Outlook Therapeutics ($OTLK): The Approval Arrived — Can the Launch Be Funded?

01 Bull, base and bear scenarios

These are reading frames, not forecasts, and none of them is a recommendation. Each is written so that a reader can say which document would confirm or break it.

What distinguishes Outlook from most companies at this stage is the order in which its risks resolved. Three complete response letters were overturned on appeal and the product is approved and priced, so the clinical and regulatory questions are closed in public. What remains is almost entirely financial and organisational: whether a company with $11.2 million of quarter-end cash, a stated going-concern doubt, a listing deficiency with the stock about 41 per cent below the $1.00 minimum and a newly appointed chief commercial officer can launch an injection that is new as a formulation but, in FDA’s words, “not the first in its class” into a market dominated by a cheaper compounded version of the same molecule — and fund it without resetting its own share count for the fourth time in three years.

What Would Falsify This Reading

The reading on this page is that the regulatory question is closed and unconditionally so, that the open risks are funding, the listing and commercial execution, and that Europe is the best available evidence for what a launch looks like here. Each of the following would damage that reading, and each is checkable.

• The approval is a full approval, not a conditional one, and the letter is short. BLA 761320 was approved on July 24, 2026 for LYTENAVA (bevacizumab-vikg) injection, 1.25 mg (25 mg/mL) single-dose vial, for “treatment of patients with neovascular (wet) age-related macular degeneration” — five days ahead of the July 29 goal date. No REMS. No postmarketing requirement. One postmarketing commitment, PMC 5036-1, verbatim: “Implement a two-tiered reference standard system by qualifying a working reference standard,” with the report due in a prior approval supplement by July 2027. No advisory committee was convened at any point in the programme’s history; the approval letter gives the reason: “this biologic is not the first in its class.” The label’s recommended dose is 1.25 mg once monthly, about every 28 days. There is no second trial obligation and no application of any kind pending.
• But the evidence package behind it is thinner than a first-cycle approval would be, and the regulator said so three times. The confirmatory trial missed. In FDA’s numbers, printed in the label, NORSE EIGHT’s week-8 mean gain was +3.3 against +4.5 letters and the LS mean difference −2.3, 95% CI (−4.0, −0.5), against a pre-specified non-inferiority margin of −3.5. In the company’s intent-to-treat figures the difference was −2.257 letters, 95% CI (−4.044, −0.470), means +4.2 against +6.3, and “the hypothesis of noninferiority was not met (p>0.025).” At week 12 the difference was −1.0 (−2.9, 0.8) in FDA’s appeal letter and −1.009 (−2.865, 0.848) in the company’s figures, which does clear the same margin. The first complete response letter found that NORSE ONE, the other trial intended as pivotal, had failed; the second and third each carried one deficiency only — lack of substantial evidence of effectiveness — the third concluding that “adequate confirmatory evidence has not been provided.” The company’s release added that the agency “has not indicated what type of confirmatory evidence would be acceptable”; at the March 2, 2026 meeting, per FDA’s appeal letter, the Division said “an additional adequate and well-controlled study demonstrating efficacy would best provide adequate confirmatory evidence.” The approval rests on NORSE TWO plus confirmatory evidence: 41.7 per cent against 23.1 per cent, p=0.0052, mean gain 11.2 against 5.8 letters, p=0.0035, against a ranibizumab regimen FDA itself calls less effective. Registry results are now posted for NORSE TWO, THREE and EIGHT; NORSE ONE and SEVEN have none.
• The commercial read is not a forecast, it is two years of European actuals, and they are poor. Germany and the United Kingdom launched in June 2025, Austria in January 2026. Fiscal 2025 net revenue was $1,413,535 for a gross profit of $57,205. The first nine months of fiscal 2026 were negative $1,071,694, the company’s explanation being a $1.1 million increase in the returns reserve on estimated returns from the United Kingdom distributor “due to lower-than-forecasted sell-through.” Over the nine months the reserve in fact fell $1.1 million net: a $1.2 million increase for United Kingdom and German sales against a $2.4 million reversal for product that became unsellable under minimum shelf-life limits, written off against receivables. The June quarter booked $8,700. Revenue is not disaggregated by country anywhere in the record. The United States equivalent does not exist yet, and one piece of its plumbing is still open: the company announced on October 1, 2026 that it had applied to CMS for a permanent HCPCS J-code — an application, not a granted code; CMS reviews such applications quarterly, and on the August 14 earnings call management said it expects a code in April 2027.
• The funding constraint is stated by management, not inferred by this page. Verbatim: cash and equivalents at June 30, 2026 together with $51.1 million of August proceeds “will only be sufficient to fund our operations into the second quarter of calendar year 2027,” and “there is substantial doubt about our ability to continue as a going concern.” Cash was $11,242,068; accumulated deficit $653,540,250; stockholders’ equity a deficit of $10,392,567. Atlas Sciences has been able to redeem up to $3.0 million per calendar quarter since September 16, 2026, every cash payment carrying a 7.5 per cent exit fee. Both at-the-market programmes are shut, and the market agrees: at the October 2, 2026 close of $0.593 the stock sits about 40.7 per cent below the $1.00 minimum and the whole equity is worth about $144.4 million, roughly 2.3 times cash plus August proceeds — both calculations — against a compliance date of March 24, 2027 and an unexercised 1-for-10 to 1-for-50 reverse-split authority that goes back to stockholders if not used before July 16, 2027.
• Or the capital position is more resilient than the bear case assumes and the dilution is the price of survival. This is the strongest argument against the rest. The company has never disclosed an event of default — the missed Avondale payment of December 31, 2025 was, on its account, a trigger event and not a default — has cleared its two older notes entirely — Streeterville repaid $32,910,027.57 in March 2025, Avondale fully extinguished in May 2026 — and has received no stockholders’-equity or market-value deficiency notice in the Capital Market era despite the equity deficit. Its largest shareholder has funded four of the last six equity financings, holds board designation rights and two designated directors, and bought again at $0.99 in August alongside the chief executive and the finance chief. Four directors bought on the open market at $0.4093 to $0.4413 in late May 2026, and no insider has sold a share in eighteen months.

None of these is a prediction. They are the observations that would make the rest of this page wrong, listed so that a reader can check them rather than take the reading on trust.

02 Executive answer: what the United States approval changed

What this page is for: Reading a company that won the approval and still has to finance the launch

Outlook sits in a position that reads backwards from most of biotech. The scientific and regulatory risk has been resolved in the company’s favour — three refusals, a formal dispute, an appeal granted by FDA letter of May 20, 2026, announced on May 26, and an approval on July 24, 2026 — while the financial risk is exactly as acute as it was before, and is stated by management rather than inferred. The page sets out the refusal-and-appeal chronology and what each refusal said, the pivotal trial that worked and the confirmatory trial that missed, the $485 price, the J-code application and the launch window, the European authorisations with the modest revenue they have produced, the Atlas note and its quarterly redemption right, the going-concern language verbatim, the Nasdaq deficiency and the unused reverse-split authority, and the litigation that the refusal history predicted.

Five facts set the frame, and the first one is the correction most readers arrive needing.

LYTENAVA is not a candidate awaiting a decision. It is an approved product. The FDA approved BLA 761320 on July 24, 2026 — LYTENAVA (bevacizumab-vikg) injection, 1.25 mg in a 25 mg/mL single-dose vial, for the “treatment of patients with neovascular (wet) age-related macular degeneration (nAMD)” — five days ahead of the July 29, 2026 goal date. No advisory committee was ever convened for the product, at any point in its eight-year regulatory life — the approval letter says it was not referred to one “because this biologic is not the first in its class” — and the letter imposes no risk evaluation and mitigation strategy. The label’s recommended dose is 1.25 mg by intravitreal injection once monthly (approximately 28 days); FDA’s appeal letter notes that a less frequent regimen after three monthly doses has not been studied. It carries one postmarketing commitment, PMC 5036-1, verbatim: “Implement a two-tiered reference standard system by qualifying a working reference standard,” with the final report due July 2027. The company anticipates twelve years of reference product exclusivity under the BPCIA.

What the approval changed is the category, not just the ticker. LYTENAVA is, on the company’s description, the first and only ophthalmic formulation of bevacizumab approved in the United States, the European Union and the United Kingdom — in the United States since July 24, 2026, in the European Union since May 2024 and in the United Kingdom since July 2024. The novelty is the formulation and the label, not the molecule or the mechanism: bevacizumab is a long-established anti-VEGF antibody, and FDA’s own letter calls LYTENAVA “not the first in its class.” The real-world incumbent is not a branded anti-VEGF drug but Roche’s Avastin, repackaged off-label by compounding pharmacies: the company’s own cited figures put off-label repackaged bevacizumab at about 66.3 per cent of new United States wet-AMD patient starts and about 34 per cent of the total anti-VEGF market; separately, the annual report counts roughly 2.7 million off-label repackaged bevacizumab injections a year in the United States across all uses.

