Kiniksa Pharmaceuticals (Nasdaq: $KNSA): Guidance Raised Twice In 2026, No Debt, And The Phase 3 That Has To Replace ARCALYST

01 July 28, 2026 update: the quarter that raised guidance for the second time

The second quarter of 2026, the three months to June 30, is the most informative document Kiniksa has published this year, because it settles three separate questions at once: how fast ARCALYST is still growing, how much of that growth reaches the bottom line after the Regeneron split, and what the company intends to do with a balance sheet that is now generating rather than consuming cash.

ARCALYST net product revenue came in at $243.6 million, against $156.8 million in the same quarter of 2025. There was no licence or collaboration revenue in the period, so total revenue and product revenue are the same number, which is itself a fact worth holding on to. Research and development expense more than doubled to $40.9 million from $18.8 million, reflecting the pivotal work on KPL-387. Selling, general and administrative expense rose to $63.9 million from $46.9 million. Cost of goods sold was $23.6 million and collaboration expense — the profit share owed to Regeneron — was $88.1 million, the single largest line in the operating cost stack. Total operating expenses of $216.4 million left income from operations of $27.2 million, and after other income and tax the company reported net income of $25.4 million, or $0.33 basic and $0.30 diluted per share.

Guidance moved for the second time in five months. Kiniksa opened the year on February 24, 2026 with a range of $900–920 million, lifted it to $930–945 million on April 28 alongside first-quarter revenue of $214.3 million, and lifted it again on July 28 to $980–995 million. Taken at the midpoint, that implies full-year growth of roughly 45 per cent on the $677.6 million reported for 2025, which itself grew 62 per cent on 2024. The company also repeated its qualitative commitment to remain, in its words, cash flow positive on an annual basis. It does not publish guidance for total revenue, operating expenses or earnings, only for ARCALYST net product revenue, and that distinction matters when reading third-party estimates.

The balance sheet closed the quarter with $525.9 million in cash, cash equivalents and short-term investments, against $414.1 million at December 31, 2025, and with no debt. Working capital was $497.4 million, total assets $896.1 million and shareholders’ equity $654.1 million, with an accumulated deficit still standing at $414.1 million — a reminder that the profitable phase of this company’s life is recent.

02 The KPL-387 Phase 2 interval data, and what it does not settle

The same July 28 release carried the first efficacy numbers on KPL-387, a fully human IgG2 monoclonal antibody that binds interleukin-1 receptor 1 and therefore blocks signalling from both IL-1α and IL-1β. The target profile is a once-monthly subcutaneous injection in a liquid formulation, against the weekly injection required by ARCALYST. That is the commercial thesis in one sentence: same pathway, twelve injections a year instead of fifty-two, and an asset Kiniksa owns without a profit split.

The data released was an interval analysis of the dose-focusing portion of the Phase 2/3 study, for the arm receiving 300 mg subcutaneously once a month — the dose the company has carried into the pivotal phase. All three figures are medians with 95 per cent confidence intervals:

The company described the effect as rapid and sustained, said efficacy was durable across the monthly dosing interval, and characterised tolerability as consistent with the known safety profile of interleukin-1 pathway inhibition. Chief executive Sanj K. Patel put commercialisation in the 2028/2029 timeframe.

PASTORALE (NCT07010159) is registered as a Phase 2/3 study with 325 participants across 63 sites in the United States, Canada and Europe, started on July 25, 2025, with a registry primary completion date of December 31, 2027. The design puts every participant on KPL-387 during a single-blind run-in while conventional oral therapies are tapered and withdrawn; clinical responders are then randomised one-to-one to KPL-387 300 mg monthly or placebo in a double-blind, event-driven withdrawal period; the study enrols up to roughly 85 participants in that portion, and only run-in responders are randomised. A second study, NCT07288216, began on March 25, 2026 with 80 participants and tests transition to KPL-387 monotherapy over sixteen weeks in patients whose disease is already well controlled, including those already on anakinra or rilonacept — in other words, on ARCALYST itself.

03 Executive summary

Kiniksa Pharmaceuticals International is a commercial-stage biopharmaceutical company built around one approved product, ARCALYST, in one principal indication, recurrent pericarditis. That product generated $677.6 million of net revenue in 2025 and is guided to $980–995 million in 2026. The company is profitable, generates cash, holds $525.9 million of it, and carries no debt. Those four facts put it in a different category from most of the names covered on this site.

