Pharma industry trends 2027+

Summary: AI has compressed target discovery and molecule design, and AI-discovered molecules pass phase 1 at high rates. Clinical efficacy remains unproven: no AI-discovered drug has completed phase 3 or won approval, and phase 2 failure rates look similar to conventional drugs. Insilico's rentosertib phase 3, which dosed its first patient in 2026, is the first large test of whether AI improves the odds where most drugs fail.

Current situation: A 2024 analysis in Drug Discovery Today found that AI-discovered molecules cleared phase 1 at rates of 80 to 90%, while phase 2 failure rates were similar to conventionally discovered drugs. In September 2026 Insilico Medicine dosed the first patient in GENESIS-IPF-3, a 320-patient, 52-week phase 3 trial of rentosertib, a TNIK inhibitor for idiopathic pulmonary fibrosis, with initial results planned for December 2029 and a potential approval in 2030. Isomorphic Labs raised a $2.1 billion Series B in May 2026 but had not dosed a patient by mid-2026.

Expected development: Until the first phase 3 readouts arrive around 2029, AI platforms will be valued on partnership deal flow and phase 1 to phase 2 transitions. A positive rentosertib result would raise the price of AI-originated assets and accelerate pharma partnerships, while a failure would reposition AI as a productivity tool for chemistry and trial design. Large pharma companies keep building internal capability alongside partnerships, which points to consolidation among independent AI biotechs.

Challenges: Most phase 2 failures stem from target biology, and novel AI-selected targets carry high validation risk. Definitions of an AI-discovered drug vary, which makes success claims hard to compare. Late-stage trials cost as much for AI-designed molecules as for others, so capital needs stay high even for AI-native companies.

Time to impact: 4-6 years

Potential impact: High

STEEP segment: Technological

Summary: Conventional antibodies barely cross the blood-brain barrier, limiting biologics in neurology. Receptor-mediated shuttles such as Roche's Brainshuttle deliver far more antibody into the brain, with early Alzheimer's data showing faster plaque clearance and less brain swelling. If phase 3 confirms this, shuttles become a platform for neurology and psychiatry.

Current situation: In Roche's phase Ib/IIa Brainshuttle AD study, 91% of patients treated with trontinemab became amyloid PET negative, with ARIA-E (brain swelling) below 5%. Roche is running the phase III TRONTIER 1 and 2 trials in early symptomatic Alzheimer's disease and has designed PrevenTRON, a phase III prevention trial in cognitively unimpaired people at high risk, using its pTau217 blood test to identify participants.

Expected development: TRONTIER readouts toward the end of the decade decide whether shuttled antibodies become standard in Alzheimer's disease. Success would carry shuttles into Parkinson's disease, ALS, neuro-oncology and lysosomal storage disorders and make shuttle platforms prime licensing targets. Prevention trials would extend treatment to people without symptoms, a far larger population.

Challenges: Receptor-mediated transport can carry off-target risks, including effects on red blood cells. Prevention trials are long and costly, and payers will question treating people without symptoms. Infusion capacity and MRI monitoring requirements already limit uptake of anti-amyloid therapies.

Time to impact: 6-8 years

Potential impact: High

STEEP segment: Technological

Summary: GLP-1-based drugs now carry outcome labels beyond weight loss, and next-generation agonists approach weight loss once reserved for surgery. Competition moves from weight loss alone to tolerability, muscle preservation and outcome data across organs. Payers face a class that touches cardiology, nephrology, hepatology and sleep medicine at once.

Current situation: Semaglutide is approved for cardiovascular risk reduction, kidney protection in type 2 diabetes and MASH, and tirzepatide for obstructive sleep apnoea. In May 2026 Lilly reported that its triple agonist retatrutide produced 28.3% average weight loss at 80 weeks and 30.3% at 104 weeks in the phase 3 TRIUMPH-1 trial. In July 2026 TRIUMPH-3 showed 22.6% weight loss in people with severe obesity and heart disease, where discontinuation reached 13.5% on the top dose against 4.8% on placebo.

