MDS 2026 Debrief: Two Genetic Bets Fall in Full Data & Juvmo Meets Its Payers

In full data shown at MDS, Biogen and Denali's LRRK2 drug let early Parkinson's progress just as it did on placebo, and more patients stopped the drug than stopped placebo.

Bial's GCase activator, built for gene carriers, fell short in its full data too, and Bial has closed the programme. Two genetic bets are down, so the question is what still stands.

Welcome to The Pharma Closeout MDS 2026 Debrief. I'm Alex Mercer.

And I'm Maya Patel. Our MDS 2026 Debrief brings you the readouts that matter, what they change, and what to watch next.

The International Parkinson and Movement Disorder Society met in Seoul from October 4 to 8, and we followed it for what it means for drugs, diagnostics and payers. The two full-data defeats come first.

Our top story from MDS. Biogen and Denali showed the full LUMA data for their oral LRRK2 inhibitor in 648 people with early Parkinson's, and the May topline held. Against placebo it failed to slow time to confirmed worsening on a combined daily-living and motor score, with a hazard ratio of 1.04.

Did the drug even get into the brain?

It reached the spinal fluid, the nearest proxy we have, because Biogen and Denali said in May that drug levels there matched expectations. The weak part is the effect. A blood marker of LRRK2 fell more than 90%, but pRab10, a pathway marker in spinal fluid, fell only about 30% at its peak, in 86 people. Our read, and it is only a read, is that the drug could have engaged its target weakly in the brain.

So the drug got there, and engagement is the open question. The Michael J. Fox Foundation relays one LUMA investigator's hedge, that enrolling mostly non-carriers could have affected the results.

The complete dataset carries a tolerability cost too, with 11.2% stopping for adverse events against 4.6% on placebo, and the secondary endpoints missed as well. Denali is carrying the drug forward in carriers of disease-causing LRRK2 variants, in BEACON, a Phase 2a measuring safety and pathway biomarkers, with data due in the first half of 2027.

So the next readout is a biology check in carriers rather than an efficacy test, and partners wait longer for an answer.

Bial's ACTIVATE miss has been public since early June, and Seoul filled in the detail. Roughly 270 people with GBA1-linked Parkinson's were randomized, about a third to placebo, and over 78 weeks neither dose of the GCase activator slowed worsening in daily living or on the key motor secondary.

Put LUMA beside it and I think the genetics-first strategy is under real pressure. Both targets came out of human genetics, the full data from both misses came out at the same congress, and I'd expect partners weighing the next genetically derived Parkinson's deal to ask much harder questions about engagement.

I read it the other way. LUMA tested LRRK2 inhibition in a broad early-Parkinson's population, while ACTIVATE tried to raise GCase activity in carriers only. Two targets, two populations, and Bial hasn't published whether GCase activity rose.

Bial had 78 weeks of results and chose to close the programme, which outweighs any reading of the biology.

For this activator, yes. Commercially the enzyme stays in play, with Genentech licensing a preclinical enzyme replacement from Alector, the deal we covered on October 6.

The bets still standing start small. Arvinas brought 28 days of data on ARV-102, its LRRK2 degrader, in 24 people who completed dosing, too brief to test benefit. Its March first look grew into exploratory changes in eye movements against placebo and in spinal-fluid proteins.

The harder fact sits in Arvinas's August quarterly filing. A planned trial in the United States in progressive supranuclear palsy, a related movement disorder, was on FDA clinical hold as of that filing, while the agency reviewed long-term primate toxicology, and Arvinas plans those PSP trials for 2027, subject to clearance.

A UCL Queen Square late-breaker points at another way to pick patients. In 379 people from the PPMI cohort, spinal fluid carried more pRab10, the pathway marker LUMA tracked, in ordinary sporadic Parkinson's than in healthy controls, and LRRK2 carriers, with or without symptoms, ran higher too.

