Viewer
AmethIntel · Oncology Intelligence Series · 2026

CEACAM5: Oncology Target and
Ramifications for Target Selection Strategy

A decision-grade intelligence review integrating clinical trial triage, failure attribution, patent landscape, payload and linker analysis, and 2026 deal structure benchmarks — structured for R&D heads, BD leaders, and oncology investment teams.

Publisher
AmethIntel
Pages
~80 pp
Chapters
8 + Appendix
Updated
May 2026
Coverage
US · EU · Global
Classification
Confidential
TOC + Executive Summary
Data
Analysis
Projection

CEACAM5 Therapeutic Landscape & Strategic Outlook

CEACAM5 (carcinoembryonic antigen-related cell adhesion molecule 5, CD66e) is emerging as one of the most strategically important solid tumor targets in precision oncology. Long recognized biologically, but historically constrained by technological limitations, the target now stands at an inflection point.


AmethIntel Analysis of CEACAM5 : A Commercial Oncology Platform with Opportunities of Unique Nature.
A study of Technological Evolution, Clinical Development, Intellectual Property, Regulatory, Licensing, and Investment Dynamics

AmethIntel™ is a Search and Value Algorithm trained to support the BD and investment teams on data backed valuations, and to the scientific, and clinical development teams to guide on the driver of such valuations. The iterative search strategy uses clinical trials as the base dataset, and on each trial builds the patents, publications, regulatory and payer (USA) outlook to identify white spaces and opportunities.

Analysis of CEACAM5 was an interesting case study, arguing convergence of various technologies with clinical development strategy to unlock clinically and commercially viable therapies that may reposition CEACAM5 within the next generation of oncology platforms.

A total of 187 therapeutic clinical trials were identified as having a bearing for the understanding of the landscape. Various moieties are tried for CEACAM5 – CAR-T, ADC, peptides, small molecules. The developments revealed the nature and basis of competition and market signals that may shape the industry. These signals as of June 2026 are analyzed.

Drug development is an iterative process running market development and clinical development hand in hand. ADCs are the leading moieties in development at the time of release of this report. Analysis of granted claims in ADC patents, analysis of payload and linker chemistries clearly establishes boundaries of the barriers to enter, which are limited. Consistent with this, the ADC space is getting crowded giving clear signals of fast followers. This demands a change in clinical development strategies, which instead of speed, which suits a field secured by strong IP barriers can corroborate with the IP strategy to secure niche segments in cases like CEACAM5.

For CAR-T as a modality, the second and third generation approaches are showing promising results. Breakthroughs are required in the economics of production, administration and after care. The constrains are not preventing strong developments and confidence, the basis of which is analyzed.

The attempted peptides show promise and no adverse events, these developments are paused for the time being, and provide qualified opportunities to potential fast followers. Diagnostics trials hold vital information in the case of a target like CEACAM5. Analysis of these trials provides clear directions to the white spaces which can be exploited.

CEACAM5 as a target is not druggable by small molecules and is a marker that leads to elimination of transformed cell, instead of being an oncogene. Analysis of small molecule drugs used in combination reveals additional applications of the platform in several tumors like KRAS, FGF, HNF-alpha, with an option to be developed as market expansion strategy.

Main programs that currently define the leading edge of the field are:

  • Precemtabart tocentecan (M9140, Merck KGaA) has emerged as a leading clinical programme, advancing across multiple PROCEADE studies in colorectal, gastric, pancreatic, and lung cancers using an exatecan payload platform.
  • SGN-CEACAM5C / PF-08046050 (Pfizer/Seagen) represents one of the most strategically important next-generation ADC programmes, combining the clinically validated tusamitamab antibody backbone with a high drug-to-antibody ratio Topoisomerase I payload platform optimized for bystander activity.
  • BMS-986490 is advancing through combination-oriented development strategies alongside bevacizumab and may contribute meaningfully to future sequencing paradigms.
  • IBI3020 (Innovent Biologics) introduces a differentiated dual-payload ADC architecture reflecting the broader movement toward increasingly engineered payload combinations.
  • Additional entrants, including linker-differentiated and regionally developed programs, continue to expand the competitive landscape and contribute to increasingly global licensing dynamics.

In therapeutic landscape analysis, shelved or discontinued programmers do not necessarily weaken a target category. In many cases, they generate valuable translational infrastructure, including validated assays, safety datasets, biomarker frameworks, manufacturing precedent, regulatory interactions, and partially de-risked development pathways. CEACAM5 has accumulated a substantial body of this latent development infrastructure. The accumulated translational history of the field may also create favorable conditions for differentiated fast-followers and best in class therapeutic strategies frequently observed in mature oncology markets.

The commercial relevance of this evolution is substantial. CEACAM5-expressing tumors — particularly colorectal cancer (CRC), non-small cell lung cancer (NSCLC), gastric and gastroesophageal cancers, pancreatic ductal adenocarcinoma (PDAC), and select breast cancer subtypes — collectively represent hundreds of thousands of new diagnoses annually across the United States and major European markets. Companion diagnostic infrastructure is already established, clinical enrollment criteria are increasingly standardized, and payer acceptance of biomarker-driven oncology therapies continues to expand. Successful CEACAM5-directed therapies therefore appear positioned for relatively rapid integration into second- and third-line treatment paradigms, with credible pathways toward earlier-line adoption.

