Traditionally, liquid biopsy methods have relied on the premise that a substantial proportion of the DNA in circulation originates from tumor cells, enabling the detection of genetic alterations that provide insight into tumor presence and composition. The difficulty arises when cancer DNA constitutes only a minor fraction amidst overwhelming healthy DNA, creating a noisy biochemical backdrop that hides crucial mutation signals. This complexity not only impedes early detection but also limits the ability to track tumor evolution and treatment response over time, especially when conventional treatment effectively reduces tumor burden and subsequently the ctDNA signal.

The newly developed approach, named BayesCNA, is a sophisticated statistical algorithm designed to distill meaningful information from low-pass whole-genome sequencing data of blood samples. Low-pass sequencing, which involves scanning the genome at a lower depth than traditional high-coverage sequencing, offers a cost-effective overview of genomic alterations but compromises detailed signal quality due to reduced data resolution. BayesCNA addresses this trade-off by leveraging classical Bayesian statistics to amplify subtle signals embedded within low-quality data, thus allowing the detection of copy number alterations—variations in the number of copies of specific DNA segments—that are hallmarks of many cancers.

This statistical ingenuity departs from the prevalent reliance on machine learning and artificial intelligence, which, although powerful, often require large datasets and high-quality inputs to perform optimally. Remarkably, the Chalmers team discovered that classical statistical modeling outperforms contemporary machine learning techniques in extracting reliable tumor-associated signals from samples heavily dominated by non-cancerous DNA. This finding demonstrates the enduring value of fundamental statistical principles in solving complex biomedical problems where data limitations challenge current computational paradigms.

The capability to accurately monitor tumor genetics from minimally invasive blood samples could revolutionize personalized oncology care. Currently, detailed tumor profiling necessitates obtaining tissue biopsies, which are invasive, sometimes risky, and infrequently performed throughout the course of treatment. In contrast, liquid biopsies can be administered frequently, offering a dynamic window into tumor biology as it responds and potentially adapts to therapy. Tracking changes in tumor genome composition via blood could inform clinicians if a treatment is diminishing tumor DNA levels or if resistance mechanisms are emerging, facilitating timely adjustments to therapeutic regimens.

Eszter Lakatos, Assistant Professor at Chalmers University of Technology and the University of Gothenburg, emphasizes the clinical implications: “When treatment is effective, the circulating cancer DNA plummets, making it difficult to detect the tumor’s signature in the blood. Our method excels in these challenging scenarios, unmasking tumor signals that would otherwise remain undetectable.” This enhanced sensitivity offers profound opportunities for early detection of relapse and refinement of treatment strategies based closely on evolving tumor profiles.

The BayesCNA method provides an analytical breakthrough by focusing on the sensitive detection of copy number alterations in complex samples, which are critical markers of tumorigenesis and disease progression. By decoding these genomic aberrations from skimmed sequencing data, researchers and clinicians gain access to deeper insights without incurring prohibitive costs or demanding ultra-high sequencing depths. This balance promises practical integration into clinical workflows and broader accessibility across healthcare systems.

Underlying the success of this method is the application of Bayesian inference, a statistical framework that updates the probability for a hypothesis as more evidence or information becomes available. In the context of low-pass genome sequencing, Bayesian strategies allow the algorithm to incorporate prior knowledge about tumor biology and sequencing noise, refining its predictions even amidst uncertainty. This contrasts with many machine learning models that may struggle to incorporate such domain-specific priors or interpret overtly noisy datasets.

Lotta Eriksson, doctoral student and study co-author, reflects on the methodological choice: “Initially, we experimented with various machine learning tools, expecting them to be superior. It was surprising and gratifying that classical statistics delivered more robust and interpretable results. This not only validates our mathematical approach but highlights the importance of matching problem-solving techniques to the nature of biomedical data.”

Looking ahead, the research team aims to expand their analytical framework to discern additional hidden features of tumors that influence patient responses to various treatments. The ultimate goal is to translate these quantitative insights into actionable clinical decision support, enabling oncologists to tailor therapies with unprecedented precision based on real-time tumor genomics.

The promise of integrating BayesCNA into clinical trials is profound. Widespread adoption could accelerate the standardization of blood-based tumor monitoring, reducing dependence on invasive biopsies and improving patient outcomes through personalized care pathways. Frequent sampling and sensitive analysis provide opportunities to preemptively identify therapeutic resistance, adapt dosing strategies, or employ combinatorial treatments to forestall disease progression.

