Pharma Sustainability 2026: The Footprint Is in the Molecule

Pharma’s sustainability problem was never a factory problem. What changed in the last twelve months, who is now asking for what, and what a small developer should do about it.

Most industries clean up by fixing their plants. Pharma can’t, or not mainly. Four-fifths of its footprint sits in a supply chain it doesn’t own, in solvents it never sees, and in molecules that carry on working after the patient has swallowed them. The pledges are still there. What’s new is that the pressure now turns up as procurement letters with dates in them, and as court orders, and it reaches a long way down the supply chain.

Two numbers from the 2025 reporting season say more than any strategy document. Novo Nordisk’s total emissions, all three scopes, rose 19% in a single year. Eli Lilly’s Scope 3 roughly doubled, from about 2.99 million tonnes of CO₂e in 2021 to 5.14 million in 2023. These aren’t laggards. They’re the two companies that grew faster than anyone in the industry’s history, and they did it the only way you can: by building plants, buying inputs and shipping product. The carbon followed the revenue.

That’s the awkward shape of the thing. Pharma isn’t failing to act. It is acting, measurably, on the part of the footprint it can see from the boardroom window, and losing ground on the part it can’t. To see why, put the sustainability reports down and follow the mass.

How big, and where

The number that first made this a topic came out of McMaster University in 2019: pharma at 48.55 tonnes of CO₂e per million dollars of revenue, around 55% above the car industry. People have argued about the method ever since, but nobody has managed to turn the direction of it round. The fullest current picture is My Green Lab’s Carbon Impact Report from December 2025, which covers 1,055 biotech and pharma companies with combined 2023 emissions of 348 million tonnes. Scope 3 is about 75% of the total at public companies and 88% at private ones. And here is the line that matters. On operational emissions alone, the sector is tracking a 1.04°C pathway. Add the supply chain back in and it’s tracking 1.9°C.

The industry has solved the part of the problem it could see.

Hospitals see it from the other end. Medicines are about a quarter of the NHS carbon footprint in England, inhalers alone around 3%. If you run a hospital trust with a net zero target, the drug cupboard is a bigger headache than the boiler house, and there’s far less you can do about it.

Challenge one: the mass behind every milligram

There’s nothing mysterious about why Scope 3 dominates. It’s a mass balance, and anyone who has closed one on a chemical plant knows how it ends. The ACS Green Chemistry Institute Pharmaceutical Roundtable benchmarks member processes on Process Mass Intensity: total kilograms in, per kilogram of active ingredient out. The Roundtable’s benchmarking put the small-molecule median at 120 kg of input per kilogram of API, with a range running from 23 to nearly 900, and the composition is the telling part: solvents around half, water around 30%, reactants under 10%. The reagents that actually make the molecule are a sliver. Solid-phase peptide synthesis is another order of magnitude again, with published estimates running into the thousands, because of the washing cycles in dimethylformamide the chemistry can’t do without.

Scale that to a dose and it stops being abstract. On the benchmark median, a 10 mg small-molecule dose has about 1.2 g of upstream material behind it, most of it solvent that was bought, heated, distilled and then either recovered or burned. At a peptide PMI of 5,000, in the middle of the published range, a 1 mg dose carries 5 g. The patient swallows the milligram. The planet gets the grams. Nobody in a plant control room would call that a rounding error, and the Roundtable doesn’t either; its 2025 grant round went on greener peptide chemistry and biocatalysis for exactly this reason.

The response that will actually shift this isn’t a technology. It’s a letter. On 21 November 2025 the chief procurement officers of AstraZeneca, GSK, Novartis, Novo Nordisk, Roche, Samsung Biologics, Sanofi, UCB and Chiesi signed an update to their joint minimum supplier targets under the Sustainable Markets Initiative. Read it as a specification, because that’s what it is. Suppliers are to have third-party validated science-based targets across all three scopes by 2027, publish product carbon footprints for their top ten products by 2028, be on at least 80% renewable electricity by 2030, cut process-heat emissions by a fifth by the same date and, in the clause most people skip, write the same terms into contracts with their own suppliers.

