CADTH Reimbursement Review

Difelikefalin (Korsuva)

Sponsor: Otsuka Canada Pharmaceutical Inc.

Therapeutic area: Chronic kidney disease

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

NOC = Notice of Compliance.

Source: Difelikefalin Product Monograph.¹

Introduction

Chronic kidney disease (CKD) is a progressive disease characterized by the gradual loss of renal function and/or abnormalities of renal structure over 3 months. CKD constitutes a major health burden worldwide and is associated with high morbidity and mortality.^2,3 CKD-associated pruritus (CKD-aP), also known as uremic pruritus, is a common, distressing, and underrecognized systemic itch CKD comorbidity that affects more than 60% of patients undergoing hemodialysis, with 20% to 40% of patients reporting moderate-to-severe pruritus.^4-9 Intense and generalized systemic itching in these patients is associated with poor sleep quality, depression, reduced quality of life, increased risk of infection, and an increased risk of death.^5,6,10-12 In Canada, the estimated overall prevalence of CKD-aP in adults undergoing hemodialysis is about 70%, according to the international, observational Dialysis Outcomes and Practice Patterns Study.⁸ More than one-third of patients with CKD who are undergoing hemodialysis experience moderate-to-severe itch.^8,13,14 There is currently no standardized lab or diagnostic testing for the diagnosis of CKD-aP. A complaint of pruritus by someone with a diagnosis of CKD is presumed to be CKD-aP unless assessment determines a different diagnosis.⁶

There is no approved therapy for CKD-aP in Canada. The standard of care for moderate-to-severe pruritus associated with CKD includes the use of topical moisturizing treatments, steroids in combination with menthol and camphor, calcineurin inhibitors, and the systemic use of gabapentin or pregabalin, naltrexone, thalidomide, biologics (i.e., dupilumab or tralokinumab), kappa opioid receptor agonists (i.e., difelikefalin), and topical phototherapy to reduce itch intensity and improve sleep quality, per feedback provided by the clinical experts consulted by CADTH.

Difelikefalin is a selective kappa opioid receptor agonist that acts in the peripheral nervous system. The drug exerts antipruritic effects by means of activation of kappa opioid receptors on peripheral neurons and immune cells. The selective activity of difelikefalin at kappa opioid receptors mostly avoids mu opioid–associated side effects, such as respiratory depression, dependence, and euphoria. Difelikefalin is anticipated to have no meaningful abuse and dependence potential.¹

The objective of this report was to perform a systematic review of the beneficial and harmful effects of difelikefalin 0.5 mcg/kg IV 3 times per week for the treatment of moderate-to-severe pruritus associated with CKD in adults undergoing hemodialysis.

Stakeholder Perspectives

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

Patient Input

One patient advocacy group, The Kidney Foundation of Canada, provided input for the treatment of adults with CKD who are undergoing hemodialysis. Patient input was gathered from independent surveys of people living with CKD and their caregivers across Canada in September 2022. In total, 19 responses were gathered from the survey (10 fully completed and 9 partially completed).

More than 90% of patients who responded to the survey reported experiencing itchy skin as part of their kidney disease, with 50% of respondents experiencing itchiness every day, 40% experiencing itchiness several times per week, and 10% experiencing itchiness occasionally. Although 60% of respondents reported living with pruritus for 1 to 2 years, 20% reported living with it for 2 to 5 years and 20% reported living with it for more than 5 years. The itchiness was described as moderate to severe by 80% of respondents. When describing disease experience, several respondents reported developing scabs and/or sores because of their itchy skin. Many respondents also reported having trouble sleeping as a result of itchiness. One-third (33%) of patients who responded to this survey reported taking medication to treat their itchiness associated with kidney disease. Although 33% of respondents reported that their treatments were covered by their provincial drug plan, 67% reported paying out of pocket. Most respondents to this survey expressed satisfaction with their current medication and/or combination of treatments, whereas 33% were neither satisfied nor unsatisfied. On the other hand, more than 66% of respondents expressed uncertainty regarding the improvement of their skin appearance in response to currently available treatment.

When describing their expectations for CKD therapies in general, patients who responded to the survey mentioned improvement in their well-being or quality of life (QoL), and 90% mentioned hoping for increased energy. Other important considerations mentioned included fewer hospital visits, less medication overall, fewer side effects, and efficacy. None of the respondents had experience with the drug under review.

Clinician Input

Input From Clinical Experts Consulted by CADTH

According to the clinical experts consulted by CADTH, treatment for moderate-to-severe pruritus associated with CKD includes the use of topical moisturizing treatments, steroids in combination with menthol and camphor, calcineurin inhibitors, and the systemic use of gabapentin or pregabalin, naltrexone, thalidomide, biologics (i.e., dupilumab or tralokinumab), kappa opioid receptor agonists (i.e., difelikefalin), and topical phototherapy to reduce itch intensity and improve sleep quality. The clinical experts stated that the goal of treatment is to reduce of itch intensity, improve QoL and sleep quality, decrease itch-related depression, and, in some cases, suicidality. They also stated that currently used off-label treatments do not adequately address these issues in all patients.

The clinical experts indicated that difelikefalin is a selective kappa opioid receptor agonist that acts in the peripheral nervous system, and thus the drug has less neuromodulation effects than a nonselective kappa opioid receptor agonist. Difelikefalin is the first treatment approved in Canada for moderate-to-severe pruritus in patients undergoing hemodialysis, although the clinical experts consulted by CADTH were uncertain about whether the drug would address the underlying disease process that causes pruritus, as the etiology of CKD-aP is not yet fully understood. The clinical experts indicated that difelikefalin would be used in combination with other therapies in a later-line setting in which existing therapies are intolerable, have failed to control symptoms, or are contraindicated. The clinical experts stated that the prior lines of therapy include a menthol-containing topical steroid and/or a topical calcineurin inhibitor, a systemic drug (gabapentin or naltrexone), and phototherapy. The clinical experts did not expect that difelikefalin would replace any treatments or cause a shift in the current treatment paradigm but, instead, would be used after other more accessible topical and systemic treatment options have failed.

The clinical experts indicated that only patients with end-stage renal disease (ESRD) on hemodialysis with moderate-to-severe pruritus who are not responsive to existing therapies would be candidates for treatment with difelikefalin. According to the clinical experts, the drug is not suitable for patients on peritoneal dialysis, patients with nondialysis CKD, or patients who have received a kidney transplant.

The clinical experts stated that clinical evaluations, such as the itch numeric rating scale, can help identify pruritus severity at baseline and assess response to treatment, but they are not typically used in clinical practice. The clinical experts indicated that it is important to consider other causes of pruritus in patients, such as dermatologic conditions, CKD mineral bone disease, hepatobiliary disorders, diabetes, and hematologic conditions, as patients with these conditions may not respond to difelikefalin. Further, the clinical experts mentioned that it is possible that misdiagnosis of the cause of pruritus may occur in clinical practice due to the various potential causes of pruritus, as well as the lack of unique laboratory findings associated with CKD-aP.

The clinical experts consulted by CADTH indicated that in clinical practice, subjective patient-reported improvement in symptoms is the primary outcome used to determine whether a patient is responding to treatment. The clinical experts highlighted that reductions in the frequency and/or severity of symptoms would be considered when evaluating response to treatment, along with other improvements in sleep quality, depression, adherence to dialysis, depression and suicidality, and overall QoL. The clinical experts consulted by CADTH stated that a patient may discontinue treatment if there is a lack of response, and that this could be assessed at 12 weeks. The development of significant and persistent side effects, such as diarrhea, dizziness, and recalcitrant nausea and/or vomiting, may also be cause for discontinuing treatment.

According to the clinical experts consulted by CADTH, difelikefalin would most often be prescribed in dialysis units or clinics setting by nephrologists or other physicians.

Clinician Group Input

Clinician group input on the review of difelikefalin for the treatment of moderate-to-severe pruritus associated with CKD in adults on hemodialysis was received from 3 clinician groups: Saskatchewan Kidney Doctors; Hemodialysis Specialty Physician Group, Division of Nephrology, The Ottawa Hospital; and Division of Nephrology, Department of Medicine, Dalhousie University and Nova Scotia Health.

The clinician groups agreed that currently only off-label medications in Canada are available for the treatment of pruritus in patients with kidney disease. There have been some unmet needs as the current treatment options are unsatisfactory and not effective in reducing the symptom burden. The debilitating symptoms in most severe cases may lead to a deterioration in QoL. The clinician groups mentioned the need for a new treatment option that could help relieve symptoms with better tolerability, affordability, and ease of administration. One clinician group in particular indicated that the most important treatment goal would be to have a therapy that would reduce or maintain the severity of itch below the threshold of clinical importance, which is known to be the level above which itch negatively impacts QoL and outcomes important to patients.

The clinician groups mentioned that difelikefalin may cause a paradigm shift in the treatment of itching, with data from randomized controlled trials (RCTs) and a pooled analysis demonstrating objective effectiveness in reducing the severity of the symptoms of CKD-associated (uremic) pruritus. Input from clinician groups provided different opinions about the place in therapy for difelikefalin. Some groups suggested that it would likely be used as first-line therapy, whereas others recommended it as an add-on or second-line treatment. Given the route of administration of difelikefalin, the clinician groups pointed out that patients on hemodialysis with moderate-to-severe pruritus would benefit the most from this treatment. The clinician groups mentioned the possibility of underreporting or underdiagnosing CKD-aP, which was also pointed out by the clinical experts consulted by CADTH. Regarding the diagnosis of this indication, the clinician groups noted the lack of diagnostic tests. One group mentioned using clinical history and exclusion criteria to identify patients, whereas other groups pointed out that the identification of patients and severity of symptoms can be done through self-administered questionnaires and screening tools (e.g., the 5-D itch scale, the Uremic Pruritus in Dialysis Patient [UP-Dial] questionnaire, and the Skindex-10 scale). The clinician groups stated that the reduction in itch severity measured on a numeric rating scale is used to evaluate symptoms. According to the clinician groups, lack of a clinically meaningful response and intolerable side effects should be considered discontinuation criteria. The clinician groups added that a meaningful response to treatment for this disease would be symptom reduction leading to better sleep, improved QoL, improved mood, and return to activities of daily living.

Drug Program Input

Input was obtained from the drug programs that participate in the CADTH reimbursement review process. The following were identified as key factors that could potentially impact the implementation of a CADTH recommendation for difelikefalin:

• considerations for relevant comparator

• considerations for initiation of therapy

• considerations for prescribing of therapy

• considerations for care provision issues

• considerations for system and economic issues.

Clinical Evidence

Pivotal Studies and Protocol-Selected Studies

Description of Studies

The KALM-1 and KALM-2 studies met the inclusion criteria for the CADTH systematic review. Both KALM trials were designed as multicentre, randomized, double-blind, phase III clinical trials to compare difelikefalin with placebo in patients being treated with hemodialysis who were experiencing moderate-to-severe pruritus. The primary objective of both KALM trials was to evaluate the efficacy of difelikefalin at a dose of 0.5 mcg/kg, compared with placebo, in reducing the intensity of itch in patients undergoing hemodialysis and experiencing moderate-to-severe pruritus. The shared key secondary objectives of the KALM-1 and KALM-2 trials were to evaluate the efficacy of difelikefalin at a dose of 0.5 mcg/kg, compared with placebo, in improving itch-related QoL and safety in patients undergoing hemodialysis and experiencing moderate-to-severe pruritus. A total of 378 patients in the KALM-1 trial and 473 patients in the KALM-2 trial were randomized, in a 1:1 ratio, to difelikefalin or placebo. The KALM-1 trial was limited to study centres in the US, and the KALM-2 trial was conducted globally, and included 5 sites in Canada. For both KALM trials, the randomization was stratified based on the use of concomitant medications to treat itch during the week before randomization (yes or no), specific medical conditions (such as a history of falls or fracture or fall-related fracture, confusional state or mental status change or altered mental status or disorientation, or gait disturbance or movement disorder [presence or absence]).

The primary efficacy end point in the KALM-1 and KALM-2 trials was the percentage of patients at week 12 who achieved an improvement of at least 3 points from baseline in the weekly mean score on the daily Worst Itch Intensity Numerical Rating Scale (WI-NRS). Key secondary efficacy end points for both trials were mean change from baseline at week 12 in itch-related QoL (as measured by the 5-D itch scale total score), mean change from baseline at week 12 in itch-related QoL (as measured by the Skindex-10 scale total score), and the percentage of patients at week 12 achieving an improvement of at least 4 points from baseline in the weekly mean score of the daily WI-NRS. In addition, the proportions of patients who achieved an improvement of at least 3 points or at least 4 points from baseline at week 4 and week 8 were considered as secondary end points in the KALM-2 trial.

The baseline patient characteristics were roughly balanced between treatment groups. The median age of all randomized patients was similar in the 2 studies, at 58.0 years (range, 22 to 88 years) in the KALM-1 trial and 60.0 years (range, 23 to 87 years) in the KALM-2 trial. Most patients in the KALM-1 and KALM-2 studies were male (61.0% and 58.2%, respectively). The predominant races in the KALM-1 and KALM-2 studies were white (48.8% and 70.3%, respectively) and Black or African American (41.6% and 19.3%, respectively). The median prescription dry body weight was 84.0 kg (range, 42.0 to 135.0 kg) in the KALM-1 trial and 78.0 kg (range, 42.0 to 135.0 kg) in the KALM-2 trial. The median baseline WI-NRS score was 7.14 (range, 4.1 to 10.0) in the KALM-1 trial and 7.13 (range, 4.5 to 10.0) in the KALM-2 trial. At baseline, 39.8% of patients in the KALM-1 trial and 36.5% in the KALM-2 trial were using antiitch medications, and 14.1% and 16.6%, respectively, reported at least 1 of the specified medical conditions (i.e., history of falls or fracture or fall-related fracture, confusional state or mental status change or altered mental status or disorientation, or gait disturbance or movement disorder). The median duration of CKD-aP for all patients was 2.5 years (range, 0.1 to 26.5 years) in the KALM-1 trial and 2.3 years (range, 0.0 to 58.4 years) in the KALM-2 trial. The median time interval since the diagnosis of CKD and of ESRD for all patients was 5.45 years (range, 0.3 to 42.9 years) and 3.92 years (range, 0.3 to 28.7 years), respectively, in the KALM-1 trial, and 7.53 years (range, 0.3 to 48.3 years) and 4.03 years (range, 0.3 to 30.2 years), respectively, in the KALM-2 trial.

Efficacy Results

A summary of key efficacy results is provided in Table 2. No evidence was identified in the KALM trials for mood, days of missed work, days of missed school, and days of missed dialysis, which were identified in the CADTH systematic review protocol and considered to be important outcomes of interest by patients and clinicians.

Pruritus Severity

Worst Itching Intensity Numerical Rating Scale

The severity of pruritus was assessed in both the KALM-1 and KALM-2 trials using the WI-NRS. The primary and key secondary end points in both trials were the proportion of patients with at least a 3-point improvement in the WI-NRS score from baseline to week 12 and the proportion of patients with at least a 4-point improvement in the WI-NRS score from baseline to week 12, respectively.

At week 12 in the KALM-1 trial, 52% and 31% of patients randomized to difelikefalin and placebo, respectively, had a WI-NRS score that improved by at least 3 points from baseline to week 12. This corresponded to an odds ratio of 2.72 (95% confidence interval [CI], 1.72 to 4.30; P < 0.001) in favour of difelikefalin. In the KALM-2 trial, 50% and 37% of patients randomized to difelikefalin and placebo, respectively, had a WI-NRS score that improved by at least 3 points from baseline to week 12. This corresponded to an odds ratio of 1.61 (95% CI, 1.08 to 2.41; P = 0.02) in favour of difelikefalin. A 3-point improvement on the WI-NRS has been validated as an appropriate threshold specifically for patients with CKD-aP in an assessment of the psychometric properties published by the sponsor that was based on data from a phase II study (CR845-CLIN2101, NCT02858726) and the KALM-1 and KALM-2 trials.¹⁵ However, the FDA report states that, “in several communications to the sponsor the Agency recommended the primary efficacy end point to be the proportion of patients achieving at least a 4-point improvement in WI-NRS score from baseline to week 12.”¹⁶ The clinical experts consulted by CADTH were aligned with the FDA guidance regarding a 4-point improvement, but input from clinician groups suggested a preference for a 3-point improvement. In summary, there is uncertainty regarding the most appropriate minimally important difference (MID) for the WI-NRS score; therefore, results for an assessment of both outcomes have been presented. With regard to the proportion of patients achieving at least a 4-point improvement from baseline to week 12 with respect to the weekly mean of the daily 24-hour WI-NRS, 41% and 21% of patients randomized to difelikefalin and placebo, respectively, in the KALM-1 trial, and 38% and 25% of patients randomized to difelikefalin and placebo respectively, in the KALM-2 trial had a WI-NRS score that improved by at least 4 points from baseline to week 12. This corresponded to an odds ratio of 2.89 (95% CI, 1.75 to 4.76; P < 0.001) in the KALM-1 trial and of 1.77 (95% CI, 1.14 to 2.74; P = 0.01) in the KALM-2 trial in favour of difelikefalin. Overall, difelikefalin demonstrated an improvement in the WI-NRS that was clinically meaningful, based on both of the reported assessments of 3- and 4-point improvements in the WI-NRS.

The supportive analyses performed for the primary end point in the KALM-1 and KALM-2 trials was conducted using the per-protocol (PP) population. The results were consistent with the primary analysis. Similarly, the sensitivity analyses for the primary end point conducted using the intention-to-treat (ITT) population showed efficacy with difelikefalin, and the results were consistent with the efficacy shown in the primary efficacy analysis.

