Gene therapies for sickle cell disease: Effectiveness and value

Sickle cell disease (SCD) is an extremely debilitating genetic disorder affecting the shape and function of red blood cells, leading to recurrent painful vaso-occlusive crises (VOCs) and a number of serious, and potentially life-threatening, health complications, both acute and chronic.¹ Patients start experiencing symptoms in early childhood and are at risk for progressive organ damage, which translates into an estimated 20-year reduction in life expectancy compared with the general US population.² The recurring pain crises and frequent hospitalizations associated with SCD interfere with patients’ ability to complete schooling or maintain employment; these and other disruptions in normal activities greatly reduces quality of life for patients and their caregivers. The associated costs of SCD in the United States are substantial with an estimated annual $3 billion in direct medical costs alone.³ Nationwide surveillance is limited, but best estimates place the prevalence at approximately 100,000 individuals, mostly of African or Hispanic descent.⁴

Standard medical care for SCD is centered on treatments intended to prevent acute and chronic complications (eg, hydroxyurea, chronic blood transfusions) and medications for symptom management (pain medications).⁵ Other novel therapies, such as crizanlizumab, voxelotor, and l-glutamine, are intended to be disease-modifying but have been shown to not be cost-effective and have low use rates.^6,7 Individuals living with SCD in the United States continue to face systemic racism, disparities in access to care, and underfunding for treatment and research.⁸ Comprehensive medical, mental health, and social support services are also lacking in many regions. Allogeneic hematopoietic stem cell transplant (HSCT) is a curative treatment for SCD, but this option is limited by the availability of matched sibling/haploidentical donors and carries risks of graft-vs-host disease, graft rejection, infertility, and increased short-term risk of mortality.^9-11

Two emerging gene therapies for SCD, lovotibeglogene autotemcel (lovo-cel) from bluebird bio and exagamglogene autotemcel (exa-cel) from Vertex Pharmaceuticals and CRISPR Therapeutics, are anticipated to receive regulatory decisions by December 2023.^12,13 Both use an autologous HSCT approach in which a patient’s own stem cells are genetically modified ex vivo to inhibit sickling of the blood cells. Lovo-cel uses a lentiviral vector to insert a modified β-globin gene variant into cells, whereas exa-cel employs CRISPR gene editing technology to increase production of fetal hemoglobin.^14,15

The Institute for Clinical and Economic Review (ICER) evaluated lovo-cel and exa-cel as 1-time gene therapies for SCD. This report presents the summary of our systematic literature review and cost-effectiveness analysis and highlights the key policy recommendations discussed at the California Technology Assessment Forum (CTAF) public meeting on July 27, 2023. The ICER Final Evidence Report on Gene Therapies for SCD is available.

Summary of Findings

CLINICAL EFFECTIVENESS

We assessed the clinical effectiveness of lovo-cel and exa-cel separately because of variations in trial enrollment and outcome measurement. In both pivotal trials, the key outcome was the proportion of trial participants who remained free of vaso-occlusive events (VOEs) or VOCs. The definition of a VOC was more broadly inclusive compared with a VOE. Medical attention at a health care facility was a shared element in both severe outcomes.

LOVO-CEL

The single-arm pivotal phase 1/2 HGB-206 study assessed lovo-cel infusion in patients aged 12-50 years with severe SCD, who were defined as having the βS/βS, βS/β0, or βS/β+ genotype and a history of at least 4 severe VOEs in the past 24 months before enrollment.^14,16 Trial participants were also required to be intolerant to or unresponsive to hydroxyurea treatment, have some independence in daily activities, and not have an available matched HLA-identical sibling hematopoietic cell donor. A total of 36 participants were followed for a median of 20.9 months, of which 31 had sufficient follow-up time required for evaluation of the primary outcome; 30 (96.8%) achieved the primary outcome of being free of severe VOEs between 6 and 18 months postinfusion.^16,17 The median annual number of severe VOEs decreased from 3 at baseline to 0 at follow-up. Moreover, lovo-cel was found to reduce the average number of annualized hospital days and hospital admissions at 24 months. Study participants experienced improvements from baseline in health-related quality of life measures, with the exception of anxiety, and in work productivity measures.¹⁸ Lovo-cel also resulted in increases in total hemoglobin and modified adult hemoglobin (HbAT87Q) levels that were sustained through month 24.¹⁷

