Talquetamab for treating relapsed and refractory multiple myeloma after 3 or more treatments

Multiple myeloma is an incurable and progressive condition that has a substantial impact on survival and quality of life. Complications of multiple myeloma can be significant, debilitating and painful. The relapsing–remitting nature of the condition has a huge psychological impact, because people are aware that treatment options and life expectancy reduce with each relapse. The patient expert at the first committee meeting explained the negative impact of frequent relapses of multiple myeloma. They spoke about the anxiety around the availability of treatment options after each relapse. They also spoke about the anxiety around having to take corticosteroids, which are associated with a considerable negative impact on quality of life. The committee recognised the substantial impact that multiple myeloma has on survival and quality of life. It acknowledged the unmet need for more treatments that are effective for people who have already had several treatments.

According to the marketing authorisation, people having talquetamab must have had 3 or more treatments including a proteasome inhibitor, an immunomodulatory drug and an anti‑CD38 monoclonal antibody. The condition must have also progressed on the last treatment. The company submission provided a comparison with teclistamab, a treatment at fourth line and beyond. Clinical advice to the EAG was that teclistamab is the most relevant comparator for this evaluation, given the company's positioning after 3 or more lines of treatment. The clinical experts at the first committee meeting agreed that teclistamab is the most frequently used fourth-line treatment option for relapsed and refractory multiple myeloma and a relevant comparator to talquetamab. But the NHS England Cancer Drugs Fund clinical lead explained that most people have pomalidomide plus dexamethasone and daratumumab at fourth line, with teclistamab mostly being used at fifth line. They also noted that the treatment pathway was rapidly evolving and that the data informing treatment use may change as newer treatments are implemented into the pathway. The committee noted that teclistamab is an available option after 3 or more lines of treatment (see NICE's technology appraisal guidance on teclistamab for treating relapsed and refractory multiple myeloma after 3 or more treatments [TA1015]). It also noted that elranatamab is available for treating relapsed and refractory multiple myeloma after 3 or more lines of treatment. But elranatamab is only recommended for use within the Cancer Drugs Fund (see NICE's technology appraisal guidance on elranatamab for treating relapsed and refractory multiple myeloma after 3 or more treatments). So, it cannot be considered a comparator to talquetamab in this appraisal. The committee concluded that teclistamab is the most relevant comparator to talquetamab for this evaluation.

The key clinical-effectiveness evidence for talquetamab came from the MonumenTAL-1 trial. This is an ongoing phase 1 and 2, single-arm, open-label, multicentre study. People in the trial have triple-class-exposed relapsed and refractory multiple myeloma that is refractory to at least 1 proteasome inhibitor, 1 immunomodulatory drug and 1 anti‑CD38 monoclonal antibody. The company presented data for cohort C only (n=154). The company stated, based on its clinical advice, that most (90%) of people in UK clinical practice will have talquetamab once every 2 weeks (cohort C) with 10% having it once weekly (cohort A; n=143). It presented data from the September 2024 data cut, with a median follow up of 31.2 months. The overall response rate was 69.5%. Median overall survival was not estimable and median progression-free survival was 11.2 months. The committee noted that the marketing authorisation allows talquetamab to be used once weekly or once every 2 weeks. So, it questioned whether the clinical-effectiveness evidence from cohort A was appropriately captured in the company's submission. It also noted that there was a difference in clinical effectiveness with a reduced median progression-free survival for cohort A (7.5 months) than for cohort C. The company said that the once every 2 weeks dosage in cohort C is more convenient for people and would likely increase the cost effectiveness of talquetamab. A clinical expert said that about 90% of people have the once every 2 weeks dosage in NHS clinical practice because of its convenience over the once weekly dosing regimen. So, it may be appropriate to model resource use according to 90% of dosing regimens being once every 2 weeks and 10% being once weekly. But the committee was unclear why the clinical efficacy would differ markedly between these 2 dosing regimens. So, the committee said that it would also like to see clinical and cost-effectiveness analyses on pooled data.

