The Beat That Wasn't: Arrowhead's Best Operating Quarter Arrives With Two Disclosures Management Left in the 10-Q

Results vs. Consensus

This was the first quarter since Arrowhead became a commercial company in which nothing episodic dominated the income statement. There was no Sarepta milestone recognition, no Novartis upfront landing, no single event that made the revenue line unmodellable. That is why the revenue estimate finally worked: consensus sat at $73.84M and the print was $73.737M, a difference of $0.1M. In the two prior quarters the revenue estimate missed the print by 23.4% and 46.3%, in both cases because providers guessed differently about when a discrete partner payment would be recognised.

The EPS line is a different story, and it is the story of the quarter. Consensus ranged from $(1.04) to $(1.18) depending on the provider, with the two most-cited figures at $(1.09) and $(1.18). Against $(0.93) reported, that reads as a beat of $0.16 to $0.25. It is neither.

FQ2 2026 Scorecard

Year-over-Year Comparison

The year-over-year comparison is close to meaningless on the top line and should be read as such. The March 2025 quarter carried more than $540M of revenue recognised on delivery of the four Sarepta C1 licences, a single accounting event that will not recur. The useful year-over-year lines are the expense lines, and they show a company that added $53.5M of quarterly operating cost in twelve months, $40.2M of it in R&D and $13.3M in the commercial organisation.

Sequential Comparison

The sequential table is where the quarter's real news sits, and it is in the expense column. Operating expenses rose in each of the three quarters before this one, from $161.5M in the March 2025 quarter to $223.2M in December. This one falls. Total operating expenses fell $8.2M, G&A fell $4.3M, and R&D fell $4.0M. The CFO characterised the line as "roughly flat with operating expenses in the first fiscal quarter", which understates it. A 3.7% sequential decline in a cost base that grew 33.1% year over year is the single most encouraging number in the release.

Revenue assessment

Revenue of $73.7M breaks into $37.4M from the Sarepta collaboration, $4.3M from the Sarepta clinical supply agreement, $20M from the Novartis upfront recognition schedule, $10.5M from the Sanofi arrangement covering Greater China, and roughly $1M of REDEMPLO product sales. The Sanofi figure is almost entirely the milestone triggered by the January approval in China and will not repeat, so the ongoing base is nearer $64M. That is a genuinely useful number and the first one Arrowhead has produced that a model can carry forward.

The Novartis schedule is the most durable component. Of the $200M upfront received in October, $54M has been recognised across the first six months of the fiscal year with $146M still deferred against preclinical obligations. That is a visible, non-lumpy tail of roughly $20M a quarter for the next several years, and it is the closest thing to recurring revenue this income statement contains. The Sarepta collaboration recognition of $28M plus $10M of cost reimbursement is the second-largest block, and it is the block the 10-Q says is now changing.

Cost assessment

R&D of $173.3M is still a record-adjacent number and remains 30.2% above the prior year, but the direction turned. The CFO said almost two-thirds of fiscal-year-to-date clinical trial spend went to the plozasiran Phase III programme, and that spend "should thereafter moderate accordingly" once the studies read out in the summer. That gives the R&D line a visible mechanical step-down, independent of any commercial or clinical outcome, beginning in the September quarter, FQ4 2026.

The G&A decline is the more surprising line. This was the item we flagged last quarter as the deterioration to watch: it had gone from $26.9M to $46.0M in four quarters with no stated ceiling and European commercialisation still to be added. It fell to $41.7M. Management gave no explanation for the sequential decline and none was requested. The prudent reading is that the fourth quarter of calendar 2025 contained non-recurring launch-preparation cost rather than that the run-rate has settled, because the company has since committed to direct commercialisation in Europe with contracted infrastructure and to a field organisation sized for an indication it has not yet been granted.

EPS assessment

Reported diluted EPS of $(0.93) on 142.4M weighted-average shares. The three components that separate the print from the operating result are worth naming individually, because two of them are non-cash and one of them is unusual. The $19.0M gain on the Visirna asset transfer added $0.13. The $6.7M equity-method loss on the resulting Bisirna stake removed $0.05. And the quarter's $23.8M of interest expense was entirely non-cash: across the six months, interest expense of $46.4M is accounted for exactly by $16.4M of accretion on the royalty-monetisation liability, $29.3M on the credit facility and $0.7M of amortised note issuance costs, while the supplemental cash-flow disclosure records interest paid of nil.

The share count deserves a note. Weighted-average diluted shares rose 5.9% year over year and outstanding shares rose 2.3% sequentially, but the reported figure understates the dilution already contracted for. At quarter end there were 140.6M shares outstanding, 2.5M shares under pre-funded warrants exercisable at $0.001, 8.0M shares issuable on conversion of the 2032 notes, and 18.2M shares reserved for options and restricted stock units, the last figure having nearly doubled from 9.9M six months earlier. Anyone modelling per-share outcomes on the current count is understating the denominator by a meaningful margin.

Revenue Composition

Arrowhead reports a single revenue line and no segments. The economically meaningful decomposition is by counterparty, because each stream has a different durability, a different recognition mechanic and a different risk. The table below reconciles to the reported total.

Component figures are as disclosed in the 10-Q and on the call and are individually rounded; the total is the income-statement figure.

