ARPA-H's BoSS Program Is Trying to End Ultra-Cold Storage for Cell and Gene Therapies. Why Its Other Transaction Structure and Phased Down-Selects Will Reshape Who Can Win.

The promise of cell and gene therapies has run, in clinical practice, into a banal logistical wall: most of them have to be stored at -150°C or colder until the moment they enter a patient. That requirement turns every CAR-T infusion into a small supply-chain miracle. Cryo-preserved doses ship from centralized manufacturing plants in vapor-shipper dewars, transit through specialty couriers, and have to be thawed at the bedside on a schedule measured in minutes. Hospitals outside major academic medical centers — and any geography that lacks a same-day cold-chain corridor — are functionally locked out.

That logistical wall is the explicit target of ARPA-H's BioStabilization Systems (BoSS) program, a four-year, multi-phase initiative now in active execution after a Proposers' Day on January 29, 2026, Solution Summaries due February 19, and Solution Pitches due March 26. The most recent BoSS deadline in the public record — Independent Verification and Validation (IV&V) partner submissions due May 15, 2026 — has just passed. The active phase of the program is now opening, and it is one of the most structurally distinctive federal biomedical opportunities of 2026.

Understanding why BoSS is structured the way it is, and who the program is designed to fund, is essential for any team thinking about how to engage with ARPA-H going forward — not just in BoSS itself, but in the dozen-plus programs that share its Other Transaction Authority architecture.

The Technical Problem: Cell Identity Has to Survive the Trip

Modern cell therapies — autologous CAR-T, allogeneic NK-cell, regulatory T-cell, induced pluripotent stem-cell-derived products — require maintaining cellular potency, identity, and viability through an extended logistics chain. The current state-of-the-art uses controlled-rate freezing into dimethyl sulfoxide (DMSO) cryoprotectants, vapor-phase liquid nitrogen storage at temperatures below -150°C, and a thaw protocol that returns the product to physiological temperature inside roughly fifteen minutes. Deviation from any step at any link reduces post-thaw viability and can trigger product release-criteria failures that force a patient's manufacturing slot to be rebuilt from scratch.

The cost of running that infrastructure is one of the main reasons CAR-T list prices sit between $400,000 and $500,000 per dose and why approved cell therapies remain concentrated in academic medical centers with on-site cryogenic capacity.

BoSS asks a more ambitious question than "can we build a better dewar." It asks whether the cells themselves can be reengineered or chemically prepared such that they survive at ambient temperatures for at least 14 days, then be "re-animated" at the bedside without loss of identity, potency, or sterility. If that target is hit, the entire downstream logistics topology changes — cell therapy distribution starts to look more like pharmaceutical distribution than like organ-transplant logistics.

The program splits its technical work into two areas. Technical Area 1 (BioPrep) funds the cellular interventions: cryoprotectant chemistries, lipid manipulation, trehalose loading, induced metabolic dormancy, lyophilization-compatible cell-line engineering. Technical Area 2 (Bioprocessing) funds the instrumentation, container systems, and aseptic-process design needed to translate any TA1 breakthrough into a scaled, GMP-compatible manufacturing format.

ARPA-H expects winning consortia to integrate across both areas, not pursue them in isolation. That requirement immediately narrows the eligible field.

The Architecture: Four Phases, Three Down-Selects

BoSS is being executed over four years against a deliberately punitive milestone structure.

• Phase 1 (15 months) demands a feasibility demonstration that meets a minimum threshold of 14+ days of ambient shelf-life, with cell identity and potency preserved.
• Phase 2a and 2b (24 months combined) push the technology toward integrated bioprocessing at scale, with explicit FDA regulatory pathway alignment.
• Phase 3 (9 months) prepares the technology for market transition — comparability validation, manufacturing scale-up, and regulatory submission readiness.

ARPA-H has made clear that the program will execute hard down-selects after Phase 1, Phase 2a, and Phase 2b. Performers who fail to hit a phase milestone are dropped from the program rather than being given extensions or modified scopes. Funding follows the survivors. This is the inheritance of the DARPA model — and a sharp departure from the typical NIH R01 paradigm in which a single PI's underperformance leads to a no-cost extension and a renewal application rather than program termination.

For teams accustomed to NIH, this is the largest cultural shift to internalize. BoSS is not buying basic research; it is buying technology delivery against a specification.

