CLDN18.2 Enters Theranostics: PeptiDream's First-in-Human Start and What It Signals

On June 11, 2026, PeptiDream dosed the first patient with 64Cu-PD-29875, a copper-64 labeled macrocyclic peptide that targets Claudin 18.2 (CLDN18.2), in a first-in-human Phase 0 imaging study in gastric and gastroesophageal junction cancer (PDRadiopharma).

Until now, theranostics has lived almost entirely inside two proven target classes: PSMA in prostate cancer and SSTR in neuroendocrine tumors. CLDN18.2 is something different. It is a tight-junction epithelial protein with no prior approved radiopharmaceutical and no prior radiopharmaceutical in humans, and its competitive landscape has been built around antibodies, ADCs, bispecifics, and CAR-T (CLDN18.2 theranostics review, NIH, CLDN18.2 competitive landscape).

The new datapoint is simple to state and large in implication: a macrocyclic-peptide radiotheranostic pair has entered the clinic against a target that already carries an FDA-approved antibody and eight ADCs in Phase 3. The imaging study is explicitly designed to feed dosimetry into a planned 225Ac-PD-29875 alpha-therapy trial (PDRadiopharma protocol approval, BioSpace).

The short version

Three things stand out:

1. A new target class entered the clinic. CLDN18.2 is the first theranostics target outside the PSMA and SSTR scaffolds to reach first-in-human, dosed June 11, 2026 (AllSci).

2. The pairing is a familiar shape with a new wrinkle. 64Cu for PET imaging and 225Ac for therapy mirrors the PSMA and SSTR playbook, but the alpha emitter means dosimetry from the imaging study has to account for high-LET energy transfer, not beta range (PDRadiopharma protocol approval).

3. The peptide may widen the eligible population. Preclinical radiopharmaceutical work showed activity in tumors with CLDN18.2 expression below current ADC eligibility thresholds, which suggests the addressable population could be larger, not smaller, than antibody-based therapy (Claudin 18.2 radiotheranostic proof-of-concept, NIH).

The operational takeaway: this is early, but it points at a future where GI cancers get a PSMA-style imaging-plus-therapy pathway, and the diagnostic becomes patient-selection infrastructure.

At a glance

Why CLDN18.2 works as a theranostics target

CLDN18.2 is normally locked inside the tight junctions of gastric cells and is essentially absent from other adult tissue. During malignant transformation, loss of cell polarity exposes the protein on the tumor surface, which is what makes it targetable by both antibodies and radiolabeled peptides (CLDN18.2 theranostics review, NIH).

Expression is high in the relevant tumors. A large European analysis in 2026 found CLDN18.2-high expression in 48.3 percent of gastroesophageal adenocarcinomas, higher than the roughly 40 percent reported in pivotal trials, which suggests broader patient-selection utility for an imaging diagnostic (Oncology News Central).

The reason a peptide matters, rather than an antibody, is pharmacokinetics. Full-length antibody imaging agents need roughly 6-day windows before tumor-to-background contrast is usable. Macrocyclic peptides clear faster, penetrate tissue better, and produce more predictable dosimetry, which is exactly what you want for clinical scheduling and for predicting an alpha-therapy dose (Claudin 18.2 radiotheranostic, NIH).

Why this could matter against the ADC field

CLDN18.2 is one of the most crowded targets in GI oncology. Zolbetuximab is FDA approved, eight ADCs are in Phase 3, and CAR-T has shown the first positive randomized result in solid tumors (MedPath, ESMO Daily Reporter).

A radiopharmaceutical can be differentiated on three fronts. Patient selection becomes non-invasive and whole-body via PET, rather than a single biopsy with sampling bias. Alpha crossfire can reach neighboring cells in heterogeneous tumors that an ADC bystander effect may miss. And resistance mechanisms differ, since DNA double-strand breaks do not face the same efflux resistance that limits auristatin payloads (NIH).

The eligibility point is the one to watch. Zolbetuximab requires high IHC positivity, which captures only a portion of gastric patients. Preclinical radiopharmaceutical work produced complete tumor regression in 5 of 7 pancreatic cancer models and was effective in tumors expressing CLDN18.2 in as few as 41 percent of cells, below current ADC thresholds (Claudin 18.2 radiotheranostic, NIH, Applied Radiation Oncology).

What still has to be proven

This is a first-in-human imaging study, not a therapy readout, and there are real open questions:

• Normal gastric mucosa expresses low-level CLDN18.2. The dosimetry has to show that 225Ac therapy will not create dose-limiting gastric toxicity (PDRadiopharma).
• Peptides clear fast, which is ideal for imaging, but they have to retain in tumor long enough to deliver a 225Ac dose given its roughly 9.9-day half-life. Chelation and peptide engineering need to balance that.
• AstraZeneca's Phase 3 CLDN18.2 ADC pivotal data is expected in the first half of 2026. A strong readout accelerates interest in the target but could also raise the efficacy bar (MedPath).
• Enrollment cap and timeline for CLAUDIRA-IIS were not disclosed, so the path to a therapeutic Phase 1 is not yet on a public clock (AllSci).

The practical bottom line

Here is how I would read this for a theranostics program:

• Watch GI as the next theranostics frontier. PSMA and SSTR built the playbook. CLDN18.2 is the first credible attempt to run it in gastric and pancreatic cancer.
• Think about imaging as the chokepoint again. If a CLDN18.2 PET agent validates, it could stratify patients across every CLDN18.2 modality, not just the radiopharmaceutical, and whoever controls that diagnostic controls a referral pathway.
• Keep this in the early-signal column. This is a Phase 0 imaging start, not a therapy result. The near-term catalyst is the dosimetry readout that decides whether 225Ac-PD-29875 becomes a viable Phase 1.
• Note the platform story. PeptiDream's PDPS engine producing a second internal radiopharmaceutical, after a CAIX program, signals that peptide discovery can systematically open new theranostics targets beyond the established two.

The bigger point for the field: theranostics is starting to look less like a two-target modality and more like a platform that can be pointed at high-value oncology targets one at a time. CLDN18.2 is the test case.

Closing CTA

If you run, refer into, or are building a theranostics program, subscribe to Theranostics Weekly. Each issue focuses on the clinical, operational, and business questions that determine whether these therapies actually reach patients.