As the authors read it, the pathway looks switched on even when the mutation is absent. Their hypothesis is that genetic and alpha-synuclein testing together could pick out sporadic patients for LRRK2 drugs, a second route into trials alongside Denali's carrier enrolment.

The bet already in Phase 3 began with a miss. Roche's prasinezumab, an antibody against clumped alpha-synuclein, missed its primary endpoint in PADOVA in December 2024, and posters in Seoul repeated analyses first shown in March that lean its way, uncorrected for testing many things at once.

Those are hints. Roche argues from the totality of evidence, and PARAISO, a Phase 3 recruiting about 900 people, now decides the antibody's future.

On the symptom side, we start with Juvmo. It is AbbVie's tavapadon, the once-daily D1 and D5 partial agonist whose approval we covered in our September 27 Week in Review, and AbbVie brought four analyses of its own to MDS, three of them post hoc or indirect, to build its prescriber story.

Payers will read ICER instead. At a reported list price of $58,800 a year, from its October 1 meeting slides, adding Juvmo for people already on levodopa runs about $10 million per quality-adjusted life year against agonists plus levodopa. For newly diagnosed people, ICER's model has it costing more than generic levodopa and doing slightly less.

What did ICER's clinical ratings give AbbVie to work with?

One favourable grade and several middling ones. Against levodopa alone, ICER rates adding Juvmo C-plus-plus, which in its own words means at least as good, and possibly somewhat to substantially better, though discontinuation ran higher. Against agonists, MAO-B or COMT inhibitors added to levodopa, it stopped at promising but inconclusive.

That favourable grade is against adding nothing, which is rarely the choice a neurologist faces, so it works as AbbVie's pitch more than a prescribing guide. In early disease, AbbVie's own post hoc rollover analysis has fewer than 8% of participants starting levodopa over 58 weeks of open-label follow-up.

That 8% includes people first randomized to placebo, so it counts who began levodopa, not whether tavapadon held it off. Early disease is also where ICER's model is least kind, so payers will make neurologists fight for it, and fights are slow.

Slower, I'll grant you, and Citi expected gradual uptake as Medicare coverage negotiations progress, in a note before approval. AbbVie's October 30 call is its first since approval, so it gives the first earnings-call read on the launch.

Our October 7 show covered Cerevance's solengepras, which cut OFF time, the hours when medication wears off, by about 37 minutes a day against placebo on the higher dose. Seoul filled in its secondary endpoints, which Cerevance reports as positive, with the testing order unstated.

Would a patient actually notice 37 minutes?

The usual bar for a change one patient would feel is about an hour, and 37 minutes is a group average, so it can't answer that by itself. Approved add-ons cut OFF time by roughly 0.4 to 1.4 hours against placebo, depending on the class and the trial pool, which puts solengepras in the lower part of that range, ahead of istradefylline and behind the other classes.

Lower shelf in a crowded aisle, then.

So what's the commercial case for it at that size?

For now it rests on mechanism and one outside analyst. GlobalData's Christie Wong argues it stands apart because GPR6 works away from the dopamine receptors, and she limits that case to add-on use. The trial's presenter, a study investigator, is quoted in Cerevance's release, so that isn't an independent view.

Cerevance's next step is an FDA meeting about a possible application. Before we go earlier in the disease, a word about the show.

If you find the show useful, subscribe to or share The Pharma Closeout on Spotify, Apple Podcasts, or wherever you listen. We appreciate you listening.

Now, to the years before diagnosis. People who act out their dreams have isolated REM sleep behaviour disorder, which often comes before Parkinson's or a related disease. Syntara randomized 41 of them in Sydney and Oxford to its pill SNT-4728 or placebo for 12 weeks, and MDS billed the result a Clinical Breakthrough.

Which result earned that billing?

The main measure was a PET scan for brain inflammation, and Syntara's preliminary analysis found a lower signal than at baseline in 20 of the 30 treated people, in the putamen on one side only. At a follow-up visit 12 weeks after the last dose, a motor score also separated from the 11-person placebo arm as a secondary endpoint, while the other motor and thinking measures stayed flat.

