Category: 3. Business

The presence of circulating tumor DNA (ctDNA) at end of treatment (EOT) was shown to be prognostic of event-free survival (EFS) outcomes for patients with lymphoma regardless of subtype, according to findings from a retrospective, real-world analysis presented at the 2025 ASH Annual Meeting and Exposition.¹

Findings showed that the median EFS was not achieved (NA) in the ctDNA minimal residual disease (MRD)–negative group (n = 49) vs 1.97 months in the ctDNA-MRD-positive group (n = 19; adjusted HR, 22.43; 95% CI, 6.76-74.45; P < .0001). The 12- and 24-month HRs in the ctDNA-MRD-negative population were 0.83 (95% CI, 0.71-0.98) and 0.79 (95% CI, 0.64-0.96), respectively. The respective HRs in the ctDNA-MRD-positive population were 0.05 (95% CI, 0.01-0.35) and 0.00 (95% CI, NA-NA), respectively.

“EOT ctDNA status can clarify ambiguous imaging results and enables earlier relapse detection,” Natalie Galanina, MD, lead study author and clinician investigator at UPMC Hillman Cancer Center in Pittsburgh, Pennsylvania, stated in during the presentation. “ctDNA kinetics offer real-time insights into treatment response during first-line therapy and can further predict response to CAR T-cell treatment,” she added.

What served as the foundation for this study?

Personalized, tumor-informed ctDNA assays have shown the ability to capture prognostic and predictive information in diffuse large B-cell lymphoma (DLBCL), but its prognostic capability has been understudied in diverse subsets of lymphoma.²

To bridge that gap, investigators prospectively collected real-world data of MRD detection and ctDNA clearance kinetics in patients with newly diagnosed or relapsed/refractory lymphoma across 14 subtypes.¹

“[Signatera is a] personalized tumor-informed assay, where both the tumor and a source of matched normal [tissue] is sequenced, either by whole exome or whole-genome sequencing. Based on the somatic mutation profile of a patient, a custom patient-specific assay is designed to track ctDNA in the plasma. This makes the test ultrasensitive while maintaining extremely high specificity,” Galanina said.

How was the trial designed to answer how ctDNA can be applied in various lymphomas?

The schema was such that 1105 prospectively collected plasma samples from 144 patients with lymphoma were subject to ctDNA assessment. Samples included aggressive (n = 123) T-cell (n = 13) and B-cell (n = 110) lymphomas, as well as indolent (n = 21) follicular (n = 10), marginal zone (n = 5), and cutaneous T-cell lymphoma (n = 6).

The demographics of the patient cohort were representative of the real-world population, Galanina said. The median age was 61 years (range, 18-84) and most patients were male (n = 77; 53%). Most patients had stage IV disease (n = 75; 56%), although those with stage I (n = 15; 11%), II (n = 29; 21%), and III (n = 16; 12%) were also included. ECOG performance status was predominantly 0 (n = 50; 46%), followed by 1 (n = 35; 32%), 2 (n = 16; 15%), 3 (n = 6; 5.6%), and 4 (n = 1; 0.9%). Revised International Prognostic Index score fell between 0 and 2 in 25.2% (n = 31) of patients and between 3 and 5 in 26.8% (n = 33) of patients; 80 scores were not reported.

With respect to therapy, patients reported receiving Pola-R-CHP (n = 6; 4.3%), R-CHOP (n = 72; 51%), R-EPOCH (n = 14; 10%), other rituximab (Rituxan; n = 17; 12%), and other (n = 31; 22%).

The median number of ctDNA MRD timepoints was 7 (range, 1-32). Median follow-up for EFS and overall survival (OS) was 20 (range, 1-108) and 21 (range, 1-108) months, respectively.

Pretreatment ctDNA was detectable in 94% of patients with lymphoma. “The median number of tumor molecules per mL was about 100 in aggressive and about 20 in indolent lymphomas, respectively, which may reflect differences in circulating tumor burden,” Galanina stated.

What else was reported on with respect to ctDNA’s validity as a prognostic tool?

Additional data revealed that ctDNA provided a better indication for treatment response than traditional imaging. Patients who had a negative PET-CT at EOT (n = 35) experienced a median EFS that was NA vs 5.16 months in those whose PET-CT was positive at EOT (n = 25; adjusted HR, 8.68; 95% CI, 2.41-31.29; P = .0010). Conversely, patients who had negative ctDNA-MRD at EOT (n = 44) experienced a median EFS that was NA vs 2.04 months in those who had positive ctDNA-MRD at EOT (n = 16; adjusted HR, 49.77; 95% CI, 9.91-250.02; P < .0001).