The proportion of patients with at least a 3-point or 4-point improvement from baseline in the WI-NRS was also reported at week 4 and week 8 in both trials. In the KALM-1 trial, the proportion of patients with at least a 3-point or at least a 4-point improvement from baseline in the WI-NRS at week 8 and week 4 were considered exploratory, but in the KALM-2 trial, they were included as key secondary end points. In the KALM-2 trial, a benefit from treatment with difelikefalin, compared to placebo, was also observed at week 4 (35% versus 22%) and week 8 (45% versus 33%) for the proportion of patients with at least a 3-point improvement from baseline in the WI-NRS. Similar results were observed in the KALM-1 trial for difelikefalin versus placebo at week 4 (33% versus 18%) and at week 8 (43% versus 28%). Similarly, with regard to the proportion of patients with at least a 4-point improvement from baseline in the KALM-2 trial, a benefit from treatment with difelikefalin, compared to placebo, was observed at week 4 (20% versus 13%) and at week 8 (31% versus 20%). Similar results were observed in the KALM-1 trial at week 4 (18% versus 8%) and at week 8 (31% versus 20%). In the KALM trials, data for the primary and secondary or other outcomes were reported up to 12 weeks. According to the clinical experts consulted by CADTH, patients with CKD-aP would receive difelikefalin beyond 12 weeks for the desired treatment effect. A pooled analysis of the 2 open-label extension (OLE) studies —KALM-1 and KALM-2 — assessed the 5-D itch scale for 52 weeks after the 12-week pivotal trials, and reported that the proportion of patients achieving at least a 5-point improvement (reduction) was maintained up to 64 weeks.¹⁷ Due to the amount of missing data, the lack of reporting for other outcome measures (i.e., WI-NRS and Skindex-10), and the absence of a control group after week 12, it is difficult to draw conclusions about whether the efficacy of difelikefalin beyond 12 weeks would be consistent with the efficacy at week 12.

Patient Global Impression of Change

At week 12 in the KALM-1 trial, ||| and ||| of patients randomized to difelikefalin and placebo, respectively, were complete responders. This corresponded to an odds ratio of |||| |||| ||| |||| || ||||| | | |||||). In the KALM-2 trial, ||| and ||| of patients randomized to difelikefalin and placebo, respectively, were complete responders. This corresponded to an odds ratio of |||| |||| ||| |||| || ||||| | | ||||||. Of note, the P values were not adjusted for multiple testing.

Health-Related Quality of Life

Health-related quality of life (HRQoL) was assessed in both the KALM-1 and KALM-2 trials using the Skindex-10 scale score and 5-D itch scale score. The change from baseline to week 12 in the Skindex-10 scale total score and the 5-D Itch scale score were included as key secondary end points in both of the trials.

Skindex-10 Scale Total Score

Skindex-10 is a modified scale that contains 10 questions to evaluate CKD-aP and QoL and relevant subdomains for patients in the hemodialysis setting, with higher scores indicating more bothered and lower scores indicating less bothered.^4,5 At the end of week 12, the least squares (LS) mean change (reduction) in the Skindex-10 scale total score was –17.2 versus –12.0 for difelikefalin versus placebo in the KALM-1 trial, –16.6 versus –14.8, respectively, in the KALM-2 trial. In the KALM-1 trial, the difference in LS mean change (reduction) from baseline to week 12 in the Skindex-10 scale total score between the difelikefalin and placebo groups was –5.1 points (95% CI, –8.0 to –2.3 points, P < 0.001). In the KALM-2 trial, the results for the change from baseline to week 12 in the Skindex-10 scale total score indicated no difference between treatment groups, with a LS mean difference of –1.8 points (95% CI, –4.3 to 0.8 points; P = 0.171). A 3-point to 12-point change on the Skindex-10 scale was identified as a clinically meaningful change in patients with moderate-to-severe CKD-aP on hemodialysis.⁵ Therefore, the reported within-group differences in Skindex-10 scores were clinically meaningful, as they met the MID identified by the literature search conducted by CADTH. With regard to Skindex-10 domain scores, the results were in favour of difelikefalin; however, there was no adjustment for multiplicity, so definitive conclusions could not be drawn with respect to the individual domains of the Skindex-10 scale: disease total, mood/emotional distress total, and social functioning total. No long-term data were reported in the KALM trials for the Skindex-10 scale.

5-D Itch Scale Total Score

The 5-D itch scale is a questionnaire that assesses CKD-aP and QoL and has relevant subdomains for patients in the hemodialysis setting, with higher scores indicating more bothered and lower scores indicating less bothered.¹⁸ At the end of week 12, the LS mean change in the 5-D itch scale total score was greater in the difelikefalin group than in the placebo group in the KALM-1 trial (–5.0 versus –3.7) and the KALM-2 trial (–4.9 versus –3.8), indicating that patients in the difelikefalin group experienced greater improvement (reduction) than those in the placebo group. Overall, the difference in LS mean change from baseline to week 12 in the 5-D itch scale total score between the difelikefalin and placebo groups was –1.3 points (95% CI, –2.0 to –0.5 points, P < 0.001) in the KALM-1 trial and –1.1 points (95% CI, –1.7 to –0.4 points; P = 0.002) in the KALM-2 trial. Of note, the analysis of the difference in LS means for the 5-D itch score at week 12 in the KALM-2 trial was at risk for a type I error, as the difference in LS means for Skindex-10 at week 12, which came first in the testing hierarchy, was not statistically significant. In addition, no published MID was identified for 5-D itch scores in patients with CKD-aP. Therefore, it is unclear whether the reported within-group differences are clinically meaningful. No long-term data were reported in the KALM trials for the 5-D itch scale.

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

CI = confidence interval; ITT = intention to treat; LS = least squares; WI-NRS = Worst Itching Intensity Numerical Rating Scale.

Note: Combined analysis for the primary efficacy end point used the separate interim and postinterim results to generate an adjusted overall estimate and P value using the Lawrence and Hung and/or Cui, Hung. and Wang (CHW) methodology.

^aCounts and percentages are based on nonmissing data.

^bEstimated percent, odds ratio, and P value used a logistic regression model with terms for treatment group, baseline WI-NRS score, region (for the KALM-2 trial only), use of antiitch medication during the week before randomization, and the presence of specific medical conditions. Missing values were imputed using multiple imputation under the missing-at-random missing data assumption for interim patients and postinterim patients separately.

^cEstimated using the Lawrence and Hung approach.

^dEstimated using the CHW approach.

^ePatient Global Impression of Change responders were those with responses of very much improved or much improved.

^eThe P value is calculated with the Cochran-Mantel-Haenszel exact test.

^gThe P value was not adjusted for multiple testing (i.e., the type I error rate has not been controlled).

^hUsed the placebo group as reference.

ⁱLS means and 95% CIs were based on an ANCOVA, with fixed effects for treatment, with baseline score, region (for the KALM-2 trial only), and the randomization stratification variables as covariates for the KALM-2 trial. Missing values were imputed using multiple imputation under the missing-at-random missing data assumption.

^jThe P value was not considered inferential and the null hypotheses was not rejected, according to the gate-keeping strategy, based on the hierarchical testing order of key secondary end points, as the test of the prior end point (change from baseline in Skindex-10 total score) in the sequence was not statistically significant.

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report;²⁰ FDA report.¹⁶

Harms Results

Selected harms outcomes from the KALM-1 and KALM-2 trials are summarized in Table 3.

The percentage of patients with any reported treatment-emergent adverse events (TEAE) was comparable in the difelikefalin and placebo groups in the KALM-1 trial (68.8% versus 62.2%) and the KALM-2 trial (68.1% versus 61.4%). In the KALM-1 trial, the most common TEAEs reported for patients randomized to difelikefalin and placebo, respectively, were diarrhea (9.5% versus 3.7%), dizziness (6.9% versus 1.1%), vomiting (5.3% versus 3.2%), and nasopharyngitis (3.2% versus 5.3%). Diarrhea, dizziness, and vomiting were reported more commonly in the difelikefalin treatment group than the placebo treatment group in the KALM-1 trial. In the KALM-2 trial, the most common TEAEs reported for patients randomized to difelikefalin and placebo, respectively, were diarrhea (8.1% versus 5.5%), vomiting (6.4% versus 5.9%), falls (6.8% versus 5.1%), dizziness (5.5% versus 5.1%), and nausea (6.4% versus 4.2%), all of which were reported more frequently in the difelikefalin treatment group than in the placebo group. Serious adverse events (SAEs) were reported in 25.9% of patients in the difelikefalin group and 21.8% of patients in the placebo group in the KALM-1 trial, and in 24.7% and 21.6%, respectively, in the KALM-2 trial. The clinical experts stated that the proportion of patients reporting an SAE was consistent with what would be expected, given the characteristics of the patients enrolled in these studies. Specific SAEs reported in the trials were hyperkalemia, sepsis, pneumonia, fluid overload, and chest pain, all of which were infrequently reported, occurring in less than 4% of patients in any treatment group.

The proportion of patients who discontinued treatment due to TEAEs was 7.9% for difelikefalin and 4.8% for placebo in the KALM-1 trial and 5.5% for difelikefalin and 3.4% for placebo in the KALM-2 trial. Dizziness was the most frequently reported TEAE that caused discontinuation for both KALM trials. Deaths were reported in 1.1% of patients in the difelikefalin group and 1.6% of patients in the placebo group in the KALM-1 trial, and in 0.9% for difelikefalin and 0.8% for placebo in the KALM-2 trial. In the KALM-1 trial, sepsis was the cause of death for 2 patients randomized to difelikefalin (but for 0 patients randomized to placebo and 0 patients in the KALM-2 trial), and septic shock was the cause of death for 2 patients randomized to placebo (but for 0 patients randomized to difelikefalin and 0 patients in the KALM-2 trial). All other reported causes of death (dyspnea and/or hypotension, cardiac arrest, unknown) were infrequently reported, with no more than 1 patient in any treatment group. No specific adverse event (AE) was identified to account for the majority of deaths in either group in the KALM-2 trial.

The following harms of particular interest were included in the CADTH systematic review protocol: diarrhea, nausea, vomiting, gait disturbance, falls, dizziness, headache, somnolence, seizures, syncope, mental status changes, mood changes, paresthesia (unusual feeling or sensation), hyperkalemia, back pain, tachycardia, and palpitation. The most common notable harms were diarrhea, dizziness, and vomiting in both KALM trials, as well as falls and nausea in the KALM-2 trial. Overall, during the 12-week treatment period of the KALM-1 and KALM-2 trials, patients who received difelikefalin reported notable harms at a similar or slightly higher frequency than patients who received placebo. There were imbalances in the proportion of patients (difelikefalin versus placebo) reporting diarrhea (9.5% versus 3.7%) and dizziness (6.9% versus 1.1%) as an AE in the KALM-1 trial.

Table 3: Summary of Key Safety Results From Pivotal and Protocol-Selected Studies

TEAE = treatment-emergent adverse event.

^aTEAEs relative to the double-blind treatment period are identified as any AE with an onset date after the first dose of the study drug up to the study end of treatment or early termination visit, the start of the discontinuation period, or 10 days after the last dose if no end of treatment or early termination visit was conducted, whichever is later.

^bReported based on TEAEs of special interest identified by the sponsor.

^cTachycardia included the following preferred terms: tachycardia, sinus tachycardia, and tachyarrhythmia.

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report.²⁰

Critical Appraisal

The overall study design of the KALM-1 and KALM-2 trials was appropriate for the objectives of the study. There was no particular concern with the methods of randomization or allocation concealment. According to the clinical experts consulted by CADTH, the frequency of hemodialysis (i.e., optimization of hemodialysis) is a potential effect modifier and prognostic factor that was not considered in either KALM trial. Patients with more frequent hemodialysis visits would have better control of the disease and more exposure to difelikefalin than patients who had less frequent hemodialysis visits. In addition, more frequent hemodialysis visits indicated better compliance with difelikefalin treatment. Therefore, the frequency of hemodialysis may have a potential impact on the validity of the study results for the treatment effect of difelikefalin; however, the magnitude of the impact is unknown, as there were no data with regard to the treatment effect in patients with different frequencies of hemodialysis in either KALM trial. The 3-point reduction in WI-NRS scores was adopted in both KALM studies as the cut-off point to define improvement and response of treatment in pruritus intensity based on the results of a phase II study of difelikefalin (CR845-CLIN2101, NCT02858726) and the 2 phase III trials (KALM-1 and KALM-2) conducted by the sponsor.¹⁵ The FDA recommended an improvement of at least 4 points as the cut-off for the primary efficacy end point (i.e., proportion of patients with a ≥ 4-point improvement in WI-NRS at week 12).¹⁶ Overall, although odds ratios at week 12 were similar with the 3-point and 4-point cut-offs, the proportion of patients in each treatment group that met the threshold was higher with the 3-point threshold, which was used for the primary end point. It is worth mentioning that odds ratios were used throughout the KALM trials for the primary and key secondary end points, and that odds ratios tend to give an inflated impression of the treatment effects, compared with relative risks.²¹ Therefore, the results of the treatment effect of difelikefalin compared to placebo estimated using odds ratios should be interpreted with caution. Although there was a large proportion of patients with at least 1 major protocol deviation (around 30%) in both studies, the sensitivity and supplemental analyses were consistent with the primary estimand. In addition, a notable response was observed in the placebo treatment groups in both KALM trials, albeit not as great as in the difelikefalin treatment groups. According to the clinical experts consulted by CADTH, the placebo response may be due to the optimized hemodialysis treatment associated with the trials, as patients enrolled in trials are more likely to attend hemodialysis visits more frequently and regularly than those in real-world settings, which could have provided a benefit to these patients. In addition, the CADTH review team considered the placebo effect to be a contributing factor to the response in the placebo group; however, the extent to which the placebo effect influenced the results is unclear.

In terms of the generalizability of the pivotal KALM studies, pruritus severity was measured using the WI-NRS and the Patient Global Impression of Change (PGIC), and HRQoL was measured using the Skindex-10 and 5-D itch scales. However, the experts consulted by CADTH stated that these outcome measures are not routinely used in clinical practice to assess itch intensity or HRQoL in patients. Therefore, there is uncertainty about how changes in pruritus severity measured with the WI-NRS and PGIC and changes in HRQoL measured with the Skindex-10 and 5-D itch scales translate to clinical practice. A limitation to note is that the studies included patients with better hemodialysis adherence and less pruritus severity than is seen in clinical practice in Canada, per feedback from the clinical experts consulted by CADTH. The clinical experts for this review also indicated that the patients in the KALM-1 and KALM-2 trials had better adherence to hemodialysis than in Canadian clinical practice. As such, the generalizability of efficacy outcomes may be overestimated and safety outcomes may be underestimated. In addition, the baseline median WI-NRS score was around 7 for both KALM trials (7.14 for the KALM-1 trial and 7.13 for the KALM-2 trial), and the clinical experts consulted by CADTH indicated the patients with CKD-aP would have a worse itch intensity (with a numeric rating scale score greater than 8) in dermatology clinical practice in Canada. Overall, the selection of patients with better hemodialysis adherence and less severe pruritus (based on the WI-NRS) may limit the applicability of the study results to the patient population in Canada and introduce selection bias, which may lead to uncertainty in the efficacy results. In the KALM trial, data for the primary and secondary or other outcomes were reported up to 12 weeks. According to the clinical expert, patients with CKD-aP would receive the difelikefalin treatment beyond 12 weeks for the desired treatment effect. Although a reduction in symptoms of CKD-aP may be observed during the first 12 weeks of treatment, it is uncertain whether the treatment effect or the safety of difelikefalin beyond 12 weeks would be consistent with the results at week 12.

Indirect Comparisons

No indirect evidence was identified for this review.

Other Relevant Evidence

Additional safety data for the open-label, phase III CR845-CLIN3101 study for up to 52 weeks was summarized in this report.

Description of the CR845-CLIN3101 Trial

One open-label, multicentre, phase III study (CR845-CLIN3101) that evaluated the long-term safety of difelikefalin at a dose of 0.5 mcg/kg administered for up to 52 weeks was included as other relevant evidence to address the gap in the long term safety of difelikefalin for this review.

The long-term safety study included patients who participated in the phase II studies for difelikefalin (CR845-CLIN2005 and CR845-CLIN2101). The long-term safety study also included de novo patients with moderate-to-severe CKD-aP undergoing hemodialysis who had not been previously exposed to difelikefalin and who had not participated in the phase II studies for difelikefalin. The open-label, phase III study consisted of a screening visit, a 52-week treatment period, an end of treatment (EOT) visit, and a follow-up visit 7 to10 days after the EOT visit. Patients received difelikefalin at a dose of 0.5 mcg/kg after each dialysis session, 3 times per week for up to 52 weeks, for a total of approximately 156 doses of the study drug. All scheduled study visits were conducted on dialysis days during the treatment period. The last dose was administered at the last dialysis visit on week 52, at or early termination (ET). The EOT visit was conducted at the dialysis visit following the last dose. A final safety follow-up was conducted 7 to 10 days after the EOT or ET visit. In the long term safety study, AEs, SAEs, withdrawal due to adverse events, and deaths were reported descriptively for each of the 3 treatment groups.

Efficacy Results

No efficacy results were evaluated in the open-label, phase III study.

Harms Results

In the open-label, phase III study, |||||| |||||, and ||||| patients experienced at least 1 AE in the placebo and difelikefalin groups in the phase II study and in the de novo group in the open-label study, respectively. The most common AEs were (frequency ≥ 5%) were nausea, diarrhea, falls, vomiting, hypotension, noncardiac chest pain, hyperkalemia, dizziness, abdominal pain, fluid overload, pneumonia, dyspnea, acute myocardial infarction, pain in extremity, arthralgia, and asthenia. The proportion of patients with at least 1 SAE was |||||| |||||, and ||||| in the placebo and difelikefalin groups in the phase II study and in the de novo group in the open-label study, respectively. The most common SAEs (frequency > 4%) among patients were acute myocardial infarction, angina pectoris, gastrointestinal hemorrhage, pneumonia, cellulitis, fluid overload, hyperkalemia, respiratory failure, pulmonary edema, and hypotension. A total of || patients discontinued the study drug due to AEs, with the proportions being 20.0%, 15.4%, and 10.2% in the placebo and difelikefalin groups in the phase II study and in the de novo group in the open-label study, respectively. A total of 16 deaths occurred during the study, with the proportions being ||||| ||||, and |||| in the placebo and difelikefalin groups in the phase II study and in the de novo group in the open-label study, respectively.