Sarepta: half the revenue, and the terms just changed

Sarepta supplied 56.5% of revenue between the collaboration and the supply agreement. The CFO's decomposition on the call put the collaboration piece at $28M of ongoing recognition of the initial consideration, $10M of preclinical cost reimbursement and $4M of clinical supply. What he did not say, and what the 10-Q filed the same day does say, is that Sarepta exercised its contractual step-in right during the quarter on certain C1 programs, taking over responsibility for their ongoing clinical trials.

The accounting consequence is specific: the modification "reduces both the scope of the Company's remaining obligations and the amount of variable consideration related to reimbursable research and development costs for the affected C1 programs." No cumulative catch-up was recorded, because the revised reimbursement budgets are still subject to mutual agreement and the variable consideration is therefore constrained. It will be recognised in a future period, in an amount and a direction Arrowhead has not indicated.

Assessment: a partner taking over trial execution is not itself bad news, and it removes cost as well as revenue. But an unsized, undated adjustment to the largest revenue stream, disclosed only in the filing and never mentioned aloud on a call that ran twelve analyst questions, is exactly the sort of item that should have been in the prepared remarks. Model the $10M reimbursement component as at risk from the September quarter.

Novartis: the only stream that behaves like an annuity

Of the $200M upfront received in October 2025, $54M has been recognised in the first six months and $146M remains deferred, recognised on an input method tied to research-and-development cost incurred against the programme budget. Arrowhead is also eligible for up to $30M of research milestones and reimbursement of certain costs. Because recognition is cost-driven rather than event-driven, this is the one line that will not produce a surprise in either direction.

Assessment: roughly $20M a quarter of visibility for several years. It is not a business, but it is a floor, and it is the reason the underlying run-rate can be described at all.

Greater China: a milestone this quarter, and a structure worth watching

The $10.5M recognised under the Sanofi arrangement is almost entirely the $10.0M milestone triggered when China's National Medical Products Administration approved REDEMPLO in January. Visirna, the majority-owned subsidiary that holds the Greater China rights, sold those rights to Sanofi in August 2025 for $130.0M upfront plus up to $265.0M of approval milestones across indications in mainland China.

Three things happened to that vehicle inside this quarter and none of them was mentioned on the call. Option holders exercised 14 million Visirna options on January 2, cutting Arrowhead's ownership from 66.25% to 56.38%. Visirna paid $94.8M of a declared $100.0M dividend in March, of which $38.4M went to noncontrolling shareholders rather than to Arrowhead. And Visirna transferred in-process R&D assets to Bisirna for equity, producing the $19.0M gain that carried the EPS line.

Assessment: the China asset is generating real value and Arrowhead's share of it is shrinking. A 10-point ownership reduction and $38.4M of cash paid out to minorities in a single quarter, against a milestone stream that runs to $265M, is a structural leak that no one has quantified for investors. Watch whether the ownership percentage keeps falling.

Commercial Performance

REDEMPLO was approved on November 18, 2025 for familial chylomicronemia syndrome, an indication of roughly 6,500 people in the United States. This was the first full commercial quarter. The headline commercial metrics accelerated, and the one metric that would let an investor value the acceleration was again withheld.

Assessment: the launch is working and the arithmetic still does not close. At the $60,000 list price in effect through the March quarter, roughly $1M of net product revenue corresponds to no more than about 67 quarterly doses even at 100% of list, and fewer at any realistic net price. The 400-plus prescriptions and the roughly 180 shipped patients are both as-of-call figures from five weeks after the quarter ended, so they cannot be divided into the revenue. That is precisely the problem. Management is offering a numerator that grows every quarter and declining, for the second consecutive quarter, to supply the denominator that would turn it into a forecast.

Key Topics & Management Commentary

Overall Management Tone: Management was the most relaxed it has been in four quarters, and the relaxation showed in what was left out rather than in what was claimed. Prepared remarks were forward-facing and catalyst-led, with the CEO opening on "the strongest footing of our history" and closing on a list of second-half readouts, while the two disclosures with the most direct bearing on the model appeared only in the filing. Analyst pushback was narrow and almost entirely about the price cut and the pancreatitis endpoint; on both, management answered directly and without hedging, which is a change from the more defensive posture of the November call.

1. The 25% list-price cut

The headline commercial decision of the quarter was to take REDEMPLO's US wholesale acquisition cost from $60,000 to $45,000, a 25% reduction taken five and a half months after launch and while the only approved indication is a 6,500-patient orphan disease. The competing APOC3 inhibitor has announced $40,000, so the residual premium is 12.5%. Management's framing was uniform across prepared remarks and Q&A: this was offensive, not defensive.

"And let's just be clear, the lowering of the WAC from $60,000 to $45,000 had nothing to do with any pushback from the payers. To the contrary, I think that those interactions have been quite positive. This is a lowering of the WAC in expectation of an expectation of, of an expansion the market into SHTG and to ensure that we are -- that we would -- the payers would not require a step through with our competitor."
— Dr. Christopher Anzalone, President and CEO

The strategic logic is the One-REDEMPLO unified pricing model: a single price across familial chylomicronemia syndrome and severe hypertriglyceridemia, set now so that payer contracting for the far larger indication does not begin with a formulary position that forces patients through the competitor first. That is a real risk, and pricing against it before the sNDA is filed is a defensible sequencing choice.

Assessment: we accept the stated motive and still think the market read it correctly. Whatever the reason, revenue per patient on the only approved product fell 25% permanently, and the option it purchases is contingent on a readout that has not happened. Management is spending certain revenue to buy a probabilistic access advantage. That is a reasonable trade for a company with $1.78B of cash and a poor one for a company that needs the FCS indication to fund anything.

