In 2016, the European Hematology Association (EHA) published the EHA Roadmap for European
Hematology Research
1
aiming to highlight achievements in the diagnostics and treatment of blood disorders,
and to better inform European policy makers and other stakeholders about the urgent
clinical and scientific needs and priorities in the field of hematology. Each section
was coordinated by 1 to 2 section editors who were leading international experts in
the field. In the 5 years that have followed, advances in the field of hematology
have been plentiful. As such, EHA is pleased to present an updated Research Roadmap,
now including 11 sections, each of which will be published separately. The updated
EHA Research Roadmap identifies the most urgent priorities in hematology research
and clinical science, therefore supporting a more informed, focused, and ideally a
more funded future for European hematology research. The 11 EHA Research Roadmap sections
include Normal Hematopoiesis; Malignant Lymphoid Diseases; Malignant Myeloid Diseases;
Anemias and Related Diseases; Platelet Disorders; Blood Coagulation and Hemostatic
Disorders; Transfusion Medicine; Infections in Hematology; Hematopoietic Stem Cell
Transplantation; CAR-T and Other Cell-based Immune Therapies; and Gene Therapy.
INTRODUCTION
Genetic engineering of hematopoietic stem cells (HSCs) has progressed from early stage
clinical trials providing evidence for substantial and durable benefits in some genetic
deficiencies to the first Advanced Therapy Medicinal Products (ATMP) approved for
the EU market (Figure 1).
2
As safety and efficacy of HSC gene therapy (HSCGT) is further established by increasingly
longer follow-up and a higher number of patients treated for different diseases, it
may become a new pillar of treatment for several inherited monogenic affections for
which allogeneic HSC transplantation (HSCT) represents a treatment option. Because
HSCGT exploits autologous patient-derived cells, it has the following advantages over
conventional allogeneic HSCT: (1) is in principle available to every patients; (2)
does not entail the risk of graft-versus-host disease; (3) can lower substantially
the risk of graft rejection, as the only potentially antigenic element in the administered
cells is the therapeutic gene product; (4) can more easily establish partial chimerism
with the administered HSC, which, according to disease and protocol, choice may allow
lowering the requirement for myeloablative conditioning regimens and still provide
substantial therapeutic benefit. Moreover, because of the genetic engineering step,
gene correction may be designed to provide an even higher benefit than obtained with
allogeneic normal or haploidentical carrier HSC, for instance by increasing therapeutic
gene dosage in lysosomal storage diseases. HSCGT may eventually be exploited also
for the treatment of some acquired affections, by instructing a new function to HSC
or some of its progeny such as in gene-based delivery of a biotherapeutic or by establishing
genetic resistance to an infectious agent.
On the other hand, HSCGT requires ex vivo culture and manipulation of the cells to
enable genetic engineering, which in turn demands for establishing a suitable GMP-compliant
multistep process from cells harvest to manufacturing and delivery at the clinical
point-of-care. Furthermore, the culture as well as the genetic engineering procedure
may affect some fundamental HSC biological properties negatively impacting the safety
and efficiency of the procedure. Although this is an active area of investigation,
most current protocols for growing HSC ex vivo fail to expand or even maintain for
more than a few days the most primitive and long-term engrafting cells and both gene
transfer and gene editing procedures have been reported to trigger innate cellular
responses that may lead to delayed or halted growth, differentiation or even apoptosis
and cell death. Consequently, if these adverse effects individually or cumulatively
surpass a threshold during cell manufacturing, they may cause delayed hematopoietic
recovery or even failure of engraftment, exposing the patient to high risk of infections,
and also compromise the clonal composition of the engineered hematopoietic graft,
which may reduce its long-term resilience and safety.
Last but not least, every meaningful genetic engineering of HSC requires stable, albeit
local, modification of the cellular DNA sequence for ensuring life-long transmission
to its cell progeny. This is a mutagenic procedure by definition, which entails a
genotoxic risk dependent on the type (gene transfer versus editing), site, extent,
and specificity of engineering. Whereas genotoxicity emerged as a significant risk
in early HSCGT trials leading to leukemia development in some treated patients, it
appears to be substantially alleviated by the new generation of lentiviral vectors
currently in broader use. In any case, long-term monitoring of HSCGT patients is still
required to establish the long-term safety and efficacy of these promising new therapies.
In the following sections, we highlight some key features of the genetic engineering
platforms under clinical development for HSCGT, the current stage and future outlook
of clinical applications of HSCGT in some major diseases families, and the regulatory
implications of these novel and revolutionary medicines entering the clinical arena.
Figure 1.
Bone marrow-resident hematopoietic stem cells and hematopoietic progenitor cells replenish
blood and tissues with new mature cells.
2
Both hematopoietic stem cells and hematopoietic progenitor cells express the cell
surface marker CD34, which is used to enrich a mixture of hematopoietic stem and progenitor
cells for transplantation and gene therapy. Hematopoietic stem cells can be classed
as long-term hematopoietic stem cells (LT-HSCs) or short-term hematopoietic stem cells
(ST-HSCs). ST-HSCs progressively acquire lineage specifications to differentiate into
lineage-committed progenitors and eventually terminally differentiated cells, which
are released into the peripheral blood. A simplified scheme of human hematopoiesis
is presented here. Alternative models have been postulated on the basis of cell surface
marker analyses, in vitro and in vivo functional assays, clonal tracking by insertion
analyses in hematopoietic stem and progenitor cell gene therapy studies, and single-cell
RNA analyses (reviewed previously
22
). Mendelian genetic disorders can affect self-renewal, differentiation, and/or the
function of different blood and immune cells. Examples of genetic diseases for which
gene therapy is under investigation or approved are represented in white boxes below
affected cell types. Wiskott–Aldrich syndrome affects platelets and other lineages.