The structure underneath the numbers is more complicated. ARCALYST was discovered by Regeneron and is licensed, not owned: Kiniksa pays a 50/50 profit share, which appears as collaboration expense and consumed $229.5 million in 2025 and $163.6 million in the first half of 2026 alone. The composition-of-matter patent expired in 2020. What protects the franchise commercially is seven years of United States orphan exclusivity granted at the March 2021 approval — running out around March 2028 — plus five method-of-use patents with a statutory term into 2038.

That is why KPL-387 matters more than any quarterly beat. It is Kiniksa’s own molecule, in the same pathway, aimed at monthly rather than weekly dosing, with an orphan designation obtained in October 2025 and a pivotal trial enrolling. If it reads out and converts patients, the franchise continues on better economics. If it does not, the company is left defending a licensed product past its exclusivity window with method-of-use patents and switching costs.

04 Why $KNSA matters now

Three things converged in 2026 to make this a name worth a dedicated hub rather than a line in a screener.

The commercial ramp has not slowed at scale. Growing 62 per cent off a $417 million base, as ARCALYST did in 2025, is one thing. Growing roughly 55 per cent year on year off a $677 million base, as it did in the second quarter of 2026, is another. Rare-disease launches usually decelerate as the prevalent pool is worked through; this one is still adding prescribers, and management’s own penetration arithmetic says most of the addressable pool is untreated.

The profit and loss account has crossed over. Kiniksa moved from a $43.2 million net loss in 2024 to $59.0 million of net income in 2025, and reported $48.0 million in the first half of 2026 alone. Cash rose $111.9 million in six months without an equity raise. For a company with a $400 million shelf registration sitting unused since June 2024, that changes what the shelf means: it is optionality rather than necessity.

The clock is visible. Unlike most single-product stories, the point at which the protection weakens is knowable in advance. Orphan exclusivity from the March 18, 2021 approval runs seven years. That is the frame within which every pipeline decision, every trial timeline and every capital allocation choice at this company should be read.

05 Company overview

Kiniksa Pharmaceuticals International, plc is a public limited company incorporated in England and Wales in April 2024, company number 15630565, with its registered office at 105 Piccadilly, Second Floor, London. It is the holding company of a group that conducts its business through subsidiaries with principal operations in the United States, the United Kingdom and Switzerland; the group’s executive offices are at 100 Hayden Avenue, Lexington, Massachusetts. The plc is the successor to Kiniksa Pharmaceuticals, Ltd., the Bermuda entity that carried the Nasdaq listing from the May 2018 initial public offering. The annual general meeting was held in London on May 19, 2026, with results filed on May 29.

The therapeutic focus is narrow and deliberate: debilitating diseases with significant unmet need, concentrated in cardiovascular and autoimmune indications where the interleukin-1 pathway is central. The commercial organisation is built around a single specialty product sold in the United States, which is why the selling, general and administrative line — $196.3 million in 2025, $125.0 million in the first half of 2026 — is large relative to revenue for a company of this size: it is a full field force covering cardiology.

Who runs it

The board is chaired by the chief executive, with Felix J. Baker as lead independent director; Patel is the only non-independent director. Two further directors, Stephen R. Biggar and M. Cantey Boyd, are also from Baker Bros. Advisors, which is the largest shareholder on an as-converted basis. The proxy is explicit that the combined chair and chief executive role was a deliberate board decision.

One executive change is on record: Eben Tessari resigned as chief strategy officer effective May 15, 2026, moving to a consulting arrangement at $450 an hour for up to twenty hours a month through May 2027, and to the science and research committee as an adviser. The 8-K states there was no disagreement with the company.

06 ARCALYST: the product, the indication, and the Regeneron economics

ARCALYST (rilonacept) is a recombinant dimeric fusion protein that acts as a cytokine trap, binding and neutralising both interleukin-1α and interleukin-1β. It is given as a weekly subcutaneous injection. It was discovered by Regeneron and carries biologics licence application 125249, now held by Kiniksa Pharmaceuticals (UK), Ltd.

