CD20×CD3 Bispecific Antibodies in Relapsed/Refractory B-Cell Lymphomas: Recent Phase III Evidence Durability, and Safety

1. Introduction

Relapsed or refractory (R/R) B-cell non-Hodgkin lymphoma (NHL) is managed with a range of options depending on histology, line of therapy, and fitness, including platinum- or gemcitabine-based salvage chemoimmunotherapy, autologous stem-cell transplantation (ASCT) in transplant-eligible patients, chimeric antigen receptor (CAR) T-cell therapy, antibody-drug conjugates (e.g., polatuzumab vedotin, loncastuximab tesirine), and, in follicular lymphoma (FL), lenalidomide-based combinations. CD20×CD3 bispecific antibodies (bsAbs) epcoritamab, glofitamab, mosunetuzumab, and odronextamab are T-cell engagers that simultaneously bind CD20 on malignant B cells and CD3 on cytotoxic T cells. This therapeutic strategy is part of the broader concept of antigen-directed cancer immunotherapy, in which identification of suitable tumour-associated targets is essential for selective immune recognition and therapeutic engagement [1]. These agents have entered the treatment landscape as an off-the-shelf option that does not require T-cell collection, manufacturing lead time, or lymphodepleting chemotherapy, in contrast to CAR T-cell therapy [2], [3].

These four agents differ in clinically relevant ways that affect how their trial results should be read. Mosunetuzumab is given as a fixed-duration regimen (up to 8–17 cycles depending on response) with treatment stopped after a defined course, whereas epcoritamab, glofitamab, and odronextamab are generally given continuously until progression or unacceptable toxicity in their currently reported regimens. Epcoritamab is administered subcutaneously; glofitamab, mosunetuzumab, and odronextamab are administered intravenously. Prior exposure to CAR T-cell therapy is an important effect modifier: some trial cohorts specifically enrolled CAR-T-naive patients, others enrolled patients after CAR-T failure, and cross-trial comparisons that do not account for this are potentially misleading. Because several randomized trials use different control arms — rituximab-gemcitabine-oxaliplatin (R-GemOx), investigator's choice chemoimmunotherapy (CIT), or lenalidomide-rituximab (R2) — a favorable hazard ratio in one trial does not establish superiority over agents or regimens used as comparators in a different trial. The effectiveness of T-cell-directed cancer immunotherapy also depends on appropriate antigen recognition and preservation of functional cellular immune responses, emphasizing the importance of tumour–host immune interactions in determining therapeutic activity [4].

This review is deliberately restricted to CD20×CD3 T-cell-engaging bsAbs in R/R B-cell lymphoma; CD30×CD16A NK-cell-engaging constructs (e.g., acimtamig/AFM13), used in CD30-positive Hodgkin and T-cell lymphoma, are mechanistically, clinically, and developmentally distinct and are outside this review's scope. The specific question addressed is: has randomized phase 3 evidence, now available for two of these four agents, changed the clinical position of CD20×CD3 bsAbs in R/R DLBCL and FL relative to the earlier single-arm era, and what does this mean for treatment duration, sequencing around CAR T-cell therapy, and infection burden with prolonged exposure? Prior reviews have summarized this drug class [2], [3]; this review's contribution is to synthesize the trial evidence reported or updated through the search date below, with explicit separation of randomized from single-arm evidence and a standardized safety comparison.

2. Review Methods

This is a narrative, non-systematic review. It does not follow PRISMA methodology, did not use dual-reviewer independent screening, and did not include a formal risk-of-bias assessment or meta-analysis. These are acknowledged limitations, stated here for transparency rather than presented as a systematic review.

Literature and regulatory documents were searched in PubMed/MEDLINE, ClinicalTrials.gov, the FDA and EMA websites (prescribing information, approval summaries, and European Public Assessment Reports), and the conference abstract archives of the American Society of Hematology (ASH), American Society of Clinical Oncology (ASCO), and European Hematology Association (EHA). The search covered literature and presentations from January 2020 through August 20, 2026 (final search date), applied consistently across all agents discussed. Search strings combined the agent generic name (epcoritamab, glofitamab, mosunetuzumab, or odronextamab) with “lymphoma,” “bispecific,” and, for safety-focused searches, “cytokine release syndrome,” “ICANS,” or “infection.”

