Survival impact of anti-CD38-based quadruplet regimens in transplant-ineligible newly diagnosed multiple myeloma: a network meta-analysis and reconstructed individual patient data meta-analysis.

Introduction

Multiple myeloma (MM) is a biologically complex hematologic malignancy characterized by clonal proliferation of malignant plasma cells, marked clinical heterogeneity, and substantial therapeutic challenges [ 1 ]. For patients with transplant-ineligible newly diagnosed multiple myeloma (TI-NDMM), the selection of an optimal frontline treatment is critical in achieving sustained disease control, minimizing toxicity, and improving survival outcomes [ 2 ]. Clinically, therapeutic decisions are guided by frailty assessments, which stratify patients according to physiological reserve and expected treatment tolerance. Consequently, fit or intermediate-risk patients typically receive intensive regimens aimed at maximizing response, whereas frail patients require less-intensive strategies to minimize adverse events and maintain quality of life [ 3 ].

Historically, treatment for TI-NDMM has been comprised of doublet or triplet combinations involving proteasome inhibitors (PI), immunomodulatory drugs (IMiD), alkylating agents, and corticosteroids [ 4 – 6 ]. Recent therapeutic advancements, notably the incorporation of monoclonal antibodies (mAb) targeting CD38, such as daratumumab and isatuximab, have significantly altered the treatment landscape. These agents, when combined with established standard-of-care regimens, have consistently demonstrated deeper therapeutic responses, prolonged disease control, and favorable safety profiles [ 7 , 8 ].

Randomized controlled trials (RCT) evaluating anti-CD38 mAb-based triplet and quadruplet regimens have reported superior overall response rates (ORR), higher minimal residual disease (MRD) negativity rates, and extended progression-free survival (PFS) compared to conventional therapies [ 9 – 12 ]. However, direct comparative evidence evaluating the impact of these regimens on overall survival (OS) remains limited. This evidence gap highlights the necessity for comprehensive comparative analyses to guide evidence-based clinical decisions and optimize patient care.

To address this unmet clinical need, we performed a systematic review (SR), network meta-analysis (NMA), and reconstructed individual patient data meta-analysis (RIPDMA) to evaluate survival outcomes of frontline daratumumab- and isatuximab-based regimens in TI-NDMM. By synthesizing evidence from existing RCT, this analysis aims to provide a clearer understanding of survival advantages of these therapeutic strategies, thereby providing critical insights to optimize therapeutic decisions in TI-NDMM.

Methods

This systematic review (SR), network meta-analysis (NMA) and reconstructed individual patient data meta-analysis (RIPDMA) were conducted and reported in accordance with the recommendations outlined in the Cochrane Handbook for Systematic Reviews of Interventions and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement guidelines [ 13 , 14 ]. The study protocol was prospectively registered on the International Prospective Register of Systematic Reviews (PROSPERO; CRD420251033401).

Databases search strategy

A comprehensive literature search was performed in PubMed/Medline, Embase and Cochrane Library databases from the inception to April 2025. The search strategy included the following terms: “multiple myeloma” AND (“newly diagnosed” OR “first line” OR “front line” OR “untreated”) AND “transplant” AND (“ineligible” OR “not eligible” OR “without intent” OR “no intent”).

Eligibility criteria

The Predefined Population, Intervention, Comparator, Outcome, Time and Study design (PICOTS) criteria guided the study selection. Eligible studies met the following inclusion criteria: (1) RCT; (2) evaluating interventional treatments consisting of quadruplet or triplet combinations of daratumumab, isatuximab, bortezomib, lenalidomide, and dexamethasone in any arrangement; (3) comparing to their backbone regimens; (4) in TI-NDMM patients; (5) reporting for PFS and OS outcomes as hazard ratios (HR) with confidence intervals (CI); and (6) providing a minimum follow-up duration of 48 months. Studies were excluded if they were observational, uncontrolled, or included patient populations that overlapped with previously selected RCT.

Study selection

All citations retrieved from database searches were imported into Zotero reference management software (Corporation for Digital Scholarship), where duplicate records were removed. Two independent authors (JTDSF and LCO) screened the titles and abstracts based on predefined inclusion criteria. Full-text evaluations were performed for studies meeting initial eligibility, and final selections were made by consensus. Any discrepancies were resolved through discussion among the investigators.

Data extraction

Relevant data were systematically extracted from each study, including study characteristics (first author, year of publication, journal, study design, sample size, treatment regimens, and follow-up duration), patient demographics and baseline characteristics (age, sex, International Staging System [ISS] distribution, Eastern Cooperative Oncology Group (ECOG) Performance Status (PS), and cytogenetic risk classification) and clinical outcome (response criteria, PFS and OS). We also evaluated whether patients were transplant-ineligible (TI) or if immediate transplant was not intended. While studies including both TI and transplant-eligible (TE) patients were not excluded, additional analyses were performed to isolate outcomes specific to the TI population. For studies with multiple publications, the version with the longest available follow-up was used to extract summary effects.

