Royalties Unstacked: Allocating the Proceeds in Combination Therapies

Disclaimer: This article is for informational purposes only and does not constitute legal, financial, or investment advice. The author is not a lawyer or financial adviser. All information is derived from publicly available sources and may not be complete or current. Details regarding transactions, royalty structures, and financial arrangements may change. Readers should conduct their own due diligence and consult with appropriate legal and financial professionals before making any decisions.

Allocating royalties from a successful drug is usually straightforward – until that drug is part of a duet or trio. In today's pharmaceutical landscape, therapies increasingly come as combinations: a blockbuster antibody paired with a chemotherapy, a drug delivered via a device, or a vaccine boosted by a proprietary adjuvant. These multi-component regimens promise enhanced efficacy, but they also pose a vexing question: how should multiple patent owners or licensors split the spoils?

Like coworkers dividing credit (and bonuses) for a joint project, companies must carefully script royalty-sharing arrangements to avoid discord. The challenge is both technical and legal, requiring precision. This report delves into the complexities of royalty allocation for combination products – from contract clauses and real-world case studies in 2024–2025 to litigation precedents across the U.S., EU, and Japan – and examines how terms like "attributable net sales," "allocable portion," and "dilution clause" come into play.

The Royalty Splitting Challenge in Combination Products

What counts as a "combination product"? Broadly, it includes any therapeutic sold or used with another active component for combined effect. It could be a fixed-dose combination (FDC) pill containing two drugs, a co-formulated biologic with an enabling enzyme, a drug-device hybrid (like a drug-eluting stent or inhaler), or simply a co-packaged/co-prescribed regimen. For example, Merck's Keytruda (pembrolizumab) is often prescribed with chemotherapy as a regimen for lung cancer, and vaccines are frequently combined with adjuvants to boost immune response.

Regulators categorize single-entity combinations (e.g. an antibody-drug conjugate, or a prefilled injector pen) and co-packaged combos (e.g. a syringe kit with drug and device) similarly in principle, but the key issue for us is economic: when multiple inventions or products unite, how do we apportion the resulting sales for royalty purposes?

From the licensor's perspective, if their patented component is a key driver of the combination's sales, they naturally argue for royalties on the entire combination's revenue. In their view, the whole "pie" exists thanks to their ingredient, so why settle for a sliver? Conversely, the licensee (or payor) insists that royalties should only apply to the licensed portion of the combination, not the full sales price. Why pay a royalty on a $100,000 combo therapy if the licensed drug is only half of it? This fundamental tension means that combination product royalty clauses are among the most intricately negotiated provisions in pharma licensing agreements.

Contractual Strategies: "Attributable Net Sales" and Royalty Allocation

The industry has converged on a few formulaic approaches to split revenues when products are combined. Most license agreements now contain a "Combination Product" clause defining how to calculate Net Sales attributable to the licensed component. A common method is to apply a proportional allocation formula based on the relative value of each component:

Relative Price Fraction: If each component of the combo is sold separately, the sales are prorated by the ratio of the standalone price of the licensed product to the sum of standalone prices. In formula form, royalty-bearing net sales = Total Combo Sales × A/(A+B), where A is the market price of the licensed product alone and B is the price of the other component(s). For instance, if Drug A (licensed) costs $50k/year and Drug B $50k, then only ~50% of a $100k combo's revenue would count toward Drug A's royalty. This A/(A+B) rule is considered the gold standard when separate pricing data exists.

Single-Component Known: If one component's standalone price is known but the other's is not (e.g. a proprietary adjuvant never sold alone), contracts use one-sided formulae. One approach is A/C, where C is the combo price – effectively assuming the unknown component accounts for the remainder. Alternatively, the formula can be expressed as 1 – (B/C) if the "B" component's price is known but "A" is not. Both yield a similar result: allocate the known part's proportion and implicitly assign the rest of the value to the unknown part.

