Low-Pass Whole Genome Bisulfite Sequencing for ctDNA Monitoring in Metastatic Colorectal Cancer: Insights from the VALENTINO Trial

Highlights

• Low-pass whole genome bisulfite sequencing (lp-WGBS) using the METER workflow offers a cost-effective, tumor-agnostic approach for longitudinal ctDNA monitoring in mCRC.
• Baseline ctDNA detection is a robust independent prognostic factor, associated with a 2.24-fold increase in the risk of death.
• CtDNA clearance at 8 weeks post-treatment initiation identifies a subgroup of patients with significantly superior progression-free and overall survival.
• METER demonstrates superior sensitivity in detecting ctDNA compared to traditional copy number alteration (CNA) and variant allele frequency (VAF) methods.

Background

The management of metastatic colorectal cancer (mCRC) has been significantly refined by the identification of molecular biomarkers, particularly RAS and BRAF mutational status, which dictate the use of anti-EGFR therapies such as panitumumab and cetuximab. However, radiological assessment remains the gold standard for monitoring treatment response, despite its inherent delays in capturing molecular-level changes and its inability to provide deep biological insights into tumor burden dynamics. Circulating tumor DNA (ctDNA) has emerged as a transformative tool for liquid biopsy, offering a non-invasive means to track disease progression and therapeutic efficacy.

Despite its promise, the widespread clinical adoption of ctDNA monitoring is hindered by high costs associated with deep-targeted sequencing and the logistical complexity of patient-specific assays (tumor-informed methods). Furthermore, current methodologies often rely on specific mutations or copy number alterations, which may not be present in all patients or may vary over time. Low-pass whole genome bisulfite sequencing (lp-WGBS) represents a novel alternative by leveraging epigenetic signatures—specifically DNA methylation—which are ubiquitous across cancer types and highly tissue-specific. The VALENTINO trial provides a unique clinical framework to explore the utility of this approach in patients with RAS wild-type mCRC receiving first-line systemic therapy.

Key Content

Methodological Advances: The METER Workflow

The primary challenge in liquid biopsy is distinguishing tumor-derived DNA from background cell-free DNA (cfDNA) derived from healthy cells. Traditional methods focus on genomic alterations. In contrast, the METER (Methylation-based Tumor Fraction Estimator) workflow utilizes lp-WGBS to infer ctDNA presence and tumor fraction by analyzing genome-wide methylation patterns. Methylation marks are significantly more abundant than individual point mutations, allowing for high sensitivity even at low sequencing depths.

In the exploratory analysis of the VALENTINO trial, METER was benchmarked against IchorCNA, a standard tool for estimating tumor fraction from copy number alterations. The study findings indicate that METER expands the detectable range of ctDNA, capturing positive cases that would be missed by CNA-based or VAF-based methods alone. This suggests that epigenetic-based liquid biopsies may provide a more comprehensive view of the tumor burden than purely genomic-based approaches.

Prognostic Significance of Baseline ctDNA

The evaluation of 154 patients from the VALENTINO trial revealed that baseline ctDNA was detectable in 72.7% of the cohort using the METER workflow. This baseline status served as a powerful prognostic indicator. Patients with detectable ctDNA at the start of first-line panitumumab/FOLFOX therapy had significantly worse outcomes compared to those with undetectable levels:

• Progression-Free Survival (PFS): Hazard Ratio (HR) of 1.65 (95% CI: 1.13–2.42; p=0.010).
• Overall Survival (OS): HR of 2.24 (95% CI: 1.37–3.66; p<0.001).

These results underscore that the absolute presence of ctDNA before treatment correlates with a higher disease burden or more aggressive biology, which radiological staging alone may not fully quantify.

Predictive Value of ctDNA Dynamics

The most significant clinical impact of the study was observed in the longitudinal monitoring of ctDNA dynamics. At the 8-week mark (coinciding with early clinical evaluation), ctDNA clearance was observed in 80.2% of patients who were initially ctDNA-positive. The persistence of ctDNA (failure to clear) at 8 weeks was a dire prognostic sign, associated with a massive increase in risk:

• PFS Risk: HR of 2.70 (95% CI: 1.63–4.49; p<0.001).
• OS Risk: HR of 3.37 (95% CI: 2.00–5.69; p<0.001).

Interestingly, while ctDNA clearance was associated with a more profound Depth of Response (DoR) (median reduction in tumor size -48.4% vs -41.2%; p=0.023), it did not significantly correlate with the frequency of Early Tumor Shrinkage (ETS). This suggests that ctDNA dynamics provide biological information that is distinct and potentially more sensitive than early radiological changes.

Expert Commentary

The VALENTINO exploratory analysis highlights a critical shift in how we might monitor mCRC. The use of lp-WGBS and the METER workflow addresses the two primary barriers to ctDNA adoption: cost and complexity. By using low-pass sequencing (~1x to 2x coverage) and focusing on epigenetic signals, this method bypasses the need for the deep sequencing (typically 5,000x to 10,000x) required for somatic mutation detection. This could reduce costs by an order of magnitude, making longitudinal monitoring economically viable for healthcare systems.

From a clinical perspective, the 8-week clearance marker is particularly compelling. In mCRC, where treatment intensity often fluctuates (e.g., de-escalation strategies), having a molecular confirmation of response could guide clinicians on when to maintain aggressive therapy or when to consider maintenance phases. However, several questions remain. The 27.3% of patients who were ctDNA-negative at baseline pose a challenge; does their negativity reflect truly low tumor burden, or a limitation in the sensitivity of the lp-WGBS approach for certain biological subtypes? Future integration of multi-modal liquid biopsies—combining methylation with fragmentomics—may further refine detection rates.

Conclusion

The implementation of METER-based ctDNA detection in the VALENTINO trial represents a significant step toward personalized, cost-effective oncology. The study proves that ctDNA dynamics measured via methylation are not only feasible but are highly predictive of long-term survival in RAS wild-type mCRC. As we move toward more nuanced treatment strategies, such as adaptive therapy based on molecular response, epigenetic-based liquid biopsies like lp-WGBS are likely to play a central role in the diagnostic and monitoring armamentarium of clinical oncologists.

Reference：

Manca P, Paoli M, Galardi F, Morano F, Di Donato S, Biganzoli L, Malorni L, Nardone A, Ambrosini M, Sciortino C, Oldani S, Ghelardi F, Nasca V, Damonte C, Villa C, Fazio R, Mohamed S, Demichelis F, Montroni I, Pusceddu S, Cremolini C, Lonardi S, Benelli M, Pietrantonio F. CtDNA detection with low-pass whole genome bisulfite sequencing in RAS wild-type metastatic colorectal cancer: an exploratory objective of the VALENTINO trial. Clin Cancer Res. 2025 Dec 22. doi: 10.1158/1078-0432.CCR-25-2773 IF: 10.2 Q1 B1