Dear Editor,
We are pleased to submit a response to the Letter to the Editor you have received by van Nimwegen et al. [ 1 ] and appreciate their interest and thoughtful analysis of our study. We welcome the opportunity to clarify certain points and provide additional context regarding the methodology and interpretation of our findings. While we acknowledge the valuable perspectives raised, we would like to address specific aspects where further explanation may help avoid potential misinterpretation.
Relative Efficacy of Elranatamab Versus Teclistamab
Uncertainty in Treatment Comparison
In relation of the mentioned differences in the timing of the clinical studies for elranatamab and teclistamab, while the emergence of the COVID-19 pandemic, including the circulation of variants such as the Delta variant of SARS-CoV-2—which was designated by the WHO in May 2021—may have influenced some of the outcomes observed in these trials, both studies recruited patients during periods of high COVID-19 waves, coincident with the initial appearance in 2020 and also several pandemic peaks during 2021 ( MajesTEC-1 enrolled from March 3, 2020 to August 13, 2021 and MagnestisMM-3 from February 9, 2021 to January 7, 2022) [ 2 ].
During these periods, access to COVID-19 treatments and vaccines varied significantly across the participating countries, and precise information is lacking regarding how and at what point during treatment—or in response to it—COVID-19 may have had an impact in each case. Consequently, post hoc analyses can provide informative insights but cannot be used for formal analyses, as they lack a validated and standardized methodology and may therefore introduce bias in their development and interpretation, especially when they are only performed over one of the options included in the indirect comparisons.
Furthermore, while it is acknowledged that there was a high initial number of COVID-related adverse events (AEs) in the teclistamab study (17.6%) and 12 (7.3%) of the patients died as a result of COVID-19 with a median follow-up of 14.1 months [ 3 ], the MagnetisMM-3 trial also reported a high incidence of COVID-19 infections (29.3%) with a median follow-up was 14.7 months [ 4 ]. However, unlike teclistamab, only 1.6% of patients in the elranatamab study died from this cause, and it remains unclear whether the treatment itself could have any direct relationship to these outcomes.
Regarding the censoring in the progression-free survival (PFS) curve for elranatamab, it has been suggested that this is likely attributable to a high discontinuation rate due to AEs. In relation to this, in a recent analysis, van Nimwegen et al. [ 5 ] assumed that all censored patients are unobserved PFS events mainly due to AEs (informative censoring), and that the totality of them will be immediate events (progressions) [ 5 ]. Nevertheless, a recent post hoc analysis of the subgroup of patients in the MagnetisMM-3 study who experienced prolonged interruptions of elranatamab treatment due to AEs estimates that the median duration of the treatment-free period was 15.5 months (range 6.5–35.4). The majority of patients in this subgroup maintained clinical response despite a prolonged dose interruption demonstrating the durability of elranatamab responses [ 6 ].
On the other hand, incorporating the lower PFS estimates from van Nimwegen et al. [ 5 ] into our analysis would have only an impact on treatment costs for those patients in the progression-free state, leading to substantially lower treatment costs for patients remaining in this state with elranatamab compared with teclistamab, resulting in additional cost savings in favour of elranatamab. For the impact on the life-years gained in the post-progression state, overall survival (OS) curves should be considered. Available evidence indicates that the influence of informative censoring is generally less pronounced for OS than for PFS, primarily because follow-up periods are longer and a greater proportion of patients experience the event (death) prior to censoring [ 7 , 8 ]. In this context, we did not identify corresponding analyses for OS curves of teclistamab and elranatamab. This is particularly relevant because OS curves have a direct impact on the differences in life-years gained between the two treatment alternatives after progression.
In addition to PFS and OS, other efficacy endpoints have also demonstrated differences between the two studies. For example, the median duration of response (DoR) in the long-term follow-up of the MajesTEC-1 trial [ 9 ] is reported to be 24 months, whereas in MagnetisMM-3, the median DoR has not yet been reached after more than 34 months of follow-up, with 61% of responders maintaining their response at 30 months [ 10 ].