In response to consultation, the company provided pooled data for cohorts A and C. This meant that the committee now had clinical-effectiveness results for:

• cohort C alone

• all pooled cohort‑A and cohort‑C data

• a weighted sample of the pooled data, of 90% cohort C and 10% cohort A.
  
  The committee noted that the company had only provided cost-effectiveness results for cohort C alone and for the pooled data using 90% cohort C and 10% cohort A. The company had not provided cost-effectiveness results for all pooled cohort‑A and cohort‑C data. The committee considered which was the most appropriate cohort for decision making. It noted the reduced clinical effectiveness in cohort A and that using cohort C alone reduced the sample size in the evidence base. The committee questioned the reasons for the lower clinical effectiveness in cohort A compared with cohort C. The clinical experts at the second committee meeting explained that there were more dose delays and modifications in cohort A. They also said that the once every 2 weeks dosing in cohort C causes less T-cell exhaustion and allows for more T-cell recovery. They explained the patient preference for once every 2 weeks dosing in clinical practice and said that this is also likely to have reduced healthcare-resource impact. The company also explained that they did not provide cost-effectiveness results using all pooled cohort‑A and cohort‑C data because only cohort C is relevant to clinical practice. It said that this dosing provides the best clinical outcomes for people with multiple myeloma. The committee agreed that most people would have the treatment once every 2 weeks in clinical practice, but it decided that the more frequent dosing in cohort A was relevant for informing the overall treatment effect. This was because there is uncertainty that the mechanistic explanations for T-cell recovery in the intervals would result in the benefits seen in cohort C. So, the committee concluded that all the available evidence should be pooled.

The clinical-effectiveness evidence for teclistamab came from MajesTEC‑1, an ongoing phase 1 and 2, single-arm, open-label, multicentre trial. People in the trial have triple-class-exposed relapsed or refractory multiple myeloma that is refractory to at least 1 proteasome inhibitor, 1 immunomodulatory drug and 1 anti‑CD38 monoclonal antibody. The company presented data from the phase 1 part 2 cohort and the phase 2 cohort A of the study (n=165). It presented data from the August 2023 data cut, with a median follow up of 30.4 months. The overall response rate was 63%. Median overall survival was 22.2 months and median progression-free survival was 11.4 months.

At the first committee meeting, the EAG explored the overall survival and progression-free survival for talquetamab and teclistamab from MonumenTAL‑1 and MajesTEC‑1 by comparing these to published real-world evidence from Europe. The real-world studies included:

• one talquetamab study from Germany (Frenking et al. 2025; n=131)

• one teclistamab study from Germany (Riedhammer et al. 2024; n=122) and one from France (Perrot et al. 2025; n=312).
  
  The EAG advised that these real-world studies showed superior overall survival and consistent progression-free survival compared with MajesTEC‑1 for teclistamab and an inferior overall survival and progression-free survival compared with MonumenTAL‑1 for talquetamab. The EAG explained that these studies likely included populations similar to UK NHS clinical practice, with its clinical advice suggesting that real-world evidence may better represent expected outcomes in UK practice.
  
  In response to consultation, the company presented an indirect treatment comparison of talquetamab and teclistamab using 2 retrospective, non-interventional real-world studies, REALiTEC (teclistamab; n=113, median age=66 years) and REALiTAL (talquetamab; n=93, median age 65 years). These evaluated real-world outcomes in people with triple-class exposed relapsed or refractory multiple myeloma and included people from the UK. The company presented unadjusted overall-survival hazard ratios with and without subsequent-treatment adjustment. They also presented results using multivariable regression to control for differences in baseline characteristics and applied subsequent-treatment adjustment. The results of this analysis showed that talquetamab had an improvement in overall survival compared with teclistamab in all the scenarios. The hazard ratio for overall survival is considered confidential by the company and cannot be reported here. The overall-survival hazard ratios in these analyses were all higher than the company's base case. The company explained that these analyses demonstrate that the results from MonumenTAL‑1 and MajesTEC‑1 are reflective of real-world clinical practice. The EAG noted that real-world evidence provides alternative overall-survival estimates to clinical trial evidence. But it said that all of the data is associated with limitations and so is unlikely to reduce the uncertainty in the overall-survival hazard ratio for talquetamab compared with teclistamab.