And a price now exists. On September 16, 2026 the company set the United States wholesale acquisition cost at $485 per vial, which sits where it said it would: “a lower-cost alternative to biosimilars and branded products, while remaining above off-label compounds.” Launch is guided before December 31, 2026, and management has put a number on the ambition — the potential, on its own estimates, to “exceed $500 million in annual U.S. sales by 2030.” On the August 14, 2026 earnings call management added a nearer figure, as company guidance: “We expect total net revenue during the first 12 months following the U.S. launch of LYTENAVA to be between $50 million and $75 million, with Europe contributing approximately 10%-15% of that total.”

Nothing is pending in front of the agency. No new United States application, no supplement and no goal date is disclosed. The one future filing in the record is a supplemental BLA for a pre-filled syringe, which NORSE SEVEN would support — and that trial has posted no registry results — FDA’s appeal letter says its 90-day visual-acuity data, like those of the open-label NORSE THREE, showed “minimal to no improvement in BCVA” — and carries an estimated primary completion of December 2027. There is no open regulatory question here. The open questions are launch execution, payer coverage and cash.

None of which the approval touched. Cash was $11.2 million at June 30, 2026, two months before the approval was monetised; the August 2026 offering added $51.1 million net; and the company’s own filing says that sum will be sufficient only “into the second quarter of calendar year 2027.” Substantial doubt about the ability to continue as a going concern is stated in the filings — in the financial-statement notes, in management’s discussion and in the liquidity section of the same 10-Q. The share count has gone from 33,568,023 on May 12, 2025 to 243,447,554 on August 14, 2026. And on September 25, 2026, two months after the approval, Nasdaq sent a fresh minimum-bid-price deficiency notice. The United States revenue line, meanwhile, is still at zero: group revenue for the nine months to June 30, 2026 was negative $(1,071,694), and $8,700 in the June quarter alone.

03 One trial that worked, one that missed, and three letters in between

The clinical record is unusually simple to state and was unusually hard to get past the agency. Five trials are registered. One of them — NORSE TWO — is the entire efficacy case, and it met its primary endpoint in August 2021. Everything that happened over the following five years was an argument about whether one positive trial was enough.

Two housekeeping facts belong under that table. Three of the five studies now have results posted on ClinicalTrials.gov — NORSE EIGHT on June 30, 2026, NORSE THREE on July 17 and NORSE TWO on August 3 — while NORSE ONE and NORSE SEVEN have none; the efficacy figures above come from the company’s filings and, where stated, from FDA’s label and appeal letter, and the NORSE EIGHT safety figures from the registry. And NORSE FOUR, FIVE and SIX do not exist as trials: the company holds three special protocol assessments covering a branch retinal vein occlusion registration trial and two diabetic macular oedema registration trials, but, verbatim, “the timing for initiating these studies has not been determined pending initial FDA approval for wet AMD.” No registry numbers, no enrolment, no dates. The approval that was the stated precondition arrived in July 2026; no initiation has been announced since.

The appeal, which is how this ended

The third letter was not answered with more data. It was answered with process. The company requested a Type A meeting on February 11, 2026, held it with the Division of Ophthalmology and the Office of Specialty Medicine on March 2, 2026, and then escalated. Per FDA’s appeal letter, the Formal Dispute Resolution Request was dated and received on March 23, 2026 — the company announced on April 7 that it had been accepted — the meeting with the deciding official took place on April 20, 2026, announced the next day, and the decision letter, dated May 20, 2026 and signed by Mary Thanh Hai, MD, Director of the Office of New Drugs, granted the appeal; the company announced it on May 26. FDA’s conclusion, verbatim: “Based on evidence from NORSE 8, the natural history of untreated nAMD, and mechanistic and pharmacodynamic evidence for bevacizumab-vikg, I believe there is strong confirmatory evidence from multiple sources to substantiate the positive results from the one adequate and well-controlled clinical investigation, NORSE 2, to conclude that SEE has been established.” The sentence usually quoted as the decision — “OND determined that the results of the NORSE TWO trial, together with confirmatory evidence including NORSE EIGHT, natural history, and mechanistic and pharmacodynamic data, establish substantial evidence of effectiveness… The response directs the Division and OSM to work with Outlook Therapeutics to reach an agreement on final labeling” — is the company’s summary, from its May 26 release.

What that paragraph means is worth stating plainly, because it is unusual: the failed trial became part of the evidence that won the approval. NORSE EIGHT missed its week-8 non-inferiority endpoint, and the deciding official counted it as confirmatory evidence anyway. The resubmission, dated May 29, 2026 in the approval letter and announced on June 1, was acknowledged on June 16 as a Class 1, sixty-day review with a July 29 goal date, and was approved on July 24. The original BLA was submitted in March 2022, withdrawn in May 2022 “to provide additional information requested by the FDA,” resubmitted that August, and accepted for filing in October 2022. Both subsequent review cycles — the Class 2 cycle with an August 27, 2025 goal date and the Class 1 cycle with a December 31, 2025 goal date — ended in a complete response letter on or about the goal date itself.

The appeal letter also drew two lines that carry into the label. Because NORSE TWO compared LYTENAVA with ranibizumab given monthly for three months and then every three months — “a less effective dosing regimen of ranibizumab” — labeling “should not have any claim, direct or implied, of superiority (or non-inferiority) to ranibizumab.” And the company “has not studied a less frequent dosing regimen of bevacizumab-vikg after a three monthly-dosing regimen”: the approved label recommends 1.25 mg once monthly, approximately every 28 days.

One consequence of that history is not regulatory but legal, and it is live. A securities class action in the District of New Jersey — Alsaidi v. Outlook Therapeutics, Inc., No. 23-cv-21862 on the court’s docket, though the company’s filings do not name it — filed November 3, 2023, runs a class period of August 3, 2021 through August 29, 2023 — ending precisely on the date of the first complete response letter — and alleges a failure to disclose “an alleged lack of evidence supporting LYTENAVA as a treatment for wet AMD” and “deficient CMC controls.” The court dismissed the second amended complaint in part with prejudice and in part with leave to amend on December 23, 2025; a third amended complaint followed, and no ruling on it is disclosed. A Delaware derivative suit is stayed behind it, and no reserve has been established for either.

04 The label, the $485 vial and a launch that has to happen this quarter

NICE TA1022 lists a UK price of £470 per 7.5 mg/0.3 mL vial of bevacizumab gamma, excluding VAT, based on the company submission accessed in September 2024. NICE also records a confidential NHS discount. That historical UK list price is not the NHS net price, a U.S. price or revenue Outlook receives per treatment. NICE technology appraisal guidance TA1022 (bevacizumab gamma).

The licence is narrow, and narrow here works in the company’s favour: the indication it covers is exactly the market the unapproved version of the same molecule already owns.

The price, and the only benchmark that matters

On September 16, 2026 the company set the United States wholesale acquisition cost at $485 per vial. That is the whole of the disclosed pricing architecture: no net price, no gross-to-net assumption, no discount or rebate structure and no payer coverage decision of any kind appears anywhere in the United States record, and reimbursement coding is at the application stage: the company applied to CMS for a permanent J-code on October 1, 2026, and on the August 14 earnings call management said it expects one in April 2027. What does appear is the stated intent, which the $485 figure is consistent with: “We intend to strategically price ONS-5010/LYTENAVA as a lower-cost alternative to biosimilars and branded products, while remaining above off-label compounds.”

The competitive map has two layers. The branded and biosimilar layer is crowded — EYLEA and EYLEA HD, LUCENTIS, BEOVU, VABYSMO, SUSVIMO, plus the biosimilars BYOOVIZ, CIMERLI and PAVBLU. The volume layer is one product used off-label: Roche’s Avastin, repackaged by compounding pharmacies, which on the company’s cited figures takes about 66.3 per cent of new United States wet-AMD patient starts and about 34 per cent of the total anti-VEGF market — the annual report separately counts roughly 2.7 million off-label repackaged bevacizumab injections a year in the United States across all uses — inside a market the company sizes at about $8.5 billion.

A further long-acting candidate is now approaching the branded layer. On September 28, 2026, Kodiak Sciences reported that its anti-VEGF candidate Zenkuda (tarcocimab tedromer) met the primary endpoint of the Phase 3 DAYBREAK trial in wet AMD, with vision gains non-inferior to aflibercept at year one (p=0.0007) and 54 per cent of patients on a six-month dosing interval under strict treat-to-dryness retreatment criteria; a second candidate, tabirafusp-ted, also met its vision endpoint. Kodiak plans a multi-indication biologics licence application for Zenkuda in the fourth quarter of 2026. As of that release no application had been filed and nothing is approved, and the figures are the company’s topline release, not a peer-reviewed publication; the release does not mention Outlook or bevacizumab. In Merlintrader’s reading, the relevance to LYTENAVA, labelled for monthly dosing, is indirect: a further long-acting branded option would widen the gap in dosing frequency, so Outlook’s case rests even more on its price position against branded products and on its quality case against compounded bevacizumab than on convenience. Kodiak 8-K, Exhibit 99.1

The pitch against that incumbent is a quality argument, not an efficacy one: repackaged bevacizumab “is not required to meet ophthalmic drug standards and can carry known risks of contamination… and inconsistent potency, with potentially severe consequences, as leading retinal societies have reported,” while LYTENAVA comes with “standardized manufacturing, FDA-approved labeling, and robust pharmacovigilance.” The supporting datapoint is old but direct: of 152 U.S. and European retina physicians surveyed in 2019, “nearly 84 per cent indicated they had an interest or high interest in an approved ophthalmic formulation of bevacizumab.”

