Abstract
Epidermal growth factor receptor (EGFR), cellular mesenchymal‐epithelial transition factor (c‐MET), and B7 homolog 3 (B7H3) are important targets for antibody‐based drug development in colorectal cancer (CRC). To overcome the limitations of single‐target antibody therapies – such as limited efficacy, widespread resistance, narrow patient populations, and treatment‐related toxicities – various bispecific antibodies or antibody–drug conjugate (ADCs) targeting two targets are being extensively validated in clinical settings. A deeper understanding of their expression profiles and co‐expression patterns of actionable therapeutic targets may guide more effective treatment strategies. We evaluated the immunohistochemical expression of B7H3, c‐MET, and EGFR in a cohort of 193 CRC patients. Among which, B7H3 exhibited the highest positivity rate (80.83%) in CRC tissues, followed by c‐MET (77.20%) and EGFR (47.15%). The positive rates of B7H3 showed no significant differences across patient gender, age, TNM stage, differentiation grade, or tumor site. In contrast, the c‐MET positivity rate was elevated in patients with stage I–II disease, and the expression of EGFR was significantly higher in female patients than in male patients. Concurrent assessment of the three targets in metastatic lesions revealed that B7H3 expression levels were comparable between primary and metastatic tumors, whereas c‐MET expression was significantly higher in primary lesions. Co‐expression analysis indicated that the group with the highest patient coverage was the B7H3/c‐MET dual‐positive group, followed by the B7H3/EGFR, and c‐MET/EGFR combinations, and the proportion of patients with at least one positive target exceeded 80%. Finally, mRNA profile from TIMER database reveals that CD276 and MET expression in CRC tumor tissues were higher than those in the majority of normal tissues. Therefore, utilizing B7H3/c‐MET combination therapies that integrate intracellular signaling inhibition with immune checkpoint modulation could potentially extend therapeutic benefits to a wider range of CRC patients.
Keywords: colorectal cancer, B7H3, c‐MET, EGFR, biomarker, digital pathology, antibody–drug conjugate, IHC
Introduction
Colorectal cancer (CRC) remains a major global public health burden, ranking as the third most common cancer worldwide and the second leading cause of cancer‐related deaths, with approximately 1.9 million new cases and over 900,000 deaths annually in 2022, imposing a significant global health burden, particularly in China which leads in new cases and deaths, underscoring the urgent need for effective detection and targeted interventions [1]. Incidence and mortality are rising in low‐ and middle‐income countries, with over 50% of patients presenting with metastatic disease at diagnosis, creating an urgent need for improved therapeutic strategies [1, 2].
Surgery is the primary life‐extending treatment for early‐stage CRC, whereas advanced disease requires adjunctive use of chemotherapy, targeted therapy, immunotherapy, or their combinations [3]. The standard treatment of CRC is stratified by disease stage, with first‐line, second‐line, and subsequent posterior‐line therapies. First‐line or second‐line treatment for advanced CRC mainly includes 5‐fluorouracil‐based chemotherapy combined with targeted agents (e.g. bevacizumab or EGFR inhibitors for RAS/BRAF wild‐type patients) [4, 5, 6]. Posterior‐line therapies are limited to palliative options due to acquired drug resistance [7]. However, current standard treatments have obvious limitations: high toxicity of chemotherapy, limited efficacy in patients with specific gene mutations, inevitable acquired resistance, and poor response in most advanced cases after multiple lines of treatment.
Targeted therapy has made progress in CRC. The potential of targeting B7 Homolog 3 (B7H3), mesenchymal‐epithelial transition factor (c‐MET), and epidermal growth factor receptor (EGFR) in various combinations is attracting considerable attention [8, 9]. B7H3 (encoded by CD276), an important immune checkpoint molecule of the B7 family, is overexpressed in CRC tumors relative to normal tissue and contributes to immune evasion [10, 11]. Compared to PD‐L1, B7H3 expression is more prevalent in CRC and is significantly associated with aggressiveness, and worse prognosis cancer immunotherapy [12]. c‐MET, a surface tyrosine kinase receptor for hepatocyte growth factor (HGF), is frequently upregulated in CRC through gene amplification or transcriptional activation [13, 14, 15, 16]. The c‐MET/HGF axis is aberrantly activated in various cancers, promoting tumor progression, metastasis, and therapeutic resistance [16]. EGFR, the first validated therapeutic target in CRC, activates RAS/RAF/MAPK and PI3K/AKT pathways upon ligand binding, regulating proliferation, survival, and metastasis [17]. However, its targeting is limited by frequent RAS/KRAS/BRAF mutations (~50%/40%/10%), acquired EGFR resistance mutations (∼25%), and compensatory pathway activation [18, 19]. EGFR inhibitors (e.g. cetuximab, panitumumab) have shown significant efficacy in RAS/BRAF wild‐type advanced CRC, but their use is restricted to specific gene subtypes, and resistance often occurs within 6–12 months due to secondary mutations [20]. c‐MET, with a mutation rate of about 1–5% in CRC, is closely associated with EGFR inhibitor resistance by activating downstream signaling pathways. Phase II trials of tivantinib combined with cetuximab showed modest efficacy, with limited response in non‐MET‐amplified patients [10]. B7H3, a tumor‐specific immunomodulatory target, has been explored in antibody–drug conjugate (ADC) development (e.g. DS‐7300) [21], but most trials focus on other tumors, with limited clinical data and unclear efficacy in CRC patients [9]. The main limitations of single‐target drugs include narrow patient eligibility, easy occurrence of resistance, and insufficient efficacy in advanced cases.
