TG Therapeutics, Inc. (NASDAQ: TGTX) — A Real Drug Whose Profits Live in the Receivables

Report date: 2026-07-25 · Price at close 2026-07-24: $56.04 · Shares outstanding: 153,093,879 (2026-03-31) Sector: Health Care · Biotechnology (Multiple Sclerosis / B-cell therapy) · SEC CIK: 0001001316

This is an independent analyst’s article. Sections 1–15 carry no recommendation and no price target, and discuss valuation only as embedded expectations and scenarios. The single exception is the Claude's Take block immediately below, which is the author’s own opinion. General information only — not investment advice.

⚡ Claude’s Take

The author’s own subjective opinion, offered as general information and not investment advice. The analytical body (Sections 1–15) below carries no position and no price target.

Verdict: HOLD / AVOID-HERE at $56.04. A genuinely good product and a genuinely well-run commercial launch, priced at an all-time high for a subcutaneous programme that has not read out — by a company that has never generated a dollar of operating cash. Accumulation zone: $33–40 (roughly 5.7–6.8x EV/2026E sales), which is both where the stock traded for the first five months of this year and where the company itself repurchased ~5% of its own equity at ~$29.

Tag: “The revenue is real. The cash is a receivable.”

The bull case here is not a fantasy, and I want to be explicit about that before dismantling the price. BRIUMVI is winning. US net product revenue has gone $92M → $314M → $594M in three years and is guided to $885–900M this year; twelve consecutive quarters of sequential growth; more than 25,000 patients prescribed; record new-patient starts; and share taken directly from a Roche franchise many times TGTX’s size. Gross margin is 84%. Management raised guidance twice inside six months rather than resetting it. The CEO bought 100,000 shares in the open market at $10.13 days after the stock halved in August 2023, and neither he nor any other insider has sold a single share into a 2026 that has nearly doubled the stock. The buyback — 6.8M shares, ~5% of the company, at an average of ~$29 against today’s $56 — is one of the better-executed repurchase programmes I have seen in commercial-stage biotech. None of that is fake.

What I will not pay $8.8bn of enterprise value for is the other half of the picture. First, the earnings are not what the screen says. FY2025’s $447M of net income and $2.77 of EPS are 78% a one-time deferred-tax valuation-allowance release; pretax income was $107M. The 19.5x “P/E” and its 33rd-percentile own-history rank are accounting artefacts and should be discarded. Second, and far more important, the cash has never arrived. Operating cash flow has been negative in every year on record — including −$24.8M in 2025 on $616M of revenue, and −$17.9M in Q1-2026 — because receivables are compounding faster than sales. Days sales outstanding have gone 93 → 109 → 145 → 175 across nine straight quarters; receivables of $392M now equal 42% of the entire 2026 revenue guide. The 10-K explains why in its own words: standard terms are 30–60 days, but “extended payment terms have been offered during the BRIUMVI commercial launch.” I do not think this is fraud, and the company says collectability is unimpaired. I do think it means the reported growth rate is flattered by credit the company is extending, and that a business three-and-a-half years into launch which still cannot fund itself is not yet the self-financing franchise the multiple assumes. Third, the price already owns the option. At 9.5x forward sales, roughly $3.7bn of the $8.8bn EV — about 42% — is payment for the subcutaneous programme, which reads out at year-end 2026 or Q1-2027 and launches, if all goes well, in 2028, into a segment Novartis’s Kesimpta already occupies and Roche is entering with an on-body device. That is a binary being underwritten at par. Fourth, the governance is a live cost. Shareholders rejected say-on-pay on 11 June 2026 by 48.9M votes to 31.9M — 39.5% support — against a CEO contractually entitled to an annual equity grant worth ten times his salary plus prior-year bonus ($25.6M of 2025 compensation on $616M of revenue), a $16.7M cash bonus tied to a $10bn market-cap threshold, and related-party arrangements with his other listed vehicle; a shareholder firm announced a self-dealing investigation on 16 July. And fifth, the clock is shorter than it sounds. Management says “patent protection into the 2040s.” The 10-K says the earliest composition-of-matter patent expires 2029 in the US and already expired in 2025 across most of Europe; the hard, uncontestable floor is BPCIA exclusivity to December 2034 — about eight years.

Framing: momentum, unambiguously — not contrarian, not value, not a falling knife. The factor evidence is emphatic. The stock is 5.1% off an all-time high made on 9 July, up 88% year-to-date, with a three-month annualised Sharpe of 8.2 and a six-month Sharpe of 4.5. The counterweight is in the same dataset: five-year annualised return of just +8.7% with a −90.2% maximum drawdown, ten-year −93.2%, and 41.6% idiosyncratic volatility on a 19% R². FactorsToday’s nearest neighbours are XBI, LABU, FDMT and STOK — the model still reads TGTX as a high-beta clinical-stage biotech, not a de-risked specialty pharma. This is a security that has twice lost nine-tenths of its value on single binary events, now trading at its highest-ever price into another binary. Buying momentum at the top of that distribution is a trade, not an investment, and it is not the trade I want.

Conviction: medium. Flips bullish: two consecutive quarters of genuinely positive operating cash flow with DSO reverting toward 100–120 days — proving the receivable is a launch artefact that inverts rather than revenue bought with credit. Positive subcutaneous Phase 3 data at a $35–40 price would make this straightforwardly interesting. Flips bearish: a subcutaneous Phase 3 miss on exposure non-inferiority, or any deterioration in receivable quality — a credit-loss allowance, a change in the extended-terms disclosure, or DSO through 200 days.

📈 Stock Price Action — Five-Year Event Map

TGTX closed at $56.04 on 2026-07-24, 5.1% below its all-time-high close of $59.06 set on 2026-07-09, at the top of a 52-week range of $26.39–$59.06. The five-year path is a near-total wipeout followed by a fifteen-fold recovery: from $34.99 in July 2021 to a low of $3.74 on 2022-06-16 (−89%), and back to today’s all-time high. Roughly two-thirds of the recovery from the 2026 low has come in the last three months.

Cycle narrative. (1) The 2021–22 collapse is the single most important fact about this security’s risk profile: TGTX lost 89% of its value when umbralisib’s benefit-risk failed at the agency, and that episode alone produces the −90.2% five-year maximum drawdown the factor model still carries. (2–3) BRIUMVI’s December 2022 approval and the 2023 Neuraxpharm ex-US deal restored a business. (4) The −49.3% session on 1 August 2023 was the market discovering that an approval is not a launch; CEO Michael Weiss bought 100,000 shares in the open market ten days later at $10.13, the only meaningful insider purchase in the five-year Form 4 corpus. (5) From late 2023 the ramp compounded — twelve consecutive quarters of sequential US revenue growth — and the stock followed. (6) 2025 was a consolidation year: the stock reached $45.51 in April, fell 18% on the Q2 print, and spent the rest of the year between $29 and $36. (7–8) The 2026 melt-up has two legs — a fundamental beat-and-raise on 6 May, then a dense six-week catalyst cluster from 27 May: ENHANCE met its bioequivalence endpoint (single Day-1 infusion), the subcutaneous Phase 1 delivered supportive PK/PD with the Phase 3 fully enrolled, subcutaneous BRIUMVI produced positive Phase 1 data in myasthenia gravis with a registration-directed Phase 2 initiated, and an exploratory Phase 2 in treatment-resistant schizophrenia began. June 2026 alone was +44.8%, achieved as a grind rather than a gap — no single June session ranks in the five-year top 25.

1. Executive Summary

TG Therapeutics is a single-product commercial-stage biotechnology company. BRIUMVI (ublituximab-xiiy), an anti-CD20 monoclonal antibody in-licensed from LFB Group and approved by the FDA in December 2022 for relapsing forms of multiple sclerosis, is approximately 98% of revenue. The commercial execution has been excellent: US net product revenue of $92.0M (2023), $310.0M (2024) and $594.1M (2025), guided to $885–900M in 2026, with twelve consecutive quarters of sequential growth, more than 25,000 patients prescribed globally, and share taken directly from Roche’s Ocrevus in the IV anti-CD20 segment. Gross margin is 83.7%; operating income reached $123.3M in FY2025 (20.0% margin) from $41.9M a year earlier.

Three things complicate that picture, and they are the substance of this memo.

First, the reported earnings are materially overstated. FY2025 net income of $447.2M ($2.77 diluted) includes a $339.8M income-tax benefit from the release of the deferred-tax-asset valuation allowance. Pretax income was $107.4M. Any multiple computed on the headline EPS — including the 19.5x P/E and its 33rd-percentile own-history rank — is meaningless.

Second, and more consequential, the business has never generated operating cash. Operating cash flow was negative in every year on record: −$295.6M (2021), −$176.2M (2022), −$31.4M (2023), −$40.5M (2024), −$24.8M (2025), and −$17.9M in Q1-2026. Cumulative 2023–2025 GAAP net income of $483.2M sits against cumulative operating cash flow of −$96.7M. The proximate cause is receivables: accounts receivable have risen from $65.0M (Q1-2024) to $392.0M (Q1-2026) — implied DSO of 93 → 175 days, rising for nine consecutive quarters — against revenue that rose 223% over the same period. The 10-K states that standard payment terms are 30–60 days but that “extended payment terms have been offered during the BRIUMVI commercial launch,” and lists the timing of collection on those terms as a principal determinant of the Company’s cash needs.

Third, the valuation already capitalises the pipeline. At $56.04, market capitalisation is $8.58bn and enterprise value approximately $8.76bn — 12.5x TTM revenue, 9.5x the 2026 revenue guide, and 58.6x TTM operating income. A mature-specialty-pharma frame on the IV franchise alone supports roughly $5bn; the residual ~42% of EV is payment for the subcutaneous programme (Phase 3 topline due year-end 2026 or Q1-2027, launch targeted 2028), the ENHANCE single-dose label, myasthenia gravis and azer-cel.

On competitive position, the honest answer is that BRIUMVI has a regulatory barrier, not an economic moat. There is no supply-side cost advantage (the molecule is in-licensed on a high-single-digit royalty; manufacturing is outsourced with sole-source exposure), no network effect, and only modest demand-side captivity from physician habit and patient persistence. Greenwald’s decisive test — market-share stability — fails: share in this class moves fast, which today favours TGTX and offers no structural reason it cannot reverse. The barrier that matters is BPCIA exclusivity to December 2034.

Capital allocation splits. Transactionally it is good: 6.8M shares (~5% of the company) repurchased at ~$29 against a $56 stock; small, optioned licensing deals; expensive but non-dilutive debt. Governance is poor and has now been formally rejected — say-on-pay failed on 11 June 2026 with 39.5% support, against a CEO contractually entitled to annual equity worth ten times salary-plus-bonus, a $16.7M market-cap-threshold cash bonus, and related-party arrangements with Fortress Biotech, where he is Executive Vice Chairman.

The stock sits 5.1% below an all-time high, up 88% year-to-date, with a three-month annualised Sharpe of 8.2 — against a five-year annualised return of +8.7% and a −90.2% five-year maximum drawdown.

2. Business Overview

2.1 What the company does

TG Therapeutics develops and commercialises therapies for B-cell-mediated diseases. Incorporated in Delaware in 1993, headquartered at 3020 Carrington Mill Blvd, Morrisville, North Carolina, it has been led since 2011 by Michael S. Weiss as Chairman, CEO and President.