Why Other Transaction Authority Changes Who Can Compete

BoSS awards are issued through Other Transaction (OT) agreements, not traditional grants and not traditional procurement contracts. OTs were originally created by Congress for DARPA in 1958 and extended to NIH and HHS components in subsequent legislation; ARPA-H inherits and uses them aggressively.

The practical consequences for applicants are substantial.

First, single-PI academic labs are structurally disadvantaged. OTs are written to "multi-party technical teams" with all contributors as signatories. A consortium of three or four organizations — typically one academic translational lab providing the science, one biotech company providing the bioprocessing engineering, one contract manufacturer providing the GMP scale-out capability, and one IV&V partner providing the standardized cell-type and characterization infrastructure — is the modal team that ARPA-H wants to fund. A single university applicant is at a near-fatal disadvantage unless it brings prefabricated industry partnerships into the proposal.

Second, OTs are not subject to the Federal Acquisition Regulation (FAR), the Bayh-Dole Act in its standard form, or the indirect-cost negotiated rates that govern grants. That means intellectual property terms, march-in rights, indirect-cost reimbursement, audit obligations, and termination rights are all negotiated per-award. Universities that have ironclad IP and indirect-cost policies for federal grants often have to renegotiate or accept terms their offices of sponsored research are uncomfortable with. Teams without OT experience routinely lose months in legal negotiation; teams with prior DARPA, ARPA-E, or ARPA-H awards have institutional muscle memory that lets them close in weeks.

Third, the IV&V partner role is unique to OT consortium structures and worth highlighting. ARPA-H separates the performer (the team building the biostabilization technology) from the IV&V partner (an independent third party providing standardized cell types and characterization assays). The IV&V role exists to prevent performer teams from grading their own homework on potency and identity preservation. It also means cell-therapy companies that already have well-characterized standard cell lines and validated assay panels — but no novel biostabilization technology of their own — can be paid by the program as IV&V partners without competing as performers. The May 15, 2026 IV&V deadline was a separate competitive call from the performer Solution Summaries.

Who BoSS Is Actually Designed For

Stripping away the program documents, BoSS funds a specific kind of team: one that has already moved a cell-therapy bioprocessing technology to at least Technology Readiness Level 3 or 4, has GMP capability or a credible partner for it, and can credibly project a regulatory pathway with the FDA's Center for Biologics Evaluation and Research within four years.

That excludes:

• Single-investigator academic labs without industrial partners
• Cryobiology research groups studying mechanisms without translational pathway
• Early biotech startups that have a novel chemistry but no bioprocessing capability
• Contract manufacturers without their own novel scientific contribution

It includes:

• Established cell-therapy companies (Allogene, Kite/Gilead, Bristol Myers Squibb's cell-therapy unit, Lonza, Bluebird, Vor Biopharma) that have internal bioprocessing capabilities and want a non-dilutive R&D vehicle
• Academic translational centers (the cell-therapy program at Penn, the bioengineering teams at MIT and Stanford, the bioprocessing groups at Georgia Tech, NCSU's BTEC) that already have industry partnerships
• Contract development and manufacturing organizations (CDMOs) with R&D arms and existing cell-therapy clients

The community of plausible BoSS performers is on the order of two dozen consortia nationally. The program is, by design, not for everyone.

What This Means for the Broader ARPA-H Funding Pipeline

BoSS is one program. But the structural choices it embodies — multi-party consortia, OT agreements, hard down-selects, IV&V separation, TA-segmented technical work — are showing up across the full ARPA-H program portfolio: IGOR (autonomous biomedical research), PROSPR (geroscience clinical trials), EVIDENT (psychedelic-assisted behavioral health), STOMP (microplastics), AIR (autonomous robotic surgery), ADVOCATE (agentic AI for cardiovascular care). The architectural similarities are not coincidental. ARPA-H has decided this is how it will operate.

For organizations that have only ever interacted with NIH R-series grants or NSF standard awards, internalizing the OT consortium model is the single highest-leverage capacity investment to make in 2026. The deadlines for the next round of ARPA-H programs — including ARPA-H's small business SBIR/STTR pathway with a Solution Summary due July 10, 2026 — are arriving on a rolling basis.

Teams that wait to encounter the model until they read a solicitation will lose. Teams that begin pre-positioning consortia, securing IV&V relationships, and negotiating template OT terms with their institutional research offices now will be on the field when the next BoSS-shaped opportunity lands.

For an ongoing list of ARPA-H solicitations and the programs that share BoSS's architecture, see Granted News.