No Separate Pricing: If neither component is sold separately (common for novel fixed-dose combos), agreements resort to good-faith negotiation or preset proxies. Some contracts specify that the parties will mutually agree on an "allocable portion" or use an expert determination of relative value. For example, a license might stipulate that if standalone prices are unavailable, the companies will negotiate a fair percentage of combo sales attributable to the licensed drug (e.g. based on manufacturing cost or contribution to efficacy).

The 2007 Boehringer–Vitae collaboration, for instance, provided that when neither component's price could be determined, "Net Sales of the Product [in the combo] shall be…a mutually agreed percentage", with unresolved disputes sent to arbitration. This prevents stalemate and avoids indefinitely fuzzy terms like "to be determined later," which both sides dislike.

Sources: Contract filings and agreements from SEC, Law Insider, and public filings.

Despite the complex formulas, the goal is simple: determine an "attributable net sales" amount for the licensed technology. In essence, parties attempt to apportion the combo's revenue in proportion to each component's contribution or value. Well-drafted clauses also clarify that no double-counting occurs – e.g. if multiple patents cover the same combo, only one royalty is paid on a given sale.

Royalty Stacking and Dilution Clauses

Combination products often implicate multiple licenses. A company marketing a combo pill might owe royalties to Licensor A for Drug X and to Licensor B for Drug Y, plus perhaps a fee to a device maker. Stacking too many royalties on a single product can make it commercially unviable, a situation licensees dread. To mitigate this, agreements include royalty-stacking (anti-stacking) provisions. These clauses typically allow the licensee to reduce the royalty rate if it must pay third-party royalties on the product.

For example, a license might state that for each 1% of net sales paid to a third-party due to necessary IP licenses, the royalty to the licensor is reduced by 0.5%, up to some cap.

Licensors, of course, seek to limit the scope of stacking reductions. Common compromises include a floor royalty rate (e.g. the royalty can't drop below half of the original rate) or permitting reductions only for certain kinds of third-party IP (e.g. only if needed to avoid infringing a patent to make the combo, and not for optional add-ons). A Bird & Bird analysis notes that while licensees often negotiate for broad stacking rights, licensors will tightly define when such reductions apply (such as only if a necessary license is obtained).

They may also require the licensee to split the pain, not dollar-for-dollar but proportionally. In practice, "dilution clauses" ensure the original licensor still gets some minimum royalty even if the product's "royalty stack" grows. For instance, Northwest Bio's 2024 in-license of a cancer vaccine reported a 4% base royalty, "potentially reduced to 3% in the event of royalty stacking" – a classic 25% reduction floor.

Royalty-stacking provisions have a twin purpose: preserving the licensee's incentive to combine therapies (so they aren't discouraged by exorbitant cumulative royalties), while ensuring each licensor gets a fair share rather than being completely diluted out. Without such clauses, a combination therapy that requires multiple IP licenses could end up burdened with an unsustainable royalty burden (imagine layering 5% + 8% + 10% on sales – the profit margin would vanish).

As one licensing survey succinctly put it, "the presence of generic or third-party royalty obligations" is routinely accounted for via royalty reduction mechanisms. In summary, stacking clauses are a pragmatic tool to balance innovation-sharing with economic reality.

Modalities and Geographies: One Size Does Not Fit All

Different therapeutic modalities and regions have developed their own twists on royalty allocation:

Biologic + Small Molecule Regimens

When a large-molecule biologic (like an antibody) is combined with a small-molecule drug, often each component is sold separately (as in Keytruda + chemotherapy). In such cases, each drug's sales are tracked individually, and a licensor of one drug would typically collect royalties on its drug's net sales only, not on the partner drug's sales.

For instance, if a biotech licensed a patent covering the use of Keytruda in combination with chemo, the license might specify that royalties are due on Keytruda's sales (since Keytruda is the licensed product) and exclude the chemo's revenue – effectively treating the regimen's sales as two pots. However, if the combination itself is the invention (say a patent on "Drug A + Drug B" as a method), royalty calculations get thornier. Contracts might then define an "allocable portion of Combination Net Sales" to Drug A, using the formulas discussed above.

