Yesterday
was a memorable day in the development of drugs for two of the main severe
monogenic diseases without treatments to change disease trajectory: Huntington’s
Disease and myotonic dystrophy type I.
A lot has
been written in the press and social media on the fantastic results by uniQure
in Huntington’s Disease (see also my preview on why I had thought exon 1-targeting,
including AMT-130, is the most promising current approach), so I focus on yet
another biotech data bomb dropped last time: single-dose data from the 15mg/kg
cohort in PepGen’s early-stage trial in myotonic dystrophy type I. In short, similar to AMT-130, the data blew
away expectations in terms of the key prognostic biomarker in DM1, the CASI splice
index, leaving the more advanced competition by Avidity and Dyne in the dust.
To knock
down or to disrupt
DM1 is an
autosomal dominant disease caused by triplet expansion in the 3’ UTR of DMPK. When triplets exceed 50 copies, they form an
RNA structure that attracts and thereby interferes with muscle blind (MBNL1)
protein to carry out its functions in RNA splicing (spliceopathy). By measuring the degree of splice disruption
of a number of MBNL1-target genes, the CASI score is therefore a key prognostic
biomarker.
People
living with DM1 experience muscle weakness, including cardiac problems, and
myotonia, in addition to developing neurological, endocrine, and other systemic
symptoms. Such systemic disease manifestation is normally observed in muscle diseases and it is often unclear whether this relates to gene expression (here DMPK1) in non-muscle cells or whether they are the result of feedback mechanisms.
Early
efforts in developing disease-modifying treatments have focused on knocking
down DMPK1 either by RNAi (Avidity Biosciences, AOC-1001) or RNaseH antisense (Dyne
Therapeutics, DYNE-101). The idea here
is that by lowering the abundance of DMPK, less MBNL1 will be sequestered. Indeed, both Avidity and Dyne have shown
convincing data that this approach works in addressing muscle disease as
measured by the vHOT myotonia assay.
Both
companies target the oligonucleotide with antibody/antibody fragments targeted to Tfr1
which is expressed on muscle cells.
Interestingly, Tfr1 is also expressed on the blood-brain-barrier and may
thus allow for addressing some of the prominent neurological symptoms as well.
The
downside to the knockdown approach is that knocking down DMPK1 too much may be
toxic based on the lack of DMPK1 knockouts in human genome databases. Also, I expect this approach to lose efficacy
with the size of the triplett expansion.
By
contrast, antisense oligos targeted to the toxic RNA structure should be soaked
up synergistically the longer the repeats and thus higher local repeat concentrations.
Binding a few repeats may unravel the structure and free MBNL1 to do its
job. There should also be no theoretical
limit as to how many target transcripts can be engaged without causing
target-related toxicity.
The two
lead candidates for the DMPK1 structure disruptors, PGN-EDODM1 by PepGen and
VX-670 from Vertex Pharmaceuticals, comprise of PMO-based oligos conjugated to
cationic peptides for broad biodistribution and enhanced cellular uptake and thus may have the
additional benefit of addressing DMPK1 toxicity in more cell types than the
Tfr1-targeting drugs.
So while
Avidity and Dyne are on track to see their DM1 knockdown drugs approved in late
2026 and 2027, respectively, the RNA structure disruptors led by PepGen and
Vertex Pharmaceuticals need watching as they represent a very differentiated approach
and the efficacy could thus be in an entirely different ballpark.
PepGen’s
Disruptive Biomarker Data
And in an
entirely different ballpark indeed they are based on data presented last night by PepGen
from the 15mg/kg single-dose portion of their early-stage trial. Though the previously reported single-dose 10mg/kg
data were numerically the highest observed in the clinic so far (29.1% vs 16.1%
for 2 doses of AOC-1001 and 25% for 2 doses of DYNE-101; discussed in a blog entry here), the newly reported 53.7%
splicing improvement based on the 28-gene panel (CASI) is opening up entirely
new horizons for what should be possible.
The
surprising increase in efficacy (I was looking for 15mg/kg mainly to confirm
the highly promising 10mg/kg results) correlated with muscle concentrations of
PGN-EDODM1 that were more than dose proportional. This is a critically valuable insight as the company is
continuing with the multiple-ascending dose portion of the study with results
expected in early 2026. But before that, expect more details from the 15mg/kg single-dose portion on October 11 at the World Muscle Society meeting.
Results
from VX-670 may also be reported over the coming months. Although I may be biased here as a PepGen shareholder, I am a bit cautious with regard to the safety and efficacy of VX-670 since
the partner from Vertex, Entrada Therapeutics, has been advertising how much
better their new delivery peptides are for muscle delivery than the one used
for VX-670.
Disclosure:
I owned close to 5%
of PepGen as the ~$35M market cap just a month ago was just a crazy disconnect
compared to the $6B market cap of Avidity Biosciences when PepGen had the
stronger CASI numbers. After reporting a
$100M offering (priced at $3.2 a share), my ownership will be diluted to about
2.5%. Frankly, the $100M pre-money
valuation of the offering is theft in broad daylight and the company should
have reported data first and then do the offering after further price discovery. The pre-money would have been at least $300M
in my opinion. Removing the ‘financing overhang’ along with
announcing data is so yesterday, the new biotech playbook allows for offerings in
the days after data. This is a new
chapter in publicly traded biotechs where stocks react to data.