Recent studies evaluating various combinations of targeted and/or immunotherapies have shown that dual‐ and triple‐targeted therapy strategies yield the best outcomes in terms of overall survival (OS) and progression‐free survival [22]. Nevertheless, not all combination regimens are beneficial, and multi‐drug combinations often lead to increased potential side effects and reduced patient compliance [23]. Therefore, a thorough assessment of co‐expression patterns of multiple targets in CRC is of significant value for guiding combination therapies.
Dual‐target therapy aims to overcome the limitations of single‐target drugs by co‐targeting complementary pathways. EGFR/c‐MET dual‐target drugs (e.g. amivantamab) have been developed to reverse EGFR inhibitor resistance caused by c‐MET amplification, with phase I/II trials showing preliminary efficacy in heavily pre‐treated CRC patients [24]. EGFR/BRAF dual‐target combinations (e.g. encorafenib plus cetuximab) have shown promising outcomes in BRAF V600E‐mutated CRC, improving OS compared to standard therapy [24]. B7H3‐based dual‐target therapies (e.g. ADC combined with EGFR inhibitor) are still in preclinical or early clinical stages, with no clear efficacy data in CRC [25]. Despite their potential to improve response rates and delay resistance, dual‐target therapies face challenges such as increased toxicity, high research costs, and unclear optimal combinations [26]. In summary, exploring more effective dual‐target strategies is crucial to addressing the unmet clinical needs of CRC patients, especially those with acquired resistance to single‐agent therapy [26].
Above all, we explored B7H3, c‐MET, and EGFR expression profile in CRC patients. In this study, new insights into therapeutic targeting of CRC by systematically analyzing the expression and co‐expression of B7H3, c‐MET, and EGFR, along with their correlations with clinical characteristics and mutational profiles may be provided.
Materials and methods
Ethics statement and patient consent
Tumor specimens from 193 Chinese patients diagnosed with pathologically confirmed colorectal adenocarcinoma at Hubei Provincial Cancer Hospital from 2016 to 2025 were retrospectively analyzed. Histological diagnoses were established using the 2026 World Health Organization classification. The pathologic stage was identified based on the American Joint Committee on Cancer TNM staging system. This study was approved by the Ethics Committee of Hubei Provincial Cancer Hospital (LCKY2024005), and all participants provided written informed consent. Demographic information (age, sex, and medical history), treatment details, and efficacy were extracted from the Hospital Information System by well‐trained research staff. Patient molecular profiling data (including KRAS, BRAF, and MMR status) were obtained from routine pathology laboratory reports. Data anonymization, privacy protection measures such as data de‐identification, access control, encryption, and data minimization and aggregation were implemented in this study.
Immunohistochemistry evaluation
Tissue samples were fixed in 10% formalin, processed through graded ethanol dehydration, paraffin‐embedded, and sectioned at 3 μm. Immunohistochemical (IHC) staining for c‐MET was performed using the CONFIRM anti‐Total c‐MET (SP44) rabbit monoclonal primary antibody (#790‐4430, Ventana Medical Systems, Tucson, AZ, USA) on the Ventana Benchmark Ultra platform (Ventana Medical Systems). Detection and visualization were carried out using the ultraView Universal DAB Detection Kit (#760‐500, Ventana Medical Systems) according to the manufacturer's instructions. Staining for B7H3 and EGFR was conducted on the Leica Bond III platform (Leica Biosystems, Mount Waverley, Victoria, Australia) using the Bond Polymer Refine Detection Kit (Leica DS9800‐CN). The primary antibodies used were: B7H3 (#14058, D9M2L, Cell Signaling Technology, Danvers, MA, USA and ab227670, SP206, Abcam, Cambridge, UK) and EGFR (#4267S, CST). Two anti‐B7H3 antibodies targeting either the intracellular (D9M2L) or extracellular (SP206) epitopes were used to assess their differential impacts on B7H3 staining outcomes (Figure 1A). In matched CRC tissue samples, SP206 demonstrated significantly higher B7H3 staining intensity and H‐score compared to D9M2L (Figure 1A,B). In addition, both antibodies exhibited low background staining (Figure 1C).