In the context of healthcare economics, the reduced sequencing depth required by BayesCNA translates into lower costs, which is a critical consideration for the scalability of genomic diagnostics. The ability to leverage low-pass whole-genome sequencing without sacrificing analytical power democratizes access to cutting-edge cancer monitoring tools, especially in resource-constrained settings.

This advancement also underscores the interdisciplinary collaboration between mathematical sciences and biomedical research, highlighting how computational innovation can drive substantial improvements in clinical applications. By harnessing statistical expertise and leveraging domain knowledge on tumor biology, the Chalmers and Gothenburg teams have charted a course toward a new paradigm in oncology diagnostics.

Ultimately, BayesCNA represents a transformative leap in liquid biopsy technology’s capability, potentially heralding a future where cancer is continuously monitored with high resolution through routine blood tests. Such progress holds the promise of not only improving survival rates through timely interventions but also enhancing the quality of life for patients by minimizing invasive procedures and personalizing therapeutic strategies.

Subject of Research: People

Article Title: Sensitive detection of copy number alterations in low-pass liquid biopsy sequencing data

News Publication Date: 16-Mar-2026

Web References: https://doi.org/10.1093/bib/bbag111

References:
Eriksson, L., & Lakatos, E. (2026). Sensitive detection of copy number alterations in low-pass liquid biopsy sequencing data. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbag111

Image Credits: Chalmers University of Technology | Marco Nikic

Keywords: Liquid biopsy, circulating tumor DNA, copy number alterations, low-pass whole-genome sequencing, Bayesian statistics, cancer monitoring, tumor evolution, personalized cancer treatment, bioinformatics, statistical modeling

Breast cancer remains one of the most prevalent malignancies worldwide, affecting millions of women annually. While hormonal therapies targeting the estrogen receptor (ER) pathway have significantly improved outcomes for ER-positive breast cancer patients, resistance mechanisms frequently evolve. Giredestrant represents a promising therapeutic agent designed to overcome these limitations by aggressively targeting and degrading the estrogen receptor, thereby inhibiting tumor growth. However, the challenge lies in early identification of responders to tailor treatment optimally and avoid unnecessary toxicity.

The acelERA study meticulously profiles ctDNA extracted from patient plasma combined with detailed molecular analysis of tumor biopsies, enabling a multidimensional view of tumor dynamics in response to giredestrant. Circulating tumor DNA, shed by malignant cells into the bloodstream, offers a minimally invasive, real-time snapshot of tumor genomic alterations. Leveraging ultra-sensitive sequencing technologies, the investigators characterized mutational landscapes and allele frequencies correlating with therapeutic efficacy.

Crucially, the report delineates distinct patterns of ESR1 mutations within the ctDNA that serve as robust predictors of giredestrant treatment response. ESR1 gene aberrations, known drivers of endocrine resistance, were observed to diminish significantly in responders, indicating effective receptor degradation at the molecular level. Conversely, persistence or emergence of certain resistance mutations heralded poor clinical outcomes, underlining the predictive power of ctDNA longitudinal monitoring.

Tumor tissue analyses complemented these findings by revealing differential expression profiles of estrogen receptor isoforms and co-regulatory proteins, establishing a biomarker signature linked with durable response. Notably, the integration of ctDNA mutational data with immunohistochemical quantifications of ER and associated pathways enhanced predictive accuracy beyond traditional clinical parameters alone, spearheading a new era of personalized therapy guidance.

Beyond pure molecular diagnostics, the study delves into mechanistic insights, illustrating how giredestrant induces conformational changes facilitating proteasomal degradation of ER, effectively dismantling estrogen-driven transcriptional programs critical for tumor cell proliferation and survival. This mechanistic validation supports ctDNA and tumor biomarker readouts as reflections of on-target drug activity, thereby providing a rigorous framework to interpret patient responses.

The importance of such biomarkers extends into the clinic, where oncologists frequently grapple with treatment decisions amid heterogeneous patient responses. Access to precise, dynamic biomarkers such as those characterized in acelERA empowers clinicians to stratify patients appropriately, escalating or de-escalating therapy in real time, and potentially circumventing resistance before overt clinical progression.

