AI-generated research brief — verify at source
DVRd Plus DR Maintenance Shows Superior Progression-Free Survival Over Current Transplant-Eligible NDMM Standards
In an indirect treatment comparison using individual patient-level data, the four-drug regimen daratumumab, bortezomib, lenalidomide, and dexamethasone followed by daratumumab-lenalidomide maintenance reduced the risk of disease progression or death by approximately 61% compared with daratumumab-VTd followed by lenalidomide maintenance (HR 0.39, 95% CI 0.26–0.59). These findings consistently favoured DVRd plus DR maintenance across all comparator regimens tested, though the analysis carries important methodological caveats inherent to unanchored indirect comparisons.
What Was Studied
This analysis asked whether the treatment strategy evaluated in the PERSEUS trial — DVRd induction and consolidation followed by DR maintenance — offers superior progression-free survival relative to regimens currently in wide clinical use for transplant-eligible patients with newly diagnosed multiple myeloma (TE NDMM). The question is clinically relevant because no head-to-head randomised trial has directly compared these approaches, leaving an evidence gap that indirect methods can partially address.
How It Was Studied
The analysis used individual patient-level data from two randomised trials: PERSEUS (NCT03710603), which evaluated DVRd, and CASSIOPEIA (NCT02541383), which evaluated DVTd and VTd with or without a second-randomisation maintenance phase. To model outcomes specifically for lenalidomide maintenance following DVTd or VTd induction — a combination not directly studied in CASSIOPEIA — data from the Myeloma XI trial (EudraCT 2009-010956-93) were incorporated. Inverse probability of treatment weighting was applied to harmonise baseline patient characteristics across trials, and inverse probability of censoring weighting addressed the second randomisation design embedded in CASSIOPEIA. This was an unanchored indirect treatment comparison, meaning no common treatment arm linked the two trials being compared.
What Was Observed
- DVRd plus DR maintenance was associated with roughly 61% lower risk of progression or death compared with DVTd followed by observation alone — a large and statistically precise advantage (HR 0.39, 95% CI 0.26–0.59). This represents the most direct comparison between the two daratumumab-containing quadruplet strategies in the absence of maintenance.
- Against VTd followed by observation, the progression-free survival benefit of DVRd plus DR maintenance was even more pronounced — approximately 83% lower risk of progression or death (HR 0.17, 95% CI 0.12–0.25), reflecting the substantial step-change that adding both a CD38 antibody and extended maintenance represents over older triplet induction.
- When lenalidomide maintenance was modelled for the comparator arms, DVRd plus DR maintenance still demonstrated meaningfully superior progression-free survival: about 38% lower risk versus DVTd plus lenalidomide maintenance (HR 0.62, 95% CI 0.41–0.92), and approximately 71% lower risk versus VTd plus lenalidomide maintenance (HR 0.29, 95% CI 0.18–0.43).
- Sensitivity analyses across all comparisons produced results consistent with the primary estimates, supporting the robustness of the directional findings even if the magnitude of effect carries uncertainty.
Why This Matters
Daratumumab-based quadruplet induction is increasingly established in TE NDMM, but the incremental value of pairing DVRd induction with prolonged DR maintenance — compared with DVTd or lenalidomide-maintained strategies — had not been quantified through comparative evidence. These results provide a structured evidence synthesis that may inform both clinical decision-making frameworks and health technology assessment processes where direct trial data are unavailable. The consistent advantage seen across all comparator arms strengthens the signal despite the methodological complexity.
How to Read This Result
Because this is an unanchored indirect comparison relying on data pooled across trials with different designs and patient populations, the possibility of residual confounding cannot be excluded, and the magnitude of the hazard ratio estimates should be interpreted with appropriate caution rather than treated as equivalent to randomised head-to-head evidence.
Limitations
The comparison was unanchored due to the absence of shared treatment arms across the included trials, which increases vulnerability to bias from unmeasured confounders. Modelling lenalidomide maintenance outcomes required external data from the Myeloma XI trial, introducing additional assumptions about cross-trial comparability. Despite the use of weighting methods to adjust for observed differences in baseline characteristics, residual confounding from unobserved factors remains a recognised limitation that the authors explicitly acknowledge.