The Role of the Conditioning Regimen for Autologous and Ex Vivo Genetically Modified Hematopoietic Stem Cell-based Therapies: Recommendations from the ISCT Stem Cell Engineering Committee

Abstract

The advent of autologous gene modified cell therapies to treat monogenic disorders has been a major step forward for the field of hematopoietic stem cell transplantation (HCT) and cellular therapies. The need for disease-specific conditioning to enable these products to provide a potential cure has required extrapolation from experience in myeloablative and non-myeloablative HCT for these disorders. In this manuscript, we review the current datasets and clinical experience using different conditioning regimens for autologous gene therapies in hemoglobinopathies, metabolic and lysosomal disorders, inborn errors of immunity (IEI) and bone marrow failure (BMF) syndromes. The disease specific and unique conditioning requirements of each disorder are considered in order to achieve maximal benefit while minimizing associated toxicities. Standardized recommendations based on these data are made for each set of disorders to harmonize treatment. Future directions and the possibility of non-genotoxic conditioning regimens for autologous gene therapies are also discussed.

Introduction

There are decades of experience using allogeneic hematopoietic stem cell transplant (HCT) as a curative option for hemoglobinopathies, inborn errors of immunity (IEI), bone marrow failure (BMF) disorders, and some inborn errors of metabolism.^1–4 Although significant improvements have been made with regard to human leukocyte antigen (HLA) compatibility matching, graft-versus-host disease (GVHD) prophylaxis and treatment, management of other immune-mediated complications, and availability of HLA-matched donors; rapid and robust immune reconstitution and transplant-related morbidity and mortality remain noteworthy challenges.⁵ Genetically modified hematopoietic stem cell therapy (GM-HCT) using gene-modified autologous hematopoietic stem cells (HSCs) has evolved over the last 30 years as an alternative approach to circumvent the limitations of donor availability, risks of excessive regimen related toxicity, prolonged immune suppression and GVHD associated with allogeneic HCT.⁶ Ex vivo GM-HCT offers the potential for lifelong cure of a growing number of monogenic blood and metabolic diseases through the correction or replacement of the defective or absent gene transcribed protein in the progenitor populations of self-renewing, long-term repopulating HSCs.⁷ Landmark studies are provided in Table 1.

Table 1.

Landmark studies of allogeneic ex vivo genetically modified hematopoietic stem cell therapy.

Indication	Pathophysiology	Landmark Studies	Recommendation
Hemoglobinopathies	Intrinsic erythropoiesis defect	Hgb 206 – Lentiviral gene addition for sickle cell disease. Busulfan targeting AUC 82mg*hr/L in groups B & C (Kanter J NEJM 2021)40 Hgb 207 & 212 – Lentiviral gene addition for thalassemia. Busulfan targeting AUC 70mg*hr/L (Locatelli F NEJM 2021)45 CLIMB THAL-111 & CLIMB SCD-121 – CRSPR/CAS9 gene editing for thalassemia and sickle cell disease. Busulfan targeting AUC 74–90mg*hr/L (Frangoul H NEJM 2020, NEJM 2024 & Blood 200)17, 34, 46	Myeloablative regimen. Therapeutic drug monitoring using pop-PK models for busulfan targeting an AUC of 80–100mg*hr/L
Lysosomal storage and Metabolic Disorders	Disorders of enzymatic pathways leading to accumulation of toxic substrates	ALD102 – Lentiviral gene addition for X-linked CALD. Busuflan targeting AUC 70–85mg*hr/L & Cyclophosphamide 200mg/kg (Eichler F NEJM 2017)62 MLD – Lentiviral gene addition. Busulfan targeting AUC 67 – 85mg*hr/L (Fumagalli F Lancet 2022)63 MPS-1 – Lentiviral gene addition. Busulfan targeting AUC 85mg*hr/L. Fludarabine 40mg/m2 × 4 and Rituximab 375mg/m2 ×1 (Genter B NEJM 2021)61	Myeloablative regimen. Therapeutic drug monitoring using pop-PK models for busulfan targeting an AUC of 80–100mg*hr/L. Disease-specific immune ablation.
Inherited Bone Marrow Failure	Intrinsic defects of one or more lineages of hematopoiesis	RP-L102 – Lentiviral gene addition. No conditioning regimen due to competitive advantage of transduced HSCs (Rio P Nat Med 2019)28	Need to balance risk of underlying disease with need to engraft. Conditioning will be disease specific.
Inborn Errors of Immunity	Genetic lesion that affects one or multiple arms of the immune system	SCID – Lentiviral gene addition. ARTEMIS SCID Busulfan targeting 20mg*hr/L (Cowan M NEJM 2022). ADA SCID single dose Busulfan 4mg/kg (Kohn DB NEJM 2021). X-SCID Busulfan targeting 22mg*hr/L (Mamcarz E NEJM 2019)67 X-CGD – Lentiviral gene addition. Busulfan targeting 70–75mg*hr/L (Kohn DB Nat Med 2020)68 WAS – Lentiviral gene addition. Busulfan 4mg/kg/day with Fludarabine 40mg/m2/day. Anti-CD20 and/or Anti-CD52 for significant autoimmunity (Magnani A Nat Med 2022).70	Need to balance risk of underlying disease with need to engraft. For SCID, genotype dependent. For other non-radiosensitive PIDs likely higher Busulfan exposure needed.