Adverse events in the HGB-206 trial, which included stomatitis, thrombocytopenia, and neutropenia, reflected complications commonly experienced by those with SCD or were consistent with expected side effects of stem cell collection and myeloablative conditioning. One death occurred in the trial but was determined to be due to the patient’s severe baseline SCD and unrelated to lovo-cel treatment.¹⁴ Additionally, 2 cases of persistent anemia were found to be related to coexisting α-thalassemia trait, which was then added as a study exclusion to ongoing studies.¹⁷ In an earlier cohort of the trial, 2 fatal cases of hematologic malignancies were thought to be related to the underlying baseline risks of SCD itself and the known risks of myeloablative conditioning chemotherapy used in conjunction with the gene therapy.^19-22

EXA-CEL

The pivotal trial of exa-cel is CLIMB-121, which is an ongoing single-arm phase 1/2/3 trial of 35 participants with severe SCD defined in this case as a minimum of 2 severe VOCs per year in the 2 years before enrollment. Efficacy and safety data from a prespecified interim analysis have been reported with a median follow-up of 11.6 months.²³

After exa-cel infusion, 16 (94%) of the 17 evaluable participants remained free of severe VOCs and no participants experienced in-patient hospitalization for severe VOCs for greater than 12 consecutive months.²³ Patients reported improvements in quality of life measures (eg, physical, social, emotional, and functional well-being) throughout 18 months of follow-up.²³ Mean hemoglobin levels increased from baseline and were maintained at a level greater than 11.0 g/dL throughout follow-up.²⁴ Patients also had clinically meaningful increases in fetal hemoglobin defined by an absolute increase to greater than 30%, which is a threshold that has been hypothesized to be a curative target.^24-27

Adverse events related to exa-cel treatment were reported by 12 (34.3%) patients, but, as with lovo-cel, many of the reported adverse events in the exa-cel trial were ascribed to baseline disease or the myeloablative conditioning regimen. There were no reported malignancies, although follow-up time is limited.²³ There was one death attributed to SARS-CoV-2 infection and potentially related to busulfan lung injury.²³ Lastly, 1 patient treated with exa-cel required therapeutic phlebotomy to manage high hemoglobin levels.

Uncertainties Because of Limitations in the Clinical Evidence. The small sample sizes and limited length of follow-up of the pivotal trials leads to uncertainty about the long-term safety and durability of benefit provided by lovo-cel and exa-cel. Additionally, the inclusion criteria of both studies required high baseline severe VOC rates, which may put the results at risk of regression to the mean and which, even if accurate, are only generalizable to a modest percentage of people living with SCD in the broader population. Although the striking hematologic response and reduction in short-term complications produced by both gene therapies makes it reasonable to assume a longer-term benefit of reduced chronic complications of SCD, the magnitude of this benefit remains uncertain.

These trials did not directly compare the gene therapies with the currently available curative treatment (allogenic HSCT). In addition, along with small sample sizes and short durations, trial eligibility criteria and primary outcomes were different enough that even qualitative comparisons of the 2 gene therapies to each other are premature.

Long-Term Cost-Effectiveness. We assessed the cost-effectiveness of lovo-cel and exa-cel gene therapies vs best standard of care for patients with severe SCD using identically structured decision analytic models. The population was adolescents and adults with severe SCD who lack donor options for HSCT or are too old for safe HSCT. Each treatment was compared with standard of care (hydroxyurea, red blood cell transfusions) over a lifetime horizon on outcomes including quality-adjusted life years (QALYs), life years (LYs) gained, equal-value LYs (evLYs), SCD complications avoided, VOCs avoided, and costs. Health sector and modified societal perspectives were taken as cobase cases. We used the ICER high-impact single and short-term therapies framework to conduct scenario analyses with optimistic and conservative assumptions regarding the magnitude and durability of the therapies’ benefits.²⁸

The models considered the incidence of acute complications (eg, stroke, myocardial infarction, acute kidney injury), chronic complications (eg, poststroke, kidney disease, chronic lung disease), and mortality events. Patients remained in the models until death based on age-specific mortality. In the absence of trial evidence on other complications, VOC data from the clinical effectiveness section were used to estimate treatment effects. Given that the existing data cannot distinguish between the clinical effectiveness of the 2 treatments, we modeled treatment success for both as identical, using the proportion of patients who were free of severe VOCs in the lovo-cel trial (96.8%) as the model input.