Figure 1.
Comparison of immunohistochemistry staining between B7H3 antibodies targeting the intracellular and extracellular domains. IHC detection of B7H3 expression in CRC patients (n = 193) using D9ML2/SP206, showing H‐scores (A), staining intensity (B), and representative images (C).
Two blinded pathologists independently evaluated the expression of these proteins by assessing the presence of partial or complete membranous staining in tumor cells. Staining intensity was categorized as negative (0), weak (1+), moderate (2+), or strong (3+). B7H3, c‐MET, and EGFR expression was analyzed semi‐quantitatively using the H‐score, which was calculated as the sum of the percentage of cell staining multiplied by the corresponding intensity value, ranging from 0 to 300 [27]. Positivity for c‐MET, B7H3, or EGFR was defined as H‐score ≥100.
Digital analysis
Slides were scanned using the Aperio® CS2 slide scanner (Leica Biosystems, San Diego, CA, USA) at 20× magnification. Digital pathology assessment was performed using HALO® and HALO AI® image analysis software (v4.1.5944.161; Indica Labs, Albuquerque, NM, USA). Tumor regions were identified using HALO AI's DenseNet V2 classifier, and all resulting annotations were manually reviewed and corrected to ensure exclusion of non‐neoplastic epithelial tissue from downstream analysis. Cell segmentation and classification were carried out using HALO AI's pretrained nuclear segmentation network, which was further refined on additional manual nuclear annotations to optimize performance for our dataset. For staining intensity estimation, default HALO® settings were applied, with the exception of slight adjustments to the DAB detection threshold (ranging from 0.1 to 0.4) to maximize detection of c‐MET positive staining.
HALO® accurately identified tumor regions and individual tumor cells, quantified membrane‐specific c‐MET staining intensity, and generated H‐scores (Figure 2A). To validate AI performance, 40 of 193 CRC samples (~20%) were randomly selected for independent pathological review. Using an a priori concordance threshold of ≤30 H‐score points, 92.5% (37/40) of cases showed agreement between AI and pathologist scores. Notably, at the predefined positivity cutoff, AI and manual scoring demonstrated 100% concordance for negative, positive, and overall classifications (Figure 2C), underscoring the reliability of AI‐assisted c‐MET assessment in CRC. Further c‐MET scoring was all performed using AI‐based assessment.
Figure 2.
Evaluation of AI‐assisted digital pathology for c‐MET assessment. (A) HALO® image analysis workflow. (B) Correlation between pathologist‐derived H‐scores and HALO‐generated H‐scores in 40 quality control samples, assessed by Pearson correlation analysis. (C) Concordance between AI‐assisted digital pathology and manual pathologist scoring for c‐MET positivity, summarized in a 2 × 2 contingency table.
Statistical methods
The sample size was determined based on the feasibility of accessing newly diagnosed cases within a specified timeframe rather than a prospective statistical power calculation. Statistical analyses related to clinical characteristics were conducted using R version 4.4.1 (R Foundation, Vienna, Austria). The χ 2 test was utilized to assess associations between B7H3, c‐MET, and EGFR expression and clinical characteristics. Statistical analyses related to molecular characteristics were performed using GraphPad Prism 10 software (Boston, MA, USA). Statistical significance was set at p < 0.05, and was determined using unpaired t‐tests, paired t‐tests, Pearson and Spearman correlation analysis. Pan‐cancer RNA expression profiles and prognostic analyses of CD276, MET, and EGFR were conducted using the TIMER database [28].
Results
Clinicopathological characteristics
Among the 193 enrolled patients, 83 (43.0%) were female and 110 (57.0%) were male, with a median age of 60 years (interquartile range 26–87). Postoperative TNM staging classified 57 (29.53%) patients as stage I/II and 136 (70.47%) as stage III/IV. The primary tumor site was in the colon (87 patients), rectum (92 patients), or other regions (14 patients). Among the 177 patients with assessable tumor differentiation, 66.67% (118/177) patients had well‐ or moderately differentiated tumors, and 30.57% (59/177) had poorly differentiated tumors. Regarding molecular characteristics, 6.90% (12/174) had BRAF mutations, 47.15% (58/123) had KRAS mutations, and 9.63% (18/187) exhibited MMR deficiency (dMMR).
