The launch is real and the addressable population is not, at least not yet. India funds roughly 70% of health spending out of pocket, so a biweekly infused biologic with per-vial pricing at Rs 21,682 sits outside any meaningful reimbursement mechanism. The binding constraint is not price, it is the diagnostic and monitoring pathway. Eligibility requires biomarker confirmation, and ARIA surveillance requires serial MRI in a system where specialist memory capacity and imaging access are concentrated in a handful of metros.

• Watch the biomarker route, not the drug. Eisai specified CSF or serum beta-amyloid rather than PET, which is the pragmatic choice for India and the same argument C2N is making in UK primary care. A blood-first pathway is the only version of this that scales here.
• The Phase IV is doing double duty. Local data supports clinical acceptance, but it also builds the health-economic file any future public procurement or insurer negotiation will need. Read it as a market access asset.
• Competitive frame: first DMT into a market with no incumbent and no reimbursement precedent means Eisai sets the pathway others inherit. Donanemab and the subcutaneous maintenance formulation follow into a route Leqembi will have defined.

Read the endpoint hierarchy before the adjectives. The prespecified primary was pharmacodynamic, a blood inflammatory marker panel, not a clinical outcome. That is a legitimate and honest design for proof of mechanism in 60 patients, and it means the MDS-UPDRS results sit downstream of the primary rather than carrying it. A 60-patient trial is not powered to establish motor benefit, and no multiplicity handling has been described for a composite of 15 measures.

• The 283 of 380 figure is directional, not inferential. A proteome-wide count of proteins moving the expected way is hypothesis-generating; without a prespecified signature and correction, it cannot support a disease-modification claim and should not be quoted as if it does.
• What would settle it: effect size and confidence intervals on MDS-UPDRS Part III, and whether the levodopa-naive design held without rescue. Both should appear on the 12 August call. If they do not, that is itself information.
• Context: bezisterim is also in ADDRESS-LC in long COVID with topline due late summer, and has prior Alzheimer’s Phase 2 and Phase 3 history. A single mechanism spanning three neurological indications is a platform claim, and platform claims need one controlled win to anchor them.

The designation is procedural. The word that matters is off-the-shelf, because it is the entire commercial argument for cell therapy in Parkinson’s. Autologous derivation is what has kept this field in academic hands: bespoke manufacture cannot serve a disease with roughly ten million patients worldwide, whatever the graft survival data show.

• Competitive frame: BlueRock’s bemdaneprocel is the reference programme and is already in a Phase 3 with allogeneic iPSC-derived cells behind Bayer’s balance sheet. XellSmart is contesting the same manufacturing thesis without that backing, so the differentiator has to be cost of goods or potency per dose, not concept.
• The unresolved question is immunosuppression. Allogeneic grafts into the striatum have historically required a year or more of immunosuppression, which materially changes the risk-benefit in a population averaging 60-plus. Any Fast Track filing should be read against whether the regimen is being shortened.
• Fast Track buys meetings, not evidence. Note it as a signal that FDA accepts serious unmet need in a disease with symptomatic therapy but no disease modification, and hold judgement until first-in-human graft survival and motor data.

The value here is trial economics rather than diagnosis. Parkinson’s trials are slow and expensive because MDS-UPDRS is rater-dependent, episodic and noisy, which forces large samples and long follow-up. A continuous, hardware-free measure that predicts decline lets a sponsor enrich for progressors and shorten follow-up, which is the same manoeuvre the ALS proteomic panel offers in a much smaller population.

• The commercial customer is a sponsor, not a clinic. Read this as a clinical trial services play. The buyers are the Parkinson’s programmes in this issue, several of which have exactly the endpoint problem this addresses.
• What to demand: test-retest reliability, performance across lighting and device conditions, and whether prediction holds independent of baseline MDS-UPDRS. A biomarker that only restates the clinical score adds nothing.
• Regulatory path is the gate. Digital endpoints require FDA qualification through the DDT programme to serve as primary outcomes. Until then this is enrichment and exploratory, which is useful but does not change a filing.

The competitive position is awkward and the company knows it. Tofersen is approved in SOD1-ALS, so RAG-17 is not entering an empty indication; it is entering one where an antisense oligonucleotide already holds accelerated approval on neurofilament reduction. A siRNA has to argue either deeper or more durable SOD1 knockdown, or a better intrathecal dosing interval, and the Phase II design should tell you which claim they are building.

• The number to demand is CSF SOD1 protein reduction with a dosing interval attached. Percentage knockdown alone is not differentiating against an approved comparator. Duration between doses is where a siRNA can genuinely beat an ASO.
• The strategic asset may not be the drug. SCAD is a delivery platform, and a controlled intrathecal dataset in a small genetically defined population is the cheapest possible validation of it. Watch whether Ractigen partners the chemistry rather than the asset.
• Financing context: $31 million funds this readout, not a Phase 3. Expect a partnership conversation to open on Phase II data rather than after it.