Closer analysis reveals opportunities in payload and linker innovation, as well as antibody design that can aid development of next generation molecules.

CEACAM5 increasingly appears to represent more than an individual therapeutic target. The field now exhibits many characteristics of an emerging oncology platform ecosystem in which biology, translational infrastructure, platform technology, intellectual property accessibility, licensing geography, and clinical strategy interact simultaneously to shape long-term competitive positioning.

The leaders in such cases will emerge by managing technological ecosystem with clinical oncology development, licensing strategy, to deliver the intended precision medicine. Successful investment and licensing bets will integrate these to develop models replicating realistic success while hedging risk.

Table of Contents

Report Chapters

Click any chapter to expand preview · unlock full content individually or purchase the complete report
01
Target Biology & Expression Landscape
What CEACAM5 is, where it is expressed, and what that means for therapeutic window design
or included in full
+
Free preview — key findings
Domain architecture resolved at 3.11 Å: the October 2024 cryo-EM structure (PDB 8BW0, Sanofi/Nature Comms) reveals tusamitamab binds a discontinuous epitope in the A3-B3 domains incorporating an N-linked mannose at Asn612 — the first atomic-resolution epitope map of any CEACAM5 antibody
Expression frequency by tumour type: CRC 90–95% any IHC; high-expression (≥2+/≥50% cells) in ~60–70% mCRC. NSCLC adenocarcinoma: 24.3% HE in primary tumours, rising to 35.3% in metastases. Gastric: 55–70%. PDAC: ~90%
The polarity shift is the therapeutic window: in normal colonic epithelium CEACAM5 is strictly apical-surface; in adenocarcinoma polarity is lost and expression becomes circumferential — basolateral surface accessible to vascular-delivered agents
Full content locked
Full chapter includes: complete IHC frequency tables with H-score distributions · KRAS-CEACAM5 molecular correlations in NSCLC · serum CEA dynamics by tumour type · shedding sink quantification · domain surface exposure by tumour type vs. normal tissue · 9 peer-reviewed citations
Single chapter licence
02
Active Clinical Trial Landscape
260+ trials — tiered by what actually matters to a BD or investment decision
or included in full
+
Free preview — key findings
Tier A — three programmes with imminent data: BMS-986490 (NCT06730750, Phase 2a with ramucirumab, CRC), IBI3020 (NCT06963281, Chinese-origin dual-payload ADC), NILK-2301 (NCT06663839). Each has a meaningfully different linker and patient selection approach from tusamitamab
The outlier with highest near-term commercial probability: SGM-101 surgical fluorescence imaging agent for CRC margin detection — bypasses systemic shedding sink, separate regulatory pathway, Phase 3 data expected 2026–26. Underrepresented in standard competitive databases due to modality categorisation
Sponsor commitment divergence: Chinese biotechs (Innovent, Henlius) are advancing capital; Western large pharma has retreated. This geographic divergence is a licensing arbitrage signal, not a biology signal
Full content locked
Full chapter includes: complete Tier A/B/C trial triage with readout timelines · cross-trial patient selection heterogeneity analysis · kinetic energy scoring by programme · sponsor commitment signal methodology · CAR-T and vaccine programme assessment · full NCT reference index (in Appendix)
Single chapter licence
03
Programme Failure Attribution
A diagnostic, not a graveyard — every terminated programme assigned to one of four root causes
or included in full
+
Free preview — key findings
Tusamitamab: two distinct failure causes, not one: the corneal DLT (keratopathy in 25% of Phase 3 patients, dose-limiting at 120 mg/m²) is a DM4 maytansinoid payload class effect — Cause II, modality-intrinsic. The Phase 3 efficacy failure is Cause III, trial design — no serum CEA stratification, PFS primary endpoint in a population where OS trend was positive
Roche's full CEA portfolio exit (cibisatamab, cergutuzumab, CEA-IL2v) is Cause IV — strategic: the termination language in public filings does not cite biology failure. The CRS events in cibisatamab are Cause II — T-cell redirector format against a target with normal tissue expression in GI epithelium
Cross-cutting finding: every programme that survived longest shared one feature — either biomarker-selected enrolment or a surgical/diagnostic application that bypasses systemic exposure entirely
Full content locked
Full chapter includes: complete four-cause taxonomy with scoring criteria · every major terminated programme classified with evidence · labetuzumab govitecan SN-38 bystander analysis · MEDI-565 BiTE immunogenicity dissection · NCI TCR-engineered PBL colitis case · full attribution table (30 programmes)
Single chapter licence
04
Patent Landscape
Where entry is blocked, where it is open, and where the real barriers now concentrate
or included in full
+
Free preview — selected signals
The antibody is commoditised: A1-B1 and A2-B2 IgV-loop space is densely claimed. The A3-B3 space — where tusamitamab binds — now has a publicly deposited cryo-EM structure (PDB 8BW0) enabling rational design of adjacent-epitope antibodies by any competitor
