Axsome Therapeutics ($AXSM) Stock Hub 2026: Q2 Revenue Of $218.4M, A $208.1M Selling Line, And The Alzheimer’s Agitation Launch

01 Executive Summary

Axsome Therapeutics, Inc. is a commercial-stage biopharmaceutical company headquartered in New York that develops and sells medicines for central nervous system conditions. It has three products approved by the U.S. Food and Drug Administration and on the market: AUVELITY (dextromethorphan hydrobromide and bupropion hydrochloride) for major depressive disorder and, since April 30, 2026, for agitation associated with dementia due to Alzheimer’s disease; SUNOSI (solriamfetol) for excessive daytime sleepiness associated with narcolepsy or obstructive sleep apnea; and SYMBRAVO (meloxicam and rizatriptan) for the acute treatment of migraine. That combination puts Axsome in an unusual position for a company of its size: it is not a clinical-stage story waiting for a single readout, but a selling organisation whose economics are still upside down.

The second quarter of 2026, reported before the U.S. market open on August 10, 2026, produced total revenue of $218.4 million, up 46% from $150.0 million a year earlier and up 14% from $191.2 million in the first quarter. AUVELITY contributed $180.3 million of net product sales, up 51% year over year. SUNOSI contributed $35.8 million of total revenue, of which $33.8 million was net product sales, $1.5 million royalty revenue from out-licensed territories and $0.5 million milestone revenue. SYMBRAVO contributed $2.3 million of net product sales.

Against that revenue the company recorded $13.6 million of cost of revenue, $46.2 million of research and development, $208.1 million of selling, general and administrative expense, $1.6 million of intangible amortisation and a $1.5 million gain on the fair value of contingent consideration, for total operating expenses of $268.0 million and an operating loss of $49.6 million. After $1.5 million of net interest expense and $0.1 million of income tax, the net loss was $51.3 million, or $(0.99) per share on 51,799,708 weighted average shares. The comparable figures for the second quarter of 2025 were a $48.0 million net loss and $(0.97) per share. The loss included $27.1 million of stock-based compensation, a non-cash item.

The balance sheet is thinner than the revenue line suggests. Cash and cash equivalents were $319.9 million at June 30, 2026 against $322.9 million at December 31, 2025 and $305.1 million at March 31, 2026, so the company consumed cash in the first quarter and rebuilt it in the second. Total assets were $761.4 million and total liabilities $678.1 million, leaving $83.3 million of shareholders’ equity against an accumulated deficit of $1.42 billion. Debt consists of a $120.0 million first lien senior secured term loan from Blackstone, carried net of unamortised discount at $118.0 million, plus $70.0 million drawn under a super senior revolving facility, which the balance sheet classifies as short-term borrowings. Contingent consideration owed to Jazz Pharmaceuticals on U.S. SUNOSI sales stood at $80.3 million between current and non-current portions.

The share count is a relatively quiet part of this story. Shares issued and outstanding went from 50,882,766 at December 31, 2025 to 52,299,889 at June 30, 2026, an increase of 1,417,123 shares or 2.8% over six months. Most of that came from option exercises and equity awards, but not all of it: the first quarter Form 10-Q discloses that the company sold 35,802 shares for approximately $6.3 million of gross proceeds under the March 2022 at-the-market sales agreement with Leerink Partners during the first quarter of 2026, against 156,484 shares for $19.7 million in the first quarter of 2025. Additional paid-in capital rose from $1.394 billion to $1.505 billion over the six months. Axsome has not announced an underwritten public offering in 2026, and the company states that current cash is sufficient to fund operations into cash flow positivity based on the current operating plan, without attaching a date to that statement.

The dated events ahead are few and specific. The FDA has set May 1, 2027 as the target action date for AXS-12 in cataplexy associated with narcolepsy. Topline results from the ENGAGE Phase 3 trial of solriamfetol in binge eating disorder are guided for the fourth quarter of 2026, and topline results from the SUSTAIN Phase 3 trial in shift work disorder for 2027. A pivotal Phase 2/3 trial of AXS-05 in smoking cessation is guided to start in the third quarter of 2026. SYMBRAVO’s New Product Exclusivity expires on January 30, 2028, and after Apotex served a Paragraph IV certification notice in August 2025 Axsome commenced a patent infringement action against it in New Jersey, which was pending with no schedule set as of the first quarter filing.

02 What The Company Actually Is Today

The description that fits Axsome in August 2026 is a single-product commercial company with two smaller franchises attached and a broad clinical pipeline behind them. AUVELITY generated 82.6% of second quarter revenue. SUNOSI, acquired from Jazz Pharmaceuticals in two steps during 2022, generated 16.4%. SYMBRAVO, launched in June 2025, generated about one per cent. Any assessment of the company is in practice an assessment of one antidepressant that has just been given a second indication.