In practice, many immuno-oncology deals simply avoid this ambiguity by structuring co-development collaborations or profit-shares rather than cross-royalties. It is notable that in Japan, companies like Ono Pharmaceuticals and BMS resolved overlapping rights to PD-1 antibody therapies by a settlement that gave Ono a percentage of worldwide sales of Opdivo/Keytruda, implicitly treating the licensed component as driving the combo therapy's value – a lump-sum and running royalty arrangement rather than a strict formula. Such deals underscore that if one component is seen as the primary driver, parties may opt to share total combo revenue in set proportions (effectively bypassing formula minutiae).

Drug–Device Combinations

These include inhalers with drug formulations, insulin pumps with insulin, drug-eluting implants, etc. A famous example is the dry-powder inhaler combining GSK's drugs with Vectura's patented particle technology. When Vectura's inhaler formulation patent was found infringed in GSK's combo products, the court had to decide a reasonable royalty. GSK argued for paying only on the patented portion of the inhaler, not the full inhaler+drug sales.

But the Federal Circuit in Vectura v. GSK (2020) upheld a 3% royalty on the entire inhaler's sales, noting that apportionment was effectively "built into" that rate given the prior license between the parties. In other words, because 3% had been the agreed royalty for the combination product in an earlier deal, it was implicitly calibrated to the tech's contribution.

Generally, for drug-device combos, licenses often treat the device or delivery mechanism as the licensed product, especially if it's proprietary (e.g. an auto-injector technology). So the royalty base might be the full combo price unless the drug component is also separately licensed. Some agreements explicitly exclude the device's cost – for instance, a transdermal patch license might say net sales exclude the value of the patch hardware if the patch tech is not part of the licensed IP.

The FDA's examples illustrate the range of drug-device combos that complicate matters: "prefilled injector pens, metered dose inhalers, drug-coated stents", etc., each requiring careful contract definitions. In Europe, one twist is the patent system: drug-device combos can involve Supplementary Protection Certificates (SPCs). A recent 2024 CJEU decision clarified that a combination of active A + active B can be eligible for an SPC extension if the basic patent claims that combo (e.g. Merck's sitagliptin + metformin combo was deemed protectable).

This means in the EU a patent on a combo can extend market exclusivity, thus prolonging royalty obligations on that combo product. Companies eyeing EU markets must therefore consider how royalty periods may be extended for combination products via SPCs – effectively geography-based royalty term differences.

Vaccines and Biologic Mixtures

Combination vaccines (e.g. MMR, or a COVID shot with multiple antigen targets) often bundle several biological actives. If different parties own each component, royalty-sharing can resemble a pooling arrangement. One real-world case: Shingrix, a shingles vaccine, uses a QS-21 adjuvant licensed from Agenus. Rather than split the price, GSK simply pays Agenus a royalty on Shingrix's total net sales (reported to be in the mid-single digits), recognizing that the adjuvant is critical but inseparable from the vaccine's value.

In contrast, for multi-antigen vaccines developed via public-private partnerships, contracts (especially in global health) sometimes assign a token share of net sales to each antigen's licensor, or set a reduced royalty if combined with another widely-licensed antigen (to avoid "double royalties" raising the cost in low-income markets). Geographically, emerging market licenses (like Gilead's voluntary licenses for HIV and HCV drugs) explicitly address combinations: they often cap the total royalty generics must pay if they combine a Gilead drug with another licensed drug, or they prorate "attributable Net Sales" of a combo based on each licensed API's share. This ensures affordable pricing goals are met by preventing royalty stacking across multiple license agreements.

Japan and Other Regions

Japanese pharmaceutical contracts typically follow the global norms on combination proration and stacking, but one notable aspect is the influence of local practice and patent law. A Japanese company might be more inclined to form a co-ownership joint venture for a combination (as Shionogi did with ViiV for HIV drugs) rather than rely on ongoing royalties.

In that Shionogi-ViiV deal restructuring, Shionogi opted for a 10% equity stake plus royalties "averaging in the high-teens" on the integrase inhibitor portfolio. Those royalties apply to any product containing the integrase, whether alone or in combination, effectively giving Shionogi a cut of combo drug sales without needing a price-apportionment formula. This approach – a broad royalty on full combination sales – can simplify matters, though it relies on strong bargaining power.