CDP = common dendritic progenitor; CID = combined immunodeficiency; CLP = common lymphoid
progenitor; CMP = common myeloid progenitor; GMP = granulomonocytic progenitor; LMMP
= lymphoid-myeloid primed progenitor; MEP = megakaryocytic–erythroid progenitor; MPP
= multipotent progenitor; NK cell = natural killer cell; preB = pre-B cell; preT =
pre-T cell; SCID = severe combined immunodeficiency. Reprinted with permission from
Nat Rev Genet. 2021;22:216–234.
GENE THERAPY PLATFORMS
Introduction
Most current genetic engineering of HSC exploits 2 major strategies, gene replacement
mediated by retroviral gene transfer vectors or targeted gene editing mediated by
artificial sequence-specific endonucleases. Gene transfer strategies exploit a couple
of well-established platforms that were developed in research laboratories between
15 and 20 years ago. Since then, progressive clinical testing in >300 patients worldwide
for the treatment of several diseases have generated 2 ATMPs approved for the EU market
with several more expected in the upcoming years (Table 1).
2
Gene editing strategies are still in the earliest stage of clinical testing, although
they are witnessing a remarkable and constant progress advancing the versatility,
precision, and scope of the technological platforms adopted.
3,4
European research contributions
The EU has represented a privileged theatre for the clinical development of HSC GT
based on retroviral gene transfer from the earliest pioneering clinical trials until
today’s most advanced stages.
Retroviral vectors are replication-defective viral particles derived either from γ-retroviruses
(γ-retroviral vectors [γRVs]), such as Moloney leukemia virus (MLV), or from the human
lentivirus HIV-1 (lentiviral vectors [LVs]). They integrate semi-randomly a reverse
transcribed RNA genome into the chromatin of the transduced cells. The vector genome
comprises cis-acting signals for packaging, reverse transcription, and integration,
lacks full length or open viral genes and comprises the therapeutic gene expression
cassette. γRVs were the earliest to be developed, having limited efficiency of gene
transfer into the more primitive HSC because productive infection is strictly dependent
on concurrent replication of the targeted cell. Moreover, their early and most commonly
used design exploited the strong enhancer/promoter sequences embedded in the long
terminal repeats (LTRs) of the viral genome to drive transgene expression. This feature,
coupled with an integration bias favoring insertion near the promoter of actively
expressed genes, result in some likelihood of altering expression of endogenous genes
nearby the insertion site through promoter insertion and enhancer-dependent trans-activation.
Sporadic insertions near proto-oncogenes may result in activation of their oncogenic
potential, thus endowing the transduced cell with a growth advantage leading to progressive
clonal expansion in vivo and, in some cases, eventual transformation by further accumulation
of mutations. Vector integration studies have allowed ascribing the origin of leukemia
emerging in treated patients even several years post HSCGT to such types of genotoxic
insertions, and longitudinal monitoring of the clonal composition of reconstituted
hematopoiesis have shown the frequent expansion of clones carrying such insertions
before any clinical signs of leukemia. Despite these hurdles, one of the earliest
application of γRV HSCGT for the treatment of severe combined immunodeficiency caused
by deficit of adenosine deaminase (ADA-SCID) has shown efficacy and better safety
outcome than observed for other diseases and became the first HSCGT ATMP approved
for the EU market in 2016.
5
Because of the lower gene transfer efficiency and clinically relevant genotoxicity,
the γRV platform has now been almost completely replaced by LV, which have improved
both aspects.
1,6
The key LV features underlying these improvements are: (1) exploitation of the HIV
core capacity to infect nondividing cells to enhance transduction of the more primitive
HSC after a short ex vivo stimulation; (2) advanced vector design which completely
eliminates transcriptional activity from the vector LTRs. Expression of the therapeutic
gene is driven from an internal promoter of choice, either reconstituted from the
cellular promoter of the gene to be replaced, thus mimicking its physiological expression
pattern, or from a moderately active house-keeping promoter, thus providing for ubiquitous
expression. These features, combined with an insertion bias that favors the body rather
than the promoter of expressed genes, strongly alleviate the risk of genotoxic effects
at the insertion site; (3) exploitation of the HIV mechanism for packaging unspliced
viral genome to transfer complex gene expression cassettes, such as used to establish
erythroid specific and robust expression of a globin transgene in hemoglobinopathies.
An ever-growing number of patients treated by LV HSCGT for several diseases have been
showing stable and robust engineering of reconstituted hematopoiesis with highly polyclonal
composition and without signs of genotoxicity such as expanding clones or enrichment
for insertions at cancer gene. Depending on the disease, conditioning applied and
product manufacturing, the engineered cell fraction can reach up to near completion,
with evidence for distinct types of progenitors contributing to early recovery or
steady-state output, which is mostly driven by engrafted self-renewing multipotent
HSC.
2
Gene editing strategies allow precise and targeted modification of a DNA sequence
of choice, opening up novel and unique opportunities to genetic engineering. In the
longest available and clinically more advanced version, it exploits an artificial
endonuclease either composed of a sequence-specific DNA binding domain made of Zinc
Finger or TALE protein modules coupled to a FokI nuclease half-domain (ZFN or TALEN),
or of a target complementary single guide CRISPR RNA assembled with a Cas family nuclease
(CRISPR/Cas).
4
Either platform can deliver a DNA double-strand break (DSB) specifically in the target
sequence, whose repair is exploited for the intended type of edit. Nontemplated DSB
repair most commonly occurs by nonhomologous end joining (NHEJ), an error-prone process
that often introduces small base insertion or deletion (indels) while sealing the
break. If the break is targeted to a coding or an essential regulatory sequence the
NHEJ outcome will most often be disruption or inactivation of the sequence.
7
This represents the most frequent and, provided that the engineered nuclease is highly
specific, better tolerated editing procedure. In most cases, electroporation is exploited
to transiently introduce the nuclease mRNA or a preassembled CRISPR/Cas nucleoprotein
into the cells to be edited. Transient but robust expression is required for efficient
break induction while limiting toxicity. Nuclease specificity is crucial because DNA
DSBs trigger a detrimental DNA damage response which, if robust and sustained, can
lead to growth arrest, senescence, or apoptosis.
