Breyer Capital 2026 Healthcare and Life Sciences Predictions
A Forecast of Where Science, Medicine, and Industry are Converging This Year
Substack
By Morgan Cheatham, M.D.
January 21, 2026
Each year at Breyer Capital, we take an intentional inventory of what has changed—from global market economies to scientific discoveries to shifts in the regulatory apparatus. This past year, we also published our updated Healthcare and Life Sciences thesis, where we explored the three most potent forces shaping the industry today: computation, precision, and prevention.
In 2026, capital, talent, and economics will continue to reorganize around these axes. The predictions that follow mark inflection points where technological maturity meets market readiness, and where we anticipate substantial activity this year.
The Great HealthTech Consolidation
Clinical AI Enters Both the Exam Room and the Boardroom
Prevention Brands Build the Clinical Backbone
Abundance Forces Precision in Metabolic Medicine
Aging Indications Move from Fringe to Fundable
Clinical Trials Enter a Necessary Reinvention
AI Drug Discovery Tests New Business Models
The In Vivo Renaissance Accelerates
Conversational Programming Unleashes Clinical and Scientific Creativity
Autonomy Expands Its Footprint, Regulators Lace Up
The Great HealthTech Consolidation
2026 will be a banner year for private market healthcare tech M&A, driven by economic necessity and strategic opportunity. The 2025 funding data tells a story of concentration, not broad recovery. Total health tech investment rose 35% to $14.2B, but mega-rounds captured over 40% of that capital—the highest share since 2021. Strip out the top nine deals and aggregate funding actually fell year-over-year. Meanwhile, M&A deal volume surged over 60% even as the IPO window stayed largely shut.
In 2025, distribution became the defining moat. We saw this across the Breyer Capital portfolio: Abridge deployed across 150+ health systems processing more than 1 million encounters weekly, OpenEvidence reached nearly half of US physicians while unlocking deeper relationships with the institutions that employ them, and Artera AI served 40% of genitourinary oncologists in the United States.
The “roll or be rolled” pressure is coming from all sides. Breakout companies are extending into adjacent products. New entrants like OpenAI and Anthropic are moving directly into healthcare with distribution advantages of their own. And the moat for building AI is fleeting: health systems and payers have gained access to tools like Claude Code, where what once required months of engineering now takes an afternoon.
Point solutions die when anyone can build them. Value accrues to those with distribution, infrastructure, or genuine IP and data moats.
This year, healthtech boardrooms face a binary: consolidate to dominate, or get acquired before commoditization sets in.
Clinical AI Enters Both the Exam Room and the Boardroom
The greatest impact of AI in medicine will be measurement. Medicine has been constrained by what it can reliably observe. Clinical AI removes this constraint, quantifying new signals from audio, video, wearables, and molecular data while uncovering latent patterns in existing clinical records, and enabling a shift from discrete events to continuous biological states.
We’re seeing this across our portfolio. Iterative Health converts colonoscopy video into inflammatory bowel disease endpoints. Cleerly quantifies plaque vulnerability beyond stenosis. Artera predicts therapeutic response from pathology slides. Large opportunities remain in neurology, autoimmune disease, and metabolic conditions—anywhere continuous, objective measurement can replace episodic, subjective observation.
As investors hunt for durable moats in an environment of collapsing software costs, clinical AI stands out: proprietary data, regulatory clearances, and clinical validation create defensibility that a weekend coding project cannot replicate.
The performance is here. The benefits are measurable. It’s imperative that the payment and business models meet the technology. CMS’s ACCESS model, launching in 2026, offers early validation: outcome-aligned payments for chronic disease management that reward the kind of continuous measurement clinical AI enables. But the broader task falls to companies and investors: This is the year we make the business case for clinical AI.
Prevention Brands Build the Clinical Backbone
This year, the pull toward consumer-centric preventive healthcare will continue to strengthen. Demographics, rising expectations, and growing distrust of reactive sick-care models all point in the same direction. Demand will not be the constraint in 2026. Structure will be.
In the coming year, consumer preventive health companies will be forced into a defining transition. They will either remain information-only businesses with fragile unit economics, or begin building clinical infrastructure that captures longitudinal value.
Information-only models will increasingly fail at scale. Delivering findings without owning what comes next will drive customer dissatisfaction, shift downstream costs onto the healthcare system, and invite regulatory and legal scrutiny. As the category crowds, acquisition costs will rise, and retention will depend on demonstrable outcomes rather than data delivery.
In 2026, the leading prevention brands will begin closing the loop. They will convert insights into actions, screening into care pathways, and one-time purchasers into longitudinal relationships. In doing so, they will move decisively away from content-driven models and toward operating as clinical systems.
