Recent Advances in Sickle-Cell Disease Therapies: A Review of Voxelotor, Crizanlizumab, and L-glutamine

Sickle-cell disease (SCD) is an inherited hemoglobinopathy, causing lifelong complications such as painful vaso-occlusive episodes, acute chest syndrome, stroke, chronic anemia, and end-organ damage, with negative effects on quality of life and life expectancy. Within the last five years, three new treatments have been approved: L-glutamine in 2017 and crizanlizumab and voxelotor in 2019. We conducted a literature search of these three medications, and of the 31 articles meeting inclusion criteria, 6 studied L-glutamine, 9 crizanlizumab, and 16 voxelotor. Treatment with L-glutamine was associated with decrease in pain crises, hospitalizations, and time to first and second crises, with a decrease in RBC transfusion rate. Barriers to filling and taking L-glutamine included insurance denial, high deductible, and intolerability, especially abdominal pain. Crizanlizumab was associated with a reduction in pain crises and time to first crisis, with reduction in need for opioid use. Adverse effects of crizanlizumab include headache, nausea, insurance difficulty, and infusion reactions. Voxelotor was associated with increased hemoglobin and decreased markers of hemolysis. Barriers for voxelotor use included insurance denial and side effects such as headache, rash, and diarrhea. These three medications represent exciting new therapies and are generally well-tolerated though price and insurance approval remain potential barriers to access. Other studies are ongoing, particularly in the pediatric population, and more real-world studies are needed. The objective of this article is to evaluate post-approval studies of crizanlizumab, voxelotor, and L-glutamine in SCD, with a focus on real-world efficacy, side effects, and prescribing data.


Introduction
Sickle-cell disease (SCD) is an inherited hemoglobinopathy, with an estimated 300,000 babies born worldwide with the disease [1]. In the United States, an estimated 100,000-120,000 people live with SCD, primarily of African American or Hispanic descent [2]. Due to a single amino acid substitution in the beta-globin chain, sickle cells are more prone to polymerization, vaso-occlusion, and hemolytic anemia in the setting of hemoglobin deoxygenation [3]. SCD is a chronic, lifelong disease characterized by acute and chronic problems, including painful vaso-occlusive episodes (VOEs), acute chest syndrome, stroke, chronic anemia, and end-organ damage in nearly all organ systems [4]. SCD also negatively affects patients' health-related quality of life (HRQOL), with an inverse relationship to medication adherence [5,6]. While advances in treatment have led to an increase in life expectancy, models still show a reduced life expectancy of 54 years compared to 76 of controls [7].
First approved for SCD in 1998, hydroxyurea, which increases the fraction of hemoglobin F and thus decreases the likelihood of cells to deoxygenate and polymerize, remained the only U.S. Food and Drug Administration (FDA)-approved therapy for SCD for decades [8]. Hydroxyurea has shown to reduce vaso-occlusive episodes and other effects of SCD, leading to an increase in quality of life and morbidity and mortality [9]. However, adherence rates to hydroxyurea remain low, with estimates between 49% and 85% [10]. Over the past five years, three new medications have been approved for treatment of sickle-cell disease: L-glutamine (in 2017), crizanlizumab (in 2019), and voxelotor (in 2019).
L-glutamine, an amino acid, reduces oxidative stress of red blood cells, thus reducing endothelial adhesion [11]. In an initial Phase 3 double-blind placebo-controlled study, L-glutamine, both alone and in combination with hydroxyurea, was shown to reduce the number of pain crises compared to placebo [12]. Crizanlizumab, a P-selectin inhibitor, reduces adhesion between the endothelium and endothelial cells, platelets, sickled red blood cells, and leukocytes [13]. In a Phase 2 double-blind placebo-controlled study (SUSTAIN trial), crizanlizumab also resulted in a statistically significant reduction of pain crises [14]. Finally, voxelotor is a hemoglobin S polymerization inhibitor, increasing the cell's affinity for oxygen and stabilizing the oxygenated hemoglobin state [15]. In its Phase 3 double-blind placebo-controlled study (HOPE trial), treatment with voxelotor significantly increased hemoglobin levels and reduced markers of hemolysis [16].
The objective of this article is to evaluate post-approval studies of crizanlizumab, voxelotor, and L-glutamine in SCD, with a focus on real-world efficacy, side effects, and prescribing data.

Literature Search and Articles Selection
A medical librarian (M.B.) performed a comprehensive literature search on 19 January 2022, of PubMed, Embase, Cochrane Library, CINAHL, Scopus, and Web of Science. The keywords "sickle cell anemia" AND "voxelotor" OR "crizanlizumab" OR "glutamine" were used, with a date range of 2017 to 2022, corresponding to the post-FDA-approval dates. This search returned 236 citations. Two citations were added through citation analysis. Following the initial collection, citations were evaluated based on the inclusion and exclusion criteria. Figure 1 depicts the flow of studies through the review.