That concentration is not a criticism of the pipeline, which is unusually broad for a company this size, but it does define where the risk sits. A commercial disappointment in AUVELITY cannot be offset by the other two products at their current scale, and the pipeline assets that could eventually diversify the revenue base are years from market. The nearest of them, AXS-12, has a target action date of May 1, 2027, which means the earliest possible commercial contribution falls in the second half of 2027.

The three marketed medicines and where each one stands

03 The Commercial Engine: Prescriptions, Coverage And The Cost Of Selling

The measurable part of Axsome is the commercial operation, because unlike the pipeline it produces numbers every ninety days. Three of those numbers moved clearly in the second quarter of 2026. Total revenue grew 46% year over year. Prescription volumes rose across all three products. And selling, general and administrative expense grew 60% year over year, from $130.3 million to $208.1 million, which the company attributes to commercialisation activities for AUVELITY including the sales force expansion and launch activities for the Alzheimer’s disease agitation indication, and to commercialisation activities for SYMBRAVO.

The sequential picture is worth reading carefully, because the run of quarters is not a straight line. Net product revenue went from $121.5 million in the first quarter of 2025 to $150.0 million, then $171.0 million, then $196.0 million in the fourth quarter of 2025, then fell 2.4% to $191.2 million in the first quarter of 2026 before recovering to $218.4 million in the second. Full year 2025 net product revenue was $638.5 million. The first quarter dip is the only sequential decline in the series and it happened in the quarter immediately before the Alzheimer’s agitation approval, which makes the second quarter rebound easier to read as a return to trend than as a step change.

On the demand side the second quarter figures were as follows. Approximately 266,000 total prescriptions were written for AUVELITY, which the company describes as an increase of 34% year over year and 12% sequentially. The first quarter release stated approximately 223,000 prescriptions, and 266,000 against 223,000 would be an increase of 19.3% rather than 12%; the company does not publish the unrounded counts, so the two rounded figures and the stated percentage cannot be reconciled from the outside and the stated percentage is the company’s own. New-to-brand prescriptions, which the company describes as a leading indicator because they represent new patient starts, rose 26% sequentially overall, and rose 126% among patients aged 65 and older during the first eight weeks after the Alzheimer’s disease agitation launch compared with the same period in the prior quarter. Approximately 61,000 prescriptions were written for SUNOSI in the United States, up 14% year over year and 8% sequentially. Approximately 23,500 prescriptions were written for SYMBRAVO, which the company describes as a 30% sequential increase.

The over-65 new-to-brand figure is the single most informative disclosure in the release, because it isolates the population the new indication targets and measures it over a defined eight-week window. It is also, by construction, a percentage change from a small base in a segment where AUVELITY previously had little presence, so a large percentage does not by itself establish a large absolute number. The company did not disclose the underlying counts.

Access improved for AUVELITY and was broadly stable elsewhere. AUVELITY payer coverage moved from about 86% of all lives at the first quarter report to about 89% at the second, with commercial coverage rising from about 78% to about 82% and government coverage remaining at about 100%. SUNOSI coverage was about 82% against about 83% a quarter earlier. SYMBRAVO coverage held at about 57% after the addition of 17 million new commercial covered lives effective May 2026, which the company disclosed in the first quarter release.

What the company has said it will do, and by when

04 The Pipeline: What Is Filed, What Is Enrolling, And What Is Only Announced

Axsome runs one of the broader late-stage central nervous system pipelines among mid-capitalisation biotechnology companies, and the register at ClinicalTrials.gov allows most of it to be checked independently rather than taken from the company’s own summary. Thirty-five studies list Axsome Therapeutics, Inc. as sponsor, and a further five involving its compounds are sponsored by academic investigators. The distinction that matters is between programmes that are filed with the FDA, programmes that are enrolling patients under a registered protocol, and programmes for which the company has announced an intention but no trial has yet appeared.

Filed with the FDA

AXS-12 (reboxetine), cataplexy associated with narcolepsy. The only asset currently under FDA review. Axsome announced on July 15, 2026 that the agency accepted the New Drug Application for filing and set a target action date of May 1, 2027, adding that the FDA does not currently plan to hold an advisory committee meeting. AXS-12 is a selective norepinephrine reuptake inhibitor and cortical dopamine modulator with Orphan Drug designation for narcolepsy. The registrational evidence is a Phase 3 trial of 90 patients (NCT05059223, primary completion March 15, 2024) and a long-term study of 68 patients (NCT05113745, primary completion November 15, 2024), both completed and both small by the standards of a chronic-therapy filing, which is normal for an orphan indication but leaves less statistical room if the agency questions a subgroup.

