Circulating Tumor DNA After Axi-Cel Stratifies Risk, Predicts Outcomes in LBCL

While many patients with relapsed/refractory large B-cell lymphoma (LBCL) respond to treatment with axicabtagene ciloleucel (axi-cel), only a small subset of these responders experience durable remissions. According to new research findings, circulating tumor DNA (ctDNA) may help clinicians risk stratify and predict outcomes of patients with LBCL who are treated with axi-cel.

The researchers, led by Matthew Frank, MD, PhD, of Stanford University, wrote that these results suggest ctDNA may be a non-invasive biomarker for disease progression following CAR T-cell therapy in clinical trials of LBCL. These findings verify previous research results that suggest ctDNA is an emerging biomarker that can assist in risk stratification and assessment of chemotherapy response in patients with LBCL and other malignancies.

In this study, Dr. Frank and colleagues explored the prognostic capability of ctDNA both prior to and following standard-of-care axi-cel, particularly in relation to the potential biomarker’s ability to predict response and survival outcomes. Patients were ≥18 years old and were diagnosed with either diffused LBCL (n=49), transformed follicular lymphoma (n=17), or primary mediastinal B-cell lymphoma (n=6).

The investigators used next-generation sequencing from the beginning of lymphodepleting chemotherapy until progression or one year following axi-cel infusion to monitor lymphoma-specific variability, diversity, and joining gene segment clonotype ctDNA sequences.

There were 33 patients who had a durable response to axi-cel and 31 who progressed on therapy.

The investigators detected a tumor clonotype in 96% of patients (n=69). There was an association between increased pretreatment ctDNA and post–axi-cel progression as well as the development of cytokine release syndrome or immune effector cell-associated neurotoxicity syndrome.

Approximately 70% of patients with a durable response (n=33) and 13% of patients who progressed (n=4) showed undetectable ctDNA at one week following infusion with axi-cel (p<0.0001). At day 90, all durable responders had undetectable ctDNA.

Additionally, patients with detectable ctDNA concentrations at day 28 had a median PFS of three months, while median PFS for those with an undetectable ctDNA had not been reached (p<0.0001). Additionally, patients with detectable ctDNA concentrations at day 28 versus those without had a median overall survival of 19 months versus not reached, respectively (p=0.008).

Among patients with concurrent undetectable ctDNA, relapse occurred in one out of 10 patients with a radiographic partial response or stable disease by day 28. In contrast, the majority of patients with concurrent detectable ctDNA (n=15) experienced a relapse (p=0.0001). Most patients had ctDNA detected either on or before a recorded radiographic relapse (94%).

As this study only assessed ctDNA concentrations for one year following infusions with axi-cel, the researchers were unable to examine the role of ctDNA monitoring over the longer term. Additionally, the study didn’t measure ctDNA at apheresis, limiting the knowledge regarding the impact of bridging therapies on ctDNA. “Patients were permitted to receive bridging treatment without confirming persistent disease thereafter,” the researchers wrote, “and therefore might have achieved MRD remissions before axi-cel infusion.”

Despite these limitations, the researchers suggest future studies may leverage the approach of ctDNA-based monitoring to identify patients requiring consolidated therapies following CAR T-cell therapy.


Frank MJ, Hossain NM, Bukhari A, et al. Monitoring of circulating tumor DNA improves early relapse detection after axicabtagene ciloleucel infusion in large B-cell lymphoma: Results of a prospective multi-institutional trial [published online ahead of print, 2021 Jun 16]. J Clin Oncol. doi:10.1200/JCO.21.00377