Moreover, the implications for drug development are profound. Pharmaceutical innovators can harness these biomarkers in adaptive clinical trial designs, enriching study populations with likely responders and accelerating regulatory approval pathways. The synergy between ctDNA and tumor-specific biomarkers exemplifies the evolution of oncology trials into biomarker-driven precision medicine approaches.

As ctDNA assays become increasingly refined and cost-effective, their integration into routine oncology practice is imminent. Combined with advanced computational algorithms analyzing complex mutational and expression data, these biomarkers provide unprecedented insights into tumor heterogeneity and clonal evolution under therapeutic pressure. This dynamic monitoring contrasts starkly with static tissue biopsies, offering longitudinal surveillance that can detect minimal residual disease and early relapse signals.

The acelERA findings also open investigational avenues for combining giredestrant with other targeted therapies. For instance, identifying co-existing pathway activations through biomarker profiling could justify rational combinations designed to thwart compensatory survival mechanisms. Such precision combinations could substantially improve durable remissions and reduce relapse rates among ER-positive breast cancer patients.

On a broader scale, the study exemplifies the power of collaborative, multi-institutional consortia uniting clinical oncology, molecular pathology, and computational biology. The multidisciplinary framework and deployment of cutting-edge next-generation sequencing technologies underpin the robustness and clinical relevance of the results. This integrative scientific model may serve as a template for biomarker discovery in other malignancies.

While these findings herald significant progress, the authors emphasize that larger validation cohorts and extended follow-up are essential to confirm long-term predictive utility and clinical utility. Real-world implementation will also require standardized assay protocols, regulatory harmonization, and clinician education to fully realize the potential of ctDNA and tumor-based biomarkers in managing breast cancer.

In conclusion, the acelERA study marks a paradigm shift in breast cancer therapeutics by establishing ctDNA and tumor molecular profiling as powerful, complementary biomarkers that predict and monitor response to giredestrant. This advancement promises to personalize endocrine therapy, maximize clinical benefit, and ultimately improve survival outcomes for patients battling this common and complex disease. As the oncology field embraces these innovations, the vision of truly precision-guided cancer care moves closer to everyday reality.

Subject of Research:
Biomarkers predicting response to giredestrant in breast cancer using circulating tumor DNA and tumor tissue analyses.

Article Title:
ctDNA and tumor-based biomarkers of giredestrant response in acelERA breast cancer.

Article References:
Collier, A.E., Hilz, S., Chibly, A.M. et al. ctDNA and tumor-based biomarkers of giredestrant response in acelERA breast cancer. Nat Commun (2026). https://doi.org/10.1038/s41467-026-70335-0

Image Credits:
AI Generated

Liquid biopsy, a technique that detects tumor-specific DNA fragments shed into bodily fluids, has emerged as a powerful tool for cancer management. Unlike traditional tissue biopsies, which are invasive and limited to accessible tumor locations, liquid biopsies provide a safer and repeatable means to capture the genetic landscape of tumors. However, the heterogeneity in ctDNA abundance and integrity across different biological fluids remains a major challenge. The research led by Richard, Maetens, Van Baelen, and colleagues addresses this gap by systematically evaluating ctDNA detection rates and representativity in plasma, urine, cerebrospinal fluid, pleural effusion, ascites, saliva, and menstrual fluid obtained from metastatic breast cancer patients.

The study involved meticulous sample collection and advanced sequencing techniques to identify tumor-specific mutations within the ctDNA extracted from the seven liquids. Plasma, traditionally regarded as the gold standard for liquid biopsy, served as the reference against which other body fluids were compared. Remarkably, the findings demonstrated that although plasma remains a highly reliable source of ctDNA, several other fluids also harbor detectable and informative tumor-derived DNA. Pleural effusions and ascitic fluids, often associated with metastatic disease, showed particularly high ctDNA concentrations, reflecting their proximity to tumor sites and potential as complementary sample types for comprehensive molecular profiling.

Urine emerged as a surprisingly informative fluid despite the anatomical distance from primary tumor sites. The data revealed that urine-derived ctDNA captures a portion of the tumor mutational landscape, enabling genetic analysis when blood samples are limited or contraindicated. Cerebrospinal fluid (CSF), critical for patients with brain metastases, offered unique insights into central nervous system tumor heterogeneity and mutations that might evade detection in the bloodstream. Saliva and menstrual fluid showed lower but nonetheless meaningful ctDNA presence, underscoring the diversity of accessible biomarkers across various biological milieus.

