Why Life Sciences, Why Now: The Thesis in Full
LeverVenture invests across five life sciences and healthcare sectors, each with one underwriting gate, and treats artificial intelligence as the accelerant inside all five.
By José Vasquéz, Managing Partner
10 min read · Life Sciences and Healthcare
In this note07 · 10 min
Life sciences and healthcare is the whole of the LeverVenture mandate because the decisive risk in the sector now sits after the science, in commercial execution, and the rules that govern execution were largely settled during 2025 and January 2026. For life sciences investors, the thesis reduces to five sectors, one underwriting gate in each, and artificial intelligence as the accelerant inside all five.
01The Mandate
LeverVenture is an operator-led, mid-market growth equity firm, and its mandate has five sectors: Biopharma & Therapeutics, Diagnostics & Precision Medicine, Devices & Robotics, Digital Health & Delivery, and Longevity & Neuro. The common purpose is saving, extending and improving lives. Artificial intelligence is the accelerant across all five sectors. It is never a sixth sector, and it is never underwritten as one.
The seat inside that mandate is defined by stage. Venture capital underwrites whether a technology can work. Private equity underwrites whether a working business can be optimized. Between them sits the passage in which a proven technology becomes a company, and that passage is where LeverVenture invests. Cambridge Associates places growth equity between late-stage venture and leveraged buyouts and describes the typical investment as a minority stake using little if any leverage at investment, often with negotiated approval rights over the annual business plan, acquisitions and new debt or equity.
The firm takes minority positions, leads or co-leads where a company wants an operator at the table, and does operating work: the commercial engine, market access, manufacturing and hiring.
The thesis line is ROI², Return on Investment plus Impact. The first return is underwritten the ordinary way. The second is named at entry as one measure, with a baseline, an owner and a reporting cadence, and it reports on the same schedule as revenue, gross margin and cash. It sits under one of three dimensions: access to care, healthy life extended, or system cost removed.
02The Timing Case
The spending base is large and still growing. The Centers for Medicare & Medicaid Services reports that national health expenditure grew 7.2 percent to $5.3 trillion in 2024, or $15,474 per person, and accounted for 18.0 percent of gross domestic product. Its constraint is the speed at which a new product can reach them.
Capital has gathered at the two ends of the company life cycle. Bain & Company estimates that disclosed healthcare private equity deal value exceeded $191 billion in 2025, surpassing the previous high in 2021, with an estimated 445 buyouts and a rise in deals above $1 billion driving the record. At the venture end, Rock Health counts $14.2 billion of funding for U.S. digital health startups in 2025 across 482 deals, fewer than the 509 deals of 2024, with the average deal at $29.3 million against a median of $12 million.
Those figures describe concentration at both ends. In LeverVenture's view, the company that has proven its product and now needs a commercial organization is served by fewer specialists, and that middle passage is where the mandate sits.
The science side of the risk falls sharply late in development. The 2021 BIO, Informa Pharma Intelligence and QLS Advisors study of 2011 to 2020 programs found a 7.9 percent likelihood of approval from Phase I, against a 90.6 percent probability of approval once a new drug application or biologics license application had been filed. Growth capital that enters after a positive pivotal readout underwrites execution, and execution can be diligenced.
The rules that govern execution also became more legible. On March 31, 2025, the Eastern District of Texas vacated FDA's rule on laboratory-developed tests in American Clinical Laboratory Association v. FDA, holding that laboratory professionals who develop test protocols are carrying out professional services and are not manufacturing devices. FDA then reverted its regulation to the prior text, effective September 19, 2025.
For devices, the Quality Management System Regulation became effective on February 2, 2026, amending 21 CFR Part 820 to incorporate ISO 13485:2016 by reference and aligning FDA's requirements with the standard used by other regulators. For therapeutics, the first Medicare negotiated prices, for 10 drugs covered under Part D, took effect on January 1, 2026. A growth investor can price a risk only when the rule behind it is known.