8
Moreover, multiple DSBs increase the risk of genotoxicity and genomic rearrangements,
such as translocations between different edited sites. Developing nuclease reagents
with stringent specificity for the target site and comprehensively assessing off-target
activity in preclinical models have been major goals of the field. A second more ambitious
editing strategy exploits DNA DSB repair by homologous recombination (HR). This requires
codelivery of the nuclease with a DNA template carrying the intended edit framed by
homologous sequences to the DNA flanking the nuclease target site. If successful,
this strategy will allow correcting mutations in situ, thus restoring both, function
and physiological expression of the affected gene, and even inserting longer sequences,
such as corrective cDNAs or transgene expression cassettes in safe genomic sites.
2
The current challenges in applying HR-mediated editing to HSCGT is the relatively
low efficiency of the process dependent on the poor permissiveness of primitive HSC
to this pathway and the cumulative detrimental impact of inducing DNA DSBs and introducing
an exogenous template DNA, which is best achieved by exploiting a viral vector such
as adeno-associated vector.
9
Proposed research for the roadmap
While application of LV gene transfer can be envisaged to be tested in an increasingly
larger number of diseases, the following areas of further HSCGT platform development
can be proposed.
Because autologous HSCGT does not require full ablation and replacement of resident
HSC, alleviation of the short-term and long-term morbidity of the procedure can be
sought by careful targeting of chemotherapeutic regimens to the minimal required dose
and through exploring emerging nongenotoxic conditioning strategies, such as those
based on antibodies or immunotoxins targeting HSC.
10,11
HSCGT may provide a favorable setting for early testing of these approaches, especially
in diseases not requiring high levels of chimerism with functional cells for correction.
Further optimization of ex vivo HSC culture conditions, including more faithful reconstitution
of signals supporting HSC self-renewal in bone marrow niches and/or their emergence
from hemogenic endothelium in the embryo, testing novel combinations of transduction
enhancers and uncovering their mechanism of action on HSC, and better standardization
and more effective process control in ATMP manufacturing should allow improving the
extent, predictability and reproducibility of gene transfer across different patients
and treatments.
Close monitoring of clonal composition and HSC activity in the engineered hematopoietic
graft of HSCGT patients will provide not only novel information on HSC biology in
living humans but also help uncover clues to the long-term safety and robustness of
therapeutic correction. As we learn more on the clinical relevance of clonal shrinking
and skewing in aging hematopoiesis, it will be important to investigate whether ex
vivo manipulation and genetic engineering may accelerate or aggravate such events.
Delayed emergence of new HSC clones after HSCGT may highlight the capacity of engineered
cells to return to long-term latency in vivo, establishing a safe reservoir for sustaining
the graft in the case of emergencies or aging. On the other hand, if the growing number
of treated patients and increasing time of follow-up were to lead to sporadic emergence
of expanding dominant clones, we could learn about the residual extent and mechanism
of genotoxicity of current vector design and introduce improved versions which are
currently being investigated in stressed experimental models but await strong rational
for clinical testing.
Gene editing strategies must be carefully monitored upon first clinical testing to
uncover any adverse outcome of the procedure in terms of hematopoietic recovery and
long-term engraftment, clonal composition of the engineered graft, and preserved long-term
maintenance and multipotency of the edited HSC. The occurrence of large-scale genomic
alterations in some edited cells and their in vivo fate will be investigated with
the goal to decrease their occurrence or purge them from the product, if necessary.
Protocols allowing for improved efficiency and tolerability of editing, especially
when exploiting HR, will be developed for instance by adopting emerging ex vivo HSC
expansion strategies and refined editors, or inhibiting or counteracting the detrimental
responses induced by the procedure.
New gene editing platforms that bypass the requirement for DNA DSB, such as base editors
and prime editors will be tested in preclinical studies for potential application
to HSCGT investigating their efficiency, tolerability, and specificity, potentially
providing for more precise and uneventful genetic engineering.
4
The potential adverse effects of pre-existing immunity to gene transfer and, in particular
gene editing reagents, some of which are of bacterial origin, will be investigated.
Residual antigenic components might be present in the cell product and trigger cellular
responses leading to clearance of the administered cells. The impact of pre-existing
effector or regulatory T cells specific for the viral/editor component at different
levels in the recipient will be investigated.
Anticipated impact of the research
We are at the beginning of a new era of medicine exploiting genes and cells as novel
therapeutics, leveraging on powerful genetic engineering tools to modify cell function
and capture fundamental biological processes, such as HSC-driven reconstitution of
intra and extravascular hematopoietic populations for therapeutic purposes. These
approaches open unprecedented opportunities and largely unexplored paths for therapeutic
interventions. Remarkable and durable benefits in until now orphan and otherwise severe
and lethal diseases have been reported and are expected to be further achieved in
growing numbers. However, this promise comes also with a tremendous challenge to the
currently established framework for developing, regulating and distributing medicines
while maintaining their economic sustainability by public and private healthcare providers
and guaranteeing fair access to the patients. To address this challenge and attain
the benefits that may come from this research, we must continue to support research
establishments that produce innovation, facilitate the creation of start-ups promoting
early development of the new technologies, advance education of biomedical trainees
to become familiar with the new technologies and treatments, update regulatory requirements
to emerging platforms and a rapidly evolving landscape, and foster engagement of pharmaceutical
industries in clinical development and commercial deployment of the new ATMPs.
GENE THERAPY FOR PRIMARY IMMUNODEFICIENCIES
Introduction
Primary immune deficiencies (PIDs) are a heterogeneous group of >400 genetic diseases
characterized by poor or absent function of one or more components of the immune system,
whose main clinical manifestations comprise increased frequency and severity of infection,
autoimmunity, and aberrant inflammation and malignancy.
12
Early diagnosis and treatment remain a mainstay for all forms of PIDs to prevent organ
damage and life-threatening infections and to improve prognosis and quality of life.