Inclusion and Exclusion Criteria
Eligible studies included full articles and abstracts of clinical trials, including both the original and post-hoc/subgroup analyses of the Phase 2 and 3 studies cited above, retrospective studies, prospective studies, case series, and case reports. We included case series and case reports that captured any reported rare or uncommon side effects of any of these therapies. All were original studies of L-glutamine, crizanlizumab, or voxelotor on children or adults with SCD, with reports of clinical efficacy, side effects, or prescribing data. Review articles, molecular and pharmacokinetic studies, cost-effective models, studies prior to the initial Phase 2/3 studies, duplicates, or initial/interim data that were later reported more fully were excluded.

Data Extraction
The following data elements were extracted from the articles retrieved and meeting inclusion criteria: author name, year of study report, country of population studied, outcome measured, study design, age of study population, sample size, and a descriptive summary of findings related to clinical efficacy, side effects, and prescribing data.

Description of Included Studies
Of the six studies investigating L-glutamine, three stemmed from the same randomized control study, including the original multi-center Phase 3 study (n = 230) [12], a subgroup analysis evaluating outcomes based on the number of vaso-occlusive crises in the year prior to the study [18], as well as a post hoc analysis of umber of transfusions for the study group [17]. One case series (n = 4, ages 9-24) was included evaluating opioid use in patients starting L-glutamine [19]. Two single-center retrospective studies were included, both (n = 50 and n = 111) investigating prescribing data, compliance, and side effects of L-glutamine [20,21]. Table 1 summarizes the findings of these included studies. Of the nine studies investigating crizanlizumab, four were from the original multicenter randomized-control SUSTAIN trial (n = 198), including the original Phase 3 data [14], and post hoc analysis of rates of and time to first hospitalizations [22], secondary endpoints [23], and days of opioid use [24]. One retrospective cohort study (n = 6) followed patients for 1 year following SUSTAIN, monitoring for pain crises, treatment patterns, and utilization of healthcare resources [25]. One pooled report between two Phase 2 trials (SUS-TAIN and SOLACE-kids, n = 111) investigated safety and side effects of crizanlizumab [26]. One retrospective multi-center study (n = 297) evaluated insurance approval and adherence [28]. One retrospective review of a safety database (n = 28) evaluated infusion-related reactions [29]. One case report was included, describing two patients experiencing acute febrile reactions during crizanlizumab infusion [27]. Table 2 summarizes the findings of these included studies.

Voxelotor
Of the sixteen studies investigating voxelotor, seven were related to the original randomized-control HOPE trial (n= 274), including the original Phase 3 data [16], post hoc analyses of hemolysis [30], leg ulcers [32], and concomitant use of hydroxyurea [41], and 72-week follow-up analyses of changes from baseline hemoglobin [31] and in VOE incidence [33] as well as long-term open-label extension (n = 178) monitoring hemoglobin response and markers of hemolysis [34]. One open-label, multicenter, multiple-dose trial (HOPE KIDS 1, n = 45) measured increase in hemoglobin and side effects in children [35]. Two Phase 2a studies evaluating increases in hemoglobin in adolescents were included at both 900 mg (n = 25) and 1500 mg (n = 15) per day [36,37]. Two single-center retrospective analyses were included investigating increases in hemoglobin (n = 17) and real-world experience of barriers to use and health outcomes (n = 54) [38,42]. Two multicenter retrospective chart reviews (n = 60 and n = 300) evaluated prescribing trends, side effects, and clinical efficacy [43,44]. One retrospective review (n = 1275) within one health system investigated the effect on anemia [40]. One pilot study (n = 9) evaluated effect on exercise capacity [39]. Table 3 summarizes the findings of these included studies.

L-glutamine
In the original Phase 3 trial, treatment with L-glutamine vs. placebo led to a statistically significant reduction in pain crises, hospitalizations, time to first and second crises, ACS rates, and days in the hospital. The cumulative number of pain crises was 25% lower compared to the placebo group over the entire 48-week treatment period. It also showed no significant between-group differences in hemoglobin level, hematocrit level, or reticulocyte count or ED visits [12]. In subsequent analyses, pain crises rate ratios were similar in all subgroups when based on or adjusted for the number of crises in the year prior to the study period [18]. In a post hoc analyses of transfusion rates, patients had decreased units of transfusions per year in the treatment arm as well compared to placebo [17]. One case series showed patients treated with L-glutamine had decreased opioid usage without having changes in average hemoglobin [19].

