Tmunity Therapeutics started as an idea in a Penn lab and ended up in Kite Pharma’s hands, with a deal that could be worth up to $1.2 billion.

That arc—from academic CAR T-cell science to a commercial acquisition—didn’t happen because someone filed a patent and waited for pharma to call. It happened because the Penn Center for Innovation (PCI) treated commercialization as a team sport: co-investing, structuring partnerships, and staying involved long after the invention disclosure.

If you’re building anything deep-tech or bio, this is the real story: not “we licensed IP,” but “we built a company that survived the valley of death because the university decided to act like a long-term partner, not a toll booth.”

Cabaletta Bio Shows CAR T Isn’t Just for Cancer

Cabaletta Bio is another Penn spinout that makes this clear. It didn’t start with “let’s build an oncology company.” It started with Penn researchers asking whether CAR T-cell technology could be adapted to treat autoimmune diseases like myasthenia gravis and lupus.

That’s a non-obvious bet. Oncology CAR T is hard enough. Moving into autoimmune means new biology, different risk profiles, and a less defined regulatory and commercial path. It’s exactly the kind of thing that dies in a traditional tech transfer model, where the office optimizes for quick licenses and low risk.

Instead, PCI helped turn that research into Cabaletta Bio, a focused company built around engineered T cells designed to selectively eliminate disease-causing B cells. Same core CAR T-cell platform, very different disease space.

What matters for founders is the infrastructure around it. PCI didn’t say, “We only do cancer.” Their model supports a range of therapeutic areas—oncology, autoimmune, rare diseases—because they’re set up to build companies, not just sign category-specific licenses.

That flexibility shows up in how they structure startups: matching founders with operators and investors, co-investing alongside external capital, and helping navigate early clinical and regulatory strategy. For a CAR T-cell therapy startup outside classic oncology, that support can be the difference between “interesting paper” and “IND-enabling program.”

From Lab Bench to Cap Table: PCI’s Numbers Tell a Story

PCI has been at this for a decade, and the data backs up that this isn’t a one-off success story. According to PCI’s FY2025 Year in Review, they launched 26 startups in that fiscal year alone, with those companies raising more than $591 million in capital.

On top of that, Penn has ranked #1 nationally in licensing revenue for five straight years. That’s not just “we have good lawyers.” It’s a sign that the underlying companies are actually creating value, de-risking assets, and closing real deals.

Underneath those numbers is a deliberate strategy: academic co-investment. Penn doesn’t just hand off IP and hope. As Penn Medicine has laid out in its discussion of academic co-investment, the institution puts its own capital into select startups, aligning incentives with founders and outside investors.

That’s a big shift from the old model where universities tried to maximize upfront license fees and royalties. Here, PCI is effectively saying: we’ll take more risk, own more equity, and work with you over the long haul. It’s closer to a seed fund embedded inside a research university than a classic tech transfer office.

You can also see the partnership strategy in who they bring to the table. Agent Capital, for example, was named PCI’s “Partner of the Year” for 2025, recognizing a multi-year collaboration where they’ve backed multiple Penn-originated biotech startups. In their own writeup of the award, Agent Capital talks about PCI as a true co-builder, not just a source of deal flow.

Geeta Vemuri at Agent has been blunt about this: what makes the relationship work is that PCI is willing to think like an investor—about milestones, capital efficiency, and syndicate quality—while still protecting the academic mission. And John Swartley at PCI has said publicly that their goal is not just to monetize patents, but to get therapies to patients faster through aligned partnerships.

For founders, this matters because it changes the negotiation. You’re not just haggling over royalty rates. You’re talking about board structure, follow-on capital, and how to share risk across a 10–15 year biotech journey.

Biotech’s Long Road, and How PCI Blunts the Worst of It

Biotech is brutal on timelines. From first in-human to approval, you’re often looking at 10–15 years, with the majority of programs failing somewhere along the way. CAR T-cell therapy startups layer on manufacturing complexity, safety concerns, and evolving regulatory expectations.

Most founders underestimate how many times they’ll need to “re-sell” the story: to new investors every 18–24 months, to pharma partners, to regulators, to internal teams who are burned out after another trial pivot. The valley of death isn’t one gap; it’s a series of them.

PCI’s model doesn’t remove that risk, but it does spread it. Academic-industry partnerships mean the science can keep moving even when a company hits a financing wall. Co-investment means the university has skin in the game and a reason to help you find new partners instead of quietly reclaiming the IP.

There are still real dependencies. If key faculty leave, if IP gets tied up, if federal funding priorities shift, your company can feel it. Regulatory shifts for novel modalities like CAR T—around safety monitoring, manufacturing controls, or long-term follow-up—can force expensive redesigns.

The resilience comes from having multiple aligned anchors: the lab, PCI, and a committed investor syndicate. When one leg wobbles, the others can carry more weight for a while. That’s very different from a pure license deal where the university steps back and you’re on your own with a ticking diligence clock.

If you’re a founder in this space, a few pragmatic moves stand out:

• Negotiate for ongoing scientific collaboration with the originating lab, not just an IP license.
• Push for the university (or PCI-equivalent) to co-invest, even modestly, to align incentives over time.
• Bring in investors early who are comfortable with platform plays and multi-indication roadmaps, not just a single-asset flip.
• Design your early milestones around de-risking both science and manufacturability—especially critical for CAR T-cell therapy startups.

None of this guarantees success. But it gives you more shots on goal and more partners who are motivated to keep the program alive when the first plan breaks, which it almost always does.

Why This Model Matters Beyond Penn

The Penn Center for Innovation CAR T-cell commercialization story is not just about one university getting good at spinning out cancer companies. It’s about a structure that consistently moves deep, risky science into the market by sharing risk, capital, and execution with founders and investors.

Co-investment, partnership awards, and a decade-long track record of startups and exits show that this isn’t charity. It’s a repeatable framework that bridges the valley of death in biotech startup formation: from early grants and lab work, through company formation, into clinical milestones that pharma and public markets actually care about.

If you’re working on deep tech—whether it’s cell therapy, gene editing, or something outside of life sciences—the lesson is the same. A pure licensing route leaves you carrying most of the risk with the least leverage. A partnership-driven model, where your institution acts like a long-term co-founder, gives you more room to navigate the inevitable setbacks.

You don’t need to be at Penn to push for that. But you do need to ask harder questions of your tech transfer office, your investors, and your academic partners: are we just signing a contract, or are we actually building a company together?