Expected development: Outcome trials in heart failure, kidney and liver disease expand labels and payer acceptance through 2028. Retatrutide and amylin combinations raise efficacy, so differentiation shifts to tolerability, dosing, lean mass and maintenance after weight loss. Trials outside metabolic disease, such as in neurodegeneration and addiction, define the limits of the class.

Challenges: Gastrointestinal side effects and discontinuation limit real-world persistence. Long-term safety of very large weight loss, including loss of lean mass, needs more data. Payers struggle to fund chronic use across so many indications.

Time to impact: 2-4 years

Potential impact: Very high

STEEP segment: Technological

Summary: In vivo CAR-T delivers genetic instructions directly into the body instead of extracting, engineering and reinfusing a patient's T cells. Early myeloma data show responses without lymphodepleting chemotherapy or individual manufacturing. If it works at scale, cell therapy becomes an off-the-shelf product that can reach autoimmune disease and community settings.

Current situation: Large pharma bought in early: AstraZeneca acquired EsoBiotec for $425 million upfront in March 2025, AbbVie acquired Capstan for up to $2.1 billion in June 2025, and Gilead's Kite acquired Interius for $350 million upfront. Phase I data for AstraZeneca's ESO-T01 in multiple myeloma, published in March 2026, showed responses without lymphodepletion, but the first patient had a hyperacute inflammatory reaction that led to a protocol change adding prophylactic steroids. Lentiviral approaches lead in oncology, while mRNA-lipid nanoparticle approaches such as Capstan's target autoimmune disease.

Expected development: Phase 1 and 2 data through 2028 show whether responses are durable and whether safety can be managed outside specialist centres. Successful programmes would undercut autologous CAR-T on cost and logistics and open lupus and other autoimmune diseases to a one-time immune reset. Vector and lipid nanoparticle manufacturing becomes the new bottleneck in place of patient-specific cell processing.

Challenges: Unintended transduction of non-target cells is a safety concern regulators will scrutinise. Dose control and persistence are harder without ex vivo selection. Pricing a one-time, off-the-shelf therapy against established CAR-Ts and bispecific antibodies is unresolved.

Time to impact: 6-8 years

Potential impact: Very high

STEEP segment: Technological

Summary: In vivo editing of liver genes such as PCSK9 and ANGPTL3 has lowered LDL cholesterol by 50 to 60% for more than a year after a single infusion. If outcome trials and long-term safety hold, cardiovascular prevention could shift from daily pills and regular injections to a one-time treatment. That would move gene editing from rare diseases into one of the largest markets in medicine.

Current situation: In May 2026 Lilly reported that its PCSK9 base editor VERVE-102 reduced PCSK9 by up to 88% and LDL cholesterol by up to 62% after a single infusion in 35 patients, with effects lasting up to 18 months, no treatment-related serious adverse events and planned follow-up of 15 years; Lilly plans to start phase 2 by the end of 2026. CRISPR Therapeutics' CTX310, which knocks out ANGPTL3, cut LDL cholesterol by 52.5% and triglycerides by 47.8% at one year at the top dose, published in the New England Journal of Medicine in August 2026. Lilly acquired Verve in 2025, while a liver safety event paused a phase 3 programme of Intellia's editing therapy for transthyretin amyloidosis.

Expected development: Phase 2 and 3 programmes into the early 2030s must show durable lipid lowering and then cardiovascular outcome benefit, likely starting with familial hypercholesterolaemia and premature coronary disease. Approval for broad prevention requires long-term safety data that take years to accumulate. Payers will weigh a one-time price against decades of statins, PCSK9 antibodies and siRNA injections.

Challenges: Permanent edits cannot be reversed if late side effects emerge. Off-target edits and delivery-related liver toxicity remain key safety questions. A one-time price for a common condition strains annual budgets and payment models.

