation; the majority (N = 25, 43.1% in Study 311 and N = 9, 42.9% in Study 312) were related to the eligibility criteria. A PP analysis (N = 44, 75.9% in Study 311 and N = 18, 85.7% for Study 312, cohort 1) was performed and showed results that were consistent with those of the FAS analysis. However, not all patients with a major protocol violation — especially those related to eligibility criteria — were excluded from the PP analysis.

It is worth mentioning that in Study 312, cohort 2, the main limitation was that the sample size (N = 10) was too small for the eculizumab-treated and TMA-stable pediatric patients with aHUS, which meant that the overall dataset was more sensitive to outliers and skewed distribution. However, this limitation is expected due to the rare nature of the disease.

External Validity

Overall, according to the clinical experts consulted by CADTH, the inclusion and exclusion criteria of 2 pivotal studies (Study 311 and 312) were reasonable, and the baseline patient characteristics, concomitant medications, and prohibited medications were reflective of patients they treat in clinical practice for the indication under review. No pediatric patients in Canada were included in Study 312. However, the clinical experts consulted by CADTH for this review indicated that they would not expect to find any important difference among different races or geographic regions in terms of the response to complement inhibitors, such as ravulizumab, for aHUS.

Patients who received plasma exchange and/or plasma infusion for 28 days or longer before the start of screening for the current TMA were excluded from the pivotal studies. Therefore, there is uncertainty as to whether the findings may be generalized to these populations. The clinical experts consulted by CADTH indicated that this would represent a small group of patients with catastrophic disease. This would be an uncommon scenario that is reasonable to exclude. There is no concern about generalizability because of this exclusion criteria.

No subgroup analysis was performed based on baseline platelet or LDH levels. No subgroup analysis based on the duration of prior plasma therapy was conducted, and patients with plasma exchange and/or plasma infusion for 28 days or longer before the start of trial were excluded from this study, as discussed.

Symptom reduction was identified as an important outcome for patients. However, symptom severity reduction at the study level was not assessed as a distinct outcome in the 2 pivotal studies. Instead, a list of patient-reported renal and extrarenal aHUS symptoms was included in the CSRs. It is understood that these symptoms can result in decreased HRQoL. Fatigue is an important symptom that patients with aHUS often experience, and it was assessed using the FACIT-F.

Furthermore, it is understood that aHUS is an extremely rare disease and that, as a result, it was not feasible to enrol large numbers of patients. However, it should be noted that the magnitude of the treatment effect estimates observed in a relatively small study sample may not be replicable in a larger study sample or generalizable to the target population in real-world clinical practice. Finally, given that all results are part of an interim analysis (i.e., at week 26 and during an extension period, with median follow-up times of 75.6 weeks for Study 311, 82.4 weeks for Study 312, cohort 1, and 52.3 weeks for Study 312, cohort 2). Based on the available data, it appears that the effects found at 26 weeks tended to be sustained through later time points.

Indirect Evidence

Objectives and Methods for the Summary of Indirect Evidence

Patients with aHUS in Canada are often treated with eculizumab or supportive care (e.g., plasma exchange or infusion, plasmapheresis). Evidence for ravulizumab is limited to single-arm trials in adult and pediatric populations; therefore, no direct comparisons are available to assess the efficacy and safety of ravulizumab relative to eculizumab. Direct comparisons between ravulizumab and eculizumab are likely to be infeasible due to the rare and severe nature of aHUS. Yet this comparison is important both clinically and economically. As such, an understanding the available ITCs may be useful to clinicians, patients, and pharmacoeconomic modelling groups.

For this submission, a systematic literature review was conducted to identify any sources of ITCs between ravulizumab and eculizumab or between ravulizumab and best supportive care. A single ITC, submitted by the sponsor and also published as a peer-reviewed publication,⁴¹ was identified.

Description of Indirect Treatment Comparison

A single sponsor-submitted study using stabilized inverse probability weighting to compare ravulizumab and eculizumab was reviewed for this submission. Although the submitted study has also been published as a peer-reviewed publication,⁴¹ the evidence for this submission was based on the sponsor-submitted report. Because the sponsor-submitted report contained greater details with respect to study rationale, sensitivity analyses, and methods, it was considered to be of greater use for this review. No discrepancies were noted with respect to the outcomes, methods, or overall interpretation between the sponsor-submitted study and the published article.