Furthermore, PET-negative patients with negative ctDNA-MRD (n = 32) had a median EFS that was NA vs 2.76 months in those with positive ctDNA-MRD (n = 3; HR, 45.29; 95% CI, 4.63-443.27; P = .0011). PET-positive patients with negative ctDNA-MRD (n = 12) had a median EFS that was NA vs 1.97 months in those with positive ctDNA-MRD (n = 13; HR, 12.26; 95% CI, 3.23-46.59; P = .0002).

“The clinical significance of this result is that EOD PET-positive patients present a significant clinical challenge, and most of them ultimately proceed to receive additional therapy. However, although the patient numbers are small, our data clearly show that the majority or 75% of PET-positive MRD-negative patients do not progress, and therefore may not require any additional therapy,” Galanina said. “Based on this, it is reasonable to integrate ctDNA as an adjunct to EOT assessment to help further risk stratify patients who are likely to relapse vs those who remain disease free.”

“For patients who are EOD PET negative, if they are ctDNA positive, this may inform post treatment surveillance, as these patients may need to be monitored more closely, and patients who are PET positive, if they’re MRD negative, may be appropriate candidates for observation only, or at the very least require pathologic confirmation of the PET-positive lesions to rule out the non-etiology of the FDG uptake, as those patients have generally good outcomes,” Galanina added.

In multi-variable analysis investigators demonstrated that ctDNA was the most significant predictor of EFS after correcting for all other factors, including age, tumor histology, and stage (P < .001).

ctDNA clearance during frontline therapy was also shown to be prognostic of outcomes in all lymphoma subtypes. The median EFS was NA in patients who cleared their ctDNA (n = 48) vs 2.05 months in those who did not (n = 12; adjusted HR, 8.57; 95% CI, 2.55-28.81; P = .0005). Moreover, the median EFS was NA, NA, and 2.05 months in patients with cycle 1 clearance (n = 14), delayed clearance (n = 34), and no clearance, respectively. The adjusted HRs for patients without clearance vs those with cycle 1 clearance and delayed clearance were 20.95 (95% CI, 2.09-21.11; P = .0097) and 7.45 (95% CI, 2.22-24.98; P = .0011), respectively.

“Early clearance by cycle 1 may have significant implications for potential de-escalation of therapy, especially for older patients and those with comorbidities,” Galanina explained.

Can ctDNA be used to predict response to CAR T-cell therapy?

“Lastly, we also looked at ctDNA clearance in patients undergoing CAR T-cell therapy [and we found that] ctDNA clearance retains its predictive value in this setting as well. Most MRD-positive patients who cleared ctDNA within 3 months post CAR T attained durable remission at 1 year. There was one patient who became ctDNA negative 1 month post CAR T, but then relapsed more than 1 year after, and this patient turned positive prior to recurrence. This suggests that single time point measurement post CAR T may not be sufficient, and longitudinal testing at various time intervals for several years of follow-up may be more optimal to detect patients who may relapse,” Galanina stated.

“MRD assessment supports the integration of ctDNA testing into routine clinical management and surveillance to personalize lymphoma care,” Galanina said in conclusion.

Disclosures: Galanina had no financial relationships to disclose.

References

1. Galanina N, Iqbal M, Nousome D, et al. Real-world evaluation of ctdna for risk stratification across the spectrum of both aggressive and indolent lymphomas. Blood. 2025;146(suppl 1):281. doi:10.1182/blood-2025-281
2. Narkhede M, Tomassetti S, Iqbal M, et al. Tumor-informed ctDNA assessment as a valuable prognostic and predictive biomarker in diffuse large B-cell lymphoma. Front Oncol. Published online July 29, 2024. doi:10.3389/fonc.2024.1407003

Carfilzomib (Kyprolis) plus lenalidomide Revlimid) and dexamethasone (KRd) delivered improvements in progression-free survival (PFS) compared with bortezomib (Velcade) plus lenalidomide and dexamethasone (VRd) in patients with newly diagnosed multiple myeloma, according to interim results from the phase 3 COBRA trial (NCT03729804).

The findings, presented during the 2025 ASH Annual Meeting, demonstrated that KRd (n = 126) reduced the risk of progression or death by 43% vs VRd (n = 124) in the intent-to-treat population (HR, 0.57; 95% CI, 0.37-0.88; P = .0095) at a median follow-up of 53 months. The median PFS was not reached (NR) in the KRd group vs 48.8 months in the VRd arm.