Time to impact: 8-10 years

Potential impact: High

STEEP segment: Technological

Summary: Both major markets are cutting the period in which a new medicine earns its return undisturbed. The EU pharma package lowers baseline regulatory protection and makes extensions conditional, while US Medicare negotiation reaches small molecules seven years after approval and biologics after eleven. Lifecycle plans built on a decade of free pricing need rework in both markets.

Current situation: On 11 December 2025 the Council and the European Parliament reached a provisional agreement on the EU pharma package. It replaces the 8+2+1 formula with eight years of regulatory data protection plus one year of market protection, conditional extensions and an overall cap of 11 years, and cuts baseline orphan market exclusivity from ten to nine years. The Council has adopted its first-reading position, and formal adoption is still pending. In the US, Medicare negotiated prices for the first ten drugs took effect in January 2026, 15 more including semaglutide follow in January 2027 (semaglutide at $274), and the 2028 cycle brings physician-administered Part B drugs into negotiation for the first time.

Expected development: Formal adoption of the EU package is expected in 2026 or 2027, and products filed before the 24-month cut-off keep the old protection periods, so the full effect lands on launches toward the end of the decade. Companies will design programmes around the extension criteria, such as unmet medical need, and reprioritise indications that qualify. In the US, each negotiation cycle pulls in more high-spend products, and the four-year gap between small molecules and biologics strengthens the tilt of R&D portfolios toward biologics.

Challenges: Protection now differs by market, modality and indication, which complicates global launch sequencing and the valuation of in-licensed assets. The details of EU extension criteria will only become clear in implementing guidance. Industry keeps lobbying to close the US small-molecule gap, so the rules may move again within the planning horizon.

Time to impact: 2-4 years

Potential impact: Very high

STEEP segment: Political

Summary: The FDA's drug centre lost close to a fifth of its staff and many divisions run under acting leaders. Approval volumes held up, but rejections rose and high-profile reversals unsettled sponsors. Regulatory risk shifts from timelines to consistency.

Current situation: The FDA's drug centre lost almost 20% of its staff to layoffs, buyouts, retirements and resignations, and the agency is seeking to hire about 2,200 people, of whom 600 had been secured by mid-2026. The FDA issued about 79 decisions in the first half of 2026 against 85 a year earlier, with both novel approvals and rejections rising. The second rejection of Replimune's melanoma therapy RP1 drew strong criticism, and analysts expect productivity to recover only from 2027.

Expected development: Sponsors invest more in pre-submission alignment, written agreements and advisory engagement to reduce decision risk. Rehiring and AI tools may restore throughput from 2027, while institutional memory takes longer to rebuild. Some companies will seek first approvals in other jurisdictions to hedge US uncertainty.

Challenges: Political leadership can override review-team recommendations. Inconsistency raises the cost of capital for small biotechs that depend on a single approval. Litigation and congressional oversight add uncertainty.

Time to impact: 0-2 years

Potential impact: High

STEEP segment: Political

Summary: The FDA has proposed a framework that lets individualised genome-editing and RNA therapies for ultra-rare diseases gain approval on a plausible mechanism and small patient numbers. It follows the 2025 treatment of a baby with a custom CRISPR therapy. Rare disease development shifts from one drug per trial to platforms that produce many variant-specific products.

Current situation: On 23 February 2026 the FDA issued draft guidance on a plausible-mechanism framework for individualised therapies targeting genetic conditions with a known biological cause, where randomised trials are infeasible. It builds on the Rare Disease Evidence Principles process announced in September 2025 and on a 2025 case in which Children's Hospital of Philadelphia and the University of Pennsylvania designed a CRISPR therapy for a baby with a rare metabolic disorder. The FDA says the concepts may apply beyond genome editing.

Expected development: Platform developers will seek approval for an editing system plus a process to create variant-specific products, with real-world evidence collected after approval. Academic centres and non-profits play a larger role in producing therapies for tiny populations. The first approvals under the framework could arrive before 2030 and test whether payers reimburse them.