The committee recalled the high level of uncertainty associated with the overall-survival hazard ratio for talquetamab compared with teclistamab based on the company's indirect treatment comparison using MonumenTAL‑1 and MajesTEC‑1 data. It specifically recalled the:

• early separation of the talquetamab and teclistamab overall-survival curves (see section 3.6)

• lack of correlation between progression-free survival and overall survival, particularly the size of the overall-survival benefit (see section 3.7)

• impact of COVID‑19 not being fully accounted for in the indirect treatment comparison (see section 3.8).
  
  So, the committee decided that the overall-survival hazard ratio from the company's indirect treatment comparison was uncertain but likely overestimates the benefit of talquetamab. The committee also noted that the company's updated base-case hazard ratio did not include all pooled cohort-A and cohort-C data from MonumenTAL‑1, which increased the hazard ratio. The committee also recalled that the company's indirect comparison using real-world studies resulted in a higher hazard ratio than in the base-case analysis (see section 3.9). The committee noted that both REALiTEC and REALiTAL were done after the COVID‑19 pandemic and that the results of the indirect treatment comparison using these studies were not impacted by COVID‑19. So, the committee decided that this analysis reduced some of the uncertainty in the overall-survival hazard ratio. The committee noted that the overall-survival hazard ratio was likely to be higher than in the company's base case. It concluded that the broad range of overall-survival hazard ratio estimates from different analyses had helped to characterise the uncertainty. So, although there was still uncertainty in the estimates, the analyses helped the committee assess the impact of the different issues.

The company used a partitioned survival model with 3 health states: pre-progression, post-progression and death. The cycle length was 1 week and the time horizon was 40 years. The EAG was broadly satisfied with the company's model structure. It noted that the company discounted costs and quality-adjusted life years (QALYs) in the first year of the model and applied a half-cycle correction to account for mid-cycle transitions in the model. The EAG advised that discounting for costs and benefits in the first year was inappropriate. It also thought that the cycle length was not long enough to apply half-cycle corrections in the model. So, the EAG discounted costs and QALYs from year 2 onward and excluded the half-cycle correction in its base-case model. The committee noted that the company's model was similar to previous models used for multiple myeloma. At the first committee meeting, the committee decided the EAG's approach to discounting from year 2 onward and the exclusion of the half-cycle correction was appropriate. In response to consultation, the company updated its base-case economic model to include discounting for costs and benefits only from year 2 onward and exclude the half-cycle correction. The committee concluded that overall, the company's model structure was appropriate for decision making.

To estimate long-term overall survival, progression-free survival and time to treatment discontinuation beyond the trial follow-up period, the company fitted parametric models to the MajesTEC‑1 individual patient data for teclistamab. The company selected a log-normal model for all 3 outcomes and calibrated it to the clinical-expert estimates for these outcomes at 10 and 15 years in line with TA1015. The hazard ratios from the indirect treatment comparison for each outcome were then applied to the calibrated log-normal models for teclistamab. This was done to predict long-term overall survival, progression-free survival and time to treatment discontinuation for talquetamab. Long-term overall survival for talquetamab was then capped to the general population mortality. With differing hazard profiles for overall survival, progression-free survival and time to treatment discontinuation, the EAG noted that the same parametric model did not need to be selected for all 3 outcomes. It advised that extrapolating using the log-normal model produces implausible long-term overall-survival estimates. It said that in order to produce plausible overall-survival curves, the company had to force it into plausibility using clinical-expert estimates. The EAG also said that it did not think that the proportional-hazards assumption was compatible with the selected log-normal models. So, the EAG selected the Weibull model which does not need to be adjusted to produce clinically plausible estimates. The committee questioned the need to calibrate the log-normal models when there are other parametric models that would not need calibration. The company explained that selecting log-normal models for all 3 outcomes and calibrating them to meet clinical-expert estimates was consistent with the approach accepted by the committee in TA1015. The committee noted that individual patient data was available for both treatments. It said that using the IPTW method in the indirect treatment comparison would allow independent models to be fitted that would not need an assumption of proportional hazards. The committee thought that the EAG's approach using the uncalibrated Weibull model was more appropriate because it had not needed calibration. But it decided that it would prefer to see independent curves modelled rather than applying the indirect treatment comparison hazard ratios. At the first committee meeting, the committee concluded that it would like to see the long-term overall survival, progression-free survival and time to treatment discontinuation extrapolations modelled independently for both teclistamab and talquetamab without calibrating to the clinical-expert estimates.