Enrolling under a registered protocol

Solriamfetol, binge eating disorder (ENGAGE). NCT06413433, Phase 3, recruiting, 450 patients planned across 45 sites, started April 26, 2024, primary completion registered as December 2026. A long-term open-label extension is enrolling by invitation (NCT06878976, 300 patients). This is the nearest clinical readout: the company guides topline results to the fourth quarter of 2026.

Solriamfetol, excessive sleepiness in shift work disorder (SUSTAIN). NCT06568367, Phase 3, recruiting, 520 patients across 49 sites, started August 1, 2024, primary completion registered as December 2026, with topline guided to 2027.

Solriamfetol, major depressive disorder with excessive daytime sleepiness symptoms (CLARITY). NCT07484217, Phase 3 randomised withdrawal, recruiting, 508 patients across 40 sites, started February 23, 2026, primary completion registered as December 2028. This is a long programme, and the registered completion date is the clearest indication that it is not a 2026 or 2027 event.

Solriamfetol, attention deficit hyperactivity disorder in children and adolescents (FOCUS-2 and FOCUS-3). Two mirror-image Phase 3 trials of 468 patients each, six-week treatment, primary endpoint change from baseline to week 6 in the ADHD Rating Scale total score. FOCUS-3 in adolescents aged 12 to under 18 started June 25, 2026 (NCT07717892); FOCUS-2 in children aged 6 to under 12 started July 6, 2026 (NCT07717879). Both register primary completion as June 30, 2028. A Phase 3 trial in adults completed in February 2025 (NCT05972044, 516 patients).

AXS-14 (esreboxetine), fibromyalgia (FORWARD). NCT07398417, Phase 3 randomised withdrawal, recruiting, 620 patients across 51 sites, started January 14, 2026, primary completion registered as March 2028, with an open-label safety study enrolling by invitation (NCT07637162, 300 patients).

Announced but not yet started or not yet in a registered trial

AXS-05, smoking cessation. The company states it is on track to initiate a pivotal Phase 2/3 trial in the third quarter of 2026. No corresponding Axsome-sponsored trial record appears in the register as of August 10, 2026, which is consistent with a study that has not yet started.

AXS-17, epilepsy. A GABA_A receptor alpha-2,3 subtype-selective positive allosteric modulator. The company describes Phase 2 trial-enabling activities as underway. There is no registered Phase 2 trial.

AXS-20 (balipodect), schizophrenia and Tourette syndrome. Axsome announced on April 1, 2026 that it had entered an asset purchase agreement with Takeda for worldwide commercial, development and manufacturing rights to TAK-063 (balipodect), an oral selective phosphodiesterase 10A inhibitor. The press release did not carry the financial terms, but the first quarter Form 10-Q does: Axsome acquired the global rights for $10.4 million inclusive of transaction costs, recorded the entire amount as a research and development charge because the transaction was accounted for as an asset acquisition, and Takeda is eligible to receive up to $260.0 million in development, regulatory and sales-based milestones plus a mid single-digit royalty on potential global net sales. Balipodect has completed a 164-patient proof-of-concept Phase 2 trial in schizophrenia and has been given to more than 360 individuals across clinical studies. Axsome describes Phase 3 trial-enabling activities in schizophrenia as underway and says it plans to evaluate the compound in Tourette syndrome. No registered Axsome-sponsored trial exists for either indication yet. A $10.4 million entry price for a de-risked Phase 2 asset with a novel mechanism is a small number against a $268.0 million quarterly cost base, and the milestone ladder does not begin until the programme advances.

05 Financials

The figures below come from the second quarter 2026 results release filed as exhibit 99.1 to a Form 8-K on August 10, 2026, and from the equivalent release for the first quarter filed on May 4, 2026. The consolidated statements in the second quarter release are unaudited and cover the three and six months ended June 30, 2026. The corresponding Form 10-Q had not been filed as of the morning of August 10, 2026, which means the cash flow statement for the quarter is not yet public.

The arithmetic reconciles cleanly. Cost of revenue of $13,564K plus research and development of $46,227K plus selling, general and administrative of $208,137K plus intangible amortisation of $1,589K, less the $1,496K gain on contingent consideration, gives total operating expenses of $268,021K. Total revenue of $218,375K less operating expenses of $268,021K gives an operating loss of $49,646K. Net interest expense of $1,540K takes the pre-tax loss to $51,186K, and income tax expense of $128K takes the net loss to $51,314K. Dividing $51,314K by 51,799,708 weighted average shares gives $0.99 per share, which matches the reported figure.

The gap between the reported loss and the cash consumed is where this quarter differs from the one before it. The $51.3 million net loss includes $27.1 million of stock-based compensation, which is non-cash, and $1.6 million of intangible amortisation, also non-cash. Cash and cash equivalents nonetheless rose from $305,106K at March 31, 2026 to $319,850K at June 30, 2026, an increase of $14.7 million, having fallen $17.8 million in the first quarter. The balance sheet shows where the movement came from without the cash flow statement being available: accounts receivable rose $52,995K over the six months and inventories rose $9,945K, both uses of cash, while accounts payable rose $13,432K and accrued expenses and other current liabilities rose $66,779K, both sources. In other words, a substantial part of the second quarter cash improvement sits in payables and accruals rather than in operating profitability, and that is a timing effect rather than a structural one.

