Cost-Effectiveness of Elranatamab Versus Teclistamab for the Management of Patients with Triple-Class Exposed Relapsed/Refractory Multiple Myeloma in Italy

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper is focusing on very important question: how should we react and choose between teclistamab and elranatamab. All countries do not have possibility to use both Bites.

The paper is well written, I only have some small comment and questions to authors.

Because elranatamab is showing more beneficial compared to teclistamab even if TTD is 5.55 vs 8.5 mo, respectively, does this mean that patients had continued their response after elra discontinuation?

46.1% of teclistamab patients were assumed to be able to switch dosing after 11.3 mo; how was this deduced? Is the finding real or only a supposition, if so, then it could have influence on the results.

Table 1. Total costs ; is this per year or for life-time?

Figure 2. In my edition the colours are not same in the image caption than in the figure.

Roe 309, there is some error intext: anemia for ....?

Figure 3. This is a bit hard to understand. Could it help if you would add the results of elranatamab as comparison?

Row 392  these is an error in word "teclistamab"

Table SI what do you mean by unit costs, please clarify

Table S3 What does cost "unit16" mean ?

Table S6  The word " Neurotossicita" , what is this?

Table S7 I can not understand following: neurotoxicity gr 3-4 , cases 0% but still some costs, Same concerns CRS

Based on this paper can we draw a conclusion that we should give patients only elranatamab instead of teclistamab?

Comments on the Quality of English Language

I have added comments on Language errors on Comments box.

Author Response

This paper is focusing on very important question: how should we react and choose between teclistamab and elranatamab. All countries do not have possibility to use both Bites.

The paper is well written, I only have some small comment and questions to authors.

 

Because elranatamab is showing more beneficial compared to teclistamab even if TTD is 5.55 vs 8.5 mo, respectively, does this mean that patients had continued their response after elra discontinuation?

• We thank the reviewer for this comment.

A shorter median TTD observed with elranatamab (5.55 months vs 8.50 months) does not imply an earlier loss of response, nor does it indicate that patients stopped responding upon treatment discontinuation. Rather, it reflects earlier treatment discontinuation while patients remained progression-free for a longer period, as demonstrated by the superior PFS and OS curves. In the partitioned survival model applied in the analysis, patients may remain in the progression-free state while off treatment, and clinical benefit (PFS/OS) is not assumed to end at the time of treatment discontinuation. Within this framework, TTD is explicitly modeled as a treatment exposure parameter to capture time on therapy within the progression-free state and is not intended as a surrogate measure of treatment efficacy or durability of response.

This interpretation is consistent with findings presented at the most recent ASH Annual Meeting on a subgroup analysis from MagnetisMM-3 (Lesokhin et al., Abstract 2269, “Prolonged elranatamab treatment interruption in patients with relapsed or refractory multiple myeloma (RRMM) is feasible: A retrospective analysis from MagnetisMM-3”). This post hoc analysis evaluated patients who experienced a prolonged interruption of elranatamab or permanently discontinued treatment while maintaining their response for more than 6 months. The authors reported that the majority of patients maintained their clinical response despite prolonged treatment interruption, demonstrating the durability of elranatamab responses. Furthermore, these data support the feasibility of dose interruptions to manage adverse events and allow treatment breaks without compromising the efficacy of elranatamab. This aspect has been incorporated into the manuscript.

46.1% of teclistamab patients were assumed to be able to switch dosing after 11.3 mo; how was this deduced? Is the finding real or only a supposition, if so, then it could have influence on the results.

• The values are not suppositions, and they are referenced by sources 20 and 26 in the previous version of the manuscript. Based on your comment, we conducted an additional check of the references and confirmed that source 26 (previous version) reported an approximate value of 12 months.

To improve accuracy, we searched for a more precise estimate in the literature and identified an abstract presented at the ASCO Annual Meeting 2023, which we have now included in the reference list. This source reports a value of 11.1 months.

To strengthen the robustness of the analysis, we have updated the manuscript using this revised value.

Table 1. Total costs ; is this per year or for life-time?

• Thank you for your comment. The 25-year time horizon and the associated costs can be considered representative of a lifetime horizon.

Figure 2. In my edition the colours are not same in the image caption than in the figure.

• Thank you for your comment. We have revised the figure: both outcomes are now represented in green, using two different shades to distinguish PFS/PPS and death, in order to improve clarity.