2. The cost base stopped rising

For three consecutive quarters the operating expense line went up, and in the last two reports we identified it as the deterioration that mattered more than any headline. This quarter it fell, from $223.2M to $215.0M, with G&A down 9.3% and R&D down 2.2%. Management did not present this as an inflection, which is either modesty or a signal that it does not consider the decline durable.

"Fiscal year-to-date, almost 2/3 of the clinical trial spend can be attributed to our plozasiran Phase III studies. As James already mentioned, the registration of SHTG studies for plozasiran should readout in the summer and clinical trial spend for these programs should thereafter moderate accordingly."
— Daniel Apel, Chief Financial Officer

Assessment: the R&D step-down is mechanical and dateable, which makes it the most reliable forward statement on the call. The G&A decline is not explained and should not be extrapolated: Arrowhead has since committed to commercialising directly in Europe with contracted infrastructure, and Andy Davis said the field organisation is being sized for both the approved population and the severe-hypertriglyceridemia opportunity. The honest expectation is R&D down and commercial spending up, with the net direction unclear until the September quarter.

3. The pancreatitis endpoint, and a change in how events are counted

The equity's value turns on whether the pooled analysis of SHASTA-3 and SHASTA-4 shows a statistically significant reduction in acute pancreatitis. Together the two studies enrolled more than 750 patients with a primary endpoint of triglyceride change and pancreatitis rates as a key secondary. MUIR-3, at more than 1,400 patients, supplies additional safety data. The blinded portion completes in mid-2026 with top-line data in the third calendar quarter and an sNDA before year end.

The most important disclosure of the call was not about the odds; it was about the definition. In March, Arrowhead presented two-year open-label extension data from the Phase IIb studies showing no adjudicated acute pancreatitis events in any patient receiving plozasiran. Asked directly whether that clean result and the pivotal analysis use the same definition, management said no.

"The answer is no. So the SHASTA-2 study, we used the strict Atlanta criteria to look at pancreatitis events, whereas in the SHASA-3, -4 pooled analysis, we're using this to modify the Atlanta criteria that has definite probable and possible pancreatitis."
— James Hamilton, Chief Medical Officer

Assessment: this cuts both ways and the direction is genuinely ambiguous. Modified Atlanta criteria capture definite, probable and possible events, which raises the event count in both arms and therefore raises statistical power, the binding constraint in a rare-event endpoint. It also means the widely circulated "zero events on plozasiran over two years" result cannot be read across to the pivotal readout. Any investor anchoring on the extension data is anchoring on a stricter and cleaner definition than the one that will determine the label.

4. Management will not characterise the blinded event rate

A recurring line of questioning tried to establish whether enough pancreatitis events have accrued for the pooled analysis to be adequately powered, and whether the company would consider extending the blind to accumulate more. Management declined the first and ruled out the second.

"So we're not giving any guidance on or sort of blow-by-blow details on the number of events we're seeing. Suffice it to say, we feel comfortable with the number of events we have seen and the big reveal will be in Q3. But no plans to extend the study or stay blinded for a longer period at this time."
— James Hamilton, Chief Medical Officer

Assessment: "we feel comfortable" plus a refusal to extend the blind is the most informative thing management can legally say, and the bull case leans on it heavily. It is still inference, not data. A company that expected the endpoint to fail would have every incentive to keep enrolling; a company that expects it to succeed would stop. The counter is that the decision not to extend also has a mundane explanation, which is that a delayed readout pushes the sNDA past the year-end target and forfeits a year of the SHTG opportunity.

5. Europe recognised the clinical-diagnosis pathway

The single most valuable regulatory event of the quarter was not an approval. The Committee for Medicinal Products for Human Use recommended EU marketing authorisation without requiring genetic confirmation, and Arrowhead expects an EMA decision in the June to July window. Alongside it, the company secured approvals from the Australian Therapeutic Goods Administration, China's National Medical Products Administration and Health Canada.

Why the absence of a genetic-confirmation requirement matters more than the approvals themselves: a meaningful share of real-world FCS patients are diagnosed clinically rather than genetically, and any policy that requires a genetic test as a prerequisite removes those patients from the accessible population. The same principle governs the far larger severe-hypertriglyceridemia opportunity, where genetic definition is not the organising diagnostic at all.

"So our interactions with payers to date have been consistent, have been positive, and we're seeing payer policies, and these are public payer policies that reflect the ability to diagnose FCS patients through multiple diagnosis pathways, in particular, through the clinical criteria that we saw in PALISADE."
— Andy Davis, SVP and Head of the Global Cardio Metabolic franchise

Assessment: this is the clean delivery on the commitment we flagged last quarter. Published payer policies recognising clinical criteria, plus a European regulator declining to require genetic confirmation, together validate the accessible-population assumption that the whole commercial model rests on. It is the item on this call we would weight most heavily in a valuation.

6. Sarepta's step-in right, disclosed only in the filing

During the quarter Sarepta exercised its contractual right to assume responsibility for the ongoing clinical trials of certain C1 programs. Under ASC 606 this is a contract modification that reduces both the scope of Arrowhead's remaining obligations and the variable consideration tied to reimbursable R&D costs. Arrowhead recorded no cumulative catch-up because the revised reimbursement budgets remain subject to mutual agreement, so the consideration is constrained until the uncertainty resolves and any adjustment lands in a future period.