The entire economic substance of the company today is one molecule: BRIUMVI (ublituximab-xiiy), a glycoengineered anti-CD20 monoclonal antibody administered as a one-hour intravenous infusion every 24 weeks following a starting dose, approved by the FDA in December 2022 for adults with relapsing forms of multiple sclerosis (RMS) — including clinically isolated syndrome, relapsing-remitting disease and active secondary progressive disease — on the basis of the ULTIMATE I & II Phase 3 trials.

The molecule is in-licensed, not owned outright. Under a January 2012 exclusive licence from LFB Biotechnologies, GTC Biotherapeutics and LFB/GTC LLC (collectively LFB Group), TGTX holds exclusive worldwide rights excluding France and Belgium. TGTX has incurred approximately $31.0M of milestone expense under that agreement, and LFB Group is entitled to high-single-digit royalties on net sales of ublituximab. This is a permanent, senior claim on the franchise’s economics.

2.2 Revenue composition

The FY2023 line is a trap for the unwary and matters for every growth calculation in this memo. Of $233.7M of 2023 revenue, only $92.0M was product — the remaining $141.7M was the one-off recognition of the Neuraxpharm ex-US commercialisation agreement signed in July 2023. Headline “revenue growth” of 40.8% in 2024 understates a product revenue increase of 241%. Conversely, anyone anchoring on 2023’s revenue base as recurring is anchoring on a licence payment.

Recurring vs. non-recurring. Post-2023 the revenue base is overwhelmingly recurring in the useful sense: BRIUMVI is dosed every 24 weeks and management reports better-than-modelled persistence, with patients tracking well at weeks 48, 72 and 96. As the Chief Commercial Officer put it, “our revenue base doesn’t reset each year, it builds. Each cohort of new patients adds to an expanding base of recurring demand.” That is a genuine and favourable business-model characteristic — a maintenance-therapy annuity, not an episodic sale — and it is the strongest argument for the durability of the ramp.

2.3 How it makes money, and to whom it sells

TGTX sells BRIUMVI in the US to specialty distributors, who supply infusion sites (hospital outpatient departments, private neurology practices with infusion capability, independent infusion centres). Because the drug is physician-administered, it is reimbursed under Medicare Part B and commercial medical benefit rather than pharmacy benefit — a “buy-and-bill” economic model.

Customer concentration is extreme, though typical of US pharmaceutical distribution. In Q1-2026, four customers accounted for 42%, 27%, 20% and 9% of gross product revenue — 98% combined — and for 38%, 19%, 28% and 13% of product accounts receivable.

Gross-to-net is guided to average approximately 65% for 2026, with Q1 seasonally worse on deductible resets and heavy co-pay-programme utilisation. TGTX must offer statutory discounts and rebates under the Medicaid Drug Rebate Program, the 340B drug pricing programme and Medicare Part D, and report prices under Medicaid and Part B.

Ex-US, TGTX does not commercialise directly. Under the July 2023 Commercialization Agreement, Neuraxpharm holds ex-US rights; TGTX earns a manufacturing/supply margin on product sold to Neuraxpharm plus milestone and royalty consideration. This structure caps ex-US economics but also caps ex-US cost — and, as the 10-K notes, leaves TGTX unable to influence ex-US pricing that could become a reference point under “most favored nation” pricing frameworks. Ildong Pharmaceutical holds a 2012 sublicence for South Korea, Taiwan and Southeast Asia.

2.4 Pipeline

Azer-cel is licensed from Precision BioSciences (January 2024) — an exclusive licence for autoimmune and non-oncology indications plus a non-exclusive manufacturing licence, with mid-single-digit royalties payable to Precision. It is early, dose-escalating, and management is candid that it is “still early.”

Verdict. This is a well-executed single-asset commercial business with a genuinely recurring dosing model, an in-licensed molecule carrying a permanent royalty, extreme distributor concentration, and a pipeline that is overwhelmingly the same molecule in new formulations and indications. The “pipeline within a product” framing management uses is accurate — which is both the efficiency of the model and the concentration of its risk.

3. Industry Dynamics

3.1 Structure of the relapsing-MS market

Multiple sclerosis therapy has consolidated, over roughly a decade, around high-efficacy B-cell depletion. The incumbent side of that transition is visible in Biogen’s own disclosure: its multiple-sclerosis franchise fell 7.1% in 2025 to $4.04bn as Tecfidera genericised, Tysabri faced a biosimilar and the interferons declined — while Biogen simultaneously collects a ~$1.86bn passive royalty on Genentech/Roche’s anti-CD20 antibodies, including Ocrevus (Biogen FY2025 Form 10-K). Once the molecule patents lapse in this market, price and share collapse toward marginal cost — the classic no-barrier commodity endgame. That is the structural backdrop against which TGTX’s growth should be read. TGTX is not creating a market; it is participating in a class that is taking one.

The anti-CD20 MS class is effectively a three-molecule oligopoly:

The class splits across two administration segments. Management sizes the patient-administered subcutaneous segment at roughly 35–40% of the anti-CD20 market — a segment in which TGTX today has zero participation. The remaining ~60–65% is HCP-administered infusion, and it is here that BRIUMVI competes and is winning.

3.2 Competitive intensity — and the direction of the capital cycle

The 10-K’s own Competition section is unusually direct, and worth quoting rather than paraphrasing: BRIUMVI “directly competes with ocrelizumab, the only other approved intravenously administered anti-CD20 monoclonal antibody (Roche Holdings AG),” and also “competes with a subcutaneous version of ocrelizumab administered by healthcare providers,” and with “ofatumumab (Novartis AG), a patient/self-administered subcutaneous anti-CD20 monoclonal antibody approved for MS” — which “would represent direct competition for any future self-administered subcutaneous formulation of ublituximab currently under development.”

Read through the Marathon capital-cycle lens, the class today is in the attractive phase: high returns, expanding volume, no biosimilar entry possible before the 2030s. But the supply-side response is fully visible and already funded:

• Roche is defending on convenience — a healthcare-provider-administered subcutaneous Ocrevus is already competing, and management confirmed on the Q1-26 call that Roche is developing “their own at-home on-body device.” Weiss himself said “I don’t think OCREVUS’ long-term plan is to participate much longer in the IV marketplace” — which is another way of saying that the segment TGTX is winning is the segment the leader is preparing to leave.
• Novartis is extending Kesimpta’s dosing interval to defend the self-administered segment.
• BTK inhibitors — fenebrutinib (Roche) at the agency, remibrutinib (Novartis) behind it — are a genuinely new mechanism. Management’s view is that “there’s still a big oral market… room for a BTK with the right profile,” but “we don’t think it’s a drug class that will have a material impact on the CD20 class.” That is a hypothesis, not evidence; each BTK to date has struggled on benefit-risk, which supports management’s reading but does not settle it.
• Anti-CD40L agents are in clinical development for MS and other autoimmune indications.

The critical structural point: regulation distorts the capital cycle here in TGTX’s favour, but only through one channel. The BPCIA 12-year exclusivity wall makes the ordinary “high returns attract capital until generic entry mean-reverts them” mechanism impossible before the mid-2030s. Competition therefore runs entirely through differentiation and convenience — which is a far faster, cheaper and more frequent form of entry than building a biosimilar. TGTX is not protected from competition; it is protected only from copying.

3.3 Regulatory and reimbursement landscape

Because BRIUMVI is physician-administered and reimbursed under Medicare Part B, its policy exposures are specific:

• IRA Part B inflation rebates penalise price increases above CPI — constraining the price lever on a product whose growth to date has been volume, not price.
• 340B and the Medicaid Drug Rebate Program impose statutory discounts; 340B in particular expands with hospital-outpatient site-of-care mix, and management has acknowledged gross-to-net varies with hospital exposure.
• “Most favored nation” / international reference pricing is called out explicitly and, unusually, with a company-specific aggravating factor: TGTX has partnered ex-US rights to Neuraxpharm and therefore “may have limited ability to influence ex-US pricing decisions that could be used as reference points under MFN or similar frameworks.” TGTX has outsourced ex-US commercialisation and the ability to defend its own reference price.
• Medicare drug price negotiation applies to biologics 13 years post-licensure — beyond this memo’s horizon for BRIUMVI, but within the life of the asset.

3.4 Verdict

Structurally good for now; structurally deteriorating on a five-to-ten-year view. A protected three-player oligopoly, growing volume, no biosimilar until at least December 2034 and a demonstrable secular shift toward the class is a good place to be, and the profit pool is real — 84% gross margins are not achieved in bad industries.

But the barrier is a regulatory clock, not an economic moat, and the two competitors on the other side of it are among the largest and best-capitalised pharmaceutical companies in the world, both of whom have already funded the convenience-based response. The specific asymmetry is uncomfortable: TGTX is taking share in the segment (IV) that the class leader is strategically de-emphasising, while it has no presence at all in the segment (self-administered SC) that is growing and where the leader is heading. The subcutaneous Phase 3 is not an upside option on top of a secure position; it is the price of continued relevance in the class over the 2030s.

4. Competitive Position

4.1 Naming the moat — or its absence

Applying the Greenwald taxonomy rigorously, there are exactly three genuine sources of competitive advantage: supply/cost advantage, demand-side customer captivity, and economies of scale combined with captivity. Testing BRIUMVI against each:

Supply / cost advantage: none. TGTX did not discover ublituximab. It in-licensed the molecule from LFB Group in 2012 and pays high-single-digit royalties on net sales in perpetuity (until the later of patent expiry, loss of regulatory exclusivity, or ten years from first commercial sale, country by country). It does not own manufacturing; the 10-K’s risk factors state that third parties supplying starting materials, intermediates, API/drug substance and drug product “are our sole source of supply, and the loss or disruption of any of these suppliers could significantly harm our business.” A company that rents its molecule and its factory has no cost advantage. Gross margin of 84% is a biologics-industry characteristic, not a TGTX characteristic — Ocrevus and Kesimpta earn similar or better.

Network effects: none. There is no mechanism by which one physician prescribing BRIUMVI makes it more valuable to the next. Management’s line — “the more patients that go on BRIUMVI, the more patients will go on BRIUMVI” — describes a reference-and-familiarity effect, which is real word-of-mouth and real prescriber-habit formation, but it is not a network effect. It creates no barrier a competitor with a better profile and a larger sales force cannot overcome.

Demand-side captivity: modest and real. This is where the actual advantage lives, and it should not be dismissed. Once a patient is established on a 24-weekly infusion at a specific site of care, with benefits verified and a treatment routine built, the friction of switching is meaningful: re-verification, a new titration, and — critically — the physician’s unwillingness to disturb a stable patient. Management reports persistence “stronger than what we originally modeled,” tracking well at weeks 48, 72 and 96. Six-year ULTIMATE open-label extension data presented at ECTRIMS in September 2025 showed ~90% of patients free from 24-week confirmed disability progression and a relapse rate equating to one relapse per 83 patient-years, with no new safety signals. That is a genuinely strong long-term dataset and it is the substance behind physician confidence.

Economies of scale with captivity: not yet. TGTX’s field organisation is expanding, not consolidating. The one genuine scale argument management makes is prospective and worth taking seriously: an ~80% overlap between the IV and subcutaneous field forces means a subcutaneous launch would carry “not a huge incremental investment.” If subcutaneous succeeds, TGTX would gain operating leverage from selling two formulations through one commercial infrastructure. That is a real future scale economy — conditional on a Phase 3 that has not read out.

