What is B7-H3 and Its Role in Cancer Therapy?

B7-H3, also known as CD276, is a protein on the surface of cells that belongs to the same family of immune-regulating molecules as PD-L1, one of the most successful targets in modern cancer treatment. What makes B7-H3 so attractive to researchers is a striking pattern: it is barely detectable on healthy tissues but heavily present on a wide range of tumors. That combination of broad tumor coverage and limited normal-tissue expression has made it one of the most actively pursued targets in cancer immunotherapy today, with dozens of clinical trials testing drugs designed to exploit it.

A Member of the B7 Family

Your immune system relies on a network of molecular signals to decide when to attack and when to stand down. The B7 family of proteins is central to that decision-making. Some members of this family, like B7-1 and B7-2, activate immune cells. Others, like PD-L1 (also called B7-H1), put the brakes on them. B7-H3 sits in an unusual position: it can apparently do both, acting as a costimulator or a coinhibitor of T cells depending on the context.¹ Under normal conditions, B7-H3 messenger RNA is broadly distributed across many tissues, but the protein itself is tightly controlled at the post-translational level, meaning the body keeps it on a short leash.² In cancer, that leash snaps. Tumors ramp up B7-H3 protein production in ways that healthy tissue does not.

In humans, B7-H3 comes in two structural forms. The shorter version has a pair of immunoglobulin-like domains, while the longer version doubles that architecture to four. The four-domain form is by far the more common one on human tumor cells and most immune cell types.³ This matters for drug design, because antibodies and other therapies need to recognize the version of the protein they will actually encounter on tumors. Most drugs in development target the four-domain form specifically.⁴

Why Tumors Love B7-H3

The reason B7-H3 has drawn so much attention is its expression profile. When researchers have stained large panels of normal human tissues with antibodies that detect B7-H3, they find only faint, scattered signals in a handful of organs, including the stomach, liver, pancreas, and adrenal gland.⁵ In contrast, B7-H3 is consistently and abundantly expressed across many different cancer types, and that overexpression tracks with worse outcomes for patients.⁶ This disparity is the foundation of the entire therapeutic strategy: if a drug zeroes in on B7-H3, it should find plenty of target on the tumor and very little on healthy organs.

B7-H3 does not simply sit passively on the tumor surface. It actively helps cancer in at least two distinct ways. First, it suppresses the immune system. Research has shown that B7-H3 inhibits T cell activation and proliferation, helping tumors evade immune detection. More recent work has revealed that B7-H3 also manipulates other immune players: it can trigger a signaling cascade in certain suppressive immune cells, driving them into a state that further shields the tumor from attack.⁷

Second, B7-H3 has direct, non-immune effects on tumor biology. It promotes tumor cell proliferation, invasion, metastasis, drug resistance, and the formation of new blood vessels that feed the tumor.⁸ In melanoma experiments, silencing B7-H3 dramatically reduced the cancer’s ability to spread. Mice injected with melanoma cells lacking B7-H3 survived more than twice as long as those injected with B7-H3-expressing cells.⁹ So blocking B7-H3 could, in theory, both unleash the immune system against the tumor and undermine the tumor’s own growth machinery.

The Missing Receptor

One of the biggest gaps in B7-H3 biology is surprisingly basic: researchers still do not definitively know what receptor B7-H3 binds to on immune cells. Most other B7 family members have clearly identified partners, but B7-H3’s binding partner has remained elusive for over two decades. Recent work using proximity-labeling techniques has suggested that CD45 (a common marker on immune cells) and EGFR (a growth factor receptor on cancer cells) may interact closely with B7-H3, but these are still classified as putative interactions rather than confirmed receptor identities.¹⁰ This uncertainty has not slowed down drug development, since therapies can target the B7-H3 protein itself regardless of what it binds, but it does mean the full picture of how B7-H3 manipulates immunity is incomplete.

What High B7-H3 Levels Mean for Prognosis

A meta-analysis pooling data from studies across multiple cancer types found that patients with high B7-H3 expression in their tumors had roughly 60 percent higher risk of death and roughly 67 percent higher risk of disease progression compared to those with low expression.¹¹ The prognostic picture is not perfectly uniform across every tumor type, though. In colorectal cancer, high B7-H3 expression was associated with worse survival at both two-year and six-year time points.¹² But in bladder cancer, a separate meta-analysis found no clear link between B7-H3 levels and overall survival.¹³ These discrepancies likely reflect real biological differences in how B7-H3 functions across different tumor environments, and they caution against treating B7-H3 as a one-size-fits-all biomarker.

