ctDNA-Based MRD Provides Prognostic Value for First-Line Treatment in DLBCL

ctDNA-Based MRD in DLBCL |

Image Credit: © v3 - Brighting

Collection - stock.adobe.com

Circulating tumor DNA (ctDNA)–based minimal residual disease (MRD; ctDNA-MRD) as assessed by phased variant enrichment and detection sequencing (PhasED-Seq) was prognostic of survival outcomes in patients with diffuse large B-cell lymphoma (DLBCL) following curative-intent frontline treatment, according to findings from a prospective study presented during the 2025 ASCO Annual Meeting.

Patients who were ctDNA-MRD negative at the end of treatment (n = 126) experienced a 36-month progression-free survival (PFS) rate of 85% compared with 15% among patients who were positive for ctDNA-MRD at the same time point (n = 34; HR, 11.03; 95% CI, 6.27-19.40; P < .0001). The 36-month overall survival rates were 92% and 41%, respectively (HR, 7.38; 95% CI, 3.72-14.62; P < .0001).

“We prospectively validated the prognostic value of ctDNA-MRD in [patients with DLBCL treated in the frontline setting],” Steven Wang, MD, said during the presentation. “We [also] demonstrated that ctDNA-MRD can detect residual disease independent of [a] PET scan.”

Wang is a visiting fellow in the Department of Hematology at Cancer Center Amsterdam, Amsterdam UMC, Vrije Universiteit, in The Netherlands.

Examining the Prospective Study Design

The study was comprised of a real-world cohort of patients with newly diagnosed DLBCL from over 50 centers in Belgium and The Netherlands. All patients received frontline treatment with R-CHOP (rituximab [Rituxan], cyclophosphamide, doxorubicin, vincristine, and prednisone) or DA-EPOCH-R (dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin, and rituximab), with curative intent.

The study included patients with adequate baseline, normal, and end of treatment samples. Those who received consolidative therapy following end of treatment blood sampling were excluded.

ctDNA-MRD status was assessed by PhasED-Seq, which identified patient-specific phased variants in baseline samples. Most patients (58%) had formalin-fixed paraffin-embedded (FFPE) samples and the remaining 42% had pretreatment plasma samples.

“PhasED-Seq is a baseline-informed, fixed-panel strategy with normal control sequencing to identify phased variants that are multiple mutations found on the same ctDNA fragment,” Wang commented. “At baseline, phased variants are identified in FFPE material or pretreatment plasma and then tracked in end of treatment plasma samples. Analytical validation shows the limit of detection to 95% at 0.7 parts per million. However, the ultimate sensitivity is also influenced by the absolute number of DNA molecules present in the plasma.”

The median age in the overall cohort (n = 160) was 67 years (range, 18-88). Most patients were male (64%), had DLBCL (89%), had stage III or IV disease (79%), had an International Prognostic Index (IPI) score of 0 to 2 (52%), received R-CHOP in the frontline setting (91%), and received 6 cycles of treatment (90%).

Further Findings

Additional data from the study showed that end of treatment ctDNA-MRD was highly sensitive in terms of predicting early disease relapse. Eighty-six percent of patients who experienced disease relapse within 0 to 6 months following the end of treatment were ctDNA-MRD positive. Among patients who relapsed between 0 and 12 months or greater than 12 months, the proportions of patients who were positive for ctDNA-MRD were 80% and 22%, respectively. The study authors noted that continued longitudinal monitoring is likely required to capture late relapses.

In patients with a complete molecular response (CMR) by PET scan, the 3-year PFS rates were 89% and 36% among those who were ctDNA-MRD negative (n = 106) or positive (n = 11) at the end of treatment, respectively (HR, 6.58; 95% CI, 2.69-16.11; P < .0001). Patients who did not achieve a CMR by PET scan displayed 3-year PFS rates of 64% and 4% in the ctDNA-MRD–negative (n = 20) and –positive (n = 23) groups, respectively (HR, 6.78; 95% CI, 2.80-16.45; P < .0001). The study authors noted that 2 of the 7 PFS events in the ctDNA-MRD–negative group were due to death unrelated to lymphoma.

Higher Ann Arbor stage disease was found to be correlated with MRD positivity; patients with stage III or IV disease were significantly more likely to be MRD positive compared with those with stage I or II disease (P < .05). Being categorized into a higher IPI risk group was also associated with increased MRD positivity.

Findings from a subgroup analysis revealed that end of treatment ctDNA-MRD was more prognostic for PFS compared with IPI score or end of treatment PET-CT scan. Patients with an IPI score of 0 to 2 experienced a higher 36-month PFS rate vs those with a score of 3 to 5 (HR, 1.61; 95% CI, 0.93-2.79; P = .086). Those with a negative PET-CT scan had a higher 36-month PFS rate vs patients with a positive scan (HR, 5.31; 95% CI, 2.87-9.82; P < .0001).

“[Findings from] our study support the integration of ctDNA-MRD by PhasED-Seq as a standard component of response evaluation in [patients with DLBCL treated in the frontline setting],” Wang said in his conclusion.

Disclosures: Wang reported no relevant relationships.

Reference

Wang S, Nijland M, Strobbe L, et al. Prospective validation of end of treatment ctDNA-MRD by PhasED-Seq in DLBCL patients from a national trial. J Clin Oncol. 2025;43(suppl 16):7000. doi:10.1200/JCO.2025.43.16_suppl.7000