The NHS has put the same logic into tender law, which is harder to ignore than a letter. Every new NHS procurement has needed a Carbon Reduction Plan since April 2024. From April 2027 that plan has to cover global Scope 1, 2 and 3. From April 2028 it reaches down to the footprint of individual products, and from 2030 a supplier that can’t show progress doesn’t qualify. Voluntary has become contractual, and it happened almost without comment.

The most important sustainability document of 2025 wasn’t a strategy. It was a procurement letter with dates in it.

Challenge two: the molecule after the dose

Carbon is the problem pharma shares with every other industry. Water is the one it owns outright. An active ingredient is designed to be potent at tiny concentrations and tough enough to survive the stomach, which are exactly the properties that make it a nuisance in a river.

The study everyone still cites is Larsson’s from 2007, on a treatment plant serving bulk-drug manufacturers outside Hyderabad: ciprofloxacin in the effluent at up to 31 mg/L, more than a thousand times the level that is toxic to some bacteria. The Leipzig group went back years later and found the area still heavily contaminated and full of multidrug-resistant bacteria. Antibiotic manufacturing is one of the rare cases where an environmental discharge and a threat to human health are the same event. In September 2024 the WHO published its first guidance on antibiotic manufacturing waste and said out loud what the sector had known for years: the area was largely unregulated and quality standards didn’t look at emissions. The AMR Industry Alliance realigned its manufacturing standard to that guidance in May 2025. The BSI Kitemark that certifies it now covers more than 60 antibiotic products across 15 geographies, and the reason generic companies pay for the badge is that NHS and Nordic antibiotic tenders now ask for it.

Then there’s the wastewater directive. In force since 1 January 2025, it requires large treatment plants to add a fourth, micropollutant-removal stage on a timetable running to 2045, and sends at least 80% of the bill to the producers of human medicines and cosmetics. Transposition by 31 July 2027; the levy bites by 31 December 2028. EFPIA went to court to annul it and on 18 February 2026 the General Court threw the case out as inadmissible without reaching the merits. Zentiva and other off-patent companies appealed in April, arguing, not unreasonably, that generics are around 70% of what Europe dispenses and are priced by governments, so there’s nowhere for a levy to go.

The clause worth reading twice is the exemption. Substances that are rapidly biodegradable in wastewater, or don’t produce micropollutants at end of life, sit outside the scheme. For the first time, what a molecule does after it leaves the body has a price at the point of sale. It’s a design incentive that happens to look like a tax. The EU pharma package agreed on 11 December 2025 closes the loop from the other side: the environmental risk assessment stops being a form you file once at authorisation and becomes something regulators apply across the life of the product, manufacturing included.

Challenge three: when the device is the emission

Sometimes the footprint isn’t the molecule or the supply chain but the thing that carries the dose. A pressurised metered-dose inhaler runs on HFA-134a, global warming potential 1,430 on the AR4 values the EU F-gas Regulation uses (AR6 puts it at 1,530), and the propellant in one inhaler can outweigh the carbon of making the drug inside it many times over. This is also the one place the industry has done something it never did before: reformulated marketed products, at scale, purely for environmental reasons. In May 2025 the MHRA approved AstraZeneca’s Trixeo Aerosphere on HFO-1234ze(E), GWP 1.37, a 99.9% cut, with the rest of the range to follow by 2030. In July 2026 it approved Chiesi’s Clenil Modulite on HFA-152a, GWP 124, the first approval of that propellant anywhere. GSK is close behind with salbutamol. The push is regulatory as much as reputational: HFA-134a sits inside the proposed EU PFAS restriction, on which ECHA’s Risk Assessment Committee adopted its opinion on 26 March 2026, with the socio-economic committee’s draft the same day, backing the restriction with targeted, time-limited derogations. A window, not a pass.

The lesson that travels is what the regulators accepted. They took a bioequivalence bridge rather than new efficacy trials. A reformulation whose only purpose was environmental got through on an equivalence argument, and that will be cited every time a company wants to change a solvent, an excipient or a pack for sustainability reasons and is nervous about a clinical re-run.

