Bug in the system

The curious challenges of gene therapy development

In software-speak a bug is a glitch and a feature is an intended functionality.
In gene therapy, however, the ‘bug’ is the virus or – more accurately – the viral vector itself, and the feature or intended functionality is a gene product that the vector carries and/or expresses.

Viral vectors are by far the most prevalent gene carriers owing to superior tissue targeting and transduction efficiency compared to non-viral vectors.

Non-viral vectors, however, like the lipid nanoparticle-based mRNA vaccines for COVID-19 – BNT162b2 (BioNTech-Pfizer) and mRNA-1273 (Moderna) – are also increasingly used because of their lower immunogenicity, easy preparation and low cost.

The importance of gene therapy, and the reason for this review, is that after many ups and downs and several stops and starts the field has reached a turning point where it may finally – lo! these many years – deliver on the promise to alleviate or even cure a range of previously untreatable diseases.

These include not only inherited genetic disorders such as sickle cell disease, cystic fibrosis and familial hypercholesterolaemia but also acquired diseases like cancer and HIV.

The different types and formats of gene therapy that each present with their own advantages and disadvantages are categorised and discussed below.

Under this category are armed oncolytic viruses. These replicate preferentially in tumours and express immunostimulatory genes that are intended to turbocharge the anti-cancer immune response.

Examples include T-VEC, an FDA-approved genetically modified herpes simplex virus-1 armed with the human cytokine granulocyte macrophage-colony stimulating factor (GM-CSF) for the treatment of melanoma.

There is also AdAPT-001, an experimental adenovirus in phase 2 clinical trials that is armed with a transforming growth factor-beta (TGF-β) trap for the treatment of sarcoma and colorectal tumors.

Also under this category is CRISPR (clustered regularly interspaced short palindromic repeat)/Cas9 (CRISPR-associated protein 9).

The most powerful gene-editing tool so far, CRISPR/Cas9 may edit, delete or deliver new genes. The main in vivo delivery method for CRISPR/Cas9 is with viral vectors like adeno-associated virus.

This is the case with gene therapy vectors based on adenoviruses, adeno-associated viruses (AAV), alphaviruses, flaviviruses, herpes simplex viruses, measles viruses, rhabdoviruses, retroviruses, lentiviruses, Newcastle disease virus, poxviruses, and picornaviruses that deliver specific cell function-altering genetic material to patients in vivo.

Most of these viruses have been ‘gutted’ or stripped of endogenous viral genes. To borrow a Texan American expression, they are ‘all hat, no cattle’, meaning, in this case, that they lack virulence properties and replicative power.

Nevertheless, even when all viral coding sequences are eliminated, gene therapies are not risk-free, as three well-known examples illustrate.

The first example occurred in 1999, when 18-year-old Jesse Gelsinger, who suffered from ornithine transcarbamylase (OTC) deficiency, died from a devastating inflammatory reaction to the adenovirus-based vector that delivered the corrective OTC gene.

The second came to light after the year 2000 when two of the 11 children in a French clinical trial that successfully received retrovirus-based treatment for the so-called ‘bubble boy’ disease, X-linked severe combined immunodeficiency (SCID-X1), developed leukaemia.

This was a tragic setback since the treatment cured 9 of the 11 children of SCID-X1.

The third happened in 2007 when Jolee Mohr, age 36, having received injections of an AAV vector that carried a gene that inhibited tumor necrosis factor α (TNF-α) for rheumatoid arthritis, died.

Even though this therapy was eventually held not responsible for her death, it cast a lasting pall over the field.

Gene therapy has largely rebounded from these tragedies as several viral vector-based success stories demonstrate.

The first of these are the COVID-19 adenovirus-based treatments such as the AstraZeneca and Johnson & Johnson vaccines.

These demonstrated excellent safety and vaccine efficacy in clinical trials even though the FDA briefly – for ten days – paused the use of the J&J vaccine due to a possible association with a rare blood-clotting syndrome called thrombosis with thrombocytopenia syndrome.

The second is the 2017 FDA approval of Luxturna for the one-time treatment of blindness due to any heritable retinal dystrophy caused by the mutated RPE65 gene.  Controversially, however, the treatment costs $425,000 per injection, or nearly $1 million for both eyes.

The approval of Luxturna paved the way for the approval of several other gene therapies.

They include Hemgenix, for the treatment of haemophilia B, Skysona for the treatment of active cerebral adrenoleukodystrophy in males between four and 17, Zynteglo for the treatment of beta-thalassaemia, Zolgensma for the treatment of spinal muscular atrophy, Vyjuvek for dystrophic epidermolysis bullosa and Elevidys for Duchenne Muscular Dystrophy.

Non-viral gene delivery of DNA and RNA is certainly less costly and potentially safer than viral-mediated delivery, but it tends to suffer from low transfection efficiency and a lack of cellular specificity.

Following the success of the non-viral delivery of mRNA for COVID-19, however, lipid nanoparticle (LNP)-delivered mRNA vaccines have reached the forefront of clinical translation.

Other viruses are all bug no feature. These include the first ever oncolytic adenovirus to reach the clinic, ONYX-015 and its successor, H101, approved in China for the treatment of nasopharyngeal carcinoma.

There are also several other viruses in development like OBP-301 (Telomelysin) from Oncolys BioPharma and DNX-2401 (Tasadenoturev) from DNAtrix.

Meanwhile, the concept of more than one bug and more than one feature is behind a so-called prime-boost strategy that involves the administration of more than one virus with more than one transgene or antigen to improve immunological outcomes.

All the rage on YouTube and other sites is ‘biohacking’, that is do-it-yourself, strategic shortcuts (that resemble Silicon Valley-type garage computer hacking) to improve health, wellness and longevity.

Perhaps the ultimate ‘biohack’, especially given the many comparisons of DNA to the ‘computer code of life’ is gene therapy. The purpose of gene therapy is to add, correct, or subtract genes for better health and wellness.

Indeed, despite notable setbacks like the Gelsinger, SCID-X1 and Mohr cases, gene therapy has recently taken off and features remarkable potential in an ever-growing number of chronic diseases such as SCID, HIV, COVID-19, muscular dystrophy, haemophilia, β-thalassaemia, sickle cell disease and cancer.

Leading the gene therapy charge are the armed oncolytic viruses like T-VEC and AdAPT-001, and viral vector gene therapies like Hemgenix, Skysona, Zynteglo, Zolgensma, Vyjuvek and Elevidys and lipid nanoparticle-based mRNA vaccines.

In conclusion, gene therapy is arguably best described both as a feature i.e., intentional and a bug i.e., a virus.

Bryan Oronsky is Chief Medical Officer at EpicentRx. Go to epicentrx.com