In the EU, competition law also lurks in the background: license terms that might extend royalties beyond patent life or "tie" unpatented combo components can be scrutinized under antitrust rules (the EU's Technology Transfer Regulation, for example, permits post-expiry royalties only if clearly tied to know-how or if genuinely optional). European licensors thus often include clauses terminating royalty obligations once the relevant patent expires or a generic version of a component grabs significant market share – similar to the generic competition clause seen in the Vitae-Boehringer contract excerpt, which cut royalties by 50% if generics took ≥25% market share.

This ensures that in Europe (and similarly in the U.S.), a licensor isn't overpaid when competition or patent expiries erode the product's exclusivity.

Recent Examples (2024–2025): Deals and Disputes

Real-world licensing in the past two years reflects these principles in action. Below are several noteworthy examples from 2024 and 2025, illustrating how companies structure royalty splits and how disputes have been resolved:

XTL Biopharma & Biossil (Canada, 2025)

In February 2025, XTL Biopharma sublicensed rights to a combination oncology therapy to Biossil. The exclusive sublicense agreement filed in Canada included a detailed combination sales clause specifying how to calculate royalties if the licensed drug was sold as part of a combo. Notably, it provided multiple fallback formulas and then a good-faith negotiation if needed.

The rest of the deal featured a modest single-digit royalty rate (exact number redacted) and a typical royalty-stacking provision allowing a reduction if Biossil had to license third-party IP. While financial specifics remain confidential, the contract language exemplifies the state-of-the-art approach to combo royalties (allocation by A/(A+B) or analogues) and shows such clauses are no longer academic – they're in active use in 2025's deals.

Merck KGaA (Ares Trading) v. Dyax Corp (USA, 2024)

This case, decided by the Third Circuit in August 2024, set an important precedent in royalty litigation. Merck's subsidiary Ares had licensed a technology from Dyax for developing a drug, agreeing to pay royalties on the drug's sales. The twist: the licensed patents were on a research process, not on the drug itself, and had expired by the time the drug (lanadelumab) hit the market.

Ares argued it was patent misuse to pay post-expiry royalties. The court disagreed – because the royalty was not based on practicing a patented invention post-expiry (the drug did not infringe the expired process patent), Brulotte's rule didn't invalidate the obligation. In essence, royalties could continue since they were tied to a non-patented product. The Third Circuit did hint that if the licensee had framed the misuse argument differently (i.e. that the licensor improperly broadened patent scope by charging for non-infringing activity), the outcome might differ.

But as it stands, the case affirmed that a royalty can be collected on combo or follow-on products even after a patent expires, so long as the royalty is not expressly for the use of the patented invention post-expiry. This nuance is highly relevant to combination therapies: often one patent in the mix might expire before the others or before the product's commercial life ends. License drafters now cite Ares v. Dyax to justify continued royalties if they have a hybrid patent/know-how license – a common scenario for combo products where proprietary know-how (formulations, etc.) supplements patents.

Acorda Therapeutics v. Alkermes PLC (USA/Ireland, 2025)

Acorda's multiple sclerosis drug Ampyra was made using Alkermes' technology, for which Acorda paid royalties. The license had an unusual term requiring royalties for a period after U.S. patent expiry, pegged to when the product launched in each country. After paying ~$82 million post-expiry, Acorda cried foul (invoking the Supreme Court's Kimble v. Marvel rule against patent post-expiration royalties).

In 2025, an arbitration panel sided with Alkermes – allowing it to keep ~$65.6 million of those royalties – and Acorda appealed. The Federal Circuit in June 2025 heard arguments, but an interesting wrinkle emerged: the key issue was actually which court had jurisdiction to review the case (patent-centric or regional circuit). This procedural quirk highlights how complex these disputes can get.

