Understanding ctDNA Biology May Help Refine Liquid Biopsy Testing and Cancer Monitoring
As liquid biopsy assays become more sensitive, understanding the biology of circulating tumor DNA (ctDNA) and matching testing methods to specific clinical needs will be essential to their use in cancer care, according to Julia V. Burnier, PhD, MBA, Assistant Professor in the Gerald Bronfman Department of Oncology at McGill University in Montreal.
At the 2026 Association for Molecular Pathology (AMP) Annual Meeting & Expo in Seattle, Dr. Burnier discussed how tumor biology, treatment effects, and specimen collection can influence ctDNA measurements. She also presented research in esophageal adenocarcinoma and human papillomavirus (HPV)-associated cancers illustrating different approaches to disease monitoring.
“I think the clinical field has actually moved faster—which is rare—than the biology field,” she said. “So, we're kind of catching up on understanding where these molecules come from, what they mean, and how they can help guide clinical decision-making.”
How Tumor Biology and Treatment Affect ctDNA Detection
Although ctDNA is among the most widely used biomarkers, it represents only the tumor-derived fraction of cell-free DNA (cfDNA) circulating in blood. The amount detected depends on tumor burden, cancer type, metastatic site, and how DNA is released from cells.
Dr. Burnier explained that apoptosis generally produces relatively short DNA fragments, whereas necrosis produces longer, more variable fragments. Cancer cells may also actively release DNA through extracellular vesicles.
In laboratory experiments involving lung and esophageal adenocarcinoma cell lines, chemotherapy increased cfDNA release, with cisplatin-sensitive esophageal cancer cells releasing more DNA than resistant cells. Cisplatin exposure also shifted DNA fragmentation toward larger fragments, suggesting that treatment-related changes in cell death and DNA release may affect liquid biopsy measurements.¹
Although these findings suggest that chemotherapy-induced cell death can influence the amount and fragmentation of DNA released by tumor cells, whether similar changes in circulating cfDNA or ctDNA can reliably distinguish treatment response from disease progression in patients requires further investigation.
Improving Detection in Esophageal Cancer
Dr. Burnier presented preliminary findings from a study examining multimodal ctDNA testing in esophageal adenocarcinoma, where conventional mutation-based assays have had limited success.
Localized tumors often shed little ctDNA, and the disease lacks recurrent hotspot mutations that can be readily targeted. Her team investigated whether combining different molecular signals could improve detection.
The researchers analyzed targeted next-generation sequencing (NGS), digital polymerase chain reaction (PCR), copy number alterations, and fragmentomics, which examines DNA fragmentation patterns. Combining mutation-based, copy-number, and fragmentomic analyses identified cancer-associated molecular signals in 39 of 46 patients (85%). Earlier studies using tumor-uninformed methods had reported detection rates of approximately 30%, although those findings were not from a direct comparison with the newer approach.
Fragmentomics contributed the most unique detections, prompting the investigators to examine whether a simpler, less expensive sequencing method could retain useful performance.
Using shallow whole-genome sequencing (sWGS), they developed a classifier incorporating several cfDNA features. When tested on a separate set of samples, a model combining multiple cfDNA features achieved an area under the receiver operating characteristic curve (AUC) of 0.84 in distinguishing disease-positive from disease-negative samples. A simpler model using only fragment size and DNA end motifs achieved a similar AUC of 0.82, with 76% sensitivity and 80% specificity.
In an exploratory prognostic analysis, the investigators also evaluated whether sWGS-derived features could identify patients at increased risk of death within 6 months of blood collection. The model achieved an AUC of 0.81, with 83% sensitivity and 86% specificity.
These results have not yet been published, and the small study populations warrant further validation. Dr. Burnier noted that the findings nevertheless suggest a less complex approach may offer useful diagnostic and prognostic information in a cancer type that has been difficult to monitor through liquid biopsy.
Monitoring HPV-Associated Head and Neck Cancer
Dr. Burnier next discussed HPV-positive head and neck cancer, which may have a favorable prognosis but whose treatment can result in substantial long-term morbidity, including dysphagia, speech difficulties, and impaired quality of life. Her group investigated whether an inexpensive, HPV-targeted digital PCR assay could help monitor treatment response using blood and saliva.
In a study involving 60 patients with HPV-positive head and neck cancer and 17 HPV-negative controls, HPV ctDNA was detected in 32 of 35 patients (91%) tested before treatment, and 4 of 50 patients (8%) tested after treatment. Blood and saliva testing showed 93% concordance across 87 paired samples.² Among patients with detectable HPV ctDNA after treatment, three of four had residual disease or subsequently experienced recurrence. In contrast, all 46 patients with undetectable posttreatment ctDNA had no evidence of active disease during the reported follow-up.²
According to Dr. Burnier, these findings suggest that HPV ctDNA testing could provide a means to monitor treatment response, with saliva offering a less invasive sampling option.
She also discussed preliminary findings suggesting that HPV ctDNA may be detected approximately 6 to 8 months before recurrence becomes apparent on imaging. In a separate analysis of patients enrolled in the NECTORS trial, longitudinal ctDNA monitoring identified molecular findings associated with recurrence in all four patients who experienced recurrence.3 Dr. Burnier suggested that ctDNA monitoring could eventually contribute to individualized treatment and surveillance strategies, although prospective studies are needed to establish whether these approaches can safely guide treatment de-escalation.
Evaluating Noninvasive Sampling in Cervical Cancer
Dr. Burnier’s group has also investigated HPV DNA–based testing in cervical cancer, examining whether blood, urine, and vaginal samples provide complementary information about cervical abnormalities and invasive disease.
In preliminary analyses, plasma testing was more useful for distinguishing invasive cervical cancer from precancerous disease. Urine and vaginal swabs, meanwhile, showed potential for differentiating high-grade from low-grade cervical abnormalities. Because urine and vaginal samples can be self-collected, these approaches could help address barriers to testing in rural and underserved communities. However, the findings do not establish that the investigational assays can replace recommended cervical cancer screening.
Taken together, the findings illustrate how liquid biopsy approaches may need to be tailored to the biology of individual cancers and their intended clinical applications, from detecting residual disease to monitoring treatment response. Whether these investigational strategies improve patient outcomes will require further prospective evaluation.
“I am not here to say that PCR is better than NGS or that multimodal fragmentomics is better than mutational testing,” Dr. Burnier said. “In fact, I think all of these are tools that we can use, and we just have to design the best test for the clinical question.”
DISCLOSURES
Dr. Burnier reported no potential conflicts of interest.
REFERENCES
1. Bartolomucci A, Kienzle L, Ferrier ST, et al: Cell-free DNA release kinetics and fragmentation reflect treatment response and resistance in vitro. Cells 15:1561, 2026.
2. Ferrier ST, Tsering T, Sadeghi N, et al: Blood and saliva-derived ctDNA is a marker of residual disease after treatment and correlates with recurrence in human papillomavirus-associated head and neck cancer. Cancer Med 12:15777-15787, 2023.
3. Kaorey N, Ferrier ST, Lefebvre C, et al: Longitudinal circulating HPV DNA kinetics predict recurrence with lead time ahead of clinical surveillance in HPV-associated oropharyngeal cancer: The NECTORS trial. 2026 International Federation of Head and Neck Oncologic Societies World Congress. Presented August 28, 2026.