Applications of Circulating Tumor DNA Liquid Biopsy Continue to Expand in Lymphomas

Get Permission

Liquid biopsies using circulating tumor DNA (ctDNA) have the potential to personalize medicine for patients with lymphoma, going beyond traditional markers and risk factors to provide dynamic assessments over time. Expanded applications of ctDNA liquid biopsy beyond diagnosis include early response to treatment, early detection of relapse, and identification of tumor evolution and mutations underlying treatment resistance, according to David Kurtz, MD, of Stanford University School of Medicine, Palo Alto, California.

“Standardizing ctDNA measurements across multiple centers and labs remains a critical and ongoing challenge.”
— David Kurtz, MD

Tweet this quote

“Multiple features of ctDNA can be used to guide therapy by quantitative assessment of disease burden, risk estimation, and qualitative identification of driver mutations and mechanisms of resistance,” Dr. Kurtz explained at the 2020 American Association for Cancer Research (AACR) Virtual Meeting: Advances in Malignant Lymphoma.1

The major hurdle for clinical application of this methodology is standardization of the technique across institutions. “Standardizing ctDNA measurements across multiple centers and labs remains a critical and ongoing challenge,” Dr. Kurtz stated.

Potential Roles of ctDNA

Liquid biopsies first gained attention as an alternative to traditional invasive tumor biopsies, as they involve obtaining plasma samples rather than invasive surgical removal. “Liquid biopsies are emerging biomarkers in lymphomas,” Dr. Kurtz explained. “There are two types of liquid biopsies: circulating tumor cells and ctDNA. The latter has received the most attention in lymphoma.”

Dr. Kurtz discussed the potential roles of ctDNA: mutational genotyping, detection of molecular disease, identification of residual disease, and early detection of relapse (attention to refractory clones that drive resistance).

ctDNA can be measured by polymerase chain reaction–based assays and sequencing-based assays. Dr. Kurtz and colleagues at Stanford have developed a sequence-based ctDNA assay called CAPP-Seq for B-cell non-Hodgkin lymphoma using targeted next-generation sequencing.

“We designed a panel of genes for diffuse large B-cell lymphoma [DLBCL] and sequenced it. We found cancer mutations in tissue and plasma along with matched germline mutations to define a patient set for malignancy. Then, we used these personalized mutations that were defined pretreatment to track patients over time for response to treatment and early detection of disease progression,” Dr. Kurtz noted.

“For example, one residual mutation in plasma predicted disease progression in a patient even after radiographic remission,” he said. “This technique allows you to track multiple mutations to increase sensitivity for disease detection and identify what drives a patient’s malignancy.”

ctDNA and Personalized Therapy

There aretwo major roles for ctDNA as personalized therapy for lymphoma:

  • Quantitative measurements to assess risk
  • Identification of drug targets and mutational mechanisms of resistance.

CAPP-Seq was studied in a discovery set of 14 patients with lymphoma to quantify disease and determine whether there was early and major molecular response.2 “We found that a simple threshold of 2 logs [decrease] could stratify responders from nonresponders,” shared Dr. Kurtz. “Given this initial signal, we further explored the dynamics of ctDNA to define features of response.”

“This technique allows you to track multiple mutations to increase sensitivity for disease detection and identify what drives a patient’s malignancy.”
— David Kurtz, MD

Tweet this quote

The validation study included more than 200 patients from 6 international centers. The investigators found that a 2-log drop after one cycle of therapy could stratify patients who achieved complete response from those with no complete response. “We termed this 100-fold drop after one cycle of therapy ‘early molecular response,’ and this was prognostic for complete response and other outcomes in a training and validation set,” Dr. Kurtz told listeners.

Next, they looked at ctDNA changes after two cycles of therapy. They identified a 2.5-log drop as being most prognostic for major molecular response and event-free survival.

Serial Assessment of Changes Over Time

“Although these initial changes were encouraging, they ignored one of the major advantages of ctDNA liquid biopsies—to assess changes in ctDNA serially over time,” Dr. Kurtz stated.

Each of the available prognostic tools in lymphoma offers a fixed assessment at a particular point in time. They include the International Prognostic Index, an interim response determined by positron-emission tomography scan, and molecular cell of origin.