Distribution and launch: self-commercialised at home, partnered abroad

The split is explicit. Outlook intends to launch and market LYTENAVA in the United States itself. Around it sits a strategic collaboration agreement with Cencora, which the 10-K describes for the EEA, the United Kingdom and other ex-United States regions — “pharmacovigilance, regulatory affairs, quality management, market access support, importation, field solutions, third-party logistics services and distribution, and medical information” — and which the amended risk factors of August 12, 2026 describe more widely, as supporting “the commercial launch of LYTENAVA globally” and as a relationship for the launch “in the EU, UK, and in preparation for our commercial launch in the US and other markets.” In the United States the company says it relies on “a select network of third party distributors and other vendors.” That list of services is the entirety of what is disclosed: no upfront, no milestones, no royalty, no fee, no territory exclusivity, no term, no termination right and no execution date — and the agreement is not filed as an exhibit.

On timing, the quarterly report filed August 18 says: “We are making preparations to commercially launch in the U.S. prior to December 31, 2026,” and the October 1 release reaffirms a launch “before the end of 2026.” The $51.1 million raised in August is earmarked “to fund the commercial launch of LYTENAVA in the United States as well as for working capital and general corporate purposes.” The long-dated figure is framed as potential: LYTENAVA “has the potential to exceed $500 million in annual U.S. sales by 2030,” first stated August 14, 2026 and not revised. On the earnings call that day management added nearer-term guidance — spoken, not filed: “We expect total net revenue during the first 12 months following the U.S. launch of LYTENAVA to be between $50 million and $75 million, with Europe contributing approximately 10%-15% of that total.” It said its segmentation finds “about a third of the retina physicians, so just under 800, that represent just below 50% of that $500 million,” and that it expects quarterly SG&A “to approximately double from current levels by the end of calendar 2026, followed by an additional increase of approximately 10% during calendar year 2027.”

What the supply chain looks like

Two names and one dependency. Bulk drug substance comes from FUJIFILM Diosynth Biotechnologies, selected to conduct “all future manufacturing” of it; fill-finish is done by PCI San Diego, of which the filings say plainly “we are heavily dependent on PCI SD.” Everything is manufactured in the United States, which the company frames both ways: it “helps mitigate exposure to global tariff volatility” but “this concentration of manufacturing may adversely affect our pricing and competitiveness in the United Kingdom and European Union.” Purchase commitments sit at about $0.9 million, “primarily related to fill-finish manufacturing services.” One wrinkle sits inside a single annual report: two passages describe single-source suppliers for drug substance and fill-finish while a third describes multiple-source suppliers for the same functions. On inspections, the company’s filings go no further than the “open observations from pre-approval manufacturing inspections” cited in the 2023 complete response letter, but FDA’s action package does: its summary review quotes the product-quality office’s executive summary of August 20, 2025 — OPQA III “recommends approval of this product”, with “no remaining List of OPQ Deficiencies” — and records that inspections of two NORSE EIGHT clinical sites found the trial “appears to have been conducted adequately.” The company’s filings disclose no Form 483 or facility classification.

Europe, which is the only place the product has ever sold

The European position is two years older than the American one and, so far, a cautionary tale about the difference between an authorisation and a business. The European Commission granted a centralised marketing authorisation on 27 May 2024, after a CHMP positive opinion on 21 March 2024, valid in all 27 Member States; the holder is Outlook Therapeutics NL B.V. in Amsterdam, and the product is under additional monitoring. The MHRA’s United Kingdom authorisation, announced by the company on 8 July 2024, came through the International Recognition Procedure. Commercial launches ran Germany and the United Kingdom in June 2025 and Austria in January 2026. The plan for the next countries has moved through 2026: the February 17 release named Ireland and the Netherlands for 2026 and France, Italy and Spain for 2027; the May 15 release named the Netherlands and Ireland “later in 2026” and “additional European markets and beyond in 2027”; the August 14 release named only the Netherlands, “later in 2026 following its Netherlands submission”, with Switzerland in 2027 through the distributor Mediconsult; and the September 2026 investor presentation shows the Netherlands as “Target Launch Early 2027” and Switzerland as “Target Launch 2027.” France, Italy and Spain have not been named since February. All of it is subject to the standing caveat that “in certain European Union Member States, LYTENAVA must receive pricing and reimbursement approval before it can be sold.”

The resulting revenue is the number to sit with. FY2024: zero. FY2025: $1,413,535 net, on which cost of revenues was $1,356,330 and gross profit was $57,205 — roughly four per cent of a figure that was itself about a tenth of one quarter’s cash burn. Then it went backwards: negative $(1,071,694) for the nine months to June 30, 2026, and $8,700 in the June quarter against $1,505,322 in the same quarter a year earlier. The cause, verbatim: “primarily due to a $1.1 million increase in the returns reserve related to estimated product returns from our UK distributor due to lower-than-forecasted sell-through… we do not anticipate further adjustments for these batches in fiscal 2026.” The financial-statement notes give the fuller picture: over the nine months the returns reserve fell $1.1 million net — a $1.2 million increase for sales in the United Kingdom and Germany, offset by a $2.4 million reversal of reserves for product that “became unsellable due to minimum shelf-life constraints,” with a corresponding write-off of accounts receivable. The company’s February 17, 2026 release tied the December-quarter charge to “short dated product used for the initial shipments into the distribution channel in June 2025” and said European unit sales “more than doubled” in that quarter against the previous three months. Accrued rebates, including a United Kingdom VPAG rebate with the NHS, rose to $1,659,057 from $317,754. There is one reportable segment and no revenue disaggregation by country, so Germany, the United Kingdom and Austria cannot be separated. Greater China is licensed royalty-free to an 80/20 joint venture with a Syntone affiliate, and the company says it “continue[s] to pursue discussions with additional potential strategic partners for LYTENAVA outside of the US,” with no guarantee of agreement. One narrower deal is signed: on February 19, 2026 the company announced an exclusive commercial distribution agreement with Mediconsult AG, which will seek and hold the Swiss marketing authorisation and import, market and distribute LYTENAVA in Switzerland, with availability expected in 2027 subject to that authorisation; no financial terms were disclosed. Business development has a named owner too: Laura Cantrell joined as Vice President of Corporate Strategy and Business Development, announced January 6, 2026.

05 Cash, the going-concern statement and the clock at Nasdaq

This is the part of the page where the approval stops helping. Every figure below comes from the 10-Q for the quarter ended June 30, 2026, filed August 18, 2026 — the most recent periodic report, filed under the extension a Form 12b-25 allows: it was due on August 14, and on August 17 the company filed a notice citing “recent turnover in its finance department” and limited accounting staff after the August offering. The report itself concludes that disclosure controls and procedures were effective. The fiscal year ended September 30, 2026 and the fourth quarter has not been reported.

Cash rose over the year — from $8,900,814 at June 30, 2025 to $11,242,068, up $2,341,254 — which tells you nothing about the business and everything about the financing calendar. Then, on August 14, 2026, the August offering added $51.1 million net. Since inception the company has raised $612.0 million from equity and debt through September 30, 2025, and accumulated a $653.5 million deficit.

The burn, quarter by quarter

Two readings of that table, and one caveat. The caveat first: the quarterly operating cash figures for the second, third and fourth quarters of each fiscal year are derived by subtraction from the disclosed year-to-date totals, because the quarterly reports present cumulative figures only; the Q1 figures and all the year-to-date and annual figures are as reported. The first reading is that total burn is falling as a programme ends — $68.8 million in FY2024, $51.8 million in FY2025, $43.8 million in the first nine months of FY2026 — and that the mix has shifted hard, with research and development down to $11.4 million for nine months from $21.2 million a year earlier while selling and administrative costs ran at $25.7 million. The second reading is the opposite, and it is the more recent: the June quarter used $21.0 million, nearly three times the March quarter’s $7.8 million, the heaviest quarter in the series. Note also that the net-loss column is nearly uninformative on its own, because it is dominated by warrant and note fair-value swings — the $17.4 million of net income in the December 2024 quarter was not a profit, and the FY2025 loss absorbed $33,521,874 of warrant inducement expense against a $43,015,831 gain on warrant fair value.

