Nobel Laureate Spinout TwoStep Harnesses Knottin Peptide That Targets Cancer Like No Drug Before

Caitlyn Miller's company is built on a serendipitous discovery. A Stanford lab set out to find a peptide that binds a single cancer receptor and ended up with one that binds 5. Today that peptide anchors TwoStep Therapeutics, an 8-person spinout that has raised $71.2 million and has FDA IND clearance to begin human trials. In an exclusive conversation with SynBioBeta founder and CEO John Cumbers, Miller, TwoStep's co-founder and CEO, explained how it happened.

Miller did her PhD and postdoc at Stanford with Jennifer Cochran, a protein engineer and former chair of Stanford's bioengineering department, and Carolyn Bertozzi, who shared the 2022 Nobel Prize in Chemistry for click and bioorthogonal chemistry. Together with oncologist Ronald Levy, whose lab Miller collaborated with, she and her mentors founded TwoStep. Miller went on to become the first entrepreneur in residence at Stanford's Innovative Medicines Accelerator who also launched the company from there.

The peptide at the center of TwoStep is called PIP, short for polyspecific integrin-binding peptide. It is 33 amino acids long and weighs about 3.3 kilodaltons, roughly 1/45 the size of an antibody. Its story starts in Jennifer Cochran's lab.

PIP is a knottin, a small peptide whose 3 disulfide bonds fold it into a tight knot. Cochran's lab was using yeast surface display to evolve peptide binders to integrins, a family of adhesion receptors on the cell surface, as new cancer imaging agents. When the lab took its top binders into mouse imaging studies, one kept beating the rest.

"And what's interesting is that at the time, the lab didn't quite know what was so special about that one peptide, but it kept showing the highest tumor uptake and also very clean biodistribution in general. So the lab kept testing it, because there was clearly something going on."

In mice, it lit up spontaneous breast, lung, pancreatic tumors, even intracranial brain tumors, Miller said.

"And it was only after some of that foundational work that the lab actually realized the reason why it had that unique profile was because the peptide was able to bind to five different tumor-associated integrin targets (ɑvβ1, ɑvβ3, ɑvβ5, ɑvβ6, ɑ5β1), even though that wasn't originally the design goal of the project. It was a serendipitous discovery."

Binding multiple targets matters because of how certain integrins behave in cancer. Several are overexpressed in tumors, where they help cancer cells proliferate, migrate and survive. That has made certain integrins become popular targets for antibody-drug conjugates, or ADCs, and radioligand therapies. One of the furthest along, Pfizer's sigvotatug vedotin, is an ADC against a single integrin, ɑvβ6. In June Pfizer reported that the drug missed its primary overall survival endpoint in a Phase 3 trial in previously treated non-squamous non-small cell lung cancer. Pfizer is still testing it in another Phase 3 trial in first-line lung cancer alongside pembrolizumab.

Due to PIP’s multi-specific binding, it has the potential to provide broader and more consistent tumor recognition across heterogeneous tumors compared to a conventional single antigen targeting approach. For example, in one study, TwoStep analyzed integrin expression across cancer types and only about 1 in 3 cancers carried enough ɑvβ6 for target-mediated activity. PIP binds ɑvβ6 and 4 other tumor-associated integrins, and Miller likened the result to an “OR” gate. A tumor needs to express only one of them. In one panel of solid tumors, the company says, every tumor expressed at least one.

"So, for instance, a colorectal cancer patient may have tumors expressing two of the five integrin targets. If you go to the next patient, other target integrins may be expressed and that's okay. We can still use the same molecule broadly across different patient populations."

Size is the other half of the bet. Antibodies circulate for weeks but often show poor penetration into dense tumors. Peptides rapidly penetrate into tumors and clear from systemic circulation within hours. PIP is fully synthetic, made by chemistry rather than in cells. For some products, TwoStep swaps in one unnatural amino acid as a handle for adding payloads, a change Miller said does not disrupt binding.

Most targeted drugs let receptor expression narrow the list of cancers worth pursuing. PIP could in principle reach almost any solid tumor, so for TwoStep’s lead program, the company prioritized indications based on the payload instead. The lead program’s payload is MMAE, the cytotoxic payload in several approved ADCs.

"MMAE is a really great drug, and it's validated, which is part of why we went after it. It works really well in many cancer types, but it doesn't work across all cancer types."

TwoStep narrowed down the list and decided to evaluate its lead drug in 6 cancer types. The first-in-human study of TS-104 (TwoStep’s PIP-MMAE conjugate) is a dose-escalation and dose-expansion Phase 1 trial designed for 74 patients with non-small cell lung, head and neck squamous cell, esophageal, gastric, endometrial or ovarian cancer. 

The second program swaps the chemotherapy for a radioactive isotope. Because an imaging isotope can ride on the same peptide, a patient can be scanned first to confirm that the tumors light up before receiving a therapeutic dose. The field calls this theranostics and sums it up as "see what you treat." TwoStep plans a human imaging study as proof of concept in 2027.

In September, the San Carlos (California) company announced an oversubscribed $62.5 million Series A, led by Insight Partners and Medical Excellence Capital with Pfizer Ventures and M Ventures, the venture arm of Merck KGaA, as co-leads. NFX, 2048 Ventures and Stanford University also took part. The round was announced alongside FDA IND clearance for TS-104.

Miller said TwoStep expects proof-of-concept clinical readouts from both programs within about 2 years, which will shape the next financing and any partnering. She is already thinking about what comes after the first 2 programs, and about how far the targeting peptide can go.

"A lot of targeted therapies are pretty limited in scope in terms of  what tumors they can address. And so we have the opportunity here to broaden the reach of targeted therapy for more cancer types with this type of technology."