Approval history

The recurrent pericarditis approval was supported by RHAPSODY (NCT03737110), the Phase 3 trial completed in 2020, and carried a breakthrough therapy designation granted in 2019. It brought seven years of United States marketing exclusivity, in the company’s own words in the 2025 annual report. The European Commission granted an orphan designation for idiopathic pericarditis in 2021.

The Regeneron agreement, and why it dominates the cost base

In September 2017 Regeneron granted Kiniksa an exclusive licence to develop and commercialise ARCALYST worldwide except the Middle East and North Africa, for all indications except oncology and local administration to the eye or ear. After the RHAPSODY results, Regeneron transferred the biologics licence application to Kiniksa. Total payments from Kiniksa to Regeneron have been $32.5 million in upfront and regulatory milestones, including a $20.0 million milestone on the pericarditis approval in the first quarter of 2021.

Termination terms are conventional but asymmetric in one respect: Kiniksa may terminate for convenience on one year’s notice, or on three months if it concludes ARCALYST is not safe; Regeneron may terminate if Kiniksa conducts no material activities for twelve consecutive months, or on a patent challenge. Regeneron also holds a right of first negotiation.

07 Commercial dynamics: prescribers, penetration and duration

Kiniksa discloses three commercial metrics consistently, and they are the most useful part of the earnings releases because they explain whether growth is coming from more patients, longer treatment, or price.

The epidemiology the company has used since the 2021 approval is a United States prevalent population of roughly 40,000 recurrent pericarditis patients, of whom around 14,000 have multiple recurrences and an inadequate response to conventional therapy. The 2025 annual report adds a second layer: approximately 26,000 further patients experiencing their first recurrence, and a stated strategy of encouraging prescribing earlier in the natural history of the disease, in biologic-appropriate patients at that first recurrence. Those two numbers frame the argument on both sides. The constructive reading is that four out of five patients in the target pool are not on the drug, and each new prescriber is a durable annuity because the mean course lasts about three years. The sceptical reading is that the core pool is fixed and modest: at full penetration of the 14,000-patient group the arithmetic ceiling of that segment is visible, and growth beyond it depends on converting first-recurrence patients — a population that has, by definition, not yet failed conventional therapy — or on a broader label, a new indication or a new product.

Gross-to-net deductions were 8.4 per cent in 2025, against 9.8 per cent in 2024. The company attributes the move to the impact of the Inflation Reduction Act through 2025 and to prior-period reserve adjustments taken in the fourth quarter, rather than to competitive discounting; the absolute level remains low, which is consistent with specialty distribution and with the absence of an approved competitor.

08 The pipeline, asset by asset

Kiniksa’s pipeline is smaller than it was two years ago, and that narrowing was deliberate. On February 25, 2025 the company discontinued development of abiprubart in Sjögren’s disease and terminated its licence for mavrilimumab, redirecting spending to the interleukin-1 franchise. What remains is two owned molecules in the same pathway and two assets sitting outside the operating story.

KPL-387 — the designated successor

A fully human IgG2 monoclonal antibody against interleukin-1 receptor 1, targeted at once-monthly subcutaneous self-administration in liquid formulation. FDA orphan drug designation for pericarditis was granted in October 2025. Phase 2 interval data was released on July 28, 2026 and is covered in section 02. Two studies are active and recruiting: PASTORALE (NCT07010159, 325 participants, 63 sites, started July 2025) and the transition-to-monotherapy posology study with long-term extension (NCT07288216, 80 participants, 41 sites, started March 2026). On the intellectual property side, one United States formulation patent is granted with a 2045 expiry, further formulation and manufacturing applications would run to 2045 if granted, and a method-of-use application in recurrent pericarditis would run to 2046. As a new biologic it would also expect twelve years of United States data exclusivity, and orphan exclusivity if the criteria are met.