One instrument outstanding, and a redemption right that is live now

The debt stack has been simplified to a single note, and the two that preceded it are gone. The December 2022 Streeterville convertible note, $31.8 million principal, was repaid in full on March 13, 2025 for $32,910,027.57 and cancelled. The March 2025 Avondale convertible note, $33.1 million face, was fully extinguished in May 2026 — $17.0 million repaid in cash on March 16, 2026 as a partial extinguishment booking a $285,600 gain, and $16,771,352 of principal and accrued interest converted into 39,404,455 shares, with extinguishment losses of $1,329,048 in the June quarter. On the way out it tripped once: the company did not make the $3.0 million Quarterly Debt Reduction Obligation due December 31, 2025, a “Major Trigger Event” that raised the outstanding balance by 10 per cent from January 1, 2026 — a $3.0 million increase in the note’s fair value — and reset the conversion price to the lesser of $2.26 or 90 per cent of the lowest closing bid of the prior three trading days, floored at $0.404. The company states that the trigger “has not resulted in an Event of Default”, and later conversions satisfied the obligation for the December and March quarters. The convertible-debt line in the dilution table is now nil, against 14,702,063 shares a year earlier. That is the good news in this section, and it was paid for in shares.

The share count, and what produced it

The dilution here is the dominant financial fact of the last two years, larger in its effect on a holder than anything the FDA did.

From 33,568,023 to 243,447,554 in fifteen months is an increase of about 625 per cent; from the 23,905,635 of September 30, 2024 it is about 918 per cent, and from the FY2025 year end alone about 352 per cent, all calculations. The drivers are all disclosed and all recent: 39,404,455 shares from the Avondale note conversions; 15,227,166 shares under the BTIG at-the-market programme in the nine months to June, implying an average gross price near $1.17; 17,005,831 shares under the H.C. Wainwright programme in the June quarter alone, near $0.86; 20,000,000 shares at $0.25 in March 2026; 16,129,033 at $0.31 in April; 8,539,709 at $0.5855 in a registered direct made exclusively to GMS Ventures in May; 55,555,556 at $0.99 in August; and 16,903,571 shares from warrant exercises in the nine months, 16,475,000 of them at $0.25. Authorised common was raised from 260,000,000 to 600,000,000 effective July 16, 2026, with 8,690,115 shares voted against.

The warrant overhang on top of that is large and now mostly struck above the recent trading range. 40,353,913 warrants were outstanding at June 30, 2026, including 5,782,496 and 1,071,429 at $7.70 (the liability-classified 2024 private placement tranches), 6,917,142 at $0.5855 and 8,571,428 at $0.5855 held by GMS after repricing, and 2,142,855 at $0.31 after a separate repricing. Issued since: 63,888,889 August 2026 warrants at $1.10, immediately exercisable to August 14, 2031, including the 8,333,333 over-allotment the underwriters took in warrants only, with no additional shares; 16,129,033 at $0.31 and 1,129,032 at $0.3875 from April, exercisable since July 16, 2026; and a further 1,129,032 contingent placement-agent warrants. Without that contingent tranche the total is 121,500,867 warrant shares, a calculation no filing makes. There are no pre-funded warrants anywhere in the record, and no tranche carries a contractual price reset — but the company has repriced warrants three times by negotiated amendment in twenty months, which is the same thing achieved by agreement. Options are 6,203,902 at a weighted $6.73 with an aggregate intrinsic value of $0 — every option outstanding is underwater — and no restricted stock units exist at all. Since June 30 the board has granted 100,000 options to the chief executive and 210,078 to the then chief financial officer on July 21, 2026 at $1.4304, vesting July 21, 2027, and a 500,000-share inducement option to the new chief financial officer at $0.63 on September 1, taking the pool to about 7.0 million before any forfeiture, a calculation. The Syntone inducement warrants for 2,142,858 shares at $2.26 were agreed in January 2025 and have never been issued, pending a Chinese regulatory approval the company says it may never receive.

The two funding taps, for completeness: the BTIG at-the-market programme was terminated effective May 12, 2026; the H.C. Wainwright programme of May 13, 2026, for up to $100 million at a 3.0 per cent commission, remains in force as an agreement but its prospectus supplement was terminated by a notice of August 10, 2026, reported on August 12, so no sales can be made “unless and until a new prospectus supplement is filed” — and as at October 3, 2026 none has been. The August 2026 prospectus supplement puts the unsold balance under that agreement at up to $85.4 million as of August 1, 2026. The $300 million universal shelf (333-278340) expires April 5, 2027, and no filing states the shelf’s own remaining capacity.

06 The rest of the pipeline, and what is missing from it

With LYTENAVA approved in the United States on July 24, 2026 under BLA 761320, the question stops being whether the company has a product and becomes whether it has anything behind one. The record gives a narrow answer: one molecule, one approved indication, one unfinished device study, three registration trials that exist only as agreements with the agency, and a licence to a joint venture in greater China. Everything else that reads like pipeline is geography.

What is left is territory rather than science. The European marketing authorisation was granted on 27 May 2024 across all 27 Member States and the United Kingdom authorisation, announced on 8 July 2024, under the International Recognition Procedure, carrying eight years of data exclusivity and ten years of market exclusivity in the EEA against twelve years of reference-product exclusivity anticipated in the United States. Launches have run Germany and the United Kingdom in June 2025 and Austria in January 2026, with the Netherlands now shown as “Target Launch Early 2027” in the September 2026 investor presentation, after “later in 2026” in the August 14 release, and Switzerland expected in 2027 through the exclusive distributor Mediconsult AG, subject to Swiss marketing authorisation; Ireland, named alongside the Netherlands as late as May, and France, Italy and Spain, named for 2027 in February, are absent from the latest materials — all subject to the company’s own caveat that in certain Member States the product “must receive pricing and reimbursement approval before it can be sold.” Launch support comes from a strategic collaboration with Cencora, covering pharmacovigilance, regulatory affairs, quality, market access, importation, logistics and distribution in Europe and described in the August 2026 risk factors as supporting the launch “globally”, including “in preparation for our commercial launch in the US”; not one economic term of it is disclosed — no upfront, no milestone, no royalty, no term, no termination right, no date and no exhibit. The United States the company intends to market itself, through “a select network of third party distributors.”

So the pipeline summary is short. One approved product, in one disease, in three jurisdictions. Behind it sit a device study running since 2021 that has posted no results, three registration trials with agency sign-off and no patients, an unsigned syringe supply agreement, a Chinese joint venture that has absorbed an initial $0.9 million and then $2,100,000 of cash contributions while disclosing no activity, and legacy biosimilars that survive mainly as a consent right held by a Mexican licensee. The breadth the company describes for bevacizumab in the eye is real as pharmacology and almost entirely unfunded as a programme.

07 Register, insiders and a short position that went up fifteen-fold

The register: one large holder diluted down, and almost everyone else is stale

There is one Schedule 13D on this company and it has been amended five times in twenty months. Everything else is a passive filing, and because the share count rose 625 per cent between May 2025 and August 2026, every percentage older than the August offering overstates the position it describes.

The arithmetic on the largest holder is the clearest picture of what has happened here. GMS was at 50.3 per cent on a Rule 13d-3 basis at the May 27, 2025 settlement, about 41.9 per cent at September 30, 2025, 22.9 per cent at May 28, 2026 and 16.4 per cent at August 12, 2026 — and it bought in every one of those windows. The percentage fell by two-thirds while the holder was adding shares. It is also not only the largest stockholder but a repeat primary financing counterparty — the January 2024 private placement, the January 2025 warrant inducement, the May 2025 underwritten offering, the May 2026 registered direct offering made exclusively to GMS and the August 2026 offering. It holds board designation rights under an amended and restated investor rights agreement and has two designated directors, Faisal G. Sukhtian and Yezan Haddadin, while Ghiath M. Sukhtian files as both director and ten-per-cent owner. And the other recent filers are financing participants rather than buyers in the market: Intracoastal’s event date is the April purchase agreement, CVI’s the August pricing, Sphera’s percentages include warrants from the company’s own placements. Avondale Capital converted $16,771,352 into 39,404,455 shares in the nine months to June 2026 and filed no 13D or 13G at all — consistent with 4.99-per-cent conversion blockers keeping it under the threshold at every instant, as with Atlas Sciences, Streeterville, Armistice and Syntone, none of which filed in 2025 or 2026. Armistice does appear in another document: the resale registration statement for the 17,258,065 April warrant shares (Form S-3, file 333-297532, effective July 21, 2026) lists Armistice Capital as beneficially owning 12,392,274 shares, 6.2 per cent, of which 11,820,274 are warrant shares — a prospectus figure, not a Schedule 13 filing.

One set of figures from outside the filings belongs beside that table, kept separate so the bases are not conflated. Paid market data as of October 2, 2026 put insider ownership at 19.21 per cent and institutional ownership at 18.46 per cent of 243.45 million shares outstanding, against a float of 196.67 million. Those are aggregated market-data figures on a different basis from the Schedule 13D and 13G record above: the 16.4 per cent attributed to GMS there is a Rule 13d-3 figure that treats 18,013,822 warrant shares as outstanding, which a share-count basis does not. The two tie in one respect — outstanding less float is 46.78 million shares and 19.21 per cent of the count is 46.77 million, a calculation, which indicates the data provider is treating closely held and insider stock as the same block. The figure that carries the most information is the institutional one. Institutional ownership below 19 per cent against a 196.67-million-share float is thin for a company with an approved product: it says the register is still held by one strategic holder, by financing counterparties sitting on warrants and by retail, rather than by long-only funds.