The conditioning regimen is a critical component of allogeneic HCT. Chemotherapeutic and radiotherapeutic agents are employed to achieve depletion of endogenous hematopoietic stem and progenitor cells (HSPCs) residing in bone marrow niches to allow engraftment of donor HSPCs.⁸ In addition immune ablative agents are required to prevent rejection of allogeneic cells, and some form of either graft manipulation or post-transplant immune suppression is required to prevent GVHD. These post-transplant allo-immune mediated complications are not typically issues in autologous GM-HCT .⁹ The underlying disease and the donor HLA compatibility dictate the degree of myeloablation and immune ablation that must be achieved to support donor engraftment. The type of donor graft also influences conditioning choices and intensity as alloreactive donor lymphocytes can assist in elimination of residual host cells.

Conventional conditioning agents used routinely in allogeneic HCT are genotoxic and are associated with short- and long-term toxicities.^{10, 11} If these same conditioning regimens are used for GM-HCT, this could represent a major challenge to widespread acceptance of either HCT or GM-HCT for certain non-malignant diseases, particularly those for which other established less toxic standard-of-care disease modifying/ameliorating therapies exist.¹² Therefore, the goal of conditioning in the setting of gene therapy is to allow sufficient engraftment of gene-modified cells to achieve a therapeutic effect while minimizing exposure and regimen-related toxicities.

Three decades of cumulative clinical experience using GM-HCT to treat inherited monogenic disorders have demonstrated the potential for significant therapeutic benefit, making it an attractive alternative to allogeneic HCT in these settings.¹³ Early GM-HCT trials for X-linked severe combined immune deficiency (X-SCID) using γ-retroviral vectors were the first to demonstrate such benefits, but were plagued with a high incidence of T cell acute lymphocytic leukemia linked to oncogenic integration of the vector. These events led to major setbacks in the field, but ultimately advanced the development of safer, third-generation self-inactivating lentiviral vectors (LVs), which are currently the dominant vector of choice in gene addition trials.^14–16 In addition, alternative strategies achieving precision gene editing with programmable nucleases have been optimized and other modalities such as shRNA are being tested.^{17, 18} These approaches are aimed at either gene disruption, targeted correction of the defective endogenous gene, or modulation of endogenous regulatory elements in HSCs.¹⁹ The benefits of GM-HCT include no risk of rejection, no GVHD and no need to find a suitable allogeneic donor.^6,7 A different risk set of GM-HCT includes failure to achieve durable correction, potential for clonal changes and potential for leukemic transformation. We review the role of pre-transplant conditioning in autologous GM-HCT to enable sufficient long-term engraftment of gene-modified HSCs, promote expression and delivery of therapeutic gene product, and approaches to minimize toxicity related to the preparative regimen.