The models adopted 2 assumptions from the prior ICER report on β-thalassemia. First, a small number of patients (1.7%) were projected to experience early mortality in the initial cycle because of the acute risks of stem cell transplantation. Second, because of the uncertainty surrounding the long-term treatment effectiveness, it was assumed that after the seventh year, 0.27% of patients undergoing gene therapy would annually revert to the costs and outcomes associated with standard care. This represents a minimal and consistent “failure” rate of gene therapy. This assumption was derived from a clinical expert’s estimate, suggesting that approximately 10% of patients might potentially return to transfusion dependence in their lifetime after treatment. A detailed description of other key assumptions can be found in the full report.

Because the models used the same structure and inputs for safety and clinical effectiveness, results are identical for the 2 gene therapies. Results of the base-case analyses and the optimistic and/or conservative assumption scenario analyses are presented in Tables 1 and 2. Treatment with lovo-cel or exa-cel resulted in fewer VOCs and a greater number of QALYs, Lys, and evLYs. At a placeholder price of $2 million per treatment, an assumption based on public analyst comments, the incremental costs per QALY, evLY, and LY gained were $170,000-$220,000 from a health care perspective and $143,000-$185,000 from a modified societal perspective. Results were most sensitive to the cost of treating a VOC, gene therapy use benefit, and baseline number of VOCs.

TABLE 1.

Results and Incremental Cost-Effectiveness Ratios for the Base Case of Lovo-cel and Exa-cel vs Standard Care

Treatment	Treatment cost a	Total cost	VOCs	Life years	evLYs	QALYs
Health care system perspective
Lovo-cel or exa-cel	$2,000,000	$2,827,000	4.18	21.87	17.31	16.38
Standard of care	—	$1,490,000	119.26	15.80	9.44	9.44
Incremental cost-effectiveness ratios	—	—	$11,600	$220,000	$170,000	$193,000
Modified societal perspective
Lovo-cel or exa-cel	$2,000,000	$2,837,000	4.18	21.87	17.31	16.38
Standard of care	—	$1,714,000	119.26	15.80	9.44	9.44
Incremental cost-effectiveness ratios	—	—	$9,800	$185,000	$143,000	$162,000

^a Placeholder price of $2 million.

evLY = equal-value life-year; exa-cel = exagamglogene autotemcel; lovo-cel = lovotibeglogene autotemcel; QALY = quality-adjusted life-year; VOC = vaso-occlusive crisis.

TABLE 2.

Scenario Analysis Results for the Optimistic and Conservative Benefit Scenarios

Treatment	Comparator	Cost per QALY gained a	Cost per life year gained a	Cost per evLY gained a	Cost per VOC averted a
Health care system perspective
Base case	Standard of care	$193,000	$220,000	$170,000	$11,600
Optimistic	Standard of care	$138,000	$154,000	$125,000	$10,600
Conservative	Standard of care	$246,000	$261,000	$206,000	$12,300
Modified societal perspective
Base case	Standard of care	$162,000	$185,000	$143,000	$9,800
Optimistic	Standard of care	$114,000	$127,000	$103,000	$8,700
Conservative	Standard of care	$208,000	$221,000	$175,000	$10,400

^a Placeholder price of $2 million.

evLY = equal-value life-year; QALY = quality-adjusted life-year; VOC = vaso-occlusive crisis.

In the optimistic scenario analysis assuming greater gene therapy treatment effectiveness and no treatment waning, the incremental cost-effectiveness ratios of this analysis were close to or below the $150,000 threshold from both health care system and modified societal perspectives. The conservative scenario, assuming lower treatment effectiveness and lower use for gene therapy, yielded ratios approaching or exceeding the $200,000 threshold under both perspectives.

Key Uncertainties in the Modeling of Long-Term Cost-Effectiveness. As noted earlier, the longer-term impact of treatment with these gene therapies is not known, including the impact of reduced short-term complications on longer-term chronic complications and early mortality. The cost-effectiveness results are also subject to uncertainty regarding the ultimate launch price of these treatments. It should be noted that a large portion of the estimated cost savings associated with treatment were from avoidance of VOCs; if the real-world number of VOCs of patients at baseline is lower than in the clinical trials, or if the cost of VOCs is lower than we have estimated, the incremental cost-effectiveness ratios of gene therapies vs standard care could be considerably higher.

Policy Discussion

The CTAF convened on July 27, 2023, to publicly deliberate on the clinical effectiveness and cost-effectiveness of gene therapies for SCD. The CTAF is an independent appraisal committee composed of medical evidence experts, including clinicians, methodologists, and patient advocates. Their deliberations included input from clinical experts and patient representatives with SCD expertise and formal comments from manufacturers and the public.