Abundance Forces Precision in Metabolic Medicine
The GLP-1 market is exploding from injectable duopoly to multi-modal competition. Novo’s oral semaglutide launched at $149–$299/month. Lilly’s orforglipron awaits approval. Compounded versions persist. Next-gen molecules enter trials. The market shifts from supply-constrained scarcity to demand-driven competition on efficacy, convenience, and price.
With abundance comes complexity. The market is bifurcating: oral pills for mild obesity and convenience-seekers, high-efficacy injectables for significant weight loss, compounded versions for price-sensitive consumers. Market projections approach $180B by 2035, but expansion beyond early adopters requires answering a question the field has largely avoided: which therapy, for which patient, and why?
This is where precision tools become essential. As products flood the metabolic market, we need infrastructure to match patients to the right molecule based on obesity severity, comorbidities, response prediction, and adherence likelihood. The clinical intuition that worked when there were two options won’t scale when there are twenty. Value shifts from production to personalization—clinical decision support, outcome tracking, longitudinal management. The companies that build this infrastructure capture durable value in a market where the drugs themselves are commoditizing.
Aging Indications Move from Fringe to Fundable
2026 is the year aging biology takes center stage in biopharma, and possibly the year FDA begins to signal openness to aging as a regulatory endpoint.
The groundwork was laid in 2025. Longevity biotech went mainstream within big pharma, as the GLP-1 era made it obvious that metabolism, inflammation, and aging biology are deeply connected. The same pathways that drive metabolic disease drive aging. The same interventions that improve cardiometabolic health extend healthspan. What was once an abstract category now has a scientific and commercial logic that mainstream investors and pharma partners can underwrite.
Capital followed. In 2025 alone, big pharma put nearly $10 billion behind FGF21: Novo Nordisk acquired Akero, Roche bought 89bio, GSK scooped Boston Pharma’s lead liver disease asset. Novartis has doubled down on its dedicated Diseases of Aging and Regenerative Medicine group. Eli Lilly is increasingly redefining its mandate as a longevity company, moving beyond discrete disease silos to target the fundamental biological drivers of aging. These aren’t speculative bets on longevity—they’re pipeline strategies.
The regulatory architecture hasn’t caught up yet. The FDA does not currently recognize aging as a disease, which means every aging drug must be approved for a specific indication, even when the mechanism is targeting aging itself. But the pressure is building. Loyal’s LOY-002, a daily pill for senior dogs, is on track this year to potentially become the first FDA-approved drug for lifespan extension in any species. With the FDA’s recent acceptance of its safety data in January 2026, the precedent is no longer theoretical. And pharma now has billions at stake in molecules that work across multiple age-related diseases through shared mechanisms. The question is no longer whether aging is a druggable target. It’s who owns the platform when the regulatory architecture catches up.
Clinical Trials Enter a Necessary Reinvention
We’ve talked about modernizing clinical trials for years. 2026 is the year we finally make headway across the entire stack as global competitive pressures mount. From reimagining the use of animal models in preclinical development, to trial design and endpoint measurement, to the data packages required for regulatory submission, long-held assumptions are being revisited.
The urgency is real. The US is hemorrhaging clinical trial volume to global competitors. China likely surpassed US trial starts in 2024. Europe’s ACT EU initiative cut approval times while the US remained mired in fragmented IRB processes and 18–24 month institutional negotiations.
FDA’s January 2026 Bayesian guidance, if finalized as proposed, signals a meaningful shift. The agency has laid out a comprehensive framework for using Bayesian methods in primary inference, enabling sponsors to borrow from prior studies, leverage real-world evidence, and augment control arms with external data. The result: smaller trials, faster timelines, and better use of existing information. Approvals like REBYOTA and platform trials like GBM AGILE have already demonstrated that the model works.
The clinical AI advances described above play a role here too. Smaller, more information-dense trials demand more precise endpoint measurement. AI that objectifies traditionally subjective endpoints makes adaptive trial designs more feasible and more powerful. Our portfolio company Atropos Health exemplifies the infrastructure required, collapsing observational research from months to minutes and enabling the kind of continuous evidence generation that Bayesian frameworks reward.
AI Drug Discovery Tests New Business Models
Many are calling 2026 “The Year of Deployment” for AI in bio. In drug discovery, we predict that it will also be the Year of the Business Model. The old binary—own drug assets or sell AI as SaaS—is dissolving. But what replaces it remains unclear.
The deals are coming fast and in new shapes. GSK paid $50 million upfront plus annual licensing fees to access Noetik’s foundation models in a subscription-based framework. Chai Discovery announced a collaboration with Eli Lilly to deploy its frontier AI platform for biologics design. Boltz launched with a $28 million seed round and a multi-year Pfizer partnership, combining open-source foundation models with exclusive, Pfizer-specific versions trained on proprietary data. And NVIDIA and Eli Lilly announced a $1 billion, five-year co-innovation lab to build next-generation foundation models for biology and chemistry.