Expression of B7H3, c‐MET, and EGFR
The expression profiles of B7H3, c‐MET, and EGFR were evaluated through IHC in all 193 patients. Samples with staining of three biomarkers displayed partial or complete membrane staining in tumor cells with intensities ranging from yellow to brown (Figure 3). Based on the definition of positivity for all targets as an H‐score ≥100, B7H3 exhibited the highest positivity rate (80.83%) in CRC tissues, followed by c‐MET (77.20%) and EGFR (47.15%) (Figure 4A–C). Co‐expression analysis indicated that the group with the highest patient coverage was the B7H3/c‐MET dual‐positive group, followed by the B7H3/EGFR and c‐MET/EGFR combinations. Among all combinations, the proportion of patients with at least one positive target exceeded 80% (Figure 4D–F). Meanwhile, we analyzed the co‐expression profiles across different stages of CRC (supplementary material, Figure S1A–D). We observed that as the CRC stage increased, the positive rate of c‐MET gradually decreased, while the proportion of B7‐H3 positivity in c‐MET‐negative samples progressively increased.
Figure 3.
Immunohistochemical staining of B7H3, c‐MET, and EGFR. Immunohistochemistry (IHC) with negative (0), weak (1+), moderate (2+), and strong (3+) staining for B7H3, c‐MET, and EGFR expression, respectively.
Figure 4.
Distribution of positive populations for target expression. (A–C) The number of patients distributed across different H‐Score ranges for B7H3 (A), c‐MET (B), and EGFR (C). (D–F) Co‐expression profiles of B7H3, c‐MET, and EGFR for specific target pairs: B7H3/c‐MET (D), B7H3/EGFR (E), and c‐MET/EGFR (F).
Correlation between specific target levels and clinical characteristics in CRC
The positive rates of B7H3 showed no significant differences across patient gender, age, TNM stage, differentiation grade, or tumor site (Table 1). In contrast, the c‐MET positivity rate was elevated in patients with stage I–II disease, and the expression of EGFR was significantly higher in female patients than in male patients (Table 1). Furthermore, no significant differences in B7H3, c‐MET, or EGFR expression were observed between patients with KRAS mutant versus wild‐type tumors, or between those with BRAF mutant versus wild‐type tumors (Figure 5A,B). In addition, expression of B7H3, c‐MET, and EGFR did not differ significantly between pMMR and dMMR patients (Figure 5C). Concurrent assessment of the three targets in metastatic lesions revealed that B7H3 and EGFR expression levels were comparable between primary and metastatic tumors, whereas c‐MET expression was significantly higher in primary lesions (Figure 5D–F).
Table 1.
Clinicopathological factors related to B7H3, c‐MET, and EGFR expression in CRC tissues [n (%)]
| n = 193 | B7H3 | p | c‐MET | p | EGFR | p | |||
|---|---|---|---|---|---|---|---|---|---|
| – | + | – | + | – | + | ||||
| Gender | |||||||||
| Female | 20 (24.10%) | 63 (75.90%) | 0.1851 | 18 (21.69%) | 65 (78.31%) | 0.8836 | 36 (43.37%) | 47 (56.63%) | 0.0319* |
| Male | 17 (15.45%) | 93 (84.55%) | 26 (23.64%) | 84 (76.36%) | 66 (60.00%) | 44 (40.00%) | |||
| Age (years) | |||||||||
| <65 | 24 (20.69%) | 92 (79.31%) | 0.6375 | 27 (23.28%) | 89 (76.72%) | 0.9848 | 68 (58.62%) | 48 (41.38%) | 0.06815 |
| ≥65 | 13 (16.88%) | 64 (83.12%) | 17 (22.08%) | 60 (77.92%) | 34 (44.16%) | 43 (55.84%) | |||
| TNM stage | |||||||||
| I–II | 11 (19.30%) | 46 (80.70%) | 1.0000 | 4 (7.02%) | 53 (92.98%) | 0.0014* | 27 (47.37%) | 30 (52.63%) | 0.4068 |
| III–IV | 26 (19.12%) | 110 (80.88%) | 40 (29.41%) | 96 (70.59%) | 75 (55.15%) | 61 (44.85%) | |||
| Differentiation | |||||||||
| Moderate/High | 21 (17.80%) | 97 (82.20%) | 0.7954 | 25 (21.19%) | 93 (78.81%) | 0.6529 | 68 (57.63%) | 50 (42.37%) | 0.1531 |
