Introduction

The current screening and diagnostic methods for colorectal cancer (CRC) primarily involve tumor markers, imaging examinations, and endoscopic evaluations, which have inherent limitations. Therefore, it is imperative to explore novel diagnostic techniques for early-stage CRC to improve patient prognosis and survival rates. Recent studies have demonstrated the significant potential of circulating exosome-associated microRNAs (miRNAs) as biomarkers in CRC; however, these investigations are limited and predominantly focused on advanced tumors. Consequently, the differences in circulating exosomal miRNAs, among different stages of CRC within the classic developmental model of 'normal mucosa, adenoma, early adenocarcinoma, advanced adenocarcinoma', remain unknown.

Aims & Methods

Plasma exosomes were isolated from colorectal adenoma (CRA), early CRC (Tis-T1), advanced CRC (T2-T4), and healthy control (NC) groups (n=5 for each group) using ultracentrifugation. The miRNA expression profiles of plasma exosomes in the four groups were compared using miRNA microarray analysis to identify differentially expressed miRNAs in the three colorectal disease groups compared to the NC group. The results of miRNA microarray detection were validated by qRT-PCR to increase the sample size (n=30 for each group) and confirm the expression of characteristic miRNAs identified. Receiver operating characteristic curves were utilized to assess the diagnostic efficacy through area under curve (AUC).

Results

The extracted exosomes exhibited consistent morphology, size, and expression of protein markers that align with the fundamental characteristics of exosomes. Microarray analysis results revealed that both early and advanced CRC groups shared 31 up-regulated miRNAs and 34 down-regulated miRNAs compared to the CRA and NC groups. By intersecting with the GSE40247 dataset (CRC cell secreted exosome-derived microarray results), we identified five up-regulated miRNAs (miR-150-3p, miR-1224-5p, miR-1246, miR-1268a, and miR-1290). The expression levels of all five miRNAs showed a progressive increase with CRC advancement at different developmental stages. Notably, significant diagnostic potential was observed for miR-1224-5p, miR-1246, and miR 1290 in distinguishing patients with early CRC from NC (AUC 0.638 to 0.719). All five identified miRNAs demonstrated relatively good diagnostic performance in discriminating patients with advanced CRC from NC (AUC 0.745 to 0.838), while none of them had clear diagnostic value in distinguishing early CRC patients from individuals with CRA (AUC 0.548 to 0.654). Logistic regression analysis was employed to construct a multi-panel diagnosis model based on three specific miRNAs: miR-1224 -5p ,miR-1246, and miR-1290. The combined AUCs were found to be as follows: for discrimination between early CRC and NC groups, AUC=0.817 (P=0.0008); for discrimination between advanced CRC and NC groups, AUC=0.907 (P<0.0001); for discrimination between early CRC and CRA groups, AUC=0.754 (P=0.0035).

Conclusion

The miRNA expression profiles of plasma exosomes in early CRC patients exhibit similarities to those observed in advanced CRC patients. Specifically, the presence of miR-150-3p, miR-1224-5p, miR-1246, miR-1268a, and miR-1290 within plasma exosomes can serve as reliable indicators for the diagnosis of CRC. Importantly, a combination of miR-1224-5p, miR-1246, and miR-1290 shows promise for the early detection of CRC.

Introduction

Colorectal cancer (CRC) is the third commonest malignancy globally & the second leading cause of cancer mortality. Prognosis is highly stage-dependent, with survival exceeding 85% for early-stage diagnoses but dropping sharply at advanced stages. While national screening programs have reduced mortality in older populations, incidence in younger individuals is rising, often accompanied by delays in diagnosis. Existing non-invasive approaches, such as quantitative faecal immunochemical (qFiT) testing & blood-based liquid biopsies, provide limited sensitivity, particularly for pre-malignant or early-stage disease, leaving colonoscopy as the invasive gold standard. Because CRC is a mucosal disease, rectal mucus represents an under-explored biospecimen enriched for exfoliated cells, DNA, & microbial material, potentially enabling early & minimally invasive detection.

Aims & Methods

We assessed rectal mucus sampling as a novel diagnostic approach for CRC & adenomatous polyps. Using the OriCol™ device, rectal mucus was retrieved from 800 participants from 4 NHS colorectal surgical outpatients clinics without the need for bowel preparation (July 2020 - October 2023). A hologenomic strategy was applied, integrating three complementary laboratory pipelines: error-corrected next-generation sequencing for somatic mutations (n=161), enzymatic methyl-sequencing for DNA methylation profiling (n=208), & whole-genome shotgun metagenomics for microbiome characterisation (n=379). Data underwent rigorous quality control, with feature selection & integrative multi-omics analysis used to identify clinically informative biomarkers, stratify pathology by site & stage, and evaluate diagnostic potential.