Regarding the use of unanchored matching-adjusted indirect comparisons (MAICs), the limitations are acknowledged in Encinas et al. [ 11 ]. Nonetheless, in the absence of head-to-head studies, unanchored MAICs are widely employed with single-arm trials without a common comparator, as demonstrated in the referenced MAIC by Gordan et al., and others that have been recently published [ 11 – 15 ]. However, the MAIC adjustment described by Gordan et al. [ 12 ] is restricted to variables that can be appropriately reweighted to align with the selected baseline characteristics of patients treated with elranatamab in MagnetisMM-3. Other important baseline factors, such as ECOG performance status (PS), could not be properly balanced because MajesTEC-1 excluded patients with ECOG PS > 1, whereas MagnetisMM-3 permitted enrolment of patients with ECOG PS = 2. Furthermore, the outcome selected for the MAIC—overall response rate (ORR)—has notable limitations compared with PFS and OS. ORR does not reflect the duration of benefit, is subject to interobserver variability, and is a poor surrogate for OS [ 16 ]. These factors contribute to considerable uncertainty in interpreting the reported MAIC, given the availability of long-term OS data for both treatment alternatives [ 17 ].
Progression-Free Survival and Overall Survival Extrapolations
As we described in the Methods section of the manuscript, extrapolating long-term survival data is a necessary part of all cost-effectiveness analyses, including ours. Of all the parametric models, the Weibull model was selected for the base case, even though it did not show the best fit according to Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). This is because visual checks showed that the Gompertz, generalized gamma, log-normal, and log-logistic models lead to overestimation. For example, the extrapolation curve yielded by Gompertz estimated that 29% of patients were alive at 25 years, and 6–8% of patients are still alive at year 25 with generalized gamma, log-normal, and log-logistic curves. This may be too optimistic compared with what is observed in a real-world setting, in that less than 10% of the patients were still alive at 3 years [ 18 ]. The same rationale was applied to the PFS, for which a Weibull model was also selected.
As stated in the UK’s National Institute for Health and Care Excellence (NICE) technical appraisal report on elranatamab [ 19 ], using generalized gamma (or Weibull) distributions for modelling both PFS and OS results in crossed curves early in the extrapolation period. To overcome this, priority was given to the PFS curve, allowing the OS curve to converge with the PFS curve. It should be noted that the exact same phenomenon occurs with teclistamab, except that the curves converge later, specifically at month 50. Experts at the committee meeting explained that multiple myeloma generally responds to new classes of therapies better than therapies that have been used previously and added that there was good reason to be optimistic about elranatamab on the basis of its newer mechanism of action [ 20 ]. In the elranatamab NICE technical appraisal report, the committee suggested a scenario in which a generalized gamma distribution was used for OS and a gamma distribution for PFS, in order to avoid convergence of the curves. We ran this scenario in our model and found that elranatamab remains dominant over teclistamab.
As outlined in our manuscript, alternative extrapolations to the Weibull models were evaluated in sensitivity analyses, and the findings remained consistent with the base case. To provide further clarity, Table 1 summarises all scenarios using different parametric models. These results indicate that the convergence of the PFS and OS curves does not alter the interpretation of the outcomes.