The affected component is the $10M of preclinical cost reimbursement the CFO itemised on the call. The four delivered C1 licences are unaffected, so the $28M of ongoing initial-consideration recognition should continue. But the largest counterparty in the revenue base has changed the shape of its commitment, and the size and timing of the accounting consequence are unknown to investors.

Assessment: operationally this may be a positive, because Sarepta absorbing trial cost removes R&D as well as revenue. Presentationally it is a failure. A same-day 10-Q disclosure that materially qualifies half the revenue line belongs in prepared remarks, and its absence is the second time in two quarters that the most model-relevant fact of the period had to be found rather than heard.

7. Ionis's patent suit over plozasiran is live and was never mentioned

The 10-Q discloses that Arrowhead filed a declaratory-judgment action in Delaware on September 10, 2025 seeking a ruling that US Patent No. 9,593,333 is invalid and not infringed by its planned commercialisation of plozasiran. On December 23, 2025 the court granted Ionis's motion to dismiss that action. Ionis's own infringement complaint, filed in the Central District of California on September 11, 2025 and seeking damages, remains outstanding. Arrowhead states it disputes the allegations and intends to defend itself, and has recorded no contingent liability.

Assessment: the dismissal of Arrowhead's chosen forum is a procedural setback, not a merits ruling, and no accrual is required until a loss is probable and estimable. But the litigation targets the commercialisation of the product that is the entire near-term thesis, the venue is now the counterparty's, and the December ruling went against Arrowhead. Silence on the call and in the release, in a quarter otherwise devoted to the competitive positioning of that same product against that same competitor, is a conspicuous omission.

8. The balance sheet after the January financing

Cash and investments closed at $1,784.1M against $919.4M at fiscal year end, a $864.7M increase in six months. The January package was $700M of 0.00% convertible senior notes due January 2032 at an $87.07 conversion price, with capped calls purchased for $47.9M lifting effective anti-dilution protection to $119.33, plus a $230M equity offering priced at $64.50 that netted $216.6M. Note issuance costs were $18.7M. Combining the capped-call premium and the issuance costs gives roughly $66.6M of upfront economic cost on $700M over six years, or about 1.6% a year undiscounted, against a zero coupon.

"To provide a little color, in this quarter alone, we brought in over $1 billion, including approximately $850 million net from our January financing transactions, inclusive of a concurrent offering of 0% convertible senior notes and common stock, along with the associated capped call transaction."
— Daniel Apel, Chief Financial Officer

Assessment: these are the best financing terms in the company's history and they were executed into strength. Net of the $700M notes, the $199.6M credit facility and the $383.8M royalty-monetisation liability, the company carries roughly $500M of net cash against a quarterly operating loss of $141.3M, and the credit facility has been paid down by $286.1M cumulatively. Funding is not a live question through the readout and well beyond it.

9. Operating cash flow is positive, and it is not what it looks like

The cash-flow statement shows $97.9M of net cash provided by operating activities for the six months, against a $109.3M net loss. That inversion is worth decomposing, because it will not repeat. Working capital contributed $126.4M, of which the deferred-revenue build alone was $154.8M, chiefly the unrecognised balance of the Novartis upfront received in October. Before working capital, operations consumed $28.4M, itself flattered by $36.0M of stock compensation, $12.8M of depreciation and $45.7M of non-cash interest accretion.

Assessment: the company is converting an upfront payment received once into an operating cash inflow recognised over years, which is correct accounting and a poor guide to burn. As the Novartis balance amortises, the deferred-revenue tailwind reverses into revenue without cash. The genuine cash question is what happens when that $146M runs down against a cost base above $200M a quarter, and the answer is that the January financing exists precisely to cover it.

10. The pipeline is now the bigger half of the story

Three clinical readouts land in the second half. ARO-DIMER-PA, the first dual-functional siRNA designed to silence PCSK9 and APOC3 with one molecule, reads out in the third quarter from a Phase I/IIa in up to 78 adults with mixed hyperlipidemia, a population management sizes at roughly 20 million in the US. ARO-MAPT, the first candidate using the subcutaneous blood-brain-barrier delivery platform, reads out around the end of the third quarter or early in the fourth. ARO-INHBE and ARO-ALK7 provide obesity updates through the second half. The CEO volunteered one unprompted line about the tau programme.

"The early ARO-MAPT data are encouraging. We expect a substantial expansion of our CNS pipeline beginning at the end of 2026."
— Dr. Christopher Anzalone, President and CEO

Assessment: an unprompted forward-looking characterisation of blinded early data is unusual for this management team and is the closest thing on the call to a positive pre-announcement. It should be discounted accordingly, but it is not nothing. The dimer readout is the more consequential of the two for the platform argument, because it is the first clinical test of whether one molecule can silence two targets, and a positive result opens a mechanism rather than a single asset.

11. Partnering discipline: Madrigal takes an asset that cost nothing

Arrowhead licensed ARO-PNPLA3 to Madrigal for $25M upfront, up to $975M of milestones and tiered royalties up to the mid-teens. The asset arrived through the Janssen collaboration, was developed and taken through Phase I at Janssen's expense, and returned when Janssen exited NASH.