The year-over-year comparison of the loss is less flattering than the revenue comparison. Revenue grew $68.3 million year over year, from $150.0 million to $218.4 million. Selling, general and administrative expense grew $77.9 million over the same period, from $130.3 million to $208.1 million. The company therefore added more selling cost than revenue in the quarter, which is a normal pattern for a launch year and an unsustainable one for a company that describes itself as heading toward cash flow positivity. The prior-year quarter also contained a $10.4 million loss on debt extinguishment from the repayment of the Hercules loan, which flatters the year-over-year loss comparison: excluding that item, the 2025 second quarter loss would have been about $37.6 million against $51.3 million in 2026.

Research and development fell year over year, from $49.5 million to $46.2 million, which the company attributes to lower costs for AXS-05 and AXS-14. Over six months, however, research and development rose from $94.3 million to $98.9 million, and the first half figure includes a $10.4 million charge taken in the first quarter for the acquisition of global rights to balipodect from Takeda, which was accounted for as an asset acquisition and expensed in full. Stripping that one-off out, underlying first half research and development was roughly $88.5 million against $94.3 million a year earlier. The quarterly decline therefore reflects programme timing rather than a reduction in development ambition, and the four Phase 3 programmes started or running in 2026 argue for the line rising again.

06 Balance Sheet And Capital Structure

The debt, and what is still available to draw

Axsome entered the Blackstone Loan Agreement on May 8, 2025 with Blackstone Alternative Credit Advisors LP and Blackstone Life Sciences Advisors L.L.C., with Wilmington Trust as administrative and collateral agent. The facility provides for up to $570.0 million in total: a $120.0 million first lien senior secured term loan funded at closing; a further $180.0 million of senior secured term loans available at the company’s option, of which $90.0 million was available until May 31, 2026 and the remaining $90.0 million is available until May 31, 2027; a $70.0 million super senior revolving facility; and up to $200.0 million of incremental term loans subject to lender consent. The loans mature five years from closing, on May 8, 2030. The proceeds were used in part to repay the earlier Hercules Capital term loan in full on the same day, producing the $10.4 million loss on debt extinguishment recorded in the second quarter of 2025.

The revolving facility deserves a note. At March 31, 2026 the full $70.0 million was drawn, and the first quarter Form 10-Q states that the company repaid the entire outstanding balance in April 2026. At June 30, 2026 the balance sheet again shows $70,000K of short-term borrowings, which means the facility was redrawn during the second quarter. The financing section of that same Form 10-Q records $70.0 million of gross proceeds from the Blackstone revolving facilities and $70.0 million of repayments within the first quarter alone, so a full draw-and-repay cycle inside a single quarter is already the established pattern. That evidence points toward a working-capital line used around quarter ends rather than $70 million of permanent debt, and the company itself classifies the balance as short-term “based on the Company’s intent and ability to repay this amount in the next twelve months”. The second quarter Form 10-Q, which would confirm the pattern for the latest period, had not been filed as of the morning of August 10, 2026.

What is owed on SUNOSI, and to whom

The $80.3 million of contingent consideration is the fair value of royalties owed to Jazz Pharmaceuticals on U.S. net sales of SUNOSI, arising from the 2022 acquisition of the product. The 10-K describes the terms as a high single-digit royalty on U.S. net sales in the current indication and a mid single-digit royalty for future indications; the royalty base stated in the filing is United States net sales. Axsome also assumed Jazz’s commitments to SK Biopharmaceuticals, the originator, which retains rights in twelve Asian markets, and to Aerial Biopharma: single-digit tiered royalties on SUNOSI sales plus up to $162.5 million payable on revenue milestones and $1.0 million on development milestones. Ex-U.S. rights in Europe and the Middle East and North Africa were licensed to Atnahs Pharma UK Limited, trading as Pharmanovia, in February 2023, which is the source of the royalty revenue line.

Dilution history

The share count used here is the figure stated on the face of the company’s own balance sheet, 52,299,889 issued and outstanding at June 30, 2026, rather than a total reconstructed by adding issuances. The weighted average used for earnings per share, 51,799,708 for the quarter, is lower because it averages the count across the period. Axsome’s dilution over the first half of 2026 was 2.8%, which for a company at this stage is modest, and the largest single external funding since 2024 has been debt rather than equity. The company has nonetheless kept an at-the-market programme active and used it in both of the last two first quarters, at a declining rate: $19.7 million in the first quarter of 2025 against $6.3 million in the first quarter of 2026. The counterpart of the debt choice is a fixed interest cost, a maturity in May 2030 and a quarterly minimum liquidity test of $30.0 million, against cash of $319.9 million at June 30, 2026.