Results

Distribution of CRC was 24% right-colon, 23% hepatic-splenic flexure, 18% left-colon, 35% rectal. Somatic mutation analysis revealed pathogenic variants in well-known CRC drivers including TP53, APC, KRAS, BRAF, ERBB2, FBXW7, PIK3CA & SMAD4, with mutation allele frequencies representing disease state. DNA methylation profiling demonstrated widespread hypermethylation at CpG islands near promoters and transcription start sites, particularly in SDC2, LRRC4, PPP2R5C & RNF217, with gradients of signal strength from rectal to right-sided cancers & from CRC to advanced adenomas (>10mm +/- HG dysplasia), to early polyps/normal controls. Microbiome profiling identified 94 species associated with CRC, including previously reported taxa such as Fusobacterium nucleatum & Hungatella hathewayi, alongside novel associations with Clostridium scindens & Anaerotruncus colihominis. Multi-omics integration markedly improved diagnostic resolution, clustering CRC & advanced adenomas polyps separately from controls, with early-stage polyps occupying an intermediate position.

Conclusion

We established rectal mucus as a rich biospecimen for minimally invasive CRC diagnostics. By combining host genetic & epigenetic markers with microbial signatures, the hologenomic approach provides sensitive detection of pre-cancerous lesions & stratification by site/stage, outperforming single-modality assays. Importantly, local mucosal sampling enabled discovery of novel microbial-pathology associations not detectable in stool-based studies. These findings demonstrate the translational potential of rectal mucus hologenomics as a scalable diagnostic platform, offering a non-invasive alternative for triaging symptomatic patients & potentially extending to screening programmes. Further clinical validation is underway to refine performance metrics & benchmark against existing diagnostic standards (ORICOL-EGI-04 Study NCT06649123).

References

Complete references for this work are listed in the accepted Nature Communications paper

Disclosure

1. This work has been accepted for publication in Nature Communications and is currently embargoed whilst final editing is completed. A letter of confirmation of acceptance for publication can be provided.
2. The study is funded and performed through Origin Sciences Ltd with many of the staff holding share options in the company. The clinical research teams were provided with educational grants from Origin Sciences Ltd to fund the research and support research fellows and nurses.
3. Rectal mucus samples and tumour tissues were collected following UK Health Regulatory Authority (HRA) and Ethics Committee approvals (IRAS 263745, 19/EM/0266). Written informed consent was obtained from all donors. All samples were collected from patients enrolled in the ORI-EGI-02 study; Exploratory Study of Rectal Mucus for Diagnosing Disease (NCT04659590) from four NHS Hospitals.
4. The presenting author is employed by the NHS as a Consultant Colorectal Surgeon and is Chief Medical Officer, Origin Sciences and holds share options with the company.

Introduction

Recently, some guidelines recommend conducting colorectal cancer (CRC) screening at ages younger than 50 years. However, there are few data on the preventive effect of colonoscopy at ages 40-49 years at the population level.

Aims & Methods

In this study, the risk of CRC according to the presence or absence of CSP in their 40s was evaluated in a retrospective cohort design using a nationwide database. Using the Korean National Health Insurance System data between 2004 and 2006, we constructed 1:5 colonoscopy cohort (CsC) and non-colonoscopy cohort (NCsC) at ages 40-49 years and followed through 2014. A total of 2,940,792 subjects were enrolled and CRC risk was evaluated after propensity score matching between CsC and NCsC based on age, sex, smoking, metabolic syndrome (MetS) and potentially relevant variables related CRC risk.

Results

Among 490,132 CsC subjects and 2,450,660 NCsC subjects at baseline periods in 40-49 ages, 689 (0.14%) and 9,440 (0.39%) cases of CRC were found during 10 years follow up period, respectively. According to age group, cumulative CRC incidences in early 40’s (40-44) were 0.10% in CsC and 0.25% in NCsC respectively. For late 40’s (45-49), those were 0.17% in CsC group and 0.49% in NCsC group, respectively. Compared with NCsC, the adjusted hazard ratios (aHRs) for CRC development in CsC group were 0.42 (men, 95% CI: 0.35-0.50) and 0.37 (women, 95% CI: 0.30-0.47) in early 40’s and 0.33 (men, 95% CI: 0.29-0.37) and 0.39 (women, 95% CI: 0.33-0.45) in late 40’s.

Conclusion

Overall, colonoscopy in their 40s significantly lowered the risk of colorectal cancer development by 58-67%. This cancer prevention effect of colonoscopy was different according to age and gender, and it was most effective in men in their late 40s, suggesting the rationales for optimization of age and gender to start colonoscopy.

Disclosure

Korea Health Technology R&D Project through the Korea Health Industry Development Institute, National Research Foundation of Korea grant and the National R&D Program for Cancer Control through the National Cancer Center funded by the Korea Ministry of Health & Welfare, Basic Science Research Program through the National Research Foundation of Korea funded by the Ministry of Education.