4.2 The market-share-stability test

Greenwald’s decisive empirical test for a barrier to entry is stability of market share. Where barriers are real, shares are sticky and change slowly; where they are absent, share moves.

Share in the anti-CD20 MS class is moving fast. TGTX has gone from zero to a claimed “#1 CD20 by dynamic share in private practices with infusion capabilities” in roughly three years, with the treatment-naïve proportion of its mix rising and, per management, no decrease in switches from Ocrevus. Management explicitly notes “the majority of the de novo business seems to be coming from Ocrevus IV.”

That is unambiguously good news for TGTX’s earnings over the next several years. It is also, read correctly, evidence that this class has no barrier to entry. A third entrant does not take the leading share of a segment inside three years in a market protected by real switching costs. The mechanism that let TGTX in — a better infusion experience, a simpler schedule, credible long-term data, and a hungrier commercial organisation — is available to anyone with a comparable molecule and more capital. Roche and Novartis both have both.

4.3 Direct comparison and the real differentiator

BRIUMVI’s differentiation is honest, narrow and operational: a one-hour infusion every 24 weeks after the starting dose, versus a materially longer Ocrevus IV infusion. Adam Waldman’s framing on the Q4-25 call is precisely right about what is driving share — “durability of the clinical profile, the 6-year data, especially the safety data… the operational advantages of BRIUMVI, the 1-hour infusion, the twice a year… relevant for patients and infusion centers.”

Two developments extend that differentiation. ENHANCE (Phase 3, primary endpoint met 27 May 2026) demonstrated bioequivalent exposure between the approved Day-1/Day-15 split initiation and a single 600mg Day-1 infusion. If approved, it removes an entire infusion visit from initiation — a genuine, if incremental, operational advantage over an Ocrevus IV start that still requires two. And in a new datapoint, TGTX presented prospective data at AAN 2026 on patients switching from a prior anti-CD20 who reported improvement in “wearing off” symptoms — a differentiated use case if it replicates.

But the same operational logic that makes BRIUMVI the better infusion is the logic that makes an infusion the wrong place to be over time. Convenience competition in this class resolves toward no infusion at all, and that end-state is Kesimpta’s today and Ocrevus’s tomorrow. TGTX’s entire long-run position depends on arriving there by 2028, three-plus years after Kesimpta and roughly contemporaneously with an on-body Ocrevus, with a molecule whose brand equity is built on infusion excellence.

4.4 Intellectual property — the real barrier, and its actual length

This deserves precision because management’s public framing and the 10-K’s disclosure diverge.

From the 10-K (FY2025, Item 1, Intellectual Property and Patents):

• The earliest composition-of-matter family for ublituximab has “expected expiration for the composition of matter patent in the U.S. [of] 2029, exclusive of patent term extension,” and “in Europe and other non-U.S. jurisdictions… 2025.” Patent term extension has been applied for in the US and granted in 13 European countries, “extending expiry into 2030 in these countries.”
• A later-filed family covering “compositions of matter comprising ublituximab, methods of manufacturing those compositions and methods for treating multiple sclerosis using those compositions” — four issued US patents and two pending applications — “extends patent protection on the composition of matter for ublituximab until 2042, not accounting for any patent term adjustment or extensions or terminal disclaimers.”
• Separately and independently, BPCIA gives 12 years of US market exclusivity from the December 2022 approval — “which extends until December 2034” — during which no referencing biosimilar can be approved.

On the Q1-2026 call, management described BRIUMVI as “supported by patent protection into the 2040s.” That is the most favourable available construction. It rests entirely on the later-filed 2042 family, whose claims — composition-of-matter comprising, manufacturing methods, methods of treatment — are the class of claims most vulnerable to inter partes review, terminal disclaimer and design-around, and which a biosimilar filer with Roche’s or Novartis’s resources would attack directly.

The defensible statement is: hard, uncontestable US exclusivity to December 2034 (~8.4 years from today), with a contested tail to 2042 that may or may not hold. That distinction is not academic — it is roughly the difference between a nine-year annuity and a sixteen-year one, and at a 10% discount rate it is worth a large fraction of the equity value.

4.5 Verdict

No durable competitive advantage — a strong product inside a regulatory window, in a class with weak barriers, against two far larger competitors.

To apply the test set out in this firm’s operating standard: if a “moat” claim cannot be tied to a financial outcome that would deteriorate without it, it is not a moat. Strip away BPCIA exclusivity and the LFB licence and ask what would deteriorate: essentially everything. Strip away only the patient-persistence and physician-habit advantages and the answer is: some near-term retention, modestly. That is the honest measure of the endogenous moat — small, real, and nothing like the size of the regulatory barrier that does the actual work.

This is not a condemnation of the investment. A protected, well-executed, high-margin annuity with eight to sixteen years of duration can be extremely valuable. It is a statement about what one is buying and therefore about what multiple is appropriate — and a company with a terminal date is worth a materially lower multiple of sales than a company with a franchise, at the same growth rate.

5. Growth History and Forward Opportunities

5.1 The record

* Derived from guidance: ~$925M revenue at ~84% gross margin less guided opex of $350M ex-SBC plus $100M of subcutaneous manufacturing and secondary-manufacturer start-up costs. Excludes stock-based compensation, which ran $64.7M in FY2025.

Twelve consecutive quarters of sequential US revenue growth since launch. Quarterly US revenue: $63.5M (Q1-24) → $73.5M → $83.9M → $108.2M → $120.9M → $141.1M → $161.7M → $182.7M → $194.8M (Q1-26). Q2-2026 is guided to approximately $220M.

The growth is entirely organic and entirely one product. There have been no acquisitions of revenue. The gross-margin decline from 94% to 84% is a mix and scale phenomenon (a 2023 base weighted to high-margin licence revenue, plus a Q4-25 inventory reserve and ex-US supply-sale timing), not price erosion.

5.2 What is actually driving it

Management identifies four drivers, and the transcript evidence is internally consistent across two quarters:

1. Record new-patient starts. March 2026 was the highest month ever; Q1-2026 set a record for new enrollments; monthly total prescribers set a new high in March.
2. Better-than-modelled persistence. Patients tracking well at weeks 48, 72 and 96, with the second treatment year — historically where drop-off occurs — running ahead of plan.
3. Rising treatment-naïve mix. Management calls this “the strongest leading indicator of long-term market position,” and it is the right metric: capturing a patient at the start of their CD20 journey is worth far more than a switch, because it front-loads the persistence annuity.
4. Commercial investment. Field-force expansion into community neurology and independent infusion centres; a direct-to-consumer campaign including a Super Bowl LX launch of “Next In MS” with Christina Applegate.

The compounding structure is the most attractive feature of the business. With 24-weekly dosing and strong persistence, each cohort “pancakes” onto the prior base. Roughly, if persistence is ~85–90% annually, a growing new-start cohort produces revenue growth that decelerates gently rather than cliffing — which is precisely the pattern observed (+237% → +92% → +63% → ~+50% guided).

5.3 Forward opportunities, honestly ranked

(1) Subcutaneous BRIUMVI — the whole ballgame. Phase 3 fully enrolled (announced 15 April 2026), evaluating every-two-month and quarterly regimens against IV, primary endpoint non-inferiority on drug exposure over 24 weeks. Topline year-end 2026 or Q1-2027; filing 2027; targeted launch 2028. Phase 1 data (3 June 2026) showed >60% bioavailability with sustained exposure, described by at least one sell-side observer as a “major de-risking event.” Management frames it as nearly doubling the addressable market at limited incremental cost given ~80% field-force overlap. This is genuinely the largest single value driver and it is unresolved.

(2) ENHANCE single-dose initiation. Phase 3 met its primary endpoint on 27 May 2026. Launch targeted 2027 on approval. Incremental rather than transformative — it removes one infusion visit — but it directly reinforces the operational-convenience positioning that is driving IV share, and market research and advisory-board feedback are reported as strongly positive.

(3) Myasthenia gravis. Phase 1 subcutaneous data (9 June 2026): 82% of subjects achieved a minimal clinically important difference in MG-ADL with a mean >4-point improvement — in n=11. A registration-directed Phase 2 has been initiated with a novel sequential design combining FcRn-inhibitor symptom control with B-cell-depletion disease modification. Real optionality; a single-digit-patient dataset.

(4) Azer-cel in progressive MS. Phase 1 dose escalation, “still early,” with management noting sites are identifying more patients than there are slots. Genuine unmet need; a long way from value.

(5) Treatment-resistant schizophrenia. Phase 2, ~60 patients, initiated July 2026 on emerging evidence of an autoimmune component in a subset. Management called it “early, but… the implications could be significant.” Free optionality; should be valued near zero.

5.4 Verdict

High-quality growth by every reasonable test — organic, volume-driven, recurring, share-taking, and margin-accretive — with one serious qualification: it does not convert to cash. Revenue growth of 223% since Q1-2024 has been accompanied by receivable growth of 503%. On the income statement this is among the best commercial biotech launches of the cycle. On the cash-flow statement the company has never funded itself. Both statements are true simultaneously, and the Financial Quality section takes up the reconciliation.

6. Financial Quality

This is the section where the investment case is decided.

6.1 The income statement is good and improving

Operating leverage is real. Revenue grew 87% in FY2025 while total costs and expenses grew 72%, lifting operating income from $41.9M to $123.3M and operating margin from 12.7% to 20.0%. Gross margin holds at 83–84%. On FY2026 guidance the operating margin should reach roughly 35% before stock-based compensation. There is nothing wrong with this income statement.

6.2 The reported net income is not real

FY2025 net income of $447.2M ($2.77 diluted EPS) includes an income tax benefit of $339.8M. Pretax income was $107.4M. The benefit arises from the release of the deferred-tax-asset valuation allowance, recognised in Q3-2025 (the Q3-25 tax line reads −$364.99M, producing a nonsensical 242% quarterly “net margin”). The CFO stated it plainly on the Q4-25 call: “2025 results including nonrecurring income tax benefit of approximately $340 million, which relates primarily to the release of our deferred tax asset valuation allowance in the third quarter.”

Mechanically, the release means management concluded it is now more likely than not that historical net operating losses will be used against future profits — which is a positive signal about management’s own forecast. But it is a non-cash accounting entry that will never recur, and it capitalises onto the balance sheet as the $347.5M of “other non-current assets” that appeared at Q3-2025.

The consequences for screening are severe and worth stating explicitly:

• The headline P/E of 19.5x is an artefact. So is its 33rd-percentile own-history rank.
• Reported ROE of ~103% is an artefact.
• Anyone comparing TGTX’s “EPS growth” from $0.15 (2024) to $2.77 (2025) is comparing a real number to an accounting release.

On a normalised basis, FY2025 pretax income of $107.4M taxed at ~21% is roughly $85M of economic net income, against a $8.58bn market capitalisation. Cash taxes remain near zero because of the NOLs — the economic benefit of which is real, and is precisely what the DTA capitalises.

6.3 The cash flow has never been positive

Cumulative operating cash flow, 2021–2025: −$568.5M. Cumulative 2023–2025 GAAP net income: +$483.2M, against cumulative operating cash flow of −$96.7M.

The FY2025 reconciliation makes the mechanism unmistakable: $447.2M of net income, less $348.0M of deferred tax (the non-cash release reversing out), plus $64.7M of stock-based compensation, less $176.4M of receivable build and $33.5M of inventory build, plus $55.9M of payables — arriving at −$24.8M.

