An ethical allocation scheme for scarce gene therapies in sickle cell disease and transfusion-dependent β-thalassemia Open Access

Novel gene therapies (GTs) are among the most exciting recent clinical advancements, offering functional cures for inherited diseases. As of January 2025, 10 conditions have approved GTs.¹ Among these, GTs for hemoglobinopathies such as sickle cell disease (SCD) and transfusion-dependent β-thalassemia (TDT) have garnered considerable attention for their potential to transform the lives of individuals with these highly morbid blood disorders.² 

SCD arises from a β-globin gene mutation that causes red blood cells (RBCs) to become misshapen upon deoxygenation, triggering painful vaso-occlusive crises (VOCs), progressive organ damage, and chronic hemolytic anemia.³ TDT stems from β-globin gene mutations impairing RBC production, leading to severe anemia, extramedullary hematopoiesis, and lifelong transfusion dependence.⁴ 

Noncurative treatments for SCD include chronic use of hydroxyurea, l-glutamine, crizanlizumab, voxelotor (recently withdrawn), RBC transfusions, iron chelation, and pain management.⁵ Treatment-related symptoms range from cytopenia, gastrointestinal problems, and headaches to serious complications such as venous thrombosis and hepatic or renal injury. TDT management involves regular RBC transfusions, iron chelation, pain control, and sometimes luspatercept or splenectomy, with risks of alloimmunization, infection, and multiorgan damage from iron overload and chronic hypoxia.⁵ These therapies often provide incomplete symptom relief and may not prevent substantial reductions in quality of life and survival. The median life expectancy is ∼53 years for SCD⁶ and 37 to 45 years for TDT.^7,8 

Allogeneic hematopoietic cell transplantation (HCT), a curative option for SCD and TDT, replaces impaired hematopoietic machinery with donor stem cells.⁵ Often pursued in younger patients to avert organ failure, HCT restores competent hemoglobin production and can prevent or even reverse complications such as renal hyperfiltration,⁹ pulmonary hypertension,¹⁰ and cerebral vascular abnormalities.^11,12 Although HCT improves overall survival, including with haploidentical grafts,¹³ it is limited by matched sibling donor scarcity and variable outcomes with alternative donors (3-year overall survival is as follows: haploidentical, 87%; matched unrelated, 81%; mismatched unrelated, 82%; vs matched siblings, 96%).¹⁴ Moreover, complications, such as veno-occlusive disease, graft rejection, severe infections, and graft-versus-host disease, can, in rare cases, produce outcomes worse than no HCT. Recent advances have lowered risks with alternative donors,¹⁵ but HCT remains a demanding, high-risk/high-reward therapy.

GTs now offer functionally curative alternatives to HCT, although long-term, randomized efficacy and safety data and data on reversal of organ dysfunction are lacking. Currently, 2 companies produce commercially available GT products for both SCD and TDT. These therapies represent significant improvements in symptoms and transfusion burden previously unattainable outside of successful HCT.^16-21 

Two GT products, lovotibeglogene autotemcel (Lyfgenia) and exagamglogene autotemcel (Casgevy), are approved by the US Food and Drug Administration (FDA) for SCD treatment. Lyfgenia involves transplanting gene-modified autologous hematopoietic stem cells transduced with a lentiviral vector encoding a modified β-globin gene (β^AT87Q), which sterically inhibits sickle hemoglobin polymerization. In trials, lovotibeglogene autotemcel demonstrated sustained hemoglobin (Hb)-AT87Q production, reduced hemolysis, and near-complete resolution of VOCs.^16,17 Casgevy, which uses CRISPR-Cas9 gene editing to upregulate fetal hemoglobin synthesis by targeting the BCL11A erythroid-specific enhancer, eliminated VOCs in 97% of patients for at least 12 months.¹⁸ 

Betibeglogene autotemcel (Zynteglo), approved for TDT, is manufactured similarly to Lyfgenia. It demonstrated transfusion independence in 91% of patients with non-β⁰/β⁰ genotype TDT.¹⁹ Preliminary data in β⁰/β⁰ genotype TDT showed promising early results, with 3 of 4 patients achieving transfusion independence.²⁰ Casgevy is also FDA approved for TDT (β⁰/β⁰, β⁰/β⁰-like, and non-β⁰/β⁰–like genotypes) with 91% transfusion independence.