Insiders: ten purchases, not one sale, and no plan behind any of it

This is the most unusual part of the record, and it cuts in the company’s favour. Across twenty-five Forms 3 and 4 covering eighteen months there is not a single disposition — no code S, no code F, no code D, no code M, not one share sold by any insider, the only cancellation entries being the bookkeeping legs of a warrant repricing, immediately offset by reissuance at the lower strike. Against that, ten code-P purchases of common stock in ten Forms 4 between May 26 and August 14, 2026; the warrants bought alongside the August offering shares carry the same code in the derivative table, and the only other entries are option grants.

Two distinctions keep this from being over-read. The May, June and July buys by four directors are genuine open-market purchases, at $0.4093 to $1.0482, and the May cluster lands immediately after the May 26 announcement that the FDA had granted the appeal. The August purchases are participation in a public offering at the offering price — meaningful, but not buying from the market. On the plan question the answer is unambiguous: there is no plan. The Rule 10b5-1 box is not checked on a single Form 4 in the window, and the quarterly report states that in the three months to June 30, 2026 “none of the Company’s directors or Section 16 officers adopted or terminated any contract, instruction or written plan” intended to satisfy Rule 10b5-1(c). No tax-withholding footnote appears anywhere either, consistent with zero restricted stock units outstanding and no option exercises in the period.

Management: new commercial leadership and a field team still being built

Robert C. Jahr has been president, chief executive and a director since July 1, 2025, recruited from the chief commercial officer role at Sobi North America, on $600,000 with a 70-per-cent target bonus and an inducement option over 800,000 shares at $1.58. Kevin M. Lundquist became chief financial officer and principal accounting officer on September 1, 2026, from CapsoVision, on $450,000 with a 50-per-cent target and an inducement option over 500,000 shares at $0.63 — on materially leaner severance than his predecessors: nine months of salary and up to four months of benefits, against twelve months plus a full target bonus for Jahr and Kenyon. On July 21, 2026 the compensation committee granted Jahr 100,000 options and Kenyon 210,078 at $1.4304, vesting July 21, 2027, and approved cash bonuses of $420,000 and $200,000 payable only if the FDA approved the application by July 31, 2026 — a condition the July 24 approval met.

The September 2026 investor presentation listed a wider leadership group, including Jennifer Kissner, PhD, EVP Medical, Clinical and Regulatory Affairs, Greg DiPasquale, then Head of Commercial Operations, and Laura Cantrell, VP Corporate Strategy and Business Development. On October 9, 2026, the company appointed DiPasquale Chief Commercial Officer. He had joined in December 2025 and now leads all commercial functions. This closes the vacancy in that title following Jeff Evanson’s September 2025 departure; it does not establish that the planned field hiring is complete. Company release.

• Lawrence A. Kenyon, chief financial officer and interim chief executive from December 3, 2024 to June 30, 2025, left by mutual agreement reached August 23, 2026; the filing states expressly that the departure is “not the result of any disagreement with the Company on any matter relating to the Company’s accounting practices, financial statements, internal controls over financial reporting, operations, policies or practices.” He was to stay in a non-executive role to September 30, 2026 and, per the 8-K, to “step down from the Board on or before” that date; Merlintrader has found no public filing confirming the board exit.
• C. Russell Trenary III, the previous chief executive, was terminated without cause on December 3, 2024; the proxy shows $1,142,906 of cash severance — twelve months of salary, a full $420,000 target bonus and $28,009 of benefit premiums.
• The board is classified into three classes and has shrunk steadily: ten to nine on Trenary’s departure, back to ten for Jahr’s election, to nine on Dr Julia Haller’s resignation of March 11, 2026 — stated to be unrelated to any dispute — and, on the board’s August resolution, to eight on Kenyon’s exit, two of the eight being GMS designees. The auditor is KPMG LLP, ratified with 35,628,284 for and 751,849 against.
• Headcount was 17 full-time employees at September 30, 2025, none unionised, and no 2026 headcount exists in any filing. Two reductions sit behind that figure: five people, about 23 per cent of headcount, on December 13, 2024 at an estimated $0.3 million charge, and a second action in September 2025 described in the annual report’s notes, expense discussion and risk factors — “the Company implemented workforce reductions to conserve capital” — with the related cash payments expected to be “substantially complete by the third calendar quarter of 2026,” but with no headcount, no charge and no Item 2.05 report filed for it.

One governance figure belongs here because it describes the shareholders rather than the managers. Say-on-pay passed at the March 10, 2026 annual meeting with 9.8 per cent against. Four months later, at the July 16, 2026 special meeting, roughly 39 and 40 per cent of votes cast went against the two warrant-issuance and warrant-repricing proposals, 26.2 per cent against raising authorised shares to 600,000,000, and 28.6 per cent — 17,765,393 shares — against the reverse-split authority. The protest in this register is aimed at dilution mechanics, not at pay.

Litigation: a securities class action three years old and still called early

Two actions are live. The filings name neither case nor docket, but the court record does: the New Jersey action is Alsaidi v. Outlook Therapeutics, Inc., No. 23-cv-21862 (D.N.J.). The securities class action was filed in the District of New Jersey on November 3, 2023 against the company and certain officers, for a class period of August 3, 2021 to August 29, 2023 — ending on the day of the first complete response letter — alleging violations of Sections 10(b) and 20(a) and Rule 10b-5 by failing to disclose “an alleged lack of evidence supporting LYTENAVA” in wet AMD and “deficient CMC controls” unresolved when the application was resubmitted. On December 23, 2025 the court dismissed the second amended complaint in part with prejudice and in part with leave to amend; plaintiffs filed a third amended complaint in February 2026 and a revised third amended complaint in April 2026, and no ruling on it is disclosed. The shareholder derivative action, filed in the District of Delaware on October 10, 2024 against officers and directors for breach of fiduciary duty, a Section 14(a) violation, contribution, unjust enrichment and waste, is stayed pending final resolution of the New Jersey case.

The accrual language is the part worth printing. The company “is not able to estimate the possible cost to it from these matters, as the pending lawsuits are currently at an early stage”, and it “has not established any reserve for any potential liability”, while allowing that amounts “could be material… if it does not prevail… or even if it does prevail.” Calling a case early stage in August 2026, nearly three years after filing and after a partial dismissal with prejudice and three amended complaints, is a characterisation a reader is entitled to weigh. Beyond those two matters the record is clean: management states there are “no additional claims or actions pending” that would be material, and no investigation, no SEC enforcement action, no patent or BPCIA litigation and no product-liability claim is disclosed anywhere.

Short interest rose fifteen-fold in eleven months

From 2,168,058 shares at the October 15, 2025 settlement to 32,220,502 at September 15, 2026 is a 14.9-fold increase, against a share count that rose roughly 4.5-fold from the 53,887,233 of September 30, 2025 — so the position grew by considerably more than the issuance behind it. The latest reading is 13.2 per cent of the 243,447,554 shares outstanding at August 14, 2026, and the peak came two weeks earlier, 32,910,138 at August 31, immediately after the offering closed. On the narrower base the position is larger: the same market data give short interest at 16.38 per cent of a 196.67-million-share float, and the two bases are consistent — 16.38 per cent of that float is about 32.21 million shares, a calculation, against the 32,220,502 the exchange published for September 15. The float figure is the more meaningful of the two here, because roughly 46.8 million shares are closely held and are not available to borrow.

The days-to-cover column needs care here, because its denominator dominates it. It sat at the 1.00 floor on nine of the twenty-three settlement dates, because reported average daily volume exceeded the short position. The 7.67 at September 15 is the highest in the series and is the product of volume collapsing to 4,202,688 shares a day from 20,062,834 two weeks earlier, not of shorts adding — the position itself fell slightly over that fortnight. The 65,462,641-share average daily volume at the May 29 settlement is the other extreme, and it brackets the May 26 appeal win and the May 29 GMS offering. The last published settlement is September 15, 2026, a lag of about 17 calendar days; the September 30 figure is due around October 9, 2026.

Analyst coverage

One consensus figure is worth printing with its provenance attached. Paid market data as of October 2, 2026 show a consensus price target of $5.03; on Nasdaq’s data the stock closed at $0.593 on October 2, 2026, up 3.1 per cent. The same data source gives 243.45 million shares outstanding, a float of 196.67 million, insider ownership of 19.21 per cent, institutional ownership of 18.46 per cent and short interest at 16.38 per cent of float — the last of which is consistent with the 32,220,502 shares published by the exchange for the September 15 settlement and corroborates the float figure. Nasdaq’s data put the 63-session average volume at 13,730,313 shares and the 20-session average at 4,510,393, both calculations, against a session volume of 5,619,583; the market value at that close is about $144.4 million, a calculation on 243,447,554 shares. Individual rating actions are listed in the next paragraph only where Dow Jones Newswires reported them and an analyst-ratings table confirms them. The consensus is a third-party market-data aggregate, shown as such, and none of the analysis is built on it.