Challenges: Small evidence packages leave questions about long-term safety and durability. Manufacturing and quality control across many variants is complex and costly. Payers may refuse to cover therapies approved on mechanism, leaving approvals without a market.

Time to impact: 4-6 years

Potential impact: High

STEEP segment: Political

Summary: The US now taxes imported patented medicines and their ingredients at up to 100% unless companies produce locally or hold pricing deals. Branded drugmakers have answered with roughly $500 billion in announced US investment, while generics and several specialty categories remain exempt for now. Network strategy, transfer pricing and launch sequencing now depend on trade policy as much as on cost.

Current situation: Proclamation 11020 of 2 April 2026 imposed Section 232 tariffs of up to 100% on patented pharmaceuticals, their active ingredients and key starting materials. Duties applied from 31 July 2026 for 17 named large manufacturers and from 29 September 2026 for all other importers. Generics, biosimilars, orphan drugs, cell and gene therapies, antibody-drug conjugates, plasma-derived therapies and nuclear medicines are exempt, and UK-origin products are exempt until January 2029 under a bilateral arrangement. Separately, the BIOSECURE Act, enacted in December 2025, requires OMB to publish a list of biotechnology companies of concern by December 2026, which law firms expect to include some China-based CROs and CDMOs.

Expected development: Much of the announced capacity is scheduled for completion by 2030, which shifts API and fill-finish for branded products toward the US and leaves European and Asian sites to serve generics and ex-US markets. Patent status becomes a customs question, so loss of exclusivity changes a product's tariff treatment. Generic tariffs, deferred until 2028, and the first BIOSECURE list will extend the pressure into the rest of the supply chain.

Challenges: New biologics plants take several years to build and validate, so tariffs bite before capacity exists. Duplicated capacity raises unit costs and leaves European sites underused. Tariff terms can change through litigation, bilateral deals or a new administration, which ties long-lived capital decisions to short political cycles.

Time to impact: 0-2 years

Potential impact: Very high

STEEP segment: Political

Summary: US drug shortages are rising again, led by cheap sterile injectables whose margins cannot fund redundancy and quality systems. Tariff exemptions keep generic imports flowing but leave their economics unchanged. Without prices that reward reliability, shortages stay structural.

Current situation: ASHP counted 227 active US drug shortages in the second quarter of 2026, the third consecutive quarterly increase after a 2025 low, down from a record 323 in early 2024. Generic sterile injectables make up the core of active shortages, and experts told lawmakers that prices are too low to justify investment in quality systems. A severe ifosfamide shortage led the second-quarter increase.

Expected development: Purchasers test reliability premiums and multi-supplier contracts, and the EU's Critical Medicines Act brings resilience criteria into public procurement. Some generic manufacturers exit further products, while others seek government support for domestic sterile capacity. Shortages persist through 2028 because new sterile plants take years to build.

Challenges: Purchasing practices favour the lowest bid and sole awards. The EU wastewater charge and other costs add pressure on the cheapest products. A quality failure at a single plant can remove a large share of supply.

Time to impact: 0-2 years

Potential impact: High

STEEP segment: Economic

Summary: Extreme weather already causes drug shortages when one plant supplies a large share of a market. Nearly two-thirds of US drug production facilities sit in counties prone to climate-related disasters, and key Indian hubs face heat, flood and water stress. Climate resilience becomes a supply and procurement requirement alongside cost and quality.

Current situation: In September 2024 Hurricane Helene flooded Baxter's North Cove plant in North Carolina, which made about 60% of the IV fluid bags used in US hospitals, forcing rationing until the shortage was declared resolved in 2025; Hurricane Maria had disrupted Baxter's saline production in Puerto Rico in 2017. A 2025 JAMA research letter found 62.8% of US pharmaceutical facilities located in counties prone to climate-related disasters. Lupin reports that key sites including Mumbai, Vadodara and Visakhapatnam face exposure to temperature extremes, flooding, cyclones and water stress.

