Shut Up, Genes (ASOs and RNAi)

Last updated on July 10, 2023

Antisense oligonucleotides (ASOs; sometimes abbreviated as AONs) are a form of gene silencing discovered in 1978. The first ASO drug was approved by the FDA 20 years later in 1998.

Another form of gene silencing is RNA interference (RNAi) and was discovered in 1998. The first RNAi drug was approved by the FDA 20 years later in 2018 (see below).

RNAi was heralded in the early 2000s as a candidate for SCA disease-modifying treatment. Both RNAi and ASOs have been successfully shown to quiet down polyglutamine SCA gene expression in non-human tests, but I think RNAi tests with SCA stopped before getting to humans. In 2015, ASOs made it to human trials for Huntington’s disease (a hop, skip, and a jump away from SCA3), with safety established in 2017.

In 2017, a form of RNAi that is worth being aware of is “short/small hairpin RNA” or shRNA.

By any other name

From my tiny perspective, which is perhaps teetering on accuracy, this is what happened:

RNAi for SCA showed promise in the early 2000s but was hampered by challenges with delivery (crossing the blood-brain barrier; BBB disruption issues) and specificity (not throwing the baby out with the bathwater). The RNAi approach to dealing with SCA fell out of the limelight.

Ten years passed.

Now researchers have achieved similar lab results after switching focus to a 20-year-older technique involving ASOs and now face similar challenges with delivery and specificity but are working through them. To bypass the blood-brain barrier, they went from intracerebroventricular injection (ICBI) to intrathecal injection (spinal canal injection).

We’ve almost been here before, and who knows what will happen next, or if successful to any degree, how long it will take? My guess is that CRISPR will be the next distraction, leaving ASOs and their limitations on the backburner, and adding years to the exploration process.

2018 progress with siRNA

Related to RNAi is siRNA (small interfering RNA). The following is not related to SCA, but the first siRNA drug was approved by the FDA on 2018-08-10 for the symptomatic treatment of polyneuropathy in adults with a neurodegenerative disease called hereditary transthyretin-mediated amyloidosis (hATTR).

Drug developer: Alnylam Pharmaceuticals, Inc.

Drug name: patisiran (trade name Onpattro).

Administration: Intravenous infusion every 3 weeks (presumably for one’s lifetime).  It uses a non-viral lipid nanoparticle delivery mechanism.

Cost: USD 450,000 per year, with a money-back guarantee, and discounts expected of USD 105,000.

Disease prevalence: Assuming accurate numbers are given, about 7 per million (2,200 in the U.S. in 2018), which is perhaps slightly more than SCA3, though fractionally less than the total SCA.

There is a nucleotide in the affairs of men

Perhaps it’s the computer scientist in me, but the term RNA interference is self-explanatory enough to not feel the need to delve into what it precisely means. But I cannot stand by idly with the term antisense oligonucleotide, which to me is at first quite cryptic.

Sense in genetics is like voltage polarity in electronics, having positive and negative, where—continuing with the analogy—voltages can combine and cancel each other out. Antisense is like that: unwanted nucleotides in a genetic sequence (e.g., CAG repeats) can be silenced (i.e., erased) by applying the negative (i.e., complement) of that sequence.

What is an oligonucleotide? We know and accept that DNA consists of strands of C-G-A-T nucleotides. Oligo– is simply the Greek prefix meaning few (i.e., a small number); think oligarchy. An oligonucleotide is a short (i.e., not long and not tiny) synthetic strand of nucleotides.

In other words, an ASO is a short strand of DNA or RNA synthesized for therapeutic purposes to cancel out an undesired nucleotide sequence in someone’s RNA. Unlike CRISPR/Cas9, which changes the original DNA, applying an ASO does not modify the original DNA. CRISPR/Cas13 is back to modifying the RNA not the DNA.

2019-01-29: This article talks about microRNA modifying cells to produce only fixed proteins, without modifying the genome/DNA.

An elephant’s faithful, one hundred percent

Back to SCA(3), I still meant what I said earlier, with some added nuance:

• The fundamental problem with gene silencing is that it can only repair some of the toxic protein as it’s created—not the existing protein and again only some of the new protein.  By definition, the best gene silencing will do is diminish the disease impact as degeneration continues; the disease won’t be prevented or stopped.
• (Non-viral) gene silencing via direct injections into the brain or spinal canal might temporarily quiet down some of the toxic protein creation within individual cells. How is that good enough? How toxic is the leftover and future toxic protein? That is not disease eradication in an individual but amelioration at best.
• New research takes decades to not succeed so it can go in another direction.
• In 2017, there are unmet challenges with ASO delivery to the brain and specificity. Years will pass; focus could continue to shift to other trendy techniques (e.g., CRISPR).
• To prevent an adult-onset genetic brain disease, better therapy than what exists in laboratories must be developed and (my guess) then applied starting in early childhood long before symptoms (though the zygotic-blastocystic-embryonic-fetal-newborn stages would be more ideal). Adults with symptoms? Way too late.
• We’re barely scratching the surface of disease amelioration. Maybe after many decades, the theoretical goals of pure gene silencing will be achieved, but then genetic defects such as SCA can and will be passed to offspring even when the parent is not symptomatic, ensuring that the disease is around forever.
• The efficacy of some trials is determined by examining the cerebrospinal fluid of participants. People are beside themselves to see toxic protein levels go down. This indicates amelioration to some degree, probably with the need for continual, life-long medication starting as soon as possible after one’s birth, and in any subsequent offspring if determined to have inherited the genetic defect—still with no long-term guarantees. If you are an SCA3 adult with decades of dead neurons in the cerebellum, it’s way too late to begin a timid, temporary, and partial repair process.
• The CGA repeat (partially) addressed by the ASO might only be the tip of the needed iceberg. For example, while we know the cerebellum is primarily affected, other parts of the nervous system might be significantly affected. We won’t really know until we try ASOs in humans. In 2021, as ASO trials for HD have started failing, we might be bumping up against reality.

I think that focus could be more on disease prevention rather than (inherently) imperfect treatment.

What’s the diff?

What’s the difference between ASOs and RNAi? Here’s a link to a very technical article on the topic.

The state of silencing and SCA3

• 2017-06-29: “ASO-Mediated Removal of the PolyQ Repeat in SCA3 Mice”
• 2017-04-12: “Evaluation of ASOs Targeting ATXN3 in SCA3 Mouse Models”
• 2018-06-20. Signs of life between SCA3 and Ionis. (The article uses strangely inaccurate prevalence information.)
• 2018-08-06: “ASO therapy mitigates disease in SCA3 mice”
• 2019-02-21 (through 2021-12): NCT03885167
  • “Identification of Biomarkers in SCA3″
• 2019-06-24: “ASO therapy rescues aggresome formation in a novel SCA3 human embryonic stem cell line”
• 2019-07-08:  “Suppression of mutant protein expression in SCA1 and SCA3 mice using a CAG repeat-targeting ASO”
• 2019-08: “ASO therapy rescues aggresome formation in a novel SCA3 human embryonic stem cell line”
• 2019-12-18: Universal RNAi.

Companies to watch: Ionis Pharmaceuticals (in partnership with Roche) and Wave Life Sciences. Update: In early 2021, their ASO trials started failing; see links for more information.

But what looks even more promising is an RNAi approach used by uniQure, the details of which became apparent in late 2018.