Skip to main content
NCBI home page
The Oncologist logoLink to The Oncologist
. 2026 Jul 31;31(9):oyag302. doi: 10.1093/oncolo/oyag302

Genomic landscape and clinical characteristics of ERBB2/HER2-mutated colorectal cancers

Muhammed Dogukan Aksu 1, Paola Zinser Peniche 2, Doga Kahramangil Baytar 3, Bhaghyasree Jambunathan 4, Yu Jen Alexander Jan 5, Ronan Hsieh 6, Shafia Rahman 7, John C Krauss 8, Christine M Veenstra 9, Ibrahim Halil Sahin 10,✉
PMCID: PMC13466721  PMID: 42536421

Abstract

Background

Molecular underpinnings and clinical characteristics of ERBB2/HER2-mutated colorectal cancers (CRCs) remain poorly understood. We aimed to comprehensively characterize the genomic landscape and clinical features of ERBB2/HER2-mutated CRC and to compare them with those of ERBB2/HER2-amplified CRC.

Patients and methods

We analyzed cBioPortal data from 4348 CRC patients with available ERBB2/HER2 profiling. Clinicodemographic features, microsatellite instability (MSI), tumor mutational burden, and alterations in 94 cancer-associated genes were evaluated using chi-square/Fisher’s exact test, Kruskal–Wallis test, and hierarchical clustering.

Results

ERBB2/HER2 putative driver mutations were identified in 95 patients (2.2%), amplifications in 87 (2.0%), and wild-type status in 4166 (95.8%). In the overall cohort, ERBB2/HER2-mutated tumors demonstrated significant KRAS enrichment (50.5%) compared to wild-type (37.7%; P = .014) and amplified tumors (18.4%; P < .001), alongside MSI enrichment (37.2% vs 1.6%; P < .001), elevated tumor mutational burden, and right-sided predominance. ERBB2/HER2-mutated tumors also showed broad driver co-alterations largely reflecting their MSI background. In contrast, ERBB2/HER2-amplified tumors exhibited left-sided predominance, with microsatellite stable (MSS)-restricted analyses revealing selective mutual exclusivity with KRAS (Log2 OR = −1.58; q = 0.019) and TP53 co-occurrence (Log2 OR = 2.04; q = 0.019). Hierarchical clustering confirmed that mutated and amplified tumors consistently occupy separate clusters across anatomical sites.

Conclusion

ERBB2/HER2 mutations and amplifications represent molecularly and clinically distinct subgroups. ERBB2/HER2-mutated tumors are characterized by an MSI-enriched, hypermutated background with frequent concurrent KRAS alterations, whereas amplified tumors exhibit selective KRAS mutual exclusivity and TP53 co-occurrence. These exploratory findings support alteration-specific therapeutic strategies in CRC.

Keywords: colorectal neoplasms, colon, rectum, mutation, genomics, ERBB2/HER2

Implications for Practice.

This large-scale genomic analysis reveals that ERBB2/HER2-mutated and ERBB2/HER2-amplified colorectal cancers represent distinct molecular and clinical entities that might require different therapeutic approaches. ERBB2/HER2-mutated tumors frequently arise on an MSI-enriched, hypermutated background with concurrent KRAS alterations, suggesting a need for combination therapies and heightened resistance potential. Conversely, ERBB2/HER2-amplified tumors show KRAS mutual exclusivity and TP53 co-occurrence on a chromosomally unstable background, making them more suitable for focused HER2-targeted therapy. These findings reinforce the critical importance of molecular stratification beyond simple HER2 positivity, suggesting alteration-specific treatment algorithms and informing clinical trial design for precision oncology approaches in colorectal cancer.

Introduction

Colorectal cancer (CRC) represents the third most common cancer and the second most common cause of cancer-related deaths globally; the 5-year survival rate for metastatic CRC remains dismal at less than 20%.1,2

The molecular underpinnings of CRC are considered to be a key determinant of clinical behavior and therapeutic response. The most commonly followed molecular classification categorizes CRCs by 3 major genomic instability patterns: microsatellite instability (MSI), the CpG island methylator phenotype (CIMP), and chromosomal instability.3 The Consensus Molecular Subtypes (CMS) classification further identifies 4 distinct subtypes: CMS1 (microsatellite instability immune, 14%), hypermutated, microsatellite unstable and strong immune activation; CMS2 (canonical, 37%), epithelial, marked WNT and MYC signaling activation; CMS3 (metabolic, 13%), epithelial and evident metabolic dysregulation; and CMS4 (mesenchymal, 23%), prominent TGF-β signaling and stromal infiltration.4 Next-generation sequencing, however, has revealed even greater complexity, identifying several molecularly distinct subgroups in which specific driver mutations influence tumor biology and treatment sensitivity.5 For example, BRAF mutations carry significant biological heterogeneity and they have distinct impacts on disease biology, treatment response, and clinical outcomes.6

Among the targetable oncogenic drivers in CRC, human epidermal growth factor receptor 2 (ERBB2, also known as HER2) represents a particularly compelling therapeutic opportunity.7,8 ERBB2/HER2 is a member of the epidermal growth factor receptor (EGFR) family and plays crucial roles in cell proliferation, survival, and migration by activating downstream PI3K/AKT and RAS/RAF/MEK pathways.9 In CRC, ERBB2/HER2 alterations occur through 2 distinct mechanisms: gene amplification and mutations.9  ERBB2/HER2 amplifications in CRC have been associated with resistance to anti-EGFR therapies and generally poor prognosis.10 However, therapeutic advances have transformed the management of patients with ERBB2/HER2-amplified CRC.11,12

In contrast, ERBB2/HER2 mutations in CRC remain far less well-characterized despite representing a similarly prevalent alteration. Unlike the relatively uniform mechanism of ERBB2/HER2 amplification, ERBB2/HER2 mutations encompass a diverse spectrum, with point mutations detectable in up to 3% of CRC patients.13 Preclinical studies suggest that different ERBB2/HER2 mutations may have distinct functional consequences and therapeutic sensitivities, with some showing resistance to trastuzumab but sensitivity to tyrosine kinase inhibitors.14 However, clinical data on ERBB2/HER2-mutated CRC remain limited to small case series and retrospective analyses. Furthermore, it remains unclear whether ERBB2/HER2-mutated and ERBB2/HER2-amplified CRCs represent distinct molecular subtypes with different co-occurring genomic alterations, pathway dependencies, and clinical characteristics.

To address these knowledge gaps, we conducted a comprehensive genomic analysis using a large, multi-institutional cohort from the cBioPortal database. In this study, we aimed to characterize the clinical, demographic, and genomic features of ERBB2/HER2-mutated CRC compared to ERBB2/HER2-amplified and wild-type disease to inform precision medicine approaches.