6.4 The receivable

Source: SEC EDGAR XBRL, us-gaap:AccountsReceivableNetCurrent and us-gaap:RevenueFromContractWithCustomerExcludingAssessedTax, CIK 0001001316. DSO computed as AR ÷ quarterly revenue × 91.25.

Nine consecutive quarters of rising DSO, from 93 to 175 days, with a 30-day step-up in Q1-2026 alone. Receivables of $392.0M equal 42% of the entire FY2026 revenue guide and 2.6x the company’s remaining unrestricted cash balance net of debt.

The 10-K gives the explanation in its own words, and it is not hidden — it sits in the Accounts Receivable accounting policy:

“The Company’s standard payment terms for invoiced amounts typically range between 30 – 60 days, however, extended payment terms have been offered during the BRIUMVI commercial launch. The extended payment terms are meant to align with the timing of reimbursement by government and commercial payers and have not adversely affected the collectability of accounts receivable.”

And in the liquidity discussion, the level of cash required for operations is stated to depend on, among other things, “the timing of collection of receivables from our customers on extended payment terms.”

How to think about this. There are two readings, and honesty requires holding both.

The benign reading, which management asserts and which is plausible: buy-and-bill infusion products create a genuine cash-timing problem for infusion sites, which must purchase drug before payer reimbursement arrives. Extending terms to specialty distributors and, through them, to sites removes a real barrier to adoption for community practices and independent infusion centres — exactly the accounts management says it has been penetrating. On this reading the receivable is a launch-financing investment that converts to cash as growth decelerates: at a stable revenue level, DSO stops rising, the working-capital drag goes to zero, and 84%-margin revenue drops straight to cash flow. The company determined that an allowance for expected credit losses was immaterial at both 2025-12-31 and 2026-03-31, and the counterparties are large, creditworthy specialty distributors, not sites directly.

The sceptical reading, which the data does not exclude: extended credit is a commercial lever, and levers that buy revenue eventually cost what they bought. The concerning feature is not the level but the direction and acceleration: DSO has risen in nine of nine quarters and jumped 30 days in the most recent one, at a point when the launch is three-and-a-half years old and management describes the base as increasingly “repeat” and “predictable.” A maturing recurring-revenue base should be shortening DSO, not lengthening it. If extended terms are still being used to win the marginal account in year four, then some part of the reported 63% growth rate is being purchased.

What would settle it: a quarter in which revenue grows and receivables do not. That has not yet happened.

6.5 Returns on capital

Meaningful ROIC is computable but must be read carefully.

TTM operating income of $149.5M, tax-effected at 21%, gives NOPAT of ~$118M. Invested capital measured as total equity ($583.1M) plus total debt ($765.5M) less cash and investments ($572.8M) is approximately $775.8M → ROIC ≈ 15.2%. Measured on operating assets only — receivables $392.0M, inventory $129.0M, other current assets $51.5M, net fixed assets $6.5M, less payables/accruals $148.8M and deferred revenue $17.2M ≈ $413.0M — ROIC ≈ 28.6%.

Both numbers flatter the business, for a structural reason: the roughly $1.5bn of cumulative R&D and commercialisation spend that created BRIUMVI was expensed as incurred and is absent from invested capital, appearing instead as a $1,062.2M accumulated deficit that reduces book equity. On a cumulative-cash-invested basis — approximately $680M of equity raised in 2020 alone, plus everything before and since — the lifetime return on capital deployed remains, on any honest accounting, close to zero and has only recently turned.

Reported ROE is unusable for FY2025 (the DTA release). On normalised FY2025 net income of ~$85M against average equity of ~$435M, normalised ROE is roughly 20% — respectable, and rising.

6.6 Balance sheet, leverage and liquidity

At 2026-03-31: cash, cash equivalents and investment securities $572.8M (excluding equity investments); receivables $392.0M; inventory $129.0M; total current assets $1,087.0M against current liabilities $187.1M (current ratio 5.8x). Loan payable $745.1M net of $4.9M of issue costs on a $750M face, plus $15.5M of legacy 5% convertible notes at a $1,125 conversion price with no cash repayment obligation. Total equity $583.1M; treasury stock $200.2M; accumulated deficit $1,062.2M.

Net debt is approximately $177M, or ~1.2x FY2026 guided EBIT before SBC — modest on its face. The uncomfortable framing is different: $750M of secured debt, at approximately 9%, on a business that does not generate operating cash. Annual interest of roughly $68M will consume about 21% of FY2026 guided operating income. The facility is secured by a lien on substantially all assets, with a leverage-based pricing grid and amortisation deferred to 2030.

Inventory of $129.0M against TTM COGS of ~$118M is roughly a year of cost of sales — defensible for a biologic with long lead times and a second-source qualification underway, but a further cash absorber. Purchase commitments of approximately $109.1M (2027) and $92.7M (2028) are disclosed.

6.7 Stock-based compensation

SBC rose 52% in FY2025 to $64.7M — more than half of pretax income — split $16.6M to R&D and $48.1M to SG&A. Management’s guided operating expenses (~$350M for 2026) are stated excluding stock-based compensation, so the guided ~35% operating margin should be read as roughly 28% on a fully-loaded GAAP basis. The ratio is declining as a share of revenue, which is the right direction, but the absolute burden is heavy and its concentration is discussed in the Capital Allocation section.

6.8 Verdict

Do economics improve with scale? On the income statement, unambiguously yes. In cash, not yet — and that is the open question on which the investment turns.

Gross margin is stable at 84%, operating margin has trebled in two years, ROIC on operating assets is high, and the balance sheet is liquid. But three-and-a-half years after launch, on a $700M revenue run-rate, this company has never produced positive operating cash flow in any year of its existence, and the gap between reported profit and cash has widened in the most recent quarter. Until receivables stop outrunning revenue, TGTX’s “profitability” is an accrual, and its growth is partly financed by the credit it extends. A business whose economics genuinely improve with scale eventually stops needing external funding. This one has just borrowed $750M at 9%.

7. Capital Allocation

7.1 Sources and uses

Equity. The formative raise was 2020: $679.7M of stock issued at the height of the umbralisib expectation cycle. It funded the failure of that programme and the launch of the next one. Weighted-average basic shares rose from 88.4M (2019) to 146.7M (Q1-2025) — roughly 66% cumulative dilution — before the buyback began reducing the count.

Debt. August 2024: a $250M Initial Term Loan from Blue Owl Capital / HealthCare Royalty, repaying and terminating the prior Hercules facility (which cost $4.6M in extinguishment charges in 2024). March 18, 2026: the Initial Term Loan was repaid in full and replaced by a $750M 2026 Term Loan from Blue Owl, maturing 18 March 2031, with a $250M uncommitted accordion. Pricing opens at SOFR + 4.75% with a 25bp step-down on a leverage threshold; quarterly amortisation of $37.5M begins only in Q1-2030 and is deferrable. It is secured by a lien on substantially all assets. A $9.2M loss on extinguishment ran through Q1-2026.

At ~4.3% SOFR that is roughly a 9.05% coupon — approximately $68M of annual interest. This is non-dilutive capital raised at a moment when the equity was at $30, which is the argument for it; it is also expensive, senior, secured debt on a business with negative operating cash flow, which is the argument against.

Buybacks. Two programmes, both fully executed:

• Prior programme (authorised August 2024, completed September 2025): 3,502,334 shares at an average $28.55.
• 2025 programme (authorised September 2025, executed in Q1-2026): “over 3 million shares… at an average price of roughly $30,” taking treasury stock from $100.2M to $200.2M.
• Cumulative: ~6.8M shares at ~$29, about 5% of shares outstanding. Shares outstanding 153,093,879 at 2026-03-31.

This is good execution, and it should be said plainly. Both tranches were bought between $28.55 and $30 against a stock now at $56.04 — roughly a 93% paper return on approximately $200M of deployed capital, and materially value-accretive to continuing holders. Judged on price paid, it is among the better-timed repurchase programmes in commercial-stage biotech, and it directly contradicts the common pattern of biotechs buying back at highs.

The legitimate objection is funding, not price. These repurchases were made by a company with negative operating cash flow, using proceeds from 9% secured debt. Weiss said so without ambiguity on the Q4-25 call: the Board would act “including adding leverage to reduce our share count.” That is a deliberate, disclosed financial-engineering choice. It works if the equity is genuinely undervalued and the debt is serviceable from future cash flow; it compounds the downside if the subcutaneous programme fails, because the debt is fixed, senior and secured while the equity absorbs the loss.

7.2 Business development

TGTX’s licensing is small, optioned and cheap — a favourable pattern.

• LFB Group (2012): the ublituximab licence. ~$31.0M of cumulative milestones; high-single-digit royalty on net sales. The foundational deal, and by any measure a spectacular one.
• Precision BioSciences (January 2024): exclusive licence to azer-cel for autoimmune/non-oncology, plus non-exclusive manufacturing rights and mid-single-digit royalties. Payments to date: a $2.5M Deferred Precision Stock Payment (January 2025, 220,712 shares) and a $7.5M Milestone 1 payment on 23 February 2026 ($5.25M cash plus $2.25M in 201,504 shares). Note the structure: TGTX buys Precision equity at 200% of the trailing 30-day VWAP, i.e. deliberately at a 100% premium — the January 2025 $2.5M payment was immediately reclassified to equity investments at a fair value of $1.4M, a 44% day-one markdown. That is a subsidy embedded in the deal terms; small in absolute dollars, and worth noting for what it reveals about negotiating posture.
• MaxCyte (February 2025): strategic platform licence.
• Neuraxpharm (July 2023) and Ildong (2012): outbound ex-US commercialisation.

R&D intensity was $160.2M in FY2025 (26% of revenue) and is guided higher in 2026 with $100M of subcutaneous manufacturing and second-source start-up costs running through R&D. SG&A was $232.0M (38% of revenue), driven by field-force expansion and the DTC campaign. The commercial spend is heavy but is producing measurable share gains; the R&D is overwhelmingly line-extension of the same molecule, which is capital-efficient and risk-concentrating in equal measure.

7.3 Insider behaviour

Across the full five-year Form 4 corpus (56 filings):

• Exactly two open-market purchases (code P): Yann Echelard 9,000 shares at $10.64 (6 January 2023) and Michael S. Weiss 100,000 shares at $10.13 (11 August 2023) — ten days after the −49.3% collapse. That is a genuine, discretionary, bottom-tick conviction purchase by the CEO.
• Total sales (code S) of 408,695 shares for ~$8.47M over five years — Power 154,190 at $19.86 avg, Charney 94,500 at $18.75, Lonial 56,785 at $23.55, Weiss 68,220 at $18.80, Echelard 35,000 at $29.20. Almost all cluster in early January, consistent with routine liquidity around restricted-stock vesting.
• The last insider sale of any kind was 24 November 2025 (Echelard, 5,000 shares at $32.57). There has been no insider selling whatsoever in 2026, through a period in which the stock roughly doubled.
• 29 grants (code A) totalling 4,337,992 shares; 11 gifts.

This is a genuinely supportive datapoint and it cuts hard against the cynical reading of the compensation structure. Whatever one thinks of how these executives are paid, they are not selling. Insiders are being enriched by grants, not by distributions into strength.