Four rating actions since the approval can be traced to Dow Jones Newswires and are confirmed by published analyst-ratings data. On July 27, 2026, three days after the FDA decision, BTIG raised the stock to Buy from Neutral with a $4.00 target, and H.C. Wainwright kept Neutral while raising its target to $1.60 from $0.50. On August 14 BTIG kept its $4.00 target, and on September 11 Ascendiant Capital kept Buy and raised its target to $10.50 from $10.00; published ratings tables round the H.C. Wainwright and Ascendiant figures to whole dollars. Two of the three firms have also acted for the company: according to the company’s filings, BTIG was the agent of the at-the-market programme terminated on May 12, 2026, and H.C. Wainwright is the agent of the replacement programme, suspended in August. The range, from $1.60 to $10.50 against a $0.593 close on October 2, is wide, and the published headlines do not give the reasoning behind each note. These are third-party opinions, not a Merlintrader view and not a recommendation. Analyst-ratings table

Separately, and not to be read alongside that consensus, the company has published two forward revenue figures of its own, both dated August 14, 2026: in the release, verbatim, “Based on current estimates, management believes LYTENAVA has the potential to exceed $500 million in annual U.S. sales by 2030”; and on the earnings call that day, as guidance, total net revenue “between $50 million and $75 million” in the first twelve months after the U.S. launch, Europe “approximately 10%-15% of that total.” That is the company speaking about its own product, not third-party research, and it carries the weight that distinction implies.

08 Catalyst map, red flags and the Merlintrader bottom line

The catalyst map

The regulatory calendar is empty, and that is the point: the approval came on July 24, 2026 and no FDA decision, goal date or pending application exists anywhere in the record. Everything near now is commercial and financial — a launch, a first revenue line, payer coverage, a listing deadline and money.

Red flags

• The accounts carry a going-concern qualification, and it is unambiguous. Verbatim, August 18, 2026: “there is substantial doubt about the Company’s ability to continue as a going concern”, and management “does not believe that the existing cash and cash equivalents as of June 30, 2026 are sufficient to fund the Company’s operations through one year” — a conclusion the report repeats including the $51.1 million of August proceeds. The same doubt was present in the December 2025 annual report, so it has been continuous across both filings, and the stated downside is explicit: the company may have to “make reductions in our workforce, scale back our plans… liquidate all or a portion of our assets, and/or seek protection under the provisions of the U.S. Bankruptcy Code.” Behind it: an accumulated deficit of $653,540,250, a stockholders’ deficit of $(10,392,567) and a working capital deficit of $(5,510,850). For scale, at the October 2, 2026 close of $0.593 the whole company is worth about $144.4 million, a calculation — so the qualification attaches to an equity valued at that against the $11,242,068 of cash behind it and the $51.1 million raised in August.
• The shares are below $1.00 and the clock is running. The stock closed at $0.593 on October 2, 2026, on Nasdaq’s data — about 40.7 per cent below the $1.00 minimum, a calculation. Nasdaq issued a minimum-bid-price deficiency notice on September 25, 2026 under Rule 5550(a)(2) after 30 consecutive business days under $1.00, with a compliance date of March 24, 2027; it is active and unresolved. This is the sixth bid-price deficiency in the company’s history: 2018 escalated to a delisting determination and was cured only by the 1-for-8 reverse split; two notices followed in 2020, dated March 27 and October 8, neither of which led to a reverse split; the 2023 notice was cured by the 1-for-20 reverse split; and the February 18, 2026 notice was cured on June 26, 2026 by ten days above $1.00 after the appeal news. The stock first closed back below $1.00 on July 28, 2026, the second session after the approval, and the 30-business-day run behind the September notice began on August 13. Two filings describe that February letter irreconcilably, one as a deficiency with “no immediate effect”, the other as notice that “Nasdaq staff had determined to delist”.
• The dilution record is the heaviest item on the record. Since January 2024: a $60 million private placement agreed January 22, 2024 and closed in two tranches — 8,571,423 shares with warrants for 12,857,133 at $7.00 on March 18, 2024 for $55.5 million net and 714,286 shares with warrants for 1,071,429 on April 15, 2024; a warrant inducement on January 16, 2025 repricing exercises to $2.51 from $7.70 and issuing 14,149,274 new warrants at $2.26 for $16.6 million; a $33.1 million convertible note to Avondale on March 13, 2025; an underwritten offering closing May 27, 2025 of 9,285,714 shares with 18,571,428 warrants at a combined $1.40; 8,681,852 shares through the BTIG at-the-market programme in FY2025 and 15,227,166 more in the nine months to June 2026; a best-efforts offering on March 25, 2026 of 20,000,000 shares with 20,000,000 warrants at a combined $0.25; the $18,360,000 Atlas note on March 16, 2026 at a $1,360,000 original issue discount; a registered direct on April 23, 2026 of 16,129,033 shares at $0.31 with matching private warrants; a registered direct to GMS alone on May 29, 2026 of 8,539,709 shares at $0.5855; 17,005,831 shares through the Wainwright programme in the June quarter; and the follow-on underwritten by Piper Sandler and BTIG and closing August 14, 2026 of 55,555,556 shares with 55,555,556 warrants at a combined $0.99 for about $51.1 million net, the over-allotment taken in warrants only for a further 8,333,333. Avondale separately converted $16,771,352 into 39,404,455 shares. The count ran 23,905,635 at September 30, 2024, 53,887,233 a year later, 187,096,998 at June 30, 2026 and 243,447,554 at August 14, 2026 — up about 918 per cent in under two years, 625 per cent in the fifteen months from May 12, 2025 and 352 per cent in fiscal 2026 to date, all calculations, with authorised common raised from 260,000,000 to 600,000,000 on July 16, 2026 to make room.
• Warrants have been repriced three times by negotiated amendment in twenty months — the January 2025 inducement to $2.51, the April 2026 amendment of 2,142,855 Armistice warrants to $0.31, and the May 2026 amendment of 15,488,570 GMS warrants from a weighted-average $1.78 to $0.5855. No tranche carries a contractual reset; the effect is the same.
• The overhang is large and no filing totals it. 40,353,913 warrants outstanding at June 30, 2026, plus 63,888,889 August warrants at $1.10, 16,129,033 April warrants at $0.31, 1,129,032 placement-agent warrants at $0.3875 and a further 1,129,032 contingent — 121,500,867 before the contingent tranche, a calculation — against 243,447,554 shares. All 6,203,902 options outstanding at June 30 were underwater, at an aggregate intrinsic value of $0, and the 810,078 granted since are struck at $1.4304 and $0.63, above the October 2 close.
• The revenue line is negative. Net revenue for the nine months to June 30, 2026 was $(1,071,694), against $1,505,322 in the June 2025 quarter alone, with the June 2026 quarter at $8,700 — caused, on the company’s explanation, by a $1.1 million increase in the returns reserve on “lower-than-forecasted sell-through” from the UK distributor; the notes add a $2.4 million reserve reversal and matching receivables write-off for product that became unsellable under minimum shelf-life limits, the reserve falling $1.1 million net over the nine months. FY2025 revenue of $1,413,535 produced a gross profit of $57,205.
• Operating cash use has not fallen with the headcount — $(43,777,688) over nine months to June 2026 against $(39,452,240) a year earlier, and $(51,829,437) for FY2025 — against cash of $11,242,068 at June 30, 2026. No expense guidance has been filed; on the August 14 earnings call management said quarterly SG&A should roughly double from current levels by the end of calendar 2026 and rise about 10 per cent more in 2027 — company guidance that points to a heavier burn, not a lighter one.
• The lender can take $3.0 million a quarter at will, from September 16, 2026, with a 7.5 per cent exit fee on each payment; on a Trigger Event Atlas may increase the outstanding balance, and on default accelerate and apply default interest of up to 22 per cent a year. No event of default is disclosed. On the retired Avondale note, a missed $3.0 million payment at December 31, 2025 triggered a 10 per cent balance increase and a conversion-price reset, which the company says did not become an event of default.
• Both at-the-market programmes are closed. BTIG was terminated May 12, 2026; the Wainwright prospectus supplement was terminated by a notice of August 10, 2026 and the programme has been suspended since, needing a new supplement to restart, with up to $85.4 million unsold under it as of August 1. The shelf behind it expires April 5, 2027 with its own remaining capacity unstated.
• A reverse split at 1-for-10 to 1-for-50 is approved but unexecuted, at board discretion until the first anniversary of the July 16, 2026 meeting, after which the proxy says the board would seek approval again, and 28.6 per cent of votes cast — 17,765,393 shares — went against it. The two previous splits, March 2019 and March 2024, were both cures for bid-price deficiencies. Shareholders also voted roughly 39 and 40 per cent against the two April 2026 warrant proposals and 26.2 per cent against the authorised-share increase, against 9.8 per cent on say-on-pay four months earlier.
• Commercial execution remains to be demonstrated. Greg DiPasquale was appointed Chief Commercial Officer on October 9, 2026. The latest previously disclosed headcount remains the September 30, 2025 figure of 17; the appointment release supplies no updated total or confirmation that the planned 30 customer-facing and 20 reimbursement roles have all been filled. The chief financial officer changed on September 1, 2026. Company release.
• The securities class action is approaching three years old and the company still calls it “at an early stage”, with no reserve established, after a partial dismissal with prejudice and three amended complaints; the derivative suit is stayed behind it, and neither case name nor docket number appears in any filing — the court docket is Alsaidi v. Outlook Therapeutics, Inc., No. 23-cv-21862 (D.N.J.).
• The approval rests on one adequate and well-controlled trial plus confirmatory evidence, after three complete response letters and a won appeal. NORSE EIGHT missed its primary endpoint — an LS mean difference of −2.3 letters (95 per cent CI −4.0 to −0.5) at week 8 in the FDA label, −2.257 (−4.044 to −0.470) in the company’s figures, against a non-inferiority margin of −3.5, and “the hypothesis of noninferiority was not met.” NORSE ONE, the other trial intended as pivotal, failed and numerically favoured ranibizumab, per FDA’s first complete response letter. The NORSE EIGHT results posted on ClinicalTrials.gov show more serious adverse events on ONS-5010 than on ranibizumab, 9 against 5 of 200, including acute coronary syndrome, myocardial infarction and a basal ganglia stroke in the ONS-5010 arm, and the label may carry no superiority or non-inferiority claim against ranibizumab.
• Manufacturing runs through two outside parties and the record contradicts itself on how many. FUJIFILM Diosynth makes all future drug substance and PCI San Diego does fill-finish, on which the company says it is “heavily dependent”; one annual report says single source in two places and multiple source in a third. The company’s filings disclose no inspection outcome beyond the first complete response letter’s “open observations from pre-approval manufacturing inspections” and no Form 483 or facility classification; FDA’s August 2026 action package fills part of the gap, its product-quality office having recommended approval in August 2025 with “no remaining List of OPQ Deficiencies.”
• The ex-United States commercial arrangement is undisclosed in every economic respect. The Cencora collaboration, described in August as supporting the launch “globally”, has no upfront, milestone, royalty, term, termination right or execution date in the record and is not filed as an exhibit; revenue is not disaggregated by country.
• One holder has outsized influence by right as well as by size. GMS holds board designation rights and two of eight directors, peaked at 50.3 per cent in May 2025, and has been the counterparty to four of the last six equity financings, including one made to it alone concurrently with the repricing of 15,488,570 of its own warrants.
• Short interest is 13.2 per cent of the shares outstanding at the September 15, 2026 settlement, up 14.9-fold in eleven months against a count up roughly 4.5-fold since September 30, 2025.
• Smaller inconsistencies in the company’s own documents: the third complete response letter dated December 30 in five press releases and December 31 — the day it was announced — in two filings, with one release calling it “December 2026”, FDA’s records settling it as December 30, 2025; warrant inducement expense as $33,521,874 or $33,856,814, the nine-month figure exceeding the full-year one; and $63,530 of a departing officer’s severance that the proxy’s own components do not add up to.

