A ‘Once-and-Done’ Approach to the Lifelong Reduction of Elevated Cholesterol Free

Comment on ‘In vivo CRISPR base editing of PCSK9 durably lowers cholesterol in primates’, which was published in Nature, doi:10.1038/s41586-021-03534-y.

Comment

Individuals with spontaneous loss-of-function PCSK9 mutations experience a significant reduction of both LDL-C levels (∼30–40%) as well as CHD risk (88%), and appear free from adverse clinical consequences.² Gene-editing technologies, which include the CRISPR–Cas nucleases and CRISPR base editor, have the potential to permanently modify disease-causing genes.³ The demonstration of durable editing of PCSK9 in target organs is a key step before in vivo administration of specific gene editors in clinical trials. In this experimental proof-of-concept study, genetic inactivation of PCSK9 gene by CRISPR editors was associated with a substantial and sustained lowering of LDL-C levels in non-human primates.¹ Although current PCSK9 targeting approaches are highly effective in reducing LDL-C levels, their effects on LDL-C are short-lived, and patients need to receive regular shots every few weeks.⁴ The recent development of RNA-based approaches blocking PCSK9 transcription (e.g. Inclisiran) has provided a significant progress by reducing the administrations to twice a year. Yet, a ‘single shot’ therapy would definitely solve the issue of poor treatment adherence, the latter being a key predictor of unsatisfactory LDL-C targets (only 20% of high-risk patients are on target) and cardiovascular mortality in statin-treated patients.⁴ Unlike currently available PCSK9-targeting drugs, gene-editing approaches offer the potential for a ‘once-and-done’ therapy. Although the long-term permanence of CRISPR-based liver editing remains to be established, no signs of attenuation of the pharmacodynamic effects of liver editing were shown after 8 months of follow-up.¹

A central aspect to consider when developing safe and effective genome editing therapies is the minimization of off-target editing and mutations.³ In this respect, our patients would likely ask a reasonable question: ‘how precise is the gene editing and what are the possible side effects after a “once-and-done” approach has been delivered’? The authors¹ show that the use of a base editor resulted in the alteration of a single base pair as the predominant editing event and had no risk of vector sequence integration. Notably, no off-target editing was observed in human hepatocytes, which is reassuring in terms of safety concerns. In contrast, a recent study using adeno-associated virus (AAV)-delivered meganuclease—defined as engineered variants of homing endonucleases that can be redirected to DNA sequences—showed off-target editing at numerous genomic sites in the non-human primate liver and in human hepatocytes.⁵ Hence, the use of mRNA (CRISPR ABE) rather than a DNA vector (meganucleases) may confer a more precise editing, mainly due to a larger degree of integration of the AAV vector sequence into the genome at the site of the break.³ The precise correction of disease-causing single-nucleotide mutations by ABE had already been shown in mouse models of genetic disorders such as phenylketonuria (through the correction of Pah mutations by a cytosine base editor) or Hutchinson–Gilford progeria syndrome (through the correction of LMNA transgene mutations by an ABE).³ Moreover, a couple of considerations suggest that off-target editing might not be associated with major risks: (i) human cells (e.g. hepatocytes) can tolerate a large number (>1000) of spontaneous point mutations during the life course and (ii) continuous expression of ABE over the period of 1 year did not increase hepatocellular carcinoma formation in Trp53 mutant mice.⁶ Lipid nanoparticle-delivered base editor was also associated with a more reproducible and pronounced editing (over 50%) as compared with meganucleases-based editing.¹ Furthermore, long-term follow-up of monkeys receiving CRISPR ABE showed only transient and moderate rises in liver enzymes (AST and ALT) that were entirely resolved after 2 weeks, with no adverse health effects. Although base editors might induce an adaptive immune response, the absence of late increases in AST and ALT observed in the present study suggests that such a response (if any) does not adversely affect the efficacy of the treatment.

That having said, further preclinical studies evaluating the long-term safety of base editing in vivo are needed before moving on with first in-human studies. Adjustment of dose levels and dosing schedule(s) could also contribute to increasing editing rates in vivo. A recent study⁷—which was published simultaneously with the present work—showed that LPNs–based delivery of ABE targeting PCSK9 induced up to 67% editing in mice and up to 34% editing in macaques, with LDL-C reductions by 58% in mice and 14% in macaques. Moreover, no off-target mutations in genomic DNA were observed.⁷ In conclusion, the landmark study by Musunuru et al.¹ paves the way for a broad application of CRISPR editors delivery for the treatment of dyslipidaemias as well as a wide panel of monogenic human diseases. The potential risks of this approach must be carefully weighed against their benefits.

Conflict of interest: F.P. is the recipient of an H.H. Sheikh Khalifa bin Hamad Al Thani Foundation Assistant Professorship at the Faculty of Medicine, University of Zürich. M.V. reports personal fees for speaker bureau and/or consulting in Advisory Board from Amgen, Astra Zeneca, Daiichi-Sankyo, Menarini Int, MSD, Novartis Pharma, Novo Nordisk, outside the submitted work.

References