07 Management And Governance

Axsome is led by its founder. Herriot Tabuteau, M.D. has served as Chief Executive Officer, President and Chairman of the Board since the company’s founding in January 2012, and holds all three roles simultaneously. The other executive officers named in the proxy statement filed on April 24, 2026 are Nick Pizzie, Chief Financial Officer since May 2018; Mark Jacobson, Chief Operating Officer since March 2020, who joined the company in April 2014; Hunter Murdock, General Counsel since June 2022; and Ari Maizel, Chief Commercial Officer since October 2024. The proxy does not name a Chief Medical Officer. Mr. Pizzie is the only named executive officer party to an employment agreement, which provides six months of base salary as severance in the event of termination without cause within twelve months of a change in control.

The board has five members divided into three classes with staggered three-year terms: Dr. Tabuteau; Mark Coleman, M.D., a director since December 2014 who serves as Lead Director; Roger Jeffs, Ph.D., a director since December 2014; Mark Saad, a director since December 2014 and chair of the audit committee; and Susan Mahony, Ph.D., a director since October 2023 and chair of the compensation committee. Four of the five are independent and sit on all three standing committees. The combination of chief executive and chairman in one person, mitigated by a lead independent director, is a governance structure that some institutional investors vote against as a matter of policy, and a five-member board is small for a company with an $11 billion market capitalisation.

The related-party royalty

One arrangement deserves separate mention because it is unusual. Axsome pays a royalty of 3.0% of AUVELITY net sales to Antecip Bioventures II LLC, an entity owned by the chief executive officer. The first quarter 2026 Form 10-Q records $4.6 million of expense under this arrangement for the quarter, against $2.9 million in the first quarter of 2025. The arrangement is disclosed as a related-party transaction, is not new, and scales directly with the success of the company’s largest product: on $180.3 million of second quarter AUVELITY net sales, a 3.0% royalty implies roughly $5.4 million for the quarter, although the second quarter figure has not yet been separately disclosed. It is legitimate and disclosed, and it means that the chief executive’s economic interest in AUVELITY is not limited to his equity holding.

Ownership

Insider transactions in the six months to August 10, 2026

Thirty-one Forms 4 were filed between February 10 and August 10, 2026, and reading them by transaction code changes the picture that a raw sales total would give. There were no open-market purchases at all in the period: not a single transaction carries code P. Sales under code S totalled 273,980 shares for gross proceeds of approximately $59.0 million, which is 0.52% of shares outstanding. Acquisitions under code A, meaning equity awards rather than purchases, totalled 187,896 shares. Exercises under code M totalled 259,272 shares, at strike prices between $2.85 and $12.95 for the executives and $4.95 for the chief executive.

Every sale in the period was made under a Rule 10b5-1 plan, and the footnotes describe the mechanism. The chief executive’s four transactions, on June 9, July 1, August 5 and August 6, 2026, covering 149,000 shares for roughly $34.4 million, are each described in the filings as a “necessary exercise of stock options set to expire due to attainment of the 10-year expiration date of such options”, with the underlying options struck at $4.95 and sold the same day. The chief financial officer’s June 9 sale of 33,000 shares is described as an exercise of options held for over eight years ahead of their ten-year expiry. The chief operating officer’s February and May sales carry equivalent language. Mark Coleman, the lead director, sold 43,180 shares across five dates between February 26 and June 10, 2026, described as sales of shares underlying previously exercised options.

The distinction is material. Options that reach their ten-year expiry must be exercised or forfeited, and an executive who exercises and sells on the same day under a pre-arranged plan is converting expiring compensation, not expressing a view on valuation. What the record does show, and what a reader can weigh independently, is that no insider bought shares on the open market during a six-month period that included an FDA approval, a pipeline acquisition and a share price that ranged from roughly $161 to roughly $249 across the transaction dates.

08 Analysts And Market Data

A note on how these figures were produced, because the differences matter. The price, market capitalisation and all five performance percentages are calculated from the daily closing series to August 7, 2026, the last complete session, and from the share count on the face of the company’s balance sheet. They are not the vendor performance figures, which on the morning of August 10 were computed against a pre-market price near $219 and therefore did not correspond to the $212.74 reference close. Short interest, beta, relative strength, price to sales and the trailing twelve-month margins are Finviz Elite screener values as of August 10, 2026 and are marked as such. The debt ratios are calculated from the June 30, 2026 balance sheet: financial debt of $187,958K, being the $117,958K term loan plus the $70,000K revolver, against $83,304K of equity gives 2.26 times, while total liabilities of $678,143K against the same equity gives 8.14 times. The 4.03 total debt to equity figure carried by the screener corresponds to an earlier equity base and is not used here.

