AC Immune ($ACIU) Stock Hub 2026: Fast Track For ACI-7104, The VacSYn Week-100 Readout And A Runway Into Q4 2027

01 Executive Summary

AC Immune SA is a clinical-stage biopharmaceutical company based at the EPFL Innovation Park in Lausanne, Switzerland, listed on Nasdaq since 2016 and reporting in Swiss francs under international accounting standards. It designs active immunotherapies, which are vaccines that train a patient’s own immune system to produce antibodies against a misfolded protein, and small molecules aimed at the same targets. The proteins in question are alpha-synuclein in Parkinson’s disease, amyloid beta and phosphorylated Tau in Alzheimer’s disease, TDP-43 in amyotrophic lateral sclerosis, and the NLRP3 inflammasome, which sits upstream of neuroinflammation in several of these diseases.

The structure of the business is unusual and worth stating plainly, because it explains both the revenue line and the risk. Everything that generates cash today is partnered: Takeda holds an option and licence on the anti-amyloid immunotherapy ACI-24, Eli Lilly funds the small-molecule Tau programme, and Johnson & Johnson’s Janssen unit holds the phospho-Tau immunotherapy ACI-35. Everything that could re-rate the company is wholly owned and unpartnered: ACI-7104 in Parkinson’s disease, the NLRP3 inhibitor ACI-19764, and the TDP-43 imaging tracer ACI-19626. The partners pay the bills; the owned assets carry the outcome.

On August 11, 2026 the U.S. Food and Drug Administration granted Fast Track designation to ACI-7104 for early-stage Parkinson’s disease and cleared the company’s Investigational New Drug application. The two decisions arrived together and do different things. The IND clearance is the operative one: it allows the ongoing Phase 2 VacSYn trial, which today runs at twelve sites across Germany, Spain and the United Kingdom, to open sites in the United States. Fast Track is a process designation that permits more frequent written and oral communication with the agency and makes a submission eligible for rolling review and, if the criteria are met at the time, for accelerated approval or priority review. Neither decision is evidence that the drug works, and neither shortens the trial.

The financial position is a small one, held together by partner payments. Total cash resources were CHF 75.4 million at June 30, 2026, about $93.0 million at the August 11 exchange rate, composed of CHF 27.8 million of cash and cash equivalents and CHF 47.6 million of short-term financial assets. That compares with CHF 91.4 million at December 31, 2025. The company states that the position funds operations into the fourth quarter of 2027 and that the statement excludes any further milestone payments. Revenue for the first half of 2026 was CHF 16.2 million against CHF 2.3 million a year earlier, entirely from partner arrangements: CHF 11.0 million recognised on the amended Lilly agreement and CHF 5.2 million on the Takeda arrangements. Research and development expense fell to CHF 22.2 million from CHF 32.7 million, which the company attributes to the pipeline focus programme announced in the third quarter of 2025 and to non-recurring manufacturing costs in the prior period, partly offset by CHF 1.9 million of new spending as the NLRP3 inhibitor entered Phase 1. The net loss narrowed to CHF 12.9 million from CHF 40.2 million, and the loss per share to CHF 0.13 from CHF 0.40.

The balance sheet carries one line that deserves attention before any judgement about equity. Total liabilities of CHF 115.2 million include CHF 89.5 million of short-term deferred contract revenue and CHF 0.8 million of long-term deferred contract revenue. That deferred revenue is the unrecognised remainder of partner payments already received, principally the $100.0 million Takeda upfront of May 2024, and it will be released into revenue as the underlying work is performed rather than being repaid in cash under normal circumstances. Shareholders’ equity of CHF 35.2 million therefore understates the funded position, while the accumulated deficit of CHF 450.4 million records what it has cost to get here.

02 What The August 11 Decision Does And Does Not Do

The announcement of August 11, 2026 contains two separate regulatory acts that are easy to conflate. Reading them apart is the difference between understanding the news and repeating a headline.

IND clearance

An Investigational New Drug application is the authorisation a sponsor needs before administering an experimental drug to patients in the United States. Until August 11, 2026, VacSYn had no such authorisation, which is why the twelve registered sites are in Germany, Spain and the United Kingdom and none is American. Clearance removes that constraint. In practical terms it allows AC Immune to add United States investigators, to recruit from a much larger patient population, and to generate a data set that the FDA has watched being produced under its own regulations rather than one imported from European practice. For a company that intends to file for approval in the United States, having no American clinical experience is a structural gap, and this is the step that closes it.

Fast Track designation

Fast Track is granted to programmes intended to treat a serious condition where non-clinical or clinical data demonstrate the potential to address an unmet medical need. What it confers is process: more frequent meetings and written communication with the review division, eligibility for rolling review of a marketing application, and eligibility to be considered for accelerated approval and priority review if the relevant criteria are met at the time of submission. What it does not confer is any conclusion about efficacy, any reduction in the evidentiary standard, or any acceleration of the trial itself. Designations are also revocable if a programme no longer meets the criteria.