Hematopoietic stem cell gene therapy

Ex vivo GM-HCT involves isolation of autologous HSCs, followed by ex vivo genetic modification to introduce a functional copy of the defective gene (i.e., gene addition), correct an underlying genetic defect or disrupt a genetic element via gene editing approaches or use of inhibitory RNA to silence genetic expression.^{17, 18, 20} Following administration of the conditioning regimen, the ex vivo gene-modified HSCs are reinfused into the patient and home to bone marrow niches. Hematopoiesis is subsequently reconstituted with new blood cells that have the altered genetic expression, ameliorating or potentially eliminating disease pathology. At the time of this writing, there are approximately fifty GM-HCT active clinical trials registered on ClinicalTrials.gov, many in advanced phases, aimed at treating more than fifteen different monogenic disease indications. Commercially available GM-HCT products have been approved by the U.S. Food and Drug Administration (FDA) and/or the European Medicines Agency (EMA) for the treatment of adenosine deaminase SCID (Strimvelis^™), transfusion-dependent β-thalassemia (TDT; Betibeglogene autotemcel, exagamglogene automcel), metachromatic leukodystrophy (atidarsagene autotemcel), X-linked cerebral adrenoleukodystrophy (elivaldogene autotemcel) and sickle cell disease (SCD; exagamglogene automcel and lovotibeglogene autotemcel).

Conditioning for GM-HCT

The majority of agents used to achieve myeloablation and immune ablation are DNA alkylating or antimetabolite/nucleoside analog chemotherapeutic drugs (ex: busulfan, melphalan, cyclophosphamide, fludarabine etc.). These are associated with a variety of early and late toxicities owing to their relatively non-selective mechanisms of DNA damage-induced cytotoxicity.^{21, 22} Toxicities such as endothelial damage (sinusoidal obstructive syndrome, transplant-associated thrombotic microangiopathy, diffuse alveolar hemorrhage), infertility and treatment related malignancy are more pronounced in intensive myeloablative regimens such as high dose busulfan and cyclophosphamide.¹¹ Non-myeloablative and reduced intensity regimens have been developed in ongoing efforts to reduce treatment-related morbidity and mortality while still allowing acceptable levels of engraftment to achieve net therapeutic benefit.^23–25 In the case of a hematologic malignancy, the lower toxicity associated with the use of reduced intensity and non-myeloablative regimens must be balanced with an increased risk of post-HCT relapse.²⁶ However, in GM-HCT for the treatment of non-malignant diseases, the goal of conditioning is to allow for engraftment of sufficient numbers of gene-modified cells to allow elimination of disease symptoms while minimizing exposure to DNA damaging, genotoxic agents with attendant risks of short- and long-term toxicities. Minimizing these toxicities is particularly important as several IEI, hemoglobinopathies and BMF syndromes are intrinsically associated with an increased risk of organ damage and neoplasms.^{27, 28}

Disease-specific characteristics impact the degree of myeloablation and/or immune suppression required for phenotypic correction. These characteristics include the proportion of normal gene expression needed, any requirement for organ specific expression, and any survival or engraftment advantage for gene corrected populations. Favorable repopulation kinetics and survival benefit of gene-modified cells during hematopoiesis over defective endogenous cells have allowed the successful use of reduced-intensity regimens in GM-HCT clinical trials for IEI.¹² In contrast, despite requiring only about 30% –50% normal hemoglobin production, full myeloablation has been necessary in GM-HCT for hemoglobinopathies to achieve sufficient engraftment of genetically modified cells to correct the disease phenotype.²⁹ Myeloablation with alkylating agents has also been required for patients with some metabolic conditions in order to ablate resident immune cells in the central nervous system for new GM-HSC-derived microglia to engraft.³⁰ Finally, the decision to incorporate transient immune suppression to the conditioning regimen may depend on (i) the need to prevent development of transgene-specific immune responses that could eradicate the gene-modified cells,³¹ or (ii) disease-specific factors, e.g., need to remove autoreactive cells and cells at risk of lymphoproliferation, as in Wiskott-Aldrich syndrome (WAS).³²

Statement:

Conditioning for GM-HCT should be indication specific and include sufficient myeloablation and potentially immune ablation to allow sufficient engraftment of gene-modified HSCs while minimizing adverse side effects of these agents.