Expected development: Through the 2030s, purchasers and regulators require climate risk assessments, dual sourcing and buffer stocks for critical medicines, building on resilience rules such as the EU Critical Medicines Act. Companies harden sites against floods and heat, invest in water recycling and spread production across climate zones. Insurance costs and capital requirements rise for plants in exposed regions.

Challenges: Redundant capacity raises costs for low-margin products. Site-level climate risk data is incomplete. Onshoring can move production into regions with their own climate exposure.

Time to impact: 6-8 years

Potential impact: High

STEEP segment: Ecological

Summary: Most-favoured-nation pricing ties US launch prices to those in other rich countries, so companies push for higher prices abroad or delay launches there. The UK raised its NICE threshold for the first time since 1999 and committed to double medicines spending as a share of GDP. Other European payers face a choice between paying more and waiting longer.

Current situation: Under the UK-US pharmaceutical deal announced in December 2025, the UK secured zero tariffs on drug exports to the US and raised NICE's cost-effectiveness threshold from £20,000 to £30,000 per QALY to £25,000 to £35,000 from April 2026. The UK committed to raise medicines spending from 0.3% of GDP toward 0.6% over ten years and to cut the VPAG rebate rate to 15% in 2026. All 17 large drugmakers in the US deals committed to launch new medicines at prices no higher than in other developed countries.

Expected development: Companies hold back or delay launches in low-price European markets unless prices rise, which pressures Germany, France, Italy and Spain to revise pricing rules or accept slower access. Confidential rebates become more important to keep list prices high while net prices stay negotiable. EU-level pricing coordination may gain political support as a counterweight.

Challenges: Higher prices strain European health budgets already under pressure. Delayed launches harm patients and provoke political backlash against industry. US policy could change after 2028, leaving Europe with commitments that outlast the pressure.

Time to impact: 0-2 years

Potential impact: Very high

STEEP segment: Economic

Summary: The price of obesity treatment is falling on three fronts at once: semaglutide generics outside the US, US cash prices from $149 a month for pills, and Medicare coverage at $245 a month. Volume grows while revenue per patient shrinks.

Current situation: Semaglutide patents expired in India on 20 March 2026, and Indian companies launched generics at prices roughly 70% or more below Novo Nordisk's brands; patents expire in about ten countries in 2026 that carry 48% of the global obesity burden, and one analysis estimates generic injectable semaglutide could be produced for $28 to $140 per person per year. In the US, Lilly and Novo agreed Medicare prices of $245 a month with $50 copays, delivered through the Medicare GLP-1 Bridge from July 2026 to December 2027. Starter doses of oral GLP-1 pills cost $149 a month for self-pay patients.

Expected development: Price competition spreads from India to Canada, Brazil, China and other markets through 2027, and Indian manufacturers export generics to Africa and Latin America. In the US, branded net prices fall toward Medicare levels while volume grows across Medicare and cash channels. Revenue growth shifts to next-generation agonists and new indications.

Challenges: US semaglutide patents run to about 2032, so the gap between US and foreign prices widens and invites gray imports. Several states have dropped Medicaid coverage for obesity drugs. Lower prices squeeze returns on the heavy investment in manufacturing capacity.

Time to impact: 0-2 years

Potential impact: Very high

STEEP segment: Economic

Summary: Medicines that lower a genetic risk factor before disease appears promise huge markets, from Lp(a) to pre-symptomatic Alzheimer's disease. In September 2026 the first Lp(a) outcomes trial lowered the biomarker but failed to reduce heart attacks and strokes. Payers will fund population-scale risk reduction only with hard outcome data, so biomarker success alone no longer carries a prevention launch.