Methods of the Sponsor-Submitted Analysis

Objectives

The purpose of the sponsor-submitted analysis was to estimate the comparative efficacy and safety of ravulizumab and eculizumab for the treatment of aHUS.

Study Selection Methods

No systematic literature review was undertaken by the sponsor to identify eligible studies. No formal inclusion or exclusion criteria were applied with respect to the selection of studies for inclusion in the ITC. No details were provided on the data abstraction process, screening process, or quality assessment of included studies. No date was provided to indicate when the trials were assessed.

In total, 5 studies were identified by the sponsor as being appropriate for inclusion in the analysis: 3 studies of eculizumab (Studies aHUS-C08-002, aHUS-C10-003, and aHUS-C10-004) and 2 studies of ravulizumab (Studies 311 and 312). Two studies were excluded from analysis: Study aHUS-C08-003 was excluded on the basis of differences in disease history and plasma exchange status at baseline; no data were provided from Study aHUS-C08-003 to verify this assumption. Data from Study aHUS-C11-003 were excluded because it was a long-term extension study. Patient-level data were available for all patients within the analysis.

Patient populations were split into 4 groups: adults without kidney transplant (primary analysis), adults with kidney transplant, children without kidney transplant, and children with kidney transplant (not analyzed due to small sample size). Patients were considered for the primary analysis if baseline data were available for dialysis status, eGFR, platelet count, and LDH, and were required to have outcome data within 56 days of the 6-month study end point. No imputation was used for missing data; instead, patients with missing data at baseline or end point were excluded.

Analysis Methods

Because no direct evidence was available, the sponsor conducted a patient-level propensity score-adjusted analysis of outcome data, using several approaches (discussed in this section) to account for between-population differences. The sponsor’s primary analysis was performed using a stabilized weights approach, with 4 clinical characteristics that were reported to be chosen on the basis of clinical input: dialysis status at baseline, eGFR at baseline, platelet count at baseline, and LDH at baseline. The sponsor noted that systolic blood pressure was observed to be of importance at baseline, but that it was not included in the statistical model owing to similarities between the 2 eligible patient populations at baseline. The justification of stabilized weights was provided because the sponsor noted that the effective sample size calculation was subject to inflation owing to patients with unexpected propensity score values. Given that LDH levels were noted to remain imbalanced following the application of stabilized weights, the sponsor refactored baseline LDH values into both terciles and halves, identifying that balance was better for LDH refactored into halves.

Separate sensitivity analyses were performed using an inverse probability of treatment weighting approach and a propensity score matching approach. For propensity score matching approaches, 1:1 matching was used. A caliper width of 0.2 times the SD of the propensity score was used in random order, with sensitivity analyses performed using a more restrictive caliper of 0.01.

Stabilized weights were also applied to several sensitivity analyses, with restrictions on populations, outcome definitions, and missing data as follows (with the provided justifications, where available):

• Outcome data to within 28 days (rather than 56 days) of the primary end point

• Cases with complete final outcome data (i.e., for eGFR, LDH, creatinine, and platelet count)

• Patients from South Korea, Taiwan, and Japan excluded

  • It was unclear whether between-country differences in overall health care might influence the comparative efficacy findings, given that these characteristics could not be accounted for in the primary model used.

• Patients who died during the study excluded

  • Patients who died during the study did not have laboratory outcome data at the final end point.

• Only patients aged under 65 years included

  • Patients who were older were considered to have worse prognoses overall.

• Adult patients with kidney transplant

  • These patients were noted by the sponsor to be considered to have substantially different prognoses compared to adult patients without kidney transplant.

• Pediatric patients without prior transplant

  • These patients were noted by the sponsor to be considered to have substantially different prognoses compared to pediatric patients with kidney transplant.

Data were provided on pediatric patients with transplant, but because of the small sample size (refer to the Results of the Sponsor-Submitted Analysis section), this analysis was not provided.