The company’s own framing is consistent with that reading. The release states that the designation “reflects the early interim Phase 2 data demonstrating safety, tolerability and immunogenicity, and providing first signals that targeting underlying a-syn pathology with an active immunotherapy could potentially impact PD”. The word doing the work in that sentence is “signals”. Interim results reported at 76 weeks met their immunogenicity targets with a 100% responder rate and showed no clinically relevant safety issues; the efficacy language throughout has been about trends and stabilisation, not about a met endpoint.

Why the sequence matters

The two decisions arriving on the same day suggests a coordinated package rather than a routine filing, and it lands roughly four weeks before the window in which the full Part 1 data set is due. Interim Phase 2 results dated December 11, 2025 have been public for eight months, and the company said then that it would discuss a development plan with regulators to move towards registration. The Fast Track grant is the first visible output of that discussion. It does not tell an outside observer what the FDA thinks of the biomarker package, what size or design of registrational trial has been contemplated, or whether the agency would accept anything short of a clinical endpoint on the Movement Disorder Society Unified Parkinson’s Disease Rating Scale.

03 What The Company Actually Is Today

AC Immune has no approved product, no product revenue and no commercial organisation. Its entire income statement is partner money: upfront payments, milestones and research funding recognised over the periods in which the associated work is performed. That makes the revenue line lumpy in a way that a reader used to commercial companies can misread. Revenue rose from CHF 2.3 million to CHF 16.2 million between the first halves of 2025 and 2026, an increase of more than six times, and none of it reflects demand for anything.

Two technology platforms sit underneath the pipeline. SupraAntigen is the active immunotherapy platform: short peptide antigens presented in a way designed to raise antibodies against a specific conformation of a misfolded protein while avoiding a T-cell response against the normal form. Morphomer is the small-molecule platform: brain-penetrant compounds designed to bind pathological protein aggregates inside the cell, where an antibody cannot reach. The distinction matters because the two platforms address the same targets from opposite sides of the cell membrane, and the company has partnered the Alzheimer’s applications of both while keeping the Parkinson’s applications.

The three wholly owned clinical programmes

ACI-7104 is the anti-alpha-synuclein active immunotherapy in Phase 2 for early Parkinson’s disease, and it is the reason most people look at this company. The optimised formulation carries the fuller designation ACI-7104.056 in trial documents; the August 2026 releases use the shorter ACI-7104 for the same candidate.

ACI-19764 is a brain-penetrant small-molecule inhibitor of the NLRP3 inflammasome, in a Phase 1 programme in healthy volunteers with single and multiple ascending dose cohorts, followed by a Phase 1b cohort in patients with cardiovascular risk factors and elevated high-sensitivity C-reactive protein. The company guides to initial Phase 1 results in the near term and to Phase 1b results later in the second half of 2026.

ACI-19626 is a first-in-class positron emission tomography tracer for TDP-43, the protein that aggregates in the great majority of amyotrophic lateral sclerosis cases and in a substantial share of frontotemporal dementia. Preliminary Phase 1 results presented in 2026 showed increased uptake in the brains of patients with ALS. A tracer is not a therapy, but a validated TDP-43 imaging agent would be the first way to see that pathology in a living patient, which is a prerequisite for running efficient trials in those diseases.

An earlier-stage small-molecule programme against intracellular alpha-synuclein is described as advancing towards the clinic, with a lead declaration guided to the second half of 2026.

Why the shape of the company changed in 2025

The pipeline focus initiative announced in the third quarter of 2025 is visible in the numbers rather than described in them. Research and development expense fell by CHF 10.5 million year on year in the first half, of which the company attributes approximately CHF 4.9 million to lower personnel and operational spend as a direct result of that initiative and CHF 5.8 million to lower clinical programme spend, including active immunotherapy manufacturing costs that did not recur. General and administrative expense fell from CHF 8.3 million to CHF 7.7 million for the same reason. A company that cuts its research spending by a third while its lead asset approaches a decisive readout is making a deliberate choice about where its remaining francs go.

04 VacSYn: The Trial That Defines The Company

VacSYn is registered as NCT06015841 and titled, in full, “An Adaptive, Phase 2, Double-blind, Randomized, Placebo-controlled, Multicenter Study to Evaluate the Safety, Tolerability, Immunogenicity, and Pharmacodynamic Effects of ACI-7104.056 in Patients With Early Stages of Parkinson’s Disease”. AC Immune is the sponsor and ICON Clinical Research is the named collaborator. The study started on July 24, 2023, its listed status is active but no longer recruiting, and both primary completion and study completion are dated January 2028.