Definition of outcomes

The primary endpoint was OS, defined as the time from randomization to death from any cause. We also evaluated the PFS, defined as the time from randomization to disease progression or death from any cause, whichever occurred first. Treatment response was evaluated according to the International Myeloma Working Group (IMWG) criteria, with complete response or better (≥CR) defined as achieving either a complete response (CR) or stringent complete response (sCR) [ 15 ]. Associations for PFS and OS were quantified using hazard ratios (HR) with 95% confidence intervals (CI), while the likelihood of achieving ≥CR was quantified using odds ratios (OR) and their corresponding 95% CI.

Quality assessment

Two independent investigators (JTDSF, LCO) assessed the methodological quality and risk of bias for each included RCT. Risk of bias was assessed using the Cochrane Risk of Bias 2 (RoB-2) tool, which evaluates study reliability based on domains, such as the randomization process, deviations from intended interventions, missing outcome data, measurement of outcomes, and selective reporting bias [ 16 ].

Statistical analysis

The NMA was conducted using a frequentist approach with a random-effects model, providing comparisons between treatment regimens. Treatment effects were pooled to estimate the HR along with their corresponding 95% CI for time-to-event outcomes, and OR with 95% CI for binary response outcomes. When the reported CI differed from a 95% CI, the provided CI was converted to a 95% CI for consistency [ 17 ]. Network plots were generated for each endpoint. Treatment-ranking probabilities were calculated using P-scores, ranging from 0 (worst) to 1 (best), with higher values indicating superior overall performance among competing treatments. P-scores closely approximate the surface under the cumulative ranking curve (SUCRA) [ 18 ]. Heterogeneity was assessed using I 2 statistics, with values of 25% or less indicating low, 50% indicating moderate, and 75% or more indicating high heterogeneity [ 19 ]. A two-tailed P-value < 0.05 was considered statistically significant. All statistical analyses were performed using R statistical software (version 4.4.1) and RStudio, and the “netmeta” package.

In the RIPDMA, individual patient data (IPD) for each treatment arm in each trial were reconstructed from published Kaplan-Meier (KM) survival curves using the IPDfromKM method. This approach estimates event times and censoring based on digitized KM curves and reported numbers at risk, generating reconstructed IPD (RIPD), also referred to as pseudo-IPD. This enables time-to-event analyses and meta-analyses when access to original IPD is unavailable [ 20 ]. KM plots were first digitized using Engauge Digitizer (version 12.1); the resulting coordinates, together with the tabulated numbers at risk, were then processed using the IPDfromKM Shiny app (version 1.2.3.0, available at https://biostatistics.mdanderson.org/ ) to impute individual event and censoring times. The algorithm accurately estimates survival probabilities, event counts, and numbers at risk, under the assumption of uniformly distributed, non‑informative censoring between reported time points [ 20 ].

To ensure the validity of the reconstructed data, we conducted visual comparisons of the original and reconstructed KM curves and confirmed exact concordance of the numbers at risk at all reported time points. Quantitative accuracy was further assessed using root mean square error (RMSE), mean absolute error (MAE), and maximum absolute error (MaxAE), with predefined thresholds of ≤0.05, ≤0.02, and ≤0.05, respectively, indicating acceptable reconstruction precision. We also applied the Kolmogorov–Smirnov (KS) test to compare the empirical distributions of original and reconstructed survival data, with high p-values (P > 0.05) indicating good agreement [ 20 ].

Once the RIPD datasets were validated, indirect comparisons of survival outcomes across treatment regimens were conducted using statistical methods analogous to those employed in RCT with actual IPD [ 21 ]. HR and corresponding 95% CI were calculated to estimate comparative efficacy. Additionally, leave-one-out sensitivity analyses were performed to identify influential studies. All statistical analyses were conducted using R Statistical Software (version 4.4.1) and RStudio, utilizing the packages “survival” (for coxph and survfit) and “survminer” (for ggsurvplot).

Results

Study selection

The initial search yielded 1727 results. After removing duplicates, 1106 articles were screened by title and abstract, leaving 40 potentially eligible studies for detailed review. Following the exclusion of 32 studies, eight reports from a total of four studies were included in the systematic review and meta-analysis (Supplementary Fig. 1 ).

Characteristics of studies and study population

Four RCTs involving a total of 2038 patients with TI-NDMM were included: SWOG S0777 [ 6 , 22 ], MAIA [ 9 , 23 – 25 ], CEPHEUS [ 10 ], and IMROZ [ 11 ]. Publication years ranged from 2017 to 2025. The SWOG S0777 trial compared bortezomib, lenalidomide and dexamethasone (VRd) with lenalidomide and dexamethasone (Rd) [ 6 , 22 ]; the MAIA trial evaluated daratumumab, lenalidomide and dexamethasone (D-Rd) versus Rd [ 9 , 23 – 25 ]; the CEPHEUS trial compared daratumumab, bortezomib, lenalidomide and dexamethasone (D-VRd) with VRd [ 10 ]; and the IMROZ trial compared isatuximab, bortezomib, lenalidomide and dexamethasone (I-VRd) with VRd [ 11 ].