Allogeneic HSCT is curative for many PIDs, but still carries a risk of mortality and
morbidity from rejection, toxicity, and graft-versus-host disease. SCIDs were the
first monogenic disorders for which HSCGT has been successfully developed. Clinical
trials with promising results are ongoing for at least 6 different PIDs due to genetic
defects of adaptive and/or innate immunity while a growing number of diseases are
at preclinical stage of development.
2,13
European research contributions
European academic research played a key role in the development of successful HSCGT
approaches for PID. In the past 2 decades, the European Commission funded collaborative
research networks such as CONSERT, PERSIST, Clinigene, CELL-PID, NET4CGD, SCIDNET,
SUPERSIST, and UPGRADE. Over 100 PID patients have been treated in early trials of
HSCGT with γRV, showing reconstitution of immunity in most patients affected by SCID-X1
due to IL2RG deficiency, ADA-deficient SCID, Wiskott–Aldrich Syndrome (WAS), and X-linked
Chronic Granulomatous Disease (X-CGD).
2
ADA-SCID GT was the first PID for which non myeloablative conditioning was introduced
and in 2016 became the first HSCGT approved in the EU (Strimvelis, Orchard Therapeutics
Ltd, London, UK), with the indication of standard treatment for ADA-SCID lacking a
compatible family donor
5
Infusion of ADA gene replaced HSC resulted in persistent (>15 y) engraftment of gene-marked
cells ranging from 1% to 10% in the myeloid compartment and reaching up to 100% in
the lymphoid compartment, correction of adenosine metabolism and improved T-cell counts,
leading to discontinuation of prophylaxis and decreased incidence of severe infections.
14
The promising clinical data obtained in γRV trials have been tempered by the later
development of T-lymphoblastic leukemia and myelodysplasia as a result of insertional
mutagenesis. The incidence varied among disease types, suggesting that disease background,
transgene function, and individual genetic predisposition influence tumorigenicity.
LV-based platform has been deployed in the past decade enabling more effective and
safe insertion of therapeutic genes into HSC, with no evidence of leukemogenesis reported
to this date. LV-HSCGT clinical trials for SCID-X1 and ADA-SCID have shown improved
lymphocyte counts with clinical benefit.
15
HSCGT for WAS resulted in amelioration of immune functions, including autoimmunity,
and reduced incidence of severe bleeding events.
3,16
Moreover, HSCGT-mediated restoration of oxidase function in X-CGD patients leading
to protection from bacterial and fungal infections in most treated patients.
17
Proposed research for the roadmap
Continuous patients monitoring, also through analyses of vector insertion sites, will
be required to confirm long-term safety and efficacy of patients who underwent HSCGT.
Clinical trials for artemis deficiency, LAD-1, and osteopetrosis have recently started.
2
Preclinical studies are actively pursued for other PIDs in which HSCGT could represent
an alternative to allogeneic HSCT, also in the most severe or adult patients presenting
with ongoing infections and/or organ damage including JAK3-SCID, PNP deficiency, RAG1/2
deficiency, ZAP70 deficiency, Munc 13-4 deficiency, and DADA2.
18
HSCGT could also become available for less severe phenotypes, in which alloHSCT would
not be indicated such as BTK deficiency. New gene editing technologies have the potential
to circumvent some of the problems associated with viral gene addition and could be
suitable for diseases requiring physiological regulation of gene expression or inactivation
of dominant alleles. Preclinical proof of concept has been attained in correcting
gene mutations for IL2RG, WAS, p47-CGD, CD40L deficiency, and IPEX, and clinical testing
of these strategies is awaited.
One of the key factors in the success of HSCGT for SCID and WAS relies on the selective
advantage of functionally corrected lymphoid cells, previously observed in patients
with somatic revertants and allogeneic HSCT. Thus, even in patients not receiving
conditioning, active thymopoiesis was shown to be maintained for many years after
treatment, suggesting durable thymic engraftment of long-lived lymphoid progenitors.
2
On the other hand, conditioning is required to achieve polyclonal engraftment of gene-corrected
HSC, restore normal B-cell lymphopoiesis and establish corrected myelopoiesis. SCIDs
seem ideal candidates to explore the use of conditioning mediated by monoclonal antibodies
or immunotoxins for HSCGT while exploiting the selective advantage in the lymphoid
lineage. Preliminary results of a clinical trial in the context of allogeneic HSCT
for SCID-X1 show sufficient degree of HSC engraftment.
11
GT with autologous mature lymphocytes or lymphoid progenitors has also been explored
preclinically with the rationale to provide in patients not eligible to allogeneic
HSCT or HSCGT, immune responses to infection in diseases such as CD40L or perforin
deficiency as well as control of immune dysregulation regulation in FOXP3 deficiency.
Anticipated impact of the research
Gene therapy for PIDs is moving from being an experimental approach to approved drug
products that are routinely beneficial. The expansion of gene addition strategies
and implementation of gene editing approaches, together with the standardization of
the technology, will be important for increasing the armamentarium of approved therapies
as an alternative to allogeneic HSCT.
HEMOGLOBINOPATHIES
Introduction
Hemoglobinopathies are genetic defects of hemoglobin chain production, caused by mutations
in the α- or β-globin gene clusters. The most frequent and severe are transfusion-dependent
β-thalassemia (TDT) and sickle cell disease (SCD), characterized by a reduced or absent
level of adult hemoglobin (HbA) and the production of an abnormal structural variant
(HbS), respectively. In TDT, survival of patients depends on chronic blood transfusion
associated to iron chelation. In SCD, sickling of deoxygenated red blood cells (RBCs)
promotes painful vaso-occlusive crises, acute chest syndrome, stroke, and eventually
death. Blood transfusion and fetal hemoglobin (HbF) induction by hydroxyurea represent
current treatments. Allogeneic HSCT is curative for both diseases, but with limited
availability of suitable donors and variability in clinical outcome depending on patient’s
age. LV-mediated HSCGT relies on the erythroid specific expression of a normal gene
copy and has been recently approved in Europe for patients over 12-years-old affected
by less severe β-thalassemia mutations, and it is still in experimental trials for
SCD.