Current situation: Novartis reported on 4 September 2026 that pelacarsen lowered lipoprotein(a) but did not significantly reduce major adverse cardiovascular events in the 8,323-patient Lp(a)HORIZON trial in people with established cardiovascular disease. The 2026 ACC/AHA multisociety dyslipidaemia guideline recommends measuring Lp(a) in all adults at least once. Amgen's olpasiran, which cut Lp(a) by more than 95% in phase 2, is in the 7,297-patient OCEAN(a)-Outcomes trial with completion listed for March 2028, and Lilly's lepodisiran is in ACCLAIM-Lp(a) with about 17,300 patients and completion listed for March 2029.

Expected development: OCEAN(a) and ACCLAIM-Lp(a) readouts in 2028 and 2029 decide whether deeper Lp(a) lowering creates a major prevention market. Universal Lp(a) testing builds a large identified population either way. Across prevention, from Lp(a) to pre-symptomatic Alzheimer's disease and type 1 diabetes, sponsors need outcome trials in primary prevention, which take longer and cost more.

Challenges: Primary prevention trials need very large populations and long follow-up. Treating millions of people without symptoms creates a budget impact payers resist. A biomarker that predicts risk may be the wrong lever to reduce it.

Time to impact: 6-8 years

Potential impact: High

STEEP segment: Economic

Summary: Manufacturer cash-pay channels have moved from niche programmes to federally promoted infrastructure. TrumpRx, launched in February 2026, routes patients to company offers priced at most-favoured-nation levels and lists more than 800 products. Direct channels give manufacturers patient relationships and data, and expose net prices that the rebate system used to hide.

Current situation: All 17 large drugmakers the White House approached in July 2025 have signed most-favoured-nation pricing deals, which the administration says cover about 86% of the branded US market. The deals combine cash-price direct-to-consumer offers through TrumpRx or company channels, MFN prices for Medicaid, and a commitment to launch new medicines at prices no higher than in other developed countries. TrumpRx launched on 5 February 2026 with around 40 products and listed more than 800 by August, including about 600 generics routed through partners such as Mark Cuban Cost Plus Drugs, Amazon Pharmacy and GoodRx. In August 2026, ten pharmacy benefit managers committed to show TrumpRx cash prices in prescribers' real-time benefit tools.

Expected development: Direct channels will extend from obesity and primary-care brands to specialty products with high out-of-pocket costs. Prospective MFN launch pricing gives companies a reason to delay or skip launches in low-price countries to protect the US reference price. PBM economics come under pressure as cash prices appear at the point of prescribing, and employers will test plan designs that steer patients to direct channels.

Challenges: Cash purchases may not count toward insurance deductibles and out-of-pocket caps, so some patients pay more overall. Reuters found many TrumpRx prices above those paid in the UK, which invites scrutiny of the MFN claims. The deals are voluntary, and the administration has asked Congress to codify MFN, so their durability depends on politics.

Time to impact: 0-2 years

Potential impact: High

STEEP segment: Economic

Summary: Clinical AI tools have become a first stop for many US doctors seeking treatment information, and some are funded by pharma advertising. How a drug appears in AI-generated answers starts to matter as much as rep visits or journal ads. Medical information and evidence must be structured for machines to read.

Current situation: OpenEvidence, a free AI search tool funded by pharmaceutical and device advertising, reports more than 757,000 verified US clinicians using it regularly and was valued at $12 billion in early 2026. A March 2026 survey of 1,165 physicians in seven countries found 83% used ChatGPT at work, 50% Gemini, 36% OpenEvidence and 32% UpToDate's AI search. Health systems including Mount Sinai and Sutter Health have embedded OpenEvidence in their electronic health records.

Expected development: Pharma shifts budget from detailing and journal ads to AI platforms, structured evidence and generative engine optimisation of medical information. Advertising inside clinical AI tools grows, and regulators examine whether ads influence answers. Field forces shrink and concentrate on complex products where AI answers fall short.

Challenges: AI answers can be wrong or outdated, which raises safety and liability questions. Advertising in clinical decision tools raises conflict-of-interest concerns and possible regulation. Companies have little control over how models summarise their evidence.