A key barrier to GT is the time-consuming and resource-intensive manufacturing process. Before stem cell collection, patients with SCD must receive RBC exchange transfusions to achieve a hemoglobin S level <30% for at least 2 months. Patients with both SCD and TDT then begin monthly collection cycles of 2 to 3 consecutive days. Patients with SCD undergo a median of 2 collection cycles,¹⁸ whereas those with TDT undergo a median of 1 cycle.^19,21 The time frame from the end of collection to infusion takes up to 6 months due to mandated testing and quality control.²² As we discuss next, production and other barriers conspire to create an imbalance in supply and demand, necessitating careful consideration of GT allocation.

Anticipated production limitations and logistical challenges in delivering GTs raise significant concerns about access and equity. Reported production estimates suggest a capacity of 85 to 105 GTs annually for either SCD or TDT in the United States per company²³ or ∼200 infusions per year across both diseases. Individual programs also face constraints based on bed availability; for example, at Dana-Farber Cancer Institute/Brigham and Women's Hospital, we expect to be able to deliver a maximum of 1 to 2 GT infusions per month (12-24 per year) for SCD and TDT combined.

Demand for GTs is elastic, but demand is expected to significantly outstrip supply. The estimated prevalence of SCD in the United States is >125 000 individuals, with >25 000 individuals experiencing ≥2 VOCs annually and >15 000 estimated to be eligible for GT.^24-27 The prevalence of β-thalassemia in the United States is estimated to be >2500 individuals, with >1500 having TDT and >800 estimated to be eligible for GT.^28-30 This high demand is reflected in our own experience, with our center seeing >5 consults per month for GT since mid-2024.

Multiple barriers compound challenges in GT access, threatening to exacerbate existing health inequities. These include significant financial, geographic, and sociocultural factors, all of which may determine who receives these therapies first or at all.

These GT products cost over $2 million per treatment, the highest of any approved therapies, creating a new order of scale for financial barriers. Zynteglo is priced at $2.8 million, Casgevy at $2.2 million for both SCD and TDT, and Lyfgenia at $3.1 million.^31,32 Approximately 50% of individuals with SCD in the United States are covered by Medicare or Medicaid. Under the 2025 Medicare Physician Fee Schedule Inpatient Prospective Payment System, the new technology add-on payment will only cover 75% of the difference between the autologous stem cell transplant or cell therapy diagnosis–related group bundled payment and the actual cost of the therapy and related HCT-level care.^27,31,33 However, for patients on Medicaid in states choosing to participate in the Centers for Medicare & Medicaid Services Cell and Gene Therapy Access Model, the cost of GT product for SCD can be carved out of the inpatient payment bundle and covered through the Medicaid Drug Rebate Program.³⁴ This would result in costs, and potential center losses, roughly equivalent to autologous stem cell transplant, while also limiting patient cost sharing.

Given the evolving GT payment landscape, many programs may incur a financial loss for each Medicare, Medicaid, or dual-enrolled patient treated, disproportionately affecting programs with higher proportions of these patients. This may require some programs to prioritize patients with private insurance to maintain solvency; however, even approval with private payers is not predictable, because prolonged prior authorization processes with variable eligibility criteria have delayed multiple GT collections at our center, also wasting manufacturing slots. Moreover, uninsured or underinsured patients may face insurmountable obstacles in accessing these life-altering treatments. Although cost is not a clinically relevant factor in allocation and feels morally at odds with the goals of medicine, the necessity of insurance approval for coverage of these expensive GTs represents a de facto consideration of cost.

The maldistribution of treatment centers relative to the affected population severely limits GT access nationally and globally. Within high-income countries such as the United States, GT programs are predominantly concentrated in specific regions, frequently overlapping with National Cancer Institute–designated Cancer Centers where HCT/GT expertise is centralized.^35,36 This restricts access for patients in rural and underserved US areas, particularly in the Midwest, Northwest, South, Southeast, and Southwest. Moreover, within large urban areas, GT access is often not found at facilities where many patients (particularly those with SCD) tend to receive routine care, such as Federally Qualified Health Centers and disproportionate share hospitals. Traveling long distances inevitably imposes additional financial burdens and logistical challenges, limiting access for individuals with more limited means. Institutions with less HCT/GT expertise may be less comfortable identifying patients most likely to benefit, have less established pathways for referral to GT programs, and possess fewer resources to help patients and communities obtain long-distance care, leading to missed opportunities for treatment.