An evidence hierarchy was applied when more than one report existed for the same trial or analysis: final peer-reviewed primary publications were used preferentially; where no peer-reviewed publication existed, primary regulatory documents (FDA prescribing information/approval summaries, EMA European Public Assessment Reports) or trial registry records were used; conference abstracts were used only where no more mature report was available and are explicitly labeled as such in-text and in the tables; and company press releases were used only as a last resort for topline results not yet reported elsewhere, and are likewise labeled. When multiple reports existed for the same trial (e.g., primary analysis and subsequent long-term follow-up), the most mature available report with the longest follow-up was used as the primary data source, with earlier reports cited only where they add information not superseded by the later report (e.g., STARGLO primary analysis and 3-year follow-up are consolidated into a single table row using the latest data cutoff).

Eligible reports were phase 1–3 interventional trials of CD20×CD3 bsAbs in patients with B-cell lymphoma reporting at least one efficacy endpoint (overall response rate [ORR], complete response [CR] rate, progression-free survival [PFS], overall survival [OS], or duration of response) or safety endpoint (cytokine release syndrome [CRS], ICANS, infection, cytopenia, treatment discontinuation, or treatment-related mortality). Preclinical studies, non-English-language reports, and reports without extractable trial-level data were excluded. Literature screening and data extraction were performed by a single reviewer, a further limitation relative to formal systematic review. Given substantial heterogeneity in trial design, patient populations, treatment combinations, and follow-up duration, results are synthesized narratively rather than pooled in a meta-analysis, and cross-trial numerical comparisons are avoided except where explicitly framed as indirect and non-equivalent.

3. CD20×CD3 Bispecifics in Aggressive B-Cell Lymphoma (LBCL/DLBCL)

3.1 Randomized phase 3 evidence

The randomized phase 3 STARGLO trial (n=274) compared glofitamab plus gemcitabine-oxaliplatin (Glofit-GemOx) against rituximab-GemOx (R-GemOx) in transplant-ineligible R/R DLBCL after ≥1 prior line. At the latest reported data cutoff (3-year follow-up, median 35.1 months), median OS was 25.5 versus 12.5 months (hazard ratio [HR] 0.60, 95% CI 0.43–0.83), median PFS was 14.4 versus 3.3 months (HR 0.41, 95% CI 0.29–0.57), and CR rate was 58.5% versus 25.3% [5]. This is a single randomized trial against a chemoimmunotherapy comparator in a transplant-ineligible population; it does not establish superiority over CAR T-cell therapy or other salvage strategies not included as comparators.

The randomized phase 3 EPCORE DLBCL-1 trial (n=483) compared subcutaneous epcoritamab monotherapy against investigator's choice chemoimmunotherapy (CIT) in R/R DLBCL. Topline results reported a PFS improvement (HR 0.74, 95% CI 0.60–0.92) without a statistically significant OS improvement (HR 0.96, 95% CI 0.77–1.20) [6]. As of the search date, this remains a company topline announcement rather than a peer-reviewed or conference-presented primary analysis, which limits independent verification of the reported effect sizes; this should be treated as provisional pending fuller reporting.

3.2 Single-arm evidence and post-CAR-T activity

In the pivotal single-arm EPCORE NHL-1 trial of subcutaneous epcoritamab monotherapy in R/R LBCL with ≥2 prior therapies (including patients previously treated with CAR T-cell therapy), the registrational efficacy cohort (n=157) showed ORR 59% and CR 41% at 3-year follow-up (median 37.1 months) [7]; the smaller FDA registrational cohort (n=148) reported ORR 61% and CR 38% with median DOR 15.6 months at a median follow-up of 9.8 months among responders [8]. A Japanese single-arm cohort (EPCORE NHL-3, n=36) reported ORR 56% and CR 47% at 3-year follow-up [9].

In the DLBCL cohort of the phase 2 ELM-2 trial of odronextamab, naive to prior CAR T-cell therapy (n=127), the peer-reviewed primary analysis reported ORR 52.0% and CR 31.5%, with median duration of response 10.2 months and median duration of CR 17.9 months at 29.9 months' efficacy follow-up [10]. Separately, in patients with DLBCL specifically progressing after prior CAR T-cell therapy, the phase 1 ELM-1 trial reported ORR 48% and CR 32%, providing direct evidence of post-CAR-T activity for odronextamab, though this is a smaller, single-arm, non-randomized cohort [11].

