CEACAM5 imaging for occult colorectal recurrence
After bowel cancer surgery, a rising blood marker often signals that the cancer is back, but scans frequently cannot say where. The one scan ever approved for that exact problem was withdrawn from the market in 2005 and nothing has replaced it.
Overview
The clinical scenario is specific and common: a treated colorectal cancer patient's serum CEA rises during surveillance while CT, MRI and colonoscopy are normal or equivocal. Salvage surgery or ablation can still cure oligometastatic recurrence, but only if the disease can be localised. Arcitumomab (CEA-Scan), a technetium-99m anti-CEA antibody fragment approved for precisely this indication, was withdrawn from both the US and EU markets in 2005; clinicians now fall back on FDG PET/CT off-label, which in one referred series found recurrence in 56 of 88 such patients.
Meanwhile the therapeutic side of CEACAM5 has grown: antibody-drug conjugates, T-cell engagers and CAR-T programmes target it, and the field has learned that CEA-directed therapy selection is not the same question as CEA-directed localisation. The most advanced CEACAM5-directed imaging agent in humans is not a PET tracer at all but SGM-101, a near-infrared fluorescent anti-CEA antibody for intraoperative use, now in a randomised phase 3 trial (NCT03659448, about 300 patients). For PET, the most advanced is the CEA-targeted nanobody 68Ga-HNI01 at first-in-human scale, nine patients.
The competitive picture changed independently of imaging. Tumour-informed ctDNA now detects recurrence a median of 5.9 months before radiological recurrence in GALAXY/CIRCULATE-Japan, and ctDNA-guided adjuvant therapy was non-inferior to standard management in DYNAMIC while halving chemotherapy use. ctDNA says whether and roughly when; it does not say where. That distinction is the whole of the case for a CEACAM5 tracer, and it is a stronger case now than when CEA-Scan was withdrawn, because more patients are being flagged as molecularly positive with nothing to localise.
When serum CEA or a ctDNA assay says a treated colorectal cancer has recurred but CT, MRI and colonoscopy cannot find it, no approved molecular imaging agent localises the disease well enough to plan salvage surgery.
- FDG PET/CT is already good enough in this scenario for many teams, so the margin a specific agent must beat is not zero.
- Shed circulating CEA can compete with tumour binding and blunt contrast, which is a formulation and format problem rather than a target problem.
- If ctDNA-guided management moves toward treating molecular recurrence systemically without localising it, the surgical rationale for a localising scan narrows.
- The randomised SGM-101 phase 3 readout, which will show whether CEA-directed imaging changes surgical outcomes at all, in any modality.
- Any CEACAM5 PET agent entering a trial with rising CEA or ctDNA positivity as the enrolment criterion rather than known measurable disease.
- Guideline language on what to do with a positive ctDNA and a negative scan, which is where the demand for this agent will be written down.
The five domains of a radiopharmaceutical due-diligence review, adapted from the Fryback-Thornbury hierarchy of diagnostic efficacy. Each carries the state of the public evidence and the fact behind it, not a score out of five: a single total hides the reason and stops telling candidates apart. Domains marked as carrying most weight are the ones that decide the case at this asset's stage.
CEACAM5 is a long-established colorectal lineage antigen; serum CEA is in every surveillance guideline, and an anti-CEA imaging agent was once approved for occult recurrence, so in-vivo targeting is proven in principle.
No CEACAM5 PET agent has gone beyond first-in-human; the nanobody 68Ga-HNI01 has nine patients. The only phase 3 CEA-directed imaging agent is a fluorescence agent for use in the operating theatre, not a tracer.
The incumbent, FDG PET/CT, detected recurrence in 56 of 88 patients referred with raised CEA and negative conventional imaging, so any new agent must beat a working off-label comparator rather than fill a void.
No published dosimetry for a current-generation CEACAM5 PET agent at a scale worth quoting; the withdrawn technetium-99m fragment is not a guide to a gallium-68 or fluorine-18 construct.
A direct precedent exists in the same indication, which is unusual: arcitumomab was approved for occult colorectal recurrence before being withdrawn in 2005, so the endpoint and claim structure have been accepted once.
Red, amber or green on each review category, each with its reason, so an obvious no-go takes seconds rather than a meeting.
Bowel and liver background is the hard part of abdominal imaging, and the antigen is shed into serum, which can blunt tumour uptake; a fragment or nanobody format is probably required.
A defined, guideline-recognised scenario with no approved agent since 2005, and a population growing because ctDNA surveillance flags recurrence earlier than imaging can localise it.
CEA binders are a crowded, decades-old space with many antibodies and fragments in therapeutic programmes; novelty would have to come from format and chelation rather than the binding site.
The localisation use stands alone, but the scan is a one-off per recurrence episode in a surveillance population, so volume depends on how widely ctDNA surveillance is adopted.
Gallium-68 or fluorine-18 on a fragment gives same-day imaging; no supply constraint.
No company has a disclosed late-stage CEACAM5 PET programme; the phase 3 activity is in fluorescence-guided surgery, which is complementary rather than competing.
- Surgimabphase 3
SGM-101, a fluorescent anti-CEA antibody for intraoperative delineation of colorectal and peritoneal disease, in a randomised phase 3 trial of about 300 patients.
https://clinicaltrials.gov/study/NCT03659448 - Academic nanobody programmesphase 1
68Ga-HNI01, a CEA-targeted nanobody, first-in-human at nine patients.
https://pmc.ncbi.nlm.nih.gov/articles/PMC7892528/ - ctDNA surveillance (Natera and others)in clinical use
Tumour-informed ctDNA detects recurrence a median of 5.91 months before radiological recurrence in GALAXY, creating the patients who need localising.
https://www.nature.com/articles/s41591-024-03254-6
Similar pages
not linked directly; found by shared links- Target5T4 (trophoblast glycoprotein)
Shares Colorectal cancer, CEACAM5, Immuno-PET, Fluorine-18.
- OpportunityAn inflammation tracer that does not need a diet
Shares FDG PET, PET/CT, Fluorine-18, Gallium-68 and the tag unmet need first.
- OpportunityFAP imaging to stop futile pancreatic surgery
Shares FDG PET, PET/CT, Fluorine-18, Gallium-68 and the tag unmet need first.
- OpportunityCAIX imaging of the indeterminate renal mass
Shares Immuno-PET, Technetium-99m, PET/CT, Gallium-68.
- OpportunityDLL3 imaging for small-cell lung cancer
Shares Immuno-PET, PET/CT, Gallium-68.
- OpportunityTROP2 imaging where no tissue test is required
Shares Immuno-PET, PET/CT, Gallium-68.
- OpportunityGPC3 imaging of the indeterminate liver nodule, led from China
Shares Immuno-PET, PET/CT, Gallium-68.
- AgentSodium fluoride F-18
Shares Technetium-99m, PET/CT, Fluorine-18.