A DSMB clearing escalation is a safety statement, and for this platform specifically that is not a formality. PepGen’s history is the context. The company discontinued its Duchenne programme on the same delivery chemistry, which means every safety gate PGN-EDODM1 clears is also read by the market as a verdict on the EDO platform rather than on this asset alone.

• The number that matters is splicing correction, not dose. DM1’s mechanism runs through sequestration of splicing factors, so the readout to demand is corrected splicing index in muscle biopsy with a dose-response, plus whether correction translates to myotonia or grip strength.
• Competitive frame: Avidity’s del-desiran is well ahead in DM1 with a different delivery approach and a Phase 3 running. PepGen is arguing delivery efficiency into a race where the leader has already shown it can reach muscle.
• Watch cohort size and follow-up duration. Escalation announcements are cheap; the informative disclosure is how many patients sit at the top dose and for how long.

The population is the point. The VA is a single integrated payer that also delivers the care, which makes it the one US environment where a device that displaces drug spend can be evaluated on total cost rather than acquisition cost. Medication-sparing is the correct endpoint for that buyer, and it is a materially easier argument than superiority against a gepant.

• Real-world means unblinded, and unblinded means expectancy. A wearable neuromodulation device carries a large placebo component, and a study of patients choosing to skip medication cannot separate device effect from that. Treat it as utilisation evidence, not efficacy evidence.
• Where it travels commercially: VA and DoD formularies first, then integrated systems and risk-bearing groups. It travels poorly to commercial fee-for-service, where the device sits under DME and the drug budget belongs to someone else.
• The adjacent argument is opioid avoidance in a veteran population, which is a policy lever rather than a clinical one. If Theranica has that analysis and has not published it, that is the number to ask for.

Acute optic neuritis is one of the few places neuroprotection can be tested cleanly, and that is why this indication keeps attracting candidates. You have a dateable injury, an accessible readout in retinal nerve fibre layer thinning on OCT, and a short window. Most neuroprotection programmes fail because the insult cannot be timed. Here it can.

• The regulatory question is whether OCT structural preservation is an acceptable endpoint on its own. Pre-IND alignment on development strategy suggests a path was discussed; the disclosure to demand is whether FDA indicated a structural primary or required visual function.
• Read the indication as a wedge. A neuroprotective agent that works in optic neuritis has an argument in MS more broadly, and that is where the value sits. Acute optic neuritis alone is a small commercial opportunity.
• Pre-IND feedback is the earliest possible disclosure. No IND has cleared, no patient has been dosed. Log it as directional and wait for the IND.

The Rare Pediatric Disease designation is the substantive one, because it carries potential eligibility for a priority review voucher on approval. For a company at this stage the voucher is frequently worth more than early revenue, and it is what makes an ultra-rare paediatric CNS programme financeable at all. Recent vouchers have transacted around $100 million.

• Check the sunset. The RPD voucher programme has been repeatedly extended and repeatedly close to lapsing, and eligibility depends on designation and approval timing. Any valuation that assumes a voucher should carry an explicit assumption about the programme still existing at approval.
• SSADH deficiency is well suited to gene therapy and badly suited to trial design. A single enzyme deficiency with a clear genetic cause is the easy part. Prevalence in the low hundreds worldwide, heterogeneous presentation and no accepted endpoint is the hard part.
• What to watch: whether Galibra secures a natural history dataset before IND. In ultra-rare CNS, the external control is usually the trial design, and the foundations that own those registries hold real leverage.

This is the second patient organisation in as many weeks to buy access to the same capsid, after the TSC Alliance completed its preclinical pilot with Apertura. The pattern is patient foundations funding the delivery layer rather than a candidate, which is a smarter use of scarce philanthropic capital than backing one asset: a validated capsid serves every programme in the disease, including ones not yet designed.

• Angelman is the right test case. The disease requires broad neuronal transduction to restore UBE3A, and intrathecal antisense approaches have shown the ceiling that limited distribution imposes. If TfR1 CapX delivers whole-brain coverage intravenously, it changes what is possible here specifically.
• Note the concentration risk. TfR1 is now the shared access route for Apertura’s capsid, Dyne’s muscle conjugates and several CNS programmes. A single receptor carrying this much of the field’s delivery strategy means one safety signal would travel a long way.
• Terms were not disclosed and probably are not the point. What to watch is whether FAST secures rights broad enough to sublicense to a commercial partner later, which is the mechanism by which the TSC and n-Lorem models convert foundation money into approved products.