7.4 Compensation and governance — where this fails

The say-on-pay vote failed. At the 2026 Annual Meeting on 11 June 2026:

39.5% support. Every other proposal passed comfortably. Director support was also notably weak: Sagar Lonial received 54.85% for / 45.15% withheld; Kenneth Hoberman 65.52% / 34.48% withheld. Hoberman chairs the Compensation Committee — and the Compensation Committee comprises all five non-executive directors, meaning the entire independent board is the compensation committee. The board is six members; four have served since 2011–2015.

What shareholders were voting against:

Cumulative three-year CEO compensation: $62.8M. CEO pay ratio 67:1; median employee compensation $306,004.

The structure, not the level, is the objection:

1. A contractual 10x formula. Weiss’s amended employment agreement obliges the Company, in 2022 “and each subsequent year during the term,” to grant restricted shares with a grant-date value equal to ten times the sum of his salary and prior-year cash bonus. This is not a discretionary award the Committee sizes annually against performance; it is a contractual entitlement that mechanically escalates with the bonus it partly determines. The 2025 grant was 750,000 shares worth $23.55M. To management’s credit, vesting is 100% performance-based on TSR exceeding the Nasdaq Biotechnology Index over a 3/5/7/9-year window plus one further year of service — a real and reasonably demanding hurdle, and better than most.
2. A $16.7M market-cap bonus with a date. Weiss is entitled to a $16.7M special cash bonus on the Company achieving a $10bn “Sustained Market Capitalization” — fully diluted market cap maintained for at least 61 days within a 120-day period — by 17 June 2026, also payable on a change of control valuing the Company above $10bn. Fully diluted market capitalisation did not reach $10bn before that date; on 160.06M diluted shares, even the all-time-high close of $59.06 on 9 July 2026 implies ~$9.45bn, and that was after the deadline. No accrual appears in the Q1-2026 balance sheet. This memo makes no allegation about the timing of the 2026 catalyst cluster; the deadline and the price path are both matters of public record and are reported as such.
3. Change-of-control economics. The proxy discloses $176.2M of accelerated equity value to Weiss on a change of control (or on termination without cause, or resignation for good reason), plus cash severance of $3.5–5.25M and the $16.7M bonus if the deal values the Company above $10bn. Full single-trigger acceleration on a change of control is a meaningful transfer of deal consideration away from shareholders.
4. Related-party arrangements. TGTX occupies ~45% of Fortress Biotech’s New York office under a 2014 Office Agreement (~$1.8M estimated average annual rent) and a 2015 Shared Services Agreement ($1.2M / $1.3M / $0.9M in 2025 / 2024 / 2023). Weiss is a director and Executive Vice Chairman of Fortress Biotech and Chairman of Mustang Bio. In February 2026 Fortress sublet the entire space to a third party; TGTX remains obligated for its share of any shortfall between the head lease and sublease income.
5. An announced investigation. On 16 and 21 July 2026, Kuehn Law PLLC issued releases stating it is “investigating whether certain officers and directors of TG Therapeutics, Inc. breached their fiduciary duties to shareholders. The investigation concerns potential self-dealing.” This is a law-firm announcement, not a filed complaint, and no material pending litigation is disclosed in the FY2025 10-K or Q1-2026 10-Q. It should be weighted accordingly — as a signal of shareholder temperature following a failed say-on-pay, not as an adjudicated fact.

Ownership. Weiss beneficially owns 11,544,752 shares (7.54%), of which 5,995,707 are unvested restricted stock and 1,766,666 are exercisable options — implying roughly 3.78M shares held outright, about 2.5% of the company. Directors and officers as a group 8.87%. BlackRock 13.15%, Vanguard 10.19%, State Street 5.29%.

7.5 Verdict

Split, and the split is unusually clean: transactional capital allocation is good; governance is poor and has been formally rejected by shareholders.

On the transactional side, this management team bought back 5% of the company at $29 against a $56 stock, in-licensed the foundational molecule for ~$31M of milestones, keeps its business-development commitments small and optioned, and raised $750M of non-dilutive capital rather than issuing equity at $30. That is a better record than most of the sector.

On the governance side, a contractually guaranteed annual equity grant of ten times pay, a $16.7M market-cap-threshold bonus, $176.2M of single-trigger acceleration, a six-person board where the entire independent complement sits on the compensation committee, related-party dealing with the CEO’s other listed vehicle, and a say-on-pay defeated at 39.5% support constitute a real and recurring claim on shareholder value. It is roughly $25M a year — about 2.7% of the FY2026 revenue guide, and roughly a quarter of normalised pretax income — flowing to one individual under terms the owners have voted down.

The mitigating fact, and it is not small: the insiders are not selling. The CEO bought the 2023 bottom and has not sold a share into a doubling. Whatever the terms of the contract, the alignment of the outcome is intact.

8. Changes and Headwinds — Last Two Years

Commercial (positive). US net product revenue $310.0M (2024) → $594.1M (2025) → guided $885–900M (2026). Twelve consecutive quarters of sequential growth; >25,000 patients prescribed globally; record new-patient starts in Q1-2026; rising treatment-naïve mix; claimed #1 CD20 dynamic share in private practices with infusion capability. Guidance raised from $825–850M (February 2026) to $885–900M (May 2026) — a raise, not a reset, and the second in two quarters.

Clinical (positive, and dense). Six-year ULTIMATE open-label extension data at ECTRIMS (September 2025): ~90% free from 24-week confirmed disability progression, one relapse per 83 patient-years, no new safety signals. Five-year data published in JAMA Neurology (early 2026). ENHANCE Phase 3 met its primary endpoint (27 May 2026) — single Day-1 600mg infusion bioequivalent to the split schedule. Subcutaneous Phase 1 PK/PD supportive with >60% bioavailability (3 June 2026); Phase 3 fully enrolled (15 April 2026), topline year-end 2026 / Q1-2027, launch targeted 2028. Subcutaneous Phase 1 in myasthenia gravis positive (9 June 2026, n=11) with a registration-directed Phase 2 initiated. Azer-cel Phase 1 first patient dosed in progressive MS (August 2025). Phase 2 in treatment-resistant schizophrenia initiated (July 2026). Post-hoc ULTIMATE analysis in treatment-naïve patients published in Frontiers in Immunology (June 2026): 56.7% relapse-rate reduction vs teriflunomide.

Financial (mixed). The DTA valuation-allowance release in Q3-2025 converted a modest profit into a $447M headline. The Blue Owl facility was upsized from $250M to $750M in March 2026 at SOFR+4.75%, secured on substantially all assets, with a $9.2M extinguishment charge. Two $100M buyback programmes completed at ~$29 average. And, continuously and unresolved, operating cash flow remained negative and DSO rose from 109 to 175 days.

Governance (negative). Say-on-pay failed on 11 June 2026 at 39.5% support. Two directors drew 34–45% withheld votes. A shareholder law firm announced a self-dealing investigation in July 2026. Fortress Biotech sublet the shared New York office in February 2026, leaving TGTX contingently liable for its share of any shortfall.

Competitive (negative and building). Roche’s HCP-administered subcutaneous Ocrevus is already competing, with an at-home on-body device in development; management concedes Roche’s “long-term plan” is not to remain in IV. Novartis is extending Kesimpta’s dosing interval. Fenebrutinib is at the agency; remibrutinib follows. Management’s assessment — that BTKs will not “have a material impact on the CD20 class” — is a hypothesis, supported so far by the class’s clinical difficulties, not settled.

Policy (watch). IRA Part B inflation rebates; 340B expansion with hospital site-of-care mix; and an explicit, company-specific “most favored nation” exposure created by having partnered ex-US rights to Neuraxpharm without retaining influence over ex-US price.

Verdict. On balance these developments strengthen the operating thesis and weaken the governance and competitive ones. The clinical and commercial newsflow over the past six months has been genuinely excellent and has de-risked the subcutaneous programme meaningfully — the Phase 1 PK/PD is the single most important positive datapoint of the period. Against that, the company has quadrupled its secured debt, failed a say-on-pay vote, drawn a self-dealing investigation, and watched its receivable balance grow by another $86M in a single quarter. Both sets of facts are large. The market has priced only the first.

9. Risk Analysis

The two that matter most. Risks 1 and 2 are the thesis. If receivables convert and subcutaneous succeeds, most of the rest is noise and today’s price will look reasonable in hindsight. If receivables prove to be purchased revenue or subcutaneous misses, the combination of a 9.5x forward sales multiple, $750M of secured debt and a 2034 exclusivity clock is a genuinely dangerous configuration. Risk 14 is low-probability but is the reason position sizing matters: this security has twice lost roughly nine-tenths of its value.

10. Valuation Discussion — Embedded Expectations

No price target. No recommendation. What follows is what the current price requires one to believe.

10.1 Where the price is

10.2 Two feed artefacts to discard first

Two widely-cited numbers on this name are wrong in ways that matter, and both are recorded rather than silently corrected.

(a) The own-history valuation percentiles are uninformative here. The AZI valuation index at 2026-07-24 reports a P/E of 19.53 at the 33rd percentile of TGTX’s own history, P/S 12.85 at the 18th, P/B 15.38 at the 64th, composite 38th. Read naively this says “cheap versus its own range.” It says nothing of the sort. The P/E denominator is a TTM EPS of $2.87 that is 78% a deferred-tax release — the “cheap” P/E is an accounting artefact. The P/S percentile is computed against a history in which revenue was $6.7M (2021) and $2.8M (2022); of course today’s price-to-sales ranks low against a period when there were no sales. Only the P/B percentile carries any signal, and book value here is itself distorted by the capitalised DTA. The composite should be ignored.

(b) The aggregator enterprise value is struck on a stale price. ROIC.ai reports EV of $5.04bn (diluted $5.63bn) at 2026-03-31 — computed on that date’s $4.80bn market capitalisation, i.e. on a $31 stock. The stock is $56.04. Any multiple pulled from that field without re-striking understates valuation by roughly 80%. Every multiple in the valuation table above is computed from the current price and the 10-Q share count.

10.3 What the price requires — the embedded-expectations bridge

The most useful way to frame $8.76bn is to ask what the existing, approved, de-risked IV franchise is worth, and treat the residual as the market’s payment for what has not happened yet.

The IV franchise, valued as a mature specialty-pharma annuity. Assume the IV business alone reaches a peak of ~$1.5bn in US net revenue in the early 2030s (versus $885–900M guided this year, implying a continued but decelerating ramp), at a sustainable 35% operating margin once commercial spend normalises, taxed at 21%, capitalised at 12x post-tax operating income to reflect a terminal exclusivity date rather than a perpetuity. That gives roughly $5.0bn of enterprise value.

The residual: ~$3.7bn, about 42% of EV. That is what is being paid for:

• subcutaneous BRIUMVI succeeding in Phase 3 and launching in 2028 into a segment representing 35–40% of the class;
• the ENHANCE single-dose label from 2027;
• myasthenia gravis becoming a second indication;
• azer-cel and the schizophrenia programme.

The conclusion is uncomfortable and is the central valuation fact of this memo: the subcutaneous programme is already substantially in the price, before it has read out. A positive Phase 3 at year-end confirms what the market has paid for; a negative one removes roughly the residual. That is a poor risk-reward asymmetry to accept at an all-time high.

10.4 Scenarios

Assumptions are stated explicitly. These are scenario EVs, not targets.

Today’s $8.76bn EV sits at the top of the base case. The bear case is roughly 45% below; the bull case roughly 60% above. The distribution is not obviously unattractive in expectation — but it requires accepting that the most likely outcome is already priced, with the residual return dependent on the bull case.