KPL-1161 — the quarterly-dosing follow-on

An Fc-modified IgG2 against the same receptor, engineered for a longer half-life with a target profile of subcutaneous dosing once a quarter. It is preclinical. The company has said, most recently on July 28, 2026, that it remains on track to start a Phase 1 first-in-human trial by the end of 2026. Drug substance is planned for in-house manufacture. A composition-of-matter application would expire in 2046 if granted.

abiprubart (KPL-404) — shelved, with strategic alternatives

A humanised monoclonal antibody that binds CD40 and blocks the CD40–CD154 co-stimulatory signal. The Phase 2 study in Sjögren’s disease (NCT06531395) was terminated with three patients enrolled; an earlier Phase 2 in rheumatoid arthritis (NCT05198310, 145 patients) completed in 2024. Direct costs fell to $6.1 million in 2025 from $59.5 million in 2024, and termination notices went to contract manufacturers in February 2025. The asset was licensed from Beth Israel Deaconess Medical Center through the 2019 acquisition of Primatope Therapeutics, with clinical and regulatory milestones capped at $1.2 million and low single-digit royalties; composition-of-matter patents run to 2036 for the Primatope estate and to 2032 for those licensed from Beth Israel Deaconess. There is no active clinical programme and no announced outcome from the strategic review.

vixarelimab (KPL-716) — out-licensed

Licensed exclusively to Genentech and F. Hoffmann-La Roche in August 2022 for $80.0 million upfront, with a further $50.0 million already recognised and up to roughly $570.0 million of contingent payments remaining, plus tiered royalties from low double digits to mid-teens before upstream obligations to Biogen. Kiniksa completed its residual obligation in 2024 and recognised no collaboration revenue from it in 2025. Two Genentech Phase 2 studies have been terminated and one Phase 1c in ulcerative colitis remains active but not recruiting. In practice this is a call option on a partner’s decisions, not a line of sight to revenue.

09 Timeline: how the story got here

10 Financial position: what self-funding actually looks like here

The 2024-to-2026 progression is the clearest way to read this business.

Figures are taken from the results releases of February 24, 2026 and July 28, 2026 and from the Form 10-Q for the quarter ended June 30, 2026. The 2024 research and development figure includes the abiprubart programme discontinued in February 2025, which is why it is higher than 2025.

Operating cash flow in the first half of 2026 was positive $97.4 million, against $50.4 million in the first half of 2025. Investing absorbed $105.5 million, largely the deployment of cash into short-term investments, and financing contributed $18.2 million, which in the absence of any equity offering is option exercises and employee plan activity. There is no debt on the balance sheet and no debt facility described in the filings.

Two caveats on how these numbers should be read. First, the company reports its cash flow statement on a six-month basis in the second-quarter filing, so a standalone quarterly cash figure does not exist in the primary sources. Second, research and development is rising sharply — from $18.8 million to $40.9 million in a single year-on-year quarterly comparison — and will keep rising while PASTORALE enrols and KPL-1161 enters the clinic. Profitability at this company is a function of a commercial product carrying a development programme, and the development programme is getting more expensive on purpose.

11 Capital structure: four classes, one listing, and a voting cap

This is the least conventional part of the story and the part most often misread by screeners. At July 24, 2026 there were 78,024,013 ordinary shares outstanding, in four classes:

The rights are otherwise identical; the classes differ in voting, transferability and conversion. Class B has been converting into Class A — it stood at 1,795,158 shares at December 31, 2025 and at 895,158 by mid-2026, with the chief executive converting 900,000 shares on June 1, 2026. Conversion of A1 and B1 into voting stock above a 4.99 per cent holding requires 61 days’ notice, and since May 21, 2026 a deed of waiver additionally prevents the Baker Brothers funds from converting if the result would take their voting rights above 49.9 per cent. That waiver can only be amended with the approval of at least 75 per cent of the ordinary shares, except for amendments that add shareholders to it or add further restrictions, which need no vote — so the exceptions can only tighten it. It is a durable governance fact rather than a revocable courtesy.

Dilution: what is outstanding

At June 30, 2026 the equity awards outstanding were 8,489,746 options at a weighted average exercise price of $21.20, of which 5,582,824 were exercisable at $16.26; 289,173 performance options at $29.70; 1,853,614 unvested restricted stock units; and 580,770 unvested performance share units. A further 7,845,077 shares remained available for grant under the 2018 plan and 765,514 under the employee share purchase plan, and the January 1, 2026 evergreen provision added 3,051,742 Class A shares to the 2018 plan and 110,000 to the purchase plan. Unrecognised compensation cost stood at $122.7 million.

What is not there

No buyback: the second-quarter 10-Q answers the issuer purchases item with a single word, none. No at-the-market programme is disclosed, and no prospectus supplement of that kind has been filed since 2020. A $400 million universal shelf registration was filed on June 28, 2024 covering shares, preferred stock, debt, warrants and units, and has not been drawn on. For a company generating cash, an unused shelf is a tool rather than a warning; it becomes a warning only if it is drawn while the operating position is deteriorating.