"We didn't spend any money on this. As we look at taking this forward, we think it's a really compelling target for a company that's focused in NASH largely. This is a genetically defined population. And that's sort of the good news and the bad news, right? The good news is it's quite specific. The challenge there is that there will be a companion diagnostic component to this."
— Dr. Christopher Anzalone, President and CEO

Assessment: this is the partnering model working exactly as advertised, and it is the pillar of the thesis that has never wobbled. The CEO also drew a useful boundary in Q&A: complement factors C3 and Factor B are the only other clinical assets he would consider out-licensing, and everything else is being retained. That is a clearer statement of portfolio intent than the company has previously given.

Guidance & Outlook

Arrowhead has never issued revenue or earnings guidance and did not start here. What the company does provide is a dated catalyst calendar, and this quarter it was unusually specific. The table below is the guidance substitute, and it is the schedule against which the next two quarters should be judged.

Implied revenue path: stripping the $10.0M China approval milestone leaves an ongoing base near $64M a quarter, comprising roughly $20M of Novartis amortisation, $28M of Sarepta initial-consideration recognition, $10M of Sarepta cost reimbursement now under revision, $4M of clinical supply and about $1M of product. Product revenue has to grow roughly twentyfold just to match the Novartis block, and on the disclosed prescription trajectory that is a fiscal 2027 event, not a fiscal 2026 one.

Guidance style: Arrowhead dates catalysts precisely and refuses to quantify anything commercial. That has been consistent across four quarters and there is no sign of it changing. The new datum this quarter is the SHTG approval target of second-half 2027 on a standard review timeline, which is the first time the company has put a year on the revenue inflection the equity is priced for.

Analyst Q&A Highlights

Whether the pancreatitis analysis will have enough events

The dominant topic in Q&A was the design and adequacy of the pooled pancreatitis analysis. The first question established the mechanics: SHASTA-3 and SHASTA-4 will be pooled and analysed for both the rate of events in individual patients and the total number of events. A later exchange pressed the question that actually matters, which is whether enough events have accrued, and offered management an out in the form of extending the blind. Management declined both the characterisation and the extension.

Q: "Just wondering if there's an updated perspective on whether the blinded AP event rates are tracking in line with your expectations and whether you'd be able to generate enough events by the time the study completes? Or could you potentially even keep the study blinded a little bit longer to get an adequate number of total events?"
— Maurice Raycroft, Jefferies

A: "So we're not giving any guidance on or sort of blow-by-blow details on the number of events we're seeing. Suffice it to say, we feel comfortable with the number of events we have seen and the big reveal will be in Q3. But no plans to extend the study or stay blinded for a longer period at this time."
— James Hamilton, Chief Medical Officer

Assessment: the refusal to extend is the informative half. A sponsor worried about power has an obvious remedy and did not take it. The offsetting explanation is that extending would push the sNDA past the year-end target and cost a year of the expansion opportunity, so the decision is not purely a statement about the data. Treat it as a positive signal with a competing motive attached.

Whether the pivotal events are counted the same way as the clean extension data

The most valuable question of the call was a throwaway clarification tacked onto a pricing question. In March the company presented two-year open-label extension data showing no adjudicated acute pancreatitis events in any plozasiran patient, a result that has been widely used to underwrite the pivotal readout. The question was whether the pivotal studies adjudicate events on the same basis. They do not.

Q: "And then just a point of clarification, are you characterizing the AP events in SHASTA-3, -4 the same way that you characterized them in the long-term of SHASTA-2 extension that was recently presented?"
— Joseph Thome, TD Cowen

A: "Why don't I take the second part first about the characterization. The answer is no. So the SHASTA-2 study, we used the strict Atlanta criteria to look at pancreatitis events, whereas in the SHASA-3, -4 pooled analysis, we're using this to modify the Atlanta criteria that has definite probable and possible pancreatitis."
— James Hamilton, Chief Medical Officer

Assessment: a broader definition raises event counts in both arms, which helps power but breaks the read-across from the extension data. The zero-event result over two years was generated under a stricter standard than the one that will decide the label. This is the single most important correction any investor should make to a bull model built on the March presentation, and it surfaced only because one analyst asked a clarifying question at the end of an unrelated one.

What actually drove the list-price cut

Two separate lines of questioning went at the price reduction, one asking whether payers had forced it and one asking why the company was not undercutting a first mover. Management's answer was consistent across both and unusually blunt: the cut was a forward-looking access decision for an indication that has not been approved, and payer conversations had been positive rather than adversarial.

Q: "Can you guys give more color on the interactions you've had with payers that led to the recent price lowering? And what has been the feedback since your competitive action last month?"
— Ron Feiner, JPMorgan

A: "This is a lowering of the WAC in expectation of an expectation of, of an expansion the market into SHTG and to ensure that we are -- that we would -- the payers would not require a step through with our competitor. As we talked about, we are priced at a premium here. We think that's appropriate given the characteristics of our drug versus the competitors' drug. And we think $45,000 actually long term is probably quite a good price in terms of maximizing access to the drug across various subpopulations within SHTG."
— Dr. Christopher Anzalone, President and CEO

Assessment: management answered the question that was asked, which is more than this section usually gets to say. The explanation is internally consistent and consistent with the payer commentary elsewhere on the call. It remains an expensive option premium paid before the option is known to be in the money, and the stock's reaction suggests the market priced the certain cost more heavily than the contingent benefit.

Gross-to-net after the cut

The natural follow-up was what the price reduction does to net realisation, since a list-price cut only matters to revenue in combination with the discount already being given. The question was passed from the CEO to the CFO, who declined to answer it as a matter of policy while volunteering that nothing beyond statutory rebates had been observed.