In response to consultation, the company updated its economic model. It independently modelled the long-term extrapolations of teclistamab and aligned them with the committee-preferred EAG approach by selecting the uncalibrated Weibull model across overall survival, progression-free survival and time to treatment discontinuation. But the company maintained that the hazard-ratio approach is more appropriate to derive these outcomes in the talquetamab arm instead of independently modelling them. So, it provided scenario analyses using independent modelling of long-term outcomes in the talquetamab arm using various parametric distributions. The company explained that the proportional-hazards assumption is valid and is associated with less uncertainty than independent modelling because it directly incorporates the results of the indirect treatment comparison. The EAG agreed with the company's hazard-ratio approach to modelling talquetamab outcomes. But it advised that the company's updated overall-survival hazard ratio used to model the long-term overall-survival outcomes was still associated with very high uncertainty (see sections 3.5 to 3.8). The committee acknowledged the uncertainty associated with the overall-survival hazard ratio from the indirect treatment comparison. But the committee concluded that independently modelling the long-term extrapolations in the teclistamab arm using the uncalibrated Weibull model was appropriate. It said that using the hazard-ratio approach to model long-term extrapolations in the talquetamab arm across overall survival, progression-free survival and time to treatment discontinuation was also appropriate.

Some people in MonumenTAL‑1 and MajesTEC‑1 had subsequent treatments not routinely available in the NHS. To account for this, the company adjusted the overall-survival hazard ratio by removing the effects of these treatments using the 2‑stage approach as per NICE Decision Support Unit's Technical Support Document 16. In its base case the company included subsequent teclistamab after talquetamab but excluded subsequent talquetamab after teclistamab. The EAG thought that the company's base-case approach was not equitable and did not allow for a fair comparison of talquetamab with teclistamab. The EAG explained that this approach affects both costs and QALYs and favours the talquetamab arm. To allow for a fair comparison, in its base case, the EAG included both talquetamab and teclistamab as subsequent treatments after initial treatment. The clinical experts at the committee meeting noted that if talquetamab is recommended, some people will start talquetamab or teclistamab and then have the other treatment after disease progression. Whether to have treatment with talquetamab or teclistamab first would be a decision between a person with multiple myeloma and their healthcare practitioner. The committee noted the lack of detail in the company submission on how it adjusted for subsequent treatments not available in the NHS. It also noted that the company only used the Weibull model when adjusting for subsequent treatments using the 2-stage adjustment approach. The committee noted that other parametric distributions could have been explored in scenario analyses. At the first committee meeting, the committee concluded that:

• the costs and benefits of both subsequent talquetamab and subsequent teclistamab after initial treatment should be modelled

• it would like the company to provide a more detailed explanation of overall methods used to adjust for subsequent treatments not available in the NHS, and

• it would like to see analyses using parametric models, other than the Weibull model, when adjusting for subsequent treatments using the 2‑stage adjustment approach.
  
  In response to consultation, the company included costs and benefits of both subsequent talquetamab and subsequent teclistamab in its updated economic model. It also provided additional methodological detail of the methods used to adjust for subsequent treatments not available in NHS clinical practice. And it provided scenario analyses using alternative parametric distributions for its 2‑stage adjustment. The EAG agreed that the company's exploration of alternative parametric distributions and the proportional-hazards Weibull model were visually consistent with the company's original accelerated failure time Weibull model. It thought that this reduced uncertainty with the adjustment. The committee concluded that the company's accelerated failure time Weibull model was appropriate to adjust for subsequent treatments using the 2‑stage adjustment approach.