The design, as registered

The structure is a screening period of up to eight weeks, a 74-week double-blind treatment period and a 26-week post-treatment follow-up, which is where the week-100 endpoint timing comes from. Up to three cohorts of sixteen patients each are foreseen, randomised three to one between ACI-7104.056 and placebo, with the second and third cohorts optional; one cohort may be expanded to bring the total study population to as many as 150 patients, and in an expanded cohort the randomisation shifts to two to one. Treatment is given by intramuscular injection.

Eligibility is narrow, which is deliberate in a disease-modification study and consequential for how the results should be read. Patients must have clinically established early Parkinson’s disease under the modified Movement Disorder Society criteria with motor symptoms present for no more than two years at screening, be aged between 40 and 75, be at modified Hoehn and Yahr stage I or II, and be on levodopa monotherapy at 300 mg per day, stable for three months before baseline. A screening DaT-SPECT scan read centrally must be consistent with Parkinson’s disease. Known carriers of PRKN, PINK1, DJ1 and LRRK2 mutations are excluded, as are patients with a history of freezing episodes or falls, and patients who have used dopamine agonists, MAO-B inhibitors, COMT inhibitors, amantadine or anticholinergics for more than ninety days in total or within ninety days of baseline.

The endpoints

The primary endpoints are safety and immunogenicity, not efficacy: adverse events by intensity and causality from screening to week 100, abnormal MRI results, clinically significant changes on physical and neurological examination to week 74, suicidal ideation or behaviour on the Columbia scale, and the level of alpha-synuclein-specific antibodies in serum generated by the vaccine. The secondary endpoints are the ones an investor watches: alpha-synuclein-related biofluid biomarkers, dopamine transporter levels in specific brain regions measured by DaT-SPECT imaging, and change from baseline in Part III of the Movement Disorder Society Unified Parkinson’s Disease Rating Scale, all to week 100. Change in motor and non-motor function on the full MDS-UPDRS is listed as an other outcome.

That hierarchy is the single most important structural fact about this readout. A Phase 2 study whose primary endpoints are safety and antibody titres can succeed on its own terms while telling the market very little about whether patients do better, and it can equally show a promising clinical trend that is not statistically powered to mean anything. Both outcomes are possible in the same data set, and the company has been consistent in describing the clinical measures as trends.

Where it runs

Twelve sites are registered: three in Germany, at Bochum, Kassel and Kiel; six in Spain, at two hospitals in Barcelona, one in San Sebastian and three in the Madrid area; and three in the United Kingdom, two in London and one in Salford. There are no United States sites, which is precisely the constraint the IND clearance of August 11, 2026 lifts.

05 The December 2025 Interim Data, Read Precisely

The interim analysis reported on December 11, 2025 is the evidentiary base for everything that has happened since, including the Fast Track grant. It covered Part 1, which randomised 34 patients three to one. All participants in the analysis had been treated for at least twelve months, and twenty of them for up to eighteen months, which is the 74-week mark.

Immunogenicity, where the result is unambiguous

All target criteria were met. The responder rate was 100%. At week 76, two weeks after the sixth immunisation, antibody titres in serum were more than 500-fold higher than in the placebo group, and antibody responses to both the immunising peptide and the native alpha-synuclein peptide were boosted after each dose from the second immunisation onward, while the placebo group showed no detectable signal. Titres in cerebrospinal fluid rose with successive immunisations and average IgG levels in CSF were also more than 500-fold higher than placebo, which is the evidence that the antibodies cross the blood-brain barrier. Changes from baseline in CSF antibody concentration correlated with changes in serum titres, with a Spearman coefficient of 0.92 at week 24 and 0.85 at week 76, both at p below 0.05.

Biomarkers, where the result is directional

Total alpha-synuclein in cerebrospinal fluid stabilised in the treatment arm while declining in the placebo arm, with a post-hoc p value of 0.018. The company’s explanation is that antibody binding stabilises the target or increases clearance from the brain, whereas untreated progression sees alpha-synuclein accumulate in tissue and fall in CSF. Neurofilament light chain, a general marker of neuronal damage, remained stable in the treatment arm and rose in the placebo arm. Plasma glial fibrillary acidic protein and DaT-SPECT imaging were described as suggesting stabilised pathology, without figures attached.

Two qualifications belong next to those results rather than in a footnote. The alpha-synuclein CSF result is explicitly post-hoc, meaning the analysis was defined after the data were seen, which lowers its evidentiary weight regardless of the p value. And the comparison group is small: in a three-to-one randomisation of 34 patients, the placebo arm holds roughly eight or nine people, so a placebo trajectory that looks like textbook progression can also be the arithmetic of a handful of patients.

Clinical measures, where the result is a trend

At week 74 the treated group showed no meaningful progression in mean total score or change from baseline on Part III of the MDS-UPDRS, while the placebo arm increased as expected in normal disease progression. Stratifying by levodopa on and off state, the difference between the arms was described as further enhanced. The company’s own word for this is “suggestive of a trend for stabilization”, and Professor Werner Poewe, quoted in the release as an outside expert, described the consistency of the trends as promising and the findings as fully supporting further development, language that stops well short of a claim of efficacy.