Critical Appraisal

The objective of the open-label, phase III study was to evaluate the long-term safety of difelikefalin, administered intravenously after each dialysis session for up to 52 weeks. Because results for the open-label, phase III trial were only reported descriptively, they should be interpreted with caution. Discontinuation rates were high in all 3 treatment groups. One of the issues with discontinuation from an open-label study, particularly when patients discontinue due to AEs, is that the summary of harms may underestimate the frequency of AEs, because those who remained in the study are more likely to have responded well to treatment. Overall, the high discontinuation rates, as well as the descriptive nature of analysis, introduce uncertainty into the long-term safety results. The lack of comparative evidence makes it difficult to interpret the safety results.

Although the patient population in the open-label study is different than the KALM-1 and KALM-2 populations, the external validity points related to demographic factors from the main report could be applicable to this study population. Although the 0.5 mcg/kg dose is consistent with the Health Canada–approved dose, the duration of the open-label, phase III trial (up to 52 weeks) is more than that in the pivotal KALM-1 and KALM-2 trials (12 weeks). Because it is expected that patients would receive this treatment beyond 12 weeks, the safety results of this open-label, phase III study may be generalizable to this time frame to some extent, but not completely, as the rates of AEs are expected to increase with longer treatment times.

Summary of Pooled Analysis of OLE of KALM-1 and KALM-2

A pooled analysis by Topf et al. (2022)¹⁷ of 2 OLE studies of the KALM-1 and KALM-2 trials assessed the 5-D itch scale for 52 weeks after the 12-week pivotal trials. The study reported the proportion of patients achieving at least a 5-point reduction in the 5-D itch scale was maintained up to 52 weeks in the open-label phase of the KALM-1 and KALM-2 trials. However, there was some uncertainty in the reported long-term treatment effect due to the amount of missing data, with 26.5% (189 of 712) of patients in the pooled analysis population contributing data at week 52 in the OLE phase. In addition, it is difficult to discern much about the long-term efficacy of difelikefalin due to a lack of reporting on other outcomes (i.e., WI-NRS and Skindex-10 scores) and the absence of a control group.

Conclusions

The CADTH systematic review identified 2 phase III, double-blind, placebo-controlled RCTs (KALM-1 and KALM-2) that compared the efficacy and safety of difelikefalin 0.5 mcg/kg with placebo over 12 weeks in patients on hemodialysis with moderate-to-severe CKD-aP. In both of the KALM trials, patients randomized to difelikefalin were more likely than patients randomized to placebo to report an improvement in WI-NRS score or a clinically meaningful reduction in the intensity of the worst itch, based on a 3-point (primary end point) or 4-point (key secondary end point) improvement in the score at week 12. This corresponded to 38% to 41% of patients randomized to difelikefalin and 21% to 25% of patients randomized to placebo with a 4-point improvement in WI-NRS score in the 2 trials. Other outcomes related to pruritus severity, including the change from baseline in WI-NRS and the PGIC scores, were consistent with the primary end point. In the analyses of other secondary end points that assessed HRQoL, measured by the Skindex-10 and 5-D itch scales, treatment with difelikefalin was associated with an improvement in QoL in the KALM-1 trial, but statistical superiority for the same outcomes were not demonstrated in the KALM-2 trial. Generally, AEs were reported at similar rates in the difelikefalin and placebo groups. Treatment with difelikefalin revealed no new safety issues in either KALM trial. No evidence was identified in the KALM trials for mood, days of missed work, days of missed school, or days of missed dialysis, which were considered to be important outcomes of interest by patients and clinicians. A pooled analysis of the OLE studies of KALM-1 and KALM-2 and the phase III, 52-week, open-label study (CLIN3101) provided evidence of long-term safety associated with difelikefalin, and no new safety signals were reported, but the study was subject to many limitations, including the lack of comparative evidence and high discontinuation rates. Overall, based on evidence from the KALM trials, difelikefalin may reduce the severity of itch and improve HRQoL, compared with placebo, but there is uncertainty regarding the magnitude of the treatment effect for patients with moderate-to-severe CKD-aP.

Introduction

Disease Background

CKD is a progressive disease characterized by a gradual loss of renal function and/or abnormalities of renal structure over 3 months. CKD constitutes a major health burden worldwide and is associated with high levels of morbidity and mortality.^2,3

CKD-aP, also known as uremic pruritus, is a CKD comorbidity that is a common, distressing, and underrecognized systemic itch that affects more than 60% of patients undergoing hemodialysis, with 20% to 40% of patients reporting moderate-to-severe pruritus.^4-9 It is marked by an unpleasant sensation and the desire to scratch. The intensity of the itch sensation can fluctuate over time, from barely noticeable to relentless and disturbing.¹³ The distribution of the itch sensation across the body is also inconsistent. Approximately 50% of patients with CKD-aP report generalized pruritus that is usually symmetrically distributed, whereas the remainder report that the itch is mainly localized to the back, face, and dialysis shunt arm.^13,22 The itch can occur either intermittently (i.e., sporadic discomfort) or persistently (i.e., constant itching), and it may present before, during, and/or after dialysis.^13,22 Intense and generalized systemic itching in these patients is associated with poor sleep quality, depression, reduced QoL, increased risk of infection, and an increased risk of death.^5,6,10-12 In Canada, the estimated overall prevalence of CKD-aP in adult hemodialysis patients is about 70%, according to the international, observational Dialysis Outcomes and Practice Patterns Study.⁸ Among patients with CKD on hemodialysis, more than one-third experience moderate-to-severe itch.^8,13,14

The pathophysiology of CKD-aP is not fully understood. a multitude of hypotheses have been proposed, including systemic inflammation and an imbalance in the endogenous opioid system (e.g., overexpression of mu opioid receptors and concomitant downregulation of kappa opioid receptors). Opioid receptors are known to modulate itch signals and inflammation, with kappa opioid receptor activation reducing itch and producing immunomodulatory effects.^11,22-25

The treatment of patients with CKD-aP involves clinical specialists such as nephrologists and dermatologists. There are currently no standardized lab or diagnostic tests for the diagnosis of CKD-aP. A complaint of pruritus in someone with a diagnosis of CKD is presumed to be CKD-aP unless assessment determines a different diagnosis.⁶

Standards of Therapy

There is no approved therapy for CKD-aP in Canada. The clinical experts stated that the goal of treatment is to reduce itch intensity, improve QoL and sleep quality, and decrease itch-related depression and suicidality.

Currently, the treatment for moderate-to-severe pruritus associated with CKD is heterogeneous, as there are no existing guidelines and/or frameworks. The standard of care for CKD-aP includes the use of topical moisturizing treatments, steroids in combination with menthol and camphor, topical calcineurin inhibitors, and the systemic use of gabapentin or pregabalin, naltrexone, thalidomide, biologics (i.e., dupilumab or tralokinumab), kappa opioid receptor agonists (i.e., difelikefalin), and topical phototherapy to reduce itch intensity and improve sleep quality, per feedback provided by the clinical experts consulted by CADTH. Generally, the clinical experts indicated that they would typically start with topical treatments and then move on to systemic drugs and phototherapy. The clinical experts stated that phototherapy is a useful treatment option in clinical practice, but only a small number of patients can receive the treatment due to limited access.

Difelikefalin is the first treatment approved in Canada for moderate-to-severe pruritus in patients on hemodialysis. The clinical experts indicated that difelikefalin would be used in combination with other therapies in a later-line setting when existing therapies are intolerable or have failed to control symptoms or contraindicated. The clinical experts stated that the prior lines of therapy include menthol-containing topical steroids and/or topical calcineurin inhibitors, systemic drugs (gabapentin or naltrexone), and phototherapy.

Drug

Difelikefalin is a selective kappa opioid receptor agonist that acts in the peripheral nervous system. The drug exerts antipruritic effects by means of activation of kappa opioid receptors on peripheral neurons and immune cells. The selective activity of difelikefalin on kappa opioid receptors mostly avoids mu opioid–associated side effects, such as respiratory depression, dependence, and euphoria. Difelikefalin is anticipated to have no meaningful abuse or dependence potential.¹

This is the first CADTH review for difelikefalin. Difelikefalin was granted a Health Canada Notice of Compliance on August 16, 2022, for the indication of treatment of moderate-to-severe pruritus associated with CKD in adults on hemodialysis. Per the Health Canada product monograph, the recommended dose of difelikefalin is 0.5 mcg/kg dry body weight (i.e., the target postdialysis weight).¹

The sponsor is seeking reimbursement of difelikefalin per the indication.

Stakeholder Perspectives

Patient Group Input

This section was prepared by CADTH staff based on the input provided by patient groups. The full original patient inputs received by CADTH have been included in the stakeholder section at the end of this report.

One patient advocacy group, The Kidney Foundation of Canada, provided input for the treatment of adults with CKD on hemodialysis. Patient input was gathered from independent surveys of people living with CKD and their caregivers across Canada in September 2022. A total of 19 responses were gathered from the survey (10 fully completed and 9 partially completed).

More than 90% of patients who responded to the survey reported experiencing itchy skin as part of their kidney disease, with 50% of respondents experiencing itchiness every day, 40% experiencing it several times per week, and 10% experiencing it occasionally. Although 60% of respondents reported living with pruritus for 1 to 2 years, 20% reported living with it for 2 to 5 years, and 20% reported living with it for more than 5 years. The itchiness was described as moderate-to-severe by 80% of respondents. When describing their disease experience, several respondents reported developing scabs and/or sores because of their itchy skin. Many respondents also reported having trouble sleeping as a result of itchiness. One-third (33%) of patients who responded to the survey reported taking medication to treat itchiness associated with kidney disease. Although 33% of respondents reported treatments being covered by their provincial drug plan, 67% reported paying out of pocket. Most survey respondents expressed satisfaction with their current medication or combination of treatments, but 33% reported being neither satisfied nor unsatisfied. However, more than 66% of respondents expressed uncertainty regarding the improvement of their skin appearance related to currently available treatment.

When describing their expectations for CKD therapies in general, patients who responded to the survey mentioned improvement in their well-being or QoL, with 90% of respondents hoping for increased energy. In addition, fewer hospital visits, less medication overall, and side effects and efficacy were important considerations. None of the respondents had experience with the drug under review.

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 moderate-to-severe pruritus associated with CKD.

Unmet Needs

According to the clinical experts consulted by CADTH, treatment for moderate-to-severe pruritus associated with CKD includes the use of topical moisturizing treatments, steroids in combination with menthol and camphor, calcineurin inhibitors, and the systemic use of gabapentin or pregabalin, naltrexone, thalidomide, biologics (i.e., dupilumab or tralokinumab), kappa opioid receptor agonists (i.e., difelikefalin), and topical phototherapy to reduce itch intensity and improve sleep quality. The clinical experts stated that the goal of treatment is to reduce itch intensity, improve QoL and sleep quality, and decrease itch-related depression and, in some cases, suicidality. They also stated indicated that currently used off-label treatments do not adequately address these issues in all patients.

Place in Therapy

The clinical experts indicated that difelikefalin is a selective kappa opioid receptor agonist that acts in the peripheral nervous system, and thus the drug has fewer neuromodulation effects than nonselective kappa opioid receptor agonists. Difelikefalin is the first treatment approved in Canada for moderate-to-severe pruritus in patients on hemodialysis, although the clinical experts consulted by CADTH were uncertain about whether the drug would address the underlying disease process that causes pruritus, as the etiology of CKD-aP is not yet fully understood. The clinical experts indicated that difelikefalin would be used in combination with other therapies in a later-line setting in which existing therapies are intolerable, have failed to control symptoms, or are contraindicated. The clinical experts stated that the prior lines of therapy include menthol-containing topical steroids and/or topical calcineurin inhibitors, systemic drugs (gabapentin or naltrexone), and phototherapy. The clinical experts did not expect that difelikefalin would replace any treatments or cause a shift in the current treatment paradigm, but instead would be used after other more accessible topical and systemic treatment options have failed.

Patient Population

The clinical experts indicated that only patients with ESRD on hemodialysis and with moderate-to-severe pruritus who are not responsive to existing therapies would be candidates for treatment with difelikefalin. According to the clinical experts, the drug is not suitable for patients on peritoneal dialysis, with nondialysis CKD, or who have received a kidney transplant.

Assessing Response to Treatment

The clinical experts stated that clinical evaluations, such as the itch numeric rating scale, can help identify pruritus severity at baseline and assess response to treatment, but they are not typically used in clinical practice. The clinical experts indicated that it is important to consider other causes of pruritus in patients, such as dermatologic conditions, CKD mineral bone disease, hepatobiliary disorders, diabetes, and hematologic conditions, as patients with these conditions may not respond to difelikefalin. The clinical experts mentioned that it is possible that underdiagnosis may occur in clinical practice due to the lack of unique laboratory findings associated with CKD-aP.

Discontinuing Treatment

The clinical experts consulted by CADTH indicated that in clinical practice, subjective patient-reported improvement in symptoms is the primary outcome used to determine whether a patient is responding to treatment. The clinical experts highlighted that reductions in the frequency and/or severity of symptoms would be considered when evaluating response to treatment, along with improvements in sleep quality, depression, adherence to dialysis, depression and suicidality, and overall QoL. The clinical experts consulted by CADTH stated that a patient may discontinue treatment if there is a lack of response, and that this could be assessed at 12 weeks. The development of significant and persistent side effects, such as diarrhea, dizziness, and recalcitrant nausea and/or vomiting may also be cause for discontinuing treatment.

Prescribing Conditions

According to the clinical experts consulted by CADTH, difelikefalin would mostly be prescribed in dialysis units or clinic settings by nephrologists or other physicians.

Clinician Group Input

This section was prepared by CADTH staff based on the input provided by clinician groups. The full original clinician group inputs received by CADTH have been included in the stakeholder section at the end of this report.

Clinician group input on the review of difelikefalin for the treatment of moderate-to-severe pruritus associated with CKD in adults on hemodialysis was received from 3 clinician groups: Saskatchewan Kidney Doctors; the Hemodialysis Specialty Physician Group, Division of Nephrology, The Ottawa Hospital; and Division of Nephrology, Department of Medicine, Dalhousie University and Nova Scotia Health.

The clinician groups agreed that, currently, only off-label medications in Canada are available for the treatment of pruritus in patients with kidney disease. There have been some unmet needs, as current treatment options are unsatisfactory and not effective in reducing the symptom burden. Debilitating symptoms in most severe cases may lead to deterioration in QoL. The clinician groups mentioned the need for a new treatment option that would help relieve symptoms, but with better tolerability, affordability, and ease of administration. One clinician group, in particular, indicated that the most important treatment goals would be to have a therapy that would reduce or maintain the severity of itch below the threshold of clinical importance that is known to be the level above which itch negatively impacts QoL and outcomes important to patients.

The clinician groups mentioned that difelikefalin may cause a paradigm shift in the treatment of itching, with data from RCTs and pooled analyses demonstrating objective effectiveness in reducing the severity of the symptoms of CKD-associated (uremic) pruritus. Input from clinician groups provided different opinions about the place in therapy for difelikefalin. Some groups suggested that it would likely be used as a first-line therapy, whereas others recommended it as an add-on or second-line treatment. Given the route of administration of difelikefalin, the clinician groups pointed out that patients on hemodialysis with moderate-to-severe pruritus would benefit the most from this treatment. The clinician groups mentioned the possibility of underreporting or underdiagnosing CKD-aP, which was also pointed out by the clinical experts consulted by CADTH. Regarding the diagnosis of this indication, the clinician groups noted the lack of diagnostic tests. Although 1 group mentioned using clinical history and exclusion criteria to identify patients, other groups pointed out that the identification of patients and the severity of their symptoms can be done with self-administered questionnaires and screening tools (e.g., the 5-D itch, UP-Dial, and Skindex-10 scales). The clinician groups stated that the reduction in itch severity measured on a numeric rating scale is used to evaluate symptoms. According to the clinician groups, lack of a clinical meaningful response, as well as intolerable side effects, should be considered as discontinuation criteria. The clinician groups added that a meaningful response to treatment for this disease would be symptom reduction leading to better sleep, improved QoL, improved mood, and return to activities of daily living.

Drug Program Input

The drug programs provide input on each drug being reviewed through CADTH’s reimbursement review processes by identifying issues that may impact 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

BSC = best supportive care; CDEC = CADTH Canadian Drug Expert Committee; CKD = chronic kidney disease; CKD-aP = chronic kidney disease–associated pruritus; ICER = incremental cost-effectiveness ratio; QALY = quality-adjusted life-year.

Clinical Evidence

The clinical evidence included in the review of difelikefalin is presented in 2 sections. The first section, the Systematic Review, includes pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as studies that were selected according to an a priori protocol. The second 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.

Systematic Review (Pivotal and Protocol-Selected Studies)

Objectives

To perform a systematic review of the beneficial and harmful effects of difelikefalin 0.5 mcg/kg IV 3 times per week for the treatment of moderate-to-severe pruritus associated with CKD in adults undergoing hemodialysis.

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 6. 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; CKD = chronic kidney disease; HRQoL = quality of life; RCT = randomized controlled trial; SAE = serious adverse event; WDAE = withdrawal due to adverse event.

^aThis outcome was identified as being of particular importance to patients in the input received by CADTH from patient groups.