The view from the excipient drum

The layer nobody talks about sits between the API and the supply chain: the excipients. Croda’s pharma group, which signed up to the SMI targets in 2023 and has committed to product carbon footprints on the Together for Sustainability method, has written one of the more candid pieces on this from the supplier’s side. Two of its points deserve a wider audience. The first is almost embarrassing: up to 90% of the excipient in a container can be thrown away after opening, because pack sizes don’t match batch sizes and once a drum is unsealed nobody will vouch for what’s left in it. The second is subtler. Sustainability has to be measured per dose, not per kilogram. A better excipient may carry a bigger footprint per kilo and still win, because it shrinks the dosage form, extends shelf life or takes a product out of the cold chain.

Croda also raises a worry it doesn’t answer, and for anyone outside the top twenty it’s the one that matters: small and emerging companies may get squeezed out, because they can’t afford the data, the expertise or the audits the targets demand. That deserves a straight answer, because the SMI letter’s last clause, the one about pushing terms down to your own suppliers, means a start-up licensing a formulation platform to any of the signatories is a supplier, and will be asked.

A medicine’s footprint is decided when its formulation is decided. Everything after that is measurement.

The squeeze, and the honest answer to it

The honest answer is that the big-company apparatus, EcoVadis scores, assured Scope 3 inventories, validated targets, is out of reach for a pre-revenue developer, and nobody sensible expects otherwise. What a licensing partner’s procurement team actually needs from a small developer is narrower. Most of it is engineering a formulation scientist already has, or could build in a week.

A mass balance of the formulation. Every input, per dose, solvent content stated. That’s the raw material of a product carbon footprint; a partner’s sustainability team can finish the job with published emission factors. A developer who hands over a clean mass balance is already ahead of most suppliers on the list.
A solvent inventory, or the absence of one. If the process is aqueous and solvent-free at the formulation stage, say so in those words, with the numbers. On the benchmark, solvent is the biggest line in the footprint, and a process without it starts somewhere else entirely.
The end-of-life position of every excipient. Ready-biodegradability data already exists for most food-grade lipids and surfactants; it’s in the supplier’s safety data sheet or REACH dossier. Pulling it together is a day’s work, and it answers the wastewater question before anyone asks it.
A cold-chain and shelf-life statement. Ambient stability is an avoided emission. Croda calls this ‘Scope 4’ and admits nobody can count it yet; the developer who can at least state it will be believed.

None of that is a sustainability programme. It’s the discipline a plant engineer applies before a line is built: know what goes in, know what comes out, know where it ends up. The companies that will find 2028 painful aren’t the small ones. They’re the mid-sized ones with twenty legacy products, no mass balances and a procurement questionnaire on the desk.

What changes next

The calendar is set, and it’s worth seeing in one place.

2027

Validated science-based targets across all three scopes (SMI). NHS Carbon Reduction Plans go global, Scope 1–3 (April). Wastewater directive transposed by member states (31 July).

2028

Product carbon footprints for top ten products (SMI) and for individual NHS-supplied products (April). Wastewater producer levy in force (31 December).

2030

80% renewable electricity and a 20% cut in process heat across the supply chain (SMI). NHS contract gate: no progress, no contract. AstraZeneca’s full pMDI range on low-GWP propellant.

2033–2045

Quaternary wastewater treatment phased in at large plants, funded at least 80% by producers. PFAS derogations for propellants expire on their own clock.

What falls due, and when. Sources: SMI Health Systems Task Force supplier targets (Nov 2025); NHS Net Zero Supplier Roadmap; Directive (EU) 2024/3019; AstraZeneca; ECHA.

Mass per dose is the number that will be asked for. Persistence is about to have a price. And an environmental reformulation now has a regulatory bridge it didn’t have five years ago.

Which brings it back to Novo Nordisk and Lilly. Their emissions went up because demand went up, and the molecules behind that demand are peptides, made at the heaviest mass intensity in the industry and delivered in disposable pens. The next generation of those medicines will be judged on more than whether they work. It’ll be judged on how much material it took to get a milligram into a patient, and on where that milligram ends up afterwards. The footprint is in the molecule. From here on, it’s in the contract too.

George Forrest is CEO and co-founder of Ibumix and a Chartered Chemical Engineer. He previously ran Stepan Company’s UK manufacturing site.

Ibumix is a UK pre-clinical pharmaceutical company developing a series of new liquid drug delivery platforms built on glyceryl caprate (GCC) lipid chemistry. This article is editorial commentary drawn from published sources and does not describe Ibumix products or make claims about their performance.