Safety

No clinically relevant or serious adverse events considered related to the study drug had been reported at weeks 50 and 76. The most common adverse events were transient injection-site reactions in 56% of patients, headache in 15% and fatigue in 12%. For an active immunotherapy, that profile is the point: the historical fear with alpha-synuclein vaccines is an autoimmune reaction against the normal protein, and nothing of that kind has appeared.

Why the field is watching this particular mechanism

Passive immunotherapy against alpha-synuclein, meaning an infused monoclonal antibody, has been tried and has not worked in the clinic. The comparison most often drawn is with Biogen’s cinpanemab and with Roche’s prasinezumab, neither of which delivered a convincing Phase 2 result on clinical endpoints. Active immunisation is a different bet: it produces a polyclonal response, it is administered a few times rather than continuously, and the antibody exposure in the central nervous system is generated by the patient rather than delivered across the barrier from outside. Whether that difference translates into clinical benefit is exactly what the week-100 data set cannot fully answer and what a registrational trial would have to.

06 The Rest Of The Owned Pipeline

ACI-19764, the NLRP3 inflammasome inhibitor

The NLRP3 inflammasome is a protein complex inside immune cells that, when activated, drives the release of interleukin-1 beta and a broader inflammatory cascade. In the brain, chronic activation of that pathway in microglia is one of the mechanisms by which protein aggregation is thought to become neuronal damage. Inhibiting it is an approach with a wide potential application and a long history of disappointment, and several large companies have programmes in the area.

AC Immune’s compound is brain-penetrant, which is the differentiating claim, and it is in a Phase 1 trial in healthy volunteers with single and multiple ascending dose cohorts plus pharmacodynamic measures. Initial results are guided to the near term. A Phase 1b cohort in patients who have cardiovascular risk factors and elevated high-sensitivity C-reactive protein on entry is designed to show that the drug reduces hsCRP, a peripheral read on whether the mechanism is engaged at tolerable doses; those results are guided to later in the second half of 2026. Reading a central nervous system drug through a peripheral inflammation marker is a reasonable early step and not a substitute for a neurological endpoint.

The programme is the reason research spending rose CHF 1.9 million in the first half even as the total fell, and it is the second of the two near-term catalysts management named.

ACI-19626, the TDP-43 PET tracer

TDP-43 aggregation is present in the overwhelming majority of amyotrophic lateral sclerosis cases and in a large share of frontotemporal dementia, and it is also found alongside Alzheimer’s pathology in many older brains. Unlike amyloid and Tau, it cannot currently be imaged in a living patient, which means trials in those diseases enrol on clinical grounds and cannot confirm the target is present. A first-in-class tracer would change that.

Preliminary Phase 1 results showed increased uptake in the brains of patients with ALS. That is an early and encouraging result rather than a validated tracer; the standard for imaging agents is demanding, and the path from a signal to a diagnostic used in trials is long. The commercial logic, if it works, is a partnership with an imaging company rather than a company-built business, which is how AC Immune has handled diagnostics before.

The intracellular alpha-synuclein small molecule

An early-stage Morphomer programme targeting alpha-synuclein inside the cell is advancing towards the clinic, with a lead declaration guided to the second half of 2026. Conceptually it is the complement to ACI-7104: antibodies raised by a vaccine act on extracellular and membrane-associated species, while a small molecule can reach aggregates inside the neuron. A company that owned both would have two shots at the same biology, which is a stronger strategic position than one.

07 The Partnered Programmes And What They Are Worth

Three partners fund the Alzheimer’s side of the pipeline. Their money is the reason the company can run a Parkinson’s programme it owns outright, and the terms are worth stating with their limits attached, because contingent milestone ceilings are the most frequently misquoted numbers in biotechnology.

Takeda and ACI-24

In May 2024 AC Immune signed a worldwide option and licence agreement with Takeda covering active immunotherapies targeting amyloid beta, including ACI-24.060. AC Immune received a $100.0 million upfront payment, recorded at CHF 92.3 million, and is eligible for an option exercise fee described as being in the low-to-mid nine-figure dollar range plus additional development, commercial and sales-based milestones of up to approximately $2.1 billion, equivalent to CHF 1.7 billion, if every milestone is achieved over the life of the agreement. Royalties are tiered in mid-to-high teens percentages of worldwide net sales. Takeda may terminate the agreement at any time on ninety days’ notice, a clause that belongs in any assessment of how firm this funding is.