Limited access is even more striking globally, with approved GT programs existing only in a few specific locations, primarily in Western countries. Although regulatory approvals for GTs for SCD and TDT have been granted in the United States, European Union, Britain, Bahrain, and Saudi Arabia; large parts of Asia and Africa, regions with some of the world's highest prevalence of SCD and TDT, are entirely excluded.³⁷ In places where such therapies receive approval, reliable mechanisms for cost coverage of GT products and related processes (eg, cell collections, exchange transfusions, product infusion, and posttherapy follow-up) are variable, creating disparities both within and between countries. Bringing GT to these high-burden and historically overlooked areas will require extensive infrastructure development, including HCT programs, cell processing facilities, and expert personnel. It is worth acknowledging that GT program infrastructure, which may only serve a relatively small proportion of a country's population, will compete directly with other community-specific health care priorities, likely leading to deprioritization, particularly within many developing nations or rural regions. This means these geographic challenges in delivery will be an ongoing concern. For now, clinical trials and international travel (which rely on patient or government resources) may be the only pathways through which these therapies become available to individuals in many low- and middle-income countries.

Language barriers, cultural beliefs, and mistrust of the health care system can significantly hinder access and uptake of GT.³⁸ For instance, the complicated nature of GT may hinder adequate understanding and contextualization for patients with limited English proficiency, even with the support of an interpreter, and inadvertently reduce their interest in pursuing GT. Cultural beliefs about medical interventions, particularly surrounding genome editing, might also influence some patients’ decisions. Additionally, historical and ongoing experiences of discrimination can lead to mistrust of the health care system, particularly among marginalized communities. Patients may lack the resources or support to navigate complex health care systems and advocate for access to GTs. These systemic challenges, rooted in historical and societal inequities, further disadvantage marginalized communities.³⁹ 

Access to GTs for SCD and TDT includes an idiosyncratic allocation problem: GT production and institutional capacity are shared between the diseases. Each company makes products for both conditions. Each program also has limited capacity to deliver GT, irrespective of the condition treated. Thus, requesting a product for SCD may decrease the capacity to treat TDT, and vice versa. Although both are inherited blood disorders with limited functionally curative treatment options, differences in their prevalence, clinical manifestations, survival, and systemic inequities of their affected populations, as well as price differences between Zynteglo and Lyfgenia (despite having the same manufacturing process and product), create additional ethical challenges in GT allocation. This shared resource demands careful consideration of patient- and population-level needs.

SCD, the most common inherited blood disorder in both the United States and the world, primarily affects historically marginalized Black and Brown individuals and communities with genetic ancestry from sub-Saharan Africa, the Middle East, and South Asia.⁴⁰ This is compounded by the debilitating nature of sickle cell VOCs, which frequently affect education and employment and lead to high health care utilization, including recurrent, prolonged hospitalizations. Reports of frequent mistreatment within the health care system, particularly through minimization of pain symptoms by physicians and other clinicians, coupled with the long-delayed development of therapies for SCD, point to a manifestation of structural racism.^39,41 These factors impose a substantial burden on patients, their communities, and the health care system. Conversely, TDT, which mostly affects individuals with Mediterranean, Middle Eastern, North African, Southeast Asian, and South Asian heritage,⁴² is substantially less prevalent but brings distinct challenges. Patients with TDT also experience significant health system challenges, including the need for regular transfusions, which can lead to alloimmunization and difficulties in finding compatible blood.⁴ Iron overload from these transfusions can cause damage to multiple organs, requiring complex chelation therapy and monitoring. Failure to receive regular transfusions can lead to extensive extramedullary hematopoiesis, related complications, and other late sequelae. Moreover, patients with TDT may face challenges in accessing specialized care, particularly in regions with limited expertise in managing this condition.

Individual centers face considerable challenges in navigating these complex factors, along with workforce capacity and financial viability. Maintaining clinical expertise and administrative experience in both SCD and TDT disease management and GT delivery are crucial. Moreover, a significantly higher proportion of patients with TDT (∼80%) have commercial insurance than patients with SCD (∼50%), introducing potentially worrisome incentives.^27,43 A core challenge lies in ensuring the needs of the smaller TDT population are not overshadowed by the larger SCD population, while simultaneously maximizing access to GTs for historically underserved individuals with SCD.

