Establishing ctDNA as a validated primary surrogate endpoint

Circulating tumour DNA (ctDNA) has emerged as a promising primary surrogate endpoint in solid tumour oncology, with the potential to accelerate drug development and expand patient access to precision medicines.

However, validation and acceptance will take robust analytical and clinical validation, prospective evidence across tumour types, and coordinated collaboration among industry.

Access to precision medicines hinges on earlier endpoints

Despite substantial advances in precision oncology, access to targeted therapies remains disproportionately concentrated in metastatic disease, with comparatively fewer approvals in early-stage settings where curative intervention may have the greatest impact. This disparity is driven, in part, by conventional clinical development paradigms that rely on late-occurring endpoints such as overall survival and progression-free survival. While clinically meaningful, these endpoints often require prolonged follow-up and large patient populations, creating significant operational and economic barriers to conducting trials in earlier disease stages.

As a result, developers are incentivised to pursue clinical studies in advanced disease settings, where endpoints are reached more rapidly and treatment effects are more readily interpretable. The availability of validated earlier endpoints could enable more efficient evaluation of novel therapies in early-stage disease, thereby supporting timely patient access to potentially transformative treatments.

ctDNA offers a sensitive measure of residual and emerging disease

Analysis of ctDNA through liquid biopsy provides a highly sensitive and minimally invasive approach for detecting molecular residual disease (MRD) and early disease recurrence, often surpassing the capabilities of conventional imaging and tissue-based assessment. Across multiple solid tumour types, ctDNA has been consistently demonstrated to be a powerful prognostic biomarker, with ctDNA positivity associated with a substantially increased risk of relapse and poorer long-term outcomes.^1,2

In several malignancies, including non-small cell lung cancer (NSCLC) and muscle-invasive bladder cancer, ctDNA clearance or changes in ctDNA burden have been associated with improvements in long-term clinical outcomes, supporting its potential role as a surrogate measure of therapeutic efficacy.^1,2These findings have intensified interest in evaluating ctDNA as a primary endpoint capable of enabling more rapid assessment of treatment benefit in clinical trials.

Regulators have a high evidentiary threshold for ctDNA acceptance

Despite growing interest in ctDNA as a biomarker of treatment response, no regulatory authority has yet accepted it as a validated primary surrogate endpoint in solid tumour drug development. Nevertheless, regulatory momentum is increasing.

In 2024, the US Food and Drug Administration (FDA) published guidance outlining considerations for incorporating ctDNA into clinical trials for early-stage solid tumours providing sponsors with a framework for incorporating ctDNA-based endpoints into trials.³ The same year, the FDA’s Oncologic Drugs Advisory Committee (ODAC) unanimously supported molecular residual disease (MRD) negativity as a potential primary endpoint in multiple myeloma.⁴

However, demonstrating that ctDNA can function as a primary surrogate endpoint remains challenging. Regulatory acceptance is likely to be highly context-dependent, reflecting differences in disease biology, therapeutic mechanism, assay performance, and endpoint definition across indications. The complexity of this challenge was highlighted by ODAC's review of a phase III breast cancer study that used ctDNA-guided treatment switching. Although the intervention was associated with substantial improvements in patient-reported outcomes, concerns regarding the absence of convincing evidence linking the strategy to improved survival outcomes underscored the regulatory emphasis on demonstrating clinically meaningful benefit.⁵

For sponsors, the key question is how to integrate ctDNA-derived endpoints in a manner that improves trial efficiency while preserving the interpretability and regulatory acceptability of study results. Addressing this challenge requires a clear understanding of regulatory expectations and a collaborative approach to evidence generation.

A surrogate endpoint must reliably predict treatment benefit

From a regulatory perspective, the consequences of approving a harmful or ineffective therapy based on an inadequate surrogate are generally considered more consequential than delaying approval of a potentially beneficial treatment pending additional evidence. Consequently, biomarkers proposed as primary surrogate endpoints must demonstrate a robust and reproducible relationship with patient outcomes.