Where the stock sits relative to its own history

The one-year gain of 103.93% and the six-month gain of 16.54% sit alongside a three-month decline of 3.40% and a one-month decline of 14.82%, and the close on August 7, 2026 was 18.2% below the June 18, 2026 intraday high of $260.19. In other words the stock doubled over twelve months and has given back roughly a sixth of that since mid-June. A relative strength index of 43.77 is neutral. Short interest of 7.01% of float with 3.57 days to cover is meaningful without being extreme for a biotechnology name of this size. The beta of 0.71 is low for the sector and reflects that AXSM has traded on company-specific regulatory and commercial events rather than on the biotechnology index.

Analyst coverage

No analyst note could be verified to the standard applied here, which requires both the identity of the house and the exact date of publication from a primary or first-tier source. Several secondary aggregators publish target changes attributed to individual houses, but the dates and in some cases the direction of the change conflict between sources, and none of those services is treated as citable. What can be reported is the aggregate: the Finviz screener recorded an analyst recommendation score of 1.27 on a scale where 1 is the most positive rating, and a mean price target of $281.10, on August 10, 2026. That figure is a vendor aggregate compiled before the second quarter release was digested, not a consensus that has been checked note by note, and it should be read as such.

Two structural points are worth holding alongside any target. First, a mean target compiled from notes written before an earnings release stops being a consensus the moment the release lands, and the second quarter figures were published at 07:00 Eastern Time on August 10, 2026 with a conference call at 08:00. Second, the practical dispersion in a name like this comes from a single modelling assumption, the peak revenue attributed to AUVELITY in Alzheimer’s disease agitation, and that assumption cannot be tested from outside until several quarters of launch data exist.

09 Retail Sentiment

The block below is a snapshot of the Stocktwits stream, with its date. These are the opinions of retail traders and non-professional investors, not of analysts, and they serve as a measure of attention and of imbalance, not as research.

The recurring themes in the stream on the morning of the release were narrow and consistent. The dominant one was that the Alzheimer’s disease agitation launch is being under-appreciated because it only began in June, with several posts pointing to the 126% increase in new-to-brand prescriptions among patients aged 65 and older as the metric that matters. A second theme, raised by more numerate posters, concerned the gap between third-party prescription tracking and the company’s own numbers, with one contributor estimating that the tracking service captured roughly 88% of prescriptions and arguing that the undercount is consistent with prior quarters. A third theme was straightforwardly defensive: an automated post at 07:03 Eastern Time noting that reported earnings per share of $(0.99) missed a consensus of $(0.83) and revenue of $218.4 million missed $221.2 million, tagged bearish, drew immediate pushback arguing that estimates had not been adjusted for a launch that started mid-quarter. That single post accounts for the 2.7% bearish share in the snapshot above.

What does not appear anywhere in the stream is the balance sheet. There is no discussion of the $70.0 million revolving facility being drawn again at June 30 after having been repaid in April, no discussion of the $80.3 million of contingent consideration owed to Jazz on SUNOSI sales, no mention of the 3.0% AUVELITY royalty paid to an entity owned by the chief executive, and no mention that three customers account for roughly 91% of gross product sales. The stream is arguing about the demand curve. The items that would change the arithmetic on the way to profitability sit in the filings, and the second quarter Form 10-Q that would update most of them had not been filed when the snapshot was taken.

10 What Bulls See

The arguments below are the ones made by investors who are positive on the shares. They are presented as their case, not as a recommendation.

A second indication arrived with the strongest possible regulatory pedigree. AUVELITY was approved for agitation associated with dementia due to Alzheimer’s disease on April 30, 2026, having been developed with Breakthrough Therapy designation and reviewed under Priority Review. The approval rests on the ADVANCE-1 trial, which met its primary endpoint on the Cohen-Mansfield Agitation Inventory at week five, and the ACCORD-2 randomised withdrawal trial, in which patients who continued treatment experienced a statistically significantly longer time to relapse than those switched to placebo. Discontinuation due to adverse events in ADVANCE-1 was 1.3%, the same rate as placebo. The bull case holds that a first-in-class mechanism with a placebo-level tolerability profile in an elderly population is a genuinely differentiated asset.

The early launch metric is the right one and it is large. New-to-brand prescriptions among patients aged 65 and older rose 126% during the first eight weeks after the launch compared with the same period in the prior quarter. New-to-brand measures new patient starts rather than refills, and the over-65 cut isolates the target population. Overall AUVELITY new-to-brand prescriptions rose 26% sequentially. On this reading the second quarter understates the run rate by construction, since the indication was live for roughly one month of it.