The confluence of inherent challenges, including the severe imbalance between GT supply and demand, production and delivery capacity being shared between SCD and TDT, the lack of curative alternatives for those without an HCT donor, the risk for progressive organ failure and death, and the scarcity of other therapeutic options, makes this moment distinct from other situations of scarcity in hematology, such as blood products or chimeric antigen receptor T-cell therapies. SCD's disproportionate impact on historically marginalized communities and the ongoing injustices of systemic racism demand a transparent pathway to GT, one that incorporates elements of input and deployment equity.⁴⁴ The consequence of inaction is a first-come, first-served approach that will disproportionately favor those with greater access and deepen existing inequities. Thus, every program that delivers these therapies should strongly consider proactively developing an ethically justified framework for GT prioritization that carefully considers patient health, risks, and need through a transparent, inclusive, and responsive process that addresses present challenges and can adapt as the situation evolves.

To address these challenges, we developed a systematic framework for GT allocation at our institution. The process was guided by a thorough review of clinical trial eligibility, FDA approval language, treatment outcomes, safety findings, and ethically justifiable approaches to allocation. The overall approach included the established ethical principles of professionalism, beneficence, nonmaleficence, stewardship, dignity, trust, respect for autonomy, distributive justice, input and deployment equity, and transparency in communication. From these, a set of prioritization principles were considered, including treating people equally (lottery or first-come, first-served), maximizing benefits (saving the most lives or years of life), prioritizing the worst off (those who are sickest, those who are youngest), and promoting or rewarding past or future action (those who suffered the most in the past and those who may do the most good in the future).

Complementing these prioritization principles was a structured process for developing the allocation framework, adhering to the accountability for reasonableness approach, which demands that any allocation framework aim to be fair, transparent, and accountable to promote trust and legitimacy in the decision-making process.⁴⁵ Accountability for reasonableness centers on the relevance, publicity, revision, empowerment, and enforcement of these processes. This approach guided the construction and enactment of the allocation framework by the multidisciplinary Gene Therapy Allocation Committee (GTAC), which includes clinicians, ethicists, social workers, patient advocates, pharmacists, nurse navigators, and other relevant parties. A GTAC working group (a subset of the larger GTAC), consisting of all GT clinicians, SCD clinicians, TDT clinicians, nurse navigators, GT physicians, social workers, and a GT administrator, is involved in the ongoing assessment and prioritization of patients on the waitlist, as well as in collecting data on the allocation process and patient outcomes. This working group was put into place to mitigate the risk of bias and duty-to-care issues that may arise when decisions are made solely by the treating GT physician. The entire GTAC met during framework development and will continue to meet quarterly to assess prioritization decisions, outcomes, need for revisions, and changes to relevant contextual features.

Our initial framework comprises several critical components: eligibility criteria, prioritization criteria, implementation plan, and engagement plan. This framework aims to address the complex ethical considerations surrounding GT allocation, considering the nuances of disease severity, potential benefits, patient safety, logistical constraints, the specific indications of each approved therapy, the limited availability and high cost of these treatments, and the ethical principles outlined above. It also aims to mitigate the potential for inequitable access due to the extreme scarcity of these therapies, while acknowledging that equity in access of any institutionally based allocation process is limited by those with access to the institution. Shared decision-making will be emphasized, empowering patients with information and support services throughout the process.

Our eligibility criteria aim to balance broadening access beyond the stringent confines of clinical trials, while ensuring patient safety. We combined elements from FDA approval criteria with specific exclusions for patients at high risk for complications from myeloablative conditioning. This approach results in criteria that are broader than clinical trial eligibility but narrower than FDA approval criteria (supplemental Figure 1). Tables 1 and 2 summarize the specific inclusion and exclusion criteria for SCD and TDT, along with justifications for these criteria.

After careful deliberation, we decided on the following lexically ranked prioritization criteria (Table 3):

1. Proportional equality: patients are prioritized in a 3:1 ratio of SCD to TDT.

2. Modified sickest first: patients with impending organ failure.

3. Lack of alternative therapy: patients without a matched donor for allogeneic HCT.

4. Lottery: random sequencing of multiple patients within the same priority category.

Other potentially relevant principles, such as highest disease burden, age, longest disease duration, and saving more life-years, were considered but not included due to limitations in current metrics and their inconsistency with the overall goals of promoting distributive justice, equity, need, and avoiding implicit bias during prioritization. Specifically, we excluded “highest disease burden” because quantifying disease severity with reliable, standardized, objective measures could lead to systematic exclusions of patients with less documented disease burden, particularly outside referrals to our institution. Moreover, there is no evidence to support that patients with more severe complications, beyond meeting the eligibility criteria and outside of impending organ failure, stand to benefit more from GTs. Similarly, we excluded age and longest disease duration because these were competing priorities with equally justifiable reasons; promoting those who would have the longest to live with a functional cure vs those who had already suffered with the disease for more years. Finally, given that GTs have no data that demonstrate efficacy in directly improving survival or reversing existing organ damage, there was no way to establish how to save more life-years.