The literature search was performed by an information specialist using a peer-reviewed search strategy in accordance with the PRESS Peer Review of Electronic Search Strategies checklist.²⁶

Published literature was identified by searching the following bibliographic databases: MEDLINE All (1946–) via Ovid and Embase (1974–) via 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 concepts were difelikefalin, Korsuva, and Kapruvia. The following clinical trials registries were searched: the US National Institutes of Health’s clinicaltrials.gov, WHO’s International Clinical Trials Registry Platform (ICTRP) 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 October 20, 2022. Regular alerts updated the search until the meeting of the CADTH Canadian Drug Expert Committee (CDEC) on February 22, 2023.

Grey literature (literature that is not commercially published) was identified by searching relevant websites from the Grey Matters: A Practical Tool For Searching Health-Related Grey Literature checklist.²⁷ Included in this search were the websites of regulatory agencies (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 with a review of bibliographies of key papers and through contacts with appropriate experts. In addition, the manufacturer of the drug 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 were identified from the literature for inclusion in the systematic review (Figure 1). The included studies are summarized in Table 6. A list of excluded studies is presented in Appendix 2.

Figure 1: Flow Diagram for the Inclusion and Exclusion of Studies

Table 6: Details of Included Studies

ESRD = end-stage renal disease; Kt/V = dialyzer clearance of urea (K), dialysis time (t), volume of distribution of urea (V); NA = not applicable; PGIC = Patient Global Impression of Change; QoL = quality of life; WI-NRS = Worst Itching Intensity Numerical Rating Scale.

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report;²⁰ Fishbane et al. (2020);²⁸ Wooldridge et al. (2020).²⁹

Description of Studies

The primary objective for both KALM trials was to compare the efficacy of difelikefalin at a dose of 0.5 mcg/kg with placebo in reducing the intensity of itch in patients undergoing hemodialysis and experiencing moderate-to-severe pruritus. The shared key secondary objectives of the KALM-1 and KALM-2 trials were to compare the efficacy of difelikefalin at a dose of 0.5 mcg/kg with placebo in improving itch-related QoL and safety in patients on hemodialysis experiencing moderate-to-severe pruritus.

Both KALM trials were designed as multicentre, randomized, double-blind, phase III clinical trials to compare difelikefalin to placebo in patients being treated with hemodialysis who were experiencing moderate-to-severe pruritus. An overview of the study schematic for the KALM-1 and KALM-2 trials is presented in Figure 2 and Figure 3, respectively. The KALM-1 and KALM-2 studies included a double-blind phase and an OLE phase. The double-blind phase consisted of a screening visit (day –28 to day –7), a 7-day run-in period, and a 12-week double-blind treatment period. During the double-blind treatment period, difelikefalin was evaluated relative to placebo over 12 weeks. In the KALM-1 trial, there was a 2-week monitored discontinuation period after the double-blind treatment period during which patients did not receive any study drug. The KALM-2 trial did not have a discontinuation period. At the end of the discontinuation period in the KALM-1 trial and at the end of the double-blind treatment period in the KALM-2 trial, patients who had received at least 30 doses of the study drug during the 12-week double-blind treatment period had the option to receive open-label difelikefalin for up to an additional 52 weeks in the OLE phase. Results for the double-blind treatment period were presented and critically appraised in this report.

A total of 378 patients in the KALM-1 trial and 473 patients in the KALM-2 trial were randomized, in a 1:1 ratio, to treatment with difelikefalin or placebo. The KALM-1 trial was limited to study centres in the US and the KALM-2 trial was conducted globally and included 5 sites in Canada. The primary efficacy end point in the KALM-1 and KALM-2 trials was the percentage of patients at week 12 who achieved at least a 3-point improvement from baseline in the weekly mean score on the daily WI-NRS. Key secondary efficacy end points for both KALM trials were mean change from baseline at week 12 in itch-related QoL (measured with the 5-D itch scale total score and the Skindex-10 scale total score) and the percentage of patients at week 12 achieving at least a 4-point improvement from baseline in the weekly mean score of the daily WI-NRS. In the KALM-2 trial, the proportion of patients with at least a 3-point improvement and at least a 4-point improvement from baseline in WI-NRS at week 8 and week 4 were considered as key secondary outcomes.

For both KALM trials, randomization was based on the following 2 stratification variables:

• the use or nonuse of concomitant medications to treat itch during the week before randomization

• the presence or absence of specific medical conditions (i.e., history of falls or fracture or fall-related fracture, confusional state or mental status change or altered mental status or disorientation, or gait disturbance or movement disorder).

Figure 2: KALM-1 Study Schematic

Figure 3: KALM-2 Study Schematic

Populations

Inclusion and Exclusion Criteria

The key inclusion and exclusion criteria applied to the KALM-1 and KALM-2 trials are summarized in Table 6.

Eligible patients for both KALM trials had ESRD, had been on hemodialysis 3 times per week for at least 3 months before the start of screening, had moderate-to-severe pruritus, and had a body weight between 40.0 kg and 135.0 kg, inclusive. For the KALM-1 trial, eligible patients were 18 years and older, and for the KALM-2 trial, eligible patients were aged 18 to 85 years. The KALM-1 trial included patients with a mean baseline WI-NRS score of at least 4 and at least 4 of 8 worksheets completed in the run-in period, whereas the KALM-2 trial included patients with a mean baseline WI-NRS score of at least 5.

Patients were excluded from both KALM trials if they were with known to be noncompliant with dialysis treatments; scheduled to receive a kidney transplant during the study; started or changed treatment for itch, including antihistamines and corticosteroids (oral, IV, or topical) in the 14 days before screening; had received a new or changed prescription for opioids, gabapentin, or pregabalin in the 14 days before screening; received another investigational drug in the 30 days before the start of screening; or planned to participate in another clinical study while enrolled in this study. Patients who had pruritus attributed to a cause other than ESRD or its complications (e.g., patients with concomitant pruritic dermatological disease or cholestatic liver disease) were not enrolled. Patients who reported pruritus only during the dialysis session or who were receiving ongoing UV B therapy and anticipated receiving such treatment during the study were also excluded from both trials. Patients who participated in a previous clinical study with difelikefalin were excluded.

Baseline Characteristics

A summary of baseline characteristics for the KALM-1 and KALM-2 trials is presented in Table 7.

Baseline demographic characteristics were well balanced between the treatment arms in both KALM trials. The median age of all randomized patients was similar in the 2 studies, at 58.0 years (range, 22 to 88 years) in the KALM-1 trial and 60.0 years (range, 23 to 87 years) in the KALM-2 trial, and the majority of patients were older than 45 years (56.8% and 51.6%, respectively, were 45 to 64 years and 20.2% and 22.1%, respectively, were 65 to 74 years). Most of the patients in the KALM-1 and KALM-2 studies were male (61.0% and 58.2%, respectively), and the predominant races were white (48.8% and 70.3%, respectively) and Black or African American (41.6% and 19.3%, respectively), with Asian patients representing 3.4% and 6.8% of patients, respectively, and other races representing less than 3%. The median prescription dry body weight was 84.0 kg (range, 42.0 kg to 135.0 kg) in the KALM-1 trial and 78.0 kg (range, 42.0 kg to 135.0 kg) in the KALM-2 trial. With the exception that a greater percentage of patients was 75 years or older in the placebo group than in the difelikefalin group (10.6% versus 5.3%) in the KALM-1 trial, the treatment groups were comparable in terms of demographic characteristics in the 2 KALM trials.

In the KALM-1 and KALM-2 trials, the median baseline WI-NRS score was 7.14 (range, 4.1 to 10.0) and 7.13 (range, 4.5 to 10.0), respectively, and the vast majority of patients (91.8% and 99.6%, respectively) had a baseline WI-NRS score of 5 or greater. The mean baseline 5-D itch scale total score was 16.9 (standard deviation [SD] = 3.5) for difelikefalin and 17.9 (SD = 3.5) for placebo in the KALM-1 trial, and 16.7 (SD = 3.5) and 16.2 (SD = 3.3), respectively, in the KALM-2 trial. The mean baseline Skindex-10 scale total score was 36.2 (SD = 14.4) for difelikefalin and 38.3 (SD = 15.4) for placebo in the KALM-1 trial, and 35.5 (SD = 15.0) and 34.2 (SD = 14.7), respectively, in the KALM-2 trial. At baseline, 39.8% of patients in the KALM-1 trial and 36.5% in the KALM-2 trial were using antiitch medications, and 14.1% and 16.6%, respectively, reported at least 1 of the specified medical conditions (i.e., history of falls or fracture or fall-related fracture, confusional state or mental status change or altered mental status or disorientation, or gait disturbance or movement disorder). The median duration of CKD-aP for all patients was 2.5 years (range, 0.1 to 26.5 years) in the KALM-1 trial and 2.3 years (range, 0.0 to 58.4 years) in the KALM-2 trial. The median time interval since the diagnosis of CKD and ESRD for all patients was 5.45 years (range, 0.3 to 42.9 years) and 3.92 years (range, 0.3 to 28.7 years), respectively, in the KALM-1 trial, and 7.53 years (range, 0.3 to 48.3 years) and 4.03 years (range, 0.3 to 30.2 years) in the KALM-2 trial.

All patients in both KALM trials reported at least 1 prior medical condition. The most frequently reported medical conditions in the KALM-1 trial were hypertension (95.8%), anemia of CKD (73.7%), diabetes (62.9%), hyperphosphatemia (51.7%), secondary hyperparathyroidism (49.6%), gastroesophageal reflux disease (45.9%), hyperlipidemia (44.3%), nausea (41.6%), hypotension (40.8%), muscle spasms (39.0%), iron deficiency anemia (37.4%), constipation (35.3%), diarrhea (33.7%), and headache (31.0%). In the KALM-2 trial, the most frequently reported medical conditions were hypertension (95.8%), anemia of CKD (77.3%), hyperphosphatemia (59.0%), diabetes (55.6%), hyperparathyroidism secondary (54.8%), hyperlipidemia (38.0%), gastroesophageal reflux disease (34.4%), iron deficiency anemia (34.2%), and nausea (32.1%).

Table 7: Summary of Baseline Characteristics (Double-Blind Safety Analysis Set)

CKD = chronic kidney disease; ESRD = end-stage renal disease; SD = standard deviation; WI-NRS = Worst Itching Intensity Numerical Rating Scale.

^aIncluding history of falls or fracture (related to fall), confusional state or mental status change or altered mental status or disorientation, or gait disturbance or movement disorder. Observed stratum values.

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report.²⁰

Interventions

In each of the included studies, the intervention employed was difelikefalin 0.5 mcg/kg administered intravenously 3 times per week, after each hemodialysis session. The duration of treatment was up to 64 weeks, which included 12 weeks of double-blind treatment and 52 weeks of open-label treatment in both KALM trials. Patients were provided with difelikefalin as an IV bolus into the venous line of the hemodialysis circuit at the end of each hemodialysis session and could be given either during or after rinse back of the hemodialysis circuit. The dose of difelikefalin to be administered throughout the double-blind treatment period was calculated using the patient’s prescription dry body weight (i.e., the target postdialysis weight, determined by the patient’s nephrologist or at the dialysis unit during screening). If a patient received additional hemodialysis during a given week for any reason, an additional dose of difelikefalin or placebo was administered after hemodialysis. A maximum of 4 doses per week was allowed. No additional doses were given for patients receiving an additional unscheduled ultrafiltration treatment.

The KALM trials were double-blind, placebo-controlled, and used matching placebo vials for IV administration. The matching placebo (0.04 mol isotonic acetate buffer, pH = 4.5) was provided in 2 mL glass vials that contained a minimum extractable volume of 1.3 mL. The placebo buffer solution was composed of acetic acid, sodium acetate trihydrate, sodium chloride, hydrochloric acid, and water for injection. The difelikefalin buffer and placebo buffer were identical in appearance and were packaged, stored, and shipped in an identical manner.

Concomitant medications were defined as medications taken any time after the first dose of difelikefalin on day 1 of the double-blind treatment period through the end of the open-label treatment period or ET (i.e., EOT or ET visit). During either the double-blind phase or the OLE phase, no new medication to treat itch could be initiated.

A summary of concomitant medication and of concomitant antiitch medications use reported by patients in the included studies is presented in Table 8 and Table 9, respectively. In the KALM-1 trial, 99.7% and 100.0% patients randomized to difelikefalin and placebo, respectively, reported any concomitant medication use. The most frequently reported concomitant medications in the KALM-1 trial were beta-blocking drugs, antiparathyroid drugs, antithrombotic drugs, other analgesics and antipyretics, plain lipid-modifying drugs, drugs for peptic ulcer and gastroesophageal reflux disease, antihistamines, selective calcium channel blockers, vitamin A and D, and insulins and analogues. With regard to the use of concomitant antiitch medications, 45.0% and 50.0% of patients randomized to difelikefalin and placebo, respectively, reported using at least 1 concomitant antiitch medication. The most common antiitch medication used was diphenhydramine, which was used 33.3% of patients in the difelikefalin group and 37.8% in the placebo group. All other antiitch medications were used by less than 10.0% of patients and were comparable in the treatment groups.

In the KALM-2 trial, all patients reported concomitant medication use. The most common concomitant medications were antithrombotic drugs, beta-blocking drugs, other antianemic preparations, iron preparations, plain lipid-modifying drugs, drugs for peptic ulcer and gastroesophageal reflux disease, vitamin A and D, selective calcium channel blockers, antiparathyroid drugs, other analgesics and antipyretics, antihistamines, insulins and analogues, and high-ceiling diuretics. With regard to the use of concomitant antiitch medications, 39.1% and 38.6% of patients randomized to difelikefalin and placebo, respectively, reported using at least 1 concomitant antiitch medication. The most common antiitch medication, diphenhydramine, was used by a higher proportion of patients in the difelikefalin group (19.1%) than in the placebo group (11.0%). All other antiitch medications were used by less than 10.0% of patients and were comparable in the treatment groups.

Table 8: Common Concomitant Medications (in > 30% of Patients) (Double-Blind Safety Analysis Set)

NR = not reported.

^aIncluding paracetamol.

^bIncluding methoxy polyethylene glycol-epoetin beta and darbepoetin alfa.

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report.²⁰

Table 9: Concomitant Antiitch Medications (in > 2% of Patients) (Double-Blind Safety Analysis Set)

NR = not reported.

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report.²⁰

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 and subsequently summarized. A detailed discussion and critical appraisal of the outcome measures is provided in Appendix 4.

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

HRQoL = health-related quality of life; PGIC = Patient Global Impression of Change; WI-NRS = Worst Itch Intensity Numerical Rating Scale.

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report.²⁰

Pruritus Severity

Worst Itching Intensity Numerical Rating Scale

Pruritus severity was measured using the WI-NRS on a worksheet that asked patients to indicate the intensity of the worst itching they experienced during the previous 24 hours by marking 1 of 11 numbers, from 0 (labelled with the anchor phrase, no itching) to 10 (labelled as worst itching imaginable), that best described it. Patients were provided with these worksheets to record their 24-hour worst itching assessment scores, both at the clinic on hemodialysis days and at home on nonhemodialysis days.^19,20 The WI-NRS has been widely used for evaluation of chronic itch, including CKD-aP.^5,10,30 The phase II CR845-CLIN2101 study showed that a 3-point difference in the WI-NRS score from baseline was a clinically meaningful improvement.³¹ This threshold was validated in other studies.^15,30

Patient Global Impression of Change

The PGIC is a global patient-reported outcome measure that assesses the overall change in itch (no change, improvement, or worsening) relative to the start of the study.³² The scale has only 1 item; the patient is asked to mark the category that best describes the change in itch, ranging from very much improved to very much worse. The PGIC has been used to evaluate the threshold of meaningful change within pruritus populations for the WI-NRS, based on the FDA recommendation.^30,33 No validity, reliability, or responsiveness outcome results in the CKD-aP population were found from the literature search. No MID results could be validated either.

Health-Related Quality of Life

Skindex-10 Scale

The Skindex-10 scale is an instrument for the measurement of QoL that correlates with itch intensity.⁵ Patients are asked to mark 1 of 7 boxes, numbered from 0 (labelled with the anchor phrase, “never bothered”) to 6 (labelled as “always bothered”) for each of the 10 questions asking how often they had been bothered by their itch and its impact over the previous week. The total score is the sum of the numeric value of each answered question. The total score is subdivided into 3 domain scores, which are sums of the scores for the following question topics: disease domain (questions 1 to 3), mood/emotional distress domain (questions 4 to 6), and social functioning domain (questions 7 to 10).^19,20 The phase II CR845-CLIN2101 study conducted by the sponsor showed that a 15-point difference in the Skindex-10 total score from baseline represented a clinically meaningful improvement.^19,20,31 In a hemodialysis setting, among 103 patients with moderate-to-severe CKD-aP, changes of at least 20% on a visual analogue scale that measured itching intensity was considered clinically meaningful, which is associated with a 3-point to 12-point change on the Skindex-10 scale in patients on hemodialysis.⁵

5-D Itch Scale

The 5-D itch scale was developed as a brief multidimensional questionnaire designed to be useful as an outcome measure in clinical studies. The scale has been validated in patients with chronic pruritus, including patients undergoing hemodialysis, and has been shown to be sensitive to changes in pruritus over time.¹⁸ The 5 dimensions of itch assessed by the scale are degree, duration, direction, disability, and distribution. Patients are asked to mark boxes that best describe the impact of their itch over the previous 2 weeks. The phase II CR845-CLIN2101 study conducted by the sponsor showed that a 5-point difference in the total 5-D itch score from baseline represented a clinically meaningful improvement.^19,20 The threshold for MID was not found from the literature search conducted by CADTH for the CKD-aP population.