The study sponsor indicated that outcome definitions across the trials were the same except for dialysis at baseline and end point. For the primary analysis, end points were considered eligible if they were recorded within 56 days of 6 months’ follow-up from baseline; and a sensitivity analysis was performed restricting the outcome eligibility window to 28 days. To harmonize definitions within the constraints of patient-follow-up, the sponsor provided the following baseline and end point dialysis definitions for the following treatments and trials:

• For ravulizumab in Study 311 and Study 312 and eculizumab in Studies C10-003 and C10-004, dialysis at baseline was recorded as “yes” for patients who received dialysis within 5 days before study drug initiation.

• For eculizumab in Study C08-002, dialysis at baseline was recorded as “yes” for patients who received dialysis within 7 days before study drug initiation.

• For ravulizumab in Studies 311 and 312, dialysis at the end points was recorded as “yes” for patients who received dialysis within 5 days before their end point measure.

• For eculizumab in Studies C08-002, C10-003, and C10-004, dialysis at the end points was recorded as “yes” for patients who received dialysis within 7 days before their end point measure.

With respect to the statistical testing of differences between populations, Welch’s 2-way t-tests were performed for continuous variables, while chi-square tests were used to obtain P values for categorical variables. Statistical significance was determined to be at a P value of less than 0.1 for between-group differences at baseline and less than 0.05 for end point comparisons.

Table 28: Analysis Methods

CKD = chronic kidney disease; eGFR = estimated glomerular filtration rate; FACIT = Functional Assessment of Chronic Illness Therapy; HRQoL = health-related quality of life; LDH = lactate dehydrogenase; SS = subscale score; TMA = thrombotic microangiopathy; VAS = visual analogue scale.

Source: Sponsor-submitted analysis.⁴²

Results of Sponsor-Submitted Analysis

Summary of Included Studies

The sponsor did not provide details on the characteristics of each individual trial included in the primary or sensitivity analyses. The application of restriction criteria for participants, as detailed in the analysis methods, resulted in a reduction in eligible sample size across all analysis populations studied, as demonstrated in Figure 6.

Figure 6: Patient Flow Chart Demonstrating Attrition Upon Application of ITC Criteria

FAS = full analysis set; ITC = indirect treatment comparison.

Note: Patient numbers for the final analysis represent patients with complete cases for propensity score variables, no more than 1 missing laboratory measure, and outcome data within 56 days of the 6-month end point. Children with kidney transplant were not included in the analysis due to low patient numbers.

Source: Sponsor-submitted analysis.⁴²

As depicted in Figure 6, across the 3 major subpopulations of interest, there were variances with respect to the eligible and eventual analysis population sizes. For adult patients without kidney transplant (primary analysis), the ravulizumab primary analysis population represented 92% of the eligible patients, and the eculizumab population represented 95% of the eligible patients. For adult patients with kidney transplant, the ravulizumab population represented 87.5% of the eligible patients, and the eculizumab population represented 93.8% of the eligible patients. For pediatric patients without kidney transplant, the ravulizumab population represented 60% of the eligible patients, and the eculizumab population represented 95% of the eligible patient population.

For the primary analyses, Study 311 represents the entire analysis set for ravulizumab, and the unweighted patient demographics represent this trial alone among those eligible for analysis. The eculizumab data on adult populations without kidney transplant are informed by 2 trials, Study CO8-002 (21% of eculizumab patients in this cohort) and Study C10-004 (79% of eculizumab patients in this cohort). Data on the unadjusted demographic differences across these 2 populations are provided in Table 29. Data for eculizumab are merged across the 2 trials for which eculizumab data were available.

Overall, the 2 unweighted populations were comparable for most baseline covariates measured and assessed, with exceptions for: the proportion of patients from Japan, South Korea, and Taiwan (eculizumab = 0%, ravulizumab = 20%; 95% CI, 8 to 31); mean age (ravulizumab = 40 years, eculizumab = 35 years; 95% CI, 0 to 12); and LDH (eculizumab = 484 [SD = 518], ravulizumab = 714 [SD = 586]; 95% CI = –7 to 469).

Table 29: Patient Baseline Demographics, Unweighted Sample, Adults Without Kidney Transplant