Table 1
Summary of sensitivity analysis scenarios using alternative parametric models
ICER incremental cost-effectiveness ratio, QALY quality adjusted life year
| Elranatamab curves converge at month | ICER (€/QALY) vs teclistamab | |
| Base case: Weibull both OS and PFS | 31 | − 169,417 (dominant) |
| S1: generalized gamma both OS and PFS | 25 | − 70,517 (dominant) |
| S2: generalized gamma OS, gamma PFS (TA1023) | Do not converge | − 76,594 (dominant) |
| S3: log-normal both OS and PFS | 66 | − 90,744 (dominant) |
| S4: log-normal OS, Weibull PFS | Do not converge | − 78,650 (dominant) |
| S5: log-logistic both OS and PFS | 45 | − 90,479 (dominant) |
| S6: log-logistic OS, exponential PFS | Do not converge | − 95,127 (dominant) |
Elranatamab Treatment Costs vs Teclistamab
Duration of Treatment Extrapolations
To estimate treatment costs over the time horizon, the base case applied exponential models adjusted to the reported median duration of treatment (DoT). This approach was necessary because only median treatment durations were available for comparators at the time of model development; therefore, the same method was applied consistently to all comparators to minimize bias. The authors acknowledge limitations related to survival curve extrapolations and the simplification of the disease’s natural history, as previously noted. Although the exponential distribution may underestimate treatment costs, this effect is symmetrical across both arms. Additionally, we evaluated alternative, more conservative scenarios in which a Weibull model (as used for PFS and OS) and a log-normal model (identified as best fit by AIC/BIC) were applied to elranatamab, adjusted to time to discontinuation (TTD) data from MagnetisMM-3. We also explored a conservative scenario incorporating spending caps on teclistamab treatment. Across all scenarios, elranatamab consistently dominated teclistamab.
Dose Switching and Relative Dose Intensity
Regarding the possibility of switching to less frequent dosing schedules, assumptions were made based on the conditions outlined in the respective Summaries of Product Characteristics (SmPC) and available public evidence, which were also part of their registrational trials.
For elranatamab, switching from weekly (QW) to a biweekly (Q2W) dosing schedule was applicable to patients who remained in the progression-free state after 24 weeks. According to elranatamab’s SmPC [ 20 ], patients who have received at least 24 weeks of treatment and achieved any response should transition to Q2W dosing. It is assumed that the patients who did not switch in the trial were either protocol deviations, on a long-term dose interruption, or non-responders per investigator or blinded independent central review (BICR). Furthermore, the recent update to elranatamab’s SmPC, in which patients who have received at least 24 weeks of treatment at the Q2W and maintained the response should transition to the every 4-week schedule (Q4W), was not considered at the time of this analysis, which could potentially result in additional cost savings with elranatamab.
For teclistamab, switching to Q2W dosing was also incorporated into the model for patients in the progression-free state. However, per the SmPC, patients must first achieve a complete response or better and maintain it for a minimum of 6 months before considering a reduction in the dosing frequency to 1.5 mg/kg subcutaneous (SC) every 2 weeks [ 21 ]. In this analysis, it was assumed that 38.2% of patients switch from QW dosing to Q2W dosing at 11.3 months, based on the available evidence [ 22 ].
While it is true that, in the context of clinical trials, certain nuances may not fully reflect real-world practice—such as dose reductions, adjustments to dosing schedules, or slightly different criteria applied to phase 1 and 2 trial populations, which do not necessarily occur in routine care—in the absence of real-world data, information from clinical trials is used. Regarding relative dose intensity (RDI), the overall impact on the global analysis is likely minimal. To address uncertainty related to other key cost drivers such as RDI, scenario and sensitivity analyses were conducted.
Subsequent Therapy Costs
In our analysis, we considered that 33.9% of patients who progress to the post-progression stage will receive subsequent treatment. The composition of these treatments and the proportion of patients receiving them were derived from the MagnetisMM-3 trial and validated by local clinical experts.
Regarding the modelling of subsequent costs, we calculated newly progressed patients in each cycle and applied the cost of subsequent treatment as a one-off input. As previously discussed, the area between the two survival curves determines the time spent in the progressed-disease state for patients receiving subsequent therapy. Sensitivity analyses, presented in Table 1 , indicate that their impact on model outcomes is not substantial.