What to watch, in order

1. Whether the United States launch happens before December 31, 2026, and what the first revenue figure is. The price is set at $485 per vial and the money is raised and earmarked for it. This is the only item here that can change the company’s financial shape by itself, and the FY2026 annual report, due by December 29, is where the first evidence lands.
2. Whether CMS grants a permanent HCPCS J-code. The company announced on October 1, 2026 that it had applied for one ahead of the launch; CMS reviews such applications quarterly, and on the August 14 earnings call management said it expects a code in April 2027, an expectation rather than a decision; until then claims run through a miscellaneous code, with manual review and a higher risk of denials. For a drug a physician buys, administers and bills, permanent coding is what turns a list price into routine reimbursement, so this sits directly in front of the revenue question above — and an application is not a grant.
3. The bid price against March 24, 2027. Ten consecutive business days at $1.00 or above cures the deficiency; the launch news flow is the only non-structural way to get there. If it does not happen, the reverse-split authority that expires July 16, 2027 is the tool already in the board’s hand, and the company has used that tool twice before for exactly this reason.
4. The next financing, and what form it takes. The stated runway ends in the second quarter of calendar 2027; the Wainwright programme needs only a new prospectus supplement to restart and the shelf expires April 5, 2027. The pattern of the last twenty months — priced equity with full warrant coverage, repeatedly to the same holder — is the base case against which any change should be read.
5. Whether Atlas starts redeeming. Up to $3.0 million a quarter has been available to the lender since September 16, 2026, each payment carrying a 7.5 per cent exit fee, against $20.3 million outstanding and a June 16, 2027 maturity. It shows up in the financing line.
6. Payer coverage and what reaches the net price. Europe is the warning: a $1.1 million returns reserve turned a revenue line negative and accrued rebates rose to $1,659,057. Gross price is disclosed; no gross-to-net assumption is.
7. Follow-through after the CCO appointment. Greg DiPasquale became Chief Commercial Officer on October 9, 2026. The next evidence is launch availability, access and reimbursement execution, and completion of the planned field team; the appointment alone does not establish those outcomes. Company release.
8. The fourth-quarter burn and the September cash figure. Operating cash use was $(21,010,522) in the June quarter on a derived basis; whether launch spending has lifted it again is what the annual report answers, against management’s own guidance on the August 14 earnings call that quarterly SG&A will roughly double by the end of calendar 2026.
9. Any ex-United States partnering agreement beyond Switzerland. The company says it “continue[s] to pursue discussions” with no guarantee any agreement would cover anticipated commercialisation costs. A signed deal with disclosed economics would be the largest change to the funding picture that does not involve selling shares.
10. Whether NORSE FOUR, FIVE or SIX gets a protocol number. Three agreed special protocol assessments, the stated gate removed on July 24, 2026, and still nothing registered.
11. A ruling on the revised third amended complaint, with no reserve recorded against it.
12. Whether insider buying continues, and whether any insider sells. Eighteen months, ten purchases, zero sales and no Rule 10b5-1 plan disclosed. The first sale would be the first break in that pattern.

Merlintrader bottom line

The useful way to read Outlook Therapeutics is that the regulatory argument is over and won, and almost nothing else is settled. The chronology is unusual enough to state plainly: an application first submitted in March 2022, withdrawn, resubmitted, and then refused three times — August 29, 2023 on manufacturing and evidence, August 27, 2025 on evidence alone, December 30, 2025 on evidence again after the company believed it had aligned with the agency. The company then did something few of its size attempt: it filed a formal dispute resolution request on March 23, 2026, met the deciding official on April 20, and by letter of May 20, 2026, announced May 26, the Office of New Drugs granted the appeal, concluding that NORSE TWO plus confirmatory evidence established substantial evidence of effectiveness. The resubmission, dated May 29, went in and BLA 761320 was approved on July 24, 2026, five days ahead of the goal date. LYTENAVA is the first and only ophthalmic formulation of bevacizumab approved in the United States, the European Union and the United Kingdom, with twelve years of reference-product exclusivity anticipated. No FDA decision is pending on anything.

What that approval buys is access to a real market with an unusual shape. Roughly 66 per cent of new United States wet-AMD patient starts use off-label repackaged bevacizumab from compounding pharmacies, by the company’s cited survey; the annual report counts about 2.7 million off-label repackaged bevacizumab injections a year in the United States across all uses and cites 2.8 million for Europe. The product is priced into that gap rather than against it: a $485 wholesale acquisition cost per vial, set September 16, 2026, with a stated intention to sit “as a lower-cost alternative to biosimilars and branded products, while remaining above off-label compounds.” The pitch is standardised manufacturing, approved labelling and pharmacovigilance against the contamination and potency risks of repackaging. That is a coherent commercial proposition and it is entirely untested in the United States, because the launch has not happened yet. The first yardstick is management’s own: $50–75 million of total net revenue in the first twelve months after launch, Europe included, as guided on the August 14 earnings call.

The balance sheet is where the story turns. Cash of $11,242,068 at June 30, 2026 plus $51.1 million of August net proceeds, against an accumulated deficit of $653,540,250, a stockholders’ deficit of $(10,392,567), a working capital deficit, $20.3 million of debt classified current and maturing June 16, 2027, and operating cash use of $(43,777,688) over nine months. The company’s own conclusion, with the August money counted, is that funding runs “only… into the second quarter of calendar year 2027” and that substantial doubt about going concern exists. The European revenue record offers no offset yet: $1,413,535 for FY2025 at a $57,205 gross profit, then a negative $(1,071,694) over nine months on a returns reserve. Both at-the-market programmes are closed, the shelf expires in April 2027, and the shares are back below $1.00 — $0.593 at the October 2, 2026 close, for a market value of about $144.4 million, a calculation — with a March 24, 2027 Nasdaq deadline and an unexercised 1-for-50-capable reverse split sitting in the board’s hand until July.