Following the discussion, the CTAF panel members deliberated on key questions raised by ICER’s report. A majority of panelists (13-1) found that current evidence is adequate to demonstrate a net health benefit for lovo-cel when compared with standard of care. The same majority of panelists (13-1) found that current evidence is adequate to demonstrate a net health benefit for exa-cel when compared with standard of care. All panelists (14-0) found that current evidence is not adequate to distinguish the net health benefit between lovo-cel and exa-cel.

The CTAF also voted on “potential other benefits” and “contextual considerations” as part of a process intended to signal to policymakers whether there are important considerations when making judgments about the long-term value for money are not adequately captured in analyses of clinical and/or cost-effectiveness. They highlight several factors beyond the results of cost-effectiveness modeling that the Comparative Effectiveness Public Advisory Council panel felt were particularly important for judgments of overall long-term value for money. The results for these votes are shown in Table 3. Value votes were not taken at the CTAF Public Meeting because net prices for lovo-cel and exa-cel were not available.

TABLE 3.

Votes on Other Benefits and Contextual Considerations for Lovo-cel and Exa-cel

When making judgments of overall long-term value for money, what is the relative priority that should be given to any effective treatment for SCD, on the basis of the following contextual considerations:
Contextual consideration	Very low priority	Low priority	Average priority	High priority	Very high priority
Acuity of need for treatment of individual patients based on short-term risk of death or progression to permanent disability	0	2	3	4	5
Magnitude of the lifetime impact on individual patients of the condition being treated	0	0	0	1	13
What are the relative effects of exa-cel/lovo-cel vs standard of care on the following outcomes that inform judgments of the overall long-term value for money of lovo-cel/exa-cel?
Potential other benefit or disadvantage	Major negative effect	Minor negative effect	No difference	Minor positive effect	Major positive effect
Patients’ ability to achieve major life goals related to education, work, or family life	0	0	0	0	14
Caregivers’ quality of life and/or ability to achieve major life goals related to education, work, or family life	0	0	0	2	12
Patients’ ability to manage and sustain treatment given the complexity of regimen	0	0	0	4	10
Society’s goal of reducing health inequities	0	0	0	4	10

exa-cel = exagamglogene autotemcel; Lovo-cel = lovotibeglogene autotemcel; SCD = sickle cell disease.

Following the discussion of the evidence, a policy roundtable was convened to deliberate on how best to apply the evidence to the use of lovo-cel and exa-cel for SCD. The policy roundtable members included 2 patient advocates, 2 clinical experts, 2 payer representatives, 1 representative from bluebird bio, the manufacturer of lovo-cel, and 1 representative from Vertex Pharmaceuticals, the manufacturer of exa-cel. The full set of policy recommendations can be found in the Final Evidence Report on the ICER website.

Select key policy recommendations are as follows:

• 1)  Given that there is insufficient evidence at present to distinguish between the safety or effectiveness of lovo-cel and exa-cel, and that clinical experts see no clinical reasons to favor one of the therapies for certain patient subgroups, payers may consider negotiating a lower price by covering only 1 of the 2 therapies. However, payers considering this coverage approach should be aware of important access and patient preference issues that may outweigh the benefit of achieving a lower price.
• 2)  If the announced prices for lovo-cel and exa-cel align with expected patient benefits and are set toward the lower edge of their estimated cost-effectiveness ranges, payers should use the US Food and Drug Association label as the guide to coverage policy without narrowing coverage by including specific clinical trial restrictions unrelated to the likelihood of benefit from treatment.
• 3)
    Payers may wish to consider the following coverage criteria (full list here):

  • a.
      Diagnosis: It is reasonable for plans to include documentation of an eligible genotype in coverage criteria. Genotype corresponds well with degree of phenotypic severity.
  • b.
      Availability of HSCT: HSCT also offers a potential cure for SCD and has a far longer clinical track record. Clinical experts suggested that the new gene therapies would offer the advantage of avoiding immunosuppression over the longer term but that most clinicians today would view a sibling-matched HSCT as a reasonable option prior to considering gene therapy. Attestation that a patient does not have a willing matched donor should suffice for coverage of gene therapy.
  • c.
      Exclusion criteria: Within the list of exclusion criteria for entry into the clinical trials, clinical experts emphasized that history of stroke should not be included as an exclusion for insurance coverage. Patients with a history of stroke were excluded from the clinical trials most likely to reduce the risk of short-term adverse events that could be difficult to ascribe to treatment as opposed to the underlying condition, but clinical experts argued that these patients are at high risk for further strokes and, therefore, have substantial opportunity to benefit from gene therapy.