How B7-H3 Compares to PD-L1

The obvious question for anyone familiar with cancer immunotherapy is how B7-H3 stacks up against PD-L1, the target behind blockbuster drugs like pembrolizumab and nivolumab. They are relatives in the same protein family, but their behavior diverges in clinically meaningful ways. In small cell lung cancer, for instance, B7-H3 was detected in about 65 percent of cases, while PD-L1 showed up in only about 7 percent, and the two markers showed limited overlap.¹⁴ A similar uncorrelated expression pattern was seen in pancreatic cancer, where B7-H3 and PD-L1 were both upregulated but seemed to operate independently.¹⁵

This independence is actually encouraging. It means B7-H3 could be relevant precisely in the tumors where PD-L1-targeting therapies fall short, since many patients whose cancers express little or no PD-L1 may still express abundant B7-H3. Rather than replacing PD-L1-directed treatments, B7-H3 therapies could extend the reach of immunotherapy to patients who currently have limited options.

Antibody-Drug Conjugates Leading the Clinical Pipeline

The most advanced therapeutic approach against B7-H3 right now is the antibody-drug conjugate, or ADC. An ADC is essentially a guided missile: an antibody that finds and sticks to B7-H3 on the tumor surface, carrying a toxic payload that gets released inside or near the cancer cell. Several B7-H3-targeting ADCs are in clinical trials, and early results have been striking.

In a phase 1/1b trial of YL201, a B7-H3-targeting ADC carrying a topoisomerase I inhibitor, about 41 percent of evaluable patients with advanced solid tumors achieved an objective response, meaning their tumors shrank meaningfully. The disease control rate, including patients whose tumors shrank or at least stopped growing, was about 84 percent. Median time before the disease worsened was about six months.¹⁶ A separate phase 1/2 trial of another B7-H3 ADC, SYS6043, reported a similar overall response rate of about 42 percent across more than 500 patients, with activity seen in lung cancer, breast cancer, gynecologic cancers, and nasopharyngeal cancers among others.¹⁷ Preclinical work with a third ADC, DS-7300a, demonstrated that it specifically killed B7-H3-positive cancer cells while sparing B7-H3-negative ones, confirming the targeting mechanism.¹⁸

These response rates are notable given that many of the enrolled patients had cancers that had already progressed through prior treatments. For context, tumor-agnostic response rates above 40 percent in heavily pretreated patients with solid tumors would be unusual for a single-agent therapy.

CAR-T Cell Therapy for Pediatric Solid Tumors

B7-H3 has become an especially important target in childhood cancers, where treatment options for solid tumors are limited and checkpoint inhibitors like anti-PD-1 drugs have been largely disappointing. B7-H3 is highly expressed in pediatric solid tumors including osteosarcoma, rhabdomyosarcoma, Ewing sarcoma, Wilms tumor, and neuroblastoma.¹⁹ Researchers have engineered chimeric antigen receptor T cells, or CAR-T cells, to recognize B7-H3, and preclinical studies have shown potent results. In animal models, B7-H3-directed CAR-T cells caused regression of established tumors including osteosarcoma, medulloblastoma, and Ewing sarcoma.²⁰

A reassuring finding from those preclinical studies is that the CAR-T cells’ killing ability depended on high levels of B7-H3 on the target cell surface. When B7-H3 expression was low, the CAR-T cells were much less active. Since normal tissues express B7-H3 at very low levels compared to tumors, this suggests a built-in safety margin where the engineered T cells would largely spare healthy organs. Multiple clinical trials testing B7-H3 CAR-T cells in children are now underway.²¹ The main challenges are ones common to all CAR-T therapy for solid tumors: getting the engineered cells to infiltrate the dense tumor mass, keeping them alive and functional inside the hostile tumor microenvironment, and preventing the tumor from developing resistance.

Radioimmunotherapy for Brain Tumors

One of the earliest clinical applications of B7-H3-targeted therapy involves a radioactive antibody called omburtamab (formerly known as 8H9). This antibody carries iodine-131, a radioactive isotope, and is injected directly into the fluid-filled spaces of the brain to treat cancer that has spread to the central nervous system. In a phase 1 study of patients with neuroblastoma that had metastasized to the brain, those who received omburtamab had a median progression-free survival of seven and a half years, a substantial improvement over what has historically been expected for that disease.²² Researchers are investigating B7-H3-targeted therapies in other brain cancers as well, including glioblastoma, a notoriously aggressive adult brain tumor, and diffuse intrinsic pontine glioma (DIPG), a devastating childhood brain cancer with almost no effective treatments.²³

Bispecific Antibodies and Combination Strategies

Another class of drugs in development uses bispecific antibodies, which are engineered to grab two different targets at once. Some are designed to bridge B7-H3-positive tumor cells with T cells by also binding CD3, a protein on T cells, physically pulling immune cells into contact with the cancer. One such molecule, CC-3, was optimized so that it binds B7-H3 tightly but has deliberately weakened affinity for CD3, which in lab models led to potent tumor killing while reducing the uncontrolled flood of inflammatory signals that can cause dangerous side effects.²⁴ Other bispecific designs pair B7-H3 with CD28, a co-stimulatory receptor, to boost T cell activation within the tumor rather than simply recruiting T cells there.²⁵