Two things sit either side of that, and both are real. On one side, the register has been diluted beyond recognition — 23,905,635 shares at September 30, 2024 to 243,447,554 by August 14, 2026, with every former five-per-cent holder except one pushed under the threshold without selling, warrants repriced three times by negotiation, 39 to 40 per cent of votes cast against the warrant proposals, and short interest up fifteen-fold to 13.2 per cent of the count. On the other, the insider record is the cleanest part of the picture: ten purchases, not one sale in eighteen months, no Rule 10b5-1 plan disclosed, four directors buying in the open market at $0.4093 to $1.0482, six of eight taking fees in options rather than cash, and the chief executive, the chief financial officer and the largest holder all subscribing to the August offering. A reader can hold both of those facts at once; they describe different things.

Which leaves a short, checkable list over the next two quarters, none of it requiring a view on retinal medicine. Does LYTENAVA launch in the United States before December 31? Does the annual report carry a real revenue figure and a fourth-quarter burn that has not grown? Does the next financing arrive, and on what terms? Does the bid price recover before March 24, or does the reverse split get used? If those land well, a company that spent four years and three refusals getting a label spends 2027 selling into a market where about two-thirds of new patient starts are treated with off-label repackaged bevacizumab. If they do not, a company with a going-concern qualification, a closed at-the-market programme, a lender entitled to $3 million a quarter and a listing deadline in March will be having a financing conversation from a weaker position than the one it had in August.

None of the above is a recommendation, a target, or a view on the share price. It is a description of what is documented, what is guided, what is calculated and labelled as such, and what the record does not say.

What the public record does not confirm. The approval, the indication, the vial presentation and the single postmarketing commitment come from the FDA approval letter and prescribing information of July 24, 2026. The European and United Kingdom authorisations come from the European Medicines Agency assessment report and the Forms 8-K of May 28 and July 8, 2024. The four-cycle review history comes from the Forms 8-K with their exhibits and, since its publication in August 2026, from FDA’s own action package for BLA 761320 — the three complete response letters, the appeal decision, the summary review and the 2023 Office Director memo — which this page follows wherever FDA’s dates or wording differ from the company’s. Balance-sheet, debt, going-concern, warrant, option and litigation detail comes from the Form 10-Q for the quarter ended June 30, 2026; licence terms, manufacturing, market figures, tax attributes and headcount from the Form 10-K for the fiscal year ended September 30, 2025. Trial figures come from the Form 8-K of November 27, 2024, the week-12 NORSE EIGHT data reported in January 2025, the annual report, the FDA label and appeal letter, and, for NORSE EIGHT safety, the results posted on ClinicalTrials.gov, with both counts given wherever the registry and the company differ. Ownership comes from the Schedules 13 at their event dates, which range from about seven weeks to about fifteen months old; short interest is as of the September 15, 2026 settlement. The closing price, session change and volume are Nasdaq historical data for October 2, 2026, and the average volumes and market value are calculations from them; float, ownership percentages and the consensus target are aggregated market data read on October 2, 2026, not company disclosure, and are labelled as such wherever they appear. The J-code application is taken from the company’s release of October 1, 2026 and does not yet appear in any periodic report. Statements attributed to the August 14, 2026 earnings call — the revenue guidance, the field-force plan, the SG&A trajectory and the J-code timing — are management’s spoken guidance as recorded in a published transcript of the call, not filed figures. Data are as of October 3, 2026; the most recent filing covered is the Form 8-K filed September 30, 2026. The fiscal year ends September 30, so fiscal 2026 closed three days before the data date and its fourth quarter has not been reported.

The following figures on this page are calculations from published data and are labelled as such where they appear: the working capital deficit of $(5,510,850), being filed current assets of $23,168,201 against current liabilities of $28,679,051; operating cash use for the second, third and fourth quarters of each fiscal year, each derived by subtraction from the cumulative year-to-date totals the quarterly reports disclose, only the first quarter being separately reported, and the fourth-quarter 2025 revenue, expense and loss lines derived the same way; the share-count increases of about 918 per cent from September 30, 2024, 625 per cent over the fifteen months from May 12, 2025 and 352 per cent over fiscal 2026 to date; the aggregation of warrants and options outstanding after June 30, 2026 to 121,500,867 warrant shares, before a further contingent 1,129,032, and about 7.0 million options, which no filing states as a combined total; the market value of about $144.4 million at the October 2 close and the 20- and 63-session average volumes; CVI Investments’ roughly 8.2 per cent from shares alone; the short interest at 13.2 per cent of shares outstanding and its 14.9-fold increase; the implied average gross price on each at-the-market programme; the average quarterly cash burn; the distance from the $0.593 close to the $1.00 minimum bid, being about 40.7 per cent below it and a rise of about 68.6 per cent to reach it; the split of the warrant tranches into roughly 39.8 million shares in the money and about 81.7 million out of it at that close, strike by strike; and the weighted total of the health score.

The following are not established on the public record: any pending FDA decision, goal date, application or supplement, because the application was approved on July 24, 2026 and nothing has been filed since — the pre-filled-syringe supplement is contemplated but not submitted; any advisory committee, none ever having been convened for this product; any United States revenue figure, unit volume, gross-to-net assumption, confirmed field headcount or launch date more precise than “prior to December 31, 2026” — what exists is the $485 per vial price set on September 16, 2026, the field plan and the $50–75 million first-twelve-month revenue guidance given on the August 14 earnings call, and a management statement that the product has “the potential to exceed $500 million in annual U.S. sales by 2030”; any granted HCPCS J-code for LYTENAVA — the company announced on October 1, 2026 that it has applied to CMS for a permanent code, and an application is not a granted code, and the April 2027 date management gave on the August 14 call is an expectation, not a CMS decision; any revenue disaggregated by country for Germany, the United Kingdom or Austria, and no customer-concentration disclosure, there being one reportable segment; any filed expense or cash-burn guidance — the only expense outlook is the SG&A trajectory given orally on the August 14 earnings call; any cash figure after June 30, 2026; any headcount after September 30, 2025, when it was seventeen; any economics of the Cencora collaboration — no upfront, milestone, royalty, fee, territory grant, term, termination right or execution date, and it is not filed as an exhibit; the counterparty or terms of the pre-filled syringe supply agreement, still being negotiated; any Form 483, establishment inspection report, warning letter or facility classification in the company’s filings — though FDA’s August 2026 action package shows the product-quality office recommending approval with “no remaining List of OPQ Deficiencies”; the remaining capacity on the $300 million shelf, which no filing states — the Wainwright programme’s own remainder, up to $85.4 million as of August 1, 2026, is stated in the August prospectus supplement — and no baby-shelf limitation stated in any prospectus supplement reviewed; the case names and docket numbers of the two lawsuits in any filing — the New Jersey class action’s public docket is Alsaidi v. Outlook Therapeutics, Inc., No. 23-cv-21862 — and any litigation accrual or estimable range, none having been recorded; any enrolment, site, date or registry number for NORSE FOUR, FIVE and SIX, which exist only as agreed protocols whose “timing for initiating these studies has not been determined”; any posted registry results for NORSE ONE or NORSE SEVEN; the split of fiscal-2026 first-quarter and second-quarter revenue; any affiliation among the three successive note holders; the first and second amendments to the December 2022 note, only the third being described; any preferred stock terms, ten million shares being authorised, none ever issued and no series designated; any restricted stock unit, none outstanding; any stockholders’-equity or market-value deficiency notice in the Capital Market era, despite the stockholders’ deficit; any certificate of amendment effecting the reverse split authorised on July 16, 2026, the authority to be used before the first anniversary of that meeting and unexercised as of October 3, 2026; any company-disclosed float, market value, share price or analyst figure — no filing gives a float or a market value, and individual analyst actions appear only where Dow Jones Newswires reported them and an analyst-ratings table confirms them, and beyond those only a labelled aggregated market-data float of 196.67 million and consensus target of $5.03 appear on this page, beside a market value calculated from Nasdaq’s closing price, and nothing is built on any of them; any short-interest figure after the September 15 settlement; any ownership position with an event date after August 12, 2026; and any strategic review, adviser appointment, takeover approach, acquisition proposal or sale process of any kind, all confirmed absent across the annual report, the quarterly report, sixty-five Forms 8-K and both proxy statements rather than merely unfound — what exists instead is ex-United States partnering discussion, consummated so far only in the Swiss distribution agreement with Mediconsult, stated as such. There is also no government or regulatory investigation, no enforcement action, no patent litigation, no product-liability claim, no biosimilar-pathway litigation, no term loan, no revolving facility, no royalty financing and no secured debt of any kind.

Several inconsistencies in the company’s own documents are printed rather than resolved: the third complete response letter dated December 31, 2025 in one Form 8-K and the quarterly report and December 30, 2025 in five successive press releases, with a sixth calling it a “December 2026” letter, FDA’s own records dating it December 30, 2025; the original resubmission dated and received August 29, 2022 by the agency and August 30, 2022 in the annual report; the February 2026 Nasdaq letter described contemporaneously as a deficiency notice with “no immediate effect” and retrospectively, four months later, as notice that staff “had determined to delist”; NORSE TWO enrolment of 227 against 228 and sites of 39 against 50; suppliers described as single source in two places and multiple source in a third within one annual report; and warrant inducement expense of $33,521,874 for a full year against a $33,856,814 nine-month comparative in a later filing.