1/12 Quick thread on $ARWR's TIDES presentation yesterday on ARO-MAPT, their Alzheimer's drug.
Most of the deck has been public since last September. But two new data points closed the last open question on the preclinical case.
Here's what they mean:
2/12 Quick primer. Alzheimer's has two bad proteins building up in the brain: amyloid and tau.
Leqembi and Kisunla go after amyloid. Modest benefit, brain bleeding side effects, no tau reduction
Next frontier is tau. Tangle burden, not amyloid, predicts how fast patients decline
Leqembi and Kisunla go after amyloid. Modest benefit, brain bleeding side effects, no tau reduction
Next frontier is tau. Tangle burden, not amyloid, predicts how fast patients decline
3/12 ARO-MAPT is $ARWR's tau drug.
MAPT is the gene that makes tau. ARO-MAPT is a small RNA that finds the gene's instructions, destroys them, and stops new tau from being made.
Less tau being made = less tau available to misfold into the toxic tangles that kill neurons.
MAPT is the gene that makes tau. ARO-MAPT is a small RNA that finds the gene's instructions, destroys them, and stops new tau from being made.
Less tau being made = less tau available to misfold into the toxic tangles that kill neurons.
4/12 The bear case on every tau-lowering drug:
Tau exists in healthy brains too. Only a chemically-modified version (phospho-tau) forms toxic tangles.
If you lower all tau, are you reducing the bad version or just the healthy pool while pathology marches on?
Tau exists in healthy brains too. Only a chemically-modified version (phospho-tau) forms toxic tangles.
If you lower all tau, are you reducing the bad version or just the healthy pool while pathology marches on?
5/12 Until yesterday, $ARWR had shown they could lower total tau. They had NOT shown the bad version specifically coming down.
That was the last open question on the preclinical case.
That was the last open question on the preclinical case.
6/12 REVEAL #1: In disease-model mice, ARO-MAPT cut the toxic phospho-tau species in the hippocampus by 53%.
The hippocampus is the memory center where Alzheimer's starts.
The bad tau, in the region the disease attacks first, is cut by half. The bear case is answered.
The hippocampus is the memory center where Alzheimer's starts.
The bad tau, in the region the disease attacks first, is cut by half. The bear case is answered.
7/12 Caveat: these are engineered mice, not aging humans. Mouse models are useful but imperfect.
Phase 1/2a is built to test this translation. One exploratory endpoint is Tau PET, a brain scan that shows tangle burden directly in living patients.
First subjects dosed Dec 2025.
Phase 1/2a is built to test this translation. One exploratory endpoint is Tau PET, a brain scan that shows tangle burden directly in living patients.
First subjects dosed Dec 2025.
8/12 REVEAL #2: Quarterly dosing is now the working case.
$ARWR extended their NHP study to 16 weeks past the last dose. Gene activity recovered, but tau protein stayed down 50-65% from normal.
Tau is a long-lived protein. The pool gets depleted and takes a long time to refill.
$ARWR extended their NHP study to 16 weeks past the last dose. Gene activity recovered, but tau protein stayed down 50-65% from normal.
Tau is a long-lived protein. The pool gets depleted and takes a long time to refill.
9/12 Why this matters commercially:
Leqembi: Q2W IV infusions for 18 months, then monthly IV or weekly SC
Kisunla: monthly IV infusions, indefinitely
ARO-MAPT (if it tracks the monkey data): one SC injection at home, every 3 months. From day one.
Leqembi: Q2W IV infusions for 18 months, then monthly IV or weekly SC
Kisunla: monthly IV infusions, indefinitely
ARO-MAPT (if it tracks the monkey data): one SC injection at home, every 3 months. From day one.
10/12 For a patient with mild cognitive impairment who's still working, driving, managing their own life, quarterly at-home is a different product than monthly clinic visits.
It's also what makes the rarer tauopathies (PSP, CBD, FTD) economically viable to treat.
It's also what makes the rarer tauopathies (PSP, CBD, FTD) economically viable to treat.
11/12 Competitive read:
Biogen's BIIB080 (CELIA Phase 2 reads out this year) also lowers tau. Gets credit for proof-of-concept.
But BIIB080 needs a spinal tap. $ARWR's own data shows spinal-tap drugs can't reach the deep brain regions where tauopathies live.
Biogen's BIIB080 (CELIA Phase 2 reads out this year) also lowers tau. Gets credit for proof-of-concept.
But BIIB080 needs a spinal tap. $ARWR's own data shows spinal-tap drugs can't reach the deep brain regions where tauopathies live.
12/12 Three numbers from Phase 1/2a will tell us if ARO-MAPT translates:
CSF tau needs ~50%+ (BIIB080 hit ~60%)
Plasma p-tau217 needs to move
Tau PET directional reduction is enough
Hit all three → re-rating event. No cognition data needed.
Full note: bioboyscout.com
CSF tau needs ~50%+ (BIIB080 hit ~60%)
Plasma p-tau217 needs to move
Tau PET directional reduction is enough
Hit all three → re-rating event. No cognition data needed.
Full note: bioboyscout.com
WOW! $BIIB's CELIA validates the tau hypothesis. Robust biomarker reductions plus cognitive signals in randomized Phase 2 = first clinical proof tau-targeting works. The primary endpoint miss and inverse dose response actually strengthen the case for $ARWR's ARO-MAPT.
Generated by Thread Navigator
Press ⌘ + S to quick-export