What’s emerging is a spectrum: pure platform licensing, bespoke model development on proprietary data, co-investment in shared infrastructure, hybrid approaches that retain optionality on assets. The hard question remains unresolved: what percentage of a successful drug’s value should accrue to the AI that helped identify it? The capital required to reach IND filing is measured in tens of millions, whereas the path from IND to approval routinely scales into the billions.
Structuring value-sharing for early-stage contributions versus late-stage development risk will define years of negotiation. By the end of 2026, we’ll have more truth about where enduring enterprise value accrues in AI-enabled drug discovery, and which of these models prove durable.
The In Vivo Renaissance Accelerates
Cell and gene therapy has long been dominated by ex vivo approaches: extract cells, engineer them externally, then reinfuse. While clinically powerful, this model carries structural constraints. Costs routinely exceed $1 million per patient, manufacturing is bespoke, and scalability is limited. In vivo delivery reverses the paradigm, shifting from “manufacture outside, deliver” to “manufacture inside,” with the patient’s body serving as the production system.
In 2026, in vivo cell therapy will remain on a steep upward trajectory, with recent transactions signaling the modality’s maturation. AstraZeneca acquired EsoBiotec for up to $1 billion in March 2025, gaining a platform that delivers cell therapies through a simple IV injection in minutes rather than the weeks required for traditional CAR-T manufacturing. Bristol Myers Squibb followed with a $1.5 billion acquisition of Orbital Therapeutics, an in vivo circular RNA platform. Umoja Biopharma’s UB-VV111, the first in vivo CAR-T candidate to reach US clinical trials, received FDA Fast Track designation for relapsed B-cell malignancies.
The economic implications are decisive. Ex vivo autologous therapies require individualized manufacturing runs, while in vivo approaches can be produced at scale like conventional biologics. By eliminating preconditioning chemotherapy and complex cell handling, in vivo CAR-T opens the door to earlier-line cancer use, autoimmune indications, and outpatient delivery. In 2026, competitive advantage will accrue to platforms that combine precise delivery and tissue targeting, disciplined indication selection, and manufacturing strategies designed for pharma-scale economics rather than boutique pricing.
Conversational Programming Unleashes Clinical and Scientific Creativity
Over the next year, some of the most meaningful productivity gains in healthcare and life sciences will not come from incremental enterprise deployments. They will come from a shift in who builds tools at all. Clinicians and scientists, long treated as end users, will increasingly become authors of software.
Conversational AI coding tools (aka “vibe coding”) have lowered the threshold for creation to near zero. As these systems mature in 2026, we expect a proliferation of micro-applications: narrowly scoped workflow automations, bespoke analysis scripts, internal dashboards, and lightweight reporting tools. Built at the point of friction, deployed immediately, and revised continuously.
The Pareto principle applies. Enterprise platforms capture most economic value by standardizing a small set of high-frequency workflows. What remains is a long tail of heterogeneous, local tasks that resist abstraction but absorb disproportionate time and cognitive effort. Vibe-coded tools make this layer tractable by allowing it to be addressed in place.
A very small subset of these tools may generalize into product-led companies, following the commercialization path of companies like OpenEvidence. Most will remain ephemeral by design, created to solve a specific problem and discarded as conditions change. Their value lies in responsiveness, not durability. Incumbents must respond by moving up-stack, pivoting from providing simple task-based tools to offering integrated, enterprise-grade infrastructure that a custom-built script cannot replicate.
The deeper shift will be institutional. Authority will move from procurement committees optimized to prevent mistakes to practitioners optimized to make progress. Governance will adapt, but the defining story of 2026 is the unlock: those closest to the work will finally be able to encode their understanding directly into software.
Autonomy Expands Its Footprint, Regulators Lace Up
Autonomous systems are moving from concept to deployment across both healthcare delivery and life sciences R&D.
In healthcare, Utah became the first US state to authorize an AI system to autonomously renew prescriptions for select chronic conditions under tightly scoped protocols. It’s a small wedge, but it points toward a larger shift: slowly chipping away at the concept of the medical appointment itself.
In drug discovery, self-driving laboratories are moving from academic curiosity to commercial deployment. Today’s most capable systems automate nearly the entire scientific method, from hypothesis generation and experimental design to execution, analysis, and updating hypotheses for subsequent rounds of discovery. NVIDIA announced collaborations with both Eli Lilly and Thermo Fisher aimed at making scientific instruments intelligent and laboratories increasingly autonomous.
The legal and regulatory apparatus is not yet prepared for either. In 2026, we expect continued experimentation, expanded sandbox programs, and a growing number of pilots testing the boundaries of what is permissible. More importantly, this year will deliver substantive data on whether autonomous systems, from the laboratory to the clinic, can produce outcomes that justify their complexity.