Q: "Could you guys how should we think about the impact here to gross to net following the price lowering?"
— Ron Feiner, JPMorgan

A: "I'm not actually going to give guidance on gross to net. We have not seen anything substantial in that regard nor do we currently expect that other than things that are statutory required such as Medicaid rebates and the manufacture discount program and the like. So -- but we have a policy not to provide any sort of guidance or -- on that at this stage."
— Daniel Apel, Chief Financial Officer

Assessment: this is the second consecutive quarter in which the gap between prescriptions written and revenue recognised has been raised and left unquantified. The substantive content is that gross-to-net erosion is described as statutory-only, which if accurate means the $45,000 list translates to a high net realisation. Until a conversion figure or a net-price assumption is provided, that has to be taken on trust.

Holding a premium against a cheaper first mover

With the competitor at $40,000 and a first-mover position, the question was whether a second entrant should be undercutting rather than charging 12.5% more. Management's defence was entirely product-attribute-based and made no appeal to access economics.

Q: "Given Ionis has announced a new price at $40,000 and have first-mover advantage, can you walk us through the rationale to continue to price for REDEMPLO at a premium"
— Shelby Hill, RBC Capital Markets

A: "We saw that from PALISADE with an 80% reduction from baseline. We saw that with the numerical decrease in acute pancreatitis from PALISADE. We've seen that also with an FDA label that has no contraindications, no warnings and no precautions. And then lastly, with a convenient dosing schedule that's only 4 injections a year. When you sum up the totality of those attributes, we just believe it's a best-in-class APOC3 inhibitor and as a consequence, should be premium priced."
— Andy Davis, SVP and Head of the Global Cardio Metabolic franchise

Assessment: the attribute list is genuine and the label really is the cleanest in the class. But the answer concedes the strategic point by omission: the premium is being defended on product quality, not on evidence that payers will pay it in a population a hundred and fifty times larger than the one where it has been tested commercially. The premium survives only if the pivotal readout supplies an outcomes claim the competitor does not have.

The obesity programmes after a competitor's setback

Sentiment on the INHBE mechanism had cooled following another company's data update, and the question invited management to say what would be needed to advance either asset for obesity as against redirecting them to liver disease. The answer restated the combination thesis rather than revisiting it.

Q: "Think investors have soured a little bit on INHBE after the WAVE data update. So just wanted to get your perspective on kind of what you need to see from these upcoming readouts to advance one or both for obesity treatment versus any alternative potentially considering for a NASH indication."
— Jason Gerberry, Bank of America

A: "We've been saying all along really that we thought that this INHBE ALK 7 access is interesting, but we thought the approach was to combine with GLP-1s. And that is still our standpoint here. We think that there's potential, particularly in the type 2 diabetics for additional weight loss on top of tirzepatide or other GLP-1s with either ALK 7 knockdown or INHBE knockdown."
— James Hamilton, Chief Medical Officer

Assessment: consistency here is worth something, because the company set the combination expectation before the competitor's monotherapy disappointment rather than after it. The forward emphasis shifted noticeably toward fat redistribution and body composition rather than weight loss alone, which is a narrower and more defensible claim. Neither asset is being advanced to Phase II, and that is the fact that should govern how much of this is in a model.

Why an asset with clinical proof of concept was licensed out

The Madrigal transaction invited the obvious capital-allocation question: if the Phase I data were compelling, why hand the programme to someone else. The answer was a clear statement of where this company believes its comparative advantage sits.

Q: "Help us understand from a capital allocation strategy perspective, why it makes sense for you to do this deal at this time as opposed to taking the drug further on your own?"
— Keay Nakae, Chardan Capital Markets

A: "It didn't really make sense for us to spend much money right now to do a Phase II. Those studies could be a bit long and more expensive than made sense for us. We've got a very large pipeline. We've got some really interesting programs that we are pushing ourselves. And I just think that our ROI is probably better by allocating capital to those programs that we are more confident that we will hold on to long term."
— Dr. Christopher Anzalone, President and CEO

Assessment: the right answer, and one that is easier to give from a $1.78B cash position than from a constrained one. The accompanying disclosure that complement factors C3 and Factor B are the only other clinical assets under consideration for partnering is more useful than the deal itself, because it tells investors which of the twenty-plus clinical programmes are considered core.

The dimer follow-on question that got a one-word answer

An analyst asked whether the next generation of the dual-silencing platform might pair one of the obesity targets with a non-overlapping mechanism, which is a question about whether the platform's most valuable application is a combination product. The exchange is the shortest of the call and the only one where management visibly declined to engage.

Q: "As you think about the next generation of the dimer platform, is a construct combining INHBE or ALK 7 with a nonoverlapping mechanism target, something you're evaluating for obesity or other indications?"
— Amine Chaherli, B. Riley Securities

A: "Yes."
— Dr. Christopher Anzalone, President and CEO

Assessment: a one-word affirmative followed immediately by a request to move to the next question. The most benign reading is that the answer is genuinely yes and the details are not ready; the less benign is that a specific programme exists and management did not want to be drawn on it before the dimer's first clinical readout. Either way it confirms that combination dimers targeting the obesity pathway are under evaluation, which is more than the pipeline disclosure says.