The pill also blocks MAO-B, an established Parkinson's target, alongside the inflammation enzyme SSAO, so a motor effect could come from either.

On our analysis, a medicine started before diagnosis could be taken for years, if regulators accept an endpoint for this stage. For Syntara, a preliminary result in a small trial, plus a plenary billing, is what it now takes to investors and partners.

In people with that same sleep disorder, a skin biopsy found phosphorylated alpha-synuclein, the misfolded Parkinson's protein, in 57 of the 70 followed for a year in the Syn-Sleep study, 81.4%, and everyone positive at the start stayed positive.

Skin testing for this protein is already offered commercially in the United States, so Seoul moves the research evidence for this sleep disorder. The one outside neurologist quoted in the MDS release wants longitudinal analyses to show what the test is worth in real-world use and to help plan trials.

Beyond Parkinson's, the nearest FDA decision we're tracking is in essential tremor. Praxis brought the full Study 1 dataset for ulixacaltamide, an investigational T-type calcium channel pill, and on a daily-living tremor scale it beat placebo by 2.57 points.

About a third were already on propranolol, the only FDA-approved pill for it, and in them the gap was 3.0 points, in line with the whole trial. Did the FDA raise anything at its mid-cycle review?

Praxis says the FDA's mid-cycle review found no major safety or efficacy concerns to date.

For children and teenagers with Tourette syndrome, Teva has an FDA decision coming on ecopipam, a dopamine D1 blocker it calls first-in-class, picked up with Emalex in June. Priority review puts the target late in the first quarter of 2027, for pediatric patients only.

The Seoul data come from an open-label extension, where everyone knew they were on the drug. Tic severity was down 45.6% from where the extension began at month 18, but only 33 had reached that visit, and 3.4% stopped for adverse events.

Teva's controlled trial is a separate randomized-withdrawal Phase 3 called D1AMOND, outside the Seoul data. Adults made up about a sixth of the extension and sit outside this filing, so any adult use waits for another decision.

Huntington's disease brings the longest follow-up yet for a VMAT2 inhibitor. In Neurocrine's five-year open-label extension of valbenazine, marketed as Ingrezza, chorea scores stayed below baseline out to week 260, then drifted back once treatment stopped.

Median time on treatment was 980 days, about 2.7 years, so the five-year mark rests on fewer people. Serious adverse events hit 24.7%, mostly judged unrelated by investigators, and 45.5% reported falls.

Ingrezza's commercial pressure comes from Teva's Austedo, the rival VMAT2 drug, which also treats tardive dyskinesia, its bigger market. Medicare's negotiated price for Austedo applies across all its uses, and on our analysis that puts pressure on Ingrezza.

Four dates to hold. Huntington's gets its next big stage at the European Huntington's Disease Network congress in Krakow, October 22 to 24, where the preliminary programme lists GENERATION HD2 results for tominersen, and votoplam's Phase 2 results with a Phase 3 update, on the closing day.

Juvmo's payer argument gets its fullest public record on November 2, when ICER publishes its final report and the summary of the October 1 panel.

No written vote tally is posted yet, so watch that summary for it.

For Ingrezza, the Austedo date to watch is January 1, 2027, when its negotiated Medicare price takes effect.

Medicare's negotiated price is $4,093 for a 30-day supply, against the 2024 list price of $6,623 that CMS used.

Then essential tremor gets its answer. Praxis carries a Breakthrough Therapy designation from December 2025 into the FDA's call on ulixacaltamide, due January 29, 2027, and the agency has told Praxis no advisory committee is planned.

The remaining bets now run through Denali's carrier study, Arvinas's degrader, whose PSP trials wait on FDA clearance, and Roche's PARAISO, with primary completion estimated for 2029. Thanks for following Seoul with us. It was a pleasure.

Follow the show on Apple Podcasts, Spotify, or wherever you listen — a quick rating or share helps other listeners find us. The daily briefing and the Week in Review carry on as usual.