The data demonstrated consistent directional benefits across cytogenetic subgroups. Among patients with standard-risk disease, KRd (n = 97) achieved a statistically significant improvement in PFS vs VRd (n = 96; HR, 0.59, 95% CI, 0.36-0.98; P = .04); the median PFS was NR in the KRd group compared with 48.8 months for VRd. In this subgroup, 27% of patients treated with KRd experienced progression or death vs 40% with VRd. In the high-risk cohort, KRd (n = 29) again demonstrated favorable outcomes vs VRd (n = 28), with 31% vs 48% of patients experiencing progression or death. The median PFS was NR vs 34.9 months, respectively (HR, 0.52; 95% CI, 0.22-1.22; P = .12).

“COBRA showed superior efficacy of KRd vs VRd in newly diagnosed multiple myeloma, achieving both co-primary endpoints of MRD-negative CR at 12 months and PFS,” noted presenting author Dominik Dytfeld, MD, PhD. “The PFS benefit of KRd was observed regardless of cytogenetic risk, and KRd produced deeper responses, with higher rates of complete response [CR] and minimal residual disease [MRD]–negativity. KRd [also] demonstrated anticipated toxicity profiles with higher rates of neutropenia and cardiac [adverse effects (AEs)], but less neuropathy. COBRA results support further evaluation of KRd-based induction regimens in newly diagnosed multiple myeloma.”

Dytfeld is an associate professor of medicine at the Poznan University of Medical Sciences in Poland and founder/chief executive officer of the Polish Myeloma Consortium

What was the design of the COBRA trial?

The COBRA trial was a multicenter, randomized, open-label phase 3 study designed to compare KRd with VRd in patients with newly diagnosed multiple myeloma who had an International Myeloma Working Group (IMWG) Frailty Score of less than 2.^1,2 Randomization was stratified by cytogenetic risk category (standard vs high risk defined by t(4;14) or del(17p)) and by history of venous thromboembolism (yes vs no). A total of 250 patients were randomly assigned in a 1:1 ratio to the KRd or VRd treatment arms.¹

Patients in the KRd arm received up to 24, 28-day cycles of therapy consisting of carfilzomib at 56 mg/m² administered on days 1, 8, and 15 (with twice-weekly dosing during cycles 1 and 2), lenalidomide at 25 mg on days 1 through 21, and dexamethasone at 40 mg weekly (or 20 mg for patients 75 years of age or older). Stem-cell collection occurred after cycle 4 in this transplant-deferred design. Following completion of induction, patients continued on lenalidomide 5 mg daily as maintenance until disease progression.

Patients in the VRd arm received eight 28-day cycles of bortezomib at 1.3 mg/m² on days 1, 4, 8, and 11; lenalidomide at 25 mg on days 1 through 14; and dexamethasone at 20 mg on days 1, 2, 4, 5, 8, 9, 11, and 12. After completion of the VRd induction phase, patients transitioned to consolidation with lenalidomide and dexamethasone for eighteen 28-day cycles, followed by lenalidomide 5 mg maintenance until progression.

The trial’s co-primary end points were the rate of MRD-negative CR rate at a sensitivity of 10⁻⁵ at 12 months and PFS. Secondary end points include MRD rate, sustained MRD negativity, overall response rate (ORR), overall survival (OS), and safety.

How did KRd and VRd compare in transplant-eligible vs transplant-ineligible patients?

The subgroup analysis by transplant eligibility showed distinct patterns in PFS favoring KRd among patients considered eligible for autologous stem cell transplant. In the transplant-eligible cohort, KRd reduced the risk of progression or death by 60% relative to VRd (HR, 0.40; 95% CI, 0.21-0.75; P = .003). Only 23% of patients treated with KRd experienced progression or death compared with 45% in the VRd arm. Median PFS was NR with KRd, whereas VRd demonstrated a median PFS of 40.1 months.

In contrast, outcomes were comparable between the two regimens in the transplant-ineligible population. In this subgroup, both KRd and VRd produced identical progression or death rates of 30%. Median PFS remained NR with KRd vs 52.9 months for VRd, resulting in no statistically significant difference between the arms (HR, 1.06; 95% CI, 0.21-0.75; P = .87).

What were the safety outcomes of the COBRA trial?

The safety profile of KRd and VRd in the COBRA trial demonstrated that both regimens were associated with high rates of AEs, although specific toxicity patterns differed between the treatment arms. Grade 3 or higher AEs occurred in 73% of patients receiving KRd and 62% of those receiving VRd, while any-grade AEs were nearly universal across both cohorts (96% with KRd and 94% with VRd). Treatment discontinuation due to AEs occurred in 11% of KRd-treated patients and 8% of VRd-treated patients.

