EyePoint EYPT layoffs news

Opinion · Devil's advocate

Fact-check-o-meter

EyePoint's releases state the primary endpoint misses but leave the vision gap and safety rates to the slides

Both wet AMD phase 3 trials missed their primary vision endpoint, and the releases state the misses but pair them with secondary measures, an ad hoc analysis and nominal p-values. The actual vision gap, the side effect imbalances and the small size of the phase 2 trial behind the DME bet are in the filed slide deck and the trial registry, not the release text. The legal record is documented: a DOJ settlement, paid in July, with a five-year integrity agreement, an FDA warning letter from July 2024, and a suit EyePoint filed against Ocular Therapeutix.

1. Press release vs filing · weight high

   LUCIA release states the miss but not the letter gap

   The company said: "In LUCIA, despite demonstrating visual improvement, DURAVYU did not achieve the primary endpoint of statistical non-inferiority in mean change in best corrected visual acuity (BCVA) versus on-label 2 mg aflibercept control." source

   The record says: The slide deck filed with the same 8-K shows the primary endpoint: "Aflibercept 2 mg q8W +6.1 letters DURAVYU 2.7 mg q24W +2.5 letters Difference -3.7 letters", on a slide marked as preliminary data pending final analysis. source

   Our view: The slide shows a 3.7 letter difference against aflibercept and says the confidence interval is outside the 4.5 letter non-inferiority margin. The release says non-inferiority was not achieved but does not give the difference or the margin, so readers of the release alone do not see how large the gap was.

2. Science · weight high

   LUGANO headline rested on an ad hoc analysis

   The company said: "DURAVYU was non-inferior to on-label aflibercept control (nominal p-value = 0.0096) in an ad hoc analysis excluding a 4% asymmetric cohort (9 of 211 patients)" source

   The record says: The registry entry for the identical LUCIA design lists the primary outcome (the LUGANO entry, NCT06668064, lists the same) as "Average change in best corrected visual acuity (BCVA)" at "Weeks 52 and 56", with no exclusion rule described there. Two months later the LUCIA release said: "LUGANO also did not meet the primary endpoint in its full dataset". source

   Our view: Excluding nine patients in an analysis the release itself calls ad hoc is a post hoc result, and the August release led with it, though the same headline says the primary endpoint was not achieved in the full dataset. LUCIA then missed, and its release and deck do not offer a similar excluded cohort, which weakens the case that the LUGANO miss was a one-off.

3. Science · weight high

   DME bet leans on a 27-patient phase 2

   The company said: "The DME program is supported by the positive Phase 2 VERONA data showing sustained visual acuity and over 70% of patients supplement-free for six months." source

   The record says: ClinicalTrials.gov lists VERONA (NCT06099184) with 27 patients enrolled across three arms, masking "Single (Participant)", and a primary outcome of "Time to first supplemental aflibercept injection" at Week 24. The phase 3 COMO trial's primary outcome is BCVA change "to the average of Weeks 52 and 56". The filed deck's VERONA slide adds: "Data excludes one outlier patient." source

   Our view: VERONA tested a different endpoint over less than half the follow-up, in a trial of about nine patients per arm. The DME phase 3 trials use the same BCVA non-inferiority test against aflibercept that both wet AMD trials failed.

4. Press release vs filing · weight medium

   "Favorable safety" next to an AE imbalance

   The company said: "Continued favorable safety profile for DURAVYU with redosing" source

   The record says: The filed deck's pooled table of ocular adverse events through Week 56 shows, for DURAVYU (n=443) versus aflibercept (n=464): vitreous floaters 8.6% vs 2.8%, neovascular age-related macular degeneration 5.9% vs 1.7%, retinal haemorrhage 4.7% vs 1.5%, vitreous haemorrhage 2.3% vs 0.0%. source

   Our view: The deck reports no cases of insert migration or retinal vasculitis, and severe intraocular inflammation and endophthalmitis under 1% in both arms, but it also shows more floaters, bleeding and neovascular AMD events on DURAVYU. A doctor choosing between this and aflibercept would want those rates, and the release summarizes them as favorable.

5. Press release vs filing · weight medium

   Runway not restated after the cut

   The company said: "EyePoint estimates that it had approximately $118 million in cash, cash equivalents, and short-term investments as of September 30, 2026." source

   The record says: The 10-Q said "cash, cash equivalents, and investments in marketable securities of $180.5 million" at June 30, 2026 would "fund our operations into the fourth quarter of 2027" and added "there is no assurance that additional funding will be achieved". The October release and 8-K give no updated runway; DME topline is guided to the fourth quarter of 2027. source

   Our view: The September 30 estimate is about $62 million below the June 30 figure, the company expects annualized savings of about $30 million, and it plans about $9 million in severance and related charges. The 10-Q runway ended in the same quarter as the DME readout, and the October release and 8-K do not restate it.

6. Legal and disclosure record · weight low

   DOJ settlement, integrity agreement, warning letter

   The record says: The 10-Q states that on July 17, 2026 the company settled a DOJ False Claims Act and Anti-Kickback Statute investigation tied to DEXYCU for $4,678,981.86 plus interest, entered a five-year Corporate Integrity Agreement with HHS on July 13, 2026, and paid in full on July 31, 2026. It also discloses the FDA warning letter received in July 2024 on YUTIQ manufacturing at the Watertown facility, and a defamation suit EyePoint filed against Ocular Therapeutix in Middlesex County Superior Court on March 20, 2026. No securities class action or derivative suit is disclosed in the latest filing. source

   Our view: These matters are disclosed and the DOJ one is paid, but the integrity agreement adds compliance cost for five years. The 10-Q says the company has fully executed its corrective and preventive action plan for the warning letter.

This is opinion. We compare the company's own announcements with its filings and the scientific record, and link every source. It is not an allegation of wrongdoing and not investment or legal advice. Legal items list only matters disclosed in filings or by courts and regulators. Checked Oct 23, 2026. Companies can respond through our right of reply.