Patients and methods

The cBioPortal for Cancer Genomics (https://www.cbioportal.org) is a comprehensive web-based platform that aggregates genomic and clinical data from major cancer sequencing studies.15 In this study, the cBioPortal for Cancer Genomics was systematically queried to identify patients with CRC and available ERBB2/HER2 profiling. Of the 26 studies listed under the bowel cancer category, 19 met eligibility criteria for further curation and analysis (Figure 1).16–32 After removing patients with multiple samples, duplicate entries, and non-colorectal primaries, a total of 4,557 unique patients remained, each with a single sample. These patients were subsequently queried for ERBB2/HER2 alteration status. Copy number alteration data were derived from either putative copy-number alterations from Genomic Identification of Significant Targets in Cancer (GISTIC) 2.0 analysis33 or copy number calls from targeted OncoPanel sequencing, depending on the contributing study. ERBB2/HER2 amplification was defined as high-level copy number gain (GISTIC score +2 or equivalent OncoPanel call), as annotated within cBioPortal and consistent with standard genomic profiling conventions used across the contributing studies. ERBB2/HER2 putative driver mutations were defined based on OncoKB annotation,34 which represents the standard variant classification framework integrated within cBioPortal, classifying variants as oncogenic or likely oncogenic based on functional and clinical evidence. Among the cohort, 274 patients harbored at least one ERBB2/HER2 alteration, while 117 patients had no profiling data available for both mutation analysis and copy number variation detection. To maximize statistical power and minimize selection bias, patients with either a known ERBB2/HER2 mutation or amplification were retained in the analysis. In the final cohort, patients were further excluded if they exhibited both driver mutation(s) and amplifications concurrently. For patients harboring both amplifications and mutations of unknown significance, classification was based on the presence of the driver alteration (amplification), and these cases were included in the amplification group (n = 4 patients with this pattern). ERBB2/HER2 wild-type patients were defined as those without any genetic alterations in ERBB2/HER2, regardless of predicted pathogenicity. The final study cohort included 4,348 patients, consisting of 95.8% (n = 4,166) ERBB2/HER2 wild-type cases, 2.2% (n = 95) cases with ERBB2/HER2 putative driver mutations only, and 2.0% (n = 87) cases with ERBB2/HER2 amplifications only.

Figure 1.

Vertical flow diagram tracing selection from 26 cBioPortal studies down to a final cohort of 4,348 patients, with exclusion counts branching off at each stage and the bottom row splitting into ERBB2/HER2 wild type, mutation only, and amplification only groups.

Open in a new tab

Study flowchart and patient selection criteria. The cBioPortal bowel cancer database was queried, yielding 26 initial studies. After excluding studies involving appendiceal cancer, pre-cancerous polyps, duplicate datasets, and those using a different reference genome (hg38), 19 studies comprising 6,800 samples were eligible for data extraction. Patients were subsequently excluded if they had multiple tumor samples (n = 239 patients, 580 samples), duplicate records (n = 1640), or non-colorectal primary sites, including small bowel cancer (n = 19), anal cancer (n = 2), colon adenocarcinoma in situ (n = 1), or tumor of unknown primary origin (n = 1), resulting in 4557 unique patients. Following the ERBB2/HER2 query, patients without profiling of mutation and copy number data (n = 117) and patients with concurrent ERBB2/HER2 driver mutations and amplifications (n = 8) or other alterations without driver events (n = 84) were excluded to ensure valid comparative analyses. The final analytic cohort comprised 4348 patients: 4166 ERBB2/HER2-wild-type, 95 with ERBB2/HER2 putative driver mutations only, and 87 with ERBB2/HER2 amplifications only.

Patients’ demographic and clinical variables, along with molecular features, were extracted directly from cBioPortal. All entries were systematically reviewed for data consistency and quality. In cases where conflicting or inconsistent data were identified, the affected observations were excluded from subsequent analyses to maintain data integrity. The total number of complete and valid observations for each variable is reported in Table 1.

Table 1.

Baseline clinical and genomic characteristics of patients with ERBB2/HER2-mutated, ERBB2/HER2-amplified, and ERBB2/HER2 wild-type colorectal cancer.

ERBB2/HER2 mutated (n = 95) ERBB2/HER2 amplified (n = 87) ERBB2/HER2 wild-type (n = 4166) P value
Female, n (%)a 36 (40.0%) 30 (34.9%) 1954 (48.2%) .017
Age (years), mean ± SDb 58.7 ± 15.0 54.4 ± 14.1 58.9 ± 14.8 .013
Race, n (%)c .098
 White 42 (71.2%) 55 (78.6%) 1824 (80.1%)
 African American 8 (13.6%) 5 (7.1%) 262 (11.5%)
 Asian 8 (13.6%) 8 (11.4%) 170 (7.5%)
 Other 1 (1.7%) 2 (2.9%) 21 (0.9%)
Cancer type, n (%)d .107
 Colon cancer 50 (56.2%) 53 (63.1%) 2668 (66.5%)
 Rectal cancer 39 (43.8%) 31 (36.9%) 1347 (33.5%)
Tumor site, n (%)e .005
 Left colon 19 (38.8%) 36 (70.6%) 1297 (51.1%)
 Right colon 30 (61.2%) 15 (29.4%) 1242 (48.9%)
Microsatellite instability, n (%)f 29 (37.2%) 1 (1.6%) 284 (9.6%) <.001
Tumor mutational burden, mean ± SDg 33.8 ± 44.8 6.1 ± 7.6 10.9 ± 20.9 <.001
Total mutation count, mean ± SDh 237.5 ± 935.2 15.4 ± 28.2 128.1 ± 427.5 <.001
Fraction of genome altered, mean ± SDi 0.14 ± 0.17 0.23 ± 0.16 0.20 ± 0.16 <.001
Open in a new tab

Patient characteristics were compared across 3 groups: ERBB2/HER2-mutated, ERBB2/HER2-amplified, and ERBB2/HER2-wild-type. Superscript letters indicate the number of patients with available data for each variable, in the respective order:

a

Gender: n = 90, 86, and 4054.

b

Age: n = 90, 87, and 3948.

c

Race: n = 59, 70, and 2277.

d

Cancer type: n = 89, 84, and 4015.

e

Tumor site (excluding rectal cancer): n = 49, 51, and 2539.

f

Microsatellite instability: n = 78, 64, and 2944.

g

Tumor mutational burden: n = 95, 77, and 3774.

h

Total mutation count: n = 95, 76, and 3764.

i

Fraction of genome altered: n = 71, 85, and 3211. Statistical analyses were performed using the chi-square test for categorical variables, the Fisher exact test for race, and the Kruskal–Wallis test for continuous variables. P values < .05 were considered statistically significant.

To contextualize ERBB2/HER2 alterations within the broader oncogenic landscape of CRC, a total of 94 genes were manually curated from core and extended oncogenic pathways based on established roles in CRC biology as defined in prior large-scale genomic studies and published pathway literature.28,35–39 These genes encompass tumorigenesis (CRC core and WNT, RAS-RAF-MAPK, TGF-β/SMAD, BMP/SMAD), therapeutic targeting (PI3K-AKT-mTOR, receptor tyrosine kinases, cell cycle, and growth control), and immune evasion (immune regulation and antigen presentation): CRC core and WNT (APC, ARID1A, AXIN2, FAT3, FAT4, FBXW7, GSK3B, RNF43, SOX9, TCF7L2, TP53); RAS-RAF-MAPK (ARAF, BRAF, HRAS, KRAS, MAP2K1, MAP2K2, MAPK1, MAPK3, MRAS, NRAS, RAF1, RRAS); PI3K-AKT-mTOR (AKT1, AKT2, AKT3, INPP4B, MTOR, PIK3CA, PIK3CB, PIK3CD, PIK3R1, PTEN, RICTOR, RPTOR); TGF-β/SMAD (SMAD2, SMAD3, SMAD4, SMAD7, TGFB1, TGFB2, TGFB3, TGFBR1, TGFBR2, TGFBR3, TGFBRAP1); BMP/SMAD branch (BMP1, BMP10, BMP15, BMP2K, BMP3, BMP4, BMP5, BMP6, BMP7, BMP8A, BMP8B, BMPER, BMPR1A, BMPR1B, BMPR2, SMAD1, SMAD5, SMAD6, SMAD9, TGFB1I1, TGFBI); receptor tyrosine kinases (EGFR, ERBB3, ERBB4, FGFR1, FGFR2, FGFR3, IGF1R, KIT, MET, PDGFRA); immune regulation and antigen presentation (B2M, CD274, CTLA4, HLA-A, HLA-B, HLA-C, LAG3, PDCD1, TAP1, TAP2); and cell cycle & growth control (CCNE1, CDK4, CDK6, CDKN2A, MYC, RB1, STK11). This gene panel was then queried across the final cohort of 4348 patients with CRC to assess pathway-level alteration patterns and visualize genomic landscapes. Specifically, for Figure 3 and Figure S3, a focused subset of 10 genes was selected for comparative driver alteration analysis based on their established role as the most frequently altered driver genes in CRC and their particular relevance as downstream convergence pathways of ERBB2/HER2 signaling, where enrichment patterns are most likely to carry biological and therapeutic significance.28,35,40,41

Figure 3.

Grid of eleven bar charts, one per gene, each showing frequency for ERBB2-mut, ERBB2-amp, and ERBB2-wt bars on its own y-axis scale, with asterisks marking significant differences in the TP53 and KRAS panels.

Open in a new tab

Comparative analysis of driver gene alterations across ERBB2/HER2-defined colorectal cancer subgroups among microsatellite stable (MSS) patients. Putative driver alteration frequencies for selected key genes are shown across ERBB2/HER2-mutated (n = 49), ERBB2/HER2-amplified (n = 63), and ERBB2/HER2 wild-type (n = 2660 ) subgroups. Genes are organized by pathway: CRC core/WNT signaling (APC, TP53, RNF43, AXIN2), RAS/RAF/MAPK signaling (KRAS, BRAF (all alterations), BRAF V600E, NRAS, MAP2K1/2 & MAPK1/3), and PI3K/AKT/mTOR signaling (PTEN, PIK3CA). Statistical comparisons were performed using Fisher’s exact test for overall group differences across all 3 subgroups, followed by pairwise comparisons with Benjamini–Hochberg false discovery rate correction applied across all pairwise tests. Significance levels indicate Benjamini–Hochberg-adjusted P values: *P < .05, **P < .01.

Statistical analysis

All statistical analyses were performed using R (version 4.4.1; R Foundation for Statistical Computing). Categorical variables were analyzed using the chi-square test or Fisher’s exact test, depending on the expected counts. Results were summarized as numbers and frequencies. Continuous variables were compared using the Kruskal–Wallis test and reported as mean ± SD. For comparative analysis of driver alteration frequencies across ERBB2/HER2-defined subgroups, an overall Fisher’s exact test was first performed across all 3 groups for each gene; pairwise comparisons were subsequently conducted when the overall test was significant, with Benjamini–Hochberg correction applied across all pairwise tests within each gene to control the false discovery rate (FDR; q values). Unsupervised hierarchical clustering was performed on a gene alteration frequency matrix representing the percentage of patients with each alteration across ERBB2/HER2-defined subgroups, computed separately for colon and rectal cancers. Euclidean distance and complete linkage were used for both row and column clustering, implemented via the “ComplexHeatmap” package in R. Associations between ERBB2/HER2 status and pathogenic gene alterations were evaluated using odds ratios (OR) from contingency tables, with Fisher’s exact test for significance testing and FDR correction as described above. Given that cBioPortal aggregates data from studies using panels with variable gene coverage, alteration frequencies for each gene were calculated using only patients with confirmed profiling data for that gene as the denominator. Similarly, co-occurrence and mutual exclusivity analyses were restricted to patients with profiling data available for both genes under comparison. All visualizations were created in R using “ggplot2,” “ComplexHeatmap,” “dplyr,” and related packages. Throughout the analyses, P-values < .05 were considered statistically significant, and q values < 0.05 (FDR-corrected) were considered significant for multiple comparison analyses. The specific statistical tests, along with corresponding P and q values where applicable, are reported in the figure legends.

Results

Patient demographics and genomic characteristics by ERBB2/HER2 status

Baseline clinical and genomic characteristics are summarized in Table 1. The frequency of ERBB2/HER2 putative driver mutations was 2.2% (n = 95), ERBB2/HER2 amplification was 2.0% (n = 87), and ERBB2/HER2 wild-type was 95.8% (n = 4166).

Among the clinical and demographic characteristics, race and cancer type (colon vs rectal) were found to be comparable across the ERBB2/HER2-mutated, ERBB2/HER2-amplified, and wild-type groups (P = .098 and P = .107, respectively). Age distribution differed significantly among the groups, with patients harboring ERBB2/HER2 amplification being younger (54.4 ± 14.1 years) compared to those with mutations (58.7 ± 15.0 years) and wild-type genotype (58.9 ± 14.8 years; P = .013). A significant sex distribution difference was also observed, with wild-type tumors being more frequent in female patients (48.2%) compared to ERBB2/HER2-mutated (40.0%) or amplified tumors (34.9%; P = .017). Tumor anatomical distribution in colon cancers differed significantly by ERBB2/HER2 status (P = .005), with ERBB2/HER2-mutated tumors more frequently in the right colon (61.2% vs 48.9% in wild-type) and amplified tumors in the left colon (70.6% vs 51.1%). MSI was present in 37.2% of ERBB2/HER2-mutated tumors and 9.6% of wild-type tumors, whereas there was only one case (1.6%) with an MSI tumor in the ERBB2/HER2-amplified cohort (P < .001). Given that MSI tumors inherently harbor increased mutational burden, we evaluated TMB among microsatellite stable (MSS) tumors only (n = 49, n = 63, and n = 2,660 for ERBB2/HER2-mutated, amplified, and wild-type, respectively). ERBB2/HER2-mutated tumors demonstrated elevated TMB compared to both amplified and wild-type groups (7.7 ± 3.9 vs 5.4 ± 2.5 and 6.8 ± 12.6, respectively; P < .001). In the overall cohort, ERBB2/HER2-mutated tumors similarly showed higher TMB (33.8 ± 44.8 vs 6.1 ± 7.6 and 10.9 ± 20.9; P < .001). Of note, TMB values were derived from cBioPortal’s pre-computed calculations without exclusion of the defining ERBB2/HER2 variant; however, a single point mutation contributes negligibly to genome-wide TMB. Regarding chromosomal instability, while ERBB2/HER2-amplified tumors showed a higher fraction of genome altered in the overall cohort (0.23 ± 0.16 vs 0.14 ± 0.17 and 0.20 ± 0.16; P < .001), this difference was not observed when restricting to MSS tumors (0.22 ± 0.19 vs 0.20 ± 0.17 and 0.20 ± 0.16; P = .914).

Genomic landscape and pathway alterations across ERBB2/HER2 subgroups

Comparative alteration frequencies between ERBB2/HER2-altered and wild-type tumors across the 30 most commonly altered genes in the overall cohort are illustrated in Figure 2. Canonical CRC drivers dominated the alteration landscape across both groups, with APC (63.2% vs 63.4%), TP53 (63.2% vs 60.6%), and KRAS (35.2% vs 37.7%) altered at broadly comparable frequencies between ERBB2/HER2-altered and wild-type tumors, respectively. Beyond these top drivers, PIK3CA (14.8% vs 15.4%) and SMAD4 (13.2% vs 11.7%) showed similarly comparable rates across the 2 groups. Notably, ERBB3 (7.7% vs 2.0%; q < 0.01) and ARID1A (14.8% vs 7.2%; q < 0.05) were significantly enriched in ERBB2/HER2-altered tumors. The comprehensive genomic alteration frequencies across all genes for the overall cohort and the MSS-restricted cohort are provided in Figures S1 and S2, respectively. These revealed broadly consistent alteration frequencies across pathways, with APC (74.7%), TP53 (74.2%), KRAS (43.4%), PIK3CA (16.4%), and SMAD4 (15.8%) maintaining similar relative dominance in MSS-restricted patients.

Figure 2.

Paired bar chart comparing alteration frequencies of select 30 genes in ERBB2/HER2-altered (including both ERBB2/HER2-mutated and amplified groups) versus ERBB2/HER2 wild type colorectal cancer, ordered by descending overall frequency, with bars colored by oncogenic pathway and asterisks marking significantly different genes.

Open in a new tab

Comparative driver genomic alteration frequencies of the 30 most commonly altered genes between ERBB2/HER2-altered and wild-type colorectal cancer. Alteration frequencies for the top 30 most commonly altered genes across the overall cohort (n = 4348) are shown for ERBB2/HER2-altered (n = 182) and wild-type (n = 4166) subgroups. Genes are ranked by descending overall alteration frequency and color-coded by oncogenic pathway. Patients without profiling data for a given gene were excluded from its frequency calculation. Statistical comparisons were performed using Fisher’s exact test with Benjamini-Hochberg false discovery rate correction. Significance levels indicate adjusted P values annotated on gene labels: *P < .05, **P < .01.

Comparative analysis of driver alterations of selected genes across ERBB2/HER2 subgroups restricted to MSS tumors is presented in Figure 3, with the overall cohort analysis provided in Figure S3. APC, the most frequently altered gene in CRC, showed similar mutation frequencies across MSS subgroups (71.4%, 73.0%, and 73.8% for mutated, amplified, and wild-type tumors, respectively; P > .05), as well as in the overall cohort (64.2%, 62.1%, and 63.4%, respectively; P > .05). Within the RAS/RAF/MAPK pathway, ERBB2/HER2-mutated MSS tumors demonstrated the highest KRAS alteration frequency (46.9%), followed by wild-type (43.7%) and amplified tumors (20.6%), with ERBB2/HER2-amplified tumors showing significantly reduced KRAS alterations compared to both mutated (P = .031) and wild-type groups (P = .004), consistent with the overall cohort analysis (mutated: 50.5%, wild-type: 37.7%, amplified: 18.4%; mutated vs amplified P < .001, amplified vs wild-type P < .001, mutated vs wild-type P = .014). BRAF driver alterations were less frequent in amplified tumors compared to mutated and wild-type tumors in both the MSS-restricted (1.6% vs 6.1% and 6.3%, respectively; P > .05) and overall cohort analyses (1.1% vs 5.3% and 8.8%, respectively; P = .018 for amplified vs wild-type). To further characterize these BRAF alterations, we separately examined BRAF V600E driver alterations across subgroups. In the overall cohort, BRAF V600E was present in 1.1% of ERBB2/HER2-mutated, 0% of amplified, and 7.2% of wild-type tumors; the depletion in ERBB2/HER2-amplified tumors relative to wild-type was statistically significant (q = 0.008). In the MSS-restricted analysis, BRAF V600E was absent in both ERBB2/HER2-mutated and amplified MSS tumors (0% each) compared to 4.2% in wild-type. TP53 alterations were significantly more prevalent in ERBB2/HER2-amplified MSS tumors (92.1%) compared to both mutated (61.2%; P = .004) and wild-type cases (73.6%; P = .004), consistent with the overall cohort (amplified: 77.0%, mutated: 50.5%, wild-type: 60.6%; amplified vs mutated P < .001, amplified vs wild-type P = .003). PIK3CA alterations were comparable across all MSS subgroups (mutated: 12.2%, amplified: 14.3%, wild-type: 15.6%; all P > .05) and the overall cohort (all P > .05). No significant differences were observed for the remaining genes presented in Figure 3, including RNF43, AXIN2, NRAS, MAP2K1/2 and MAPK1/3, and PTEN, in the MSS-restricted analysis.

Anatomical site-specific genomic patterns and co-occurrence analysis

To evaluate whether tumor anatomical site influences the genomic similarity patterns of ERBB2/HER2-defined subgroups, we performed unsupervised hierarchical clustering based on subgroup-level gene alteration frequencies for both MSS-restricted tumors (Figure S4) and the overall cohort (Figure S5), understanding that this approach is descriptive in nature and, by design, precludes de novo group discovery. In MSS colon cancers, ERBB2/HER2-mutated tumors formed a distinct cluster separate from both wild-type and amplified groups, with wild-type and amplified tumors clustering together, consistent with the overall cohort analysis. In MSS rectal cancers, ERBB2/HER2-mutated and wild-type tumors clustered more closely, while amplified tumors formed a separate cluster. Importantly, across both anatomical sites in both MSS-restricted and overall cohort analyses, ERBB2/HER2-mutated and amplified tumors occupied distinct clusters. However, transcriptomic analyses will be further needed to validate expression level differences and characterize these subgroup distinctions.

We next assessed co-occurrence and mutual exclusivity patterns between ERBB2/HER2 alterations and the 94-gene panel in both the overall cohort and MSS-restricted tumors (Figure 4). In the overall cohort analysis, the ERBB2/HER2-mutated subgroup demonstrated significant co-occurrence with ten genes: ARID1A (Log2 OR = 1.96, q < 0.001), ERBB3 (Log2 OR = 2.67, q < 0.001), BMPR1A (Log2 OR = 3.08, q = 0.006), AXIN2 (Log2 OR = 2.01, q = 0.022), B2M (Log2 OR = 1.74, q = 0.034), RNF43 (Log2 OR = 1.57, q = 0.044), RB1 (Log2 OR = 2.38, q = 0.044), TCF7L2 (Log2 OR = 1.37, q = 0.044), HLA-A (Log2 OR = 2.36, q = 0.044), and CDKN2A (Log2 OR = 2.09, q = 0.044). Among ERBB2/HER2-amplified tumors, significant mutual exclusivity with KRAS was observed (Log2 OR = −1.42, q = 0.017), while TP53 co-occurrence showed a consistent directional effect (Log2 OR = 1.12, q = 0.082). In the MSS-restricted analysis, ERBB2/HER2-amplified tumors retained selective mutual exclusivity with KRAS (Log2 OR = −1.58, q = 0.019) and significant co-occurrence with TP53 (Log2 OR = 2.04, q = 0.019), while no significant associations were identified for the ERBB2/HER2-mutated subgroup. This likely reflects the reduced statistical power of the smaller MSS subgroup (n = 49 for mutated) and the known enrichment of several co-altered genes, including ARID1A,42  B2M,43  HLA-A,44  RNF43,45  TCF7L2,46  ERBB3,47 and AXIN2,46 in MSI tumors.

Figure 4.

Four volcano plots in two rows, ERBB2 mutations left and amplifications right, overall cohort above and MSS-restricted below, each plotting log2 odds ratio against negative log10 q-value with dashed significance and neutral-odds lines and gene names labeling the significant points.

Open in a new tab

Associations of ERBB2/HER2 alterations with selected oncogenic gene alterations. Volcano plots displaying statistical associations between ERBB2/HER2 mutations or amplifications and putative driver alterations across the curated 94-gene panel in the overall cohort and in the microsatellite stable (MSS)-restricted cohort. In the overall cohort, sample sizes were n = 95 for ERBB2/HER2-mutated and n = 87 for ERBB2/HER2-amplified tumors; in the MSS-restricted analysis, n = 49 and n = 63, respectively. Each point represents a gene, with color coding indicating association patterns: red denotes relative co-occurrence (q < 0.05, Log2 OR > 0), blue indicates relative mutual exclusivity (q < 0.05, Log2 OR < 0), and gray represents non-significant associations. The horizontal dashed line marks the significance threshold (q = 0.05), and the vertical dashed line indicates a neutral odds ratio (Log2 OR = 0). Statistical analyses employed Fisher’s exact test with Benjamini–Hochberg correction for false discovery rate control.

Discussion

To our knowledge, this comprehensive genomic analysis of 4,348 colorectal cancer patients represents one of the largest systematic characterizations of ERBB2/HER2 alterations in CRC. Our findings demonstrate that while ERBB2/HER2 alterations collectively affect approximately 4.2% of CRC patients, mutations (2.2%) and amplifications (2.0%) exhibit fundamentally different biological behaviors.

ERBB2/HER2-mutated tumors arise within a hypermutated, MSI-enriched genomic context and frequently harbor concurrent KRAS driver alterations. In the overall cohort, those tumors show co-occurrence with ten genes spanning chromatin remodeling (ARID1A), WNT signaling (RNF43, AXIN2, TCF7L2), cell cycle regulation (RB1, CDKN2A), immune pathways (B2M, HLA-A), TGF-β superfamily signaling (BMPR1A), and receptor tyrosine kinase networks (ERBB3). However, this pattern did not persist in the MSS-restricted analysis and likely reflect the known prevalence of several of these genes within the MSI-high background. In contrast, ERBB2/HER2-amplified tumors represent molecularly discrete events with a sparse driver interaction profile. They demonstrate selective mutual exclusivity with KRAS and co-occurrence with TP53, consistent with a chromosomally unstable, KRAS-independent route of MAPK pathway activation. Hierarchical clustering further reinforced these distinctions across both anatomical sites and regardless of MSI status.

Our observed frequencies of ERBB2/HER2 putative driver mutations (2.2%) and amplifications (2%) are consistent with previous large-scale studies reporting ERBB2/HER2 mutations in 2-3% of CRCs and amplifications in 1.3-6% of cases, with some studies documenting combined ERBB2/HER2 alterations in up to 7% of patients.13,48–50 Additionally, the observed age and sex distribution differences are consistent with previous studies in ERBB2/HER2-positive colorectal cancer, which have reported median ages ranging from 54 to 56 years and male predominance.51,52

As a landmark novel finding, our study revealed that, in contrast to ERBB2/HER2-amplified tumors, ERBB2/HER2-mutated CRCs are highly enriched with KRAS driver alterations (46.9%), unlike the study by Loree et al., in which KRAS mutations were noted to be approximately 25-35% in their institutional cohorts.47 The high rates of concurrent KRAS mutation noted in our study suggest that the “addiction” of ERBB2/HER2-mutated CRC cells to ERBB2/HER2 activation might be limited by the requirement of a secondary oncogenic driver mutation of its own downstream signaling. Interestingly, ERBB2/HER2-altered tumors showed a depleted phenotype of concurrent BRAF V600E driver alterations, which is also notably different from previously published data.47 Targeting strategies directed solely at pathogenic ERBB2/HER2 mutations might potentially result in a relatively modest benefit or even treatment failure, as the tumor cells might be capable of forming bypass loops in the MAPK pathway. Thereby, ERBB2/HER2 mutations appear to be challenging targets due to high incidence rates. RAS oncogene mutations likely will require consideration of targeted approaches for RAS and HER2 oncogenes concurrently. Notably, the HER2-specific tyrosine kinase inhibitors have shown activity in ERBB2/HER2-mutated cancers, including breast and biliary cancers,53 and targetability of ERBB2/HER2-mutated RAS WT CRC can be addressed in similar basket trials in the future. It is also important to note that whether ERBB2/HER2 mutations may in fact confer resistance to KRAS- and BRAF-directed precision therapies remains an important unanswered question that warrants further investigations to better define their predictive role.

Our finding that 37.2% of ERBB2/HER2-mutated tumors exhibit MSI aligns with recent reports showing strong associations between ERBB2/HER2 mutations and hypermutated tumor phenotypes across gastrointestinal cancers.54 In contrast, the near absence of MSI in ERBB2/HER2-amplified tumors (1.6%) suggests these alterations arise through distinct mutational processes consistent with the chromosomal instability pathway that characterizes the majority of CRCs.55 The anatomical distribution patterns we observed, with ERBB2/HER2-mutated tumors enriched in the right colon and ERBB2/HER2-amplified tumors predominating in the left colon, align with previous reports demonstrating that ERBB2/HER2 amplification occurs more frequently in left-sided and rectal tumors.10,56 These patterns parallel the established molecular distinctions between right- and left-sided CRCs. Right-sided tumors are enriched for CMS1 and CMS3 subtypes, while left-sided tumors predominantly exhibit CMS2 and CMS4 patterns.4,57

The co-occurrence analysis revealed distinct pathway interaction patterns that provide insights into potential therapeutic vulnerabilities. Although the co-occurrence patterns observed in overall ERBB2/HER2-mutated tumors were not demonstrated in the MSS-restricted cohort, likely due to reduced cohort size, the overall genomic landscape of this subgroup still carries potential biological and therapeutic implications worth considering. For example, ARID1A has been linked to resistance to anti-EGFR therapies in CRC.58 The ERBB3 association may be more directly relevant to ERBB2/HER2 biology, given that ERBB3 serves as the preferred heterodimerization partner for ERBB2/HER2.48  B2M and HLA-A co-alterations are consistent with the MSI background and may reflect altered antigen-presenting mechanisms that could influence responses to anti-PD-1 therapy.44

In contrast, ERBB2/HER2-amplified tumors demonstrated a genomically sparse interaction profile, with only 2 significant associations: relative mutual exclusivity with KRAS and co-occurrence with TP53. The KRAS mutual exclusivity aligns with established patterns in CRC, where ERBB2/HER2 amplification might serve as an alternative mechanism of MAPK pathway activation. The TP53 co-occurrence, together with the higher fraction of genome altered observed in this subgroup, is in line with the chromosomal instability pathway of colorectal tumorigenesis. Early TP53 loss enables the tolerance of widespread copy number alterations, including ERBB2/HER2 amplification.55 Collectively, our findings suggest that ERBB2/HER2-targeted therapeutic strategies should be tailored according to the specific type of alteration. Mutations are associated with a multi-pathway dysregulated phenotype possibly requiring comprehensive targeted approaches with extensive potential resistance mechanisms, while amplifications could benefit from focused ERBB2/HER2 inhibition in a KRAS wild-type context.