10.5 The terminal-value problem

Whatever the peak, the hard BPCIA exclusivity wall is December 2034 — approximately 8.4 years from today. Beyond it, value depends on the later-filed 2042 patent family surviving inter partes review and design-around challenge from competitors with effectively unlimited litigation budgets.

A 9.5x forward sales multiple implicitly assumes a 15-to-20-year cash-flow horizon. This asset has a hard 8-to-9-year floor and a contested tail. That mismatch is the single most under-discussed element of the valuation, and it argues that TGTX should trade at a discount to a comparably-growing specialty pharma with an unencumbered franchise — not, as it currently does, at a premium multiple justified by growth alone.

10.6 A note on the comparable set

FactorsToday’s factor-similar peers are XBI, LABU, FDMT, STOK, PTGX, RYTM, CYTK and IDYA — that is, the model still classifies TGTX as a high-beta clinical-stage biotech, not a de-risked commercial specialty pharma. This is informative in both directions. It suggests the tape has not yet re-rated the name to reflect $900M of approved-product revenue (a bull argument for multiple expansion), and that its trading behaviour remains binary-event-driven with 41.6% idiosyncratic volatility (a bear argument about the path). Both are true.

11. Variant Perception

11.1 Consensus

The visible consensus is bullish and momentum-reinforcing. Sell-side and financial-media coverage from the past ninety days runs “Strong Buy,” “flawless execution,” “beat-and-raise trends to continue,” “Phase 1 results are a de-risking event,” and “up 70% and climbing.” Estimate revisions are positive; the Q1 EPS “miss” (17c vs 23c) was ignored in favour of the revenue beat and guidance raise. The consensus view is: a superior anti-CD20 taking share from Roche, with a subcutaneous formulation about to roughly double the addressable market and a management team that guides conservatively and raises.

Most of that is defensible. It is also, on the evidence assembled here, incomplete in three specific ways: it treats a deferred-tax release as earnings, it does not discuss the receivable, and it dates the exclusivity from management’s “into the 2040s” rather than the 10-K’s December 2034.

11.2 The strongest bull case

Stated as forcefully as the evidence permits:

BRIUMVI is a structurally better anti-CD20 for the infusion setting — one hour, twice yearly, with six-year data showing ~90% freedom from confirmed disability progression and no new safety signals — and physicians have voted with their prescriptions twelve quarters running. The business has crossed into real profitability: 84% gross margins, ~35% operating margins on 2026 guidance, and a recurring, compounding patient base with better-than-modelled persistence. The receivable is a launch artefact of buy-and-bill economics that will invert into a large one-time cash release as growth decelerates — at which point an 84%-margin, $1bn+ revenue business converts almost entirely to free cash flow. Subcutaneous BRIUMVI has now been substantially de-risked by Phase 1 PK/PD, is fully enrolled, and would let TGTX compete across 100% of the class rather than 60%, on an existing commercial infrastructure with ~80% field-force overlap — the definition of operating leverage. Management has proven it can execute a launch, guide conservatively and raise, and it repurchased 5% of the company at $29. The CEO bought the 2023 bottom and no insider has sold into a doubling. And the market still classifies this as a clinical-stage biotech, which means the re-rating to specialty pharma has not happened yet.

11.3 The strongest bear case

Equally forcefully:

This company has never generated a dollar of operating cash in its existence, and the gap is widening, not closing — receivables grew $86M in a single quarter to $392M, DSO hit 175 days, and the company’s own filing tells you why: it is selling on extended payment terms. Growth of 63% is not comparable to growth of 63% funded by collections. Meanwhile the “profit” the multiple is computed on is 78% a deferred-tax release. To fund the gap, management has put $750M of 9% secured debt with a lien on substantially all assets onto a business with negative operating cash flow, and used part of it to buy stock. The product’s competitive position is winning the segment the class leader is exiting, while it has zero presence in the segment that is growing — and its entry there is a 2028 event, three years behind Kesimpta and roughly level with an on-body Ocrevus. The hard exclusivity date is December 2034, not the 2040s. And the owners of this business voted down the pay of the man running it by 49 million shares to 32 million six weeks ago, against a contract entitling him to annual equity worth ten times his pay. All of this trades at an all-time high, 9.5x forward sales, in a security with a −90% five-year drawdown and a three-month annualised Sharpe of 8.

11.4 The 3–5 assumptions that actually matter

1. Does the receivable convert? The single highest-information question. Everything about the quality of reported growth and the sustainability of the balance sheet turns on it.
2. Does subcutaneous BRIUMVI hit non-inferiority on exposure? Roughly 42% of enterprise value is riding on it. Phase 1 PK/PD is supportive but Phase 3 is Phase 3.
3. If it hits, what share does it take? Entering a segment against an entrenched Kesimpta and an incoming on-body Ocrevus in 2028 is a very different proposition from “nearly doubling the addressable market.” Access to a market is not share in it.
4. How long is the annuity — 2034 or 2042? Worth a large fraction of the equity value at any sensible discount rate, and currently resolved by management’s framing rather than by the 10-K’s.
5. Does the board respond to the failed say-on-pay? An unchanged 2027 grant under the existing contract would signal the formula is untouchable, and would properly widen the governance discount.

11.5 Falsification evidence

11.6 Positioning read

The factor evidence supports the bear case on timing, not on fundamentals. TGTX sits 5.1% below an all-time high, +88% year-to-date, with a three-month annualised Sharpe of 8.19 and six-month of 4.50 — statistically extreme. Against that: a five-year annualised return of just +8.7%, a five-year maximum drawdown of −90.2%, a ten-year drawdown of −93.2%, and 41.6% idiosyncratic volatility on a 19% R². The tape says the marginal buyer today is buying strength, not value, in a security whose historical distribution is violently fat-tailed. Consensus is not wrong about the company. It is offsides on the price and silent on the cash.

12. Fact vs. Interpretation

| # | Statement | Classification | Source | |:-:|:----------|:-------| | 1 | FY2025 net income of $447.2M includes a $339.8M income-tax benefit from the DTA valuation-allowance release; pretax income was $107.4M | Fact | 10-K FY2025 MD&A; Q4-25 call | | 2 | Operating cash flow was negative in every fiscal year on record, including −$24.8M in FY2025 and −$17.9M in Q1-2026 | Fact | EDGAR XBRL; 10-K; 10-Q | | 3 | Accounts receivable rose from $65.0M (Q1-24) to $392.0M (Q1-26); implied DSO 93 → 175 days | Fact | EDGAR XBRL us-gaap:AccountsReceivableNetCurrent | | 4 | Standard payment terms are 30–60 days; extended terms have been offered during the BRIUMVI launch | Fact (quoted) | 10-K FY2025 & Q1-26 10-Q, AR policy note | | 5 | Some portion of reported revenue growth is effectively financed by the credit TGTX extends | Interpretation | Derived from #3 and #4 | | 6 | The receivable will invert into a one-time cash release as growth decelerates | Assumption (management’s, and the bull case’s) | Q4-25 / Q1-26 calls; not yet observed | | 7 | US BRIUMVI net product revenue: $92.0M (2023), $310.0M (2024), $594.1M (2025); FY26 guided $885–900M | Fact | 10-K FY2025; Q1-26 release | | 8 | FY2023 total revenue of $233.7M included $141.7M of one-off Neuraxpharm licence revenue | Fact | 10-K FY2025 MD&A comparison table | | 9 | BPCIA exclusivity runs to December 2034; earliest composition-of-matter patent expires 2029 US / 2025 ex-US; a later family runs to 2042 | Fact | 10-K FY2025, IP and Patents | | 10 | Management’s “patent protection into the 2040s” is the most favourable available construction and rests on contestable later-filed claims | Interpretation | Q1-26 call vs. #9 | | 11 | Say-on-pay failed on 11 June 2026: 31,905,837 for vs 48,858,169 against (39.5% support) | Fact | 8-K 2026-06-12, Item 5.07 | | 12 | Weiss’s contract entitles him to annual restricted stock worth 10× salary plus prior-year bonus; 2025 total compensation $25,571,250 | Fact | DEF 14A 2026-04-30 | | 13 | Weiss is entitled to a $16.7M cash bonus on a $10bn sustained fully-diluted market cap by 17 June 2026; the threshold was not met by that date | Fact | DEF 14A 2026-04-30; computed from share count and AZI price series | | 14 | The compensation structure represents a recurring claim of ~$25M/yr on shareholder value, ~2.7% of FY26 revenue | Interpretation | Derived from #12 | | 15 | $750M Blue Owl term loan at SOFR+4.75%, maturing 2031, secured on substantially all assets | Fact | 8-K 2026-03-20; 10-Q Q1-26 Note 7 | | 16 | ~$68M of annual interest will consume ~21% of FY2026 guided operating income | Interpretation (arithmetic on #15 and guidance) | Computed | | 17 | ~6.8M shares (~5% of the company) repurchased at ~$29 average against a $56.04 price | Fact | 10-K FY2025; Q1-26 call; 10-Q Q1-26 | | 18 | Exactly two open-market insider purchases in five years; CEO bought 100,000 shares at $10.13 on 2023-08-11; no insider selling in 2026 | Fact | SEC Form 4 corpus, 56 filings | | 19 | Four specialty distributors = 98% of gross product revenue (Q1-26) | Fact | 10-Q Q1-26 Note 2 | | 20 | Enterprise value ≈ $8.76bn = 9.5x FY2026E revenue and ~27x FY2026E operating income before SBC | Fact (computed from primary sources) | Valuation section | | 21 | Roughly 42% of EV is payment for the subcutaneous programme and adjacent optionality | Interpretation | Valuation bridge, stated assumptions | | 22 | BRIUMVI has a regulatory barrier, not an economic moat | Interpretation | Competitive Position, Greenwald tests | | 23 | The stock is 5.1% off an all-time high; 3-month annualised Sharpe 8.19; 5-year max drawdown −90.2% | Fact | AZI price series; FactorsToday leaderboard | | 24 | The AZI own-history valuation percentiles are uninformative for TGTX | Interpretation (with stated mechanism) | Valuation section | | 25 | Kuehn Law announced an investigation into potential self-dealing (16 and 21 July 2026); no complaint has been filed and no material litigation is disclosed | Fact | Kuehn Law releases; 10-K/10-Q legal proceedings |

13. Open Questions

1. What is the actual ageing of the $392.0M receivable? The company discloses concentration by customer and states that credit losses are immaterial, but publishes no ageing schedule. Current vs. 30/60/90/120+ days would settle the central question of this memo in one table.
2. What proportion of gross product revenue is currently sold on extended terms, and how extended? “Extended” relative to 30–60 days could mean 90 days or 180. The blended 175-day DSO implies a very substantial proportion, a very long extension, or both.
3. Is the extended-terms programme being wound down? Management has never been asked about it on a call in the two transcripts reviewed. It has not been raised by a single sell-side analyst.
4. What is the primary endpoint margin in the subcutaneous Phase 3, and what happens if only the every-two-month arm hits? Management has said “both would be a very strong profile” but “we subscribe to less is more.” A two-month-only outcome is a materially weaker competitive position against Kesimpta.
5. How will the board respond to the failed say-on-pay, and can the 10x formula be amended without Weiss’s consent? The 2025 grant was made “with Mr. Weiss’ consent,” implying the Committee needed it.
6. Was the $16.7M market-cap bonus amended, extended or forfeited after 17 June 2026? No accrual appears in Q1-2026 and the deadline has now passed; the next 10-Q or proxy should clarify.
7. What is TGTX’s contingent exposure on the Fortress Biotech office sublease shortfall? Disclosed qualitatively, not quantified.
8. What is the gross-to-net trajectory as hospital-outpatient and 340B mix grows? Guided to ~65% for 2026; the direction beyond that is unaddressed.
9. What are the terms of the $250M uncommitted accordion, and what would trigger drawing it? A further $250M would take secured debt to $1bn.
10. Is there any read-through from the six-year ULTIMATE safety data to long-term hypogammaglobulinaemia and serious-infection risk? Chronic B-cell depletion carries a known cumulative risk profile; six years of data is reassuring but not definitive for a therapy patients may take for decades.