Revenue is compounding across six quarters and the one dip has been recovered. Net product revenue ran $121.5 million, $150.0 million, $171.0 million and $196.0 million through 2025, dipped 2.4% to $191.2 million in the first quarter of 2026, then reached $218.4 million in the second, giving 46% year-over-year growth and full year 2025 revenue of $638.5 million. AUVELITY alone grew 51% year over year. Payer coverage for AUVELITY rose from about 86% to about 89% of all lives within the quarter, and commercial coverage from about 78% to about 82%, so access is still widening rather than saturating.

Cash rose in the quarter despite the reported loss. Cash and cash equivalents went from $305.1 million at March 31, 2026 to $319.9 million at June 30, 2026. The reported net loss of $51.3 million includes $27.1 million of non-cash stock-based compensation. Against a $30.0 million minimum liquidity covenant, a cash balance ten times that level and a debt maturity in May 2030 leaves no near-term refinancing pressure, and the company reiterated that current cash is sufficient to fund operations into cash flow positivity.

Dilution is modest and the reliance on the equity market is shrinking. The share count rose 2.8% over six months, from 50,882,766 to 52,299,889, mostly through option exercises and equity awards. Use of the at-the-market programme fell from $19.7 million of gross proceeds in the first quarter of 2025 to $6.3 million in the first quarter of 2026, and no underwritten offering has been announced. A biotechnology company generating $409.6 million of half-year revenue while raising six million dollars of equity in a quarter is an unusual profile, and the $90.0 million Blackstone tranche available until May 31, 2027 provides a non-dilutive option if it is needed.

The pipeline gives multiple independent shots without a single binary. AXS-12 has a PDUFA date of May 1, 2027 with no advisory committee currently planned. Solriamfetol is in Phase 3 across binge eating disorder, shift work disorder, major depressive disorder with excessive sleepiness, and both adolescent and paediatric ADHD, with the ADHD programmes each enrolling 468 patients. AXS-14 is in Phase 3 in fibromyalgia with 620 patients. Balipodect, acquired from Takeda in April 2026, adds a first-in-class PDE10A mechanism with a completed 164-patient Phase 2 in schizophrenia. No single readout carries the company.

The patent position on the lead asset is long. The 10-K describes more than 150 issued United States patents covering AXS-05 products with protection extending through 2043, and the Teva settlement concluded in February 2025 permits generic entry no earlier than March 31, 2039 with pediatric exclusivity or September 30, 2038 without. SUNOSI patent litigation was fully settled per the June 3, 2026 announcement, with generic entry no earlier than September 1, 2040 with pediatric exclusivity or March 1, 2040 without.

11 What Bears See

The arguments below are the ones made by investors who are negative on the shares, presented on the same terms.

The selling line has grown faster than revenue, in absolute dollars. Year over year, revenue increased $68.3 million while selling, general and administrative expense increased $77.9 million. At $208.1 million against $218.4 million of revenue, the selling line consumed 95.3% of every dollar collected. Over six months the pattern is the same: revenue of $409.6 million against selling expense of $393.1 million. A company can spend through a launch, but the bear case notes that the crossover point keeps moving and that the company has never published a revenue target or an operating expense ceiling against which to measure it.

Fourteen years in, the company has never reported an annual profit. The accumulated deficit stood at $1.42 billion at June 30, 2026 and grew $115.9 million in the first half of 2026 alone. Shareholders’ equity is $83.3 million against $678.1 million of liabilities, a ratio of 8.14 times, and financial debt alone of $188.0 million is 2.26 times equity. The statement that cash is sufficient to reach cash flow positivity carries no date and no defined metric, and has been made in substantially the same form in successive releases.

SYMBRAVO revenue fell while its prescriptions rose. Net product sales for the migraine product fell from $4.1 million in the first quarter of 2026 to $2.3 million in the second, while prescriptions rose approximately 30% sequentially to about 23,500. The release does not reconcile the two. For a product now more than a year past launch, with about 150 sales representatives assigned to it and payer coverage of only about 57% of lives, the bear reading is that gross-to-net economics are worse than the headline prescription growth implies.

Revenue depends on three customers and one product. The first quarter Form 10-Q discloses that the three largest customers represented approximately 39%, 28% and 24% of gross product sales, roughly 91% in total, and the full-year 2025 figures were similar. Accounts receivable stood at $277.5 million at June 30, 2026 against quarterly revenue of $218.4 million. Meanwhile AUVELITY is 82.6% of revenue, so a single commercial or reimbursement setback has no offset.

The balance sheet improvement in the quarter came largely from payables. Cash rose $14.7 million in the second quarter, but over the six months accrued expenses and other current liabilities rose $66.8 million and accounts payable rose $13.4 million, while receivables rose $53.0 million and inventories $9.9 million. Without the second quarter Form 10-Q, which had not been filed as of August 10, 2026, the cash flow statement cannot be examined, and the bear case treats a working-capital-driven cash increase as a timing effect rather than evidence of operating leverage.