Grade 5 AEs were infrequent but observed in both arms, occurring in 5 patients (4%) in the KRd arm and 7 patients (6%) in the VRd arm. Reported fatal AEs in the KRd cohort included COVID-19, stroke, pneumonia, sepsis, and acute kidney failure, whereas the VRd arm included deaths from COVID-19, pneumonia, respiratory failure, and two AEs of unknown cause. Neutropenia of any grade occurred in 29% of KRd-treated patients and 17% of VRd-treated patients, with grade 3 or higher neutropenia reported in 21% and 11%, respectively.

Neuropathy, a known bortezomib-associated toxicity, was more common with VRd: 56% of patients experienced neuropathy of any grade compared with 17% in the KRd arm, although grade 3 or higher neuropathy remained low in both groups (2% in each arm). Cardiac AEs occurred more frequently with KRd (18% any grade; 6% grade ≥3) than with VRd (10% any grade; 2% grade ≥3), consistent with the known cardiovascular risk profile of carfilzomib. Infection rates were high in both groups but were more common in the KRd arm, where 75% experienced infections of any grade and 25% developed grade 3 or higher infections, compared with 60% and 23% in the VRd arm, respectively.

References

1. Dytfeld D, Kubicki T, Tyczynska A, et al. Carfilzomib, lenalidomide, and dexamethasone (KRd) versus bortezomib, lenalidomide, and dexamethasone (VRd) in patients with newly diagnosed multiple myeloma: results of the randomized phase III COBRA trial. Blood. 2025;146(suppl 1):99. doi:10.1182/blood-2025-99
2. Carfilzomib, lenalidomide, and dexamethasone versus bortezomib, lenalidomide and dexamethasone (KRd vs. VRd) in patients with newly diagnosed multiple myeloma (COBRA). ClinicalTrials.gov Updated October 3, 2025. Accessed December 6, 2025. https://clinicaltrials.gov/study/NCT03729804

The CD20 and CD3 bispecific antibody odronextamab showed compelling early efficacy in combination with standard CHOP chemotherapy as a treatment for patients with previously untreated diffuse large B-cell lymphoma (DLBCL), according to the first results from a small cohort of patients enrolled in part 1 of the phase 3 OLYMPIA-3 study (NCT06091865) presented at the 2025 ASH Annual Meeting.¹

In the early results, the objective response rate (ORR) was 78% with the weekly 80 mg dose of odronextamab plus CHOP and was 100% for the weekly 160 mg dose of odronextamab plus CHOP. The complete response rates were 44% and 100% for each dose, respectively. The median duration of response, duration of complete response, and progression-free survival were not yet reached at the early analysis. Based on the combination of efficacy and safety, the 160 mg dose of odronextamab was selected for further investigation in the randomized portion of the study comparing the bispecific with rituximab (Rituxan).

“Data from part 1a of OLYMPIA-3 suggest that when combining odronextamab with CHOP in previously untreated patients with DLBCL, rituximab was not required to achieve deep and durable responses,” lead investigator Jean-Marie Michot, MD, from Institute Gustave Roussy, said during a presentation of the results. “The safety profile of fixed duration odronextamab-CHOP treatment was generally manageable in patients with previously untreated DLBCL with high-risk features, with no new safety signals compared with previous reports.”

OLYMPIA-3 Study Design and Patient Characteristics

The open-label study was designed with 2 parts. In part 1, the dose of odronextamab was escalated and optimized. Standard CHOP was given on day 1 and 8 of each cycle and odronextamab was administered starting on day 8, initially at a step-up dose of 0.7/4/20 mg and then at varying dose levels including 80 mg or 160 mg weekly and 160 mg and 320 mg every 2 weeks, with data only available for the weekly doses. Part 2 of the study will continue CHOP with patients randomly assigned to receive odronextamab (Odro-CHOP) or rituximab (R-CHOP).

Across all of part 1, the median age of patients was 66 years (range, 24-81), with nearly a third aged 75 or older (32%). ECOG performance status was 0 (40%), 1 (45%), and 2 (14%). The primary cell of origin was non-GCB (59%), and all patients had de novo DLBCL. IPI score was 3 for 36% and 4 to 5 for 27% of patients. The Lugano stage was III to IV for 95% of patients.

At the time of the analysis, 77.8% of patients enrolled to the 80 mg dose had completed cycle 1 to 6 (7 of 9). The remainder of patients in this group had discontinued early, due to physician decision (n = 2). In the 160-mg arm (n = 13), all patients had completed cycle 1 and 84.6% had completed cycle 6. Two discontinued early due to physician decision. The relative dose intensity was 87% in the 80-mg group and 77% in the 160-mg group.