We chose not to solely rely on a lottery system due to the expected bottleneck in treatment availability, which might prevent patients with impending organ failure from receiving GT in a timely manner. In addition, a lottery system would minimize opportunities to develop expertise in treating TDT with GT. There are many other potentially valid approaches to prioritization, and we strongly encourage other centers to engage in a similar deliberative process to develop allocation frameworks tailored to their unique circumstances and patient populations. As outcome data evolve, these principles and their application will be reassessed through the following implementation plan.

The prioritization criteria translate into a 5-tier prioritization list that is then sorted into a 3:1 ratio of SCD:TDT to determine the sequence of patients who will receive GT (supplemental Table 2). This list is updated monthly by the GTAC.

This includes handling internal and external referrals, record acquisition, obtaining insurance authorization for consultation, and timely scheduling with an HCT/GT specialist. Only formally referred patients are evaluated for eligibility and prioritization. The consultation visit will discuss both HCT and GT options within the context of all treatment modalities for SCD or TDT and initiate a donor search if appropriate. The institution will continue to perform outreach to communities and practices in its National Cancer Institute–designated catchment area, especially historically underserved communities, to maximize equitable referral and mitigate mistrust.

All patients eligible for GT based on the eligibility criteria, and who have been ruled out for allogeneic HCT by donor availability or patient preference, will be placed on the GT waitlist, discussed in recurring planning meetings, and prioritized based on the prioritization criteria. GT consultations and waitlists are reviewed on a recurring basis by the GTAC working group to ensure all patients who are eligible are included and to maintain an up-to-date priority list. Patients in the highest priority tier will be reviewed, and treatment urgency will be discussed among them. If there is no consensus decision to select 1 patient due to impending organ failure–related urgency, a lottery will be implemented among patients in the highest priority tier to determine the next patient to receive therapy. Patients on the waitlist are regularly monitored by their SCD/TDT team at the GT program or at the referring center to reassess disease burden and progression and recalculate their prioritization category. GT slots will be allocated in a 3:1 ratio between SCD and TDT. Patients eligible for clinical trials or other functionally equivalent alternative therapies will be discussed among the GTAC working group and managed transparently. Patients will be informed of the waitlist at the time of consultation and their estimated place in the prioritization on a rolling basis. Once a patient is selected from the prioritization process, they will be submitted for insurance approval of GT collection.

If conflicts in eligibility or prioritization arise that cannot be resolved within the GTAC working group, an independent non-GT transplant physician will be involved. If the referring physicians, providers, or patients have concerns about prioritization process or extended wait times for GT collection, they can contact the GTAC for a full committee review, with the potential for an individual patient revision of prioritization.

The expected timeline from referral to infusion is ∼14 months for patients with SCD (supplemental Figure 2) and 11 months for patients with TDT (supplemental Figure 3), not including waitlist time. This lengthy process involves eligibility testing, vein check and line placement, exchange transfusions (for patients with SCD not already receiving them), cell collection, manufacturing, quality control, conditioning, infusion, and engraftment.

Key partners in this process include TDT and SCD specialists, transplant specialists, cell manufacturing and processing specialists, ethicists, social workers, patients and patient advocates, and pharmacists. Patients’ cultures and wishes will influence the allocation of these therapies. We are currently discussing formal collaboration with other institutions to develop a unified process for referral and allocation.

Our eligibility and prioritization frameworks will need to be periodically updated because more data are being collected, treatable populations, patient perceptions, and preferences are changing, cost and reimbursement models are evolving, and more programs are gaining access to GT programs. Specifically, data tracking is needed for time from consult to listing to collection and infusion, clinical outcomes, and reasons for patient delays and/or delisting. Quarterly meetings are planned to discuss ethical prioritization scheme revisions.