03The Underwriting Gates
For each sector, the table names the one gate that most often decides the investment, the condition a company must already meet at entry, and the governing authority. A company that fails its gate is declined, and the reason is written into the file.
| Sector | Already true at entry | The one underwriting gate | Governing authority |
|---|---|---|---|
| Biopharma & Therapeutics | An approved product in early launch, or a late-stage asset with positive pivotal data and a defined filing plan | The exclusivity and negotiation calendar: the revenue window from approval to generic or biosimilar entry and to Medicare negotiation eligibility must fund the launch and the manufacturing it requires | 21 U.S.C. 355(j)(5)(F)(ii), five years for a new chemical entity; 42 U.S.C. 262(k)(7)(A), twelve years for a reference biologic; 21 U.S.C. 360cc(a), seven years of orphan exclusivity; 42 U.S.C. 1320f-1(e)(1), negotiation eligibility at 7 years for a drug and 11 for a biologic |
| Diagnostics & Precision Medicine | A validated test run under a CLIA certificate, with paying customers | Coverage evidence: analytical validity, clinical validity and clinical utility at the level Medicare treats as reasonable and necessary, modeled on the contractor timeline | 42 U.S.C. 263a(b); 42 U.S.C. 1395y(a)(1)(A); Palmetto GBA LCD L35025 (MolDX); ACLA v. FDA (E.D. Tex. 2025) |
| Devices & Robotics | An authorized device, through 510(k), De Novo or premarket approval, built under a working quality system | Payment per procedure, settled separately from the clearance; for a surgical robot, what an installed system earns across the procedures performed with it | 21 U.S.C. 360c, device classes I to III; 21 CFR Part 820, as amended by the Quality Management System Regulation; CMS guidance that a new code does not imply coverage |
| Digital Health & Delivery | A proven paying channel with a renewal history | Channel durability: which payer pays, under which code, contract or model, and how much revenue rests on temporary federal flexibilities | 42 CFR 410.78, Medicare telehealth services; 90 FR 61301, the fourth temporary extension of telemedicine prescribing flexibilities for controlled medications |
| Longevity & Neuro | Technical risk retired and a named indication in view | Indication discipline: a named disease, population and endpoint that can support a label, with the Medicare evidence requirement modeled beside the FDA pathway | 21 U.S.C. 355(d), substantial evidence; 21 CFR 314.510, accelerated approval; CMS NCD 200.3, coverage with evidence development |
In biopharma, the statute sets the clocks. No abbreviated application may reference a new chemical entity for five years after its approval, a biosimilar approval cannot take effect until 12 years after the reference product was first licensed, and an orphan drug receives seven years of exclusivity for the same use. The Inflation Reduction Act added a second clock: a drug becomes eligible for negotiation once 7 years have elapsed since approval, and a biologic once 11 years have elapsed since licensure.
In diagnostics, certification comes first: no laboratory may accept human specimens for examination unless a CLIA certificate is in effect. Coverage then decides the business. Medicare excludes items that are not reasonable and necessary for diagnosis or treatment, and the MolDX program states that it will cover only tests that demonstrate analytical and clinical validity, and clinical utility, at that level.
In devices, the statute assigns every device type to one of three classes, from general controls to premarket approval, and the class sets the authorization route. Authorization is only half the gate. CMS states that coding is distinct from coverage and that a new code does not automatically imply coverage by any payer, so payment is underwritten as its own question.
In digital health, Medicare defines the core modality as two-way, real-time interactive communication between patient and practitioner, and much of the sector's revenue rests on flexibilities that Congress and the agencies extend for fixed terms. The Drug Enforcement Administration and HHS issued a fourth temporary extension of telemedicine prescribing flexibilities, effective January 1, 2026, and a clinical model that depends on them carries a dated risk.
In longevity and neuro, a therapy aimed at the biology of aging still reaches patients through a labeled use, because approval requires substantial evidence from adequate and well-controlled investigations. FDA may approve on a surrogate endpoint reasonably likely to predict clinical benefit, subject to confirmatory study, and Medicare has its own evidence bar: since April 7, 2022, NCD 200.3 has covered anti-amyloid antibodies for Alzheimer's disease only under coverage with evidence development.
The gate is the first filter and never the whole file. Every opportunity is also scored on the six LeverRating dimensions, Team, Market, Product, Traction, Financial and Thesis fit, which expand to ten for a regulated asset so that regulatory position, clinical evidence, reimbursement and freedom to operate each carry their own score.