What They're NOT Saying

1. The $19.0M gain that produced the EPS beat. A related-party transfer of in-process R&D from the Greater China subsidiary to an entity Arrowhead owns 25.29% of, worth $0.13 a share, absent from the press release and from every minute of the call. Without it the quarter is in line, not ahead.
2. Sarepta's step-in right. Exercised during the quarter, reducing reimbursable R&D revenue on the largest revenue stream, with an unsized cumulative catch-up deferred to an unspecified future period. Disclosed in the 10-Q filed the same day and never spoken aloud.
3. The Ionis patent litigation. Arrowhead's declaratory-judgment action was dismissed in December; the counterparty's infringement suit over plozasiran commercialisation is live and seeks damages. Not mentioned in a quarter otherwise spent discussing competitive positioning against that same counterparty.
4. Prescription-to-paid-claim conversion. Second consecutive quarter without a figure. A shipment count of roughly 180 patients was offered for the first time, but as an as-of-call number that cannot be divided into a quarter-end revenue figure.
5. Gross-to-net. Explicitly declined as a matter of policy. The company will not say what proportion of a $45,000 list price it actually collects, in the quarter it cut that list price by 25%.
6. The Visirna ownership decline. Arrowhead's stake fell from 66.25% to 56.38% on January 2 when option holders exercised 14 million options. A ten-point reduction in a subsidiary entitled to up to $265M of milestones passed without comment.
7. The $38.4M dividended to minority shareholders. Visirna paid $94.8M of a $100.0M declared dividend in March, of which $38.4M left the group. Real cash, disclosed only in the filing.
8. Where G&A settles. The line fell 9.3% sequentially and no explanation was given for the decline, no ceiling was offered, and direct European commercialisation plus a field force sized for an unapproved indication are still to be layered on.
9. What happens to SHASTA-5 and at what cost. Management will "reassess whether there is added value in continuing" the dedicated pancreatitis study if the pooled analysis succeeds. The cost of running it, and the cost of stopping it, were not discussed.
10. Peak-sales or penetration framing for severe hypertriglyceridemia. The population was sized at more than one million high-risk US patients and the price at $45,000. No penetration assumption, no launch-curve analogue, and no revenue framing were offered for the opportunity the entire equity is priced on.
11. Internal expectations for the ARO-MAPT readout. Asked directly alongside a question about a competitor's tau programme, management discussed the competitor and the breadth of tauopathies and never returned to its own expectations, despite the CEO having volunteered in prepared remarks that the early data are encouraging.

Market Reaction

• Pre-print setup: the shares closed at $77.95 on May 7, up 496.9% over trailing twelve months from $13.06, up 27.4% over trailing thirty days from $61.18, and up 17.4% year to date from $66.39. The 52-week closing range entering the print was $13.03 to $79.14, so the stock went into the report 1.5% below its highest close of the past year. The S&P 500 was up 7.2% year to date.
• Reaction session: the shares opened at $76.50 on May 8, a 1.9% downward gap, traded a range of $72.61 to $77.92, and closed at $72.69, down 6.7% or $5.26. The S&P 500 rose 0.8% on the same session, so the relative move was 7.5 points.
• Volume: 2.7 million shares against a 1.8 million thirty-day average, 1.5 times normal.
• Attribution: the two facts a reader of the release would have registered first are the 25% list-price cut and a headline swing from $0.22 of earnings to a $0.93 loss. The accelerating prescription metrics sat below both.

The most useful thing about this session is the volume. A 6.7% decline on 1.5 times average turnover is a repricing by holders who stayed, not a liquidation by holders who left. Compare it with the February session, when the stock closed unchanged on 10.3 times normal volume as the January offering settled: that was an ownership transfer with no change of view, and this is a change of view with no change of ownership. The people who own this stock marked it down and kept it.

What they marked it down for is the price cut, and the reaction is better understood as a valuation adjustment than a verdict on the quarter. A 25% reduction in the list price of the only approved product mechanically reduces every revenue-per-patient assumption in every model, and it does so immediately, while the compensating benefit is contingent on a readout one quarter away. A holder who does not want to underwrite the pancreatitis endpoint has no reason to accept the trade, and the 7.5-point underperformance against a rising market is roughly what a 25% price reduction applied to the near-term revenue line and not the terminal one should produce.

The setup amplified it. This equity entered the print having risen almost fivefold in twelve months and 27.4% in the previous thirty days, one and a half percent below its 52-week closing high. At that level the print had to be unambiguously good to hold the price, and it was not unambiguous. It was a good operating quarter reported alongside a price cut, with the strongest line on the income statement supplied by an accounting gain that nobody explained.

Street Perspective

Debate: Was the price cut offensive or defensive?

Bull view: management stated flatly and repeatedly that payers never pushed back, that interactions have been positive, and that the reduction is a forward-looking move to prevent a step-through requirement in the far larger severe-hypertriglyceridemia indication. A company with $1.78B of cash and no funding need does not cut price under duress. Setting a single unified price before the expansion sNDA is filed removes a negotiation risk at the exact moment it can still be removed cheaply, and the residual 12.5% premium to the competitor preserves the best-in-class positioning.

Bear view: the sequence is a second entrant cutting price 25% within six months of launch, shortly after the first mover repriced to $40,000, and describing it as strategy. Whatever the motive, revenue per patient on the approved product fell by a quarter, permanently, and the company simultaneously declined to say what proportion of list it actually collects. A pricing action taken in expectation of an indication that has not been approved is an option premium paid before anyone knows the option is in the money.