Safety

TEAEs included any AE that occurred after the start of the study drug administration and any AE that was present at baseline but worsened in severity after the start of the study drug administration. SAEs included any event that resulted in death, was life-threatening, required hospitalization or prolongation of existing hospitalization, resulted in persistent or significant disability, was a congenital anomaly, and was an important medical event that may have jeopardized the patient or required intervention to prevent any of those outcomes.

Statistical Analysis

A summary of the statistical analyses of efficacy outcomes in the KALM-1 and KALM-2 trials is shown in Table 11.

Sample Size and Power Calculation

The sample size calculations for the KALM-1 and KALM-2 trials are based on the results of a completed phase II, double-blind, placebo-controlled study of difelikefalin in patients with ESRD on hemodialysis who had moderate-to-severe pruritus.³¹ In that phase II study, 30% of patients randomized to the placebo group reported at least a 3-point improvement from baseline with respect to the 24-hour WI-NRS score at the EOT (week 8). The proportion of patients who received difelikefalin and reported a similar improvement in itch scores ranged from approximately 60% to 45% (i.e., a 30% to 15% difference from placebo), depending on the dose of the active study drug (0.5 mcg/kg, 1.0 mcg/kg, or 1.5 mcg/kg).³¹

Assuming a true response rate of 30% for the placebo group and a true response rate of 50% for the difelikefalin group (defining response as an improvement from baseline of ≥ 3 points with respect to the WI-NRS at week 12), the KALM-1 and KALM-2 trials required a sample size of 350 patients (175 per treatment group) to be randomized in a 1:1 ratio to difelikefalin or placebo to detect a treatment difference of 0.2 at the alpha of 0.05 with a power of 96% using a 2-sided continuity corrected Chi-square test. The power of this test statistic would be 84% or greater for differences from placebo as low as 0.16 in both trials.

Interim Analysis and Sample Size Re-estimation

There was 1 planned unblinded interim analysis conducted by an Independent Data Monitoring Committee for both KALM trials for sample size re-estimation. The sample size in the KALM-1 and KALM-2 trials could be increased to 500 patients (250 per treatment group) or keep the original sample size based on the results of the unblinded interim assessment. The interim analysis was conducted when 175 (50%) of the first 350 patients had been randomized and had either completed the 12-week treatment period or had discontinued treatment early. The re-estimation of the sample size was conducted using a conditional power approach, which used the primary end point (the proportion of patients achieving a ≥ 3-point improvement from baseline with respect to the weekly mean of daily 24-hour WI-NRS scores at week 12 of the double-blind treatment period) and 1 key secondary end point (the proportion of patients achieving a ≥ 4-point improvement from baseline with respect to the weekly mean of daily 24-hour WI-NRS scores at week 12 of the double-blind treatment period) as conditions.³⁴ A weighted test statistic, the Cui, Hung. and Wang (CHW) approach, in which the z score is a weighted average of the z score at the interim and the z score observed for data collected after the interim was used to calculate the final P value.³⁵ The Lawrence and Hung approach was used to calculate the point estimates and CIs.³⁶ After the interim analysis, the Independent Data Monitoring Committee recommended that the original sample size be kept.

Multiplicity Adjustment

In both KALM trials, testing of the primary efficacy end point was 2-sided and conducted at the 5% error level. The study was considered positive if the null hypothesis of no treatment difference in the primary efficacy analysis of the primary end point was rejected (i.e., deemed statistically significant) in favour of the alternative, in which patients randomized to difelikefalin experienced significantly less itching than patients randomized to placebo.

To protect against type I errors, a sequential gate-keeping strategy was implemented in both KALM trials. Although the P values corresponding to the hypothesis testing of the secondary variables were reported, they were only considered inferential if the primary analysis was statistically significant. Testing of the secondary efficacy end points was performed sequentially at a 2-sided 5% error level. If the test of an end point in the sequence was not statistically significant, the P value for the tests corresponding to the remaining end points in the sequence would not be considered inferential and the null hypotheses for the subsequent tests would not be rejected.

The testing order for secondary efficacy end points in the KALM-1 trial is as follows:

1. Difference between patients assigned to difelikefalin and placebo with respect to the 5-D itch scale total score.

2. Difference between patients assigned to difelikefalin and placebo with respect to the Skindex-10 scale total score.

3. The proportion of patients achieving at least a 4-point improvement from baseline with respect to the weekly mean of daily 24-hour WI-NRS scores at week 12 of the double-blind treatment period.

The testing order for secondary efficacy end points in the KALM-2 trial is as follows:

1. The proportion of patients achieving at least a 4-point improvement from baseline with respect to the weekly mean of daily 24-hour WI-NRS scores at week 12 of the double-blind treatment period will be tested first.

2. The proportion of patients achieving at least a 3-point improvement from baseline with respect to the WI-NRS at week 8 will be tested next, followed by week 4 testing.

3. Testing will continue with the proportion of patients achieving at least a 4-point improvement at week 8, followed by week 4, in an identical manner.

4. Change from baseline in the Skindex-10 scale total score at week 12 of the double-blind treatment period.

5. Change from baseline in 5-D itch scale total scores at week 12 of the double-blind treatment period.

Primary Efficacy Analysis

In both KALM trials, the primary end points, the proportion of patients who have an improvement from baseline with respect to the weekly mean of daily 24-hour WI-NRS scores of 3 points or greater, were calculated using the ITT datasets. Differences between difelikefalin and placebo were compared using a logistic regression model that contained terms for treatment group, baseline WI-NRS score, use of antiitch medication during the week before randomization, and the presence of specific medical conditions. For the KALM-2 trial, region was also included in the logistic regression model for the primary efficacy analysis. A summary of the statistical analysis of primary efficacy end points is included in the description provided in Table 11.

Data Imputation

In both KALM trials, patients had to report at least 4 daily values for a week for the weekly value to be considered as nonmissing. Missing WI-NRS data for the primary efficacy analysis at the end of week 12 was imputed using a multiple imputation approach, under the assumption that patients who discontinue double-blind treatment early would have similar WI-NRS scores as other patients in their respective treatment arm who have complete data. The multiple imputation with the missing-at-random assumption was performed using logistic regression within treatment group, with covariates for baseline WI-NRS score, randomization stratification factors (i.e., the use or nonuse of concomitant antiitch medications and the presence or absence of a specific medical condition), and the nonmissing WI-NRS scores for each week. For the KALM-2 trial, region was also included in the multiple imputation. In addition, multiple imputation was performed separately for the interim analysis cohort and the postinterim analysis cohort. The missing data were imputed 20 times.

Sensitivity Analyses

Three sensitivity analyses of the primary efficacy end point were conducted to evaluate the robustness of the study results under different assumptions and imputation algorithms (described here) for both KALM trials. For each of the following sensitivity analyses, the final P value was calculated using the CWH procedure.

Sensitivity analysis under assumption 1: Early discontinuations as nonresponders. Patients who discontinue the study drug early were considered to be nonresponders (including patients who discontinued the study drug but continued to report WI-NRS scores). Patients who did not discontinue but who had missing week 12 data were imputed via multiple imputation, as was done in the primary analysis. The imputed data were analyzed using a logistic regression model similar to the 1 used in the primary analysis.

Sensitivity analysis under assumption 2: Multiple imputation with the missing-not-at-random assumption. This sensitivity analysis is an implementation of a pattern mixture model that draws from different populations based on the reason for withdrawal.

• Intermittent missing WI-NRS scores were first imputed using the Markov chain Monte Carlo method.

• For patients who discontinued due to AEs, WI-NRS scores missing after discontinuation were imputed using the distribution of the baseline value of all patients’ daily worst itching score, assuming a trimmed normal (from 4 to 10).

• For patients who discontinued for reasons other than an AE, missing WI-NRS scores after patients discontinued were multiply imputed using data from patients in the same treatment group who had complete data at that time.

Sensitivity analysis under assumption 3: Tipping point analysis. Multiple imputation with mixed missing data mechanisms (missing not at random for difelikefalin and missing at random for placebo) were used to assess the robustness of the missing-at-random assumption. This sensitivity analysis investigates the departure from the missing-at-random assumption by progressively decreasing the treatment differences with respect to WI-NRS scores over the missing visits in the active treatment group until a conclusion from the primary analysis is overturned. This sensitivity analysis was applied to only week 12 values.

• Markov chain Monte Carlo methodology was used to impute the intermittent missing data to a monotone missing pattern.

• A standard missing-at-random–based multiple imputation approach was used to impute data from monotone missing data.

• For patients in the active treatment group, a shift parameter was progressively applied to impute the missing data, until the P value > 0.05.

Secondary Efficacy Analyses

For both KALM trials, the secondary efficacy analyses were performed in the ITT and PP populations. The 5-D itch scale and the Skindex-10 scale scores were analyzed using an analysis of covariance (ANCOVA) and/or a mixed effects model with repeated measures (MMRM). The ANCOVA contains treatment as fixed effects, with baseline score and randomization stratification variables as covariates. The MMRM model contains treatment, week, and treatment-by-week interaction as fixed effects, and baseline score and randomization stratification variables as covariates. The baseline 5-D Itch scale score and the Skindex-10 scale total score were defined as the value collected on day 1, before randomization. Repeated measures included values assigned at the end of weeks 4, 8, 10, and 12 (EOT). A summary of the statistical analyses of secondary efficacy end points is included in the description provided in Table 11.

For both KALM trials, standard descriptive statistics were reported at each time point for the values and changes from baseline, along with the LS means, standard errors, 95% CIs, and differences between treatment groups reported with LS means, standard errors, and 95% CIs.

Missing data were handled implicitly in the MMRM or explicitly with multiple imputation, where intermittent missing WI-NRS scores were first imputed using the Markov chain Monte Carlo method and the monotone missing WI-NRS values were then imputed using the regression method for both KALM trials. For each domain in the 5-D itch scale and the Skindex-10 scale, missing values at week 12 were imputed using a multiple imputation approach, assuming that patients who discontinued double-blind treatment early would have 5-D itch and Skindex-10 scores similar to other patients in their respective treatment arm who have complete data. All available visits were included in the multiple imputation to better inform the week 12 imputed results. The multiple imputation approach was used in the ITT population.

Other Efficacy Analysis

In both pivotal trials, other efficacy analyses were tested in a nonhierarchical fashion without adjustments for multiplicity. Methods used and data reporting 5-D itch scale scores and Skindex-10 scale scores were identical to the secondary efficacy analyses previously mentioned. The treatment difference between difelikefalin and placebo was tested using the Cochran-Mantel-Haenszel exact test with respect to the PGIC, adjusted for the randomization stratification variables. The Mantel-Haenszel estimate of common odds ratio, the exact 95% CI for the common odds ratio, and the Cochran-Mantel-Haenszel exact test P value were reported for this treatment comparison. Additionally, the exact Clopper-Pearson 95% CIs for the proportion of patients who rated their itch condition as very much improved or much improved were reported.

Subgroup Analyses

The subgroup analyses for both KALM trials were prespecified and based on age (< 65 years and ≥ 65 years), sex, race (white, Black or African American, and other), prior antiitch medication use, specific medical conditions, and country (US and outside the US) for the primary end point (i.e., proportion of patients with a ≥ 3-point improvement in WI-NRS from baseline at week 12) and key secondary end point (i.e., proportion of patients with a ≥ 4-point improvement in WI-NRS from baseline at week 12). Subgroup analyses by use of concomitant medications (e.g., antihistamines, gabapentin, pregabalin) to treat pruritus, comorbid conditions, or pruritus severity were included in the CADTH systematic review protocol. None of the subgroups analyzed in the KALM-1 and KALM-2 trials were aligned with those of interest to this review.

Table 11: Statistical Analysis of Efficacy End Points

ANCOVA = analysis of covariance; CMH = Cochran-Mantel-Haenszel; MMRM = mixed effects model with repeated measures; PGIC = Patient Global Impression of Change; WI-NRS = Worst Itching Intensity Numerical Rating Scale.

Analysis Populations

Six analysis sets were used for the KALM-1 study: the enrolled set, the ITT set, the double-blind safety set, the double-blind discontinuation set, the double-blind discontinuation safety set, and the PP set.

Four analysis sets were used for the KALM-2 study: the enrolled set, the ITT set, the double-blind safety set, and the PP set.

The definitions of each analysis set are as follows:

• The enrolled set is defined as the group of patients who sign an informed consent form.

• The ITT set is defined as the group of patients randomized to a treatment group. Following the ITT principle, patients in the ITT set were analyzed according to their randomized treatment, regardless of the actual treatment received. The ITT set was used to analyze all efficacy end points collected during the double-blind phase.

• The double-blind safety set is defined as the group of randomized patients who received at least 1 dose of the double-blind study drug during the double-blind treatment period. Patients in the double-blind safety set were analyzed according to the actual treatment received. The double-blind safety set was used to analyze all safety end points collected during the double-blind phase.

• The double-blind discontinuation safety set is defined as the subset of patients in the double-blind safety set who have at least 1 visit in the discontinuation period. The double-blind discontinuation safety set was used to analyze all end points collected during the discontinuation period. The double-blind discontinuation safety set was not reviewed in this report due to limited relevance.

• The double-blind discontinuation set is defined as the subset of patients in the double-blind safety set who have completed 12 weeks of treatment, have received at least 6 doses in the 2 weeks before the start of the discontinuation period, and have at least 1 visit in the discontinuation period. The double-blind discontinuation set repeated analyses on the end points related to drug withdrawal that are also presented for the double-blind discontinuation safety population. The double-blind discontinuation set was not reviewed in this report due to limited relevance.

• The PP set is defined as the subset of patients in the ITT population who do not have any major protocol deviations that could affect the efficacy analyses of the double-blind data.

Results

Patient Disposition

A summary of patient disposition for the KALM-1 and KALM-2 trials is shown Table 12.

Of the 503 screened patients in the KALM-1 trial, 378 patients were randomized, 189 to the difelikefalin group and 189 to the placebo group. The proportion of patients who completed the double-blind treatment period was 85.7% for the difelikefalin group and 90.4% for the placebo group. A greater proportion of patients discontinued early from the double-blind treatment period in the difelikefalin group than in the placebo group (14.3% versus 9.6%). The most common reasons for early discontinuation from the double-blind treatment period were AEs (6.1%) and patient withdrawal of consent (3.7%). The most frequently reported TEAEs leading to study drug discontinuation were septic shock (0.8%), dizziness (0.8%), and pneumonia (0.5%). With respect to treatment group differences in reasons for early discontinuation, only the proportion of patients who discontinued due to an AE was greater in the difelikefalin group than in the placebo group (7.4% versus 4.8%).

Of 620 screened patients in the KALM-2 trial, 473 patients were randomized, 237 to the difelikefalin group and 236 to the placebo group. The proportion of patients who completed the double-blind treatment period was 87.7% for the difelikefalin group and 94.5% for the placebo group. A greater proportion of patients discontinued early from the double-blind treatment period in the difelikefalin group than in the placebo group (12.3% versus 5.5%). The most common reason for early discontinuation from the double-blind treatment period was AEs (4.2%), followed by other (1.9%) and patient withdrawal of consent (1.3%). The most frequently reported TEAEs leading to study drug discontinuation were anxiety (0.9%) and insomnia (0.9%) in the difelikefalin group. A higher proportion of patients in the difelikefalin group than in the placebo group discontinued due to AEs (5.5% versus 3.0%), withdrew consent (2.1% versus 0.4%), or discontinued for other reasons (2.6% versus 1.3%).

Table 12: Patient Disposition (All Patients, Enrolled Analysis Set)

ITT = intention to treat; PP = per protocol.

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report.²⁰

Protocol Deviation

The incidence of major and minor protocol deviations in the KALM-1 and KALM-2 trials is shown in Table 13. The proportion of patients who reported at least 1 major protocol deviation was ||||| for the difelikefalin group and ||||| for the placebo group in the KALM-1 trial, and ||||| for the difelikefalin group and ||||| for the placebo group in the KALM-2 trial. In the KALM-1 trial, the most frequently identified categories of major deviations were informed consent |||||||, investigational product accountability management |||||||, delegation of authority |||||||, more than 25% of WI-NRS scores missing |||||||, tests and/or assessments and/or procedure |||||||, no dosing in either week 11 or week 12 || ||||||, and receiving less than ||| of planned study doses |||||||. In the KALM-2 trial, the most frequently identified categories of major deviations were dosing noncompliance ||||||||, procedure not performed ||||||||, and procedure performed out of window |||||||. The proportion of patients who reported at least 1 minor protocol deviation was ||||| for the difelikefalin group and ||||| for the placebo group in the KALM-1 trial, and ||||| for the difelikefalin group and ||||| for the placebo group in the KALM-2 trial, which is comparable between the treatment groups in the 2 KALM trials.

Table 13: Protocol Deviations (Double-Blind Safety Analysis Set)

AE = adverse event; IP = investigational product; IRB = Institutional Review Board; NR = not reported; SAE = serious adverse event; WI-NRS = Worst Itching Intensity Numerical Rating Scale.

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report.²⁰

Exposure to Study Treatments

Exposure to study treatments for the KALM-1 and KALM-2 trials is summarized in Table 14. The median duration of treatment during the double-blind treatment period was 85.0 days (range, 5 to 93 days) in the difelikefalin group and 85.0 days (range, 3 to 94 days) in the placebo group in the KALM-1 trial, and 85.0 days (range, 3 to 90 days) in the difelikefalin group and 85.0 days (range, 7 to 91 days) in the placebo group in the KALM-2 trial. The median average dose per administration was 0.5 mcg/kg in both groups for KALM-1 and KALM-2 trials, which is consistent with the dose in the product monograph. During the 12 weeks of the double-blind treatment period, the majority of patients in the safety population (60.2% in the KALM-1 trial and 69.4% in the KALM-2 trial) had 34 to 36 doses (times) administered. Most patients (54.6% in the KALM-1 trial and 41.6% in the KALM-2 trial) missed 1 to 3 doses during the 12 weeks of the double-blind treatment period. Overall, median dosing compliance was 94.44% (range, 66.7% to 100.0%) for patients in the difelikefalin group and 97.22% (range, 71.4% to 100.0%) for patients in the placebo group in the KALM-1 trial, and 97.22% (range, 51.4% to 100.0%) for patients in the difelikefalin group and 97.22% (range, 50.0% to 100.0%) for patients in the placebo group in the KALM-2 trial.