14. What Must Be True

14.1 For the bull case

14.2 For the bear case

15. Source Appendix

See Appendix B below for the full source list with URLs and access dates.

Sections 1–15 contain no recommendation and no price target; the Claude's Take block is the author’s own subjective view. General information only — not investment advice. Every material claim in this article is sourced to a public primary document listed in Appendix B.

APPENDIX A — Standard Diligence Questionnaire

Report date: 2026-07-25 · Price: $56.04 (2026-07-24 close) A supplemental due-diligence questionnaire covering the standard checklist. Answers are labelled Fact / Interpretation / Assumption where the distinction matters.

General

What thoughtful questions have other investors asked about this company?

From the two most recent earnings calls, the sell-side (Goldman Sachs, JPMorgan, TD Cowen, Jefferies, Evercore ISI, H.C. Wainwright, Cantor Fitzgerald, B. Riley) has concentrated on five themes: (i) the size and accessibility of the subcutaneous opportunity and whether every-two-month versus quarterly dosing is commercially decisive; (ii) the sustainability of persistence and the new-versus-repeat patient mix; (iii) gross-to-net dynamics and Q1 seasonality; (iv) capital allocation between buybacks and business development; and (v) the competitive threat from BTK inhibitors (fenebrutinib, remibrutinib) and from Roche’s at-home on-body Ocrevus device. One analyst (Cantor) pushed usefully on the disconnect between “record new patient starts” and guidance implying modest new-start growth — the only genuinely adversarial question asked in either call.

The most important question no investor has asked: in two full transcripts, across roughly a dozen analysts, not one question was asked about accounts receivable, days sales outstanding, extended payment terms, or the fact that operating cash flow has been negative in every year of the company’s existence. (Fact — transcript review.) That is the central analytical gap in the consensus view and, in this analyst’s judgement, the reason the variant perception exists at all. (Interpretation.)

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Neither — TGTX is not cyclical. It is at an early-franchise-life ramp point: FY2026 will be the fourth full year of a launch that is still compounding, with operating margin having gone 8.8% → 12.7% → 20.0% → ~35% (guided, ex-SBC). Earnings are at a structural low relative to franchise potential and a structural high relative to history. (Fact/Interpretation.)

Driven by the external environment or internal actions? Overwhelmingly internal: field-force expansion, DTC investment, formulary and site-of-care penetration, and clinical-data generation. The one large external tailwind is the secular shift of the MS market toward high-efficacy anti-CD20 therapy, visible independently in Biogen’s disclosed MS-franchise decline. (Fact/Interpretation.)

How stable are revenues? Structurally stable and improving in stability. BRIUMVI is dosed every 24 weeks with reported persistence better than modelled at weeks 48/72/96; each new cohort layers onto a recurring base. Management describes an increasingly “repeat” and “predictable” revenue mix. The offsetting instability is gross-to-net variability (guided ~65% for 2026, seasonally worse in Q1 on deductible resets and co-pay-programme utilisation) and site-of-care mix shifts affecting 340B exposure. (Fact.)

Outlook for products/services? IV BRIUMVI: continued but decelerating growth (+237% → +92% → +63% → ~+50% guided). ENHANCE single-dose initiation is incrementally supportive from 2027 on approval. Subcutaneous BRIUMVI is the swing factor — Phase 3 fully enrolled, topline YE2026/Q1-2027, targeted 2028 launch. Myasthenia gravis, azer-cel and treatment-resistant schizophrenia are early-stage optionality. (Fact.)

How big will this market be — growing, shrinking, domestic or international? The anti-CD20 MS class is growing in both volume and share of the MS market as legacy platform and oral therapies decline. Management sizes the patient-administered subcutaneous segment at 35–40% of the class, with a view that “over time… that space can expand.” TGTX’s economics are overwhelmingly domestic: US BRIUMVI was $594.1M of $616.3M FY2025 revenue (96%). Ex-US is licensed to Neuraxpharm (Europe and beyond) and Ildong (Korea/SE Asia), so international upside accrues to TGTX only as supply margin, milestones and royalties — and the company correspondingly has no control over ex-US price, which the 10-K flags as an aggravating factor under “most favored nation” reference-pricing proposals. (Fact.)

Business Quality & Competitive Moat

Is the industry getting more or less competitive? More. Roche is defending with a healthcare-provider-administered subcutaneous Ocrevus and an at-home on-body device in development; Novartis is extending Kesimpta’s dosing interval; two BTK inhibitors (fenebrutinib, remibrutinib) are in late regulatory/clinical stages; anti-CD40L agents are in development. TGTX’s own subcutaneous programme is itself capital entering the segment. (Fact — 10-K Competition section and management commentary.)

How profitable is the business (ROIC, ROE)? TTM operating income $149.5M on $700.3M of revenue (21.3%). NOPAT at 21% ≈ $118M. Against invested capital of ~$775.8M (equity + debt − cash), ROIC ≈ 15.2%; against operating assets only (~$413.0M), ROIC ≈ 28.6%. Reported FY2025 ROE of ~103% is meaningless — the DTA release. Normalised FY2025 ROE on ~$85M of economic net income against ~$435M of average equity is roughly 20%. All these figures are flattered because ~$1.5bn of cumulative R&D that created the asset was expensed and sits as a $1,062.2M accumulated deficit rather than as invested capital. (Fact/Interpretation.)

How profitable is the industry — how many competitors, what barriers to entry? Highly profitable: 84% gross margins are the class norm, not a TGTX distinction. Three molecules, two of them owned by Roche and Novartis. Barriers to copying are very high (BPCIA 12-year exclusivity, biologics manufacturing complexity); barriers to competing are low (a differentiated formulation or dosing schedule is a fast, cheap entry — as TGTX itself demonstrated by taking leading IV private-practice share in three years). (Fact/Interpretation.)

Can the business be easily understood? Yes — unusually so for biotech. One approved molecule, one indication, one geography of economic consequence, one distribution channel, one large binary catalyst. The complexity is in the working capital, not the business model. (Interpretation.)

Can it be undermined by foreign low-cost labour? No. Not a labour-cost-exposed business. The relevant analogue is biosimilar entry, which is legally barred in the US until December 2034. (Fact.)

Do brands matter? Moderately, and in a specific way. In physician-administered specialty pharmaceuticals the “brand” is the aggregate of clinical data, safety experience and operational reliability. BRIUMVI’s six-year ULTIMATE extension data (~90% free from 24-week confirmed disability progression; one relapse per 83 patient-years) is the brand. TGTX’s DTC investment — including a Super Bowl LX launch of “Next In MS” with Christina Applegate — is an explicit attempt to build consumer brand equity in a shared-decision-making market. Whether that produces durable pricing power is unproven. (Fact/Interpretation.)

What is the nature of competition? Convenience and operational simplicity, layered on broadly comparable efficacy. BRIUMVI wins on a one-hour infusion twice yearly; ENHANCE removes one initiation visit; Kesimpta wins on no infusion at all; Ocrevus wins on incumbency, scale and an on-body device roadmap. Price is not the primary battleground — which is favourable while it lasts. (Fact/Interpretation.)

Customers’ switching costs? Real but modest. Once a patient is established on 24-weekly infusions at a given site with benefits verified, physicians are reluctant to disturb a stable regimen — hence the strong persistence. But TGTX’s own success in switching patients off Ocrevus demonstrates the switching cost is far from prohibitive. Note the direct evidence of low customer stickiness at the account level: extended payment terms were required to win adoption, which is a form of price/credit concession, not a switching-cost moat. (Interpretation.)

Financial Condition & Balance Sheet

Assets not fully recognised on the balance sheet? Yes, materially. The BRIUMVI franchise itself — approximately $1.5bn of cumulative R&D and launch expense — is entirely expensed, so a business with a ~$8.6bn market capitalisation carries $583.1M of book equity and a $1,062.2M accumulated deficit. The pipeline (subcutaneous, MG, schizophrenia, azer-cel) carries no balance-sheet value. Remaining NOLs beyond the $347.5M capitalised as a deferred tax asset have economic value. (Fact/Interpretation.)

Off-balance-sheet liabilities? Three worth naming. (i) Purchase commitments of approximately $109.1M (2027) and $92.7M (2028), described as not representing the Company’s entire anticipated purchase requirements. (ii) Contingent milestone and royalty obligations — high-single-digit royalties to LFB Group on all ublituximab net sales in perpetuity, mid-single-digit royalties and future milestones to Precision on azer-cel. (iii) The Fortress Biotech Office Agreement: Fortress sublet the entire New York space to a third party in February 2026, but TGTX “may be required to fund its proportionate share of any shortfall between the head lease obligations and sublease income” — disclosed but not quantified. (Fact.)

How conservative is the accounting? Broadly conservative on the expense side — R&D fully expensed, including ~$100M of 2026 subcutaneous manufacturing and second-source start-up costs that will produce saleable inventory with little associated future COGS (a genuine future margin tailwind, and a conservative present treatment). Aggressive, or at minimum optimistic, on the revenue and receivable side: revenue is recognised on shipment to specialty distributors on extended payment terms, with no financing component imputed (justified on the basis that expected time to payment is under one year), and no expected-credit-loss allowance recognised on a $392.0M receivable balance at 175 days DSO. The auditor is KPMG LLP; ratification passed with 99.4% support and no adverse ICFR opinion or late filings appear in the corpus. (Fact/Interpretation.)

How CapEx-hungry is the business? Almost not at all — capital expenditure was $214,000 in FY2025 and $45,000 in FY2024, with net fixed assets of $6.5M. Manufacturing is outsourced. But this is precisely the wrong metric for TGTX. The business is not capex-hungry, it is working-capital-hungry: FY2025 absorbed $176.4M into receivables and $33.5M into inventory. The correct statement is that TGTX requires roughly 45–55 cents of incremental working capital for every incremental dollar of revenue — which is why a 20%-operating-margin business generates negative cash. (Fact/Interpretation.)

Capital Allocation & Management

How much FCF does the business generate, how does management use it, what is the philosophy? It generates none. Free cash flow was −$25.0M in FY2025 and has been negative in every year on record; cumulative 2021–2025 operating cash flow is −$568.5M. Management’s philosophy, stated on the Q1-26 call, is to “deploy capital where we see the best risk-adjusted long-term return, whether that’s in the business, repurchasing shares or pursuing external opportunities.” In practice, the funding has come from equity ($679.7M in 2020) and, since 2024, from secured debt — with $100M of the March 2026 $750M facility recycled into buybacks in Q1-2026. Weiss was explicit on the Q4-25 call about “adding leverage to reduce our share count.” (Fact.)