The clinical vehicle is the Phase 1b/2 ABATE trial in prodromal Alzheimer’s disease. In February 2026 the agreement was amended to initiate a final cohort, AD4, which evaluates ACI-24 with an additional adjuvant intended to boost immunogenicity. First patients in that cohort were dosed on April 30, 2026, triggering a $12.0 million milestone payment recognised at CHF 9.5 million at the transaction date. On June 30, 2026 the company reported twelve-month interim data from the first three cohorts, covering 74 prodromal Alzheimer’s patients: generally safe and well tolerated, no evidence of amyloid-related imaging abnormalities with oedema, and anti-amyloid antibody responses detected at every dose level. The absence of ARIA-E is the differentiating claim for an active immunotherapy against a background of approved anti-amyloid antibodies where that complication is a labelled risk.

Eli Lilly and the Tau Morphomer programme

The licence and collaboration agreement dated December 11, 2018 was amended in April 2026 to extend research into new lead Tau Morphomer candidates and potential back-up compounds. AC Immune received a CHF 10.0 million upfront payment and a further CHF 1.0 million milestone in the first half of 2026 on delivery of a preclinical data package, and a subsequent milestone is tied to Phase 1 dosing. Remaining development, regulatory and commercial milestones are disclosed as over CHF 1.7 billion with tiered royalties in the low double digits. Those two payments are the CHF 11.0 million that dominates the half-year revenue line.

Johnson & Johnson’s Janssen unit and ACI-35

The phospho-Tau active immunotherapy ACI-35.030 is partnered with Janssen. The company continues to list the collaboration among its partnered Alzheimer’s programmes in the corporate description accompanying its August 2026 releases, and no new financial terms or clinical milestones for that programme were disclosed in the half-year materials.

How to read the aggregate

The “more than $4.5 billion in potential milestone payments” that appears in every AC Immune release is a sum of maximum contingent amounts across agreements, payable only if candidates are approved and reach specified sales thresholds. It is not a receivable, not a valuation and not a forecast. The number that describes the partnerships in cash terms is what has actually been paid: a $100.0 million Takeda upfront in 2024, a $12.0 million Takeda milestone in April 2026, and CHF 11.0 million from Lilly in the first half of 2026. The unrecognised remainder of those payments sits on the balance sheet as CHF 89.5 million of short-term deferred contract revenue and CHF 0.8 million of long-term deferred contract revenue.

08 Financial Position

The company reports in Swiss francs and files with the SEC as a foreign private issuer, which means half-year rather than quarterly financial statements and a 20-F rather than a 10-K. The interim statements for the six months ended June 30, 2026 were authorised for issue by the audit and finance committee on August 3, 2026 and furnished on Form 6-K on August 5.

Revenue of CHF 16.2 million against CHF 2.3 million, research and development of CHF 22.2 million against CHF 32.7 million, general and administrative of CHF 7.7 million against CHF 8.3 million, and other operating income of CHF 0.1 million produce total operating expenses of CHF 29.8 million and an operating loss of CHF 13.6 million, against CHF 38.8 million a year earlier. The finance result swung from a CHF 1.5 million loss to a CHF 0.7 million gain, driven by CHF 2.9 million of higher foreign exchange gains partly offset by CHF 0.7 million less interest income on short-term financial assets. There is no income tax charge. The loss for the period is CHF 12.9 million, or CHF 0.13 per share on 101,963,177 weighted average shares.

The runway statement and what it assumes

The company’s own words are that cash resources “are expected to provide sufficient capital to last into Q4 2027, excluding potential milestone payments”. The going-concern note states separately that the company believes it can meet its obligations for at least twelve months from the filing date after considering CHF 27.8 million of cash and CHF 47.6 million of short-term financial assets.

The quarterly path is more informative than the half-year total. Cash resources fell CHF 16.6 million in the first quarter and were flat in the second, and the reason the second quarter held is that both large partner payments landed inside it. Stripping them out, the underlying consumption is in the region of CHF 15 to 17 million a quarter, which is consistent with a runway into late 2027 and leaves very little room for the trial expansion that the new IND makes possible. Adding United States sites to a Phase 2 study costs money that is not in the historical run rate.

The balance sheet lines that get misread

Total assets are CHF 150.4 million, of which CHF 50.4 million is an intangible asset carried unchanged from December 31, 2025, CHF 47.6 million short-term financial assets, CHF 27.8 million cash, CHF 9.9 million accounts receivable that were nil at year end, and CHF 4.4 million of prepaid expenses. Total liabilities are CHF 115.2 million, of which CHF 89.5 million is short-term deferred contract revenue and CHF 0.8 million long-term deferred contract revenue, CHF 9.1 million net employee defined benefit liabilities, CHF 7.1 million accrued expenses, CHF 4.1 million of lease liabilities across current and non-current, and CHF 3.1 million of deferred income. Shareholders’ equity is CHF 35.2 million against CHF 44.9 million at year end, with share premium of CHF 483.6 million and accumulated losses of CHF 450.4 million.

A reader who sees CHF 115.2 million of liabilities against CHF 150.4 million of assets and concludes the company is heavily indebted has misread the largest line. AC Immune has no bank debt, no convertible notes and no term loan. The CHF 89.5 million of short-term deferred contract revenue is partner money already banked and not yet recognised in the income statement, and it converts into revenue as work is performed rather than into a cash outflow. The defined benefit pension liability of CHF 9.1 million is a Swiss employer obligation, real but long-dated. The genuine claims on cash in the next twelve months are the CHF 1.5 million of trade payables, CHF 7.1 million of accrued expenses and CHF 0.8 million of current lease liabilities.

09 Capital Structure And Dilution

The company-declared share count is 101,742,231 shares outstanding at December 31, 2025, stated on the cover of the Form 20-F filed on March 13, 2026. The equivalent figures were 100,410,377 at December 31, 2024 and 99,197,829 at December 31, 2023, so the count has grown by roughly 1.3% a year over two years. For a clinical-stage company with no product revenue, that is a remarkably quiet share register, and it is a direct consequence of the partnering strategy: the Takeda upfront of $100.0 million in May 2024 is money that did not have to be raised from shareholders.

The half-year statements do not restate a point-in-time outstanding count at June 30, 2026. What they do give is the weighted average used for the loss per share, 101,963,177 shares for the six months against 100,519,884 a year earlier, and the treasury position, which fell from 10,899,773 shares at December 31, 2025 to 10,673,993 at June 30, 2026. Share capital at par rose from CHF 2,253 thousand to CHF 2,263 thousand and share premium from CHF 481.9 million to CHF 483.6 million. Proceeds from the sale of treasury shares in public offerings, net of fees, were CHF 0.6 million in the period. Treasury shares held against future issuance are a normal feature of Swiss capital structures and function as authorised capital ready for use; the movement in the period is small.

An at-the-market programme with Jefferies LLC exists and is described in note 12 of the 2025 Form 20-F. The half-year statements refer to it only by cross-reference and disclose no at-the-market activity for the first half of 2026. That is a fact worth holding lightly: the absence of disclosed usage in a six-month interim statement is not the same as a commitment not to use it, and a company approaching a binary readout with a runway that ends in the fourth quarter of 2027 has an obvious reason to keep the facility available.

What dilution would look like from here

The arithmetic is uncomfortable at the current market value. With roughly 101.7 million shares outstanding and a market capitalisation near $252 million on August 11, 2026, raising the equivalent of one year of operating spend, call it CHF 60 million or about $74 million, at a price close to the market would issue roughly thirty million shares and expand the count by nearly thirty per cent. That is the mechanical reason a positive week-100 readout matters beyond the science: it changes the price at which any future financing happens, and therefore how much of the company existing holders keep. The reverse is equally mechanical.

10 Governance: A Company Between Chief Executives

Andrea Pfeifer co-founded AC Immune in 2003 and led it as chief executive for more than twenty years, including through the 2016 Nasdaq listing. On May 12, 2026 the company announced her retirement at the annual general meeting, and she signed her last filing as chief executive on December 11, 2025 alongside chief financial officer Christopher Roberts. At the annual general meeting of June 11, 2026 all agenda items were approved and the board appointed its chair, Dr Martin Zügel, to serve as interim chief executive while a search for a permanent successor continues. The 6-K of August 11, 2026 carrying the Fast Track announcement is signed by Zügel as interim chief executive and by Roberts as chief financial officer.

The situation is worth weighing without dramatising it. A founder-led company that loses its founder after two decades faces a genuine transition, and the chair stepping into the executive seat is a common bridging arrangement that also concentrates board and management roles in one person for its duration. What makes the timing consequential here is what falls inside it: the full VacSYn Part 1 readout, the regulatory conversations about a registrational design that follow it, and any decision about partnering ACI-7104 or financing the company through 2027. Those are the decisions a permanent chief executive would normally own.

The counter-argument is equally factual. The strategy that produced the current position, meaning the pipeline focus initiative of the third quarter of 2025 and the partnering of the Alzheimer’s assets, was set before the transition and is visibly being executed: research spending down a third, two partner amendments signed in April 2026, a regulatory package delivered in August. Continuity of execution is visible in the filings; continuity of leadership is not yet settled.

11 Market Data And How This Stock Trades

$ACIU closed the reading of August 11, 2026 at $2.48 on Nasdaq, up 8.7% on the day the Fast Track designation was announced, on volume of about 289,000 shares, with a market capitalisation near $252 million. Finviz classifies the company under healthcare and biotechnology with Switzerland as the country of domicile. There is no price-to-earnings ratio because there are no earnings.

Three characteristics define the trading profile and none of them is a judgement about the business. The first is size: a market capitalisation around a quarter of a billion dollars places this among the smaller Nasdaq-listed biotechnology names, which is the range where a single institutional decision moves the price. The second is that the company is Swiss and reports in Swiss francs on a half-year cycle, so there are only two financial reporting events a year rather than four, and the dollar figures a United States investor sees embed a currency movement that has nothing to do with the pipeline. The third is that the entire value of the equity rests on unpartnered assets whose next data point is a window rather than a date, which is the classic setup for a stock that drifts and then gaps.

Two figures from the balance sheet are useful as reference points against the market value, and neither is a valuation. Cash resources of CHF 75.4 million are about $93.0 million at the August 11 rate, which is roughly 37% of the market capitalisation. The intangible asset carried at CHF 50.4 million, about $62 million, is a further balance-sheet item, though intangibles in development-stage biotechnology carry accounting values that need not correspond to what anyone would pay.

12 Retail Sentiment

The block below is a snapshot of the Stocktwits stream with its timestamp. These are the opinions of retail traders and non-professional investors, not of analysts or institutions, and they measure attention and imbalance rather than research quality.

The recurring themes on the stream over the preceding weeks were the Fast Track and IND headline itself, the presentations at the Alzheimer’s Association International Conference in London in July 2026, and links to third-party commentary describing the second half of 2026 readout as decisive. None of that constitutes evidence about the data set. What the volume spike does establish is that a small, previously quiet stream became busy on a single regulatory headline, which is the pattern that precedes sharp moves in both directions around a readout.

13 The Case Made By Those Who Are Positive

What follows is the argument as its holders make it, not a Merlintrader view.

The mechanism has cleared the bar that killed the competition. Passive antibodies against alpha-synuclein have failed to show clinical benefit. Active immunisation produces a polyclonal response, and the interim data show that response reaching the cerebrospinal fluid at more than 500 times placebo levels with a statistically significant correlation between compartments. Whatever else is uncertain, target engagement in the central nervous system is demonstrated, which is more than several better-funded programmes achieved.

The safety profile is what an active immunotherapy needs. The theoretical objection to vaccinating against a normal human protein is autoimmunity. Across weeks 50 and 76 there were no clinically relevant or serious adverse events considered drug related, and the common events were injection-site reactions, headache and fatigue. That profile supports the long dosing intervals and the multi-year follow-up the Goradia grant now funds.

Three biomarkers moved in the same direction. CSF alpha-synuclein stabilised, neurofilament light stabilised where placebo rose, and plasma GFAP and DaT-SPECT both pointed the same way. Consistency across independent measures is harder to explain by chance than a single positive marker, which is the point Professor Poewe made when he called the consistency of the trends remarkable.

The FDA has now engaged twice. Fast Track designation and IND clearance on the same day, four weeks or so before the guided readout window, is consistent with a regulator that has seen the interim package and is prepared to work with it. It also gives the company the frequent-interaction rights that make agreeing a registrational design faster.

Someone else pays for the Alzheimer’s half. Takeda funds ACI-24, Lilly funds the Tau small molecules, Janssen holds ACI-35. That structure produced CHF 16.2 million of first-half revenue, a $100.0 million upfront in 2024 and a $12.0 million milestone in April 2026, and it is why the share count has grown about 1.3% a year rather than the double digits typical of a company at this stage.

The market value is close to the tangible resources plus a modest premium. Cash resources of about $93.0 million sit against a market capitalisation near $252 million, so the entire owned pipeline, the two platforms and the partnership stream are being carried at roughly $160 million by the market on August 11, 2026. Holders of this view argue that a Parkinson’s disease-modification asset with demonstrated CNS target engagement is worth more than that if the week-100 data hold.

14 The Case Made By Those Who Are Sceptical, And The Red Flags

Thirty-four patients. Part 1 randomised 34 people three to one, so the placebo arm is roughly eight or nine patients. Every biomarker and clinical comparison rests on that comparator. Trends that look clean at this size have repeatedly failed to replicate in Parkinson’s disease, where symptomatic variability is large and progression over eighteen months is modest even in untreated patients.

The efficacy analysis that matters most is post-hoc. The stabilisation of total CSF alpha-synuclein carries a p value of 0.018 from a post-hoc analysis. An analysis specified after seeing the data is a hypothesis, not a result, and regulators treat it accordingly.

The primary endpoints are not efficacy endpoints. VacSYn’s primary endpoints are safety and antibody titres. The clinical measure that would matter to a patient, MDS-UPDRS Part III, is secondary and the study was not designed to power it. A readout can therefore be reported as successful while leaving the central commercial question open, and the market may not price that distinction consistently.

Guidance has already moved. On December 11, 2025 final Part 1 data were expected in mid-2026. In the August 5, 2026 release the same data set is guided to the second half of 2026. A slip of that size in a trial that stopped recruiting some time ago is not alarming on its own, but it is a fact, and the two August documents describe the endpoint as week 104 and week 100 respectively, which is an inconsistency in the company’s own disclosure.

The company is being run on an interim basis. The founder chief executive retired at the June 11, 2026 annual general meeting and the chair holds the executive seat while a search continues. The registrational strategy, any partnering of ACI-7104 and any financing decision fall inside that interval.

The runway ends before a registrational trial could finish. Cash resources fund operations into the fourth quarter of 2027 on the company’s own statement, and that statement excludes milestone payments. A registrational Parkinson’s trial in early disease would take years and cost multiples of the current balance sheet. Whatever happens with the readout, this company cannot self-fund the next stage, so the outcome is a partnership, a large equity raise, or both.

Partner money can stop. Takeda may terminate the option and licence agreement at any time on ninety days’ notice. The revenue line that carried the first half of 2026 is composed of two non-recurring payments from two counterparties, and there is no product revenue underneath it.

Currency and reporting cadence work against a United States reader. The accounts are in Swiss francs and published twice a year. Between February and August, an investor has no interim financial statement to check the burn against, and every dollar figure carries an exchange-rate movement unrelated to the pipeline.

15 Scenario Framework

The two paths below are descriptions of what would have to happen, not forecasts, and they carry no probabilities and no price levels.

The variable that separates the two is not the Fast Track designation and not the IND. It is whether a biomarker separation observed against a placebo arm of roughly eight patients survives contact with the complete Part 1 data set. Everything else in the model is downstream of that. The checkpoints the company has actually published are: full Part 1 VacSYn results in the second half of 2026, initial NLRP3 Phase 1 results in the near term, NLRP3 Phase 1b results later in the second half of 2026, and a lead declaration on the Morphomer alpha-synuclein aggregation inhibitor in the second half of 2026. The next scheduled financial disclosure is the full-year 2026 report, which in 2026 was furnished on March 13.

16 Bottom Line

Two accurate and opposite descriptions of AC Immune can be held at the same time. The first is a company that has demonstrated, for the first time in Parkinson’s disease, that an active immunotherapy raises antibodies which reach the central nervous system at more than 500 times placebo levels with a 100% responder rate, that three independent biomarkers moved consistently in the direction of stabilisation, that the FDA has now granted Fast Track and cleared an IND, and that three large pharmaceutical partners fund the rest of the pipeline while the share count grows about 1.3% a year. The second is a company with no product, no revenue that is not a partner payment, a placebo arm of roughly eight patients underneath its central efficacy claim, a post-hoc p value on its most quoted biomarker result, an interim chief executive, a readout window that has already moved once, and cash that runs out in the fourth quarter of 2027. Both descriptions come from the same filings.

What is verified. Fast Track designation and IND clearance for ACI-7104 in early Parkinson’s disease, announced August 11, 2026 and furnished on Form 6-K the same day. Interim Part 1 results at week 76 in 34 randomised patients: 100% immunogenicity responder rate, serum and CSF antibody titres more than 500-fold above placebo, Spearman correlation between compartments of 0.92 at week 24 and 0.85 at week 76, CSF alpha-synuclein stabilisation with a post-hoc p of 0.018, stable neurofilament light against a rising placebo arm, no clinically relevant drug-related serious adverse events, injection-site reactions in 56% of patients. Cash resources of CHF 75.4 million at June 30, 2026, about $93.0 million, composed of CHF 27.8 million cash and CHF 47.6 million short-term financial assets, funding operations into Q4 2027 on the company’s own statement and excluding milestone payments. First-half revenue of CHF 16.2 million, all from Lilly and Takeda, against CHF 2.3 million a year earlier. Research and development of CHF 22.2 million, down from CHF 32.7 million. Net loss of CHF 12.9 million, CHF 0.13 per share. 101,742,231 shares outstanding at December 31, 2025 per the Form 20-F, and 10,673,993 treasury shares at June 30, 2026.

What is not verified. Whether the week-100 data set reproduces the week-76 separation. What registrational design, if any, the FDA has indicated it would accept. Whether the endpoint of the final Part 1 analysis is week 100 or week 104, since the company used both descriptions in documents six days apart. The point-in-time share count at June 30, 2026, which the interim statements do not restate. Any use of the at-the-market programme in 2026, which is referred to only by cross-reference to the 20-F. Who will be the permanent chief executive, and when. And the analyst consensus, since no individual note could be confirmed with both house and exact date.

The dates that settle the question: the second half of 2026, for full Part 1 results from the Phase 2 VacSYn trial, the single readout on which the equity turns; the near term and then later in the second half of 2026, for initial Phase 1 and then Phase 1b results on the NLRP3 inhibitor ACI-19764; the second half of 2026, for the lead declaration on the Morphomer alpha-synuclein aggregation inhibitor; and the fourth quarter of 2027, the end point of the company’s own runway statement, which is when the financing question stops being theoretical.