12 Exclusivity and patents: the 2028 question

This section is the one to read twice, because the protection around ARCALYST is layered and each layer expires differently.

• Composition of matter. The United States patent on the molecule expired in 2020; the corresponding patents outside the United States expired in 2023. This protection no longer exists.
• Orphan exclusivity. The March 18, 2021 approval in recurrent pericarditis carried orphan drug exclusivity, described by the company as seven years of United States marketing exclusivity. On that arithmetic it runs to around March 2028. The company states the seven years but does not publish an expiry date, so the date here is a calculation from the approval date and should be treated as such.
• Method of use. Five United States patents covering methods of using ARCALYST in recurrent pericarditis carry a statutory term into 2038, before any patent term adjustment, and are licensed to Kiniksa on a field-specific exclusive basis through the Regeneron agreement.
• Biologic status. ARCALYST is a biologic, not a small molecule, so the competitive threat at exclusivity expiry is a biosimilar programme rather than a generic filing — a longer, costlier and more visible process for any challenger, against a patient pool of roughly 14,000 in the principal indication.

The honest framing is that method-of-use patents are a weaker form of protection than composition of matter, because they can in principle be designed around through labelling, but that a biosimilar of a fusion protein for a rare cardiovascular indication is not an obvious commercial target. The risk is real and dated; it is not the same as a small-molecule patent cliff.

13 Supply chain: one supplier, and a transfer in progress

For most of the franchise’s life Regeneron has been the sole supplier of ARCALYST drug substance, having begun a technology transfer of the manufacturing process in 2023. That changed during the second quarter of 2026: the Form 10-Q states that in June 2026 the FDA approved Samsung Biologics as Kiniksa’s replacement contract development and manufacturing organisation for drug substance. The single-source dependency is therefore in the process of being unwound rather than merely planned, and $20.9 million of Samsung-manufactured inventory had already been capitalised by the end of 2025.

The commitments attached to that arrangement have also grown. At June 30, 2026 minimum purchase commitments stood at $51.0 million to Regeneron, all falling due within twelve months, and $140.4 million to Samsung, of which $53.1 million falls within a year — against $24.6 million and $147.7 million respectively, with only $18.9 million of the Samsung figure due within a year, at December 31, 2025. In other words the totals are broadly stable but the near-term cash call has roughly tripled.

Two things follow. The first is that regulatory approval of a second manufacturer removes the sharpest version of the supply risk, which is the one where the licensor is also the only source. The second is that these commitments are cash obligations to be funded whether or not demand meets the guidance, and they are among the few large fixed obligations on a balance sheet that otherwise carries none.

14 Competition

In its 2025 annual report Kiniksa states that it is not aware of any other therapy approved by the FDA for recurrent pericarditis. That is the starting point, and it explains both the gross-to-net figure and the pricing power. What is coming is nonetheless visible.

The nearest dated competitive event is the MAVERIC registry primary completion of September 21, 2026. It is a competitor’s trial and a registry field rather than a guided readout, but if that study succeeds it creates an oral option for patients coming off interleukin-1 blockade, which is precisely the population Kiniksa is trying to keep on therapy for three years.

15 Analyst positioning

Coverage is small and uniformly positive, which is itself a piece of information: there is no published bear case from a covering bank, so the sceptical arguments in this hub come from the filings rather than from the sell side. As reported by the data providers on August 31, 2026, eight analysts cover the stock, the consensus rating is at the most positive end of the scale, and the average target is $98.00 with a range of $87 to $108.

The two figures above do not reconcile, and the discrepancy is worth flagging rather than smoothing over. The provider’s consensus of $98.00 with a floor of $87 cannot be the arithmetic mean of the eight actions listed in the table, which averages $93.4 and has a floor of $71 because of the Jefferies target set in April, before the second quarter. The most likely explanation is that the provider’s consensus uses only refreshed estimates and drops or updates stale ones, but that cannot be verified from the published data, so both series are shown as they stand.

The pattern is worth stating plainly rather than interpreting: six of the eight targets listed moved in the four days around the second-quarter release, which means the consensus is largely a reaction to one set of disclosures. Targets are third-party estimates, they are not forecasts by this publication, and they are reproduced here as a record of what the sell side published and when.