| Low | 13 (22.03%) | 46 (77.97%) | 14 (23.73%) | 45 (76.27%) | 25 (42.37%) | 34 (57.63%) | |||
| Unknown | 3 (18.75%) | 13 (81.25%) | 5 (31.25%) | 11 (68.75%) | 9 (56.25%) | 7 (43.75%) | |||
| Site | |||||||||
| Left colon | 10 (21.28%) | 37 (78.72%) | 0.9635 | 12 (25.53%) | 35 (74.47%) | 0.8715 | 29 (61.70%) | 18 (38.30%) | 0.1355 |
| Right colon | 7 (17.50%) | 33 (82.50%) | 9 (22.50%) | 31 (77.50%) | 15 (37.50%) | 25 (62.50%) | |||
| Rectum | 17 (18.48%) | 75 (81.52%) | 19 (20.65%) | 73 (79.35%) | 51 (55.43%) | 41 (44.57%) | |||
| Others | 3 (21.43%) | 11 (78.57%) | 4 (28.57%) | 10 (71.43%) | 7 (50.00%) | 7 (50.00%) | |||
Positivity for B7H3, c‐MET, or EGFR was defined H‐Score ≥100.
p < 0.05.
Figure 5.
Association of B7H3, c‐MET, and EGFR expression with patient molecular characteristics, metastasis, and prognosis. (A–C) Association of B7H3, c‐MET, and EGFR expression levels with KRAS mutation status (A), BRAF mutation status (B), and MMR status (C). (D–F) Comparison of B7H3 (D), c‐MET (E), and EGFR (F) expression levels between metastatic and primary lesions.
Association between specific target RNA levels and survival outcomes in CRC
To systematically explore the expression characteristics of B7H3, c‐MET, and EGFR in CRC, the RNA levels of these three targets across pan‐cancer tissues were analyzed. The levels of CD276 and MET in CRC tumor tissues were higher than those in the majority of adjacent normal tissues. However, EGFR levels were lower in CRC tumor tissues compared to adjacent normal tissues (Figure 6A). Additionally, CRC patients with high CD276 expression exhibited significantly worse OS compared to those with low CD276 expression (Figure 6B).
Figure 6.
Expression characteristics of CD276, MET, and EGFR. (A) Pan‐cancer expression levels of CD276, MET, and EGFR. (B) Association between CD276, MET, and EGFR expression levels and overall survival in colorectal cancer patients. The HR for OS was 1.406 (95% CI, 1.033–1.914) (CD276), 1.033 (95% CI, 0.809–1.319) (MET), and 1.269 (95% CI, 0.996–1.616) (EGFR), separately. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.
Discussion
An in‐depth assessment of the co‐expression patterns of multiple targets in CRC holds significant value for guiding combination therapies. Our study is the first to comprehensively analyze the co‐expression profiles of B7H3, c‐MET, and EGFR in CRC patients and their associations with clinical characteristics and treatment outcomes. Using an H‐score ≥100 as the cutoff, we found that B7H3 exhibited the highest positivity rate in CRC tissues, followed by c‐MET and EGFR. Moreover, our findings reveal that in CRC, the proportion of patients with B7H3/c‐MET double‐positive exceeds that of B7H3/EGFR or c‐MET/EGFR double‐positive groups. A combinatorial therapeutic strategy targeting B7H3/c‐MET may thus offer broader patient coverage. Additionally, expression profiling reveals that B7H3 and c‐MET are significantly overexpressed in most malignancies at both protein and mRNA levels, while they are limitedly expressed in normal tissues, thus a dual‐targeting strategy against B7H3 and c‐MET is expected to exhibit limited off‐target toxicity.
Consistent with several previous studies [10, 14, 29, 30, 31], we observed no significant association between B7H3, c‐MET, and EGFR expression and most clinical or molecular features in our cohort. Therefore, the combination strategy targeting these three markers may not be restricted to patient populations with specific clinical or molecular characteristics. Some reports have suggested that B7H3 positivity correlates with distant metastasis and tumor differentiation, while c‐MET expression is elevated in female patients [30, 32]. In contrast to our findings, these discrepancies may be attributed to variations in sample size or population heterogeneity.