Considerations for selection of conditioning for ex vivo GM-HCT

Requirement for myeloablation

The unique biological characteristics of each disease must be considered when selecting an appropriate conditioning strategy for GM-HCT. Since the use of autologous cells theoretically eliminates the risk of graft rejection and GvHD complications, stable mixed chimerism of gene-modified and unmodified cells can be tolerated for some diseases and can be achieved using reduced-intensity regimens; however, these approaches are typically inadequate to achieve sufficiently high levels of engraftment of GM-HSC required to ameliorate disease phenotypes in hemoglobinopathies or lysosomal storage diseases.²⁹ The need for high level of GM-HSC engraftment is compounded by the fact that most genetic engineering methods are not very efficient and that genetically modified long-term HSCs likely decrease in number and proportion over time.^{6, 7, 29} The optimization of real-time PK adjusted and/or population (pop)-PK modeled conditioning agents in allogeneic HCT has shown that a fully myeloablative regimen can be used while minimizing associated toxicities.¹¹ Pop-PK modeled busulfan with a target of 80–100 mg*h/L is considered a myeloablative regimen in allogeneic HCT that reduces toxicity from higher exposure levels.^{10, 11} Similarly, pop-PK modeled fludarabine is associated with improved outcomes and robust immune reconstitution in allogeneic HCT for various disease settings.³³ PK-adjusted busulfan to reach pre-determined target exposures has been incorporated in recent GM-HCT trials with excellent and uniform effect.³⁴ Thus, real-time PK-adjusted and/or pop-PK modeled conditioning agents should be used when feasible to ensure appropriate ablation and minimize associated toxicities.

Hemoglobinopathies

Beta-thalassemia and sickle cell disease (SCD) are caused by defects in the HBB gene, resulting in either low, mutant or absent β-globin. Studies of engrafted donor-derived nucleated cells and circulating donor-derived red blood cells (RBCs) in SCD patients with stable mixed chimerism following allogeneic HCT suggest that 15–25% donor engraftment results in 73–90% donor RBCs and reversal of the sickle phenotype.^35–38 However in the setting of GM-HCT for SCD the ratio of non-sickle to sickle hemoglobin (HbS) produced within each gene-modified mature erythrocyte must also be considered. For lentiviral GM-HCT in SCD to be curative, sufficient vector copy number must be achieved within individual gene-modified HSCs to prevent sickling in red blood cells.²⁰ Similarly, for gene editing technologies sufficient efficiency is necessary to ameliorate clinical phenotype.¹⁷ The same challenges apply in the setting of GM-HCT for TDT, where heterocellular correction of autologous HSCs in the ex vivo gene therapy product requires high levels of engraftment to ensure sufficient numbers of gene-modified/gene-corrected long term- HSCs to restore normal erythropoiesis.³⁹

The largest GM-HCT clinical study performed to date for the treatment of SCD is HGB-206 sponsored by Bluebird Bio, Inc. (ClinicalTrials.gov Identifier: NCT02140554), a phase I/II study using a LV encoding β^A-T87Q-globin (BB305) to modify autologous CD34⁺ HSPCs transplanted under myeloablative busulfan conditioning.⁴⁰ The first seven patients (Group A) received a busulfan exposure target area under the curve (AUC) 65–74mg*hr/L and cell dose of ≥1.5 × 10⁶ CD34⁺/kg. Despite achieving durable expression, suboptimal HbA^T87Q concentrations of 0.51–1.17 g/dl prompted refinements to the protocol, including increased target AUC busulfan to 82mg*hr/L in Groups B and C without significant increase in unexpected toxicity⁴⁰

The majority of clinical experience in GM-HCT for TDT has been derived from the Bluebird Bio sponsored trials HGB-204, −205, −207, and −212, using the same BB305 product.^41–43 Twenty-two subjects were treated in HGB-204 and −205 (phase I/II). Subjects received myeloablative intravenous busulfan conditioning targeting an AUC of 59–82mg*hr/L.⁴¹ RBC transfusion requirements were reduced or eliminated in all 22 patients. Forty-one subjects have been treated in HGB-207 and −212 (phase III) and were conditioned with a similar myeloablative busulfan regimen.^{44, 45} Transfusion independence was achieved in 20 of 22 evaluable subjects from HGB-207, including 6 out 7 subjects under the age of 12.⁴⁵ There were no reports of unexpected increased toxicity attributed to busulfan exposure levels.⁴⁵