16 Ownership, insiders and retail sentiment

Who owns the company

From the proxy statement of April 16, 2026, on a record date of April 6, 2026, holders above five per cent of the listed Class A were FMR LLC with 10.50 per cent, Rubric Capital Management with 7.20 per cent, and entities managed by Baker Bros. Advisors with 6.53 per cent of Class A plus 100 per cent of both the A1 and B1 classes. Because A1 and B1 do not vote, Baker Brothers controlled only 4.71 per cent of the voting power despite being, in the proxy’s own words, the largest shareholder on an as-converted basis. That gap is the whole point of the May 2026 waiver.

Chief executive Sanj K. Patel held 4.99 per cent of Class A and 85.02 per cent of the high-vote Class B, giving him 26.64 per cent of total voting power; executive officers, directors and entities affiliated with certain directors held 15.20 per cent of Class A and 33.91 per cent of the voting power, although the proxy states that they do not vote together as a group. Institutional ownership was reported at 55.83 per cent on August 31, 2026, against a float of 40.20 million shares. Among the largest thirteen-F positions at June 30, 2026, BlackRock’s holding rose sharply to 2.74 million shares, Invesco, State Street and Geode all added materially, while Acadian and D. E. Shaw each cut roughly half of their positions.

Insider transactions

In the twelve months to August 19, 2026 the Form 4 record shows roughly $69.4 million of gross share disposals across fourteen insiders, against about $17.2 million of associated option exercise cost, and no open-market purchases. The largest single day was April 28, 2026, when the chief executive exercised 483,654 options at a weighted average strike near $12.56 and sold the same number of shares. By person, the chief executive accounted for around $40.6 million of the total, the chief medical officer about $8.3 million, the outgoing chief strategy officer about $8.0 million and the chief financial officer about $5.9 million.

Retail sentiment — non-professional opinion, not research

On Stocktwits, read on September 1, 2026, the sentiment score for $KNSA sat at 50 out of 100, a neutral reading, with 2,534 watchers. Message volume tells a more interesting story than the score: activity on a one-month view was up about 95 per cent against its own baseline and up on three and six-month views, while the last week cooled. Across the six-month sentiment history the peaks line up exactly with disclosure: late April 2026, when the first guidance raise and the target increases landed, and July 28 to 30, the second-quarter release. The troughs are mid-July and early August, with no disclosure attached to them.

The recurring themes in retail discussion are round-number price objectives stated without argument, repetition of the second-quarter revenue and profit figures as the bull thesis, complaints about how thinly the shares trade, and generic anticipation of unspecified news. A material share of the message count is automated technical signals and promotional posts rather than opinion. These are the views of non-professional traders on social platforms; they are not analyst research, they are not verified, and they are recorded here only as a measure of attention. No identifiable, dated Reddit or X discussion on this ticker was found in the checks run for this hub, so no claim is made about sentiment on those platforms.

17 Catalysts and watchpoints

Alongside the dated items, four things are worth monitoring in each quarterly release because they are where the thesis breaks or holds: the cumulative prescriber count and the penetration percentage; the collaboration expense line, which shows what proportion of gross profit leaves the company; research and development expense, which is now rising fast and is the price of the succession plan; and any disclosure on the ramp of Samsung-manufactured supply now that the FDA has approved that site, together with the near-term purchase commitments attached to it.

18 Bull, base and bear scenarios

These are descriptions of how the same set of disclosed facts could develop. They are not predictions, not probabilities and not advice.

Bull scenario

Penetration keeps climbing from the roughly 21 per cent reported at mid-2026, prescriber additions continue at the recent pace, and the mean three-year course of therapy compounds the installed base. Guidance is raised a third time, and 2027 revenue moves past a billion dollars with the cost base growing more slowly than revenue, so operating leverage becomes visible. PASTORALE reads out positively, KPL-387 is approved with monthly dosing and no profit split, and the posology study demonstrates that patients can be transitioned from ARCALYST onto it, converting the exclusivity expiry from a cliff into a handover. KPL-1161 enters the clinic on schedule and gives the franchise a third generation with quarterly dosing.

