Founder · Scientist · CEO
Yang Liu
He spent three decades as a professor of immunology. Then he decided to build the drugs himself.
The professor who kept the lab coat in the drawer and picked up a balance sheet.
He spent three decades as a professor of immunology. Then he decided to build the drugs himself.
The professor who kept the lab coat in the drawer and picked up a balance sheet.
Why does a single bout of measles grant lifelong immunity, while a tumor sitting in plain sight can grow for years without the immune system raising a hand? Yang Liu has chased that asymmetry for thirty years, and at some point the answers stopped being papers and started being patents.
Today he is Chairman, Chief Executive, and Chief Scientific Officer of OncoC4, a clinical-stage biotech in Rockville, Maryland, that turns immune checkpoint biology into actual antibodies. The lead program, gotistobart (also known as BNT316/ONC-392), is a next-generation anti-CTLA-4 antibody now in late-stage trials with BioNTech. It is built on a deceptively simple idea: let CTLA-4 keep doing its day job protecting the body from autoimmunity, while still unleashing the immune system on cancer. Decouple the cure from the collateral damage.
That distinction matters because the first wave of CTLA-4 drugs worked, but came with toxicity that limited how widely they could be used. Liu's wager is that a smarter antibody, one that allows CTLA-4 to recycle rather than degrade, can broaden who gets to benefit. In his words, ONC-392 has "the potential to broaden the reach of CTLA-4-targeting immunotherapy." It is the rare sentence from a CEO that is also a falsifiable scientific claim.
What makes the arc unusual is the order of operations. Most academics consult for biotech and keep the tenured seat warm. Liu left the endowed chairs behind and ran the company. He had held named professorships at four different institutions before he decided the bench was no longer the fastest path from idea to patient.
OncoC4 itself is a spinout of OncoImmune, the company Liu and Pan Zheng built starting in 2000. When Merck acquired OncoImmune in 2020, it was largely for CD24Fc, an immunotherapeutic Liu had developed to calm inflammation from tissue injury. The rest of the portfolio, the cancer-focused antibodies, went into OncoC4. So the new company did not start from a blank page. It started from two decades of discovery, with the inflammation asset carved out and the oncology engine kept whole. That is a founder collecting on a long bet, then immediately placing the next one.
The Rockville company describes itself plainly as a privately held, clinical-stage biotech engaged in discovering and developing novel biologics for cancer, built on the scientific discoveries of Liu and his team. The phrasing is modest for what it represents: a working translation layer between a particular scientist's lifetime of questions and the regulatory machinery that turns molecules into approved medicines.
Because of its specific mechanism of action, we believe ONC-392 has the potential to broaden the reach of CTLA-4-targeting immunotherapy.— Yang Liu, on OncoC4's lead antibody gotistobart
CTLA-4 is one of the immune system's brake pedals. Old antibodies tend to wear the pedal out. Liu's design lets it spring back.
The antibody engages CTLA-4 but allows it to recycle to the cell surface instead of being destroyed, preserving its protective role against autoimmunity.
It depletes regulatory T cells inside the tumor microenvironment, turning an immunologically "cold" tumor "hot" where it counts.
By uncoupling efficacy from the toxicity that limited earlier CTLA-4 drugs, the goal is to make the therapy usable for far more patients.
Read the pipeline as a sentence about Liu's thesis: the immune system already knows how to kill cancer, so the job is to remove the specific brakes the tumor leans on, one checkpoint at a time.
Gotistobart, the BNT316/ONC-392 program co-developed with BioNTech, is the flagship. It is a next-generation anti-CTLA-4 antibody designed to maintain CTLA-4's protective function against autoimmune disease while improving anti-tumor activity, and it has moved into pivotal Phase 3 development as a monotherapy in metastatic non-small-cell lung cancer, alongside combination work in other solid tumors. China's regulator has granted it Breakthrough Therapy Designation. The collaboration with BioNTech, announced in 2023, brought $200 million upfront with development, regulatory, and commercial milestones plus royalties layered on top.
Behind it, OncoC4 has won FDA clearance to begin human testing of AI-081, a PD-1/VEGF bispecific antibody aimed at advanced solid tumors and positioned as a potential best-in-class entrant in one of the hottest categories in oncology. The company has also expanded through a 2024 merger with AcroImmune, broadening its immunotherapy footprint. The common thread is Liu's conviction that precision in mechanism beats brute force: a drug that knows exactly which signal to silence will reach more patients than one that simply hits harder.
When Merck acquired OncoImmune, it did not just write a check and walk away. It put $50M into Liu's brand-new venture. OncoC4 kept the original assets except CD24Fc. That is not an exit. That is a vote of confidence.
OncoC4 was founded by two renowned immunologists and serial entrepreneurs: Liu and Pan Zheng. They have co-authored across hundreds of papers and built two companies side by side. The science and the business carry the same fingerprints.
Liu was first to identify a major CTLA-4 ligand, CD86, and later mapped the CD24-Siglec-10 innate immune checkpoint, a "don't eat me" signal cancers exploit. Both became platforms, not just publications.
Liu pioneered human gene knock-in mouse models so that human-targeting antibodies could be tested in a meaningful preclinical setting. When the tools did not exist, he made them.
OncoC4 is not a one-molecule company. Alongside gotistobart sit a PD-1/VEGF bispecific (AI-081) cleared by the FDA and additional Siglec and checkpoint programs. The breadth is the strategy.
Colleagues describe a leader who runs on data-driven decisions, openness to many methods, and stubborn persistence, bringing the rigor of academia into the corner office without losing the science.
One question - why microbes grant lifelong immunity and cancer does not - has powered a thirty-year career and two companies. He never really changed the subject.
He trained under Charles Janeway, one of the towering figures of modern immunology. The lineage shows up in the way he treats every drug claim as a hypothesis.
Founder twice over with the same co-founder, Pan Zheng. Most partnerships do not survive one company. Theirs is on its second.
There is a familiar division of labor in biomedicine. Academics discover; companies develop; the two camps rarely share a body. Liu's career is a quiet argument against that split. He spent years building the human knock-in mouse models that let human-targeting antibodies be tested honestly. He identified CD86 as a CTLA-4 ligand and later the CD24-Siglec-10 axis, the "don't eat me" signal that lets tumors hide from the immune system's cleanup crew. Each of those was a discovery another founder might have licensed out and watched from the sidelines. He kept them, and built the companies that would carry them into trials.
The ambition is not abstract. It is to take immune checkpoint biology, the field he helped found, and engineer drugs that separate benefit from harm well enough that immunotherapy stops being reserved for the patients healthy enough to tolerate its side effects. If gotistobart works the way the design intends, the win is not a single approval. It is a wider door. That is the kind of goal that reads like marketing until you remember the man saying it has 200 papers and two companies backing the claim.
He came a long way to say it. From Wuhan University to a master's in Beijing, a PhD in Canberra, a postdoc in New Haven, and a string of American medical schools, the route was anything but direct. What stayed constant was the question he started with and the refusal to leave its answer to someone else. The professor put the lab coat in a drawer not to leave the science, but to follow it all the way to the patient.