4. CD20×CD3 Bispecifics in Follicular Lymphoma

4.1 Randomized phase 3 evidence

The randomized phase 3 EPCORE FL-1 trial compared epcoritamab plus lenalidomide-rituximab (R2) against R2 alone in R/R FL (n=243 vs 245). ORR was 89% versus 74% (p<0.0001) and CR was 74% versus 43% (p<0.0001) favoring the triplet; PFS was not reached with epcoritamab+R2 versus 11.2 months with R2 (HR 0.21, 95% CI 0.13–0.33) [12]. This is a chemotherapy-free combination compared against a chemotherapy-free doublet, not against chemoimmunotherapy, and represents evidence in previously treated rather than untreated FL.

4.2 Single-arm and fixed-duration evidence

In the pivotal single-arm phase 2 study of fixed-duration mosunetuzumab monotherapy (up to 8 cycles) in R/R FL after ≥2 prior therapies (n=90), 3-year follow-up (median 37.4 months) showed ORR 77.8% and CR 60.0%, with median PFS 24.0 months; no new CRS events or fatal, serious, or grade ≥3 adverse events emerged with extended follow-up beyond the prior data cutoff [13]. In the non-randomized US extension cohort of the phase 3 CELESTIMO study, mosunetuzumab plus lenalidomide (Mosun-Len) in R/R FL showed ORR 96.3% and CR 87.0% at 12.7 months' follow-up; this is an uncontrolled cohort reported at a conference, not yet published as a peer-reviewed randomized comparison [14].

In the FL cohort of ELM-2 (n=128), the peer-reviewed primary analysis reported ORR 80.0% (95% CI 72.5–86.9%) and CR 73.4% at 20.1 months' median follow-up, with median duration of CR 25.1 months [15]; a corrigendum to this publication corrected a dosing-regimen typographical error in the supplementary material without altering the reported efficacy or safety results [16].

5. Safety and Implementation

Table 2 standardizes safety data using consistent categories (CRS, ICANS, infection, neutropenia, discontinuation, treatment-related death) across trials, with the grading system and population (monotherapy, combination, or pooled cohort) specified for each row; qualitative labels such as “rare” or “low” are avoided in favor of exact reported percentages where available.

CRS is the most common early toxicity of CD20×CD3 bsAbs and is concentrated in the first 1–2 treatment cycles across agents. In the EPCORE NHL-1 registrational LBCL cohort, CRS occurred in 51% of patients (37% grade 1, 17% grade 2, 2.5% grade 3) [8]. ICANS, a distinct and separately graded toxicity from CRS, occurred less frequently but is clinically important: ICANS was included among serious adverse reactions occurring in ≥2% of patients and among fatal adverse reactions (0.6%) in the same registrational population [8]. This exact incidence and severity distribution, rather than a qualitative descriptor, should be used when characterizing epcoritamab's neurotoxicity profile in this population.

In cohorts using optimized step-up dosing regimens (for example, the 0.7/4/20 mg odronextamab regimen or epcoritamab's cycle-1 dose-optimization schedule), CRS incidence and severity were lower than in earlier, non-optimized cohorts within the same development programs. Because these are before/after or across-cohort comparisons rather than randomized comparisons, this should be described as an observed association with lower incidence rather than a proven causal reduction.

Infection is a persistent risk with prolonged CD20×CD3 bsAb exposure, reflecting sustained B-cell depletion and consequent hypogammaglobulinemia; this immunologic consequence is distinct from hematologic cytopenias (neutropenia, anemia, thrombocytopenia) and both should be tracked separately from CRS/ICANS. In the ELM-2 DLBCL cohort, grade ≥3 infections occurred in 37% of patients, including 11% grade 5 (fatal) infections [10]; in the pooled ELM-1/ELM-2 population, COVID-19 infection was a specific and clinically significant contributor to infection-related mortality. “No new adverse events” reported at long-term follow-up (e.g., mosunetuzumab 3-year data) means no new events emerged after the prior data cutoff; it does not mean serious or grade ≥3 events did not occur earlier in the same trial, and should not be read as an overall safety guarantee.

Practically, fixed-duration regimens (mosunetuzumab) limit cumulative infection and hypogammaglobulinemia exposure by design, while continuously dosed regimens (epcoritamab, glofitamab, odronextamab in their currently reported schedules) require ongoing monitoring for infection and consideration of immunoglobulin replacement or antimicrobial/antiviral prophylaxis with prolonged exposure; head-to-head data comparing these strategies are not yet available.