Conclusions
In summary, we acknowledge the methodological limitations inherent to this type of study, particularly those arising from the use of indirect comparisons, in which the supporting evidence is drawn from two separate studies conducted at different time periods. However, the shortcomings in the arguments provided regarding relative effectiveness and those related to treatment costs presented by van Nimwegen et al. are unlikely to alter the overall direction of the results. This is supported by the extensive scenario and sensitivity analyses conducted, which consistently confirm the robustness of the results.
Author Contributions
Virginia Lozano, Julia Llinares, Patrick Halavacek, Olivia Ashman, David Carcedo, Cristina Encinas and Joaquín Martínez-López contributed to the letter development. The first draft was written by Virginia Lozano and Julia Llinares. All authors reviewed on previous versions of the document. All authors read and approved the final version.
Funding
No funding or sponsorship was received for the publication of this letter. This response letter is based on previously conducted study that was funded by Pfizer SLU, Madrid, Spain. David Carcedo is employee of Hygeia, which received funding from Pfizer in connection with the development of the original manuscript. Joaquín Martínez-López and Cristina Encinas were paid consultant to Pfizer in connection with the development of the original manuscript.
Declarations
Conflict of Interest
Virginia Lozano, Julia Llinares, Patrick Hlavacek and Olivia Ashman are employees of Pfizer and may hold Pfizer stock. David Carcedo is employee of Hygeia Consulting. Joaquín Martínez-López has received honoraria derived from research projects from Amgen, Incity, Janssen and Pfizer, and from lectures and participation in advisory boards from Amgen, BMS, Gilead, Incity, Janssen, Novartis, Pfizer and Roche. Cristina Encinas has received honoraria derived from lectures and participation in advisory boards from Amgen, BMS, GSK, Johnson and Johnson, Menarini Stemline, Sanofi and Pfizer.
Ethics Declarations
This response letter is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.
References
1. Nimwegan K, et al. Letter to the editor. A response to “Cost-effectiveness of elranatamab versus current therapies for the management of patients with triple-class exposed, relapsed and refractory multiple myeloma, including other bispecific and physician’s choice of treatment in Spain”. Oncol Ther. 2026. 10.1007/s40487-026-00420-3.
2. Van de Donk N, et al. Preliminary recommendations for prevention and management of infections, hypogammaglobulinemia, and neutropenia during treatment with teclistamab based on experience from the MajesTEC‑1 study. Presented by NWCJ van de Donk at the 20th International Myeloma Society (IMS) Annual Meeting: September 27–30, 2023; Athens, Greece.
3. Moreau P, Garfall AL, van de Donk NWCJ, et al. Teclistamab in relapsed or refractory multiple myeloma. N Engl J Med. 2022;387:495–505.
4. Lesokhin A, Tomasson MH, Arnulf B, et al. Elranatamab in relapsed or refractory multiple myeloma: phase 2 MagnetisMM-3 trial results. Nat Med. 2023. 10.1038/s41591-023-02528-9.
5. Van Nimwegen KJM, Van Sanden S, Ghilotti F, et al. Evaluating the impact of censoring on progression-free survival comparisons of BCMA-targeting bispecific antibodies in relapsed/refractory multiple myeloma. Value Health. 2025;28(12):S63. 10.1016/j.jval.2025.09.210.
6. Lesokhin A, et al. Prolonged elranatamab treatment interruption in patients with relapsed or refractory multiple myeloma (RRMM) is feasible: a retrospective analysis from MagnetisMM-3. ASH Annual Meeting; 2025 Dec; San Diego, CA, USA. Abstract abs25-8338.
7. Templeton AJ, Amir E, Tannock IF. Informative censoring—a neglected cause of bias in oncology trials. Nat Rev Clin Oncol. 2020;17:327–28.
8. Burke JM. Addressing informative censoring bias in clinical trials. Target Oncol. 2024. https://www.targetedonc.com/view/addressing-informative-censoring-bias-in-clinical-trials . Accessed 20 Jan 2026.