Voxelotor
In the HOPE trial, treatment with voxelotor resulted in a statistically significant increase in hemoglobin level, with 51% of patients in the 1500 mg group having a response, with a mean change in hemoglobin of 1.1 g/dL (SD ± 0.13). This was also associated with a decrease in markers of hemolysis (i.e., indirect bilirubin and reticulocyte percentage). The annual incidence of VOE was not significantly different between the treatment and control groups [16]. In post hoc analyses of the HOPE data, those patients that had greater improvements in hemoglobin had the greatest reduction in markers of hemolysis, and nearly all patients receiving voxelotor in the trial had clinical improvement in their leg ulcers [30,32]. In 72-week follow-up of HOPE, the hemoglobin level increase was sustained, both with a mean increase in hemoglobin of 1.0 g/dL (SD ± 0.15) in up to 80% of patients, and it was associated with a decrease in the markers of hemolysis. They also had lower number of VOEs, but it was not statistically significant [31]. VOE incidence rates were lowest in patients who had achieved the highest hemoglobin levels, specifically higher than 10 g/dL [33]. In long-term extension of HOPE, the mean change in hemoglobin was 1.3 g/dL, with sustained decreases in markers of hemolysis [34]. Similar increases in hemoglobin of about 1 g/dL (ranging from 1-1.6) with decreases in markers of hemolysis were also found in the open-level multiple-dose trial HOPE KIDS 1, Phase 2a studies in adolescents, and multiple retrospective chart reviews [35][36][37][38][39][40]44]. Recent real-world data suggest potential clinical benefits related to reduction in VOE events, yet there are some inherit methodological limitations in these studies [40,43].

Crizanlizumab
In the SUSTAIN trial, treatment with higher-dose crizanlizumab showed a statistically significant reduction in the rate of vaso-occlusive episodes and longer time to first and second crises, with a decreased median rate of crises per year [14]. In post hoc studies of the SUSTAIN data, there was also a decrease in hospitalizations, increase in patient crisis-free time or days across multiple subgroups studied (i.e., higher baseline crises, HbSS genotype, concomitant use of HU), and reduction in days of opioid use [22][23][24]. In a 1-year follow-up from SUSTAIN, patients with higher-dose crizanlizumab had lower VOEs, but the study was underpowered to detect statistical significance. All patients still required opioids, one-third required transfusions, and all but one patient still required healthcare resources, such as emergency room visits or hospital admissions [25].

L-glutamine
There have been no published data or studies related to the effects of l-glutamine on HRQOL outcomes in SCD.

Crizanlizumab
There have been no published data or studies related to the effects of crizanlizumab on HRQOL outcomes in SCD.

Voxelotor
Recent real-world data suggest that voxelotor was associated with improvement in HRQOL outcomes, as measured by the Clinical Global Impression of Change (CGI-C) and Patient Global Impression of Change (PGI-C) rating scales (Idowu, 2022 #6495). These observations are consistent with the CGI-C data reported earlier in the HOPE clinic trial.

L-glutamine
In the original Phase 3 trial, adverse events were higher in the treatment group with l-glutamine compared to placebo, with nausea, chest, and musculoskeletal pain and fatigue being the most common side effects. About 2.7% of subjects discontinued therapy due to side effects [12]. In retrospective studies, since L-glutamine's approval, most frequent side effects cited included abdominal pain, nausea, constipation, or taste of the medication [20,21]. When looking at l-glutamine prescribing data, at one pediatric center, 70% of prescriptions were dispensed, and 24% had good compliance (defined as ≥70%) [20]. At another adult center, at the end of a study period, 35% never filled the prescription, 42% discontinued, and only 19% were still actively taking L-glutamine. Barriers to initiating l-glutamine following prescription included denial of a prior authorization, high deductible, or other insurance issues. Reasons for stopping therapy included poor adherence, side effects, with a median day to stopping of 47 days [21].

Crizanlizumab
In the original SUSTAIN trial, adverse effects occurring more often in the treatment crizanlizumab group compared to the placebo included arthralgias, diarrhea, pruritis, vomiting, and chest pain [14]. In pooled data among the SUSTAIN and SOLACE trials, 21.6% of patients suffered a serious adverse event, but only 5.4% were thought to be due to crizanlizumab. The most common adverse events reported included headache, nausea, and back pain [26]. Infusion-related reactions, such as pain or fever, were rare but cited, with a rate of 1.67 cases per 100 patient-years in a review of the safety database, and 1.8% having a reaction in the pooled trial data [26,27,29]. Infusion reactions usually occurred on the 1st or 2nd infusions, with most patients stopping treatment after developing a reaction [29]. Barriers to crizanlizumab included insurance denial and transportation issues to get the infusion [28]. Reasons for stopping, with 25-32% stopping treatment prematurely, included side effects, lack of feeling pain was improving, or that pain was worsening [26,28]. The use of premedication prior to crizanlizumab infusions varied, with 64% of patients not automatically using premedication [28].