6. Regulatory Status

As of the search date, epcoritamab (Epkinly) and glofitamab hold FDA accelerated approval and EMA authorization for R/R DLBCL after ≥2 prior lines, with epcoritamab additionally approved in combination with R2 for R/R FL following the EPCORE FL-1 results; mosunetuzumab holds approval for R/R FL after ≥2 prior lines [2], [8], [12], [13].

Odronextamab's regulatory history is more complex and region-specific. The European Commission granted conditional marketing authorization for odronextamab (Ordspono) for R/R FL and R/R DLBCL after ≥2 prior lines in August 2024, following a positive CHMP opinion on 27 June 2024, based on the ELM-1 and ELM-2 trials [17]. In the United States, the FDA issued a complete response letter (CRL) in March 2024 related to enrollment status of the confirmatory phase 3 trial, not to efficacy, safety, trial design, or manufacturing [18]. Following attainment of the relevant enrollment target, a BLA resubmission for the FL indication was accepted in February 2025 with a PDUFA target date of July 30, 2025; the FDA subsequently issued a second CRL for the FL indication related to observations from a general site inspection at a third-party manufacturing facility, not to efficacy or safety of odronextamab itself [19]. As of this review's search date, odronextamab therefore remains approved in the EU but not yet approved in the US.

7. Discussion

Randomized phase 3 evidence is now available for two of the four CD20×CD3 bsAbs discussed here, in specific settings: glofitamab plus chemotherapy improved OS versus chemoimmunotherapy in transplant-ineligible R/R DLBCL (STARGLO), and epcoritamab plus R2 improved PFS and response rates versus R2 alone in previously treated FL (EPCORE FL-1). Epcoritamab's DLBCL-1 topline results improved PFS but not OS versus chemoimmunotherapy, and remain unpublished beyond a company announcement. These are agent-, trial-, and setting-specific findings and do not establish class-wide superiority of CD20×CD3 bsAbs over other treatment strategies, including CAR T-cell therapy, which was not a comparator arm in any of the trials reviewed here.

On sequencing relative to CAR T-cell therapy: available data indicate that CD20×CD3 bsAbs retain meaningful activity after CAR-T failure (ELM-1 post-CAR-T cohort, ORR 48%; EPCORE NHL-1 included CAR-T-pretreated patients within its overall 59% ORR), supporting their use as a subsequent-line, off-the-shelf option without the manufacturing delay of a second CAR-T product. Conversely, whether bsAb exposure affects the efficacy of a subsequent CAR-T attempt — through T-cell exhaustion or antigen-escape mechanisms — is not well characterized in the trials reviewed and represents an important gap given the increasing use of both modalities in sequence.

On treatment duration: mosunetuzumab's fixed-duration design (up to 8 cycles) is a structurally different exposure strategy from the continuous dosing used in the epcoritamab, glofitamab, and odronextamab regimens reported here, with plausible implications for cumulative infection risk, hypogammaglobulinemia, cost, and patient burden; however, no trial in this review directly randomized fixed-duration against continuous dosing, so this remains a hypothesis-generating observation rather than an established comparative advantage.

On resistance and biomarkers: CD20 loss or antigenic downregulation, T-cell dysfunction or exhaustion after prolonged engager exposure, and tumor microenvironmental factors have been proposed as mechanisms of acquired resistance to CD20×CD3 bsAbs, though the trials reviewed here were not designed to characterize resistance mechanisms directly. This underscores the general importance of sustained tumour-associated antigen expression for the success of antigen-directed therapies: reduced or lost CD20 expression removes the target that these bsAbs depend on, echoing the broader principle that identification and persistence of a suitable tumour-associated target is a prerequisite for durable immune-mediated tumour control [1]. Preservation of functional T-cell responses is equally relevant, since bsAb activity depends on host cytotoxic T cells remaining capable of recognizing and eliminating antigen-bearing tumour cells rather than on an engineered cellular product, linking durability of response to the broader tumour–host immune interaction [4]. Minimal residual disease (MRD) status by circulating tumor DNA was associated with PFS in exploratory ELM-2 analyses [10], suggesting a possible role for MRD-guided treatment duration or discontinuation decisions; prospective validation is needed before this can inform practice.

On infection burden specifically: the combination of grade ≥3 infection rates in the range of one-third of patients and infection-related fatal events (including COVID-19) in some cohorts represents a clinically important, not merely theoretical, risk that should be weighed against the convenience of an off-the-shelf agent, particularly in older or more heavily pretreated patients, and argues for systematic infection-prophylaxis strategies as these regimens move into earlier lines of therapy.