Co‐expression analysis revealed a significant correlation between EGFR and c‐MET at protein levels, consistent with previous findings [30]. This association may be driven by the substantial overlap in intracellular signaling pathways utilized by both receptors (e.g. PI3K/AKT) [33]. While c‐MET/EGFR bispecific antibody–drug conjugates (BsADCs) are being actively investigated in lung cancer, their development in CRC remains limited. Current B7H3/EGFR BsADCs (IBI3001, KA‐3123‐LD38, TJ101) demonstrate promising clinical development. These agents leverage dual‐targeting to enhance tumor selectivity, reduce on‐target toxicity (e.g. IBI3001 minimizes EGFR‐related toxicities), and enable potent bystander killing [34, 35, 36]. All candidates exhibit strong anti‐tumor activity in EGFR/B7H3 co‐expressing CRC models, highlighting their potential to overcome resistance mechanisms while improving therapeutic indices [34, 35, 36]. Despite fewer developmental candidates for the B7H3/c‐MET bispecific pairing compared to c‐MET/EGFR or EGFR/B7H3, with SCR‐A0011 being the only reported agent, it demonstrates superior binding and internalization in B7H3/c‐MET co‐expressing tumor cells relative to parental monoclonal antibody‐based ADCs [37]. Considering both patient coverage and target safety, the B7H3/c‐MET combination strategy may represent the optimal option among the three combinations. We found that from stage I to stage IV, the positive rate of c‐MET gradually decreased from 94.73% to 67.04%. Additionally, we observed that from stage I to stage IV, the percentage of B7‐H3 positivity in c‐MET‐negative samples progressively increased. Currently, there is no direct evidence in the published literature that the loss or inhibition of c‐MET leads to upregulation of B7‐H3, despite evidence of a close functional interaction between B7‐H3 and c‐MET. The specific mechanism remains to be investigated. However, it has been reported that the binding of c‐Met to B7‐H3 can directly activate the c‐Met/STAT3 signaling pathway, thereby promoting stemness maintenance in CRC‐derived tumor cells [38]. We hypothesize that the loss of c‐MET may lead to B7‐H3 upregulation through a compensatory mechanism, a phenomenon that warrants further investigation.
Several limitations exist in this study. Firstly, the sample size of this study is relatively limited, and a larger‐scale study is necessary to get more accurate conclusions. Besides, this study only analyzed the correlation between mRNA expression and efficacy, without performing analysis at the protein level. The absence of data on patient responses to therapy prevents a definitive assessment of whether the B7H3, c‐MET, and EGFR expression correlates with therapeutic outcomes for advanced CRC. Additionally, IHC analysis offers a semi‐quantitative evaluation of a relatively limited portion of the tumor, which may not fully capture the overall protein expression. The inherent heterogeneity within tumor and benign tissues can further obscure expression differences.
In conclusion, the B7H3/c‐MET co‐expression profile could provide new insights for utilizing B7H3/c‐MET combination therapies and may potentially extend therapeutic benefits to a wider range of CRC patients.
Author contributions statement
JY, XW and RT contributed equally and are considered co‐corresponding authors of this work. They had full access to all the data in the study and took responsibility for the integrity and accuracy of the data analysis. YY, HL and SX contributed equally and are considered co‐first authors of this work. Conceptualization and study design were led by the corresponding and first authors. HL performed the statistical analysis. CA provided administrative, technical and material support. All authors contributed to data acquisition, analysis and interpretation. All authors contributed to drafting the manuscript and provided critical revisions for important intellectual content. All authors reviewed and approved the final version of the manuscript.
Supporting information
Figure S1. Co‐expression profiles across different stages of CRC
Acknowledgements
This research was funded by Health commission of Hubei Province scientific research project (WJ2019H124), 2020 Jingjian·Tongshu Microsatellite Instability Research Fund Project (Beijing Jingjian Foundation for The Advancement of Pathology, No. JJTS2020‐017), and Research Projects of Biomedical Center of Hubei Cancer Hospital (2022SWZX24).
Conflict of interest statement: MJ, HL, CA, SY, SS, CY, RT, and XW were employees of State Key Laboratory of Neurology and Oncology Drug Development at the time of day. Other authors declare no competing financial interest.
Contributor Information
Renhong Tang, Email: renhong.tang@zaiming.com.
Xiyuan Wang, Email: wendywang@zaiming.com.
Junqiu Yue, Email: yuejunqiu@hotmail.com.
Data availability statement
The datasets supporting the conclusions of this article are included within the article.