Substantively, the case underscores that even in recent years, parties sometimes push the envelope on royalty terms, risking later fights. Here the combination aspect was a timing one – Ampyra's formulation was covered by an Alkermes patent that expired, yet the contract attempted to extend payments beyond that. The outcome will likely reinforce the Kimble principle: royalties tied purely to expired patent rights are unenforceable in the U.S. (a stark reminder when structuring combination product deals to segregate patent royalties from know-how royalties, and sunset the former appropriately). For now, the Acorda-Alkermes saga is a cautionary tale: what seemed like a creative contract structure led to years of litigation and uncertainty.

Vectura v. GSK (USA/UK, 2020 & ongoing)

While just outside our 2024–25 window, the ripple effects of this case are being felt now. Vectura (a UK company) had licensed inhaler powder technology to GSK, including combination use in Ellipta® inhalers that contain multiple drugs. That license expired, but GSK kept using the tech. Vectura sued and won ~$90 million in 2020, with U.S. courts affirming a 3% royalty on the full net sales of GSK's combination inhalers.

GSK's plea for further apportionment (to pay only on the portion of the inhaler's value attributable to Vectura's one patent) was rejected – because the prior license's 3% rate was deemed to already reflect apportionment across the combo. This established the idea of "built-in apportionment": if a comparable license covered a multi-component product, the agreed royalty rate can be assumed to factor in the component's value, obviating the need to slice the royalty base in court.

Why is this relevant in 2025? Because many current combination product licenses reference earlier deals or industry benchmarks. Vectura's case encourages licensors to set a single royalty rate on the entire product but at a level that represents their contribution (e.g. a lower percentage than if their component were the whole product). It also gives lawyers a precedent to cite in any dispute: "this rate is comparable to X license, so we don't need to relitigate exact allocation – it's baked in." In licensing negotiations now, we see more explicit recitals that the royalty rate "reflects the relative contribution of the licensed component to the combination product" – essentially trying to preempt arguments like GSK's.

Sources: Court decisions and company disclosures from Smith Law, Fitch Even, Bloomberg Law, Kluwer Patent Blog, and Fierce Biotech.

Conclusion

Crafting royalty allocations for combination therapies is part science, part art, and part armistice. The science lies in the economic formulas – dividing the revenue pie by price, value, or contribution. The art is in negotiating terms that both sides deem "fair" – or at least livable – often using creative clauses to hedge against future unknowns (generic competition, new indications, shifting prices). And the armistice is what these clauses represent: a peace treaty between licensor and licensee, ensuring neither feels short-changed when their products join forces.

Recent licensing events in 2024–2025 show that companies are navigating this terrain with increasing sophistication, borrowing from past lessons. We see contracts explicitly delineating how "Net Sales…shall be calculated by multiplying by the fraction A/(A+B)" or similar, leaving little to chance. We also see "dilution" safeguards so that even in a stack of royalties, everyone gets something and the product can still thrive.

Litigation precedents have simultaneously sharpened the rules of engagement – courts will enforce clear contract language (even post-patent payments) when legitimately tied to know-how or non-patented contributions, but they will also strike down overreach that conflicts with public policy on patent limits. In multiple jurisdictions, from Delaware to Luxembourg to Tokyo, the message is: clarity and balance are paramount. A combination can be more than the sum of its parts, but its revenue cannot be double-counted or endlessly extended.

For the pharmaceutical dealmaker, the task is to anticipate every scenario – "what if component B's price is unknown?", "what if a third license is needed?", "what if a patent expires early?" – and address it in the contract. As combination regimens and co-formulated cures become the norm in therapy (think oncology cocktails, gene therapy with device delivery, multi-strain vaccines), these royalty allocation provisions are moving from the contract's fine print to center stage. The tone may be polite and technical, but behind each percentage and formula lies a hard-fought compromise.

Splitting pills is easy; splitting the bill for the pill is where it gets interesting. With thoughtful contracts and informed precedent, however, even this most complicated of group projects can reach a fair accounting – ensuring that inventors remain incentivized, partners remain cooperating, and life-saving combinations reach patients unencumbered by corporate squabbles over who gets how big a slice of the pie.