Time to impact: 0-2 years

Potential impact: High

STEEP segment: Social

Summary: Large studies in England and New York now sequence newborns for hundreds of treatable genetic conditions. Most confirmed cases are disorders that standard newborn screening misses, so children can be treated before symptoms appear. If health systems adopt genomic screening, rare disease companies gain earlier, larger and better-defined patient populations.

Current situation: The GUARDIAN study in New York screened 4,000 newborns in its first analysis, published in JAMA: 3.7% screened positive, and 92% of the 120 confirmed cases had disorders outside standard newborn panels, such as long QT syndrome, severe combined immunodeficiency and Wilson disease. GUARDIAN has enrolled more than 13,000 babies and expanded its panel to 446 genes covering more than 460 conditions. England's Generation Study, run by Genomics England with the NHS, aims to sequence up to 100,000 newborns for more than 200 treatable conditions and has returned more than 13,000 results to families.

Expected development: Health systems decide in the early 2030s whether to add genome sequencing to routine newborn screening, with England and New York as the first test cases. Rare disease developers design trials and launch plans around pre-symptomatic cohorts, where early treatment may show larger benefit. Gene and bespoke therapies gain patients identified at birth, which strengthens platforms that can address many variants.

Challenges: Variants of uncertain significance can cause anxiety and over-treatment. Consent, data storage and later use of genomes raise ethical questions. Health systems must fund confirmatory testing, specialist follow-up and the therapies themselves.

Time to impact: 6-8 years

Potential impact: High

STEEP segment: Social

Eight trends carry very high potential impact. Four of them hit within two years: China becoming big pharma's largest external source of new molecules, US tariffs that tie market access to US production, US pressure on European drug prices and the collapse in obesity drug prices. The other four follow later: shorter exclusivity in Europe and the US, a patent cliff worth more than $200 billion, incretin drugs growing into a multi-organ franchise and in vivo CAR-T.

Since 29 September 2026, Section 232 tariffs of up to 100% apply to imported patented medicines, their active ingredients and key starting materials. Generics, biosimilars, orphan drugs, cell and gene therapies and antibody-drug conjugates are exempt for now. Branded API and fill-finish move toward the US as new plants open toward 2030, which leaves many European and Asian sites serving generics and markets outside the US. A product's tariff treatment also changes once its patents expire.

AI already speeds up target discovery, molecule design and regulatory writing, and many doctors look up treatments in AI tools. Its clinical value is still an open question. The first phase 3 trial of an AI-discovered drug, Insilico's rentosertib, expects initial results around December 2029. Regulators want proof that AI models are credible too, as set out in FDA draft guidance from January 2025 and joint FDA and EMA principles from January 2026.

Eighteen of the 71 trends sit six to ten years out. The highest rated include in vivo CAR-T, one-time gene editing for cholesterol, digital twins in clinical trials, AI-run labs and virtual cells, brain shuttles for antibody drugs, genome sequencing at birth and the pressure ageing societies put on drug budgets. Decisions on these trends belong in today's discovery and early clinical portfolios.

They rate each trend against their own assets and link it to the programmes it affects, so a change in pricing, regulation or science shows up where budget decisions are made.

A shorter EU data protection period, for example, changes the value of assets that launch toward the end of the decade, and a failed outcomes trial weakens the case for prevention programmes in the same class.

Pharma teams do this in ITONICS: R&D, strategy and business development rate the trends with their own experts, connect each one to the projects it affects and bring that view into portfolio and stage-gate reviews.

Big pharma now sources a large share of its new clinical assets from outside, and Chinese partners alone account for 38% of its licensing deals. Competition for good assets is pushing prices up, with average upfront payments for China-sourced deals rising from $52 million in 2022 to $172 million in early 2026.

Scouting teams screen biotechs, academic groups and licensing opportunities against the targets and modalities in their R&D strategy and rate each candidate before business development engages.

ITONICS gives these teams one place to track trends, technologies and startups, rate them with internal experts and connect each one to the projects it affects.