Efficacy

Only efficacy outcomes and analyses of subgroups identified in the review protocol are reported here. Refer to Appendix 3 for the hierarchical testing order of key secondary end points and detailed efficacy data.

Table 14: Study Drug Exposure (Double-Blind Safety Analysis Set)

NA = not applicable; SD = standard deviation.

^aIncluding discontinuation period.

^b% compliance = actual doses / planned doses.

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report.²⁰

Pruritus Severity

Worst Itching Intensity Numerical Rating Scale

The primary efficacy end point for both KALM trials was the proportion of patients achieving at least a 3-point improvement from baseline with respect to the weekly mean of daily 24-hour WI-NRS scores at week 12 of the double-blind treatment period. Both of the KALM trials met the primary end point. Table 15 summarizes these results for the ITT population, based on the combined imputed data from patients in the interim and postinterim analyses. The results were adjusted for the interim analysis using the CHW methodology. In the KALM-1 trial, 52% and 31% of patients randomized to difelikefalin and placebo, respectively, had a WI-NRS score that improved by at least 3 points from baseline to week 12. This corresponded to an odds ratio of 2.72 (95% CI, 1.72 to 4.30; P < 0.001) in favour of difelikefalin. In the KALM-2 trial, 50% and 37% of patients randomized to difelikefalin and placebo, respectively, had a WI-NRS score that improved by at least 3 points from baseline to week 12. This corresponded to an odds ratio of 1.61 (95% CI, 1.08 to 2.41; P = 0.02) in favour of difelikefalin.

The supportive analyses performed for the primary end point in the KALM-1 and KALM-2 trials were conducted using the PP population. The results were consistent with the primary analysis (P < 0.001 with CHW adjustment for the interim analysis and P ≤ 0.001 without CHW adjustment for the interim analysis in the KALM-1 trial, and P = 0.019 with CHW adjustment for the interim analysis and P = 0.027 without CHW adjustment for the interim analysis in the KALM-2 trial). Similarly, the sensitivity analyses conducted using the ITT population showed that efficacy results with difelikefalin were consistent with the efficacy shown in the primary efficacy analysis, with P values less than 0.01 across 3 sensitivity analyses (P < 0.001 for patients who discontinued the study drug early designated as nonresponders, P < 0.001 for multiple imputation followed a missing-not-at-random approach, and P = 0.009 for the tipping point analysis) in the KALM-1 trial. In the KALM-2 trial, the P value was 0.168 for the sensitively analysis with patients who discontinued the study drug early designated as nonresponders, P = 0.029 for the multiple imputation followed a missing-not-at-random approach, and P = 0.044 for the tipping point analysis.

The percentage of patients with at least a 3-point improvement from baseline in WI-NRS score at week 4 and at week 8 was considered to be part of other efficacy end points in the KALM-1 trial, but was considered to be a key secondary efficacy end point in the KALM-2 trial. At week 8, 43% and 28% of patients randomized to difelikefalin and placebo, respectively, in the KALM-1 trial, and 45% and 33% of patients randomized to difelikefalin and placebo, respectively in the KALM-2 trial had a WI-NRS score that improved by at least 3 points from baseline. This corresponded to an odds ratio of 2.33 (95% CI, 1.40 to 3.55; P < 0.001, with the P value not adjusted for multiple testing) in the KALM-1 trial and of 1.69 (95% CI, 1.13 to 2.53; P = 0.01) in the KALM-2 trial in favour of difelikefalin. At week 4, 33% and 18% of patients randomized to difelikefalin and placebo, respectively, in the KALM-1 trial and 35% and 22% of patients randomized to difelikefalin and placebo, respectively, in the KALM-2 trial had a WI-NRS score that improved by at least 3 points from baseline. This corresponded to an odds ratio of 2.5 (95% CI, 1.50 to 4.17; P < 0.001, with the P value not adjusted for multiple testing) in the KALM-1 trial and 1.99 (95% CI, 1.13 to 2.53; P = 0.02) in the KALM-2 trial in favour of difelikefalin.

Table 15: Proportion of Patients Achieving a ≥ 3-Point Improvement From Baseline in WI-NRS Score at Weeks 4, 8, and 12 (ITT Analysis Set)

CI = confidence interval; ITT = intention to treat; LS = least squares; WI-NRS = Worst Itching Intensity Numerical Rating Scale.

Note: Combined analysis used the separate interim and postinterim results to generate an adjusted overall estimate and P value using the Lawrence and Hung and/or CHW methodology.

^aEstimated %, odds ratio, and P value used a logistic regression model with terms for treatment group, baseline WI-NRS score, region (for the KALM-2 trial only), use of antiitch medication during the week before randomization, and the presence of specific medical conditions. Missing values were imputed using multiple imputation under the missing-at-random missing data assumption for interim patients and postinterim patients separately. Counts and percentages were based on nonmissing data.

^bOdds ratio was estimated using the Lawrence and Hung approach. P value was estimated using the CHW approach.

^cThe P value was not adjusted for multiple testing (i.e., the type I error rate has not been controlled).

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report;²⁰ FDA report.¹⁶

The percentage of patients with at least a 4-point improvement from baseline in WI-NRS score at week 12, week 8, and week 4 is summarized in Table 16. At week 12, 41% and 21% of patients randomized to difelikefalin and placebo, respectively, in the KALM-1 trial and 38% and 25% of patients randomized to difelikefalin and placebo respectively, in the KALM-2 trial had a WI-NRS score that improved by at least 4 points from baseline. This corresponded to an odds ratio of 2.89 (95% CI, 1.75 to 4.76; P < 0.001) in the KALM-1 trial and 1.77 (95% CI, 1.14 to 2.74; P = 0.01) in the KALM-2 trial in favour of difelikefalin.

The percentage of patients with at least a 4-point improvement from baseline in WI-NRS score at week 4 and week 8 was considered to be part of other efficacy end points in the KALM-1 trial, but was considered to be a key secondary efficacy end point in the KALM-2 trial. At week 8, 31% and 19.7% of patients randomized to difelikefalin and placebo, respectively, in the KALM-1 trial and 31% and 20.4% of patients randomized to difelikefalin and placebo, respectively, in the KALM-2 trial had a WI-NRS score that improved by at least 4 points from baseline. This corresponded to an odds ratio of 2.11 (95% CI, 1.26 to 3.53; P = 0.005, with the P value not adjusted for multiple testing) in the KALM-1 trial and 1.82 (95% CI, 1.16 to 2.86; P = 0.01) in the KALM-2 trial in favour of difelikefalin. At week 4, 18% and 8% of patients randomized to difelikefalin and placebo, respectively, in the KALM-1 trial and 20% and 13% of patients randomized to difelikefalin and placebo, respectively, in the KALM-2 trial had a WI-NRS score that improved by at least 4 points from baseline. This corresponded to an odds ratio of 2.75 (95% CI, 1.40 to 5.39; P = 0.003, with the P value not adjusted for multiple testing) in the KALM-1 trial and 1.76 (95% CI, 1.04 to 2.98; P = 0.036) in the KALM-2 trial in favour of difelikefalin.

Table 16: Proportion of Patients Achieving a ≥ 4-Point Improvement From Baseline in WI-NRS Score at Weeks 4, 8, and 12 (ITT Analysis Set)

CI = confidence interval; ITT = intention to treat; LS = least squares; WI-NRS = Worst Itching Intensity Numerical Rating Scale.

Note: Combined analysis used the separate interim and postinterim results to generate an adjusted overall estimate and P value using the LH and/or CHW methodology.

^aMultiple imputation with missing-at-random assumption.

^bCounts and percentages were based on nonmissing data.

^cSecondary end point for both the KALM-1 and KALM-2 trials.

^dEstimated %, odds ratio, and P value used a logistic regression model with terms for treatment group, baseline WI-NRS score, use of antiitch medication during the week before randomization, region (KALM-2 trial only), and the presence of specific medical conditions. Missing values were imputed using multiple imputation under the missing-at-random missing data assumption for interim patients and postinterim patients separately.

^eEstimated using the Lawrence and Hung approach.

^fEstimated using the CHW approach.

^gSecondary end points for the KALM-2 trial only.

^hThe P value was not adjusted for multiple testing (i.e., the type I error rate has not been controlled).

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report;²⁰ FDA report.¹⁶

Results for the LS mean change from baseline in the weekly WI-NRS score at week 12, week 8, and week 4 in both KALM trials have been summarized in Table 17, Figure 4, and Figure 5. The LS mean change from baseline to week 12 was –3.2 points for difelikefalin versus –2.0 points for placebo in the KALM-1 trial and –3.1 points for difelikefalin versus –2.5 points for placebo in the KALM-2 trial. Similar results were observed at week 8, when the LS mean change from baseline was –2.7 points for difelikefalin versus –1.8 points for placebo in the KALM-1 trial and –2.8 points for difelikefalin versus –2.1 points for placebo in the KALM-2 trial. At week 4, the LS mean change from baseline was –2.2 points for difelikefalin versus –1.3 points for placebo in the KALM-1 trial and –2.3 points for difelikefalin versus –1.6 points for placebo in the KALM-2 trial. The difference in LS means between difelikefalin and placebo was –1.1 points (95% CI, –1.6 to –0.6 points; P < 0.001) at week 12, –0.9 points (95% CI, –1.4 to –0.6 points; P < 0.001) at week 8, and –0.9 points (95% CI, –1.3 to –0.5 points; P < 0.001) at week 4 in the KALM-1 trial, and –0.6 points (95% CI, –1.1 to –0.2 points; P = 0.008) at week 12, –0.7 points (95% CI, –1.1 to –0.3 points; P = 0.001) at week 8, and –0.7 points (95% CI, –1.1 to –0.3 points; P = 0.001) at week 4 in the KALM-2 trial. Of note, the P values for the difference in LS means between difelikefalin and placebo at week 12, week 8, and week 4 were not adjusted for multiplicity in either KALM trial.

Table 17: MMRM Analysis of Weekly WI-NRS Score Changes From Baseline (ITT Analysis Set)

CI = confidence interval; ITT = intention to treat; LS = least squares; MMRM = mixed effects model with repeated measures; WI-NRS = Worst Itching Intensity Numerical Rating Scale.

^aLS means, P values, and CIs were based on an MMRM analysis with effects for treatment, visit, treatment-by-visit interaction, baseline score, use of antiitch medication during the week before randomization, region (the KALM-2 trial only), and the presence of specific medical conditions. The model was fit using an unstructured covariance structure. Missing values were imputed using multiple imputation under the missing-at-random missing data assumption.

^bPlacebo group was used as reference.

^cThe P value was not adjusted for multiple testing (i.e., the type I error rate has not been controlled).

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report.²⁰

Figure 4: KALM-1 Mean Change From Baseline in WI-NRS Score by Study Week (Primary Efficacy Imputation, ITT Analysis Set) — Redacted

Note: This figure has been redacted at the request of the sponsor. LS means, standard errors, and CIs were based on an MMRM analysis with effects for treatment, visit, treatment-by-visit interaction, baseline score, use of antiitch medication during the week before randomization, and the presence of specific medical conditions. The model was fit using an unstructured covariance structure. Missing values were imputed using multiple imputation under the missing-at-random missing data assumption.

Source: KALM-1 Clinical Study Report.¹⁹

Figure 5: KALM-2 Mean Change From Baseline in WI-NRS Score by Study Week (Primary Efficacy Imputation, ITT Analysis Set) — Redacted

CI = confidence interval; ITT = intention to treat; LS = least squares; WI-NRS = Worst Itching Intensity Numerical Rating Scale; SE = standard error.

Note: This figure has been redacted at the request of the sponsor. LS means, standard errors, CIs, and P values were based on an MMRM analysis with effects for treatment, visit, treatment-by-visit interaction, baseline score, region, use of antiitch medication during the week before randomization, and the presence of specific medical conditions. The model was fit using an unstructured covariance structure. Missing values were imputed using multiple imputation under the missing-at-random missing data assumption.

Source: KALM-2 Clinical Study Report.²⁰

Patient Global Impression of Change

The proportion of responders on the PGIC scale at week 12 is summarized in Table 18. At week 12 in the KALM-1 trial, ||| and ||| of patients randomized to difelikefalin and placebo, respectively, were complete responders. This corresponded to an odds ratio of |||| |||| ||| |||| || ||||| | | |||||||. At week 12 in the KALM-2 trial, ||| and ||| of patients randomized to difelikefalin and placebo, respectively, were complete responders. This corresponded to an odds ratio of |||| |||| ||| |||| || ||||| | | ||||||.

Table 18: Summary of PGIC Responders (ITT Analysis Set)

CI = confidence interval; PGIC = Patient Global Impression of Change.

Note: The denominator for all percentages is the number of patients with available PGIC scores. Clopper-Pearson CIs for proportions are reported. Odds ratio and CI were based on the Mantel-Haenszel estimate, stratified by the use of antiitch medication during the week before randomization and the presence of specific medical conditions.

^aPGIC responders were those with responses of very much improved or much improved.

^bThe P value is calculated with the Cochran-Mantel-Haenszel exact test and was not adjusted for multiple testing (i.e., the type I error rate has not been controlled).

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report.²⁰

Symptom Reduction (e.g., Pruritus, Pain, Sleep Disturbance)

Evidence of outcomes related to an improvement in symptoms of pruritus were reported in the previous section (Pruritus Severity). Evidence of outcomes related to symptom reduction, such as pain and sleep disturbance, was not identified for this review.

Health-Related Quality of Life

Itch-Related Quality of Life (Skindex-10 Scale Score)

Summary of the Main Analyses (Total Score and Domain Scores)

The change from baseline in itch-related QoL, assessed by the Skindex-10 scale total and domain scores, at the end of week 12 was summarized in Table 19. At the end of week 12, the LS mean change in Skindex-10 scale total score was –17.2 for difelikefalin versus –12.0 for placebo in the KALM-1 trial –16.6 for difelikefalin versus –14.8 for versus placebo in the KALM-2 trial. Overall, the difference in LS mean change from baseline to week 12 on the Skindex-10 scale total score between the difelikefalin and placebo groups was –5.1 points (95% CI, –8.0 to –2.3 points, P < 0.001). In the KALM-2 trial, results for the change from baseline to week 12 in the Skindex-10 scale total score indicated no difference between treatment groups (LS mean difference = –1.8 points; 95% CI, –4.3 to 0.8 points; P = 0.171).

The findings on the Skindex-10 scale by domain scores (disease domain, mood/emotional distress domain, and social functioning domain) demonstrated trends of change similar to those in the preceding analysis of Skindex-10 scale total scores, in which the difelikefalin group showed a greater reduction in scores from baseline at week 12 than the placebo group in the KALM-1 trial. For the disease total, the LS mean difference was –1.8 points (95% CI, –2.8 to –0.9 points; P < 0.001); for the mood/emotional distress total, the LS mean difference was |||| |||||| |||| ||| |||| || ||||||||||| | ||||||); and for the social functioning total, the LS mean difference was –||| |||||| |||| ||| |||| || ||||||||||| | | |||||||.

In the KALM-2 trial, the LS mean difference for the disease domain was –0.9 points (95% CI, –1.7 to 0.0 points; P = 0.047); a difference between treatment groups was not observed in the other analyses of the Skindex-10 scale by the domains of mood/emotional distress total (LS mean difference = |||| ||||||| ||| ||| |||| || ||| ||||||| ||||||| in the KALM-2 trial) or social functioning total (LS mean difference = |||| ||||||| ||| ||| |||| || ||| ||||||| ||||||| in the KALM-2 trial).

Summary of Supportive Analyses

The findings for the supportive analysis of the Skindex-10 scale total score in the PP population were consistent with the analysis of the ITT population in the KALM-1 trial, but in the KALM-2 trial, the results were not statistically significant (LS mean difference = |||| ||||||| ||| ||| |||| || |||| ||||||; | |||||| in the KALM-1 trial and LS mean difference = |||| ||||||| ||| ||| |||| || ||| ||||||| | || |||| || ||||||).

Summary of Sensitivity Analyses

The results of the MMRM sensitivity analysis (no imputation) of the Skindex-10 scale total score at week 12 for the ITT population were similar to the preceding ANCOVA with multiple imputation for the KALM-1 trial, but for the KALM-2 trial, the results were not statistically significant. In the KALM-1 trial, the treatment group difference in LS means (difelikefalin minus placebo) was |||| |||| ||| |||| || ||||| | ||||||| in favour of difelikefalin. In the KALM-2 trial, the treatment group difference in LS means was |||| |||| ||| |||| || |||| | |||||||, which was not statistically significant.

In the KALM-2 trial, sensitivity analyses were also performed for Skindex-10 scale scores using ANCOVA with multiple imputation of missing data and control distribution and using ANCOVA with multiple imputation of missing data and baseline distribution. The treatment group difference in LS means was |||| |||| ||| |||| || |||| | ||||||| using ANCOVA with multiple imputation of missing data and control distribution and |||| |||| ||| |||| || |||| | | ||||| using ANCOVA with multiple imputation of missing data and baseline distribution).

Table 19: Change From Baseline in Skindex-10 Scale Score at Week 12 (ITT Analysis Set)

CI = confidence interval, ITT = intention to treat; LS = least squares.

^aLS means and 95% CIs were based on ANCOVA with fixed effects for treatment, with baseline score, region (for the KALM-2 trial only), and the randomization stratification variables as covariates for the KALM-2 trial. Missing values were imputed using multiple imputation under the missing-at-random missing data assumption.