9. Van de Donk NWCJ, et al. Long-term follow-up from MajesTEC-1 of teclistamab, a B-cell maturation antigen (BCMA) x CD3 bispecific antibody, in patients with relapsed/refractory multiple myeloma (RRMM). J Clin Oncol. 2023;41(16 suppl):Abstract 8011.
10. Prince HM, et al. MagnetisMM-3: long-term update and efficacy and safety of less frequent dosing of elranatamab in patients with relapsed or refractory multiple myeloma. Presented at the 66th ASH Annual Meeting and Exposition; 2024 Dec 7–10; San Diego, CA, USA. Abstract P4738.
11. Encinas C, Lozano V, Hlavacek P, Llinares J, Toribio-Castelló S, Carcedo D, Asís-Montalt J, Martínez-López J. Cost-effectiveness of elranatamab versus current therapies for the management of patients with triple-class exposed, relapsed and refractory multiple myeloma, including other bispecific and physician's choice of treatment in Spain. Oncol Ther. 2025;13(2):381–395. 10.1007/s40487-025-00333-7
12. Gordan LN, Bensimon AG, Mu F, et al. Cost per responder for teclistamab and elranatamab in relapsed or refractory multiple myeloma in the United States. J Med Econ. 2025;28(1):910–20.
13. Reece D, Diels J, Van Sanden S, et al. Matching-adjusted indirect comparison of talquetamab vs selinexor-dexamethasone and vs belantamab mafodotin in patients with relapsed/refractory multiple myeloma. Curr Med Res Opin. 2024;40(10):1769–83.
14. Lee HC, Bumma N, Richter J, et al. Indirect comparison of linvoseltamab versus teclistamab for the treatment of triple-class exposed relapsed/refractory multiple myeloma. Clin Lymphoma Myeloma Leuk. 2025. 10.1016/j.clml.2025.06.014.
15. Martin T, Usmani SZ, Schecter JM, et al. Updated results from a matching-adjusted indirect comparison of efficacy outcomes for ciltacabtagene autoleucel in CARTITUDE-1 versus idecabtagene vicleucel in KarMMa for the treatment of patients with relapsed or refractory multiple myeloma. Curr Med Res Opin. 2023;39(1):81–9.
16. Merino M, Kasamon Y, Theoret M, Pazdur R, Kluetz P, Gormley N. Irreconcilable differences: the divorce between response rates, progression-free survival, and overall survival. J Clin Oncol. 2023;41(15):2706–12 (Erratum in: J Clin Oncol, 2023; 41(27): p. 4449).
17. Mol I, Hu Y, LeBlanc TW, et al. Matching-adjusted indirect comparison of elranatamab versus teclistamab in patients with triple-class exposed/refractory multiple myeloma: updated results. J Blood Med. 2025;16:233–9.
18. National Institute for Health and Care Excellence. Isatuximab with pomalidomide and dexamethasone for treating relapsed or refractory multiple myeloma: TA658. 2020.
19. National Institute for Health and Care Excellence. Elranatamab for treating relapsed and refractory multiple myeloma after 3 or more treatments: TA1023. 2024.
20. European Medicines Agency. Elrexfio: summary of product characteristics [November 2025]. https://www.ema.europa.eu/en/documents/product-information/elrexfio-epar-product-information_en.pdf . Accessed 20 Jan 2026.
21. European Medicines Agency. Tecvayli: summary of product characteristics. [November 2025]. https://www.ema.europa.eu/en/documents/product-information/tecvayli-epar-product-information_en.pdf . Accessed 20 Jan 2026.
22. Usmani SZ, Karlin L, Benboubker L, et al. Durability of responses with biweekly dosing of teclistamab in patients with relapsed/refractory multiple myeloma achieving a clinical response in the majesTEC-1 study. J Clin Oncol . 2023;41(16):8034. 10.1200/JCO.2023.41.16_suppl.8034.