Roe 309, there is some error intext: anemia for ....?

• Thank you for your comment. We have corrected the typographical error.

Figure 3. This is a bit hard to understand. Could it help if you would add the results of elranatamab as comparison?

• Thank you for your suggestion. We have revised the figure to explicitly indicate that it presents a comparison between elranatamab and teclistamab. We have also added axis labels to make this comparison clearer.

Row 392  these is an error in word "teclistamab"

• Thank you for your comment. We have corrected the typographical error.

Table SI what do you mean by unit costs, please clarify

• This refers to the cost per cycle. To facilitate understanding, we have revised the table accordingly. Thank you for your comment.

Table S3 What does cost "unit16" mean ?

• Thank you for your comment. This was a typographical error. We have removed the number 16 and, for completeness, added the reference to the source (already cited in the manuscript).

Table S6  The word " Neurotossicita" , what is this?

• Thank you for your comment. This was a typographical error, as the term had been inadvertently reported in Italian. We have corrected it to English (“neurotoxicity”).

Table S7 I can not understand following: neurotoxicity gr 3-4 , cases 0% but still some costs, Same concerns CRS

• Thank you for your comment. We acknowledge that the table, as previously presented, may have been confusing. Since no grade 3 or 4 neurotoxicity or CRS events were observed, these do not impact the overall cost. The reported cost refers to the cost per episode. Unit cost estimates were derived from the Tariffario Nazionale delle Prestazioni Ambulatoriali and DRG tariffs. We have revised the table accordingly.

Based on this paper can we draw a conclusion that we should give patients only elranatamab instead of teclistamab?

• Thank you for your comment.

This cost-effectiveness analysis shows that elranatamab represents a financially favorable option compared with teclistamab for adult patients with TCE/RRMM in Italy and provides pharmacoeconomic evidence.

Therefore, considering the heterogeneity of treated patients, it would not be appropriate to draw an absolute clinical conclusion that elranatamab should be used instead of teclistamab in all cases.

Reviewer 2 Report

Comments and Suggestions for Authors

An original article “Cost‑Effectiveness of Elranatamab Versus Teclistamab for the Management of Patients with Triple‑Class Exposed, Relapsed and Refractory Multiple Myeloma in Italy” reported that the cost-effectiveness balance of elranatamab was better than those of teclistamab. This result was consistent with previous analysis and so seems to be reasonable. However, the clinical outcome of elranatamab in IFM experiences was much poorer than those in the Megestic-MM3 trial, suggesting that the authors should pay a tension to misleading from statistical analysis using the data from pivotal studies. Additionally, there were several issues in this manuscript.

 

1. In MAIC, what parameters were matched between ELRA and TEC groups? The authors should describe clearly.
2. Regarding figure1, why was TTD in TEC longer than that in ELRA? Additionally, why was difference between PFS and TTD in ELRA much higher than that in TEC? Is it affected by the parameters in MAIC? Figure1 should have a potential risk to mislead. The author should explain those clearly.
3. Subsequent treatments could contribute the cost directly. The author should describe the disposition for subsequent treatment in ELRA and TEC groups, clearly.

Author Response

An original article “Cost‑Effectiveness of Elranatamab Versus Teclistamab for the Management of Patients with Triple‑Class Exposed, Relapsed and Refractory Multiple Myeloma in Italy” reported that the cost-effectiveness balance of elranatamab was better than those of teclistamab. This result was consistent with previous analysis and so seems to be reasonable. However, the clinical outcome of elranatamab in IFM experiences was much poorer than those in the Megestic-MM3 trial, suggesting that the authors should pay a tension to misleading from statistical analysis using the data from pivotal studies. Additionally, there were several issues in this manuscript.

• We thank the reviewer for this comment.

The IFM (Intergroupe Francophone du Myélome) data, largely derived from the French compassionate use program, provide a valuable perspective on treatment performance in routine clinical practice. However, these cohorts generally include patients who are more heavily pretreated and clinically more complex than those enrolled in the MagnetisMM-3 registrational study, with a higher prevalence of adverse prognostic factors. Such differences in baseline characteristics may reasonably contribute to less favorable outcomes observed in some real-world settings and should be carefully considered when making cross-study comparisons.