People in MonumenTAL‑1 and MajesTEC‑1 could have intravenous immunoglobulin (IVIg) to prevent or treat infections. In the company's base-case analysis, IVIg use was modelled as a one-off cost in the first cycle in the talquetamab and teclistamab arms. This was in line with the observed IVIg use in the respective clinical trials. The proportion of people in the trials having IVIg is considered confidential and cannot be reported here. The company assumed 9 doses of IVIg in both treatment arms, which aligns with TA1015. The company noted that because of the novel G protein-coupled receptor class 5D target, people having talquetamab have less severe infections compared with teclistamab and so need less IVIg. The EAG thought that modelling IVIg use as a one-off cost underestimates IVIg costs in favour of talquetamab. It said that this is because it does not account for IVIg use by people having subsequent talquetamab and teclistamab after disease progression on the initial treatment. So, the EAG included IVIg costs associated with subsequent talquetamab and teclistamab in its base case. It based the proportion of IVIg use for subsequent talquetamab and subsequent teclistamab on the observed IVIg use in the respective clinical trials. These proportions are considered confidential and cannot be reported here. The EAG assumed 6 doses of IVIg for subsequent talquetamab and 9 doses of IVIg for subsequent teclistamab. The clinical experts said that there is a considerable difference in IVIg use between people having talquetamab and teclistamab in clinical practice. This is likely because there are fewer infections associated with talquetamab because of its novel target compared with teclistamab. The experts agreed that the proportions used in the company's base case are broadly reflective of NHS practice. The clinical experts also agreed that people having subsequent talquetamab and teclistamab would also need IVIg. At the first committee meeting, the committee decided that the proportion of IVIg use in the talquetamab arm based on MonumenTAL‑1 and in the teclistamab arm based on MajesTEC‑1 was appropriate. It also decided that IVIg use should be modelled for subsequent talquetamab and subsequent teclistamab.

In response to consultation, the company updated its economic model using the MonumenTAL‑1 and MajesTEC‑1 trials to inform the proportions of people having IVIg with talquetamab and teclistamab. The company also updated its economic model to include IVIg use for subsequent talquetamab and teclistamab, informed by MonumenTAL‑1 and MajesTEC‑1. The committee concluded that the company's updated IVIg modelling, with the proportions having IVIg in the talquetamab arm being based on MonumenTAL‑1 data and in the teclistamab arm being based on MajesTEC‑1 data, was appropriate. It also concluded that basing IVIg use for subsequent talquetamab on MonumenTAL‑1 data and basing IVIg for subsequent teclistamab on MajesTEC‑1 data was appropriate.

In its base case, the company applied a one-off utility decrement in the first cycle of the model for grade 3 or higher treatment-related adverse events. These events happened for at least 5% of people who had talquetamab or teclistamab. A higher utility decrement was applied in the teclistamab arm compared with the talquetamab arm because teclistamab was associated with more grade 3 or higher treatment-related adverse events. The EAG advised that including a utility decrement for adverse events may lead to double counting. This is because the effect of adverse events on health-related quality of life was likely captured in the patient-reported outcomes data collected in the trial, and this data was used to estimate the health-state utility values in the model. So, the EAG excluded modelling the utility decrement for adverse events separately to the health-state utility values. The patient experts explained that talquetamab is associated with altered taste in the first few months of starting it, which may result in subsequent weight loss. But the patient and clinical experts explained that these side effects are manageable, with weight loss being managed with dietitian support. A patient expert also explained that other treatments have more profound side effects, particularly fatigue and gastrointestinal side effects, which impact day-to-day life. The committee noted that altered taste and subsequent weight loss is likely to impact quality of life. But it noted that this had not been accounted for in the utility decrement applied for adverse events in the talquetamab arm because changes in taste are not classed as a grade‑3 or 4 adverse event. The committee said that it would like the company to provide more explicit modelling of adverse-event disutilities and include the impact of altered taste and associated weight loss in the adverse-event disutility for talquetamab.