2
Gene editing strategies in the field of hemoglobinopathies have reached the stage
of clinical testing, with ongoing phase 1/2 trials using ZFN or CRISPR/Cas technologies
in SCD and in TDT. Reactivation of HbF synthesis by disruption of Bcl11a suppressor
led to initial encouraging clinical results
19
and longer follow-up will fully disclose the real potentiality as well the caveats
of this novel approach.
20
European research contributions
European researchers have been prime actors in nonclinical and clinical research for
the cure of TDT and SCD, contributing to the discovery of the disease molecular cause
and the molecular mechanisms of globin genes regulation, introducing allogeneic HSCT
and translating basic research to clinical application of gene therapy. Two successful
clinical trials in TDT have been conducted in France
21
and Italy,
22
paving the way for application also in SCD.
23
The results showed correction of the disease in most adult patients carrying nonsevere
mutations and in young patients with severe mutations. The level of marked engrafted
cells positively correlates with the clinical benefit, highlighting a threshold of
genetically corrected cells to produce sufficient HbA per cell to rescue anemia and
ineffective erythropoiesis in TDT, and to dilute abnormal HbS and preventing sickling
in SCD. The functional status of HSC and the BM microenvironment is particularly relevant
in TDT, where an impaired HSC-niche cross-talk has a negative impact on HSC functionality.
24
Proposed research for the roadmap
Gene therapy for hemoglobinopathies poses unique challenges, including high level
of transgene expression for therapeutic correction and high global incidence worldwide.
A single ATMP on the market (Zynteglo) will not be sufficient for such a prevalent
disease but demonstration of dissimilarity of drug products might be challenging,
thus discouraging biotech and/or pharma investments in further academic preclinical
and clinical research. New rules governing the ATMP market should be introduced to
favor the treatment of large patients’ population with biologically similar products.
As far as for other genetic diseases, the myeloablative conditioning required to favor
the engraftment of genetically modified HSPCs imposes a high burden on patients, with
general toxicity and impairment of fertility. Non-genotoxic biological conditioning
specifically targeting resident HSC would be highly desirable, although its efficiency
needs to be tested and modeled in the context of a BM engulfed by an expanded erythroid
component. Additionally, amelioration of BM microenvironment will be instrumental
in favoring engraftment and expansion of HSCs in the context of both, allogeneic HSCT
and HSCGT.
Anticipated impact of the research
The long-term effort in GT research for TDT has recently resulted in the approval
of a first ATMP (Zynteglo) and poses the challenge of its manufacturing and distribution
to a large number of affected patients. The development and optimization of gene editing
approaches will offer an additional opportunity of intervention, with potentially
therapeutic levels of Hb production exploiting the power of chromosomal regulatory
sequences. In both cases, the complex manufacturing and associated costs represent
major obstacles for fulfilling the current medical need, leaving the door open to
the development of novel and potentially more feasible approaches of in vivo gene
therapy.
METABOLIC DISORDERS
Introduction
Inborn errors of metabolism (IEMs) are a large class of genetic disorders characterized
by the absence/dysfunction of an enzyme or its co-factor leading to disruption of
cellular biochemical functions. Subgroups characterized by permanent, progressive
symptoms often involving the nervous system and the skeleton encompass peroxisomal
and lysosomal storage diseases (LSDs), according to the subcellular location of the
defects. LSD pathology results from accumulation of waste materials, causing cellular
dysfunction, alteration of cell morphology, impaired autophagy, oxidative stress,
neuroinflammation, and impaired organ function in the brain, bones, viscera, and connective
tissue.
25
Enzyme replacement therapy (ERT) and allogeneic HSCT (allo-HSCT) have been exploited
as potential treatments for LSDs, but preclinical and clinical studies have shown
limited efficacy in ameliorating cardinal disease manifestations. GT using autologous
HSC is an attractive alternative to allo-HSCT, as it promises not only an improved
safety profile, but also enhanced efficacy, due to its potential to turn HSC progeny
into hyperfunctional enzyme factories capable of more effectively cross-correcting
nonhematopoietic cells, systemically and locally within the affected tissues.
European research contributions
X-ALD (X-linked adrenoleukodystrophy): Successful treatment of 2 boys affected by
the childhood cerebral form of adrenoleukodystrophy (CALD) with LV-transduced HSC
marked a milestone ushering in a new era of more effective and less genotoxic engineering
of hematopoiesis.
26
Since then, >30 patients have been treated with HSCGT,
27,28
which arrested disease progression and kept children treated early in their disease
course free from major functional disabilities, similar to what can be achieved by
a successful Allo-HSCT.
MLD: Metachromatic leukodystrophy put LV HSCGT to a stress test, as allo-HSCT is not
effective in early onset forms of the disease and, based on preclinical models, benefit
could only be expected from supraphysiologic expression of ARSA enzyme from the progeny
of transduced HSC. A clinical trial of HSCGT in presymptomatic or early symptomatic
early onset MLD
29
showed unprecedented levels of stable gene marking throughout hematopoiesis, effective
prevention of disease onset in presymptomatic patients and slowing of disease progression
in early symptomatic ones, thereby representing the first treatment capable of modifying
the natural history of this devastating neurologic disease. This MLD HSCGT has recently
been granted market authorization by EMA (commercial name “Libmeldy”).
MPSIH (mucopolysaccharidosis type 1-Hurler syndrome): Capitalizing on the increased
therapeutic potential of supraphysiologic enzyme reconstitution by HSCGT in MLD, a
phase I/II study for Hurler disease, the most severe form of MPSIH, has recently been
conducted, introducing a shortened ex vivo manipulation protocol with PGE2 as transduction
enhancer that allowed optimal conservation of repopulation potential. Rapid biochemical
correction and supraphysiologic IDUA activity was obtained in all 8 treated patients,
with an excellent safety profile.
MPSIIIA (mucopolysaccharidosis type IIIA): Similarly encouraging preliminary results
have been obtained in a phase I/II trial for MPSIIIA.
30
HSCGT allowed achieving supraphysiologic enzyme levels in hematopoietic lineages and
effective substrate reduction, consolidating the potential of gene therapy in this
group of LSDs.