“These assessments at a fixed point in time largely do not integrate multiple variables. They are used to divide patients into risk groups rather than to identify a personalized prediction,” Dr. Kurtz explained. “The advantage of ctDNA as a prognostic tool includes the ability to integrate multiple variables over time.”

To take advantage of these potential applications of ctDNA, Dr. Kurtz and colleagues have constructed a new model to tract ctDNA over time called the Continuous Individualized Risk Index (CIRI).3 The CIRI includes various predictors learned at key time points that are integrated in a Bayesian framework.

“Instead of trying to identify risk status prior to therapy, CIRI attempts to provide an assessment over time using the individual’s course of disease,” said Dr. Kurtz. “For example, you may learn risk factors before therapy, but other factors are learned during therapy (ie, interim risk factors). CIRI attempts to update this over time.”

In a cohort of patients, the CIRI outperformed each separate predictor of event-free survival. These patients were stratified by risk as favorable, intermediate, or high.3

Drivers and Mechanisms of Resistance

In addition to serial (dynamic) monitoring during therapy, another important use of ctDNA is identification of drivers and mechanisms of resistance. By tracking multiple genetic loci, CAPP-Seq may provide additional genotype information, not only mutations in individual genes. This tool can identify molecular subtypes and classify individual lymphomas.

“Not only does CAPP-Seq allow you to obtain information on individual genes, but this method can [identify] germinal center B-cell [GCB] subtypes and non-GCB subtypes in DLBCL, which are traditionally identified by gene expression from tumor samples,” Dr. Kurtz continued. “GCB tumors are associated with better progression-free survival compared with non-GCB tumors,” he reminded the virtual audience.


  • Liquid biopsies using ctDNA have a number of potential expanded applications beyond diagnosis for malignant lymphomas and other cancers.
  • Tools have been developed to utilize ctDNA for serial monitoring during lymphoma therapy and for identifying mutations that drive resistance.
  • The use of liquid biopsy using ctDNA needs standardization before it can be adapted routinely to the clinic.

“Mutational genotyping may also lend insight into resistant subclones,” Dr. Kurtz continued. “Both target and genome-wide assessments can reveal alterations in specific mutations that may drive resistance. In one example, we observed emergence of a MYC translation and BCL2 amplifcation after only 4 days of treatment. This gives you a flavor of why we care about this type of genotypic information.”

“More recently, we applied ctDNA monitoring in a patient with lymphoma undergoing chimeric antigen receptor T-cell therapy. Perhaps not unsurprising, patients with residual ctDNA had significantly worse progression-free survival than those without it. This is quantitative information. To look into the biology of the disease, we looked at specific mutations in patients whose disease had progressed and in those whose disease had not,” Dr. Kurtz noted.

At the point in time of relapse, a number of emergent mutations were identified. In one patient, the researchers identified the emergence of a CD19 stop mutation; in another patient, there were not only on-target mutations, but also emergent immunologic mutations, which may potentiate resistance. 

DISCLOSURE: Dr. Kurtz has served as a consultant or advisor to Roche Molecular Diagnostics; has ownership equity in Foresight Diagnostics; and has patents pending related to methods for analysis of cell-free nucleic acids and methods for treatment selection based on statistical frameworks of clinical outcome.


1. Kurtz D: Approaches for personalized medicine in lymphoma through liquid biopsies. 2020 AACR Virtual Meeting: Advances in Malignant Lymphoma. Presented August 19, 2020.

2. Kurtz DM, Scherer F, Jin MC, et al: Circulating tumor DNA measurements as early outcome predictors in diffuse large B-cell lymphoma. J Clin Oncol 36:2845-2853, 2018.

3. Kurtz DM, Esfahani MS, Scherer F, et al: Dynamic risk profiling using serial tumor biomarkers for personalized outcome prediction. Cell 178:699-713, 2019.

Related Articles

Expert Point of View: David Scott, MBChB, PhD

“In 2015, papers showed that ctDNA could suggest relapse noninvasively in serial plasma samples. In those studies, ctDNA was positive prior to relapse detected by computed tomography scans,” explained session moderator David Scott, MBChB, PhD, of the Center for Lymphoid Cancer, BC Cancer,...