In addition to the gene-addition studies described above, gene-editing has also been used to treat hemoglobinopathies.¹⁷ Currently available data on gene-editing GM-HCT for TDT/SCD come primarily from the CLIMB THAL-111 and CLIMB SCD-121 trials, which use CRISPR-Cas9 to reactivate HbF production by editing the enhancer region of the BCL11A gene in autologous CD34⁺ HSPCs (DP termed CTX001^™).^{17, 46} Subjects received myeloablative intravenous busulfan targeting an AUC of 74–90 mg*hr/L. Outcomes have been similar to those reported with the lentiviral approach.^{34, 46} An additional study reports 3 patients that received a gene edited product with busulfan targeting an AUC of 80–100 mg*hr/L without any unexpected toxicity.⁴⁷ Other ongoing gene-editing GM-HCT trials, ST-400⁴⁸ and PRECIZN-1⁴⁹, have also incorporated myeloablative busulfan conditioning into their study designs.

In contrast to these studies, several lentiviral-based trials for hemoglobinopathies have used reduced intensity conditioning with either sub-myeloablative busulfan (39.8–59.7mg*hr/L) or melphalan.^50–52 In both instances, there was symptom improvement but cure was not achieved. This is likely due to a number of factors including an inability to express sufficient lentiviral based globin product. Thus, until gene engineered cell products are more efficient and able to induce higher levels of globin expression, a myeloablative regimen appears critical to providing cure for patients with TDT and/or SCD.

Interestingly, in SCD there have been reports of patients with cytopenias, MDS and/or AML after GM-HCT.^53–55 Studies showed that patients with SCD have a higher intrinsic rate of clonal hematopoiesis, potentially predisposing them to these types of complications.^56–58 Whether these events are caused by exposure to conditioning agents, prior disease directed treatment, ex vivo HSC manipulation or some combination thereof remains an area of active research.⁵⁹ Since clonal hematopoiesis increases with age, current clinical trials evaluating approved gene therapy products in younger patients (NCT05329649, NCT04293185) may help further minimize complications related to clonal hematopoiesis by allowing curative therapies to be delivered earlier. As GM-HCT for SCD becomes more widespread, careful monitoring of clonal hematopoiesis, hematologic toxicity and leukemogenesis will be critical in this patient population to help optimize conditioning regimens and GM-HCT processes to minimize these events.

Statement:

Myeloablative conditioning is likely necessary in GM-HCT for hemoglobinopathies to allow adequate gene modified HSC to engraft to provide a durable and sufficient effect. Therapeutic drug monitoring (TDM) using pop-PK models for busulfan to a target of 90 mg*hr/L (range: 80–100mg*hr/L) should be used to minimize conditioning related toxicities. Monitoring of clonal hematopoiesis and hematologic toxicities of GM-HCT in SCD will be critical for further optimizations.

Lysosomal Storage & Metabolic Disorders

One advantage of GM-HCT compared to traditional allogeneic HCT for lysosomal storage and/or metabolic disorders is the ability to deliver supra-physiologic enzyme levels in gene corrected cells.^{60, 61} Due to the need for high enzyme expression levels as well as the risk of an immune response to the enzyme that is produced, fully myeloablative and immune ablative regimens are generally used to optimize outcomes. An additional complicating factor for patients with lysosomal storage and/or metabolic disorders is the need for corrected enzyme producing cells to cross the blood brain barrier. The ALD102 trial for boys with X-linked cerebral adrenoleukodystrophy used a lentiviral GM-HCT conditioned with busulfan targeting AUC 70–85mg*hr/L and cyclophosphamide 200mg/kg. At time of interim analysis 15/17 patients were free of major functional disabilities.⁶² Similarly in metachromatic leukodystrophy, a lentiviral-based GM-HCT was given after busulfan conditioning. In this study the first 9 patients received submyeloablative busulfan targeting an AUC of 67.2mg*hr/L which was then increased to a myeloablative target AUC of 85.0mg*hr/L for the remainder of the cohort.⁶³ Though there were no reported increased toxicities due to the higher busulfan exposure target, there also was no statistically significant difference in transduced cell engraftment or arylsulfatase activity in peripheral blood mononuclear cells between the submyeloablative and myeloablative cohorts. Given the small numbers, the authors of the trial recommend a myeloablative regimen to maximize therapeutic effect.⁶³ At a median follow up of 3.16 years, 26/29 treated patients were alive with all patients maintaining engraftment of GM-HCT modified stem cells.⁶³ 5 patients developed anti-arylsulfatase-A antibodies that either spontaneously resolved or were treated with rituximab. For lysosomal storage disorders the enzyme delivered by GM-HCT can induce an immunological response that may be antibody or T-cell mediated. To prevent this, some trials (MPS-1) incorporated fludarabine and rituximab as a preventative strategy with good effect and minimal anti-enzyme antibodies detected.⁶¹