Investigational HSCGT is currently being studied in clinical trials in Fabry disease,
Gaucher disease, and cystinosis in the United States and Canada.
Proposed research for the roadmap
HSCGT has been demonstrated to represent a promising treatment modality for previously
refractory LSDs and other metabolic disorders. Ongoing and future clinical trials
will provide essential insights into the mechanisms of correction through HSCGT in
specific tissues/organs (ie, nervous system, skeleton) and will allow refinement of
indications and procedures of treatment (ie, conditioning regimens, cell dose, modulation
of antitransgene immune response and, potentially, intrathecal administration of corrected
HSPC). LSDs, and in particular MPSs, represent an area of further HSCGT development
with more diseases potentially amenable to cross-correction mechanisms and benefitting
from supraphysiologic levels of the missing enzyme.
HSCGT-based therapeutic strategies have provided the clinical basis for the implementation
of newborn screening (NBS) for IEMs which allows early diagnosis and timely treatment,
this being a key factor for better preserving tissue and organ function and improving
treatment outcome. NBS programs for LSDs are being conducted in Europe and the United
States to this purpose.
Anticipated impact of the research
The last 10 years represent unprecedented times for HSCGT in IEMs with several clinical
approaches being developed and tested in Phase I/II clinical trials. So far, HSCGT
in LSDs has provided evidence of metabolic correction in several diseases (MLD, MPSIH,
and MPSIIIA), while demonstration of robust and sustained clinical efficacy has been
obtained so far in ALD and MLD. Longer observation of HSCGT-treated IEM patients will
give support to the clinical efficacy of these strategies in the long-term, together
with additional proof of safety. Further clinical development of HSCGT approaches
in these and other IEMs will possibly transform the outcome of some of them.
OTHER EMERGING APPLICATIONS
A long-sought application of HSCGT has been the correction of Fanconi anemia (FA),
caused by loss-of-function mutations in any one of at least 17 genes in the FA pathway
and resulting in the inability to repair interstrand DNA crosslinks. HSCs are most
sensitive to this defect and progressively exhaust, giving rise to severe pancytopenia
as well as a high risk of myelodysplasia. Whereas GT has been particularly challenging
for this disease because of the paucity and fragility of FA HSC, the growth advantage
of corrected HSC provides the opportunity for in vivo expansion of even a limited
engrafted input. A recent trial of LV–based HSC GT conducted in Spain for FA complementation
group A reported successful engraftment of corrected HSC in the absence of conditioning
followed by sustained and progressive expansion of the corrected hematopoiesis, reaching
therapeutic relevance over several months.
31
Preclinical studies of HSC gene editing are also ongoing in FA, aging leveraging on
the selected advantage of correction.
Another recent development of HSC GT is to broaden gene addition beyond the replacement
of a defective gene and encompassing the delivery of a biotherapeutic to disease sites
or establishing resistance to an ineradicable infectious agent. These approaches are
based on engineering hematopoietic progenitors with vectors designed to selectively
express in certain hematopoietic lineages, differentiation or maturation stages. For
gene-based delivery of biotherapeutics, the engineered mature cells act as smart agents
selectively distributing the gene product to extravascular disease sites. Emerging
application of this approach are being tested to target immunostimulatory cytokines,
such as alpha-interferon, to tumors by tumor-infiltrating macrophages or immunomodulatory/growth-promoting
agents to neuroinflammatory/degenerative lesions by brain-infiltrating macrophages.
Overall, these early stage studies may open up new avenues for the treatment of some
prevalent affections, such as cancer, chronic infections and neurodegenerative diseases.
REGULATORY CHALLENGES
HSCGT ATMPs have unique attributes which differentiate them from standard pharmaceuticals
and biologics. They bring the potential to offer a durable curative therapeutic effect
upon a single administration to patients who may have few or no alternative treatment
options. However, the complexity and novelty of these ATMPs also present several challenges
that need to be overcome to ensure that these products reach all those in need.
Regulators have established dedicated pathways and expert committees to ensure appropriate
and expedited registration of ATMPs. In EU, specific regulatory tools are available,
such as classification of the products as ATMP and certification of ATMP quality and
nonclinical data for small and medium-sized enterprises (SMEs). In addition, the use
of more general tools such as scientific advice/protocol assistance, parallel EU/US
scientific advice, the priority medicines or PRIME scheme and qualification of novel
methodologies, is recommended to discuss as early as possible the strategy under development
with the central (ie, EMA in EU or FDA in the United States) and national regulatory
authorities that will assess the investigational drug application. However, it should
be recognized that these novel and revolutionary medicines entering the clinical arena
pose great challenges for the developers as there is still lack of harmonization of
regulatory requirements across geographic areas in terms of quality standards, genetic-modified
organisms (GMOs) manipulation, preclinical data packages to support the first in human
and clinical evidence to be generated for registration and access to the market. Moreover,
most products are developed to treat rare and orphan diseases with the associated
challenges of small patient population and complex clinical trial design. HSCGTs are
innovative products which bring complexity in the manufacturing process and in the
comprehensive analytical panel needed to characterize, control and release the product,
which require highly specialized manufacturing equipment, processes and skills. Ensuring
consistent standards and adequate characterization across starting materials, processes
and infrastructure is a common challenge for HSCGT manufacturing. As HSCGT are highly
personalized medicines they pose even higher challenges in terms of standardization
and reproducibility. Overall, these hurdles together with the associated high costs
and regulatory burden may jeopardize early steps toward clinical testing of new GT
products, especially if developed within the context of academic institutions. Thus,
a critical balance should be maintained when manufacturing products for early stage
clinical testing by adopting quality standards sufficient to ensure patient safety
as well as compliance with regulatory requirements for clinical testing while postponing
more stringent requirements and comparability studies to later stages of clinical
development. At the same time, it should be realized that entering the clinic with
a nonoptimized product may compromise its effective development and, even if shown
to be safe and efficacious, may delay the registration due to the further time needed
for process scale-up and full validation for commercialization.