04Crossing Sectors
The strongest companies inside the mandate rarely sit in one sector, and the thesis is built for them. A healthcare-driven company that is also deep tech in its instrumentation, that is also digital in its delivery, or that carries a clean tech dimension, with an artificial intelligence layer that makes it work, is the company LeverVenture is looking for. Each of those disciplines is a characteristic of a company inside the mandate, never a sector of the mandate, and the firm does not invest in any of them apart from a health outcome.
A care platform may ship its own regulated device, and a surgical robot is a device, a software system and a training business at once. In each case the health outcome is the core of the business, and the other discipline is how the company delivers it.
The underwriting consequence is that the whole company is diligenced as a single object. A care platform with a regulated device inside it is scored under both the six-dimension and the ten-dimension rulers, and a rating that holds under only one of them is reported as unresolved. A specialist that can read only one half of such a company will either pass on it or price the unread half as a guess. Reading both halves is part of the edge the thesis claims.
05Artificial Intelligence as the Accelerant
Artificial intelligence compresses the slowest steps inside each sector, from trial enrollment and image reading to surgical planning and clinical documentation. It changes how fast a company reaches its gate. It does not change what the gate requires.
The market has already priced the label. Rock Health reports that AI-enabled digital health companies captured 54 percent of total funding in 2025, up from 37 percent the year before, and commanded a roughly 19 percent premium in average deal size. When the label commands a premium on its own, the record behind the model is what separates one company from the next.
LeverVenture credits a compressed timeline only when it appears in a dated record: an authorization, a coverage decision or a signed contract. For a model that will change after authorization, typically one regulated as software as a medical device, the statute offers a path. Under section 515C of the Federal Food, Drug, and Cosmetic Act, a change to an approved device that is consistent with an approved predetermined change control plan does not require a supplemental application. Diligence therefore asks whether planned retraining sits inside an authorized plan, who holds the rights to the training data, and whether performance held at sites other than the development sites.
06ROI² and the Exclusions
The second return is held to the same standard as the first. A claim of system cost removed survives diligence only if a payer or a provider is measurably better off. A claim of healthy life extended is tracked against a named clinical or real-world endpoint. A claim of access to care is measured as reach into a defined population, which is a different number from an addressable market.
In several of the five sectors, the impact case and the commercial case point at the same patients. The Global Burden of Disease Study 2021 found that conditions affecting the nervous system were the leading group cause of disability-adjusted life-years worldwide in 2021, affecting 3.40 billion people, or 43.1 percent of the global population. A company that changes outcomes for even a defined fraction of that population is building revenue and impact on the same evidence.
The thesis also states what it rules out. LeverVenture does not underwrite unproven science, because a question about whether the biology works belongs to venture capital. It does not take passive positions, because the product is operating support inside the company. It does not treat artificial intelligence as a sector. It does not publish the names of companies it has reviewed or the files behind its ratings.
07Frequently asked questions
What does LeverVenture invest in?
LeverVenture makes minority growth equity investments in life sciences and healthcare companies across five sectors: Biopharma & Therapeutics, Diagnostics & Precision Medicine, Devices & Robotics, Digital Health & Delivery, and Longevity & Neuro. It enters after technical risk is retired, when the next constraint on the company is commercial.
Is artificial intelligence one of LeverVenture's sectors?
No. Artificial intelligence is the accelerant across all five sectors and is never a sector of its own. The firm underwrites a model through the record of the product it sits in: its authorization, its coverage and payment, its change control plan and the rights to its training data.
Does a company have to fit only one sector?
No. A healthcare-driven company that is also deep tech in its engineering, digital in its delivery or clean tech in its footprint, with an artificial intelligence layer, is the company the thesis is built for. The health outcome must be the core of the business, and the whole company is diligenced as a single object.
Nothing in this piece is investment, legal, tax or accounting advice, and nothing in it is an offer to sell or a solicitation of an offer to buy any security.
Former Chairman of JumpStart Inc. He has founded, scaled and exited companies across media, nutrition, real estate and human-resources technology, and owned the revenue line in every one of them. What he brings to a position is that operating experience at scale, investment acumen, an advanced diligence capability, and technology and artificial-intelligence expertise. He invests the way he operated: inside the business, on the plan, alongside the management team.