“Most patients completed 6 cycles of odronextamab-CHOP at both dose levels,” said Michot. “There were few dose reductions of odronextamab and no permanent treatment discontinuations due to TEAEs related to odronextamab. There were no clinically important differences in safety between dose levels.”

Additional Odronextamab Efficacy Findings

The median duration of follow-up was 9.2 months for those enrolled in the 80 mg dose and was 7.8 months for those in the 160 mg dose. At the assessment, most responses remained ongoing. “CRs appeared durable,” Michot said.

In a biomarker analysis, B cell counts declined quickly following the initiation of therapy. There was an initial drop with CHOP administration, with B cells being completely cleared with the initiation of odronextamab.

There was slight T cell margination following the initiation of therapy, but these were transient and like prior reports with odronextamab, Michot said. T cell findings were similar for each dose.

Odronextamab Safety Profile in OLYMPIA-3

Grade 3 or higher treatment emergent adverse events (TEAEs) were experienced by all patients treated with the 80 mg and 160 mg doses of odronextamab. Serious TEAEs were seen in 77.8% of those treated with the 80 mg dose and for 92.3% of those administered the 160 mg dose. TEAEs led to treatment interruption or delay for 66.7% of those in the 80-mg arm and for 84.6% of those in the 160-mg group.

TEAE led to an odronextamab dose reduction for no patients in the 80 mg-arm and for 1 in the 160-mg group. Dose results in CHOP due to TEAEs were needed for 1 patient in the 80-mg group and for 5 in the 160-mg group. TEAEs led to treatment discontinuation for 1 patient in each dose level arm. There was 1 TEAE that led to death in the 160-mg arm. “Of note, there were no dose-limiting toxicities reported,” Michot said.

Across both doses, the most common TEAE was neutropenia (81.8%), cytokine release syndrome (CRS; 54.5%), anemia (45.5%), and nausea (40.9%). The most common treatment-related adverse events were similar with neutropenia seen in 77.3% of patients, CRS in 54.5%, anemia in 45.5%, and nausea in 36.4%.

CRS was solely grade 1/2 in severity, with 40.9% of patients having a grade 1 event and 13.6% having a grade 2 event. Tocilizumab was administered to manage CRS for 27.3% of patients and steroids were given for 18.2%. The median CRS duration was 3.8 months and the median time to onset was 9 hours. CRS mostly occurred during the step-up dosing phase at the lowest dose of 0.7 mg, after this initial step-up the rates of CRS were low. There were no cases of immune effector cell–associated neurotoxicity syndrome or tumor lysis syndrome.

Infections were seen in 81.8% of patients treated across both levels. Of these, 31.8% were grade 3 in severity and 9.1% were grade 4. Opportunistic infections were experienced by 50% of patients, of which only 1 patient had a grade 3 or higher opportunistic infection. The most commonly reported events were CMV infection or reinfection (27% for both) or COVID-19 and oral candidiasis (18% for each).

Odronextamab Regulatory History and Further Study

In August of 2025, the FDA issued a complete response letter (CRL) for a biologics license application for odronextamab for the treatment of relapsed/refractory follicular lymphoma following 2 or more lines of systemic therapy.² Additionally, in March of 2024,³ the agent received 2 CRLs for DLBCL and follicular lymphoma. In both cases, the applications were based on phase 2 findings. The CRL issued in August noted concerns with site inspections completed at a plant ran by Catalent Indiana LLC.

Odronextamab is the subject of several clinical trials across several disease settings, either as monotherapy or in various combination regimens, including the phase 3 OLYMPIA-2 study (NCT06097364) for follicular lymphoma and the phase 3 OLYMPIA-5 study (NCT06149286) looking at odronextamab plus lenalidomide for follicular lymphoma.

REFERENCES

1. Michot J-M, Yagci M, Kargus K, et al. Odronextamab plus chemotherapy in patients with previously untreated diffuse large B-cell lymphoma (DLBCL): First Results from part 1 of the Phase 3 Olympia-3 study. Blood. 2025;146 (Supplement 1):abstract 65. doi:10.1182/blood-2025-65

2. Regeneron Reports Second Quarter 2025 Financial and Operating Results. News release. Regeneron. August 1, 2025. Accessed December 6, 2025. https://tinyurl.com/bdz4e7ex

3. Regeneron provides update on biologics license application for odronextamab. News release. Regeneron. March 25, 2024. Accessed December 6, 2025. https://tinyurl.com/mr2w4j8x