Despite different regulatory requirements across regions still exist for this kind
of therapies, there is an emerging paradox between regulators’ approaches implemented
to ensure rapid approval and early access to the ATMP, and payers’ and health technology
assessment (HTA) bodies’ current hesitancy to ensure access until the long-term safety
and efficacy profile has been fully characterized. Payers/HTA bodies need to establish
more effective mechanisms to capture the full benefits of these therapies and overcome
current potential barriers to timely patient access postapproval. Early dialogue with
HTA bodies and future payers during ATMP development using tools like the parallel
consultation with HTAs might help to address the challenges posed by these transformative
medicines and develop new evaluation tools and payment models to support financial
sustainability for both payers and manufacturers and make HSCGT fairly and promptly
available to patients.
Table 1.
Most Recent HSPC Gene Therapy Available on clinicaltrails.gov (Recruiting or Completed)
Disease (Gene)
Trial phase
Vector
Conditioning
Preliminary Outcomes
Clinical Trial Registry Number(Reference)
SCID-X1 (IL2RG)
I/II
G2SCID LV
Low-dose busulfan
Still recruiting
NCT03311503
SCID-X1 (IL2RG)
I/II
G2SCID LV
Low-dose busulfan
Still recruiting
NCT03601286
SCID-X1 (IL2RG)
I/II
TYF-IL-2Rg self-inactivating LV (TYF-IL-2Rg)
Not known
Still recruiting
NCT03217617
SCID-X1 (IL2RG)
I/II
LV VSV-G pseudotyped CL20- 4i-EF1a-hyc-OPT
Low-dose busulfan
Sustained marking levels and restoration of humoral responses to immunization
NCT01306019(De Ravin,
Sci Transl Med. 2016;8:335ra57)
ADA-SCID (ADA)
I/II
EFS-ADA LV
Low-dose busulfan
Sustained engraftment of genetically modified HSPCs in all the patients with long-term
metabolic detoxification from deoxyadenosine nucleotides after stopping ERT
NCT02999984(Kohn,
Blood (2019) 134 (Supplement_1): 3345)
ADA-SCID (ADA)
I/II
EFS-ADA LV
Low-dose busulfan
Sustained engraftment of genetically modified HSPCs in 9/10 patients
NCT01852071(Kohn,Blood (2019) 134 (Supplement_1): 3345)
ADA-SCID (ADA)
I/II
EFS-ADA LV
Low-dose busulfan
Still recruiting
NCT03765632
ADA-SCID (ADA)
II/III
EFS-ADA LV
Low-dose busulfan
Suspended (recruitment on hold for business reasons)
NCT04140539
ADA-SCID (ADA)
Not Applicable
a
Self-inactivating LV TYF-ADA
Not known
Still recruiting
NCT03645460
WAS (WAS)
I/II
w1.6_hWASP_WPRE (VSVg) LV
Reduced-intensity conditioning regimen with busulfan and fludarabine
Sustained engraftment of genetically modified HSPCs with reduction of bleeding events
and restoration of WASP expression in lymphocytes and plateles,
NCT01515462(Ferrua et al
16
)
WAS (WAS)
I/II
w1.6_hWASP_WPRE (VSVg) LV
b
Myeloablative conditioning regimen with busulfan and fludarabine
Sustained multi-lineage vector gene marking over time. All subjects had improvement
or resolution of eczema and none had intercurrent severe infectious events
NCT01410825(Labrosse, Blood (2019) 134 (Supplement_1): 4629)
WAS (WAS)
II
w1.6_hWASP_WPRE (VSVg) LV
b
Reduced-intensity conditioning regimen of busulfan and fludarabine
No results available
NCT03837483
WAS (WAS)
I/II
w1.6_hWASP_WPRE (VSVg) LV
b
Myeloablative conditioning regimen with busulfan and fludarabine
Stable engraftment of genetically and functionally corrected lymphoid and myeloid
cells in all patients with lack of severe adverse events or clonal expansion
NCT02333760 (Magnani, et al. Mol Therapy 2020;28:4S1)
X-CGD (gp91phox)
I/II
G1XCGD LV
Myeloablative conditioning regimen with busulfan
Stable vector copy numbers and persistence of oxidase-positive neutrophils in 6/7
surviving patients. No new CGD-related infections
NCT01855685and NCT02234934(Kohn et al
17
)
X-CGD (gp91phox)
I/II
G1XCGD LV
Myeloablative conditioning regimen with busulfan
2/4 patients showed clinical and biological benefits
NCT02757911 (Magnani, et al. Mol Therapy 2020;28:4S1)
Transfusion-dependent β-thalassemia (HBB)
I/II
GLOBE LV
Myeloablative conditioning with treosulfan and thiotepa
Robust and persistent engraftment of genetically modified HSPCs in 7/9 patients and
achievement of transfusion independence in 4/6 children
NCT02453477(Scaramuzza, et al. Mol Therapy 2020;28:4S1)
Transfusion-dependent β-thalassemia (HBB)
III
LV βA-T87Q-Globin
Myeloablative conditioning regimen with busulfan
Transfusion independence was observed in most of the patients. HbAT87Q stabilized
approximately 6 months after treatment and patients who stopped RBC transfusions had
improved erythropoiesis
NCT02906202(Thompson, et al. Blood, 2019; 134(Supplement_1): 3543)
Transfusion-dependent β-thalassemia (HBB)
III
LV βA-T87Q-Globin
Myeloablative conditioning regimen with busulfan
In 3/4 patients with ≥ 6 months follow-up have stopped transfusions and one patient
has achieved transfusion independence
NCT03207009(Lal, Blood, 2019, 134(Supplement_1): 815)
SCD (HBB)
I/II
LV βA-T87Q-Globin
Myeloablative conditioning with busulfan
Improvement in hematologic parameters and disease-related symptoms
NCT02151526(Magrin, Blood, 2019, 134 (Supplement_1): 3358)
SCD (HBB)
I/II
LV βA-T87Q-Globin
Myeloablative conditioning with busulfan
Reduction in the annualized rate of disease-related symptoms. Patients maintained
HbAT87Q production, demonstrating the durability of gene therapy-derived β-globin
gene expression
NCT02140554(Walters, Blood, 2019, 134 (Supplement_1): 2061)
SCD (HBB)
I/II
Gamma-globin LV
Reduced intensity conditioning with melphalan
Still recruiting