Statement:

Myeloablative conditioning (busulfan target: 90mg*h/L) is likely optimal for GM-HCT for lysosomal storage and/or metabolic disorders. Immune ablation is beneficial for many of these disorders to reduce the risk of developing B cell or T cell mediated anti-enzyme immunity (ex: busulfan/fludarabine +/− rituximab). Therapeutic drug monitoring (TDM) using population-PK models should be used.

Inborn Errors of Immunity and Bone Marrow Failure Syndromes

The requirement for myeloablation in IEI and/or BMF syndromes is more complex and disease specific. In many of these disorders (ex: Fanconi anemia (FA), ARTEMIS SCID etc.) the use of any myeloablation must be carefully balanced with the increased risks of toxicity from the underlying diseases. A phase II trial by Rocket Pharmaceuticals is using RP-L102, a lentiviral based gene therapy delivered to CD34+ HSCs for patients with Fanconi anemia due to FANCA mutations. The GM-HCT product is infused without conditioning given the survival benefit of gene corrected HSCs. In the 9 patients that were evaluable to date, 5 show increasing signs of engraftment over time indicating that this method has some success, but additional modifications may be necessary.^{28, 64} These modifications may include increased modified hematopoietic stem cell dose, earlier harvest of HSCs (which are particularly susceptible to dysfunction over time in FA) and/or the incorporation of some type of conditioning regimen.

Similarly, in patients with SCID, the genotype dictates susceptibility to myeloablative agents. In these patients, toxicity risks of conditioning must be balanced with the need to engraft all cell lineages including B cells. Patients with ARTEMIS and ADA deficient SCID have inferior survival after allo-HCT compared to other infants with SCID including increased toxicity from alkylating agents. In the lentiviral-based GM-HCT for ARTEMIS SCID sub-myeloablative busulfan targeting an AUC of 20mg*hr/L is used and facilitates engraftment of all lineages. In 5/6 evaluable patients GM T cell immune reconstitution was demonstrated by 12 months post-infusion. GM B cells were detected in all patients at a median of 6-weeks post-GM-HCT.²⁷ For patients with ADA SCID, several retroviral and lentiviral-based GM-HCT trials were conducted in the US and UK.^{65, 66} In the lentiviral trials sub-myeloablative busulfan (single dose 4mg/kg) was used to allow for sufficient B cell engraftment. Persistence of lentiviral transduced HSCs was demonstrated in 48/50 treated patients; 90% of US participants and 100% of UK participants were able to discontinue IgG replacement therapy and had rising numbers of T-lymphocytes by 24 and 36 months post-GM-HCT. Similarly in X-linked SCID lentiviral based GM-HCT a sub-myeloablative busulfan conditioning regimen targeting 22mg*hr/L was used.⁶⁷ At a median follow-up of 16 months, 7/8 patients had robust immune reconstitution and 4/8 were able to discontinue IVIG infusions.⁶⁷ These data show that a carefully selected busulfan exposure can balance the risk of toxicities and the goal of protective multilineage engraftment.