^bPlacebo group was used as reference.

^cThe P value was not adjusted for multiple testing (i.e., the type I error rate has not been controlled).

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report.²⁰

The proportion of patients with at least a 15-point improvement from baseline in Skindex-10 scale total score at week 12 is summarized in Table 20. At week 12, the LS mean percentage of patients with at least a 15-point improvement from baseline in the Skindex-10 scale total score was ||||| in the difelikefalin group and ||||| in the placebo group for the KALM-1 trial, and ||||| for difelikefalin and ||||| for placebo in the KALM-2 trial. The odds ratio for at least a 15-point improvement from baseline with difelikefalin versus placebo was |||| |||| ||| |||| || ||||| |||||||| in the KALM-1 trial and |||| |||| ||| |||| || ||||| |||||||| in the KALM-2 trial.

Table 20: Proportion of Patients With a ≥ 15-Point Improvement in Skindex-10 Scale Total Scores (ITT Analysis Set)

CI = confidence interval; ITT = intention to treat; LS = least squares.

^aCounts and percents are based on nonmissing data.

^bEstimated percent, odds ratio, and P value use a logistic regression model with terms for treatment group, baseline score, use of antiitch medication during the week before randomization, and the presence of specific medical conditions. No imputation was performed.

^cThe P value was not adjusted for multiple testing (i.e., the type I error rate has not been controlled).

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report.²⁰

Itch-Related Quality of Life (5-D Itch Scale Score)

Summary of the Main Analyses (Total Score and Domain Scores)

The change from baseline in itch-related QoL, assessed by the 5-D itch scale total and domain scores at the end of week 12 is summarized in Table 21. At the end of week 12, the LS mean change in total 5-D itch scale score was greater in the difelikefalin group than in the placebo group for the KALM-1 and KALM-2 trials (–5.0 versus –3.7and –4.9 versus –3.8, respectively). Overall, the difference in LS mean change from baseline to week 12 for the 5-D itch scale total score between the difelikefalin and placebo groups was –1.3 points (95% CI, –2.0 to –0.5 points, P < 0.001) in the KALM-1 trial and –1.1 points (95% CI, –1.7 to –0.4 points; P = 0.002) in the KALM-2 trial. Of note, the P value for the difference in LS means on the 5-D itch scale at week 12 for the KALM-2 trial was not adjusted for multiple testing, as the preceding test in the testing hierarchy (i.e., difference in LS means for Skindex-10 score at week 12 in the KALM-2 trial) was not statistically significant.

In the KALM-1 trial, the findings of the 5-D itch domain scores were consistent with the analysis of the 5-D itch scale total score for duration (|| |||| |||||||||| | |||| ||||||| ||| ||| |||| || |||| ||||||| | | |||||), degree (|| |||| |||||||||| | |||| ||||||| ||| ||| |||| || |||| ||||||| | | |||||), direction (|| |||| |||||||||| | |||| ||||||| ||| ||| |||| || |||| ||||||| | ||||||), disability(|| |||| |||||||||| | |||| ||||||| ||| ||| |||| || ||| ||||||| |||||||), and distribution (|| |||| |||||||||| | |||| ||||||| ||| ||| |||| || |||| ||||||| | | |||||). In the KALM-2 trial, the findings of the 5-D itch scale domain scores did not show a difference in duration (|| |||| |||||||||| | |||| ||||||| ||| ||| |||| || ||| ||||||| |||||||) or disability (|| |||| |||||||||| | |||| ||||||| ||| ||| |||| || |||| ||||||| |||||||). The remainder of the analyses by domain were consistent with the analysis of the 5-D itch scale total score for degree (|| |||| |||||||||| | |||| ||||||| ||| ||| |||| || ||| ||||||| |||||||), direction (|| |||| |||||||||| | |||| ||||||| ||| ||| |||| || |||| ||||||| |||||||), and distribution (|| |||| |||||||||| | |||| ||||||| ||| ||| |||| || |||| ||||||| |||||||).

Summary of Supportive Analyses

The findings for the supportive analysis of the 5-D itch scale total score in the PP population were consistent with the analysis in the ITT population for both KALM trials (LS mean difference = –1.4 points; 95% CI, –2.2 to –0.6 points; P < 0.001 in the KALM-1 trial and LS mean difference = –1.3 points; 95% CI, –2.0 to –0.6 points; P < 0.001 in the KALM-2 trial).

Summary of Sensitivity Analyses

The results of the MMRM sensitivity analysis (no imputation) of the5-D itch scale total score at week 12 for the ITT population were similar to the preceding ANCOVA with multiple imputation for both KALM trials (LS mean difference = |||| ||||||| ||| ||| |||| || |||| ||||||| | | ||||| in the KALM-1 trial and LS mean difference = |||| ||||||| ||| ||| |||| || |||| ||||||| | ||||||| in the KALM-2 trial).

In the KALM-2 trial, sensitivity analyses were also performed for the 5-D itch scale total score using ANCOVA with multiple imputation of missing data and control distribution and using ANCOVA with multiple imputation of missing data and baseline distribution: the treatment group difference in LS means was –1.2 (95% CI, –1.9 to –0.5; P = 0.004) using ANCOVA with multiple imputation of missing data and control distribution and –0.7 (95% CI, –1.4 to 0.0; P = 0.046) using ANCOVA with multiple imputation of missing data and baseline distribution).

Table 21: Change From Baseline in 5-D Itch Scale Score at Week 12 (ITT Analysis Set)

CI = confidence interval; ITT = intention to treat; LS = least squares.

^aLS means and 95% CIs were based on ANCOVA with fixed effects for treatment, with baseline score, region (for the KALM-2 trial only), and the randomization stratification variables as covariates for the KALM-2 trial. Missing values were imputed using multiple imputation the missing-at-random missing data assumption.

^bThe P value was not considered inferential and the null hypothesis was not rejected, according to the gate-keeping strategy based on the hierarchical testing order of key secondary end points, as the test of the prior end point (change from baseline in Skindex-10 scale total score) in the sequence was not statistically significant.

^cThe P value was not adjusted for multiple testing (i.e., the type I error rate has not been controlled).

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report.²⁰

The proportion of patients with at least a 5-point improvement from baseline in 5-D itch scale total score at week 12 of the double-blind treatment period in the ITT analysis set is summarized in Table 22. At week 12, the LS mean percentage of patients with at least a 5-point improvement from baseline in the 5-D itch scale total score was greater in the difelikefalin group than in the placebo group for the KALM-1 and KALM-2 trials (||||| ||| ||||| for difelikefalin versus placebo and ||||| ||| ||||| for difelikefalin versus placebo, respectively).The odds ratio for at least a 15-point improvement from baseline for difelikefalin versus placebo was |||| |||| ||| |||| || ||||| |||||||) in the KALM-1 trial and |||| |||| ||| |||| || ||||| |||||||) in the KALM-2 trial. Of note, the P values were not adjusted for multiple testing.

Table 22: Proportion of Patients With a ≥ 5-Point Improvement in 5-D Itch Scale Total Score at Week 12 (ITT Analysis Set)

CI = confidence interval; ITT = intention to treat; LS = least squares.

^aCounts and percents are based on nonmissing data.

^bEstimated percent, odds ratio, and P value use a logistic regression model with terms for treatment group, baseline score, use of antiitch medication during the week before randomization, and the presence of specific medical conditions for the KALM-1 trial. Estimated percent, odds ratio, and P value used a logistic regression model with terms for treatment group, baseline score, region, use of antiitch medication during the week before randomization, and the presence of specific medical conditions for the KALM-2 trial. For both trials, no imputation was performed.

^cThe P value was not adjusted for multiple testing (i.e., the type I error rate has not been controlled).

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report.²⁰

Mood (e.g., Depression)

Evidence of outcomes related to mood, such as depression, was not identified for this review.

Days of Missed Work, Missed School, or Missed Dialysis

Evidence of outcomes related to days of missed work, missed school, or missed dialysis was not identified for this review.

Harms

Only harms identified in the review protocol are reported here. Refer to Table 23 or detailed harms data.

Adverse Events

The percentage of patients with any reported TEAE was similar in the difelikefalin and placebo groups (68.8% versus 62.2% in the KALM-1 trial and 68.1% versus 61.4% in the KALM-2 trial). The TEAEs more commonly reported in the difelikefalin group than in the placebo group were diarrhea (9.5% versus 3.7%), dizziness (6.9% versus 1.1%), and vomiting (5.3% versus 3.2%); the TEAE more commonly reported in the placebo group was nasopharyngitis (3.2% versus 5.3%) in the KALM-1 trial. In the KALM-2 trial, the TEAEs more commonly reported in the difelikefalin than in the placebo groups were diarrhea (8.1% versus 5.5%), vomiting (6.4% versus 5.9%), falls (6.8% versus 5.1%), dizziness (5.5% versus 5.1%), and nausea (6.4% versus 4.2%). Overall, the frequency of TEAEs was roughly balanced in the difelikefalin and placebo groups in both KALM trials, with the exception of dizziness (6.9% versus 1.1%) and diarrhea (9.5% versus 3.7%), which were notably higher in the difelikefalin group than in the placebo group in the KALM-1 trial.

Serious Adverse Events

Serious TEAEs were reported in 25.9% of patients in the difelikefalin group and 21.8% of patients in the placebo group in the KALM-1 trial, and in 24.7% and 21.6%, respectively, in the KALM-2 trial (Table 23). The most common SAEs (difelikefalin versus placebo) were hyperkalemia (2.1% versus 2.1% in the KALM-1 trial and 1.7% versus 1.3% in the KALM-2 trial), sepsis (1.6% versus 2.1% in the KALM-1 trial and 1.3% versus 1.3% in the KALM-2 trial), and pneumonia (1.6% versus 2.7% in the KALM-1 trial and 0.9% versus 0.0% in the KALM-2 trial), as well as fluid overload (1.1% versus 2.1% in the KALM-1 trial), and chest pain (3.4% versus 0.4% in the KALM-2 trial). Overall, the frequency of SAEs was roughly balanced between treatment groups in both KALM trials.

Withdrawals Due to Adverse Events

Withdrawals due to TEAEs were comparable in the 2 treatment groups in both KALM trials. The proportion of patients who discontinued treatment due to TEAEs was 7.9% for difelikefalin and 4.8% for placebo in the KALM-1 trial and 5.5% for difelikefalin and 3.4% for placebo in the KALM-2 trial. Dizziness was the TEAE that most frequently caused discontinuation (difelikefalin versus placebo) in both KALM trials (1.6% versus 0.% for in the KALM-1 trial and 0.4% versus 0.4% in the KALM-2 trial).

Mortality

Deaths were reported in 1.1% of patients in the difelikefalin group and 1.6% of patients in the placebo group in the KALM-1 trial, and in 0.9% for difelikefalin and 0.8% for placebo in the KALM-2 trial. In the KALM-1 trial, sepsis was the cause of death for 2 patients randomized to difelikefalin (and for 0 patients randomized to placebo and 0 patients in the KALM-2 trial), and septic shock was the cause of death for 2 patients randomized to placebo (and for 0 patients randomized to difelikefalin and 0 patients in the KALM-2 trial). All other causes of death (dyspnea and/or hypotension, cardiac arrest, unknown) were infrequently reported, with no more than 1 patient in any treatment group. No specific AE was identified to account for the majority of deaths in either group in the KALM-2 trial.

Notable Harms

The following notable harms were included in the CADTH systematic review protocol: diarrhea, nausea, vomiting, gait disturbance, falls, dizziness, headache, somnolence, seizures, syncope, mental status changes, mood changes, paresthesia (unusual feeling or sensation), hyperkalemia, back pain, tachycardia, and palpitation. The most common serious notable harms (difelikefalin versus placebo) were diarrhea (9.5% versus 3.7% in the KALM-1 trial and 8.1% versus 5.5% in the KALM-2 trial), dizziness (6.9% versus 2.1% in the KALM-1 trial and 5.5% versus 5.1% in the KALM-2 trial), and vomiting (5.3% versus 3.2% in the KALM-1 trial and 6.4% versus 5.9% in the KALM-2 trial), as well as falls (6.8% versus 5.1% in the KALM-2 trial) and nausea (6.4% versus 4.2% in the KALM-2 trial. Overall, during the 12-week treatment period of the KALM-1 and KALM-2 trials, patients who received difelikefalin reported notable harms at a similar or slightly higher frequency than patients who received placebo. There were imbalances in the proportion of patients (difelikefalin versus placebo) reporting diarrhea (9.5% versus 3.7%) and dizziness (6.9% versus 1.1%) as an AE in the KALM-1 trial.

Table 23: Summary of Harms (Double-Blind Safety Analysis Set)

NR = not reported; TEAE = treatment-emergent adverse event.

^aTEAEs relative to the double-blind treatment period are identified as any AE with an onset date after the first dose of the study drug up to the EOT or ET visit or the start of the discontinuation period or 10 days after the last dose if no EOT or ET visit was conducted, whichever is later.

^bFrequency > 5%.

^cFrequency > 2%.

^dFrequency > 1%.

^eReported based on the TEAEs of special interest identified by the sponsor.

^fTachycardia included the following preferred terms: tachycardia, sinus tachycardia, and tachyarrhythmia.

Sources: KALM-1 Clinical Study Report;¹⁹ KALM-2 Clinical Study Report.²⁰

Critical Appraisal

Internal Validity

The KALM-1 and KALM-2 trials appear to have used accepted methods for blinding, allocation concealment, and randomization with stratification. Overall, study baseline demographic and disease characteristics were well balanced between treatment groups in both KALM trials. In the KALM-1 trial, there was an imbalance between the treatment groups in years since diagnosis of ESRD (median = 3.67 years [range, 0.3 to 26.5] versus 4.10 [range, 0.3 to 28.7) for difelikefalin versus placebo], indicating patients in the difelikefalin group had less disease burden and may have had a better response to treatment than those in the placebo group. This difference in patients may introduce a high risk of selection bias and lead to uncertainty in the results. In the KALM-2 trial, there was a difference in the proportion of patients who had diabetes (||||| versus ||||| for difelikefalin versus placebo), and the proportion of patients with hyperphosphatemia was higher in the difelikefalin treatment group than in the placebo group (||||| versus |||||). The clinical experts consulted by CADTH indicated that diabetes and mineral disorders (i.e., hyperphosphatemia) have direct associations with pruritus. The impact of the differences in disease history on efficacy results is uncertain due to limited data reported with respect to the duration of medical history and the severity of these issues. The frequency of hemodialysis (i.e., optimization of hemodialysis) is a potential effect modifier and prognostic factor that was not considered in either KALM trial. Patients with more frequent hemodialysis visits would have better control of the disease and more exposure to difelikefalin than patients with less frequent hemodialysis visits. In addition, more frequent hemodialysis visits indicated better compliance with difelikefalin treatment. Therefore, the frequency of hemodialysis may have a potential impact on the validity of the study results for the treatment effect of difelikefalin; however, the magnitude of the impact is unknown, as there were no data with regard to the treatment effect in patients with different frequencies of hemodialysis in either KALM trial.

Approximately one-third of patients in the studies had at least 1 major protocol deviation (||||| to ||||| across treatment arms at 12 weeks). Particularly in the KALM-2 trial, there were high rates of major protocol deviation due to noncompliance in both treatment groups (||||| versus ||||| for difelikefalin versus placebo) and procedure not performed (||||| versus ||||| for difelikefalin versus placebo). Although the number of major protocol deviations is a limitation, these events were roughly balanced between treatment arms in each study, and the results of the sensitivity and supplementary analyses (including PP analysis) were consistent with the primary estimand.

The primary end point in the KALM-1 and KALM-2 studies was the proportion of patients who achieved a 3-point reduction in WI-NRS scores at week 12, which was used to define improvement and response to treatment for pruritus intensity. The 3-point improvement in WI-NRS scores was considered the MID in patients with CKD-aP, based on a study that used data from a phase II study (CR845-CLIN2101, NCT02858726) and the KALM-1 and KALM-2 trials conducted by the sponsor.³⁰ However, the FDA multidiscipline review for difelikefalin recommended an improvement of at least 4 points as the cut-off for the primary efficacy end point (i.e., proportion of patients with a ≥ 4-point improvement in WI-NRS at week 12) in the target population for both KALM trials.¹⁶ In both the KALM-1 and KALM-2 trials, although odds ratios at week 12 were similar based on the 3-point and 4-point cut-offs, the proportion of patients in each treatment group who met the threshold was higher with the 3-point threshold, which was used for the primary end point. The CADTH review team observed that the sponsor used odds ratios to estimate the treatment effect in the primary and key secondary outcomes. It is important to note that odds ratios tend to give an inflated impression of the treatment effects compared with relative risks.²¹ Therefore, the results of the treatment effect of difelikefalin, compared to placebo, estimated using odds ratios should be interpreted with caution.

There was an imbalance in the proportion of patients (difelikefalin versus placebo) who discontinued the double-blind treatment period (14.3% versus 9.6% in the KALM-1 trial and 12.3% versus 5.5% in the KALM-2 trial). The most common reason for discontinuation in both KALM trials was AEs. Discontinuation due to AEs was slightly higher in the difelikefalin group than in the placebo group in both KALM trials (7.4% versus 4.8% in the KALM-1 trial and 5.5% versus 3% in the KALM-2 trial). Although the clinical experts consulted by CADTH indicated that imbalances in discontinuation were not seen as having a differential effect on blinding or outcomes, as the majority of patients (88.1% in the KALM-1 trial and 91.1% in the KALM-2 trial) completed the study treatment in both KALM trials, there is a potential risk that patients may have been aware of their treatment assignment as a result of the differential rate of reported AEs. Moreover, the WI-NRS scale used in the KALM trials to collect and assess pruritus severity in patients with CKD-aP is a subjective patient-reported outcome measure, so the awareness of treatment may introduce performance bias in the reporting of the results.