References
- 1. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024; 74: 229–263. [DOI] [PubMed] [Google Scholar]
- 2. National Health Commission of the People's Republic of China , Chinese Society of Oncology . Chinese protocol of diagnosis and treatment of colorectal cancer (2023 edition). Zhonghua Wai Ke Za Zhi 2023; 61: 617–644. [DOI] [PubMed] [Google Scholar]
- 3. Guo H, Miao L, Yu C. The efficacy of targeted therapy and/or immunotherapy with or without chemotherapy in patients with colorectal cancer: a network meta‐analysis. Eur J Pharmacol 2025; 988: 177219. [DOI] [PubMed] [Google Scholar]
- 4. Benson AB, Venook AP, Al‐Hawary MM, et al. NCCN guidelines insights: colon cancer, version 2.2018. J Natl Compr Canc Netw 2018; 16: 359–369. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Boccaccino A, Borelli B, Intini R, et al. Encorafenib plus cetuximab with or without binimetinib in patients with BRAF V600E‐mutated metastatic colorectal cancer: real‐life data from an Italian multicenter experience. ESMO Open 2022; 7: 100506. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Hafliger E, Boccaccino A, Lapeyre‐Prost A, et al. Encorafenib plus cetuximab treatment in BRAF V600E‐mutated metastatic colorectal cancer patients pre‐treated with an anti‐EGFR: an AGEO‐GONO case series. Eur J Cancer 2022; 168: 34–40. [DOI] [PubMed] [Google Scholar]
- 7. Yoshino T, Cervantes A, Bando H, et al. Pan‐Asian adapted ESMO clinical practice guidelines for the diagnosis, treatment and follow‐up of patients with metastatic colorectal cancer. ESMO Open 2023; 8: 101558. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Shao W, Liu Y, Huang L, et al. Clinical trials of bispecific antibody therapy for colorectal cancer: advanced and next steps. Front Oncol 2026; 16: 1758251. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Ott JM, Gassenmaier V, Bitzer M, et al. B7‐H3: a consistent marker in metastatic colorectal cancer with potential for targeted treatment. Pathol Oncol Res 2025; 31: 1612186. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Mielcarska S, Dawidowicz M, Kula A, et al. B7H3 role in reshaping immunosuppressive landscape in MSI and MSS colorectal cancer tumours. Cancer 2023; 15: 3136. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Peuker K, Strigli A, Tauriello DVF, et al. Microbiota‐dependent activation of the myeloid calcineurin‐NFAT pathway inhibits B7H3‐ and B7H4‐dependent anti‐tumor immunity in colorectal cancer. Immunity 2022; 55: 701–717.e7. [DOI] [PubMed] [Google Scholar]
- 12. Lu Z, Zhao ZX, Cheng P, et al. B7‐H3 immune checkpoint expression is a poor prognostic factor in colorectal carcinoma. Mod Pathol 2020; 33: 2330–2340. [DOI] [PubMed] [Google Scholar]
- 13. Wan X, Li Z, Zhou L. c‐MET is an important marker for acid‐driven metastasis and anti‐immune in colorectal cancer. Int Immunopharmacol 2024; 142: 113153. [DOI] [PubMed] [Google Scholar]
- 14. Resnick MB, Routhier J, Konkin T, et al. Epidermal growth factor receptor, c‐MET, beta‐catenin, and p53 expression as prognostic indicators in stage II colon cancer: a tissue microarray study. Clin Cancer Res 2004; 10: 3069–3075. [DOI] [PubMed] [Google Scholar]
- 15. Zeng ZS, Weiser MR, Kuntz E, et al. c‐Met gene amplification is associated with advanced stage colorectal cancer and liver metastases. Cancer Lett 2008; 265: 258–269. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Gherardi E, Birchmeier W, Birchmeier C, et al. Targeting MET in cancer: rationale and progress. Nat Rev Cancer 2012; 12: 89–103. [DOI] [PubMed] [Google Scholar]
- 17. Herbst RS. Review of epidermal growth factor receptor biology. Int J Radiat Oncol Biol Phys 2004; 59: 21–26. [DOI] [PubMed] [Google Scholar]
- 18. Tardito S, Matis S, Zocchi MR, et al. Epidermal growth factor receptor targeting in colorectal carcinoma: antibodies and patient‐derived organoids as a smart model to study therapy resistance. Int J Mol Sci 2024; 25: 7131. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Khelwatty SA, Puvanenthiran S, Seddon AM, et al. Advancements in targeted therapies for colorectal cancer: overcoming challenges and exploring future directions. Cancer 2025; 17: 2810. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Rimassa L, Bozzarelli S, Pietrantonio F, et al. Phase II study of tivantinib and cetuximab in patients with KRAS wild‐type metastatic colorectal cancer with acquired resistance to EGFR inhibitors and emergence of MET overexpression: lesson learned for future trials with EGFR/MET dual inhibition. Clin Colorectal Cancer 2019; 18: 125–132.e2. [DOI] [PubMed] [Google Scholar]
