First in Human Gene Editing for an Inherited Metabolic Disease

Abstract

Last month, Musunuru et al. [1] reported the first in vivo gene editing therapy for a baby with an inherited metabolic disease (IMD) in the New England Journal of Medicine. The ‘n of 1’ study comprised a well planned and executed therapeutic pipeline culminating in delivery of a customised novel gene editing treatment to an infant less than 7 months after their diagnosis with neonatal onset carbamoyl phosphate synthase (CPS1) deficiency. This groundbreaking work impressed by its demonstration of feasibility and early successful data in a human but also by the rapidity by which this was achieved, 7 months from diagnosis to a novel personalised therapy designed from scratch.

The patient presented soon after birth with severe hyperammonaemia requiring nitrogen scavengers and renal replacement therapy. Targeted genome sequencing enabled a rapid diagnosis of CPS1 deficiency, identifying compound heterozygous pathogenic variants, one of which was particularly amenable to base editing. By 1 month, the patient's mutation had been introduced into a cell line and by 2 months screening had been completed for the most efficacious base editing strategy. By 3 months, a mouse model had been created incorporating the patient-specific mutation using CRISPR/Cas9 technology. Off-target analysis revealed low-level synonymous bystander editing in the CPS1 gene in keeping with a single base mismatch in the guide mRNA. After an initial meeting with the FDA at 4 months, a toxicology batch of the base editor was manufactured and delivered to the mouse model and to non-human primates (NHPs) (5–6 months).

Satisfied by the lack of toxicity in the mice and NHPs, a clinical batch of the lipid nanoparticle mRNA gene editing treatment, named k-abe, was manufactured (6 months). Off-target editing analyses of the clinical batch were acceptable, and an IND application was approved by the FDA at 7 months. Two doses were administered to the infant, at 7 and 8 months of age, together with concomitant prophylactic immunosuppression (sirolimus and tacrolimus) to prevent immunisation against the transgenic enzyme. Partial efficacy was evidenced by good control of plasma ammonia and glutamine levels during a period of clinical stability and during two viral illnesses, an increase of orotic acid allowing a well-tolerated normalisation of dietary protein intake and a halved dose of nitrogen scavenger. No safety issues were reported, but only two low doses were delivered (0.1 and 0.3 mg/kg respectively), and clearly longer-term follow-up is needed.

This was an ideal candidate disease to trial this type of innovative therapy for the first time—a severe neonatal presentation of a disease with predictable consequences and lack of alternative therapeutic options other than liver transplantation in infancy. The patient remained stable with no major hyperammonaemic decompensation before dosing. As the authors of the article stated, mortality of neonatal onset CPS I deficiency is approximately 50% in infancy and whilst liver transplantation may be curative, many infants are too unwell or not large enough to receive a liver transplant. It is possible that gene editing may be used as a bridge to liver transplant, although the hope is that it will act as a standalone, definitively curative therapy.

This may be a taste of a distant future, but it is difficult to envisage how such a detailed operation dedicated to cure a single child could be upscaled to treating large numbers of children with multiple different rare genetic diseases. It is essential that targeted IMDs should have a ‘window of opportunity’, where the phenotype remains stable enough to wait for the therapy to be designed, tested preclinically and formulated according to Good Manufacturing Practice (GMP) standards. There are enormous implications in terms of costs and manpower needed to deliver such therapies at scale. It is also unlikely that all mutations will be amenable to such editing success, nor is it known how frequently off-target editing will occur, potentially leading to adverse effects. A mutation specific gene editing approach may be more successful for common mutations, but the reality for urea cycle defects and many other IMDs is that there are hundreds of private mutations. This is a competitive space, with other possible solutions including PRIME editing [2], whole gene insertion approaches mediated by nucleases such as ARCUS (which is being used in the OTC-HOPE study, an iECURE-sponsored phase I/II clinical trial to treat another urea cycle disorder, ornithine transcarbamylase deficiency [3]) and in vivo lentiviral gene therapy [4], in addition to AAV mediated gene therapy [5] and mRNA therapies [6] that are already in the clinic, in clinical trials or in IND enabling stage.

The key message is that this whirlwind delivery of a patient-specific base editing therapy is a significant milestone for IMD, heralding the possibility of rapid targeted personalised medicine therapies for children affected by IMDs going forward. A well-anticipated and designed pipeline and the involvement of a rapidly responding regulatory agency were essential to pave the way for a record-breaking individualised therapy. Developing a widely available therapeutic platform for patients would be an ideal scenario. Whether this can be translated into reality remains to be seen.

The authors declare no conflicts of interest.