Discussion
As the only new agents in decades, L-glutamine, crizanlizumab, and voxelotor represent exciting new therapies for SCD with potential clinical benefits to improve patients' health outcomes. L-glutamine was found to have decreased pain crises, hospitalizations, opioid use, and rates of ACS but without affecting hemoglobin rates or markers of hemolysis. Side effects include nausea, pain, and fatigue but is generally well-tolerated. Crizanlizumab similarly resulted in decreased VOEs, hospitalizations, and opioid use, without major changes in hemoglobin. However, with it being an infusion, it has side effects in addition to headache, nausea, and back pain. Voxelotor was consistently found to increase hemoglobin levels and decrease markers of hemolysis. It is generally well-tolerated, with mild side effects related to headache, GI symptoms, and rash.
All three medications have significant cost associated with them, with insurance approval and co-pay of the medication and, for crizanlizumab, transportation to the infusions as barriers to care. However, in a cost-effective analysis for L-glutamine, treatment alongside HU resulted in a decrease in overall cost due to hospitalizations averted [45]. Similarly, for crizanlizumab, whereas crizanlizumab alone was not cost-effective, when added to HU, the decrease in VOEs and hospitalizations made crizanlizumab more costeffective than the cost of the medication itself [46]. There is growing evidence for real-world data of all these medications, which would be key to optimize their utilization and be able to overcome barriers and other logistical considerations. Moreover, taking a patient-centered approach is critical to better understand how to select the right disease-modifying therapy that would benefit patients the most based on their clinical phenotype and genotype.
Ongoing trials are further looking into the effects of these medications beyond the initial pivotal studies. The STEADFAST trial is investigating crizanlizumab versus standard of care, with respect to chronic kidney disease (CKD) and SCD nephropathy [47]. In addition, while some of the trials included adolescents, further studies are now looking further into the pediatric population, including crizanlizumab in the STAND and SOLACE-Kids trials [48,49]. The HOPE KIDS 1 and 2 studies are investigating voxelotor's effects on pediatric patients as well as its specific effect on TCDs [50,51]. It is worth noting that voxelotor was recently FDA-approved for children aged 4-11 years old based on clinical benefits related to hemoglobin level increase. There also are multiple other therapies in the pipeline with promising early data, such as inclacumab (p-selectin inhibitor), GBT021601 (next-generation HbS polymerization inhibitor), etavopivat and mitapivat (pyruvate kinas activators), FTX-6058 (fetal hemoglobin induction), IMR-687 (inhibitor of phosphodiesterase 9), and CSL889 (plasma-derived hemopexin), among others. There are several other active studies related to curative approaches for SCD, such as gene therapy and gene editing, which have shown encouraging results so far; yet, long-term follow-up is still needed to ensure sustained clinical benefits and favorable safety.
Strengths of this literature review include its broad and thorough approach, including not just established and fully published clinical trials but also abstracts of retrospective studies and case reports, many of which published within the last year, which allows it to be more up to date on the most recent data coming out on these therapies. It also included not just clinical effects but also prescribing data and real-world data, which are helpful and practical for patient care.
Limitations of this review include the fact that several of the included studies stem from the original three pivotal trials for the medications, with many publications having post hoc analyses or limited single-center chart review studies. Additionally, many of the most recent studies are still abstracts, without full trials finishing up since the approval of the medications. Finally, there have yet to be many publications on the effects of these medications on other effects of SCD beyond VOE, such as end organ damage and healthrelated quality of life.
In conclusion, SCD a lifelong disease with risk for several complications and significant impairment of quality of life. These three disease-modifying therapies, namely l-glutamine, crizanlizumab, and voxelotor, represent the start of a new era for novel therapies for SCD with several others in the pipeline SCD. L-glutamine and crizanlizumab have both shown impact on reducing VOEs and days in the hospital, with voxelotor improving anemiareducing hemolysis. All three therapies are generally safe and effective for SCD patients with potential clinical benefits, and they are considered valuable additions to our treatment arsenal, especially when hydroxyurea is not well-tolerated or not desired by patients or caregivers. Funding: This project was supported by grant (K23HL150232, PI: Badawy) from the National Heart, Lung, and Blood Institute of the National Institutes of Health. The content is solely the responsibility of the authors and does not necessarily represent the National Institutes of Health.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
No new data were created or analyzed in this study. The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors have no relevant conflict or competing interests.