Scientists create antibody that makes tumors glow during PET scans

Researchers have engineered a miniature antibody that targets a protein called EphA2, which is overproduced by many tumor types, and causes those tumors to produce a visible signal during PET imaging. The study was published on March 16. In mouse tests, tumors with elevated EphA2 expression lit up clearly on PET scans after the antibody was administered, while surrounding healthy tissue did not. The result is a potential diagnostic tool that could help clinicians detect cancer earlier and with greater spatial precision than current imaging agents allow.

PET scanning already exists as a diagnostic method, and cancer imaging agents are not new. What makes this development different is the target. EphA2 is overexpressed in a wide range of solid tumors, including lung, breast, ovarian, prostate, and pancreatic cancers, but is present at low levels in most healthy adult tissue. That specificity is exactly what a good imaging agent needs: strong signal where the tumor is, minimal background noise where it is not.

What EphA2 is and why it makes a useful imaging target

EphA2 is a receptor tyrosine kinase, a class of proteins embedded in cell membranes that receives and transmits growth signals. In normal tissue, EphA2 is expressed at low levels and plays a role in cell organization and wound repair. In many cancers, the gene that encodes EphA2 is overactivated, causing the protein to appear at concentrations ten to fifty times higher than in surrounding normal cells, according to data from The Cancer Genome Atlas. That overexpression is associated with more aggressive tumor behavior, worse prognosis, and higher rates of metastasis in several cancer types.

Because of those properties, EphA2 has been studied for years as a potential therapeutic target. Several drug companies have worked on antibody-drug conjugates designed to deliver chemotherapy directly to EphA2-expressing cells. Most of those programs focused on treatment rather than detection. The new study took a different approach, using a miniaturized antibody fragment called a nanobody, which is smaller than a full antibody and clears from the bloodstream faster, to build an imaging probe rather than a drug carrier.

How the nanobody imaging probe works

Nanobodies are derived from the antibodies found in camelid animals, including llamas and alpacas, which produce a type of antibody made of a single heavy chain rather than the two-chain structure of human antibodies. The single-chain format produces a smaller binding fragment that can penetrate tumor tissue more efficiently and exits the bloodstream more quickly, which reduces background signal in PET images. The researchers attached a radioactive tracer to the EphA2-targeting nanobody so that wherever it accumulated in the body, the PET scanner could detect and image it.

In the mouse experiments, the probe reached peak tumor concentration approximately two hours after injection and produced signal-to-noise ratios that the research team described as sufficient for clinical imaging purposes. Importantly, the kidneys, which filter nanobodies from circulation, showed the expected accumulation of tracer during clearance, but the liver and lungs showed minimal uptake, which means the probe did not produce confounding signals in organs where false positives would complicate interpretation.

Which cancers could benefit most from this imaging approach

Pancreatic cancer is one of the most compelling potential applications. It is typically diagnosed at late stage because it produces no early symptoms and sits in a location where standard imaging has limited resolution for small tumors. Pancreatic ductal adenocarcinoma overexpresses EphA2 in approximately 70 percent of cases, according to published immunohistochemistry studies. An imaging agent that could reliably detect EphA2-positive pancreatic lesions at earlier stages would address one of the most persistent diagnostic gaps in oncology.

Ovarian cancer presents a similar problem. It is frequently diagnosed at stage III or IV because early-stage tumors do not cause distinctive symptoms and are not caught by standard screening protocols. EphA2 overexpression is documented in roughly 60 percent of ovarian cancer cases in data from the Human Protein Atlas. An EphA2-targeted PET probe could serve as a complement to CA-125 blood testing and transvaginal ultrasound, the two methods currently used for ovarian cancer surveillance, both of which have well-documented sensitivity limitations for early-stage disease.

Path to clinical use and what comes next

The study was conducted entirely in mouse models, which means there is a substantial distance between these results and a clinical imaging agent approved for human use. The next step the research team identified is a toxicology and biodistribution study in larger animals, likely non-human primates, which is a standard preclinical requirement before an investigational new drug application can be filed with the FDA. That process typically takes 18 to 36 months depending on funding and the manufacturing readiness of the probe.

One advantage the program has is that nanobody-based imaging agents have already completed Phase I clinical trials in other cancer targets, including HER2 and PD-L1, establishing that the nanobody format is tolerated in humans at imaging doses. That prior work reduces some of the regulatory uncertainty around the format itself. The research team stated in the paper that they have filed a provisional patent on the EphA2 nanobody and are in early discussions with two academic medical centers about a Phase I imaging trial, which they hope to initiate by the second half of 2027.