Limitations of the evidence base, and of this review, include: this is a narrative rather than formal systematic review without dual-reviewer screening or risk-of-bias assessment; most included trials outside STARGLO and EPCORE FL-1 are single-arm phase 1/2 studies without concurrent controls; EPCORE DLBCL-1 is available only as a company topline announcement at the time of this review; cross-trial heterogeneity in patient populations, prior therapy exposure, CRS/ICANS grading conventions, and prophylaxis protocols limits direct comparison even where numerical results are presented together; and follow-up remains comparatively short for several combination and frontline regimens.

8. Conclusion

Within the above limitations, randomized phase 3 evidence now supports specific benefits for CD20×CD3 bsAbs in defined R/R settings: glofitamab plus chemotherapy for OS in transplant-ineligible R/R DLBCL (STARGLO), and epcoritamab plus R2 for PFS and response in previously treated FL (EPCORE FL-1). Evidence for odronextamab, for mosunetuzumab combinations, and for epcoritamab in DLBCL-1 remains either single-arm, topline/unpublished, or based on non-randomized extension cohorts, and should be described accordingly rather than placed on equal evidentiary footing with the randomized results. CRS is well characterized and appears reducible with step-up dosing in non-randomized comparisons; ICANS and infection, including fatal infection, are less frequent but clinically important and require exact trial-specific reporting rather than qualitative reassurance. The most important unresolved clinical questions are how to sequence CD20×CD3 bsAbs around CAR T-cell therapy, whether fixed-duration and continuous-dosing strategies produce meaningfully different long-term infection and efficacy outcomes, what resistance mechanisms limit durability, and whether MRD-guided approaches can safely inform treatment duration.

9. Abbreviations and Acronyms

ASCT: autologous stem-cell transplantation; bsAb: bispecific antibody; CI: confidence interval; CIT: chemoimmunotherapy (investigator's choice); CR: complete response; CRL: complete response letter; CRS: cytokine release syndrome; DLBCL: diffuse large B-cell lymphoma; DOR/mDOR: (median) duration of response; mDOR-CR: median duration of complete response; FL: follicular lymphoma; HR: hazard ratio; ICANS: immune effector cell-associated neurotoxicity syndrome; ICR: independent central review; IRC: independent review committee; LBCL: large B-cell lymphoma; NE: not estimable; NHL: non-Hodgkin lymphoma; NR: not reported; ORR: overall response rate; OS: overall survival; PFS: progression-free survival; R/R: relapsed/refractory; R2: lenalidomide plus rituximab; R-GemOx: rituximab plus gemcitabine-oxaliplatin.

10. Acknowledgement

None declared. No specific funding was received for this work.

11. Use of Generative AI Disclosure

AI tool used: Claude (Anthropic), accessed via Claude.ai.

Task performed: The AI tool was used to conduct literature and regulatory-document searches, and to draft and organize the synthesized manuscript text, tables, and reference list, including revisions responding to editorial and peer review.

Author confirmation: The author has reviewed, verified against original sources, and edited the AI-assisted output for accuracy, completeness, and potential bias, and takes full responsibility for the content, interpretation, and conclusions of this manuscript.

