The emerging clinical utility of ctDNA to guide adjuvant treatment in cancer

Circulating tumor DNA (ctDNA) testing has the potential to dramatically impact our approach to cancer management. This minimally invasive test, with potential for repeated testing, can track dynamic changes in tumour volume and molecular profile. Studies in early-stage colorectal cancer have confirmed utility in guiding adjuvant therapy decisions and studies across multiple tumour types, most notably lung cancer, have demonstrated utility in molecular characterisation.1, 2, 3 Along with confirming the clinical utility through carefully conducted prospective studies, in parallel there needs to be an ongoing focus on optimising testing methodology and confirming cost effectiveness, both being essential to ensuring broad uptake and maximal gains for patients that are affordable to the health system.

The most robust data supporting a role for ctDNA in the adjuvant setting has emerged from studies in colorectal cancer.^1,2 Early observational studies being followed by the first ever randomised study of ctDNA informed management, the DYNAMIC study, with recently published mature data confirming that ctDNA testing versus standard of care reduced the use of adjuvant chemotherapy by almost half, without compromising risk of recurrence.¹ Also demonstrated were the prognostic significance of initial ctDNA burden and the adverse prognosis associated with persistence of detectable ctDNA despite chemotherapy. The PEGASUS trial,⁴ has demonstrated the potential for early intervention in these patients, reporting that treatment with FOLFIRI was associated with ctDNA clearance in 11 of 23 patients, none of whom had relapsed early in follow-up. Notably, based on this emerging data, the FDA has recently published guidance supporting the use of ctDNA as a molecular residual disease (MRD) marker, to enrich for high-risk populations for novel drug development as well as using ctDNA dynamics or clearance in early phase trials to aid signal finding of drug activity.⁵

The utility of ctDNA to molecularly characterise more advanced cancer cases, particularly where obtaining a tumor sample can be challenging, such as lung cancer.³ While tumor tissue based testing remains the gold standard, recent data suggests that testing both tumour tissue and peripheral blood could be advantageous, even where tumour samples are available.⁶ In the FIRE-4 study that enrolled previously untreated patients with RAS wild-type metastatic colorectal cancer based on tissue testing, 13% were found to be have a RAS mutation (mt) in the liquid biopsy, with outcomes consistent with patients who were RAS mt on tissue testing.⁷ Repeat ctDNA testing over time can also identify dynamic changes, including emerging resistance mutations, which can meaningfully inform further therapy such as has been demonstrated with later line RAS testing in colorectal cancer patients undergoing an epidermal growth factor receptor (EGFR) inhibitor rechallenge, where RAS mutation detection defines a treatment resistant population.⁸

Along with the importance of further prospective studies, further refinement of testing methodology could further enhance clinical validity. Broadly, two types of ctDNA testing are emerging, tumor informed (where molecular testing on a patient tissue sample identifies the targets for plasma testing for that individual patient) and tissue agnostic (tailored to the mutational profile of the cancer type tested without tissue testing of individual patient tumor).⁹ Tumor informed assays have greater sensitivity, but come with the financial cost and time delays associated with sequencing of tumour tissue samples. Within the categories of tumour informed and agnostic testing there is a proliferation of laboratories offering testing, with varying methodologies and varying amounts of supporting clinical data. As such, clinicians should carefully examine the evidence base for any ctDNA assay that they are planning to use in the real-world setting.

There are other challenges to the routine rollout of ctDNA testing in solid cancers, with demonstrating cost-effectiveness in the community setting an important goal. In the adjuvant setting there are calculable costs savings if ctDNA negative patients that would have received treatment based on standard clinicopathologic criteria can avoid such treatment, and the associated healthcare costs, without compromising their recurrence risk. In the advanced setting avoiding treatments unlikely to be effective (such as the use of an EGFR inhibitor rechallenge in a patient with liquid RAS mt status) could also drive savings. A recent study of US practice found that the application of ctDNA testing in stage II colon cancer was cost effective, modelling on the robust trial data generated from the DYNAMIC study.¹⁰ Further studies will need to assess cost-effectiveness in the context of long-term clinical outcomes and ensure that any gains in quality-adjusted life years are achieved at an affordable cost.

The DYNAMIC study is the first to report long term data from a prospective study of ctDNA informed management in early-stage colorectal cancer, with many similar studies expected to report in the next few years. Further studies are exploring the utility of integrating ctDNA testing into adjuvant therapy studies. The value of this approach has already been demonstrated in the IMvigor010 adjuvant urothelial cancer study, which was negative overall but did demonstrate a survival benefit in the (high-risk) patients with detectable ctDNA.¹¹ Similarly serial ctDNA testing is being integrated in studies in advanced disease settings, which will expand the evidence base for testing in this context.

It is worth noting that ctDNA faces several challenges in cancer medicine despite its promising potential as a non-invasive biomarker. One significant hurdle is the low abundance of ctDNA in the bloodstream, which can make detection and quantification difficult, particularly in early-stage cancers, with certain types of genomic abnormalities such as fusion and amplification, and particular sites of metastasis such as the lung and peritoneum. Additionally, the heterogeneity of tumours means that ctDNA may not consistently reflect the genetic makeup of all tumour cells, potentially leading to incomplete or misleading information. Standardization of assay testing protocols is another critical issue, given variations in methodology can impact the reproducibility of results. Finally, the high costs and technical complexity of ctDNA assays may restrict their widespread adoption in routine clinical practice.

Contributors

ES, PG and JT drafted the original manuscript, reviewed and edited the final version.

Declaration of interests

The authors have no conflict to disclose.