N-terminal domain remains relatively open: limited assignee concentration, NCI public domain contributions available as FTO starting points — specific sequences and method-of-use claims identified in full chapter
Full content locked
Full chapter: complete domain-level assignee map · Synaffix/BI $1.3B linker deal analysis · biomarker and CDx IP layer · NCI FTO pathways · white space identification by domain and modality · strategic entry pathway for new entrant
Single chapter licence
05
Payload Landscape
Why maytansinoid gave way to Topo1i — and what the data says about the next transition
or included in full
+
Free preview — selected signals
DM4 corneal toxicity is a class effect, not a CEACAM5 effect: identical keratopathy/keratitis profile documented in mirvetuximab soravtansine (FOLR1-DM4) and trastuzumab emtansine (HER2-DM1). Corneal epithelial cells take up maytansinoid ADCs via non-specific endocytosis independent of target expression
Topo1i bystander killing is mechanistically superior for CEACAM5: heterogeneous expression in CRC and NSCLC means not every tumour cell expresses CEACAM5. DXd and exatecan derivatives with larger bystander killing radius compensate — DM4's bystander radius is insufficient for this target profile
Full content locked
Full chapter: DXd vs. SN-38 vs. exatecan clinical comparison · patent landscape by payload class · dual-payload platform IP (IBI3020) · combination payload hypotheses with mechanistic rationale · payload selection framework for next-generation CEACAM5 ADC
Single chapter licence
06
Linker Landscape
The linker is where this field will be won — the clinical evidence and the deal data agree
or included in full
+
Free preview — selected signals
Tusamitamab's SPDB linker is cleavable — but at the wrong site: lysine conjugation produces DAR heterogeneity (DAR 3.8 average, wide distribution). Site-specific conjugation at engineered cysteines or glycan anchors produces homogeneous DAR — directly improving the shedding sink survival ratio
The $1.3B signal: Boehringer Ingelheim acquired Synaffix specifically for glycan-anchor site-specific conjugation — not for an antibody, not for a target. The linker platform itself commanded the deal value. This is the most important pricing signal in CEACAM5-adjacent IP in 2024
Full content locked
Full chapter: shedding sink PK competition modelling · plasma half-life data across all CEACAM5 ADC programmes · site-specific conjugation IP landscape · DAR optimisation for high-shedding targets · white space in linker IP · patient selection solution derived from linker analysis
Single chapter licence
07
Differentiation for Long-Term Dominance
The three-variable framework — and the population sizing that determines the commercial ceiling
or included in full
+
Free preview — selected signals
All three errors must be corrected simultaneously: a programme with Topo1i payload but no serum CEA stratification repeats the patient selection error. A programme with site-specific conjugation but A3-B3 epitope targeting remains vulnerable to shed ectodomain competition. Sequential optimisation is insufficient
SGM-101 is the underappreciated exception: surgical fluorescence guidance for CRC margin detection bypasses every systemic problem. Highest near-term commercial probability in the field. Distinct licensing model from ADC programmes — closer to medical device than oncology drug
Full content locked
Full chapter: dual-criterion patient population sizing (IHC + serum CEA) in CRC, gastric, NSCLC · commercial ceiling by indication · four-scenario risk matrix with probability weights · decision triggers by scenario · combination strategy with mechanistic rationale
Single chapter licence
08
2026 Deal Structure Analysis
Where capital is moving — and what a CEACAM5 ADC at Phase 1 POC would actually license for
or included in full
+
Free preview — selected signals
Licensing overtook M&A as primary value driver in 2026: $250B+ across 516 transactions. For CEACAM5, this means a Phase 1 POC asset is a licensing play, not an acquisition target — and the deal structure must account for the Sanofi Phase 3 failure discount in narrative positioning
Chinese-origin out-licensing trajectory: Innovent/Takeda $11.4B deal structure is the template. Three active Chinese-origin CEACAM5 programmes (IBI3020, DNP002, NILK-2301) are all potential licensing candidates to Western pharma within 24–36 months of Phase 2 data
Full content locked
Full chapter: complete 2024–25 ADC deal comparables table · upfront/milestone/royalty benchmarks by stage · geo-split norms · the Sanofi failure discount — how to position against it · surgical fluorescence licensing model · non-ADC deal angles
Single chapter licence

Licensing & Pricing

Chapter licence
per chapter
Access any individual chapter. Purchase only what your analysis requires. Chapters 1–3 are most relevant for target validation; 4–6 for platform and IP strategy; 7–8 for commercial and deal decisions.
Single chapter, unlimited internal use
No redistribution outside purchasing organisation
PDF + web access
Peer-reviewed citation list included
Enterprise + Bespoke
custom
Custom target analysis applying the AmethIntel methodology to a target of your choosing, or multi-report series subscription including microbiology intelligence series (2026–26).
Bespoke target analysis — your choice of target
Series subscription: oncology + microbiology
Quarterly briefing calls
Multi-user enterprise licence
Competitive Kinetic Energy Scorecard for custom target