In response to consultation, the company provided further information on its modelling of adverse-event disutilities. It also included the impact of altered taste and associated weight loss in its updated economic model. The committee concluded that the company's updated adverse-event disutilities modelling was appropriate for decision making.

NICE's manual on health technology evaluations notes that, above a most plausible incremental cost-effectiveness ratio (ICER) of £20,000 per QALY gained, judgements about the acceptability of a technology as an effective use of NHS resources will take into account the degree of certainty around the ICER. The committee will be more cautious about recommending a technology if it is less certain about the ICERs presented. But it will also take into account other aspects including uncaptured health benefits. The committee noted that there remained a high level of uncertainty, specifically the:

• reduced clinical-effectiveness outcomes for cohort A in MonumenTAL‑1 (see section 3.3)

• impact of COVID‑19 on the clinical evidence (see section 3.8)

• indirect treatment comparison hazard ratio for overall survival (see sections 3.5 to 3.10), including the:

  • early separation of the talquetamab and teclistamab overall-survival curves (see section 3.6)

  • lack of correlation between progression-free survival and overall survival, particularly the size of the overall-survival benefit (see section 3.7)

  • impact of COVID‑19 not being fully accounted for in the indirect treatment comparison (see section 3.8).
    
    The committee thought that there was considerable remaining uncertainty, especially in the magnitude of overall-survival benefit. But it decided that the introduction of real-world evidence and the company's updated modelling of subsequent treatments, IVIg use and adverse-event disutilities had reduced some of this uncertainty. So, the committee concluded that an acceptable ICER would be around the middle of the range NICE considers a cost-effective use of NHS resources (£20,000 to £30,000 per QALY gained).

Because of confidential commercial arrangements for talquetamab, teclistamab and some of the subsequent treatments, the exact cost-effectiveness results are confidential and cannot be reported here.

The committee's preferred assumptions included:

• pooling data from cohort A and cohort C in MonumenTAL‑1 to inform the clinical and cost effectiveness of talquetamab (see section 3.3)

• a range of overall-survival hazard ratios for talquetamab compared with teclistamab that were higher than the results of the company's indirect treatment comparison (see section 3.10)

• discounting for costs and QALYs from the second year of the economic model (see section 3.11)

• excluding half-cycle correction of costs and QALYs in the economic model (see section 3.11)

• independently modelling the long-term extrapolations in the teclistamab arm using the uncalibrated Weibull model and using the hazard-ratio approach to model long-term extrapolations in the talquetamab arm across overall survival, progression-free survival and time to treatment discontinuation (see section 3.12)

• including in the model the costs and benefits of using subsequent talquetamab and subsequent teclistamab after initial treatment (see section 3.13)

• using the accelerated failure time Weibull model to adjust for subsequent treatments using the 2-stage adjustment approach (see section 3.13)

• the proportions having IVIg use in the talquetamab arm being based on MonumenTAL-1 data and in the teclistamab arm being based on MajesTEC-1 data, including subsequent talquetamab and teclistamab (see section 3.14)

• including the impact of altered taste and associated weight loss in the adverse-event disutility for talquetamab (see section 3.15).
  
  The exact cost-effectiveness estimates using the committee's preferred assumptions cannot be reported here because there are confidential discounts for talquetamab, teclistamab and some of the subsequent treatments.

A professional-organisation submission noted that there is an increased incidence of multiple myeloma and development of more severe disease in African and African‑Caribbean people. People with more severe multiple myeloma are more likely to need additional lines of treatment because it is associated with an increased risk of relapse with earlier lines of treatment. Race is a protected characteristic under the Equality Act 2010. A patient organisation also said that as with all treatments the costs incurred by hospital visits and time off work will have a more significant impact on people with lower incomes. The committee decided that its recommendations do not restrict access to treatment for some people over others. So, the committee agreed that these were not potential equality issues in this evaluation. The committee did not identify any other equality issues.

The committee noted that the clinical-effectiveness evidence for talquetamab was uncertain, including the indirect comparison results for overall survival compared with teclistamab. But the cost-effectiveness estimates using its preferred modelling assumptions are within the range NICE considers a cost-effective use of NHS resources. So, talquetamab can be used.