The allo-HCT experience with X-linked chronic granulomatous disease (X-CGD) and WAS, supports the use of myeloablative regimens to achieve sufficient and durable engraftment. In contrast, the GM-HCT studies for these disorders have used varied busulfan dosing. In the initial reports of lentiviral-based GM-HCT for X-CGD myeloablation with busulfan targeted an AUC = 70–75mg*hr/L, 6/7 evaluable patients met the primary endpoint at 12 months and were able to discontinue all CGD-related antibiotic prophylaxis.⁶⁸ Initial lentiviral-based GM-HCT for WAS incorporated busulfan conditioning (4mg/kg/day) with fludarabine (40mg/m²/day). Anti-CD20 and/or anti-CD52 was added for patients with significant autoimmunity. In this study 6/7 patients were alive at a median of 27 months follow-up. All had resolution of eczema, autoimmunity and immune deficiency, and 3/6 were able to discontinue prophylactic antibiotics.⁶⁹ Recent long-term follow up confirmed the benefit of this approach and provided data on an additional 2 patients that received this therapy, though the continued thrombocytopenia should be further investigated.⁷⁰

Statement:

The need for myeloablation in IEI and BMF is disease and genotype specific. Exposure should carefully balance risks of increased toxicities for certain disorders. It is currently unclear what the optimal myeloablation is for patients with IEI . For diseases with DNA repair issues this is likely around a busulfan AUC of 20mg*h/L, for other diseases it is less clear, but likely higher. For some BMF disorders such as FA, conditioning is not being used though other BMF disorders likely need some degree of myeloablation.

Future Perspectives

As the field of GM-HCT evolves, one of the major hurdles remaining is overcoming the need for genotoxic conditioning regimens. To that goal, several new technologies are being piloted to provide myelodepletion and/or ablation without genotoxicity. The most advanced of these technologies are agents that target CD117, a receptor tyrosine kinase expressed on HSCs that interacts with stem cell factor (SCF) and signals for HSCs to remain in the perivascular niche.

Briquilimab^® (JSP191), is a humanized monoclonal anti-CD117 antibody that interrupts the interactions of CD117 and SCF.⁷¹ It is currently being studied as a myelodepleting conditioning agent for allogeneic transplantation in SCID (NCT02963064), BMFs (NCT04784052), hemoglobinopathies (NCT05357482), IEIs (NCT05907746, NCT05600907) and myeloid malignancies (NCT05903274, NCT04429191).⁷² A prior study using MGTA117 (NCT05223699), an antibody-drug conjugate (ADC) that similarly targets CD117 with an amantinin payload was terminated due to toxicity in a phase 1 dose escalation trial in adult patients with transplant ineligible acute myeloid leukemia or myelodysplastic syndrome with excess blasts. More refined CD117 ADCs are currently being developed.⁷³ CD45 targeting ADCs are also in development to target autoimmune, autoinflammatory and other disorders.⁷⁴ The use of these precision biologics will transform the field of GM-HCT; however the path for development will be complicated. PK models will need to be developed in order to ascertain optimal use. The effects of target antigen expression on non-HSCs will also need to be understood (ex: CD117 is expressed in the gonads). Thus, while these products may ameliorate serious complications associated with genotoxic myeloablative regimens, they may also present their own risks of side effects.

As the role of non-genotoxic antibody-based conditioning continues to be refined, other optimizations can also be explored. Recent data has shown the critical importance of adequate fludarabine exposure in allogeneic HCT as well as in CAR-T cell efficacy.^{33, 75, 76} Thus, extrapolating these data and identifying optimal exposures to fludarabine in GM-HCT that require immune ablation may improve engraftment. Establishing optimal exposure levels for other conditioning agents like melphalan may increase options for preparative regimens that reduce associated toxicities without compromising GM-HCT engraftment and cure.

Summary Statement:

Conditioning for GM-HCT needs to be tailored by the underlying diagnosis to enable engraftment that will provide cures. Precision PK adjusted or population-PK modeled agents should be used whenever possible to maximize engraftment while minimizing associated toxicities. Newer targeted biologics will allow for continued refinement of GM-HCT while minimizing genotoxic conditioning associated side effects. Until we get to that point, use of busulfan with TDM using population-PK models should be considered standard of care (SOC). For hemoglobinopathies and most metabolic diseases myeloablative exposure 90mg*h/L is SOC. For IEI and BMF the optimal busulfan exposure level is still being identified and is likely different for each specific disorder.