This is harder to establish than prognostic value, which predicts the disease’s natural history independent of specific treatment. Although ctDNA has consistently demonstrated strong prognostic value across multiple tumour types, prognostic associations alone are insufficient to establish surrogate validity.⁶

The distinction is clinically important. A treatment could reduce tumour burden and produce an apparent short-term decline in ctDNA levels without improving long-term outcomes. In such a circumstance, reliance on a poorly defined ctDNA endpoint could result in incorrect conclusions regarding therapeutic benefit. More stringent endpoints, such as complete ctDNA clearance, may provide stronger evidence of treatment effect, but remain dependent on assay sensitivity and analytical performance. Accordingly, endpoint definition and assay characteristics are inseparable components of surrogate validation.

In settings where mature clinical outcome data is not yet available, ctDNA-based endpoints may support accelerated approval pathways if they are considered "reasonably likely" to predict clinical benefit. However, such approvals generally require subsequent confirmation of treatment benefit in post-marketing studies. For some development programmes, the logistical and financial burden associated with confirmatory trials may limit the practical advantages of pursuing a surrogate-based regulatory strategy.⁶

Endpoint definitions must be data-informed and consensus-driven

The first ctDNA-based approval in a solid tumour setting will likely set an important precedent, so, endpoint definitions will need to be biologically meaningful, analytically robust, and clinically reproducible. This is particularly challenging because ctDNA dynamics vary with assay design, analytical sensitivity, sampling strategy, tumour biology, and disease burden. As a result, endpoint definitions developed in one setting may not generalise to others.

Establishing ctDNA as a primary surrogate endpoint will therefore require consensus on both assay performance standards and disease-specific endpoint definitions. Multi-stakeholder collaborations that integrate data across trials, platforms, and patient populations will be critical for defining clinically relevant thresholds and response criteria.

Payer acceptance requires stronger evidence

Regulatory approval does not guarantee reimbursement. Payers often require stronger evidence than regulators, placing greater emphasis on demonstrated improvements in patient outcomes and comparative effectiveness.⁶ As a result, surrogate endpoints may face scepticism when not supported by mature clinical outcome data.

For sponsors, this highlights the importance of building a robust evidence base. Incorporating ctDNA as an exploratory or secondary endpoint using standardised methodologies can help establish reproducible associations between ctDNA dynamics and clinical outcomes across studies and populations. Such evidence will be essential for both regulatory and payer acceptance.

Advancing validation through collaboration

Validation of ctDNA as a primary surrogate endpoint will require coordinated efforts across sponsors, regulators, diagnostic developers, and academic researchers. This need is particularly acute in settings where conventional endpoints are impractical, such as early-stage cancers with long recurrence timelines.

Collaborative initiatives are already helping to build the necessary evidence base. For example, the Friends of Cancer Research ctMoniTR Project has proposed standardised ctDNA endpoint approaches and demonstrated associations between ctDNA clearance and favourable outcomes in NSCLC.¹ Similarly, international efforts have proposed liquid biopsy response criteria (LB-RECIST) to complement established radiographic assessment frameworks.⁷

Ultimately, progress toward ctDNA-based surrogate endpoints will depend on the generation of standardised, cross-study evidence. Development teams should view ctDNA strategy both through the lens of individual trials, and as part of a broader effort to establish a reliable and widely accepted framework for molecular endpoints in oncology.

References

About the author

Dr Bea Mann, PhD, is senior director and oncology therapeutic expert at ICON. As the head of the Oncology Drug Development Services group, Dr Mann is responsible for overseeing the scientific and medical oncology clinical development strategies globally. She brings to ICON more than 16 years of pharmaceutical industry experience (Medical Affairs) having worked as a medical & scientific advisor in the field of oncology and haematology at Roche and GSK. She has extensive experience across all phases of development from pre-clinical to commercialisation. Dr Mann joined ICON in 2016 and in her current role has an expert understanding of the oncology research environment and a thorough understanding of the rapidly changing guidelines for the development of anticancer drugs from study design to protocol development. She has a PhD in Medicinal Chemistry & Drug Development from Nottingham University, UK and was a major contributor to the identification, development, and commercialisation of two linker molecules as part of her oncology research.

About ICON

ICON plc is a world-leading clinical research organization. Offering deep operational and medical expertise we accelerate innovation, driving emerging therapies forward to improve patient outcomes. From molecule to medicine, we deliver integrated consulting, clinical development, commercialization and post-marketing solutions to pharmaceutical, biotechnology, medical device, government and public health organizations worldwide. With headquarters in Dublin, Ireland, ICON employed approximately 40,350 employees in 97 locations in 55 countries as at March 31, 2026. For further information about ICON, visit: www.iconplc.com.