Our take: the bulls have the better argument on motive and the bears have the better argument on consequence, and both can be true. We accept the explanation as given, because it is coherent, was volunteered rather than extracted, and is consistent with the payer commentary elsewhere on the call. But accepting the motive does not change the arithmetic. The cut is a certain cost incurred to buy a contingent benefit, and until the third-quarter readout there is no way to know whether it was cheap or expensive. That is a reason to wait, not a reason to sell.

Debate: Does the pancreatitis endpoint hit?

Bull view: the accumulating circumstantial case is stronger than at any prior point. Two years of open-label extension produced no adjudicated pancreatitis events in any plozasiran patient. The pivotal studies were enriched at enrolment toward patients with very high triglycerides and prior pancreatitis history. Management says it is comfortable with the number of events accrued, has declined to extend the blind, and has made continuation of the dedicated pancreatitis study contingent on the pooled analysis succeeding. Companies do not sequence that way when they expect to fail.

Bear view: every one of those signals is inference and none is data. The extension result that anchors the bull case was adjudicated under strict criteria while the pivotal pool uses a broader definition, so the read-across is invalid. Acute pancreatitis is a rare event and the studies were powered on triglyceride change, not on it. The decision not to extend the blind has an obvious commercial explanation that has nothing to do with confidence, which is that delay would forfeit a year of the expansion opportunity.

Our take: the bear framing of the evidence is correct and the bull framing of the incentives is correct, which is why we are not positioned for either. The criteria change is the fact that moved us this quarter, and it moved us toward humility rather than toward pessimism: a broader definition raises event counts in both arms and therefore raises power, so it may well help. What it forecloses is the clean extrapolation from the March presentation, which is what a great deal of the sell-side enthusiasm has been built on. Anyone whose model assumes the pivotal readout inherits the extension's zero-event profile should rebuild it.

Debate: Is the launch readable yet?

Bull view: the headline commercial metrics accelerated. Prescriptions quadrupled in a quarter to more than 400, grew nearly threefold within the period, and added more than 40% in the final four weeks alone. Roughly 180 patients have taken delivery. Around 85% are new to the class, which is the hard part in a poorly-identified rare disease, and more than 10% are switches from a competitor that has been selling for longer. Payer policies are publishing with clinical diagnosis as a valid pathway and Europe's regulator declined to require genetic confirmation.

Bear view: product revenue is approximately $1M against a cost base of $215M a quarter, and the conversion rate from prescription to paid claim has been withheld for four consecutive quarters. The company will not give a gross-to-net assumption in the quarter it cut list price by 25%. Every metric offered is a numerator; the denominator that would turn any of them into a revenue forecast is the one thing management has consistently declined to supply.

Our take: unchanged from last quarter in structure, improved in degree. The mix data remains more informative than the count, and a class-naive share near 85% means the diagnostic funnel is working, which is the part of a rare-disease launch that is hard to fix if it breaks. The new fact this quarter is a shipment figure, which is the first denominator management has ever offered, even if it arrives as an as-of-call number that cannot be reconciled to quarter-end revenue. The launch will not be readable until the September quarter, and by then the readout will have made the question secondary.

Model Update Needed

Valuation: at the $72.69 reaction close, 140.6M shares outstanding plus 2.5M pre-funded warrants gives an economic market capitalisation of roughly $10.40B. Adding $700M of convertible principal, $199.6M of credit facility and the $383.8M royalty-monetisation liability, less $1,784.1M of cash and investments, gives an enterprise value near $9.90B. Against a base-case peak revenue assumption of $2.0B that is 4.9 times peak sales, which is approximately where a de-risked commercial-stage single-product biotech trades. In other words the market is currently paying for the base case and nothing beyond it: neither the bull outcome in which a pancreatitis label claim supports a premium price across a million-patient population, nor the bear outcome in which a triglyceride-only label meets a cheaper first mover.

Fair-value range: $70 to $100, probability-weighted approximately $79, raised from $50 to $68. The increase reflects the balance-sheet transformation, the first sequential decline in operating expenses in four quarters, the European regulator's acceptance of clinical diagnosis, and the commercial acceleration. Scenarios: bull at $110 (30%), in which the pooled analysis is statistically significant, the label carries a pancreatitis-risk claim, the $45,000 price holds and peak revenue approaches $3.5B; base at $77 (45%), in which triglyceride reduction is decisive, the pancreatitis secondary is directionally positive without changing the label, the sNDA files on time for a second-half 2027 approval and peak revenue is near $2.0B; bear at $47 (25%), in which there is no pancreatitis claim, price competition compresses the premium and an education-paced ramp caps peak revenue near $1.2B.

Thesis Scorecard Post-Earnings

Overall: the thesis is stronger on execution and weaker on disclosure. Three of the four bear points that concerned us most in February improved materially, and two of them, the cost base and the milestone-dependence of revenue, improved for reasons that look structural rather than seasonal. Against that, the quarter introduced two new bear points that did not exist before, both of them about what management chose not to say rather than about what the business did. The net is a thesis that has advanced in substance and lost ground in trust, which is an unusual combination and one that argues for patience rather than for conviction in either direction.

Action: hold. The equity has done most of the work a good quarter could justify and the decisive information arrives in roughly ninety days. At $72.69 against a probability-weighted $79, the risk-adjusted upside is approximately 9%, which does not compensate for a binary readout in a stock that has already risen almost fivefold in twelve months. The case for adding is a view on the pancreatitis endpoint, and this quarter gave investors less basis for that view than the March data presentation appeared to, not more. We would revisit on a meaningful de-rating ahead of the readout, or immediately after it in either direction.