- 21. Owonikoko TK, Byers L, Cheng Y, et al. IDeate‐Lung02: a phase 3 study of second‐line ifinatamab deruxtecan in patients with relapsed small cell lung cancer. Future Oncol 2025; 21: 3275–3282. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Elez E, Yoshino T, Shen L, et al. Encorafenib, cetuximab, and mFOLFOX6 in BRAF‐mutated colorectal cancer. N Engl J Med 2025; 392: 2425–2437. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Ocean AJ, Polite B, Christos P, et al. Cetuximab is associated with excessive toxicity when combined with bevacizumab plus mFOLFOX6 in metastatic colorectal carcinoma. Clin Colorectal Cancer 2010; 9: 290–296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Tomasini P, Wang Y, Li Y, et al. Amivantamab plus lazertinib in atypical EGFR‐mutated advanced non‐small cell lung cancer: results from CHRYSALIS‐2. J Clin Oncol 2026; 44: 54–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Chen LC, Yang PC, Chen CY, et al. Dual inhibition of B7‐H3 and EGFR overcomes acquired chemoresistance in colon adenocarcinoma. J Cancer 2024; 15: 1750–1761. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Underwood PW, Pawlik TM. Precision medicine for metastatic colorectal cancer: where do we stand? Cancer 2024; 16: 3870. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Zhou Q, Li K, Lai Y, et al. B7 score and T cell infiltration stratify immune status in prostate cancer. J Immunother Cancer 2021; 9: e002455. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Li T, Fan J, Wang B, et al. TIMER: a web server for comprehensive analysis of tumor‐infiltrating immune cells. Cancer Res 2017; 77: e108–e110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Sun J, Chen LJ, Zhang GB, et al. Clinical significance and regulation of the costimulatory molecule B7‐H3 in human colorectal carcinoma. Cancer Immunol Immunother 2010; 59: 1163–1171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Lai X, Dong Q, Xu F, et al. Correlation of c‐MET expression with clinical characteristics and the prognosis of colorectal cancer. J Gastrointest Oncol 2021; 12: 2203–2210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Ceccarelli C, Piazzi G, Paterini P, et al. Concurrent EGFr and Cox‐2 expression in colorectal cancer: proliferation impact and tumour spreading. Ann Oncol 2005; 16: iv74–iv79. [DOI] [PubMed] [Google Scholar]
- 32. Zhang T, Wang F, Wu JY, et al. Clinical correlation of B7‐H3 and B3GALT4 with the prognosis of colorectal cancer. World J Gastroenterol 2018; 24: 3538–3546. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Tong CWS, Wu WKK, Loong HHF, et al. Drug combination approach to overcome resistance to EGFR tyrosine kinase inhibitors in lung cancer. Cancer Lett 2017; 405: 100–110. [DOI] [PubMed] [Google Scholar]
- 34. Qi L, Takeda S, Zhang Y, et al. Abstract 2955: preclinical development of TJ101, a potent bispecific ADC targeting EGFR and B7‐H3 for the treatment of solid cancers. Cancer Res 2025; 85: 2955. [Google Scholar]
- 35. Wu G, Xu C, Nian W, et al. Abstract 2132: KA‐3123‐LD38: a novel bispecific antibody‐drug conjugate (ADC) targeting EGFR and B7H3 with a DNA topoisomerase I inhibitor payload. Cancer Res 2025; 85: 2132. [Google Scholar]
- 36. Guan J, Zhang X, Wu W, et al. Abstract LB055: IBI3001: a potentially first‐in‐class site‐specifically conjugated B7‐H3/EGFR bispecific ADC for multiple solid tumors. Cancer Res 2024; 84: LB055. [Google Scholar]
- 37. Fu Y, Cheng G, Zheng H, et al. Abstract 1560: SCR‐A0011, a novel bispecific ADC targeting cMET and B7H3, exhibits potent anti‐tumor efficacy. Cancer Res 2025; 85: 1560. [Google Scholar]
- 38. Cao L, Xu Y, Hu Y, et al. Identification of c‐met on tumor cells as a novel receptor for B7‐H3 entails implications for cancer cell stemness and targeted therapy. MedComm 2025; 6: e70332. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1. Co‐expression profiles across different stages of CRC
Data Availability Statement
The datasets supporting the conclusions of this article are included within the article.