Significant acquisitions recently? No acquisitions. Licensing only, and small: Precision BioSciences (azer-cel, January 2024 — $2.5M deferred stock payment plus a $7.5M Milestone 1 payment in February 2026), MaxCyte platform licence (February 2025). Note the structure of the Precision equity payments — struck at 200% of the trailing 30-day VWAP, i.e. a deliberate 100% premium; the January 2025 $2.5M payment was reclassified to equity investments at $1.4M fair value on the payment date, a 44% day-one markdown. Small in absolute terms; informative about negotiating posture. (Fact/Interpretation.)

Buying back shares? Yes, and well. Two $100M programmes fully executed: 3,502,334 shares at $28.55 average (completed September 2025) and “over 3 million shares… at roughly $30” in Q1-2026. Cumulative ~6.8M shares at ~$29, about 5% of shares outstanding, against a current $56.04. On price paid this is materially value-accretive and among the better-executed repurchase programmes in commercial-stage biotech. The objection is funding — negative operating cash flow, financed by 9% secured debt. (Fact/Interpretation.)

Issuing large amounts of new shares to insiders? Yes. 29 code-A grants totalling 4,337,992 shares across the five-year Form 4 corpus. Stock-based compensation was $64.7M in FY2025 (10.5% of revenue), up 52% year-on-year, of which $48.1M sat in SG&A. Weiss’s 2025 grant alone was 750,000 restricted shares ($23.55M grant-date fair value). Weiss beneficially owns 11,544,752 shares (7.54%), of which 5,995,707 are unvested restricted stock and 1,766,666 are exercisable options. (Fact.)

Compensation policy of directors/management? This is the clearest failing in the file, and shareholders have said so. The advisory say-on-pay vote FAILED at the 11 June 2026 annual meeting: 31,905,837 for versus 48,858,169 against — 39.5% support. Every other proposal passed. Comp-committee chair Kenneth Hoberman drew 34.5% withheld votes; director Sagar Lonial drew 45.2%. The board is six members; all five non-executive directors sit on the compensation committee, so there is no independent sub-set reviewing pay. Weiss’s contract obliges the Company to grant him restricted shares annually worth ten times the sum of salary and prior-year cash bonus; 2025 total compensation was $25,571,250 (three-year cumulative $62.8M) on a company with $616M of revenue and negative operating cash flow; a $16.7M special cash bonus was payable on a $10bn sustained fully-diluted market cap by 17 June 2026 (not achieved by that date); and $176.2M of equity accelerates on a change of control. In mitigation, the annual grant is 100% performance-vesting on TSR exceeding the Nasdaq Biotechnology Index over 3/5/7/9 years plus a further service year — a genuinely demanding hurdle, better than most peers. (Fact/Interpretation.)

Motivations of management? Mixed, and the evidence pulls both ways honestly. Aligned: Weiss bought 100,000 shares in the open market at $10.13 ten days after the August 2023 collapse — one of only two open-market purchases in the entire five-year corpus — and no insider has sold a single share in 2026 while the stock roughly doubled. The buyback was executed at $29. Misaligned: a contractual pay formula the owners have voted down, a market-cap-threshold cash bonus with a hard date, single-trigger acceleration, related-party arrangements with Fortress Biotech (where Weiss is Executive Vice Chairman) covering office space and shared services, and a July 2026 shareholder-firm investigation into “potential self-dealing.” The fair summary: management is genuinely invested in the outcome and simultaneously extracting an unusually large and contractually protected share of it. (Fact/Interpretation.)

Valuation & Market Data

Is the stock an ADR, MLP, or K-1 issuer? No. TG Therapeutics, Inc. is a Delaware C-corporation listed on the Nasdaq Capital Market under “TGTX.” Ordinary Form 1099 treatment; no K-1. (Fact.)

Dividend policy? None. TGTX has never paid a dividend, and the $750M Blue Owl financing agreement contains customary covenants; capital return is via buyback only. Trailing annual dividend yield 0%. (Fact.)

How profitable is the business? FY2025: gross margin 83.7%, operating margin 20.0%, pretax margin 17.4%. Reported net margin of 72.6% is an artefact of the $339.8M DTA release and should be disregarded. TTM: operating margin 21.3%. FY2026 guidance implies roughly 35% operating margin before stock-based compensation, or roughly 28% fully loaded. (Fact.)

Is net income diverging from cash from operations? Yes, and this is the single most important finding in the file. FY2025: net income +$447.2M against operating cash flow of −$24.8M — a $472.0M divergence, of which $348.0M is the non-cash deferred-tax release and $209.9M is the receivable and inventory build. Cumulative 2023–2025: GAAP net income +$483.2M against operating cash flow of −$96.7M. Q1-2026 continued the pattern: +$19.8M of net income, −$17.9M of operating cash flow. Accounts receivable have risen from $65.0M (Q1-2024) to $392.0M (Q1-2026) — implied DSO 93 → 175 days, rising in nine consecutive quarters. The 10-K’s explanation is that “extended payment terms have been offered during the BRIUMVI commercial launch.” (Fact.)

Risks & Downside

What factors would cause the stock to decline? In rough order of expected impact: (1) a subcutaneous Phase 3 miss on exposure non-inferiority at year-end 2026 / Q1-2027 — roughly 42% of enterprise value is attributable to that programme and adjacent optionality; (2) any deterioration in receivable quality — a credit-loss allowance, expanded extended-terms disclosure, or DSO through 200 days; (3) US revenue growth decelerating below ~25% year-on-year while receivables continue to build; (4) an adverse IPR outcome on the 2042 patent family, shortening the valuation horizon toward the December 2034 BPCIA floor; (5) Roche’s on-body Ocrevus launching ahead of subcutaneous BRIUMVI; (6) an approved BTK inhibitor with a clean benefit-risk profile; (7) escalation of the governance situation into filed litigation or a proxy contest; (8) simple momentum reversal — the stock is 5.1% off an all-time high with a three-month annualised Sharpe of 8.19. (Interpretation.)

Risk of a catastrophic loss? Yes, and it is not theoretical for this issuer. TGTX lost 89% of its value between November 2021 and June 2022 ($34.99 → $3.74) when umbralisib’s benefit-risk failed at the FDA and the product was withdrawn. The five-year maximum drawdown is −90.2% and the ten-year is −93.2%; idiosyncratic volatility is 41.6% annualised on a 19% factor R². The plausible catastrophic mechanism today is the same one: a post-approval safety signal in a chronically dosed B-cell-depleting immunosuppressant — cumulative hypogammaglobulinaemia, serious or opportunistic infection, or PML — emerging as the treated population moves past six years of continuous exposure. Six-year ULTIMATE extension data show no new safety signals, which is reassuring but not dispositive for a therapy patients may take for decades. Secondary catastrophic paths: a manufacturing failure at a sole-source supplier, or a receivable write-down of a size that triggers the leverage covenants on $750M of secured debt. (Fact/Interpretation.)

Chance of a total loss? Low. There is an approved, growing, 84%-gross-margin product generating $900M of revenue with hard regulatory exclusivity to December 2034, $572.8M of cash and investments, and $392.0M of receivables from large creditworthy distributors. Equity would be wiped out only by a catastrophic safety withdrawal combined with $750M of senior secured debt — a real but low-probability compound event. Note that the debt is secured by a lien on substantially all assets, so in a severe adverse scenario the recovery waterfall favours Blue Owl decisively over equity. (Interpretation.)

Recent News & Events

Has the business environment changed recently? Yes, in both directions, and materially — the past twelve weeks have been the densest period of newsflow in the company’s post-launch history.

Positive: Q1-2026 revenue of $205M against $185–190M guidance with FY26 US guidance raised from $825–850M to $885–900M (6 May); Phase 3 ENHANCE met its primary endpoint, supporting a single Day-1 600mg initiation (27 May); positive Phase 1 subcutaneous BRIUMVI PK/PD with >60% bioavailability supporting the quarterly regimen (3 June); positive Phase 1 subcutaneous BRIUMVI in myasthenia gravis with a registration-directed Phase 2 initiated (9 June); Phase 2 in treatment-resistant schizophrenia initiated (6 July); subcutaneous Phase 3 fully enrolled (15 April).

Negative: the say-on-pay vote failed (11 June); Kuehn Law announced an investigation into “potential self-dealing” by officers and directors (16 and 21 July).

Structural: the Blue Owl facility was upsized from $250M to $750M at SOFR+4.75%, secured on substantially all assets (18 March).

Significant acquisitions? None. See Capital Allocation above.

Change in accounting policies? No change in accounting policy, but a material change in accounting estimate with a very large earnings effect: the release of the deferred-tax-asset valuation allowance in Q3-2025, producing a $339.8M income-tax benefit and $347.5M of capitalised deferred tax assets. Anyone comparing FY2025 EPS of $2.77 to FY2024’s $0.15 is comparing a real number to an accounting release. (Fact.)

Recent changes — new markets, facilities, management? No change to the executive team (Weiss CEO, Sean Power CFO, Adam Waldman CCO, Jenna Bosco CCO-Communications) and no change to the six-person board, which stood for re-election in June 2026 and was returned in full despite two directors drawing 34–45% withheld votes. New segment entry is prospective, not actual: the patient-administered subcutaneous segment (2028 target) and new indications (myasthenia gravis, schizophrenia, progressive MS). On facilities: Fortress Biotech sublet the entire shared New York office to a third party in February 2026, reducing TGTX’s net rent expense prospectively while leaving it contingently liable for its share of any shortfall; a second contract manufacturer is being qualified at roughly $100M of 2026 expense running through R&D. (Fact.)

APPENDIX B — Source Appendix

Report date: 2026-07-25. All sources accessed 2026-07-25 unless otherwise stated. Primary sources are listed first. Every material claim in this article traces to one of the public documents below. All SEC filings are freely available on EDGAR at https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001001316.

A. SEC filings — primary (CIK 0001001316)

The trailing 60-month filing corpus (172 filings since 2021-07-25) was enumerated from EDGAR and read in full: 74 primary documents plus 56 Form 4/5 ownership filings retrieved as raw XML. Form breakdown: 55 Form 4, 35 8-K, 17 Form 144, 16 SC 13G/D, 15 10-Q, 6 SCHEDULE, 5 DEF 14A, 5 10-K, 4 DEFA14A, 3 PRE 14A, 2 S-8, 2 S-3ASR, 2 ARS, 1 Form 5, 1 10-K/A.

SEC EDGAR XBRL company-concept API (https://data.sec.gov/api/xbrl/companyconcept/CIK0001001316/us-gaap/{tag}.json), used as the authoritative source for the multi-period series in the Financial Quality section:

• us-gaap:AccountsReceivableNetCurrent — the DSO series
• us-gaap:RevenueFromContractWithCustomerExcludingAssessedTax — quarterly and annual revenue
• us-gaap:NetCashProvidedByUsedInOperatingActivities — the operating-cash-flow series

B. Management commentary — earnings-call transcripts

Retrieved via the ROIC.ai MCP transcript tools (list_earnings_calls, get_earnings_call_transcript). Treated throughout as hypothesis, not evidence, and validated against filings.

C. Company press releases (public, primary)

D. Third-party quantitative data feeds

All third-party aggregated data; not primary. Every material figure reconciled to the filings. Where a feed and a filing disagreed, the filing governs and the discrepancy is noted.

E. Analytical frameworks

F. Press and secondary sources

Used for event triage and consensus characterisation only; every material claim traced to a primary source.

G. Reproducibility

Every figure in this article can be reproduced from public sources without paid data: