Original Article Transcriptomic validation and clinical translation of CCDC78 as a prognostic biomarker in colorectal cancer
Jiang Gong, Binsong Xia, Lei Qian, Yingchang Cai
Department of Anal and Pelvic Floor Surgery, The Quzhou Affiliated Hospital of Wenzhou Medical University, Quzhou People’s Hospital, Quzhou 324000, Zhejiang, China
Received June 12, 2025; Accepted July 23, 2025; Epub July 25, 2025; Published July 30, 2025
Abstract: This study investigates CCDC78 as a potential prognostic biomarker in colorectal cancer (CRC), incor- porating both clinical correlation and functional validation. Analysis of 135 paired tumor and adjacent tissues re- vealed significantly elevated CCDC78 expression in tumor tissues (P<0.001), and higher expression levels were associated with markedly lower 5-year survival rates (P=0.001). Time-dependent ROC curves demonstrated robust prognostic performance at 12, 36, and 60 months (AUCs of 0.85, 0.84, and 0.82, respectively). In vitro assays showed that CCDC78 overexpression significantly enhanced cell proliferation, migration, and invasion (P<0.05), whereas siRNA-mediated knockdown suppressed these phenotypes and increased apoptosis (P<0.01). Cox regres- sion analyses identified CCDC78 as an independent prognostic factor (P=0.02). Notably, despite similar baseline expression across CRC cell lines, SW480 cells were more sensitive to knockdown, while HCT116 cells more strongly recapitulated the overexpression phenotype. TCGA pan-cancer analysis showed upregulated CCDC78 in various tu- mors, including CRC, adrenocortical carcinoma (ACC), bladder cancer (BLCA), and kidney renal clear cell carcinoma (KIRC), reinforcing its broad oncogenic relevance. Correlation analyses linked high CCDC78 expression to older age, poor tumor differentiation, advanced TNM stage, lymph node metastasis, and distant metastasis. Immune profil- ing revealed negative associations with 11 immune cell types but a positive correlation with NK CD56 bright cells. Gene set enrichment analysis (GSEA) implicated CCDC78 in interferon-JAK-STAT, RIG-I/NFKB, and WNT signaling pathways. Altogether, these findings suggest that CCDC78 promotes CRC progression through enhancing tumor cell aggressiveness and modulating the immune microenvironment, underscoring its potential as a prognostic bio- marker and therapeutic target.
Keywords: Colorectal cancer, CCDC78, prognostic biomarker, immune microenvironment, GSEA, survival analysis
Introduction
Colorectal cancer (CRC) is one of the most com- mon malignancies globally, ranking highest in both incidence and mortality among digestive system cancers [1]. According to the World Health Organization (WHO), approximately 1.93 million new cases and 935,000 CRC-related deaths were reported worldwide in 2020, establishing CRC as a major public health con- cern [2]. Rising incidence rates, particularly in developing nations, are attributed to lifestyle modifications, dietary shifts, and population aging [3]. Despite advances in diagnosis and treatment, encompassing endoscopic screen- ing, targeted therapeutics, and immunothera- py, early detection remains suboptimal [4]. Patients diagnosed at advanced stages contin-
ue to experience poor prognosis, with five-year survival rates below 50% in specific popula- tions [5]. This clinical challenge underscores the urgent need for novel biomarkers to facili- tate early diagnosis, prognostic stratification, and personalized therapeutic strategies.
CRC pathogenesis involves a multifaceted inter- play of genetic, epigenetic, and environmental factors [6]. Aberrant gene expression, dysregu- lated signaling cascades, and tumor microenvi- ronment alterations are critical in the tumor initiation and progression [7]. The advent of high-throughput sequencing has enabled com- prehensive profiling of tumorigenesis-associat- ed genes and their interaction with the microen- vironment [8]. However, established biomark- ers including carcinoembryonic antigen (CEA)
CCDC78 as a prognostic biomarker in colorectal cancer
and carbohydrate antigen 19-9 (CA19-9) dem- onstrate limited sensitivity and specificity for early detection and prognostic evaluation [9]. Consequently, identifying novel molecular bio- markers with both prognostic and therapeutic relevance has become a research priority in CRC.
Coiled-Coil Domain Containing 78 (CCDC78) encodes a protein featuring coiled-coil motifs, which participate in cytoskeletal organization, signal transduction, and cellular differentiation [10]. These coiled-coil domains commonly mediate protein-protein interactions and play crucial roles in intracellular signaling networks [11]. Nevertheless, CCDC78 expression pat- terns, functional mechanisms, and prognostic implications in CRC remain largely unexplored. As an emerging candidate in cancer research, elevated CCDC78 expression may be associat- ed with tumor invasiveness, metastatic poten- tial, and adverse patient outcomes. Whether CCDC78 contributes to CRC progression through tumor microenvironment modulation, immune response regulation, or critical path- way activation remains unclear. Additionally, the relationship between CCDC78 and immune cell infiltration, along with potential immuno- therapeutic roles, warrants systematic investi- gation. Therefore, a comprehensive analysis of CCDC78 expression characteristics, prognostic significance, and underlying molecular mecha- nisms may clarify its tumorigenic role and iden- tify novel diagnostic and therapeutic targets.
This investigation aims to comprehensively evaluate the expression profiles, prognostic significance, and potential biological mecha- nisms of CCDC78 in CRC, thereby laying the theoretical foundations for clinical application as a biomarker. Our study addressed several key objectives: 1) to analyze CCDC78 expres- sion across various cancer types using pan- cancer datasets to assess its universality and specificity in tumor biology; 2) to evaluate CCDC78’s prognostic relevance in CRC patients using univariate and multivariate Cox regres- sion analyses and Kaplan-Meier survival analy- ses; 3) to explore correlations between CCDC78 expression and clinical parameters, including age, tumor differentiation, tumor-node-metas- tasis (TNM) staging, lymph node metastasis, and distant metastasis; 4) to examine its involvement in immune regulation and signal- ing pathway activity through immune cell infil-
tration analysis and Gene Set Enrichment Analysis (GSEA).
Methods and materials
Sample size calculation
The required sample size was estimated based on the findings of Matsuyama et al. [12], which reported a hazard ratio (HR) of 2.23 for ITGBL1 in predicting overall survival (OS) and a 5-year mortality rate of 29%. Using the formula for sample size calculation (n = (log(HR)2×Px(1-P)) ), with a=0.05 (Za/2= (Za/2+ZB)2 1.96), B=0.2 (ZB=0.84), HR=2.23, and P=0.29, a minimum of 59 samples was required.
Sample collection
Clinical data and tissue samples were obtained from 135 CRC patients treated at Quzhou People’s Hospital between January 2017 and March 2020. All patients provided written informed consent prior to participation. The study protocol and sample collection proce- dures were reviewed and approved by the Ethics Committee of Quzhou People’s Hospital.
Inclusion criteria: pathologically confirmed co- lorectal adenocarcinoma (colon or rectum); rad- ical surgical resection performed with intraop- erative collection of paired tumor and adjacent non-tumor tissues, which were immediately preserved for subsequent qRT-PCR analysis of CCDC78 expression; availability of complete TNM staging based on AJCC/UICC criteria; a minimum follow-up duration of 5 years; expect- ed survival time exceeding 6 months; and age between 18 and 85 years. Exclusion criteria: presence of other concurrent malignancies; receipt of neoadjuvant chemotherapy or radio- therapy prior to surgery; severe cardiovascular, hepatic, or renal dysfunction; or unclear origin of distant metastases.
Data acquisition and analysis
This investigation utilized The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) datasets to assess CCDC78 expression, prognostic significance, and potential molecu- lar mechanisms in both pan-cancer and CRC contexts. TCGA data (https://portal.gdc.cancer. gov/), accessed in October 2024, encom- passed RNA-sequencing (RNA-Seq) data in
CCDC78 as a prognostic biomarker in colorectal cancer
fragments per kilobase of transcript per million mapped reads [FPKM] format and correspond- ing clinical information across 33 cancer types. Sample selection criteria included complete CCDC78 expression data, definitive pathologi- cal diagnosis, clinical staging documentation, and follow-up duration exceeding six months. This yielded approximately 10,000 samples, comprising 643 CRC cases (colon adenocarci- noma [COAD] and rectum adenocarcinoma [READ]). The GEO GSE30378 dataset, based on teh Affymetrix Human Genome U133 Plus 2.0 Array platform (https://www.ncbi.nlm.nih. gov/geo/), facilitated Gene Set Enrichment Analysis (GSEA) to investigate CCDC78-asso- ciated signaling pathways.
GSEA analysis
GSEA was conducted using GSEA software (ver- sion 4.3.2) based on GSE30378 dataset analy- sis. The c2.cp.kegg and c5.go gene sets were analyzed. A total of 1,000 permutations were conducted, with statistical significance thresh- olds defined as P<0.05 and false discovery rate (FDR) q-value <0.25.
Clinical data collection
Clinical and pathological data from 135 CRC patients who underwent radical surgery between January 2017 and March 2020 were retrieved from hospital electronic medical records. Collected parameters included age, sex, tumor differentiation grade, TNM staging (American Joint Committee on Cancer [AJCC]/ Union for International Cancer Control [UICC] criteria), lymph node metastasis status, pres- ence of distant metastasis, tumor diameter, anatomical location (left colon, right colon, sig- moid colon/rectum), KRAS mutation status, P53 mutation status, and Ki67 proliferation index. All data were independently verified by two trained personnel to ensure accuracy and completeness.
qRT-PCR analysis
qRT-PCR was used to assess CCDC78 expres- sion levels in tumor and adjacent non-tumor tis- sues from all 135 CRC patients. Total RNA was extracted from liquid nitrogen-preserved tis- sues using the EasyPure® RNA Kit (TransGen Biotech, Beijing, China). RNA purity was evalu- ated using a NanoDrop 2000 spectrophoto- meter (Thermo Fisher Scientific, USA), with
OD260/280 ratios ranging from 1.8 to 2.0. Complementary DNA (cDNA) was performed with 1 µg of total RNA using TransScript® All-in- One First-Strand cDNA Synthesis SuperMix (TransGen Biotech) at 42℃ for 30 minutes and 85℃ for 5 seconds.
qRT-PCR was performed on and Applied Bio- systems 7500 system (Thermo Fisher Scientific, USA). Each 20 uL reactions contained: 10 uL TransStart® Top Green qPCR SuperMix, 0.5 uL forward primer (10 uM, 5’-CTTGGGAGACGG- CCTAGTGG-3’), 0.5 uL reverse primer (10 µM, 5’-GCCTCAGGCGCTAAAAGCAG-3’), 2 uL cDNA, and 7 uL nuclease-free water. The expected amplicon size was 193 bp. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) served as an internal control (forward: 5’-GAGTCC- ACTGGCGTCTTCAC-3’; reverse: 5’-ATCTTGAGG- CTGTTGTCATACTTCT-3’). Thermal cycling condi- tions included initial denaturation at 95℃ for 3 minutes, followed by 40 cycles of 95℃ for 15 seconds, 60℃ for 30 seconds, and 72℃ for 30 seconds, with fluorescence collection per cycle. Relative expression levels of CCDC78 were calculated using the 2^(-44Ct) method, normalized to GAPDH. All reactions were per- formed in triplicate, and results were analyzed using ABI 7500 software (version 2.3).
Follow-up protocol
Five-year follow-up data for the 135 CRC patients were collected through March 2025, with documentation of survival status and time of death. Follow-up was conducted via electronic medical record reviews, telephone interviews, and outpatient consultations. During the first postoperative year, patients were monitored monthly for survival, recur- rence, and metastasis. From the second to the fifth year, follow-up assessments were conduct- ed quarterly in conjunction with outpatient evaluations.
Outcome measures
Primary outcomes: CCDC78 expression in CRC tissues was determined using qRT-PCR, and its association with five-year overall survival was assessed using Kaplan-Meier analysis. Univariate and multivariate Cox regression analyses were conducted to determine whether CCDC78 served as an independent prognostic indicator. Additionally, TCGA data were used to analyze CCDC78 expression across 33 cancer
CCDC78 as a prognostic biomarker in colorectal cancer
types to evaluate its tumor-type specificity and pan-cancer relevance in CRC.
Secondary outcomes: Correlations between CCDC78 expression and clinicopathological characteristics (age, differentiation grade, TNM staging, lymph node metastasis, distant metas- tasis) were analyzed. Immune correlation analy- ses were performed using CIBERSORT to esti- mate 23 immune cell subsets, and Spearman correlation was applied to examine relation- ships with CCDC78 expression. GSEA was uti- lized to explore signaling pathways associated with elevated CCDC78 expression, including type I interferon-Janus kinase-signal transduc- er and activator of transcription [JAK-STAT] and wingless-related integration site [WNT] path- ways. Stratified and regression analyses were conducted to evaluate CCDC78 prognostic sig- nificance across subgroups defined by age and TNM stage, and to explore potential interac- tions with other clinical variables.
Cell line sources
The human colorectal cancer cell lines SW480 and HCT-116 were obtained from the American Type Culture Collection (ATCC), while the normal human colonic epithelial cell line FHC was pur- chased from BeNa Culture Collection (China). All cell lines were cultured in in Roswell Park Memorial Institute (RPMI)-1640 medium con- taining 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin, maintained at 37℃ in a humidified incubator with 5% CO2.
Cell transfection
Cell transfections were performed using Lipofectamine® 3000 reagent (Thermo Fisher Scientific, Catalog No. L3000008) following manufacturer protocols. For CCDC78 overex- pression, cells were transfected with the pcDNA3.1-CCDC78 plasmid (overexpression group) or an empty vector (pcDNA3.1-NC, nega- tive control group). CCDC78 knockdown was achieved using a specific siRNA (si-CCDC78 group), with a non-targeting siRNA used as the negative control (si-NC group). Cells were har- vested 48 hours post-transfection for subse- quent experiments.
EdU assay
Cell proliferation was assessed using EdU Cell Proliferation Detection Kit (RiboBio, Catalog
No. C10310-1). Transfected cells were incubat- ed with EdU solution for 4 hours, followed by staining per manufacturer instructions. EdU- positive cells were visualized and quantified using a Leica DMI8 fluorescence microscope. The proportion of EdU-positive cells was calcu- lated to assess proliferative activity.
Transwell assay
Cell migration and invasion were assessed using Transwell chambers (Corning, Catalog No. 3422). Transfected HCT-116 cells in serum-free medium (50,000 cells/well) were seeded into the upper chambers. The lower chambers were filled with RPMI-1640 medium containing 10% FBS to serve as a chemoattractant. Following 24-hour incubation, migrated cells on the lower membrane surface were fixed and stained with crystal violet (Solarbio, Catalog No. G1010). Cells were then imaged and counted under a Leica DMI8 microscope.
Wound healing assay
Cells were seeded in 6-well plates (Corning, Catalog No. 3516) and cultured to approximate- ly 90% confluence. A linear scratch was gener- ated in the cell monolayer using a 10 uL pipette tip. Cells were then washed with phosphate- buffered saline (PBS) to remove debris and incubated in serum-free medium. Wound clo- sure was monitored and imaged at 0, 12, and 24 hours. Migration velocity quantified by mea- suring wound area reduction using ImageJ software.
Flow cytometry
Apoptosis assessment utilized flow cytometry (BD Biosciences, FACSAria™ Fusion). Cell stain- ing employed Annexin V-fluorescein isothio- cyanate/propidium iodide (FITC/PI) Apoptosis Detection Kit (BD Biosciences, Catalog No. 556547) per manufacturer instructions. Flow cytometry detected early and late apoptosis, with FlowJo software analysis calculating apop- totic cell proportions.
Western blot analysis
Total cellular proteins were extracted using RIPA flysis buffer (Beyotime, Catalog No. P0013C), and protein concentration were determined using the BCA Protein Assay Kit (Beyotime, Catalog No. P0012). Equal amounts
CCDC78 as a prognostic biomarker in colorectal cancer
of protein were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electropho- resis (SDS-PAGE) and transferred onto polyvi- nylidene fluoride (PVDF) membranes (Millipore, Catalog No. IPVH00010). Membranes were blocked and incubated overnight at 4℃ with the following primary antibodies: CCDC78 (1:1000, Abcam, Catalog No. ab124767), phos- phorylated JAK1 (p-JAK1; 1:1000, Cell Signal- ing Technology, Catalog No. 3371S), p-JAK2 (1:1000, Cell Signaling Technology, Catalog No. 3776S), p-STAT1 (1:1000, Cell Signaling Technology, Catalog No. 14994S), and p-STAT3 (1:1000, Cell Signaling Technology, Catalog No. 9131S). After washing, membranes were incu- bated with Horseradish peroxidase (HRP)- conjugated secondary antibodies (1:2000, Cell Signaling Technology, Catalog No. 7074S) for 1 hour at room temperature. Protein bands were visualized using an enhanced chemilumines- cence (ECL) detection kit (Thermo Fisher Scientific, Catalog No. 32209) and quantified using ImageJ software.
Statistical analysis
Statistical analyses were performed using SPSS 20.0 and R software (version 4.3.3). Categorical data were expressed as percent- ages and compared using the chi-square test. The normality of continuous data was assessed using the Kolmogorov-Smirnov test. Normally distributed data were presented as mean ± standard deviation (SD) and compared using independent-sample t-tests, while non-normal- ly distributed data were compared using the Mann-Whitney U test. Survival analysis was conducted using Cox regression models, with univariate and multivariate analyses performed to estimate hazard ratios (HR) and 95% confi- dence intervals (CIs).
Differential gene expression analysis was con- ducted with DESeq2. Kaplan-Meier survival curves were generated using the survival and survminer R packages, statistical significance was assessed by the log-rank test (P<0.05). Time-dependent receiver operating character- istic (ROC) curves were plotted using the pROC package. Forest plot for Cox regression were created using forestplot. ggplot2 was utilized for generating scatter and lollipop plots, and clusterProfiler was used for GSEA (1,000 per- mutations, P<0.05, FDR q<0.25), Correlation
analysis was conducted using Spearman’s rank correlation with visualization via the corrplot package. Statistical significance was estab- lished at P<0.05. Results were visualized employing forest plots, survival curves, ROC curves, lollipop plots, and scatter plots, as appropriate.
Results
CCDC78 expression in pan-cancer analysis
Pan-cancer analysis was performed to assess CCDC78 expression across multiple cancer types, revealing significant differential expres- sion between normal and tumor tissues. Notably, CCDC78 expression was significantly upregulated in tumor tissues of bladder urothe- lial carcinoma (BLCA), breast invasive carcino- ma (BRCA), colon adenocarcinoma (COAD), esophageal carcinoma (ESCA), kidney renal clear cell carcinoma (KIRC), liver hepatocellu- lar carcinoma (LIHC), rectal adenocarcinoma (READ), stomach adenocarcinoma (STAD), and uterine corpus endometrial carcinoma (UCEC) (all P<0.001).
Elevated tumor expression of CCDC78 was also observed in kidney chromophobe (KICH) and prostate adenocarcinoma (PRAD) (P<0.01), as well as in cervical squamous cell carcinoma (CESC), cholangiocarcinoma (CHOL), kidney renal papillary cell carcinoma (KIRP), pheochro- mocytoma and paraganglioma (PCPG), and thy- roid carcinoma (THCA) (P<0.05).
Conversely, lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) demon- strated significantly lower CCDC78 expression in tumor tissues compared to normal tissues (P<0.05). No significant differences were ob- served in glioblastoma multiforme (GBM) and pancreatic adenocarcinoma (PAAD) (P>0.05). See Figure 1.
Prognostic value of CCDC78 in pan-cancer analysis
The prognostic significance of CCDC78 across various cancer types was evaluated using uni- variate Cox regression analysis, followed by Kaplan-Meier survival validation. Forest plot analysis showed that high CCDC78 expression was significantly associated with overall surviv- al (OS) in adrenocortical carcinoma (ACC),
”
The expression of CCDC78 Log2 (TPM+1)
8
m
6
J
A
JE
M
Normal
A
4
5
7
Tumor
¥
2
0
ACC
BLCA
BRCA
CESC
CHOL
COAD
DLBC
ESCA
GBM
HNSC
KICH
KIRC
KIRP
LAML
LGG
LIHC
LUAD
LUSC
MESO
OV
PAAD
PCPG
PRAD
READ
SARC
SKCM
STAD
TGCT
THCA
THYM
UCEC
UCS
UVM
BLCA, COAD, and KIRC (all P<0.05), but not in other cancer types (P>0.05, Figure 2A).
Kaplan-Meier survival analysis further con- firmed that patients with elevated CCDC78 expression exhibited significantly shorter OS in ACC (Figure 2B), BLCA (Figure 2C), COAD (Figure 2D), and KIRC (Figure 2E), compared to those with low expression levels (P<0.05). These find- ings suggest elevated CCDC78 expression is associated with poor prognosis in select can- cers, highlighting its potential utility as a can- cer-specific prognostic biomarker.
CCDC78 expression in CRC tissues and five- year survival association
qRT-PCR analysis of tumor and adjacent nor- mal tissues from 135 CRC patients revealed significantly higher CCDC78 expression in tumor tissues compared to adjacent non-tumor tissues (P<0.001, Figure 3A). Patients were stratified into high- and low-CCDC78 expres- sion groups based on the optimal cutoff values determined by X-tile analysis. Kaplan-Meier survival analysis demonstrated that patients in high-expression group had significantly lower five-year survival rates compared to those in low-expression groups (P<0.001, Figure 3B). Time-dependent ROC analysis was performed to evaluate the prognostic accuracy of CCDC78 expression. The AUC values were 0.951 at post- operative 12 months, 0.935 at 36 months, and 0.743 at 60 months (Figure 3C), indicating robust short- to medium-term predictive perfor-
mance, with moderate long-term prognostic value. Collectively, these findings suggest that high CCDC78 expression is strongly associated with unfavorable CRC patient prognosis, em- phasizing its potential prognostic evaluation value.
Association between CCDC78 expression and CRC clinicopathological features
Analysis of CCDC78 expression in relation to CRC clinicopathological features revealed sig- nificant differences between high- and low- expression groups across multiple variables. Elevated CCDC78 expression was significantly correlated with advanced age (P<0.001), poor tumor differentiation (P=0.026), advanced TNM staging (P=0.001), presence of lymph node metastasis (P=0.006), and distant metas- tasis (P=0.002). These associations suggest high CCDC78 expression is linked to more ag- gressive tumor phenotypes. In contrast, no sig- nificant associations were observed between CCDC78 expression and gender, perineural invasion, tumor diameter, anatomical location, KRAS mutation status, P53 mutation status, or Ki67 proliferation index (P>0.05) (Table 1).
CCDC78 expression across clinicopathological feature subgroups
Further subgroup analysis demonstrated con- sistent trends in CCDC78 expression across key clinicopathological variables. Patients aged >65 years exhibited significantly higher
A
CCDC78 - Overall Survival
B
C
0
1
2
3
4
5
| Group | Total(N) | HR(95% CI) | P value | 1.0 | CCDC78 | 1.0 | CCDC78 | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| ACC | 79 | 2.378 | (1.092 - 5.180) | 0.029 | Low | Low | |||||
| BLCA | 411 | 0.736 | (0.549 - 0.988) | 0.041 | 0.8 | High | 0.8 | High | |||
| BRCA | 1086 | 0.797 | (0.578 - 1.099) | 0.165 | probability | probability | |||||
| CESC | 306 | 0.850 | (0.534 - 1.351) | 0.491 | |||||||
| CHOL | 35 | 0.623 | (0.236 - 1.648) | 0.340 | 0.6 | 0.6 | |||||
| COAD | 477 | 1.837 | (1.230 - 2.742) | 0.003 | Survival | ||||||
| DLBC | 48 | 0.745 | (0.177 - 3.131) | 0.687 | Survival 0.4 | 0.4 | |||||
| ESCA | 163 | 0.942 | (0.578 - 1.535) | 0.811 | Overall Survival | Overall | Survival | H | |||
| GBM | 168 | 1.208 | (0.860 - 1.696) | 0.275 | HR = 2.38 (1.09 - 5.18) P = 0.029 | HR = 0.74 P = 0.041 | (0.55 - 0,99) | ||||
| HNSC | 503 | 0.865 | (0.661 - 1.130) | 0.287 | 0.2 | 0.2 | |||||
| KICH | 64 | 0.433 | (0.108 - 1.739) | 0.238 | 0 1000 2000 3000 4000 | 0 1000 | 2000 3000 | 4000 5000 | |||
| KIRC | 541 | 1.868 | (1.376 - 2.536) | <0.001 | Time (days) | Time (days) | |||||
| KIRP | 290 | 0.860 | (0.476 - 1.554) | 0.617 | Low | 39 23 14 5 | Low 206 | 44 18 5 | 2 1 | ||
| LAML | 139 | 0.768 | (0.501 - 1.178) | 0.227 | High | 40 23 8 3 | High 205 | 56 20 7 | 2 1 | ||
| LGG | 530 | 1.026 | (0.731 - 1.440) | 0.883 | |||||||
| LIHC | 373 | 1.147 | (0.813 - 1.618) | 0.436 | |||||||
| LUAD | 530 | 0.964 | (0.723 - 1.284) | 0.801 | |||||||
| LUSC | 496 | 0.786 | (0.598 - 1.033) | 0.084 | D 1.0 | CCDC78 | E 1.00 | CCDC78 | |||
| MESO | 86 | 1.156 | (0.726 - 1.840) | 0.541 | - Low | Low | |||||
| OV | 379 | 1.245 | (0.960 - 1.614) | 0.098 | High | High | |||||
| PAAD | 179 | 0.868 | (0.574 - 1.312) | 0.502 | probability 0.8 | probability 0.75 | |||||
| PCPG | 184 | 0.611 | (0.146 - 2.567) | 0.502 | |||||||
| PRAD | 501 | 1.718 | (0.408 - 7.240) | 0.461 | |||||||
| READ | 166 | 0.949 | (0.426 - 2.112) | 0.898 | 0.6 | 0.50 | |||||
| SARC | 263 | 1.028 | (0.692 - 1.527) | 0.892 | Survival | Survival | |||||
| SKCM | 457 | 1.119 | (0.856 - 1.463) | 0.410 | 0.4 | Overall Survival | 0.25 Overall | Survival | |||
| STAD | 370 | 0.883 | (0.637 - 1.225) | 0.456 | HR = 1.60 (1.13 - 2.28) | HR = 1.87 | (1.38 - 2.54) | ||||
| TGCT | 139 | 0.759 | (0.101 - 5.671) | 0.788 | P= 0.009 | P < 0.001 | |||||
| THCA | 512 | 1.355 | (0.506 - 3.631) | 0.545 | 0 1000 2000 3000 | 4000 | 0 1000 | 2000 3000 | 4000 | ||
| THYM | 119 | 1.334 | (0.356 - 5.004) | 0.669 | Time (days) | Time (days) | |||||
| UCEC | 553 | 0.690 | (0.458 - 1.040) | 0.076 | 20 6 | 171 75 29 | |||||
| UCS | 57 | 1.922 | (0.949 - 3.894) | 0.070 | Low | 321 109 | 2 | Low 270 | 3 | ||
| UVM | 80 | 2.335 | (0.956 - 5.703) | 0.063 | High | 322 95 20 10 | 4 | High 271 | 139 47 11 | 0 | |
Figure 2. Prognostic value of CCDC78 in pan-cancer analysis. A: Univariate Cox regression forest plot of CCDC78 in pan-cancer. B: OS curves for ACC patients strati- fied by high- or low-CCDC78 expression. C: OS curves for BLCA patients stratified by high- or low-CCDC78 expression. D: OS curves for COAD patients stratified by high- or low-CCDC78 expression. E: OS curves for KIRC patients stratified by high- or low-CCDC78 expression. Note: CCDC78, Coiled-coil domain containing 78; ACC, Adrenocortical Carcinoma; BLCA, Bladder Urothelial Carcinoma; COAD, Colon Adenocarcinoma; KIRC, Kidney Renal Clear Cell Carcinoma; OS, Overall Survival; CI, Confidence Interval.
A
4
B
K-M - CCDC78
C
expression of CCDC78
100%
1.0
3
75%
0.8
Relative
3-year survival rate (%)
Sensitivity (TPR)
2
0.6
50%
Low (= 2.14 )
High (> 2.14)
0.4
1
25%
P < 0.001
0.2
CCDC78
12-months (AUC = 0.951)
36-months (AUC = 0.935)
0
0%
60-months (AUC = 0.743)
0
10
20
30
40
50
60
0.0
Adjacent tissues
Cancer tissue
Time (months)
0.0
0.2
0.4
0.6
0.8
1.0
Number at risk
1-Specificity (FPR)
Low (=2.14) 78
78
78
78
77
71
62
High (> 2.14)
57
49
33
23
23
22
22
0
10
20
30
40
50
60
CCDC78 expression compared to those ≤65 years (P<0.001, Figure 4A). TNM staging analysis demonstrated significantly elevated CCDC78 expression in patients with advanced TNM stage (III+IV) compared to those in early- stage (I+II) (P<0.01, Figure 4C). Patients with lymph node or distant metastasis exhibited sig- nificantly higher CCDC78 expression versus those without metastasis (P<0.05, Figure 4D, 4E). However, no significant difference in CCDC78 expression was detected between poorly differentiated and moderately to well- differentiated tumors (P>0.05, Figure 4B). These findings further support the role of CCDC78 as a potential indicator of CRC inva- siveness and disease progression.
CCDC78 as an independent prognostic factor in CRC
Univariate Cox regression identified age (P< 0.001), tumor differentiation (P=0.002), TNM staging (P<0.001), and CCDC78 expression (P<0.001) as factors significantly associated with CRC prognosis, while gender, perineural invasion, tumor diameter, anatomical location, KRAS mutation, P53 mutation, and Ki67 prolif- eration index were not (P>0.05). Multivariate Cox regression confirmed age (P=0.012), TNM staging (P<0.001), and CCDC78 expression
(P=0.017) as independent CRC prognostic fac- tors, whereas tumor differentiation lost statisti- cal significance after adjustment for other vari- ables (P>0.05) (Table 2).
Similarly, TCGA-based univariate Cox regres- sion identified age (P<0.001), pathological T stage (P=0.004), and CCDC78 expression (P<0.001) as significant prognostic factors, while gender and body mass index (BMI) showed no significant association (P>0.05). Multivariate Cox regression further confirmed age (P<0.001), pathological T stage (P=0.007), and CCDC78 expression (P=0.003) as indepen- dent prognostic factors (Table 3). These results collectively demonstrate that elevated CCDC78 expression is consistently associated with poorer prognosis in both clinical and TCGA datasets and remains an independent prog- nostic factor after adjustment for confounding variables, suggesting its potential as a CRC prognostic biomarker.
Prognostic significance of CCDC78 across clinical subgroups
The prognostic value of CCDC78 was further evaluated in subgroups stratified by age and TNM stage, with patients divided into high- and low-expression groups using a 2.14 cutoff
CCDC78 as a prognostic biomarker in colorectal cancer
| Variable | Total | CCDC78 Expression Level | Chi-Square Value | P-Value | |
|---|---|---|---|---|---|
| High (n=57) | Low (n=78) | ||||
| Age | |||||
| >65 years | 55 (40.74%) | 40 (70.18%) | 15 (19.23%) | 33.325 | <0.001 |
| ≤65 years | 80 (59.26%) | 17 (29.82%) | 63 (80.77%) | ||
| Gender | |||||
| Male | 91 (67.41%) | 43 (75.44%) | 48 (61.54%) | 2.298 | 0.130 |
| Female | 44 (32.59%) | 14 (24.56%) | 30 (38.46%) | ||
| Differentiation | |||||
| Poorly Differentiated | 21 (15.56%) | 14 (24.56%) | 7 (8.97%) | 4.962 | 0.026 |
| Moderately/Well Differentiated | 114 (84.44%) | 43 (75.44%) | 71 (91.03%) | ||
| TNM Stage | |||||
| I-II | 69 (51.11%) | 39 (68.42%) | 30 (38.46%) | 10.661 | 0.001 |
| III-IV | 66 (48.89%) | 18 (31.58%) | 48 (61.54%) | ||
| Perineural Invasion | |||||
| Yes | 59 (43.70%) | 26 (45.61%) | 33 (42.31%) | 0.043 | 0.836 |
| No | 76 (56.30%) | 31 (54.39%) | 45 (57.69%) | ||
| Tumor Diameter | |||||
| >4 cm | 70 (51.85%) | 32 (56.14%) | 38 (48.72%) | 0.460 | 0.498 |
| ≤4 cm | 65 (48.15%) | 25 (43.86%) | 40 (51.28%) | ||
| Tumor Location | |||||
| Left Colon | 18 (13.33%) | 8 (14.04%) | 10 (12.82%) | 0.557 | 0.757 |
| Right Colon | 41 (30.37%) | 19 (33.33%) | 22 (28.21%) | ||
| Sigmoid Colon/Rectum | 76 (56.30%) | 30 (52.63%) | 46 (58.97%) | ||
| Lymph Node Metastasis | |||||
| Yes | 63 (46.67%) | 35 (61.40%) | 28 (35.90%) | 7.614 | 0.006 |
| No | 72 (53.33%) | 22 (38.60%) | 50 (64.10%) | ||
| Distant Metastasis | |||||
| Yes | 16 (11.85%) | 13 (22.81%) | 3 (3.85%) | 11.333 | 0.001 |
| No | 119 (88.15%) | 44 (77.19%) | 75 (96.15%) | ||
| KRAS Mutation | |||||
| Yes | 31 (22.96%) | 15 (26.32%) | 16 (20.51%) | 0.342 | 0.559 |
| No | 104 (77.04%) | 42 (73.68%) | 62 (79.49%) | ||
| P53 Mutation | |||||
| Yes | 81 (60.00%) | 37 (64.91%) | 44 (56.41%) | 0.669 | 0.413 |
| No | 54 (40.00%) | 20 (35.09%) | 34 (43.59%) | ||
| Ki67 Proliferation Index | |||||
| >60% | 105 (77.78%) | 49 (85.96%) | 56 (71.79%) | 3.050 | 0.081 |
| ≤60% | 30 (22.22%) | 8 (14.04%) | 22 (28.21%) | ||
Note: CCDC78, Coiled-coil domain containing 78; TNM, Tumor Node Metastasis; KRAS, Kirsten rat sarcoma viral oncogene homolog; P53, Tumor protein 53; Ki67, Proliferation marker protein Ki-67.
value. In the age-stratified analysis, patients >65 years with high CCDC78 expression had significantly poorer prognosis compared to those with low expression (P=0.029, Figure 5A), and this prognostic difference was more pronounced in patients ≤65 years (P<0.001, Figure 5B).
Similarly, TNM stage subgroup analysis showed stage III+IV patients with high CCDC78 expres- sion was associated with significantly worse survival (P=0.024, Figure 5C), with similar trends in stage I+II patients (P<0.001, Figure 5D). These findings indicate that high CCDC78 expression is consistently associated with
D
A
B
4
ns
Relative expression of CCDC78
3
w
Relative
expression of CCDC78
2
2
1
1
0
0
>65
≤65
Low
Mid + High
D
4
*
E
4
Relative
expression of CCDC78
Relative
expression of CCDC78
O
00
2
2
1
1
0
0
Transferred
Non-transferred
Distant Metastasis
No Distant Metastasis
C
4
expression of CCDC78
3
Relative
2
1
0
III + IV
I + II
worse prognosis across different age and TNM stage subgroups, reinforcing its potential as a robust CRC prognostic marker.
Interaction analysis of CCDC78 with prognostic variables
Interactions between CCDC78 expression and key CRC prognostic variables (age, tumor differ- entiation, TNM staging) were examined using regression models. Among patients >65 years, high CCDC78 expression (cutoff: 2.14) was associated with a 3.51-fold increased mortality risk (P=0.038, Figure 6A), whereas in patients ≤65 years, the risk was markedly higher at
8.37-fold (P=0.004). In the differentiation sub- group, high CCDC78 expression conferred a 4.15-fold increased risk in highly/moderately differentiated tumors (P=0.005, Figure 6B), and a 15.06-fold risk in poorly differentiated tumors (P=0.025). TNM stage subgroup analy- sis associated high CCDC78 expression with significantly elevated risk in stage I+II patients (P<0.001, Figure 6C), but no significant effects was observed in stage III+IV patients (P=0.072). These results suggest that the prognostic impact of CCDC78 is particularly pronounced in younger patients, poorly differentiated tumors, and early-stage CRC, highlighting its potential as prognostic biomarker.
CCDC78 as a prognostic biomarker in colorectal cancer
| Variable | Univariate | Multivariate | ||||
|---|---|---|---|---|---|---|
| β | P Value | HR (95% CI) | β | P Value | HR (95% CI) | |
| Age | ||||||
| >65 years | ||||||
| ≤65 years | -1.678 | <0.001 | 0.187 (0.102-0.343) | -0.887 | 0.012 | 0.412 (0.205-0.826) |
| Gender | ||||||
| Male | ||||||
| Female | 0.295 | 0.304 | 1.343 (0.765-2.358) | |||
| Differentiation | ||||||
| Poorly Differentiated | ||||||
| Moderately/Well Differentiated | -0.956 | 0.002 | 0.384 (0.207-0.712) | -0.343 | 0.284 | 0.709 (0.379-1.329) |
| Perineural Invasion | ||||||
| Yes | ||||||
| No | 0.144 | 0.614 | 1.154 (0.661-2.017) | |||
| Tumor Diameter | ||||||
| >4 cm | ||||||
| ≤4 cm | -0.089 | 0.750 | 0.914 (0.528-1.585) | |||
| Tumor Location | ||||||
| Left Colon | ||||||
| Right Colon | 0.366 | 0.475 | 1.442 (0.528-3.936) | |||
| Sigmoid Colon/Rectum | 0.403 | 0.405 | 1.496 (0.58-3.855) | |||
| TNM Stage | ||||||
| I-II | ||||||
| III-IV | -2.323 | <0.001 | 0.098 (0.042-0.231) | -1.886 | <0.001 | 0.152 (0.063-0.365) |
| KRAS Mutation | ||||||
| Yes | ||||||
| No | -0.459 | 0.135 | 0.632 (0.346-1.154) | |||
| P53 Mutation | ||||||
| Yes | ||||||
| No | -0.124 | 0.670 | 0.884 (0.501-1.559) | |||
| Ki67 Proliferation Index | ||||||
| >60% | ||||||
| ≤60% | -0.207 | 0.557 | 0.813 (0.407-1.623) | |||
| CCDC78 | ||||||
| ≥2.14 | ||||||
| <2.14 | -1.646 | <0.001 | 0.193 (0.106-0.35) | -0.84 | 0.017 | 0.432 (0.216-0.863) |
Note: CCDC78, Coiled-coil domain containing 78; TNM, Tumor Node Metastasis; KRAS, Kirsten rat sarcoma viral oncogene homolog; P53, Tumor protein 53; Ki67, Proliferation marker protein Ki-67; HR, Hazard Ratio; CI, Confidence Interval; BMI, Body Mass Index.
Correlation between CCDC78 and immune cell types
The relationship between CCDC78 expression and immune cell infiltration in CRC tissues was analyzed across 23 immune cell types. CCDC78 expression significantly correlated with 11 immune cell types (P<0.05, Figure 7A). Negative correlations were identified between
CCDC78 expression and eosinophils (r =- 0.158, P<0.001, Figure 7B), immature dendritic cells (iDC, r =- 0.092, P=0.019, Figure 7C), macro- phages (r =- 0.158, P<0.001, Figure 7D), mast cells (r =- 0.180, P<0.001, Figure 7E), neutro- phils (r =- 0.105, P=0.008, Figure 7F), T helper cells (r =- 0.132, P=0.001, Figure 7G), central memory T cells (Tcm, r =- 0.138, P<0.001, Figure 7H), follicular helper T cells (TFH,
CCDC78 as a prognostic biomarker in colorectal cancer
| Characteristics | Total (N) | Univariate | Multivariate | ||
|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | ||
| Age | 643 | ||||
| ≤65 | 276 | Reference | Reference | ||
| >65 | 367 | 1.939 (1.320-2.849) | <0.001 | 1.974 (1.331-2.926) | <0.001 |
| Gender | 643 | ||||
| Female | 301 | Reference | |||
| Male | 342 | 1.054 (0.744-1.491) | 0.769 | ||
| BMI | 329 | ||||
| ≤25 | 107 | Reference | |||
| >25 | 222 | 0.649 (0.394-1.069) | 0.090 | ||
| Pathologic T stage | 640 | ||||
| T1-T2 | 131 | Reference | Reference | ||
| T3-T4 | 509 | 2.468 (1.327-4.589) | 0.004 | 2.343 (1.257-4.367) | 0.007 |
| CCDC78 | 643 | 1.343 (1.127-1.601) | <0.001 | 1.303 (1.093-1.553) | 0.003 |
Note: CCDC78, Coiled-coil domain containing 78; BMI, Body Mass Index.
r =- 0.102, P=0.010, Figure 71), yō T cells (Tgd, r =- 0.142, P<0.001, Figure 7J), and Th2 cells (r =- 0.164, P<0.001, Figure 7K), suggesting that elevated CCDC78 expression may may be associated with reduced activity or infiltration of these immune cells. Conversely, CCDC78 expression was positively correlated with natu- ral killer (NK) CD56 bright cells (r=0.189, P<0.001, Figure 7L), indicating high CCDC78 expression may enhance the activity or recruit- ment of this NK cell subset. Overall, these find- ings suggest that CCDC78 may contribute to CRC progression through immune microenvi- ronment modulation.
GSEA analysis of CCDC78-associated path- ways in CRC
GSEA was performed to identify signaling path- ways potentially associated with CCDC78 in CRC and to elucidate its role in tumorigenesis and progression. GSEA identified five signifi- cantly enriched signaling pathways associated with high CCDC78 expression (P<0.05, false discovery rate [FDR] q-value <0.441).
The most strongly enriched pathways were the type I interferon-Janus kinase-signal transduc- er and activator of transcription (JAK-STAT) signaling pathway (set size =21, normalized enrichment score [NES]=1.825, P<0.001, q= 0.044) and the variant mutation-induced retro- grade axonal transport pathway (set size =27, NES=1.783, P<0.001, q=0.040). Additional sig-
nificantly enriched pathways included the RIG-I- NFKB signaling pathway (set size =18, NES= 1.761, P=0.001, q=0.039), WNT signaling modulation via WNT inhibitor (set size =25, NES=1.597, P=0.003, q=0.175), and the G pro- tein-coupled receptor (GPCR)-phospholipase C beta (PLCB)-inositol 1,4,5-trisphosphate recep- tor (ITPR) signaling pathway (set size =49, NES=1.457, P=0.011, q=0.441).
Among these, the type I interferon-JAK-STAT signaling pathway and retrograde axonal trans- port pathway exhibited the highest enrichment scores with statistical significance (q<0.05), suggesting that CCDC78 may promote CRC pro- gression through interferon-mediated immune signaling and cytoskeletal transport regulation. Enrichment of RIG-I/NFKB and WNT modulation pathways indicates potential CCDC78 involve- ment in inflammatory responses and WNT sig- naling inhibition, while enrichment of the GPCR- PLCB-ITPR pathway further supports potential regulation of the tumor microenvironment via GPCR-mediated signaling. Collectively, these findings suggest that CCDC78 may contribute to CRC progression by regulating immune sig- naling, cytoskeletal transport, and WNT-related pathways (Figure 8).
CCDC78 expression in cell line models
CCDC78 expression levels were examined in colorectal cancer cell lines and normal colonic epithelial cells. qRT-PCR demonstrated signifi- cantly higher CCDC78 expression in HCT-116
A
K-M - CCDC78
B
K-M -CCDC78
C
K-M -CCDC78
D
K-M -CCDC78
100%
100%
100%
100%
3-year survival rate (%)
75%
3-year survival rate (%)
75%
3-year survival rate (%)
75%
3-year survival rate (%)
75%
50%
High
50%
High
50%
High
50%
High
Low
Low
Low
Low
25%
25%
25%
25%
P = 0.029
P < 0.001
P = 0.024
P < 0.001
0%
0%
0%
0%
0
10
20
30
40
50
60
0
10
20
30
40
50
60
0
10
20
30
40
50
60
0
10
20
30
40
50
60
Time (months) Number at risk
Time (months) Number at risk
Time (months) Number at risk
Time (months) Number at risk
High
40
34
21
13
13
12
12
High
21
17
12
10
10
10
10
High
34
23
15
15
14
14
High
20
15
10
8
8
8
8
Low
15
15
15
15
14
10
7
Low
59
59
59
59
59
57
51
Low
32
32
32
31
25
16
Low
46
46
46
46
46
46
46
0
10
20
30
40
50
60
0
10
20
30
40
50
60
10
20
30
40
50
60
0
10
20
30
40
50
60
A
CCDC78: Age
B
CCDC78: Degree of differentiation
C CCDC78: TNM staging
Age
>65
$65
Degree of differentiation
High + Mid
Low
TNM staging
I + II
III + IV
4
4
15
log(Hazard ratio)
log(Hazard ratio)
log(Hazard ratio)
10
0
0
.
5
.
-4
-4
0
1.0
1.5
2.0
2.5
3.0
1.0
1.5
2.0
2.5
3.0
1.0
1.5
2.0
2.5
3.0
CCDC78
CCDC78
CCDC78
CCDC78 as a prognostic biomarker in colorectal cancer
A
B
C
D
E
0.6
CCDC78
0.4
0.4
NK CD56bright cells
R = 0.175 …
Eosinophils
:
0.4
Macrophages
0.4
TReg
R = 0.052ns
Mast cells
8
.
NK CD56dim cells
R = 0.029ns
0.2
S
0.2
Tem
R = 0.021ns
0.2
Cytotoxic cells
R = 0.018™s
-0.158
-0.092
= - 0.158
-0.180.
NK cells
R = 0.005”s
0.0
p = 0.000
p = 0.019
p = 0.000
P = 0.000
·
0
1
2
3
4
0.0
Q
1
2
3
4
0.0
0
1
2
3
4
0.0
0
1
2
3
4
CD8 T cells
R = 0.002”s
CCDC78
CCDC78
CCDC78
CCDC78
aDC
Th17 cells
R = - 0.008”s
P value
R = - 0.024ns
0.75
F
G
H
PDC
R = - 0.059ns
0.50
0.6
.
0.6
:
B cells
R = - 0.064”s
.
Th1 cells
NK.CD56bright cells
R = - 0.066”$
0.25
04
0.4
0.4
T.helper.cells
T cells
Neutrophils
R = - 0.069ns
0.4
Tcm
:
DC
R =- 0.073”s
|Cor]
0.2
0.2
R =- 0.082
0.05
:
0.2
0.2
T helper cells
.
TFH
R =- 0.086
=- 0.105
= 0: 189
=- 0.132
=- 0.138
iDC
R = - 0.087
0.10
0.0
P = 0.008
0.0
P = 0.000
0.0
P = 0.001
P = 0.000
0.15
0
1
2
3
4
0
1
2
3
4
0
1
2
3
4
0.0
0
1
2
3
4
Neutrophils
R = - 0.095
CCDC78
CCDC78
CCDC78
CCDC78
Tcm
Th2 cells
R = - 0.112
R =- 0.132
J
K
L
…
Mast cells
Macrophages
R = - 0.148
Eosinophils
R = - 0.149
0.4
C
0.4
R =- 0.150
Tgd
R =- 0.171
Th2.cells
…
TFH
Tgd
0.2
T
T
0.2
0.2
-0.1
0.0
0.1
Correlation
-0.102
-0.142
-0.164
0.0
-0.010
2
3
0.0
P = 0.000
p = 0.000
0
1
4
2
CCDC78
Q
1
3
4
0.0
0
2
3
4
CCDC78
CCDC78
Figure 7. Correlation between CCDC78 and immune cell types in CRC tissues. A: Lollipop plot of the correlations between CCDC78 and 23 immune cell types. B: Scatter plot of the correlations between CCDC78 and eosinophils. C: Scatter plot of the correlations between CCDC78 and immature dendritic cells (iDCs). D: Scat- ter plot of the correlations between CCDC78 and macrophages. E: Scatter plot of the correlations between CCDC78 and mast cells. F: Scatter plot of the correla- tions between CCDC78 and neutrophils. G: Scatter plot of the correlations between CCDC78 and NK CD56 bright cells. H: Scatter plot of the correlations between CCDC78 and T helper cells. I: Scatter plot of the correlations between CCDC78 and central memory T cells (Tcm). J: Scatter plot of the correlations between CCDC78 and follicular helper T cells (TFH). K: Scatter plot of the correlations between CCDC78 and yo T cells (Tgd). L: Scatter plot of the correlations between CCDC78 and Th2 cells. Note: CCDC78, Coiled-coil domain containing 78.
GSEA Bubble Plot
TYPE ___ INTERFERON_TO_JAK_STAT_SIGNALING_PATHWAY
RIG ___ NFKB_SIGNALING_PATHWAY
Set Size
☒
20
☒
30
☒
40
Pathway
WNT_SIGNALING_MODULATION_WNT_INHIBITOR
NES
High
RETROGRADE_AXONAL_TRANSPORT
Low
GPCR_PLCB_ITPR_SIGNALING_PATHWAY
0.5
0.6
0.7
Enrichment Score (ES)
A
**
B
*
*
3
2.5
expression of CCDC78
HCT-116
SW480
FHC
Relative expression level of CCDC78 protein
*
2.0
Relative
2
CCDC78
1.5
1
1.0
ß-actin
0.5
0
0.0
HCT-116
SW480
FHC
HCT-116
SW480
FHC
and SW480 cells compared to FHC cells at the mRNA level (P<0.01, Figure 9A). Western blot
analysis further validated these findings at the protein level (P<0.05, Figure 9B).
A
B
si-NC
si-CCDC78
3
SW480
HCT-116
3
SW480
HCT-116
expression of CCDC78
SW480
Relative EdU-positive Cell
2
2
Relative
1
1
pcDNA3.1-NC
pcDNA3.1-CCDC78
0
0
si-NC
si-CCDC78
pcDNA3.1-NC
pcDNA3.1-CCDC78
HCT-116
si-NC
si-CCDC78
pcDNA3.1-NC
pcDNA3.1-CCDC78
C
si-NC
si-CCDC78
1-1 : P1
40
SW480
HCT-116
9
1-2 : P1
Q1-UL(0.30%)
Q1-UR(3.97%)
“e
Q1-UL(0.22%)
Q1-UR(8.42%)
SW480
10°
105
*
PIPE-A
PIPE-A
Apoptosis rate (%)
30
e
0
20
0
o
Q1-LL(93.32%)
Q1-LR(2.41%)
Q1-LL(77.08%)
Q1-LR(14.28%)
102
103
104
105
Annexin V FITC-A
10ª
102
103
104
105
Annexin V FITC-A
10º
pcDNA3.1-NC
pcDNA3.1-CCDC78
10
O
2-1 : P1
Q1-UL(0.39%)
Q1-UR(4.23%)
0
2-2: P1
Q1-UL(0.30%)
Q1-UR(2.62%)
0
HCT-116
si-NC
si-CCDC78
pcDNA3.1-NC
pcDNA3.1-CCDC78
10
“e
PI PE-A
PI PE-A
10
:
0
0
Q1-LL(92.76%)
Q1-LR(2.62%)
Q1-LL(95.14%)
Q1-LR(1.94%)
102
10ª
104
105
10ª
102
103
104
105
Annexin V FITC-A
Annexin V FITC-A
10º
Functional analysis of CCDC78 in transfected cells: expression, proliferation, and apoptosis
qRT-PCR validated transfection efficiency in HCT-116 cells. CCDC78 mRNA expression was significantly increased in the pcDNA3.1- CCDC78 overexpression group compared to the pcDNA3.1-NC control group (P<0.0001, Figure 10A), while expression was markedly reduced in the si-CCDC78 knockdown group relative to the si-NC control group (P<0.001, Figure 10A). EdU staining revealed a significant increase in cell proliferation in the pcDNA3.1- CCDC78 group compared to pcDNA3.1-NC group (P<0.001, Figure 10B), while si-CCDC78 group demonstrated significantly reduced pro-
liferation relative to si-NC group (P<0.001, Figure 10B). Flow cytometry showed that the si-CCDC78 group had significantly higher apop- totic rates than the si-NC group (P<0.0001, Figure 10C), while pcDNA3.1-CCDC78 group exhibited significantly lower apoptosis rates versus pcDNA3.1-NC group (P<0.05, Figure 10C).
Effects of CCDC78 transfection on cellular in- vasion and migration
Transwell assay demonstrated significantly in- creased invasion ability in pcDNA3.1-CCDC78 transfected cells compared to pcDNA3.1-NC transfected cells (P<0.01, Figure 11A), where-
A
si-NC
si-CCDC78
300
SW480
HCT-116
SW480
Cell invasion number
200
**
**
100
pcDNA3.1-NC
pcDNA3.1-CCDC78
0
HCT-116
si-NC
si-CCDC78
pcDNA3.1-NC
pcDNA3.1-CCDC78
B
Relative cell migration rate
1.5
SW480
HCT-116
si-NC
si-CCDC78
pcDNA3.1-NC
pcDNA3.1-CCDC78
1.0-
*
0h
0h
SW480
HCT-116
0.5
0.0
48h
48h
si-NC
si-CCDC78
pcDNA3.1-NC
pcDNA3.1-CCDC78
as si-CCDC78 group showed significantly reduced invasion ability compared to the si-NC group (P<0.01, Figure 11A).
Similarly, wound healing assays showed signifi- cantly higher migration rates in the pcDNA3.1- CCDC78 group compared to the pcDNA3.1-NC group (P<0.001, Figure 11B), while the si- CCDC78 group demonstrated significantly lower migration rates relative to si-NC group (P<0.05, Figure 11B). These results indicate that CCDC78 promotes both invasion and migration in CRC cells.
CCDC78 regulates JAK-STAT signaling pathway proteins
Based on GSEA results, the JAK-STAT signaling pathway demonstrated significant enrichment in high CCDC78 expression profiles. Western blot analysis validated pathway activation at protein level. Knockdown of CCDC78 (si- CCDC78 group) significantly reduced the
expression of phosphorylated JAK1 (p-JAK1), p-JAK2, p-STAT1, p-STAT3, and CCDC78 in si- CCDC78 group compared with the si-NC group (P<0.0001, Figure 12). Conversely, CCDC78 overexpression (pcDNA3.1-CCDC78 group) led to significant upregulation of these phosphory- lated proteins compared to control (pcDNA3.1- NC group) (P<0.0001, Figure 12). These find- ings support CCDC78’s roles in JAK-STAT signal- ing pathway modulation.
Discussion
CRC remains one of the leading causes of can- cer-related morbidity and mortality worldwide, with persistently high mortality rates, despite advances in diagnostic and therapeutic tech- nologies [13]. Major clinical challenges, includ- ing low early detection rates and poor prog- nosis in advanced stages, continue to limit effective disease management [14]. CCDC78, a gene involved in cytoskeletal organization and signal transduction, has been suggested to
si-CCDC78
pcDNA3.1-NC
pcDNA3.1- CCDC78
Si-NC
JAK1
3
p-JAK1
protein expression level
si-NC
si-CCDC78
pcDNA3.1-NC
pcDNA3.1-CCDC78
JAK2
Relative
2
p-JAK2
STAT1
1
p-STAT1
STAT3
0
p-JAK1
p-JAK2
p-STAT1
p-STAT3
CCDC78
p-STAT3
CCDC78
ß-actin
regulate tumor cell behavior in preliminary investigations, yet its role in CRC has not been comprehensively characterized. This investiga- tion employed qRT-PCR, TCGA, and GSE30378 datasets to systematically analyze CCDC78 expression profiles, prognostic significance, and associated molecular mechanisms in CRC. The objective was to determine the potential of CCDC78 as a prognostic biomarker and to explore novel targets for precision diagnosis and therapy. The application of comprehensive methodologies has provided multidimensional insights into the role of CCDC78 in CRC progression.
Our findings revealed significantly elevated CCDC78 expression in CRC tumor tissues com- pared to adjacent non-tumor tissues, suggest- ing tis pivotal roles in promoting tumor cell proliferation, invasion, and metastasis during CRC initiation and progression. These findings align with previous reports demonstrating high CCDC78 expression in CRC [15]. Similarly, Fan et al. utilized machine learning approaches and reported elevated CCDC78 expression in pros- tate cancer and several other malignancies [16], indicating potential cancer-type specifici- ty. Mechanistically, CCDC78 overexpression may promote tumor aggressiveness through regulation of cytoskeletal dynamics or intra- cellular signaling pathways. Supporting this,
Lopergolo et al. demonstrated CCDC78 interac- tions with sarcoplasmic reticulum proteins that influence cytoskeletal regulation in muscle cells [17], while Hong et al. reported its colocal- ization with microtubules at ciliary tips, empha- sizing its role in cytoskeletal organization [18].
Pan-cancer analysis demonstrated that CC- DC78 is overexpressed in multiple cancer types, including CRC, ACC, BLCA, and KIRC, with the highest expression observed in CRC. Fan et al. reported that CCDC78 was closely associated with prognosis in castration-resis- tant prostate cancer (CRPC) [16], while Wu et al. identified CCDC78 as a microtubule-associ- ated marker significantly correlated with sur- vival in diffuse large B-cell lymphoma (DLBCL) [19]. These expression differences may stem from CCDC78’s distinct functions in diverse tumor microenvironments, such as cytoskele- tal organization or signal transduction roles, which vary by cancer type. In CRC, high CCDC78 expression was significantly associated with reduced five-year survival rates and exhibited high sensitivity and specificity for short- to medium-term prognosis prediction, highlighting its potential utility as an early risk stratification biomarker [15]. However, its long-term prog- nostic predictive performance declined, possi- bly due to tumor heterogeneity or therapeutic interventions (e.g., chemotherapy or targeted
CCDC78 as a prognostic biomarker in colorectal cancer
therapies) that modulate CCDC78 expression or function. Literature suggests single biomark- er predictive power may diminish in long-term follow-up [20], consistent with the observed CCDC78 patterns. Univariate and multivariate Cox regression analyses confirmed CCDC78 as an independent CRC prognostic factor, retain- ing robustness after adjusting for confounders including age and TNM stage [20]. TCGA data further supported its prognostic value across multiple cancers [16], suggesting CCDC78 may represent a candidate pan-cancer prognostic biomarker.
High CCDC78 expression was significantly associated with advanced age, poor tumor dif- ferentiation, advanced TNM staging, lymph node metastasis, and distant metastasis, indi- cating close association with aggressive tumor biology. For instance, patients with high CCDC78 expression were more likely diagnosed with advanced TNM stages, potentially reflect- ing its role in promoting tumor progression through pathways related to cell migration or invasion. In contrast, no significant associa- tions were observed with gender, KRAS muta- tion, or P53 mutation. This pattern suggests CCDC78 may drive tumor invasiveness through specific molecular pathways rather than being directly regulated by common oncogenic muta- tions. Larger, multicenter studies are warranted to validate these interactions and explore the molecular mechanisms by which CCDC78 con- tributes to CRC CRC progression in specific clinical subgroups.
Our investigation found CCDC78 expression was negatively correlated with multiple immune cell types, including eosinophils, macrophages, and T helper cells, suggesting it may promote tumor immune evasion by suppressing the infil- tration or activity of these cells. Previous reports noted negative correlations between CCDC78 expression and immune cells includ- ing macrophages and T helper cells [21]. Additionally, Gao et al. demonstrated that Gab2 influences M2 macrophage polarization, sup- porting potential immunosuppressive roles for CCDC78 [22]. Conversely, CCDC78 exhibited a positive correlation with NK CD56 bright cells, which are primarily involved in immune regula- tion. This association may reflect a role for CCDC78 in modulating cytokine secretion and inflammatory signaling, thereby indirectly fos-
tering a pro-tumorigenic microenvironment. These findings highlight the dual regulatory roles of CCDC78 in the CRC immune microenvi- ronment, offering new perspectives on its immunotherapy potential.
GSEA analysis indicated significant enrichment of high CCDC78 expression in several key sig- naling pathways, including type I interferon- JAK-STAT, RIG-I/NFKB, WNT, and GPCR-PLCB- ITPR signaling pathways. Among these, the most pronounced enrichment was observed in the type I interferon-JAK-STAT pathway, sug- gesting CCDC78 may promote tumor cell sur- vival by enhancing inflammatory and immune responses, potentially through STAT transcrip- tion factor overactivation. Chen et al. reported that CXCL1/miR-302e regulates CRC cell prolif- eration and metastasis via JAK-STAT pathway [23], while Ghasemian et al. noted that lncRNAs contribute to CRC progression via similar mech- anisms [24]. Enrichment of the RIG-I/NFKB pathway suggests that CCDC78 may enhance tumor cell anti-apoptotic capacity by upregulat- ing NFKB downstream anti-apoptotic genes, aligning with evidence that clAP2/NFKB activa- tion promotes metastasis and chemoresis- tance in CRC [25]. Furthermore, WNT signaling pathway enrichment indicates a potential role for CCDC78 in regulating tumor cell prolifera- tion and stemness through modulation of WNT inhibitors, potentially linked to aberrant B-catenin activity. This observation is support- ed by studies implicating ß-catenin as a critical mediator of SNTB1-mediated CRC progression [26]. GPCR-PLCB-ITPR pathway activation fur- ther supports CCDC78’s roles in reshaping tumor microenvironments via G protein-cou- pled receptor signaling, potentially affecting calcium flux and cell migration. Moy et al. dem- onstrated critical roles of ITPR3/Ca2+/RELB axis in CRC liver metastasis [27], while Lee et al. linked low PLCB4 expression to CRC drug resistance and MAPK/vascular endothelial growth factor (VEGF) pathway dysregulation [28]. Collectively, these pathways may act syn- ergistically to enhance CRC invasiveness and metastasis, underscoring CCDC78 as a poten- tial therapeutic target. Future investigations should employ functional experiments, such as RNA interference or CRISPR-mediated knock- out, to validate CCDC78’s specific roles in these pathways and assess its feasibility as a target- ed therapy candidate.
CCDC78 as a prognostic biomarker in colorectal cancer
By integrating qRT-PCR, TCGA, and GSE30378 datasets, this study systematically validated CCDC78 expression profiles, prognostic signifi- cance, and molecular mechanisms in CRC, addressing a critical literature gap. To our knowledge, this is the first comprehensive investigation to elucidate the role of CCDC78 in CRC [15]. It is also the first demonstrating CCDC78’s negative correlation with 11 immune cell types and positive correlation with NK CD56 bright cells, providing novel evidence of its involvement in tumor immune microenviron- ment. GSEA analysis further clarified CCDC78’s associations with key pathways including type I interferon-JAK-STAT and WNT, expanding our understanding of the molecular networks driv- ing CRC progression. Moreover, CCDC78 was validated as an independent prognostic bio- marker, providing theoretical support for its application in precision diagnosis and risk stratification in CRC. Its dual roles in immune regulation and signal transduction suggest potential applications in targeted and immuno- therapies. These findings not only deepen our understanding of CCDC78’s biological signifi- cance in CRC but also lay foundations for future clinical translation studies.
This investigation further explored the expres- sion of CCDC78 in different cell lines and its functional impacts. Through qRT-PCR and Western blot analysis, results showed signifi- cantly elevated CCDC78 expression in HCT-116 and SW480 cells compared to FHC cells, sug- gesting thar CCDC78 upregulation may be closely related to in CRC progression. Functional assays revealed that CCDC78 overexpression significantly enhanced cell proliferation, apop- tosis, migration, and invasion capabilities, whereas CCDC78 knockdown exerted the opposite effects. Flow cytometry analysis con- firmed that CCDC78 overexpression inhibited cell apoptosis, while its inhibition enhanced apoptosis. These results indicate that CCDC78 may play key roles in CRC cell biological behav- ior. Further analysis showed CCDC78 might regulate tumor invasiveness through JAK-STAT signaling pathways. Western blot results indi- cated that in CCDC78 overexpression groups, phosphorylation levels of JAK1, JAK2, STAT1, and STAT3 were significantly increased, while in inhibition groups, these levels were significant- ly decreased. This suggests CCDC78 may pro- mote tumor cell growth and metastasis by acti-
vating JAK-STAT signaling pathways. This find- ing provides new molecular targets for CRC therapy. Therefore, CCDC78 is not only a poten- tial prognostic biomarker for CRC but also a promising molecular target for future targeted therapeutic strategies.
This investigation provides valuable insights into the expression and functional roles of CCDC78 in CRC. Our findings indicate that CCDC78 plays critical roles in cell proliferation, invasion, and immune regulation. While this study presents compelling results, further validation in larger and multicenter cohorts would be beneficial to strengthen findings and confirm CCDC78’s clinical applicability as a prognostic biomarker. Additionally, the precise molecular mechanisms through which CCDC78 drives tumor progression and modulates immune responses remain incompletely under- stood. Future investigations should incorporate more comprehensive functional experiments, including in vivo animal models and detailed analysis of CCDC78 interactions with key CRC pathways. Furthermore, exploring relationships between CCDC78 and factors including micro- satellite instability (MSI) or tumor mutational burden (TMB) would enhance understanding of its immunotherapy roles and provide impor- tant insights into its role in immunotherapy responsiveness. Finally, integrating CCDC78 with other biomarkers, and exploring its thera- peutic application will be critical for advancing its clinical utility and guiding precision manage- ment of CRC.
Conclusion
CCDC78 is significantly overexpressed in CRC and strongly associated with poor prognosis, establishing its value as an independent prog- nostic biomarker. By modulating the immune microenvironment and key signaling pathways, including type I interferon-JAK-STAT and WNT, CCDC78 contributes to tumor progression and offers promising avenues for precision diagno- sis and targeted therapeutic development.
Disclosure of conflict of interest
None.
Address correspondence to: Yingchang Cai, De- partment of Anal and Pelvic Floor Surgery, The Quzhou Affiliated Hospital of Wenzhou Medical
CCDC78 as a prognostic biomarker in colorectal cancer
University, Quzhou People’s Hospital, No. 100 Minjiang Avenue, Kecheng District, Quzhou 324000, Zhejiang, China. E-mail: cyc0202@126.com
References
[1] McPhail S, Barclay ME, Johnson SA, Swann R, Alvi R, Barisic A, Bucher O, Creighton N, Denny CA, Dewar RA, Donnelly DW, Dowden JJ, Down- ie L, Finn N, Gavin AT, Habbous S, Huws DW, May L, McClure CA, Møller B, Musto G, Nilssen Y, Saint-Jacques N, Sarker S, Shack L, Tian X, Thomas RJS, Thomson CS, Wang H, Woods RR, You H and Lyratzopoulos G; ICBP Module 9 Chemotherapy Group. Use of chemotherapy in patients with oesophageal, stomach, colon, rectal, liver, pancreatic, lung, and ovarian can- cer: an International Cancer Benchmarking Partnership (ICBP) population-based study. Lancet Oncol 2024; 25: 338-351.
[2] Sung H, Ferlay J, Siegel RL, Laversanne M, So- erjomataram I, Jemal A and Bray F. Global Can- cer Statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 can- cers in 185 countries. CA Cancer J Clin 2021; 71: 209-249.
[3] Han B, Zheng R, Zeng H, Wang S, Sun K, Chen R, Li L, Wei W and He J. Cancer incidence and mortality in China, 2022. J Natl Cancer Cent 2024; 4: 47-53.
[4] Ju W, Zheng R, Zhang S, Zeng H, Sun K, Wang S, Chen R, Li L, Wei W and He J. Cancer statis- tics in Chinese older people, 2022: current burden, time trends, and comparisons with the US, Japan, and the Republic of Korea. Sci Chi- na Life Sci 2023; 66: 1079-1091.
[5] Cao W, Qin K, Li F and Chen W. Socioeconomic inequalities in cancer incidence and mortality: an analysis of GLOBOCAN 2022. Chin Med J (Engl) 2024; 137: 1407-1413.
[6] Schmitt M and Greten FR. The inflammatory pathogenesis of colorectal cancer. Nat Rev Im- munol 2021; 21: 653-667.
[7] Saadh MJ, Allela OQB, Kareem RA, Baldaniya L, Ballal S, Vashishth R, Parmar M, Sameer HN, Hamad AK, Athab ZH and Adil M. Prognos- tic gene expression profile of colorectal cancer. Gene 2025; 955: 149433.
[8] Lv X, Li X, Chen S, Zhang G, Li K, Wang Y, Duan M, Zhou F and Liu H. Transcriptional dysregula- tions of seven non-differentially expressed genes as biomarkers of metastatic colon can- cer. Genes (Basel) 2023; 14: 1138.
[9] Cai M, He H, Hong S and Weng J. Synergistic diagnostic value of circulating tumor cells and tumor markers CEA/CA19-9 in colorectal can- cer. Scand J Gastroenterol 2023; 58: 54-60.
[10] Mohar NP, Cox EM, Adelizzi E, Moore SA, Mathews KD, Darbro BW and Wallrath LL. The
influence of a genetic variant in CCDC78 on LMNA-associated skeletal muscle disease. Int J Mol Sci 2024; 25: 4930.
[11] Majczenko K, Davidson AE, Camelo-Piragua S, Agrawal PB, Manfready RA, Li X, Joshi S, Xu J, Peng W, Beggs AH, Li JZ, Burmeister M and Dowling JJ. Dominant mutation of CCDC78 in a unique congenital myopathy with prominent internal nuclei and atypical cores. Am J Hum Genet 2012; 91: 365-371.
[12] Matsuyama T, Ishikawa T, Takahashi N, Yama- da Y, Yasuno M, Kawano T, Uetake H and Goel A. Transcriptomic expression profiling identi- fies ITGBL1, an epithelial to mesenchymal transition (EMT)-associated gene, is a promis- ing recurrence prediction biomarker in colorec- tal cancer. Mol Cancer 2019; 18: 19.
[13] Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I and Jemal A. Global can- cer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 can- cers in 185 countries. CA Cancer J Clin 2024; 74: 229-263.
[14] Liu Y and Zheng Z. Understanding the global cancer statistics 2022: growing cancer bur- [15] den. Sci China Life Sci 2024; 67: 2274-2276. Jin W, Lu Y, Lu J, Wang Z, Yan Y, Liang B, Qian S, Ni J, Yang Y, Huang S, Han C and Yang H. Identification of cancer-associated fibroblast signature genes for prognostic prediction in colorectal cancer. Front Genet 2025; 16: 1476092.
[16] Fan C, Huang Z, Xu H, Zhang T, Wei H, Gao J, Xu C and Fan C. Machine learning-based identifi- cation of co-expressed genes in prostate can- cer and CRPC and construction of prognostic models. Sci Rep 2025; 15: 5679.
[17] Lopergolo D, Gallus GN, Pieraccini G, Boscaro F, Berti G, Serni G, Volpi N, Formichi P, Bianchi S, Cassandrini D, Sorrentino V, Rossi D, San- torelli FM, De Stefano N and Malandrini A. CCDC78: unveiling the function of a novel gene associated with hereditary myopathy. Cells 2024; 13: 1504.
[18] Hong J, Lee C, Papoulas O, Pan J, Takagishi M, Manzi N, Dickinson DJ, Horani A, Brody SL, Marcotte E, Park TJ and Wallingford JB. Molec- ular organization of the distal tip of vertebrate motile cilia. bioRxiv 2025.
[19] Wu W, Liu S, Tian L, Li C, Jiang Y, Wang J, Lv Y, Guo J, Xing D, Zhai Y, Sun H, Li Y, Zhang L, He X, Luo K, Zhan H and Zhao Z. Identification of microtubule-associated biomarkers in diffuse large B-cell lymphoma and prognosis predic- tion. Front Genet 2023; 13: 1092678.
[20] Tang Z, Wu Y, Sun D, Xue X and Qin L. A novel prognostic immunoscore based on The Cancer Genome Atlas to predict overall survival in colorectal cancer patients. Biosci Rep 2021; 41: BSR20210039.
CCDC78 as a prognostic biomarker in colorectal cancer
[21] Ding TT, Zeng CX, Hu LN and Yu MH. Establish- ment of a prediction model for colorectal can- cer immune cell infiltration based on the can- cer genome atlas (TCGA) database. Beijing Da Xue Xue Bao Yi Xue Ban 2022; 54: 203-208.
[22] Gao X, Long R, Qin M, Zhu W, Wei L, Dong P, Chen J, Luo J and Feng J. Gab2 promotes the growth of colorectal cancer by regulating the M2 polarization of tumor-associated macro- phages. Int J Mol Med 2024; 53: 3.
[23] Chen B, Song L, Nie X, Lin F, Yu Z, Kong W, Qi X and Wang W. CXCL1 regulated by miR-302e is involved in cell viability and motility of colorec- tal cancer via inhibiting JAK-STAT signaling pathway. Front Oncol 2021; 10: 577229.
[24] Ghasemian A, Omear HA, Mansoori Y, Man- souri P, Deng X, Darbeheshti F, Zarenezhad E, Kohansal M, Pezeshki B, Wang Z and Tang H. Long non-coding RNAs and JAK/STAT signaling pathway regulation in colorectal cancer devel- opment. Front Genet 2023; 14: 1297093.
[25] Wang R, Wang Y, Liu X, Liu M, Sun L, Pan X, Hu H, Jiang B, Zou Y, Liu Q, Gong Y, Wang M and Sun G. Anastasis enhances metastasis and chemoresistance of colorectal cancer cells through upregulating clAP2/NFKB signaling. Cell Death Dis 2023; 14: 388.
[26] Liang Y, Wang B, Chen S, Ye Z, Chai X, Li R, Li X, Kong G, Li Y, Zhang X, Che Z, Xie Q, Lian J, Lin B, Zhang X, Huang X, Huang W, Qiu X and Zeng J. Beta-1 syntrophin (SNTB1) regulates colorec- tal cancer progression and stemness via regu- lation of the Wnt/ß-catenin signaling pathway. Ann Transl Med 2021; 9: 1016.
[27] Moy RH, Nguyen A, Loo JM, Yamaguchi N, Kaj- ba CM, Santhanam B, Ostendorf BN, Wu YG, Tavazoie S and Tavazoie SF. Functional genetic screen identifies ITPR3/calcium/RELB axis as a driver of colorectal cancer metastatic liver colonization. Dev Cell 2022; 57: 1146-1159, e7.
[28] Lee CC, Lee AW, Wei PL, Liu YS, Chang YJ and Huang CY. In silico analysis to identify miR- 1271-5p/PLCB4 (phospholipase C Beta 4) axis mediated oxaliplatin resistance in metastatic colorectal cancer. Sci Rep 2023; 13: 4366.