In the KALM trials, between ||||| and ||||| of the data were missing for assessments of the primary end points at week 12. Of note, the rate of missing data was higher in the difelikefalin group than in the placebo group (||||| versus ||||| in the KALM-1 trial and ||||| versus ||||| in the KALM-2 trial) for the primary end point of both KALM trials. There were no data reported regarding the demographic and disease characteristics of the patients with missing WI-NRS scores. Given that AEs were the most common cause for discontinuation in both of the KALM trials, there is a high possibility that patients with missing WI-NRS scores dropped out due to AEs or lack of response. The sponsor performed 3 sensitivity analyses with different assumptions (i.e., early discontinuations as nonresponders, missing not at random, and mixed missing data mechanisms with missing not at random for difelikefalin and missing at random for placebo) to assess the robustness of the missing-at-random assumptions. The results of the sensitivity analyses were generally similar, irrespective of the method used to impute the missing data, indicating that the missing-at-random assumption for the multiple imputation in the ITT population was reasonable.

In addition, whether patients with missing WI-NRS scores would satisfy the multiple imputation assumption, which assumes that patients who discontinue double-blind treatment early would have WI-NRS scores similar to other patients in their respective treatment arm who have complete data, is uncertain, given the limited available data. Thus, there was a potential risk of attrition bias, as there was a disproportionally high rate of patients with missing WI-NRS scores in the treatment group, compared with the placebo groups, and it is uncertain whether those patients would systematically differ from those who remained in their respective study arms. The direction of the observed attrition bias is uncertain, as there were no data available with regard to the demographics and disease characteristics for patients with missing WI-NRS scores in either KALM trial.

The clinical experts consulted by CADTH indicated that the clinical relevance of the subgroup variables in the KALM-1 and KALM-2 trials were limited. Subgroups of particular relevance, such as the use of concomitant medications to treat pruritus (e.g., antihistamines, gabapentin, pregabalin), comorbid conditions (e.g., diabetes, neuropathies, liver disease, dermatologic conditions), and pruritus severity (moderate versus severe) were not assessed in the KALM trials. It is not clear whether the effect of difelikefalin would be consistent across these potentially relevant subgroups.

In the KALM trials, only the primary (i.e., proportion of patients with a ≥ 3-point improvement from baseline in WI-NRS score at week 12) and key secondary outcomes (i.e., proportion of patients with a ≥ 3-point improvement from baseline in WI-NRS score at week 4 and week 8, proportion of patients with a ≥ 4-point improvement from baseline in WI-NRS score at week 4, week 8, and week 12, and change from baseline in Skindex-10 scale total score and 5-D itch scale total score) were controlled for multiple statistical testing using a gate-keeping procedure. The lack of adjustments for multiplicity for the domain scores for the Skindex-10 and 5-D itch scales and the PGIC makes it challenging to draw conclusions surrounding any difference between treatment groups for these end points in the KALM trials. Moreover, neither the KALM-1 trial nor the KALM-2 trial were powered to detect a change in secondary end points. Therefore, these outcomes should be interpreted with caution.

The WI-NRS is a valid, reliable, and responsive questionnaire for assessing pruritus severity in patients with CKD-aP, and a 3-point improvement in Wi-NRS scores in both KALM trials was considered clinical meaningful, according to the reported MID threshold for symptom reduction (≥ 3 points). The PGIC used to evaluate improvement in pruritus severity relative to baseline status in the KALM trials is commonly used in clinical trials; however, the validity, reliability, and responsiveness of the instrument has not been studied in patients with CDK-aP. In addition, the Skindex-10 and 5-D itch scales used for HRQoL assessment are generally appropriate for patients with CKD-aP; however, no evidence of MIDs for these 2 measurements in patients with CKD-aP were identified in the literature search conducted by CADTH.

External Validity

The median age of all randomized patients in both KALM trials was around 60 years (58.0 years in the KALM-1 trial and 60.0 years in the KALM-2 trial), and majority of patients were aged 45 years to 64 years (56.8% in the KALM-1 trial and 51.6% in the KALM-2 trial). The clinical experts consulted by CADTH felt that patients on hemodialysis in Canada would be older, on average, with most patients 65 years and older. The KALM-1 trial included study sites in the US only, whereas the KALM-2 trial included 5 study sites in Canada; however, the clinical experts indicated that the demographics of the patient population in both trials was consistent with patients living with CKD in Canada. Thus, the differences between the study population and the patient population in Canada would not likely bias the generalizability of the study results to patients with CKD-aP in Canada.

In the KALM trials, difelikefalin was administered with an IV bolus injection into the venous line of the dialysis circuit at the end of each hemodialysis session. The majority of patients (||||| in the KALM-1 trial and ||||| in the KALM-2 trial) had 34 to 36 doses (times) administered, and ||||| of patients in the KALM-1 trial and ||||| in the KALM-2 trial missed 1 to 3 doses during the 12 weeks of the double-blind treatment period. The clinical experts for this review indicated that the patients in the KALM-1 and KALM-2 trials had better hemodialysis adherence than patients in Canadian clinical practice. The clinical experts stated that better hemodialysis adherence (i.e., hemodialysis optimization) may have contributed to the observed improvement in pruritus severity, and thus the observed treatment effect of difelikefalin may have been overestimated in both the treatment and placebo groups in the KALM trials. In addition, the baseline median WI-NRS score was around 7 in both KALM trials (7.14 in the KALM-1 trial and 7.13 in the KALM-2 trial). The clinical experts consulted by CADTH indicated that patients with CKD-aP would be expected to have a worse itch intensity (with a numeric rating scale greater than 8) in dermatology clinical practice in Canada. Overall, the selection of patients with better hemodialysis adherence and less pruritus severity may limit the applicability of the study results to the real-world patient population in Canada, and may have introduced selection bias, which may lead to uncertainty about the efficacy results.

In the KALM trials, data for the primary and secondary or other outcomes were reported up to 12 weeks. According to the clinical experts, patients with CKD-aP would receive the difelikefalin treatment beyond 12 weeks for the desired treatment effect. As there were no long-term data reported in the KALM trials, it is uncertain whether the treatment effect or the safety of difelikefalin beyond 12 weeks would be consistent with the results at week 12.

The comparator used for both KALM trials was placebo. The clinical experts consulted by CADTH indicated that other relevant treatment options, such as topical treatments, gabapentin, and phototherapy, are commonly used in clinical practice. The lack of active comparators for the KALM trials does not allow conclusions to be drawn about the efficacy of difelikefalin relative to commonly used antiitch therapies in clinical practice.

In the pivotal KALM studies, pruritus severity was measured using the WI-NRS and PGIC, and HRQoL was measured using the Skindex-10 and 5-D itch scales. However, the experts consulted by CADTH stated that these outcome measures are not routinely used in clinical practice to assess itch intensity or HRQoL in patients. Therefore, there is uncertainty about how the changes in pruritus severity measured with the WI-NRS and PGIC and HRQoL measured with the Skindex-10 and 5-D itch scales translate to clinical practice.

Indirect Evidence

No indirect evidence was identified for this review.

Other Relevant Evidence

One open-label, multicentre, phase III study (CR845-CLIN3101) that evaluated the long-term safety of difelikefalin at a dose of 0.5 mcg/kg administered for up to 52 weeks was included as other relevant evidence to address the gap in the long term safety of difelikefalin for this review.

Description of Studies

The long-term safety study included patients who participated in the phase II studies of difelikefalin (CR845-CLIN2005 and CR845-CLIN2101). The long-term safety study also included de novo patients with moderate-to-severe CKD-aP undergoing hemodialysis who had not been previously exposed to difelikefalin and had not participated in the phase II studies of difelikefalin. Figure 6 shows the study design and key features of the open-label, phase III study, which consisted of a screening visit, a 52-week treatment period, an EOT visit, and a follow-up visit 7 to 10 days after the EOT visit. To confirm eligibility, the screening visit was scheduled in the 14 days before administration of the first dose of the study drug. The eligibility criteria for de novo patients required moderate-to-severe pruritus, measured with the 24-hour WI-NRS. Scores for the WI-NRS were collected at each dialysis visit in the week before administration of the first dose (including the score on day 1 of dosing). At least 1 WI-NRS score greater than 4 was needed to be eligible for participation. Patients remained unaware of these eligibility criteria.

Patients received difelikefalin at a dose of 0.5 mcg/kg 3 times per week after a dialysis session for up to 52 weeks, for a total of approximately 156 doses of the study drug. All scheduled study visits were conducted on dialysis days during the treatment period. The final dose was administered at the final dialysis visit on week 52, or at the ET visit. The EOT visit was conducted at the dialysis visit after the final dose. A final safety follow-up was conducted 7 to 10 days after the EOT or ET visit.

Figure 6: Study Schematic for the Open-Label, Phase III Study

EOT = end of treatment; ET = early termination; WI-NRS = Worst Itching Intensity Numerical Rating Scale.

Source: Clinical Study Report for the open-label, phase III study.³⁷

Populations

Inclusion and Exclusion Criteria

Patients meeting the following inclusion criteria were included in this study:

• being 18 years or older and participating in either of the phase II studies or being a de novo patient

• being currently on hemodialysis for ESRD and being categorized as experiencing moderate-to-severe CKD-aP as part of either of the phase II studies or as a de novo patient

• having continued to experience CKD-aP since the previous phase II studies or, for de novo patients, having experienced CKD-aP at the time of screening

• having a dry body weight of at least 40.0 kg at screening (prescription target dry body weight)

• having adequacy of dialysis, defined as meeting any 1 of the following criteria during the 3 months before screening —

  • at least 2 single-pool measurements of (dialyzer clearance of urea × dialysis time) / (volume of distribution of urea) of ≥ 1.2 or

  • at least 2 urea reduction ratio measurements of at least 65% or

  • 1 single-pool measurement of (dialyzer clearance of urea × dialysis time) / (volume of distribution of urea) of at least 1.2 and 1 urea reduction ratio measurement of at least 65%

• for de novo patients, having recorded up to 4 WI-NRS scores at dialysis visits over the week before the first dose (including the score on day 1 of dosing) and having at least 1 of the WI-NRS scores be above 4.

Patients meeting the following criteria were excluded from the study:

• having received an investigational drug (other than the study drug while participating in the phase II studies) in the 30 days before the first dose of the study drug, or planning to participate in another interventional clinical study while enrolled in the study

• having a concomitant disease or any medical condition that could have posed undue risk to the patient, impeded completion of the study procedures, or compromised the validity of the study measurements

• having abnormal liver function, defined as —

  • serum alanine aminotransferase or aspartate aminotransferase more than 2.5 times the reference upper limit of normal at screening and/or

  • total bilirubin more than 2 times the upper limit of normal at screening.

De novo patients meeting the following criteria were excluded from the study:

• a known history of allergic reaction to opiates, such as hives or anaphylaxis

• having pruritus attributed to a cause other than ESRD or its complications (e.g., patients with concomitant pruritic dermatological disease or cholestatic liver disease), determined by the investigator

• having localized itch restricted to the palms of the hands.

Baseline Characteristics

Table 24 Summarizes the baseline characteristics of participants in the open-label, phase III trial.

Table 24: Summary of Baseline Characteristics in the Open-Label, Phase III Study (Safety Population)

CKD = chronic kidney disease; ESRD = end-stage renal disease; SD = standard deviation; WI-NRS = Worst Itch Intensity Numerical Rating Scale.

Note: The Pbo/CR845 group consists of patients randomized to the placebo group in a previous phase II study; the CR845/CR845 group consists of patients randomized to the difelikefalin group in a previous phase II study; and the de novo group consists of patients who did not participate in a previous phase II study.

Source: Clinical Study Report for the open-label, phase III study.³⁷

Results

Patient Disposition

A summary of patient disposition for the open-label, phase III study is summarized in Table 25.

A total of 315 patients were enrolled in the study, of which 288 (91.4%) received the study treatment. Among these 288 patients, 30 (10.4%) had been randomized to the placebo group, 52 (18.1%) had been randomized to the active treatment group in 1 of the phase II studies, and the remaining 206 (71.5%) were de novo patients.

A total of 133 (46.2%) patients out of 288 completed the study treatment. Overall, 57 (19.8%) patients discontinued the study treatment because the sponsor stopped the study for administrative reasons unrelated to safety concerns (0.0%, 1.9%, and 27.2% for the placebo and difelikefalin group in the previous studies and the de novo group in the open-label, phase III study, respectively), whereas 98 (34.0%) patients discontinued the study treatment for other reasons (40.0%, 32.7%, and 33.5%, respectively). For the group of patients who discontinued for other reasons, the most common reasons for discontinuation were AEs, withdrawal of consent, and other reasons.

Table 25: Patient Disposition (All Patients, Enrolled Analysis Set) for the Open-Label, Phase III Study

Note: The Pbo/CR845 group consisted of patients randomized to the placebo group in a previous phase II study; the CR845/CR845 group consisted of patients randomized to the difelikefalin group in a previous phase II study; and the de novo group consisted of patients who did not participate in a previous study.

^aThe safety population included any patient who received the study treatment.

Source: Clinical Study Report for the open-label, phase III study.³⁷

Protocol Deviation

The incidence of major and minor protocol deviations in the open-label CR845-CLIN3101 study is presented in Table 27. The proportion of patients reporting at least 1 major protocol deviation was ||||| and at least 1 minor deviation was ||||||. The most frequently identified categories of major deviations (≥ 2% of all patients) were investigational product accountability management (|||||), good clinical practice noncompliance (||||), and informed consent (||||).

Efficacy

No efficacy outcomes were assessed in the open-label, phase III study.

Harms

Table 28 summarizes the harms outcome for the safety population in the open-label, phase III study. The safety population was defined as all enrolled patients who had received at least 1 dose of the study drug during the open-label phase. The harms results were reported descriptively.

Adverse Events

The percentage of patients with any reported TEAEs in the open-label, phase III study was ||||| ||| ||||| in the placebo and difelikefalin groups in the previous studies, respectively, and ||||| in the de novo group. The most common TEAEs (> 10%) were nausea ||||||| |||||| ||| |||||||, diarrhea ||||||| |||||| ||| |||||||, falls ||||||| |||||| ||| |||||||, vomiting ||||||| |||||| ||| |||||||, hypotension ||||||| |||||| ||| |||||||, noncardiac chest pain |||||| |||||| ||| |||||||, hyperkalemia (||||| |||||| ||| ||||||), dizziness ||||||| ||||| ||| |||||, abdominal pain ||||||| ||||| ||| |||||, and acute myocardial infarction ||||||| |||||| ||| ||||| in the placebo and difelikefalin groups in the previous studies and in the de novo group in the open-label, phase III study, respectively. The percentage of patients with any TEAEs in the open-label, phase III study was higher than the percentages seen in the pivotal KALM-1 and KALM-2 studies.

Serious Adverse Events

Serious TEAEs (SAEs) were reported in ||||| and ||||| in the placebo and difelikefalin group in the previous studies, respectively, and in ||||| in the de novo group. The most common SAEs (> 4%) were acute myocardial infarction (|||||| ||||| ||| ||||||), angina pectoris |||||| ||||| ||| ||||||, gastrointestinal hemorrhage |||||| ||||| ||| ||||||, pneumonia ||||||| ||||| ||| ||||||, cellulitis |||||| ||||| ||| ||||||, fluid overload |||||| ||||| ||| ||||||, hyperkalemia |||||| ||||| ||| ||||||, respiratory failure ||||||| ||||| ||| ||||||, pulmonary edema |||||| ||||| ||| |||||, and hypotension |||||| ||||| ||| ||||| in the placebo and difelikefalin groups in the previous studies and in the de novo group in the open-label, phase III study, respectively. The percentage of patients with any SAEs in the open-label, phase III study was higher than the percentages seen in the pivotal KALM-1 and KALM-2 studies.

Withdrawals Due to Treatment-Emergent Adverse Events

The proportion of patients who discontinued treatment due to TEAEs was |||||| |||||| ||| ||||| in the placebo and difelikefalin groups in the previous studies and in the de novo group in the open-label, phase III study, respectively. Cardiac arrest was the most frequently reported TEAE (> 3%) to cause discontinuation in the difelikefalin group |||||||, whereas myocardial infarction, acute myocardial infarction, chills, fatigue, Clostridium difficile colitis, decreased blood sugar, visual hallucinations, and insomnia were the most frequently reported TEAEs to cause discontinuation in the placebo crossover group |||||||. The de novo group had the least reported TEAEs leading to study drug discontinuation compared with other 2 groups. The percentage of patients with any TEAE leading to study drug discontinuation in the open-label, phase III study was higher than the percentage seen in the pivotal KALM-1 and KALM-2 studies.

Mortality

Deaths were reported in |||| of patients in the placebo crossover group, |||| in the difelikefalin group, and |||| in the de novo group. No specific AE was identified to account for the majority of deaths in any group in the open-label, phase III study.

Notable Harms

The most common serious notable harms were diarrhea ||||||| |||||| ||| |||||||, nausea ||||||| |||||| ||| |||||||, falls ||||||| |||||| ||| |||||||, vomiting ||||||| |||||| ||| |||||||, dizziness ||||||| ||||| ||| ||||||, hyperkalemia |||||| |||||| ||| ||||||, mental status changes |||||| |||||| ||| ||||||, unusual feeling or sensation |||||| |||||| ||| ||||||, syncope |||||| ||||| ||| ||||||, and altered mood |||| ||||| ||| ||||| in the placebo crossover, difelikefalin, and de novo groups, respectively. The percentage of patients with the serious notable harms in the open-label, phase III study was higher than the percentage seen in the pivotal KALM-1 and KALM-2 studies.

Table 27: Summary of Harms (Safety Population) for the Open-Label, Phase III Study