Combining B7-H3-targeted therapy with existing checkpoint inhibitors is also being explored. In an early-phase trial, the B7-H3 antibody enoblituzumab was combined with the anti-PD-1 drug pembrolizumab in patients with head-and-neck or non-small-cell lung cancers who had never received a checkpoint inhibitor. About a third of patients in each cancer type had objective tumor responses, and the combination was well tolerated.²⁶ Preclinical data in prostate cancer models have shown that combining B7-H3 blockade with PD-L1 or CTLA-4 inhibitors achieved durable, potentially curative responses in tumors carrying specific genetic vulnerabilities.²⁷ The rationale for combination makes intuitive sense: if tumors use multiple independent brake pedals to slow the immune system, releasing two brakes at once should free up immune cells more effectively than releasing just one.

Safety Signals So Far

The limited expression of B7-H3 on normal tissues was supposed to translate into fewer side effects than drugs targeting more widely expressed proteins, and the early data partly bears that out, though with some caveats. In the YL201 ADC trial, nearly all patients experienced treatment-related side effects, most commonly drops in white blood cell counts and anemia. About 55 percent of patients had grade 3 or higher adverse events, with low neutrophil counts being the most frequent serious issue. However, the rate of interstitial lung disease, a worrying complication seen with some other ADCs, was strikingly low at about 1 percent.¹⁶ Whether these side effects stem from the B7-H3 targeting, the toxic payload, or a combination is still being sorted out. As these drugs move into larger trials, understanding exactly which toxicities are target-related and which are payload-related will be critical for choosing the right patients and managing treatment safely.

The broader challenge, as one review puts it, is that the molecular mechanisms behind B7-H3’s tumor-promoting properties are still not fully understood, which makes predicting and managing therapy toxicity harder than it would be with a perfectly characterized target.²⁸

Using B7-H3 to Find and Track Tumors

Beyond therapy, B7-H3 is also being developed as a way to visualize tumors and monitor disease. In preclinical prostate cancer models, researchers labeled a B7-H3-targeting antibody-drug conjugate with zirconium-89, a radioactive tracer, and used PET scanning to image tumors. The B7-H3-positive tumors lit up brightly and clearly, with signal increasing over time, while B7-H3-negative tumors showed minimal uptake.²⁹ An early clinical study using a different B7-H3 PET tracer in 20 patients with malignant tumors found that B7-H3-targeted imaging actually outperformed standard FDG-PET in overall diagnostic efficiency for certain cancers, including some liver and gastric cancers. It detected B7-H3 expression with about 84 percent sensitivity and 100 percent specificity at an optimized threshold.³⁰

The practical appeal here is twofold. First, imaging could tell oncologists in advance whether a patient’s tumor expresses enough B7-H3 to be a good candidate for B7-H3-targeted therapy, sparing non-responders from futile treatment. Second, the same tracer concept could be adapted for therapy by swapping the imaging isotope for a therapeutic one, combining diagnosis and treatment in a single platform.

Blood-Based Biomarkers

Beyond imaging, researchers are investigating whether B7-H3 can be measured from a simple blood draw. Tumors shed fragments and tiny vesicles into the bloodstream, and some of these carry B7-H3. In non-small-cell lung cancer patients, blood levels of soluble B7-H3 were significantly higher than in healthy controls and correlated with disease stage, lymph node spread, and PD-L1 status. Elevated levels were also an independent predictor of worse survival in advanced disease.³¹

In prostate cancer, a newly developed assay measuring B7-H3-positive extracellular vesicles in blood found that patients with more advanced, treatment-resistant disease had higher levels. These levels tracked with treatment response over time and predicted worse survival. The researchers propose this assay as a potential companion diagnostic that could be used alongside B7-H3-targeted drugs to select the right patients and monitor whether treatment is working.³² If validated in larger studies, blood-based B7-H3 measurement could offer a less invasive alternative to repeated tissue biopsies for tracking how tumors respond to therapy.

Resistance and the Tumor Microenvironment

One emerging concern is that tumors may find ways to resist B7-H3-targeted therapy, and the mechanisms might be more complex than simply losing B7-H3 expression. Recent research in animal models has revealed a feedback loop: when B7-H3 on tumor cells triggers suppressive immune cells, those immune cells, in response to B7-H3 blockade, can secrete factors that activate support cells called fibroblasts in the tissue surrounding the tumor. These activated fibroblasts remodel the physical structure of the tumor’s neighborhood in ways that may dampen the effectiveness of treatment.⁷ This kind of compensatory resistance, where blocking one pathway inadvertently strengthens another, is a recurring theme in cancer treatment. Understanding these escape routes early in the development process could help researchers design combination strategies that close multiple exits at once, rather than pushing the tumor toward alternative survival pathways.