One of the pivotal revelations of this research lies in the concept of ctDNA representativeness: how well ctDNA from different body fluids reflects the genetic heterogeneity of metastatic breast cancer lesions. The authors demonstrated that combining analyses from multiple liquid types enhanced the detection of subclonal mutations and minimized sampling bias inherent to single-fluid biopsies. This multidimensional approach may enable oncologists to better monitor tumor evolution, therapeutic resistance mechanisms, and metastatic progression in real time.

The techniques employed in this study included ultra-sensitive next-generation sequencing platforms tailored to detect low-frequency mutations, digital droplet PCR assays for variant validation, and bioinformatics pipelines designed to dissect tumor clonal architectures from ctDNA. By comparing mutational profiles and allele frequencies across fluids within the same patient, the researchers established a robust methodological framework to assess detectability thresholds and biological representativeness.

This study also posed important questions about the biological origins and trafficking pathways of ctDNA. It highlighted that different body fluids likely capture distinct fractions of the tumor burden, shaped by tumor microenvironment, vascularization, and organ-specific dissemination patterns. These biological nuances emphasize the need for fluid-specific preanalytics and analytic optimization to maximize the clinical utility of liquid biopsies in metastatic breast cancer.

Importantly, the research has immediate translational implications. Multifluid liquid biopsy strategies could enhance personalized treatment paradigms by offering more comprehensive molecular portraits without subjecting patients to repeated invasive biopsies. This is particularly valuable in metastatic breast cancer, a highly heterogeneous disease where dynamic genetic information can inform personalized therapeutic decisions, anticipate resistance, and improve prognostication.

Moreover, this multifaceted liquid biopsy approach may accelerate drug development by facilitating biomarker-driven clinical trials. Researchers can longitudinally track tumor genetic shifts using easily accessible fluids, providing real-time feedback on treatment efficacy and guiding adaptive therapeutic interventions. Such precision monitoring has the potential to reduce treatment-related toxicities and improve patient outcomes.

The findings also underscore the promise of integrating liquid biopsy into routine oncological practice. Current clinical guidelines primarily emphasize plasma ctDNA testing; however, this study advocates expansion to include body fluids such as pleural effusions, ascites, and CSF in appropriate clinical contexts. This integration will require harmonization of sample collection, processing protocols, and standardized reporting to ensure reproducibility and clinical validity.

In summary, this extensive analysis of ctDNA across seven body fluids presents a paradigm shift in how tumor genomic information can be sourced non-invasively from metastatic breast cancer patients. It enriches our understanding of tumor biology and ctDNA kinetics while charting a roadmap for refining diagnostic assays that capture the full spectrum of metastatic heterogeneity. It elevates liquid biopsy from a single-fluid diagnostic test to an integrative multi-fluid molecular surveillance platform.

Future research building on these findings will likely explore the longitudinal dynamics of ctDNA in multifluid compartments during treatment, investigate the prognostic and predictive value of multifluid ctDNA signatures, and develop machine learning models that integrate multifluid molecular data for personalized medicine. This study thus provides a foundational resource that inspires innovative clinical protocols and technological advances aimed at conquering metastatic breast cancer.

The implications of this work resonate beyond breast cancer, as similar multifluid ctDNA analyses may transform diagnostics and treatment monitoring for diverse cancers with complex metastatic patterns. By revealing the strengths and limitations of various body fluids as reservoirs of tumor DNA, the research invites a reevaluation of liquid biopsy paradigms and stimulates the oncology field toward more holistic and precise tumor monitoring approaches.

This landmark study represents a critical step toward realizing the full potential of liquid biopsies as a universal, minimally invasive tool to unravel cancer’s genetic complexity, optimize therapeutic decisions, and ultimately improve survival outcomes for patients battling metastatic breast cancer worldwide.

Subject of Research: Detectability and representativeness of circulating tumor DNA (ctDNA) in multiple body fluids from patients with metastatic breast cancer.

Article Title: ctDNA detectability and representativeness in seven body liquids from patients with metastatic breast cancer.

Article References:
Richard, F., Maetens, M., Van Baelen, K. et al. ctDNA detectability and representativeness in seven body liquids from patients with metastatic breast cancer. Nat Commun 16, 10826 (2025). https://doi.org/10.1038/s41467-025-65838-1

Image Credits: AI Generated

DOI: https://doi.org/10.1038/s41467-025-65838-1