In the present analysis, pivotal trial evidence was used as the primary source of comparative effectiveness, consistent with standard health technology assessment (HTA) methodology. In the absence of head-to-head randomized data or appropriately adjusted real-world comparative analyses, registrational trials remain the most robust and internally valid source for estimating relative treatment effects. Our objective was not to predict effectiveness across all real-world contexts, but to estimate the relative cost-effectiveness of elranatamab versus teclistamab using the best available comparative evidence at the time of model development.

Importantly, reliance on pivotal trial data does not inherently favor elranatamab. Any potential overestimation of effectiveness associated with trial conditions would be expected to apply to both treatments. Moreover, the robustness of our findings was confirmed through extensive deterministic and probabilistic sensitivity analyses.

It is also worth noting that recent real-world evidence presented at the latest ASH Annual Meeting provides a more nuanced picture. While some datasets reported outcomes inferior to those of MagnetisMM-3, others were consistent with the registrational trial. In particular, the international analysis by Popat et al. (ASH 2025, Abstract 4588: “Real-world efficacy and safety of elranatamab, a BCMA bispecific antibody for patients with relapsed and refractory multiple myeloma: An International Myeloma Working Group Immunotherapy Database analysis”) reported a response rate of 67% and a 12-month PFS of 67% in BCMA-naïve patients—results comparable to MagnetisMM-3—despite nearly half of the patients not meeting the original trial eligibility criteria. This large international cohort, which included European centers, suggests that real-world outcomes may approximate those observed in the pivotal study even in a less selected population.

Taken together, these considerations indicate that, while differences between clinical trial and real-world populations must be acknowledged, the use of registrational data for comparative economic modeling remains methodologically appropriate, and currently available real-world evidence does not consistently demonstrate substantially poorer effectiveness of elranatamab relative to trial results. This aspect has been incorporated into the manuscript.

In MAIC, what parameters were matched between ELRA and TEC groups? The authors should describe clearly.

• Thank you for your comment.

The parameters included age, sex (for OS only), time since initial diagnosis, ISS stage, cytogenetic risk, extramedullary disease, number of prior lines of therapy, ECOG performance status, creatinine clearance, refractory status (penta-exposed; penta-refractory), and type of myeloma. This clarification has been added to the manuscript.

Regarding figure1, why was TTD in TEC longer than that in ELRA? Additionally, why was difference between PFS and TTD in ELRA much higher than that in TEC? Is it affected by the parameters in MAIC? Figure 1 should have a potential risk to mislead. The author should explain those clearly.

We thank the reviewer for this comment. Reviewer 1 also raised a similar point; therefore, we provide below a clarification, which has been shared in response to both reviewers’ comments.

A shorter median TTD observed with elranatamab (5.55 months vs 8.50 months) does not imply an earlier loss of response, nor does it indicate that patients stopped responding upon treatment discontinuation. Rather, it reflects earlier treatment discontinuation while patients remained progression-free for a longer period, as demonstrated by the superior PFS and OS curves. In the partitioned survival model applied in the analysis, patients may remain in the progression-free state while off treatment, and clinical benefit (PFS/OS) is not assumed to end at the time of treatment discontinuation. Within this framework, TTD is explicitly modeled as a treatment exposure parameter to capture time on therapy within the progression-free state and is not intended as a surrogate measure of treatment efficacy or durability of response.

This interpretation is consistent with findings presented at the most recent ASH Annual Meeting in a subgroup analysis from MagnetisMM-3 (Lesokhin et al., Abstract 2269, “Prolonged elranatamab treatment interruption in patients with relapsed or refractory multiple myeloma (RRMM) is feasible: A retrospective analysis from MagnetisMM-3”). This post hoc analysis evaluated patients who experienced a prolonged interruption of elranatamab or permanently discontinued treatment while maintaining their response for more than 6 months. The authors reported that the majority of patients maintained their clinical response despite prolonged treatment interruption, demonstrating the durability of elranatamab responses. Furthermore, these data support the feasibility of dose interruptions to manage adverse events and allow treatment breaks without compromising the efficacy of elranatamab.

Subsequent treatments could contribute the cost directly. The author should describe the disposition for subsequent treatment in ELRA and TEC groups, clearly.

• Subsequent treatment, in terms of both composition and costs, is identical for both treatments (elranatamab and teclistamab). We have clarified this aspect in the supplementary tables. Thank you for your comment.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript was revised well following the reviewer's comments, and so I have no additional comments.