There are several limitations to our study, including the inherent limitations of the retrospective nature of the study; potential selection bias due to preexisting CRC datasets in the cBioPortal with specific inclusion/exclusion criteria; the manual composition of the 94-gene panel, and the absence of detailed clinical and demographic information, treatment history, and treatment-related survival outcomes. Notably, the heterogeneous nature of the aggregated datasets precluded meaningful analysis of clinical outcomes, therapeutic responses to HER2-targeted agents or tyrosine kinase inhibitors, and survival endpoints. Prospective studies specifically designed to capture these endpoints in ERBB2/HER2 mutated vs amplified CRC populations are warranted. Additionally, ERBB2/HER2 copy number alteration data were derived from heterogeneous genomic profiling platforms across contributing studies, which may introduce variability in amplification calls. Finally, MSS-restricted subgroup analyses, including hierarchical clustering and co-occurrence and mutual exclusivity patterns, were limited by reduced sample sizes within ERBB2/HER2-mutated and amplified subgroups, which may have reduced statistical power to detect associations observed in the overall cohort. The major strengths of our study include the novelty of the scientific questions investigated, the large cohort size curated from cBioPortal, and the ability to evaluate comprehensive molecular data that encompass multi-institutional genomic datasets.

Conclusion

Our large-scale genomic profiling suggests that ERBB2/HER2 alterations in colorectal cancer might encompass 2 biologically distinct entities. ERBB2/HER2-mutated tumors represent a distinct subgroup of CRC characterized by an MSI-enriched, hypermutated background with frequent concurrent KRAS driver alterations that may carry implications for resistance and may require combination targeted therapy approaches including RAS inhibitors, due to extensive co-mutation status. In contrast, ERBB2/HER2-amplified tumors exhibit chromosomally unstable patterns with depleted KRAS and BRAF driver alterations, TP53 co-occurrence, and otherwise limited driver co-alterations, which may make them more amenable to focused ERBB2/HER2-targeted therapy. These exploratory findings highlight potentially different biological backgrounds of ERBB2/HER2-mutated and ERBB2/HER2-amplified CRC. Prospective studies are essential to validate our findings in these pathway interactions and evaluate treatment responses to targeted combinations.

Supplementary Material

oyag302_Supplementary_Data
oyag302_supplementary_data.zip (1.1MB, zip)

Contributor Information

Muhammed Dogukan Aksu, Division of Medical Oncology, Department of Oncology, Mayo Clinic, Rochester, MN, 55905, United States.

Paola Zinser Peniche, Department of Medicine, Division of Hematology and Oncology, University of Pittsburgh School of Medicine, Pittsburgh, PA, 15213, United States.

Doga Kahramangil Baytar, Department of Medicine, Division of Hematology and Oncology, University of Pittsburgh School of Medicine, Pittsburgh, PA, 15213, United States.

Bhaghyasree Jambunathan, Department of Medicine, Division of Hematology and Oncology, University of Pittsburgh School of Medicine, Pittsburgh, PA, 15213, United States.

Yu Jen Alexander Jan, Department of Medicine, Division of Hematology and Oncology, University of Pittsburgh School of Medicine, Pittsburgh, PA, 15213, United States.

Ronan Hsieh, Division of Hematology and Oncology, Swedish Cancer Institute, Seattle, WA, 98104, United States.

Shafia Rahman, Department of Internal Medicine, Division of Medical Oncology, The Ohio State University Comprehensive Cancer Center, Columbus, OH, 432120, United States.

John C Krauss, Division of Hematology and Oncology, University of Michigan Medical School, Ann Arbor, MI, 48109, United States.

Christine M Veenstra, Division of Hematology and Oncology, University of Michigan Medical School, Ann Arbor, MI, 48109, United States.

Ibrahim Halil Sahin, Division of Hematology and Oncology, University of Michigan Medical School, Ann Arbor, MI, 48109, United States.

Author contributions

Muhammed Dogukan Aksu (Data curation, Formal analysis, Investigation, Project administration, Writing—original draft, Writing—review & editing), Paola Zinser Peniche (Investigation, Writing—original draft, Writing—review & editing), Doga Kahramangil Baytar (Investigation, Writing—original draft, Writing—review & editing), Bhaghyasree Jambunathan (Investigation, Writing—original draft, Writing—review & editing), Yu Jen Alexander Jan (Investigation, Writing—original draft, Writing—review & editing), Ronan Hsieh (Investigation, Writing—original draft, Writing—review & editing), Shafia Rahman (Investigation, Writing—original draft, Writing—review & editing), John C. Krauss (Investigation, Writing—original draft, Writing—review & editing), Christine M. Veenstra (Investigation, Writing—original draft, Writing—review & editing), and Ibrahim Halil Sahin (Conceptualization, Investigation, Methodology, Supervision, Writing—original draft, Writing—review & editing)

Supplementary material

Supplementary material is available at The Oncologist online.

Funding

None declared.

Conflicts of interest

I.H.S. reports Advisory Board fees from Pfizer, Johnson & Johnson, Amgen, Seattle Genetics, GSK, Guardant Health, and Lumanity, and a research grant from BAYER.

Data availability

All data used in this study are publicly available through cBioPortal for Cancer Genomics (https://www.cbioportal.org). Specific datasets and studies analyzed are referenced within the manuscript and can be accessed directly via the cBioPortal website.

Ethics approval and consent to participate

This study utilized publicly available, de-identified genomic data from cBioPortal for Cancer Genomics (https://www.cbioportal.org/). The original studies from which these data were derived obtained appropriate ethics approval and patient consent at the time of data collection. This study was conducted in accordance with the Declaration of Helsinki principles for research involving human subjects.