NCT02186418
SCD (HBB)
I/II
GLOBE1 LV expressing the βAS3 globin gene
Myeloablative conditioning with busulfan
Still recruiting
NCT03964792
SCD (HBB)
I/II
Lenti/G-βAS3-FB LV
Myeloablative conditioning with busulfan
Still recruiting
NCT02247843
SCD (HBB)
I
LV encoding human γ-globinG16D and short-hairpin RNA734 for selection of hypoxanthine
guanine phosphoribosyltransferase
Reduced intensity conditioning with melphalan
Still recruiting
NCT04091737
MLD (ARSA)
I/II
LV ARSA
Myeloablative conditioning with busulfan
Sustained multilineage engraftment of genetically modified HSPCs and clinical improvement
compared to natural history patients
NCT01560182(Fumagalli WORLD symposium 2020)
MLD (ARSA)
II
LV ARSA
Myeloablative conditioning with busulfan
Preliminary results showed multilineage engraftment of genetically modified HSPCs
and restoration of ARSA activity
NCT03392987(Fumagalli WORLD symposium 2020)
X-ALD (ABCD1)
II/III
SIN LV MNDprom-ABCD1 (Lenti-D) encoding human adrenoleukodystrophy protein
Myeloablative conditioning with busulfan and cyclophosphamide
Gene marked cells after engraftment and measurable ALD protein in all the patients
NCT01896102(Eichler et al
27
)
X-ALD (ABCD1)
III
Autologous CD34+ cells transduced with SIN LV MNDprom-ABCD1 (Lenti-D) encoding human
adrenoleukodystrophy protein
Myeloablative conditioning with busulfan and fludarabine
Still recruiting
NCT03852498
MPSI (IDUA)
I/II
LV IDUA
Myeloablative conditioning with busulfan and fludarabine
Preliminary results showed multilineage engraftment of genetically modified HSPCs
and restoration of IDUA activity
NCT03488394 (Gentner, WORLD symposium 2020)
MPSIII (SGSH)
I/II
CD11b LV vector encoding for human SGSH
Myeloablative conditioning with busulfan
Preliminary results showed multilineage engraftment of gene-modified cells and sustained
vector copy number
NCT04201405(Kinsella et al. Mol Therapy 2020;28)
Fabry disease (GLA)
I
LV AVR-RD-01
Myeloablative conditioning
Durable engraftment. Sustained plasma and leukocyte enzyme activity
NCT02800070(AvroBio data update, ASGCT 2020)
Fabry disease (GLA)
I/II
LV AVR-RD-01
Myeloablative conditioning
Durable engraftment. Sustained plasma and leukocyte enzyme activity
NCT03454893(AvroBio data update, ASGCT 2020)
aNot applicable is used to describe trials without FDA-defined phases.
bThe same vector design but performed in different manufacturing sites and with different
transduction protocols.
Adapted from: Tucci F, Scaramuzza S, Aiuti A, Mortellaro A. Update on clinical ex
vivo hematopoietic stem cell gene therapy for inherited monogenic diseases. Mol Ther.
2021;29:489-504.
Summary box: Main research & policy priorities
Expand administration of HSCGT to increasing number of patients and continue long-term
monitoring of the treated ones to establish safety and efficacy of the treatment and
make it first-line option for those diseases in which an autologous source of gene-corrected
cells can lower the risks and improve the benefits of allogeneic HSCT.
Broaden application of HSCGT to disease families where strong proof-of-principle of
the therapeutic potential of HSCGT has been demonstrated, such as LSD.
Testing the therapeutic potential of emerging nongenotoxic conditioning regimens for
which HSCGT may provide a favorable setting.
Improving the extent, predictability, and reproducibility of lentiviral gene transfer
across different patients and treatments.
Monitoring of clonal composition and HSC activity and fate in the long-term genetically
engineered graft.
Investigate the multifactorial contribution and longitudinal stepwise evolution of
genotoxicity in HSCGT and devise strategies to further alleviate the risk of progression
to malignancy.
Closely monitor new gene editing strategies as they enter clinical testing to uncover
any adverse outcome of the procedure in terms of hematopoietic recovery and long-term
engraftment, clonal composition of the engineered graft, and preserved long-term maintenance
and multipotency of the edited HSC.
Investigate and ameliorate any disease-specific alterations of the BM microenvironment
to facilitate engraftment and expansion of gene corrected HSC.
Assess the potential adverse effect of pre-existing or newly developed immunity to
gene transfer and editing tools as well as to transgene products as HSCGT is applied
to an increasing number of diseases and lymphodepleting preconditioning is further
alleviated or bypassed.
Establish more supportive framework for developing, regulating and distributing ATMP
while maintaining their economic sustainability by public and private healthcare providers
and guaranteeing fair access to the patients, also by adapting quality standards to
the stage of development and allowing development of biologically similar products.
DISCLOSURES
MPC is PI or Co-PI of studies sponsored by Orchard Therapeutics, and consulting for
Orchard Therapeutics, Exafield S.r.l, LSC Italy S.r.l, Atheneum S.r.l. MEB is PI of
a clinical trial sponsored by Orchard Therapeutics and participated in an Advisory
Board for Orchard Therapeutic. BG is a founder, stockholder and consultant of Genenta
Science. MG is consulting for Genespire. AA is the PI or Co-PI of clinical trials
sponsored by Orchard Therapeutics. LN is a founder, owns equity, and is consultant
and member of the scientific advisory board of Genenta Science, Genespire, Epsilen
Bio/Chroma, Tessera Therapeutics, Magenta Therapeutics. All the other authors have
no conflicts of interest to disclose.