On the final day of AACR 2026, the German Cancer Consortium presented first-in-human PET/CT imaging data from pancreatic-cancer patients who had received a single injection of EphA2 BIA, a gallium-68-labeled bicyclic peptide from Bicycle Therapeutics. Across 18 patients imaged to date, the tracer detected liver, bone, lymph node, and peritoneal metastases in 15. The scans ran up to three hours post-injection, with rapid tumor uptake and primarily renal clearance in six of seven patients from the latest cohort. All per the April 23 peptide-news-digest ↗.

What a bicyclic peptide is

A bicyclic peptide is a short peptide (usually 9 to 15 residues) locked into two rings by a central small-molecule scaffold that its cysteines cover. The ring constraint gives the molecule a defined three-dimensional shape, the way an antibody's binding loop has one. What you get is a binder roughly the size of a macrocycle (around 1,500 to 2,000 Da) that can hit protein targets with antibody-like selectivity, penetrate tissue faster than an antibody, and clear through the kidneys within a few hours.

The fast clearance is the imaging trick. A tracer that binds its target tightly but washes out of everything else in the body within hours produces a clean PET image. Too slow a clearance and the background buries the signal. Too fast and there is not enough tracer on the tumor to detect. The EphA2 BIA data says the clearance curve lands in the useful band.

Why EphA2 in pancreatic cancer

EphA2 is a receptor tyrosine kinase overexpressed in most pancreatic ductal adenocarcinomas. PDAC is one of the hardest solid tumors in oncology. Five-year survival sits near 12 percent. Imaging for metastasis is one of the weakest points in PDAC staging: CT misses small peritoneal and lymph-node deposits, and MRI struggles with bone. A PET tracer that hits a tumor-specific receptor directly is, in principle, the right tool. That is what Bicycle and the German Cancer Consortium are testing.

Two points the readout does not settle. First, the comparison against FDG-PET (the standard oncology PET tracer) has not been reported. FDG picks up any metabolically active tissue, including inflammation. EphA2 BIA should be more specific but may miss EphA2-negative tumor clones. Second, whether detection of small metastases actually changes treatment decisions in pancreatic cancer is a clinical question, not an imaging question. A better scan is only better if it changes what the oncologist does next.

The companion-diagnostic frame

The bigger story is what this tracer is a scout for. Our AACR map this week ↗ noted Bicycle's therapeutic bicyclic peptide-drug conjugate BT5528, which targets the same EphA2 receptor and showed a 40 percent objective response in head-and-neck cancer. The imaging tracer and the therapeutic share the target. In practice that means a patient can be scanned first, and only patients whose tumors light up on EphA2 BIA PET get the matched therapeutic. That is the companion-diagnostic model that made Lutathera (¹⁷⁷Lu-DOTATATE) possible in neuroendocrine tumors, preceded by ⁶⁸Ga-DOTATATE imaging. Both of those are peptides. The pairing is the pattern.

If it works for EphA2 in PDAC, the same pattern can extend to the other disease areas where Bicycle is advancing bicyclic PDCs. The readout to watch is whether the next Bicycle therapeutic trial in PDAC uses EphA2 BIA PET as an enrollment biomarker. That is the test of whether imaging and therapy converge into one pipeline, or stay as two separate ones.

Platform context

Peptidemodel's anticancer target ↗ collects 196 cards on peptide-based oncology candidates. Macrocyclic and bicyclic peptides are the modality layer immediately above conventional peptide drugs in molecular weight and constraint. They sit in the 500 to 2,000 Da window where antibodies cannot reach (too big to penetrate dense tumor stroma) and small molecules cannot selectively engage large protein-protein interfaces. PDAC, with its dense stroma and few targetable antigens, is exactly where that gap shows up in clinic.

The 15-of-18 metastasis detection number is not an efficacy endpoint. It is a feasibility endpoint, and the first real one for EphA2 bicyclic peptide imaging. If the Phase 2 imaging trial reads out cleanly, the next PDAC therapeutic that makes it to Phase 2 will probably carry an EphA2 BIA scan in the entry criteria.