References

  • 1. Wang J, Li S, Liu Y, Zhang C, Li H, Lai B.  Metastatic patterns and survival outcomes in patients with stage IV Colon cancer: a population-based analysis. Cancer Med. 2020;9:361-373. 10.1002/cam4.2673 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Biller LH, Schrag D.  Diagnosis and treatment of metastatic colorectal cancer: a review. JAMA. 2021;325:669-685. 10.1001/jama.2021.0106 [DOI] [PubMed] [Google Scholar]
  • 3. Simons CCJM, Hughes LAE, Smits KM, et al.  A novel classification of colorectal tumors based on microsatellite instability, the CpG island methylator phenotype and chromosomal instability: implications for prognosis. Ann Oncol. 2013;24:2048-2056. 10.1093/annonc/mdt076 [DOI] [PubMed] [Google Scholar]
  • 4. Guinney J, Dienstmann R, Wang X, et al.  The consensus molecular subtypes of colorectal cancer. Nat Med. 2015;21:1350-1356. 10.1038/nm.3967 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. De Nicola F, Goeman F, Pallocca M, et al.  Deep sequencing and pathway-focused analysis revealed multigene oncodriver signatures predicting survival outcomes in advanced colorectal cancer. Oncogenesis. 2018;7:55. 10.1038/s41389-018-0066-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Sahin IH, Xiu J, Baca Y, et al.  Clinical and molecular characteristics of class II and III BRAF mutations in colorectal cancer. NPJ Precis Oncol. 2026;10. 10.1038/s41698-026-01329-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Pectasides E, Bass AJ.  ERBB2 emerges as a new target for colorectal cancer. Cancer Discov. 2015;5:799-801. 10.1158/2159-8290.CD-15-0730 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Connell CM, Doherty GJ.  Activating HER2 mutations as emerging targets in multiple solid cancers. ESMO Open. 2017;2:e000279. 10.1136/esmoopen-2017-000279 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Nowak JA.  HER2 in colorectal carcinoma: Are We there yet?  Surg Pathol Clin. 2020;13:485-502. 10.1016/j.path.2020.05.007 [DOI] [PubMed] [Google Scholar]
  • 10. Strickler JH, Yoshino T, Graham RP, Siena S, Bekaii-Saab T.  Diagnosis and treatment of ERBB2-positive metastatic colorectal cancer: a review. JAMA Oncol. 2022;8:760-769. 10.1001/jamaoncol.2021.8196 [DOI] [PubMed] [Google Scholar]
  • 11. Meric-Bernstam F, Hurwitz H, Raghav KPS, et al.  Pertuzumab plus trastuzumab for HER2-amplified metastatic colorectal cancer (MyPathway): an updated report from a multicentre, open-label, phase 2a, multiple basket study. Lancet Oncol. 2019;20:518-530. 10.1016/S1470-2045(18)30904-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Siena S, Di Bartolomeo M, Raghav K, et al.  DESTINY-CRC01 Investigators. Trastuzumab deruxtecan (DS-8201) in patients with HER2-expressing metastatic colorectal cancer (DESTINY-CRC01): a multicentre, open-label, phase 2 trial. Lancet Oncol. 2021;22:779-789. 10.1016/S1470-2045(21)00086-3 [DOI] [PubMed] [Google Scholar]
  • 13. Vaghi C, Mauri G, Agostara AG, et al.  The predictive role of ERBB2 point mutations in metastatic colorectal cancer: a systematic review. Cancer Treat Rev. 2023;112:102488. 10.1016/j.ctrv.2022.102488 [DOI] [PubMed] [Google Scholar]
  • 14. Robichaux JP, Elamin YY, Vijayan RSK, et al.  Pan-cancer landscape and analysis of ERBB2 mutations identifies poziotinib as a clinically active inhibitor and enhancer of T-DM1 activity. Cancer Cell. 2019;36:444-457 e7. 10.1016/j.ccell.2019.09.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Cerami E, Gao J, Dogrusoz U, et al.  The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov. 2012;2:401-404. 10.1158/2159-8290.CD-12-0095 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Wala J, de Bruijn I, Coy S, et al. Integrating spatial profiles and cancer genomics to identify immune-infiltrated mismatch repair proficient colorectal cancers. bioRxiv. 2024. Preprint. 10.1101/2024.09.24.614701 [DOI]
  • 17. Chatila WK, Walch H, Hechtman JF, et al.  Integrated clinical and genomic analysis identifies driver events and molecular evolution of colitis-associated cancers. Nat Commun. 2023;14:110. 10.1038/s41467-022-35592-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Cercek A, Chatila WK, Yaeger R, et al.  A comprehensive comparison of early-onset and average-onset colorectal cancers. J Natl Cancer Inst. 2021;113:1683-1692. 10.1093/jnci/djab124 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Chatila WK, Kim JK, Walch H, et al.  Genomic and transcriptomic determinants of response to neoadjuvant therapy in rectal cancer. Nat Med. 2022;28:1646-1655. 10.1038/s41591-022-01930-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Giannakis M, Mu XJ, Shukla SA, et al.  Genomic correlates of immune-cell infiltrates in colorectal carcinoma. Cell Rep. 2016;15:857-865. 10.1016/j.celrep.2016.03.075 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Seshagiri S, Stawiski EW, Durinck S, et al.  Recurrent R-spondin fusions in colon cancer. Nature. 2012;488:660-664. 10.1038/nature11282 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Brannon AR, Vakiani E, Sylvester BE, et al.  Comparative sequencing analysis reveals high genomic concordance between matched primary and metastatic colorectal cancer lesions. Genome Biol. 2014;15:454. 10.1186/s13059-014-0454-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Li C, Sun Y-D, Yu G-Y, et al.  Integrated omics of metastatic colorectal cancer. Cancer Cell. 2020;38:734-747 e9. 10.1016/j.ccell.2020.08.002 [DOI] [PubMed] [Google Scholar]
  • 24. Yaeger R, Mezzadra R, Sinopoli J, et al.  Molecular characterization of acquired resistance to KRASG12C-EGFR inhibition in colorectal cancer. Cancer Discov. 2023;13:41-55. 10.1158/2159-8290.CD-22-0405 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Mondaca S, Walch H, Nandakumar S, Chatila WK, Schultz N, Yaeger R.  Specific mutations in APC, but not alterations in DNA damage response, associate with outcomes of patients with metastatic colorectal cancer. Gastroenterology. 2020;159:1975-1978 e4. 10.1053/j.gastro.2020.07.041 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Gularte-Mérida Rodrigo, Smith Shaleigh, Bowman Anita S, et al.  Same-cell co-occurrence of RAS hotspot and BRAF V600E mutations in treatment-naive colorectal cancer. JCO Precis Oncol.  2022; 6: e2100365. 10.1200/PO.21.00365 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Alatise OI, Knapp GC, Sharma A, et al.  Molecular and phenotypic profiling of colorectal cancer patients in West Africa reveals biological insights. Nat Commun. 2021;12:6821. 10.1038/s41467-021-27106-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Yaeger R, Chatila WK, Lipsyc MD, et al.  Clinical sequencing defines the genomic landscape of metastatic colorectal cancer. Cancer Cell. 2018;33:125-136 e3. 10.1016/j.ccell.2017.12.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Ganesh K, Wu C, O’Rourke KP, et al.  A rectal cancer organoid platform to study individual responses to chemoradiation. Nat Med. 2019;25:1607-1614. 10.1038/s41591-019-0584-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Guda K, Veigl ML, Varadan V, et al.  Novel recurrently mutated genes in African American colon cancers. Proc Natl Acad Sci U S A. 2015;112:1149-1154. 10.1073/pnas.1417064112 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Vasaikar S, Huang C, Wang X, et al.  