Base scenario

Growth continues but decelerates as the easier part of the pool is treated, and the company lands in or near its guided range for 2026. Research and development keeps rising while PASTORALE enrols, so reported profitability flattens even as revenue grows. KPL-387 stays on its stated 2028/2029 track, which means there is an interval where ARCALYST is past or near its exclusivity date and the successor is not yet approved. The company covers that interval with method-of-use patents, switching costs and the fact that no biosimilar developer has publicly targeted the molecule. Cash keeps building; the shelf stays unused.

Bear scenario

PASTORALE disappoints on the randomised withdrawal endpoint, or the timeline slips past 2029, and the succession plan loses its anchor. Meanwhile the competitive picture changes: Cardiol’s MAVERIC succeeds and gives physicians an oral option for patients coming off interleukin-1 blockade, or the NLRP3 programme that Eli Lilly agreed to acquire accelerates. Growth decelerates faster than expected as the 14,000-patient pool is worked through, collaboration expense keeps taking half the profit, and the manufacturing commitments of $191.4 million across the two suppliers have to be honoured into softer demand. In that combination, a company whose entire revenue line is one licensed product with an expiring exclusivity has very little to fall back on: abiprubart is shelved, vixarelimab is somebody else’s decision, and KPL-1161 has not yet dosed a human.

19 Red flags and open questions

• One product is the whole revenue line. There was no licence or collaboration revenue at all in the second quarter of 2026. Total revenue and ARCALYST revenue are the same number.
• Half the product profit is contractually somebody else’s. Collaboration expense of $229.5 million in 2025 and $163.6 million in the first half of 2026 is not a cost the company can optimise away; it scales with success.
• The exclusivity date is close and the successor is not. Orphan exclusivity runs to about March 2028 on the seven-year arithmetic; the company’s own commercialisation window for KPL-387 is 2028/2029. Those two ranges do not comfortably overlap.
• Supply concentration, now easing. The licensor was the sole source of drug substance until the FDA approved Samsung Biologics as a replacement manufacturer in June 2026. The commitments attached to the two suppliers were $51.0 million and $140.4 million at June 30, 2026, with $118.5 million of total minimum commitments falling due within twelve months.
• The Phase 2 data released was partial. An interval analysis of an open-label dose-focusing portion, with no per-arm numbers, no full safety table and no placebo comparison, and with the company’s own caution that final data may differ.
• Insider selling without insider buying. Around $69.4 million of gross disposals in twelve months and no open-market purchases. Predominantly exercise-and-sell, but the 10b5-1 status is not confirmable from aggregated data.
• Concentrated control and a thin float. Directors and officers hold about a third of the voting power, the chief executive alone more than a quarter, and only the Class A shares are listed.
• Coverage is uniformly positive. Eight analysts, no published bear case, and six of the eight targets set within four days of one release.
• An unexplained labelling supplement. The FDA approved a labelling supplement to the ARCALYST application on March 10, 2026. It is not classified as a new indication and the company has not commented on it in any release; what it changed has not been established here.
• No announced date for the next results. As of September 1, 2026 the third-quarter date has not been published.

20 Merlintrader bottom line

Kiniksa is the unusual case of a mid-cap biotech that has already answered the question most of its peers are still asking: can the product sell. ARCALYST sells, it sells faster than the company expected twice in five months, and the money it generates stays in the business rather than being replaced by equity issuance. Two consecutive guidance raises, six consecutive profitable quarters, more than half a billion dollars of cash and no debt are the facts on the record.

What the company has not yet answered is the succession question, and that is the whole investment debate compressed into one sentence. The asset producing all of the revenue is licensed on a 50/50 split, its composition-of-matter protection is long gone, and the orphan exclusivity that replaced it runs out around the same time the designated successor is due to reach the market. Everything Kiniksa is doing — the pivotal PASTORALE study, the transition-to-monotherapy posology trial that enrols patients already on rilonacept, the second-generation quarterly molecule, the manufacturing transfer to a second supplier — is an attempt to make that handover continuous rather than abrupt.

So the reasonable way to hold this hub in mind is as a two-clock problem. One clock is commercial and is running in the company’s favour: penetration, prescribers, duration of therapy. The other is regulatory and is running against it: March 2028, against a stated 2028/2029 arrival for the replacement. Each quarterly release moves both. That is what to watch, and it is why the prescriber count and the collaboration expense line deserve more attention than the headline revenue number.