Colorectal cancer: detecting disease before the tumor appears?

By Ana Espino | Published on March 17, 2026 | 3 min read


Colorectal cancer (CRC) is the third most common cancer worldwide, with nearly 2 million new cases each year. Despite the existence of screening programs, mortality remains high, with 5-year survival dropping from 91% at early stage to 15% at stage IV. At the same time, the concerning rise in early-onset cases (<50 years) is reshaping the epidemiological landscape.  

Current strategies rely on the fecal immunochemical test (FIT) and colonoscopy, the diagnostic gold standard. However, FIT has limited sensitivity for early lesions and advanced adenomas. Colonoscopy, being invasive and costly, is difficult to implement at scale for population-wide screening.  

The current challenge is to identify minimally invasive, cost-effective tools that are sensitive to early-stage disease and capable of detecting not only tumor signals but also pre-malignant metabolic and immune alterations. This review, published in Clinical and Translational Medicine in 2025, examines the opportunities offered by multi-omics molecular profiling to improve early CRC detection.  

Can cancer be detected before it develops?  

The authors conducted a comprehensive review of epidemiological data, molecular pathways involved in tumorigenesis, and emerging diagnostic technologies.  

Biologically, CRC develops through the adenoma–carcinoma sequence, involving key mutations such as APC (≈80% of cases), followed by alterations in KRAS, TP53, or SMAD4. Three major pathways are described: chromosomal instability (70%), microsatellite instability (MSI), and the CpG island methylator phenotype (CIMP). These alterations define four consensus molecular subtypes (CMS1 to CMS4), highlighting tumor heterogeneity.  

Regarding screening, FIT shows variable sensitivity, which may drop to around 50% for stage I disease, limiting its effectiveness for early detection. Recently approved blood-based tests achieve sensitivities above 80% across all stages but remain less effective for precancerous lesions.  

Innovation lies in the integration of multi-omics liquid biopsies, combining genomics, proteomics, and metabolomics. High-throughput proteomic platforms (Olink®, SomaScan®, Seer®) enable the identification of plasma signatures associated with early-stage disease. Some studies report areas under the curve (AUC) above 0.80 for CRC detection.  

Infrared spectroscopy represents an original alternative. This technology analyzes the global biochemical profile of serum, integrating both tumor and non-tumor signals. A prospective study reported 91% sensitivity for CRC, with 100% detection of stage I–II disease in an exploratory cohort.  

The article also emphasizes the importance of non-tumor information, including chronic inflammation, gut dysbiosis, and metabolic reprogramming, which are present from premalignant stages.  

Toward integrated biological screening  

Colorectal cancer remains a major public health issue, exacerbated by the rise in early-onset cases. The key challenge is detection before symptoms or advanced lesions appear.  

This review aimed to assess the potential of integrated molecular profiling to transform screening. The data suggest that multi-omics approaches—particularly those combining genomic, proteomic, and metabolic signatures—could improve sensitivity for early-stage disease.  

Current limitations include cost, large-scale clinical validation, and reproducibility in real-world settings.  

Ultimately, integrating multi-omics tools capable of simultaneously analyzing tumor-derived and microenvironmental signals could enable personalized risk stratification, reduce unnecessary colonoscopies, and support truly early detection of colorectal cancer. 

Read next: CAR-T cells in colorectal cancer: the challenge of solid tumors