The revolving facility was repaid and then drawn again. The first quarter Form 10-Q states that the entire $70.0 million outstanding under the revolver at March 31, 2026 was repaid in April 2026. At June 30, 2026 the balance sheet again shows $70.0 million of short-term borrowings. A company describing itself as approaching cash flow positivity while keeping a super senior revolver fully drawn at consecutive quarter ends invites the question of what the line is funding.

Insiders sold and no insider bought. Across the six months to August 10, 2026, insiders sold 273,980 shares for approximately $59.0 million and purchased none on the open market. The sales were made under Rule 10b5-1 plans and are largely the exercise of options approaching their ten-year expiry, which is the correct context, but the absence of any open-market purchase across a period containing an FDA approval and a pipeline acquisition is the bear’s observation.

The chief executive collects a royalty on the lead product. Axsome pays 3.0% of AUVELITY net sales to Antecip Bioventures II LLC, an entity owned by the chief executive officer, recorded at $4.6 million in the first quarter of 2026. The arrangement is disclosed and long-standing, and the bear case is not that it is improper but that it sits alongside a combined chief executive and chairman role, a five-member board, and no employment agreement for the chief executive.

12 Scenario Framework

The scenarios below organise what would have to happen for each path to unfold. They are not forecasts, they carry no probabilities and no price levels, and they are not recommendations.

The outcome between the two poles is defined by a single unresolved variable: how much of the $208.1 million selling line is permanent. Everything else in the model follows from that number, and it is not disclosed. The way to keep score is the dates the company has published: the third quarter release, whose date has not yet been announced; the ENGAGE topline guided to the fourth quarter of 2026; the AXS-05 smoking cessation start guided to the third quarter of 2026; and the May 1, 2027 target action date for AXS-12. Those are the checkpoints. The news cycle in between is not one.

13 Bottom Line

Two accurate and opposite descriptions of Axsome Therapeutics can be held at the same time. The first is a company with three approved central nervous system medicines, $409.6 million of half-year revenue growing 51% year over year, a 93.8% gross margin in the latest quarter, patent protection on its lead asset running toward 2043, dilution of 2.8% over six months, a broad late-stage pipeline and a newly approved second indication in a condition with no comparable treatment. The second is a company that has never earned an annual profit in fourteen years, carries a $1.42 billion accumulated deficit against $83.3 million of equity, spends 95.3 cents of selling and administrative expense for every dollar of revenue it collects, keeps a super senior revolver drawn at consecutive quarter ends, still sells shares at the market, and asks investors to accept an undated promise of cash flow positivity. Both descriptions come from the same filing.

What is verified. Second quarter 2026 total revenue of $218,375K, up 46% year over year and 14% sequentially, of which AUVELITY $180.3 million, SUNOSI $35.8 million and SYMBRAVO $2.3 million. Operating loss of $49,646K and net loss of $51,314K, or $(0.99) per share on 51,799,708 weighted average shares, including $27.1 million of stock-based compensation. Cash of $319,850K at June 30, 2026, up from $305,106K at March 31 and against $322,933K at December 31, 2025. Total liabilities of $678,143K including a $120.0 million Blackstone term loan maturing May 8, 2030 and $70.0 million drawn on the revolver. 52,299,889 shares outstanding, up 2.8% in six months. AUVELITY approved for Alzheimer’s disease agitation on April 30, 2026 and launched in June, with about 266,000 prescriptions in the quarter and payer coverage of about 89%. AXS-12 accepted for FDA filing on July 15, 2026 with a target action date of May 1, 2027.

What is not verified. The permanent level of selling, general and administrative expense, because the company does not separate launch cost from sales-force cost. The trajectory of the Alzheimer’s agitation launch, because the quarter contains roughly one month of it and the disclosed metric is a percentage change without an underlying count. The reason SYMBRAVO net revenue fell 44% sequentially while its prescriptions rose about 30%. The purpose and duration of the redrawn revolving facility, because the second quarter Form 10-Q had not been filed as of the morning of August 10, 2026 and the cash flow statement is therefore not public. The date the company associates with cash flow positivity. And the analyst consensus, since no individual note could be confirmed with both house and exact date.

The dates that settle the question: the third quarter 2026 results, date not yet announced, the equivalent report having been filed on November 3, 2025 and November 12, 2024, which will be the first full quarter of the Alzheimer’s disease agitation indication and the first clean read on whether the selling line flattens; the second quarter Form 10-Q, due within days of this release, which will publish the cash flow statement and explain the revolver; the fourth quarter of 2026 for ENGAGE topline results in binge eating disorder; the third quarter of 2026 for the start of the AXS-05 smoking cessation trial; and May 1, 2027, the FDA target action date for AXS-12 in cataplexy associated with narcolepsy.