12. References

1. Tawel, H. M., Ali Hasen, Y. M., Miles, A., (2022), Cancer testis antigen (CTA) T21, as a potential diagnostic, prognostic, and immune-therapeutic target for malignant tumours., Journal of Medical Sciences, 17(1), 1-7. https://doi.org/10.51984/joms.v17i1.1892
2. Bayly-McCredie, E., Treisman, M., Fiorenza, S., (2024), Safety and efficacy of bispecific antibodies in adults with large B-cell lymphomas: A systematic review of clinical trial data., Int J Mol Sci, 25(17), 9736. https://doi.org/10.3390/ijms25179736
3. Liu, X., Zhao, J., Guo, X., Song, Y., (2023), CD20×CD3 bispecific antibodies for lymphoma therapy: Latest updates from ASCO 2023 annual meeting., J Hematol Oncol, 16, 90. https://pubmed.ncbi.nlm.nih.gov/37537626/
4. Alhoderi, J. A. H., Fernandez, N., (2022), Prospective cancer vaccines for prevention and treatment of cancer., Journal of Medical Sciences, 17(2), 14-23. https://doi.org/10.51984/joms.v17i2.2187
5. Abramson, J. S., Ku, M., Fox, C. P., et al., (2026), Glofitamab plus gemcitabine and oxaliplatin for relapsed/refractory DLBCL: 3-year follow-up of STARGLO., Blood Adv. https://doi.org/10.1182/bloodadvances.2026019764
6. Genmab A/S, AbbVie Inc., (2026, January 16), Genmab announces topline results for epcoritamab from phase 3 EPCORE DLBCL-1 trial [company press release; topline results, not yet peer-reviewed]., Available at: https://ir.genmab.com/news-releases/news-release-details/genmab-announces-topline-results-epcoritamab-duobodyr-cd3xcd20/
7. Karimi, Y. H., Cheah, C. Y., Clausen, M. R., et al., (2026), Efficacy and safety of epcoritamab in relapsed or refractory large B-cell lymphoma: 3-year update from the EPCORE NHL-1 trial., Ann Hematol, 105(3), 79. https://doi.org/10.1007/s00277-026-06798-4
8. U.S. Food and Drug Administration, (2025), EPKINLY (epcoritamab-bysp) prescribing information [primary regulatory document]., Available at: https://www.accessdata.fda.gov/drugsatfda_docs/label/2025/761324s009lbl.pdf
9. Izutsu, K., Kumode, T., Yuda, J., et al., (2025), Efficacy and safety of epcoritamab in Japanese patients with relapsed or refractory diffuse large B-cell lymphoma: 3-year follow-up from the EPCORE NHL-3 trial., Int J Clin Oncol. https://doi.org/10.1007/s10147-025-02788-0
10. Kim, W. S., et al., (2025), Odronextamab monotherapy in patients with relapsed/refractory diffuse large B-cell lymphoma: Primary efficacy and safety analysis in phase 2 ELM-2 trial., Nat Cancer. https://doi.org/10.1038/s43018-025-00921-6
11. Topp, M. S., Matasar, M., Allan, J. N., et al., Crombie, J. L., (2025), Odronextamab monotherapy in R/R DLBCL after progression with CAR T-cell therapy: Primary analysis of the ELM-1 study., Blood, 145(14), 1498-1509. https://doi.org/10.1182/blood.2024027044
12. AbbVie Inc., Genmab A/S, (2026), EPCORE FL-1: Epcoritamab plus lenalidomide and rituximab (R2) versus R2 in relapsed/refractory follicular lymphoma — primary results. https://doi.org/10.1016/S0140-6736(25)02360-8
13. Sehn, L. H., Bartlett, N. L., Matasar, M. J., et al., (2025), Long-term 3-year follow-up of mosunetuzumab in relapsed or refractory follicular lymphoma after ≥2 prior therapies., Blood, 145(7), 708-719. https://pubmed.ncbi.nlm.nih.gov/39447094/
14. Sano, D., Bartlett, N., Budde, E., et al., (2025), Promising response rates and manageable safety with mosunetuzumab plus lenalidomide (Mosun-Len) in patients with relapsed/refractory follicular lymphoma: US extension cohort from the phase III CELESTIMO study [conference abstract]., Blood, 146(Suppl 1), 1800. https://doi.org/10.1182/blood-2025-1800
15. Kim, T. M., Taszner, M., Novelli, S., et al., (2024), Safety and efficacy of odronextamab in patients with relapsed or refractory follicular lymphoma., Ann Oncol, 35(11), 1039-1047. https://doi.org/10.1016/j.annonc.2024.08.2239
16. ELM-2 Investigators, (2025), Corrigendum to “Safety and efficacy of odronextamab in patients with relapsed or refractory follicular lymphoma.”, Ann Oncol, 36(3), 352. https://doi.org/10.1016/j.annonc.2024.11.009
17. European Medicines Agency, (2024), Ordspono (odronextamab): An overview of Ordspono and why it is authorised in the EU. EMA/313527/2024 [primary regulatory document]., Available at: https://www.ema.europa.eu/en/documents/overview/ordspono-epar-medicine-overview_en.pdf
18. Regeneron Pharmaceuticals, Inc., (2024, March 25), Regeneron provides update on Biologics License Application for odronextamab [primary regulatory/company source]., Available at: https://investor.regeneron.com/news-releases/news-release-details/regeneron-provides-update-biologics-license-application
19. OncLive, (2025), FDA issues CRL for odronextamab in relapsed/refractory follicular lymphoma [secondary report of primary FDA regulatory action, manufacturing-site inspection finding]., Available at: https://www.onclive.com/view/fda-issues-crl-for-odronextamab-in-relapsed-refractory-follicular-lymphoma