Clinical Proteomic Tumor Analysis Consortium. Proteogenomic analysis of human colon cancer reveals new therapeutic opportunities. Cell. 2019;177:1035-1049 e19. 10.1016/j.cell.2019.03.030 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Roelands J, Kuppen PJK, Ahmed EI, et al.  An integrated tumor, immune and microbiome atlas of colon cancer. Nat Med. 2023;29:1273-1286. 10.1038/s41591-023-02324-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Mermel CH, Schumacher SE, Hill B, Meyerson ML, Beroukhim R, Getz G.  GISTIC2.0 facilitates sensitive and confident localization of the targets of focal somatic copy-number alteration in human cancers. Genome Biol. 2011;12:R41. 10.1186/gb-2011-12-4-r41 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Chakravarty D, Gao J, Phillips SM, et al.  OncoKB: a precision oncology knowledge base. JCO Precis Oncol. 2017;2017. 10.1200/PO.17.00011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Cancer Genome Atlas N. Comprehensive molecular characterization of human colon and rectal cancer. Nature. 2012;487:330-337. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Korkut A, Zaidi S, Kanchi RS, et al.  Cancer Genome Atlas Research Network. A pan-cancer analysis reveals high-frequency genetic alterations in mediators of signaling by the TGF-beta superfamily. Cell Syst. 2018;7:422-437.e7. 10.1016/j.cels.2018.08.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Kawazu M, Ueno T, Saeki K, et al.  HLA class I analysis provides insight into the genetic and epigenetic background of immune evasion in colorectal cancer with high microsatellite instability. Gastroenterology. 2022;162:799-812. 10.1053/j.gastro.2021.10.010 [DOI] [PubMed] [Google Scholar]
  • 38. Nunes L, Li F, Wu M, et al.  Prognostic genome and transcriptome signatures in colorectal cancers. Nature. 2024;633:137-146. 10.1038/s41586-024-07769-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Li Q, Geng S, Luo H, et al.  Signaling pathways involved in colorectal cancer: pathogenesis and targeted therapy. Signal Transduct Target Ther. 2024;9:266. 10.1038/s41392-024-01953-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Appleyard JW, Williams CJM, Manca P, Pietrantonio F, Seligmann JF.  Targeting the MAP kinase pathway in colorectal cancer: a journey in personalized medicine. Clin Cancer Res. 2025;31:2565-2572. 10.1158/1078-0432.CCR-25-0107 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Lustig B, Jerchow B, Sachs M, et al.  Negative feedback loop of Wnt signaling through upregulation of conductin/axin2 in colorectal and liver tumors. Mol Cell Biol. 2002;22:1184-1193. 10.1128/MCB.22.4.1184-1193.2002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Zhao S, Wu W, Jiang Z, et al.  Roles of ARID1A variations in colorectal cancer: a collaborative review. Mol Med. 2022;28:42. 10.1186/s10020-022-00469-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Battaglin F, Naseem M, Lenz HJ, Salem ME.  Microsatellite instability in colorectal cancer: overview of its clinical significance and novel perspectives. Clin Adv Hematol Oncol. 2018;16:735-745. [PMC free article] [PubMed] [Google Scholar]
  • 44. Grasso CS, Giannakis M, Wells DK, et al.  Genetic mechanisms of immune evasion in colorectal cancer. Cancer Discov. 2018;8:730-749. 10.1158/2159-8290.CD-17-1327 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Giannakis M, Hodis E, Jasmine Mu X, et al.  RNF43 is frequently mutated in colorectal and endometrial cancers. Nat Genet. 2014;46:1264-1266. 10.1038/ng.3127 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Kim MS, Kim SS, Ahn CH, Yoo NJ, Lee SH.  Frameshift mutations of Wnt pathway genes AXIN2 and TCF7L2 in gastric carcinomas with high microsatellite instability. Hum Pathol. 2009;40:58-64. 10.1016/j.humpath.2008.06.006 [DOI] [PubMed] [Google Scholar]
  • 47. Loree JM, Bailey AM, Johnson AM, et al.  Molecular landscape of ERBB2/ERBB3 mutated colorectal cancer. J Natl Cancer Inst. 2018;110:1409-1417. 10.1093/jnci/djy067 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Ross JS, Fakih M, Ali SM, et al.  Targeting HER2 in colorectal cancer: the landscape of amplification and short variant mutations in ERBB2 and ERBB3. Cancer. 2018;124:1358-1373. 10.1002/cncr.31125 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Kavuri SM, Jain N, Galimi F, et al.  HER2 activating mutations are targets for colorectal cancer treatment. Cancer Discov. 2015;5:832-841. 10.1158/2159-8290.CD-14-1211 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Siena S, Sartore-Bianchi A, Marsoni S, et al.  Targeting the human epidermal growth factor receptor 2 (HER2) oncogene in colorectal cancer. Ann Oncol. 2018;29:1108-1119. 10.1093/annonc/mdy100 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51. Yang L, Li W, Lu Z, et al.  Clinicopathological features of HER2 positive metastatic colorectal cancer and survival analysis of anti-HER2 treatment. BMC Cancer. 2022;22:355. 10.1186/s12885-022-09447-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52. Wei Q, Zhang Y, Gao J, et al.  Clinicopathologic characteristics of HER2-positive metastatic colorectal cancer and detection of HER2 in plasma circulating tumor DNA. Clin Colorectal Cancer. 2019;18:175-182. 10.1016/j.clcc.2019.05.001 [DOI] [PubMed] [Google Scholar]
  • 53. Nakamura Y, Mizuno N, Sunakawa Y, et al.  Tucatinib and trastuzumab for previously treated human epidermal growth factor receptor 2-Positive metastatic biliary tract cancer (SGNTUC-019): a phase II basket study. J Clin Oncol. 2023;41:5569-5578. 10.1200/JCO.23.00606 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54. Uchida S, Sugino T.  ERBB2-mutant gastrointestinal tumors represent heterogeneous molecular biology, particularly in microsatellite instability, tumor mutation burden, and co-mutated genes: an in silico study. Curr Issues Mol Biol. 2023;45:7404-7416. 10.3390/cimb45090468 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55. Pino MS, Chung DC.  The chromosomal instability pathway in colon cancer. Gastroenterology. 2010;138:2059-2072. 10.1053/j.gastro.2009.12.065 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56. Lee SM, Oh H.  RAS/RAF mutations and microsatellite instability status in primary colorectal cancers according to HER2 amplification. Sci Rep. 2024;14:11432. 10.1038/s41598-024-62096-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57. Loree JM, Pereira AAL, Lam M, et al.  Classifying colorectal cancer by tumor location rather than sidedness highlights a continuum in mutation profiles and consensus molecular subtypes. Clin Cancer Res. 2018;24:1062-1072. 10.1158/1078-0432.CCR-17-2484 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58. Johnson RM, Qu X, Lin C-F, et al.  ARID1A mutations confer intrinsic and acquired resistance to cetuximab treatment in colorectal cancer. Nat Commun. 2022;13:5478. 10.1038/s41467-022-33172-5 [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

oyag302_Supplementary_Data
oyag302_supplementary_data.zip (1.1MB, zip)

Data Availability Statement

All data used in this study are publicly available through cBioPortal for Cancer Genomics (https://www.cbioportal.org). Specific datasets and studies analyzed are referenced within the manuscript and can be accessed directly via the cBioPortal website.

Articles from The Oncologist are provided here courtesy of Oxford University Press

RESOURCES