An ethical analysis is needed at every institution and program, in every country, that treats patients with commercially available GTs or clinical trial products. We anticipate that individual programs will develop differing eligibility and prioritization criteria based on their unique region- or center-specific circumstances, such as different disease prevalence, catchment areas, patient travel times, payer mix, and reimbursement models. Explicit public reporting of eligibility and prioritization schemes will increase transparency and fairness and may prevent future gamesmanship by more well-resourced patients (a well-documented challenge in other scarce-resource situations such as solid organ transplant lists).⁴⁶ As more programs implement such ethically driven prioritization processes, collaboration between centers may develop, creating a path toward nationwide guidelines. Moreover, programs must explicitly consider the ethical implications of how clinical trial availability, eligibility, and referral processes will prioritize patients, because participation in clinical trials offers a potential pathway for treatment, particularly in regions with limited access to commercially available GTs or where the financial components of treatment are more challenging. Prioritization schemes must consider both the benefits (expanded number of GT slots and reduced dependence on complex reimbursement models) and risks (unintentional coercion into receiving an experimental product and variable efforts to obtain regulatory approval in low- and middle-income countries after clinical trial completion)⁴⁷ of clinical trial involvement.

Whenever transformative therapies emerge in medicine, they open avenues of both hope and the risk of inequity, which, if left unaddressed, can easily become entrenched. These inequities must be anticipated and mitigated by transparent, ethically grounded processes. GT for SCD and TDT exemplifies this dilemma: the potential of these therapies is profound, yet supply constraints, high costs, and uneven geographic availability threaten to widen existing gaps in care.

The stark mismatch between the modest number of GT doses available and the large, diverse pool of eligible patients, particularly in SCD, in which prevalence is high, yet geographic and sociocultural barriers exacerbate access challenges, creates a moral imperative for equitable allocation. Further complicating matters is the shared demand for the same GT products by patients with TDT, who have different mix of payers and clinical presentations. These conditions demand a structured management process. Our multidisciplinary approach, guided by principles of equity, accountability, and transparency, culminated in a plan that sets out core eligibility and prioritization criteria using a combination of proportional equality, clinical urgency due to impending organ failure, lack of alternative curative options, and a lottery, and is being implemented attendant to referral pathways, waitlist management, and conflict resolution. We encourage other programs to adopt a similarly structured approach. Each program’s specific circumstances, such as patient demographics and local health system realities, will shape the details of their scheme, but a deliberate, ethically justifiable process can guard against implicit or systemic biases.

However, even the most carefully conceived plans have limitations. At the program level, referral is not controlled, and there are known disparities in who is comfortable seeking an evaluation for GT as well as who has the financial and social resources to travel and remain engaged over the lengthy treatment timeline. Consequently, even with a prioritization scheme that seeks to promote equity, those with more means inevitably retain an advantage in accessing and completing GT. As more safety and efficacy data emerge, including quality-of-life outcomes, prevention of end-organ damage, and survival, this prioritization plan will require regular reassessment and possible recalibration. Evolving patient and community preferences, competing or complementary therapies, and changes in regulatory or reimbursement landscapes will similarly warrant ongoing monitoring. Ultimately, preemptive attention to fairness and equity in GT delivery is not merely best practice, it is the only path forward if these remarkable scientific breakthroughs are to fulfill their potential without perpetuating or exacerbating disparities in care.

The authors thank the members of the Gene Therapy Allocation Committee for their contributions to the framework and their ongoing work on the behalf of patients with sickle cell disease and transfusion-dependent β-thalassemia.

A.H.K. is supported by a 2024 American Society for Transplantation and Cellular Therapy New Investigator Award. A.H. is supported by the National Cancer Institute of the National Institutes of Health (award number K08 CA273043) and the American Society of Clinical Oncology under a Career Development Award.

Contribution: A.H.K. and A.H. contributed to the methodology; A.H.K. wrote the original draft and visualized the study; M.O.A. and A.H. supervised the study; and all authors conceptualized the study and reviewed and edited the manuscript.

Conflict-of-interest disclosure: A.H. reports personal fees from Bristol Myers Squibb for advisory boards on clinical trial equity. M.O.A. reports clinical trial support from Pfizer and consultancy fees from Vertex Pharmaceuticals. A.H.K. declares no competing financial interests.

Correspondence: Amar H. Kelkar, Department of Medical Oncology, Dana-Farber Cancer Institute, 450 Brookline Ave, Boston, MA 02215; email: amarh_kelkar@dfci.harvard.edu.