Among the included studies, five distinct intervention arms were analyzed, comprising two quadruplet regimens, two triplet regimens, and one doublet regimen. All regimens included lenalidomide and dexamethasone. Specifically, the treatment groups consisted of D-VRd (n = 197, 9.7%), I-VRd (n = 265, 13.0%), D-Rd (n = 368, 18.1%), VRd (n = 614, 30.1%), and Rd (n = 598, 29.1%). Median follow-up durations reported across these trials ranged from 58.7 to 89.3 months. The weighted median patient age was 70.3 years, 55.1% were male, 13.3% had an Eastern Cooperative Oncology Group (ECOG) Performance Status scale greater than one, 26% had the International Staging System (ISS) stage III disease and 19.1% had a high-risk cytogenetic profile. The main characteristics of the included studies are reported in Table 1 .

Not all participants in the SWOG S0777 and CEPHEUS trials were TI-NDMM. Both studies enrolled patients for whom immediate autologous stem cell transplant was not intended, resulting in a mixed population that included TI patients as well as TE patients who declined or deferred transplant. Since age is a primary criterion for transplant eligibility, we used data from age-based subgroups (patients aged ≥65 years) in the SWOG S0777 trial as a proxy for the TI population, representing 197 patients (43% of the total 460 patients), with 91 in the VRd arm and 106 in the Rd arm [ 22 ]. For the CEPHEUS trial, we extracted data from the reported subgroup analysis specific to TI patients, comprising 289 individuals (73% of the total 395 patients), with 144 receiving D-VRd and 145 receiving VRd [ 26 ].

Clinical outcomes

The OR for achieving ≥CR were derived from the most recent available data, either directly reported in the primary studies, such as CEPHEUS (OR 2.73; 95% CI, 1.71-4.34) [ 10 ] and IMROZ (OR 1.66; 95% CI, 1.10-2.50) [ 11 ], or calculated from published results, as in the MAIA (OR 2.43; 95% CI, 1.79-3.28) [ 23 ] and SWOG S0777 (OR 2.32; 95% CI, 1.38-3.91) [ 22 ] trials. The proportion of patients achieving responses of ≥CR and ≥VGPR is summarized in Table 1 .

The HR for PFS, based on the latest follow-up data from included studies, were reported as follows: CEPHEUS study, HR 0.57 (95% CI, 0.41-0.79) [ 10 ]; IMROZ study, HR 0.60 (98.5% CI, 0.41-0.88) [ 11 ]; MAIA study, HR 0.55 (95% CI, 0.45-0.67) [ 25 ]; and SWOG S0777, study HR 0.742 (96% CI, 0.594-0.928) [ 22 ]. To maintain consistency, confidence intervals originally reported as 98.5% (IMROZ) and 96% (SWOG S0777) were recalculated to the standard 95% CI, resulting in adjusted intervals of 0.44-0.82 for IMROZ and 0.599-0.919 for SWOG S0777.

The HR for OS, based on the most recent follow-up data, were reported as follows: CEPHEUS study, HR 0.85 (95% CI, 0.58-1.24) [ 10 ]; IMROZ study, HR 0.78 (99.97% CI, 0.41-1.48) [ 11 ]; MAIA study, HR 0.67 (95% CI, 0.55-0.82) [ 24 ]; and SWOG S0777 study, HR 0.709 (96% CI, 0.543-0.926) [ 22 ]. Confidence intervals from IMROZ and SWOG S0777 were similarly recalculated to the standard 95% CI, yielding adjusted intervals of 0.55-1.10 for IMROZ and 0.550-0.915 for SWOG S0777.

Network meta-analysis

The network geometry exhibited a linear structure, with each treatment comparison connected by a single direct evidence path and no clustering of nodes (Supplementary Fig. 2 ). Figure 1 summarizes the efficacy outcomes (≥CR, PFS and OS) from all included trials. Treatments were ranked according to their efficacy, using Rd as the reference comparator, as it demonstrated the least favorable OR and HR. For evaluations of ≥CR, an OR > 1 indicates that the comparator treatment is favored over Rd. Conversely, for PFS and OS, an HR < 1 indicates a benefit of the comparator treatment relative to Rd.

In achieving ≥CR, D-VRd demonstrated the highest efficacy (OR 6.33, 95% CI 3.28-12.25), followed by I-VRd (OR 3.84, 95% CI 1.98-7.46), D-Rd (OR 2.43, 95% CI 1.80-3.29), and VRd (OR 2.32, 95% CI 1.38-3.91), all compared to Rd. Regarding PFS, D-VRd showed the greatest benefit (HR 0.42, 95% CI 0.29-0.63), followed by I-VRd (HR 0.45, 95% CI 0.31-0.65), D-Rd (HR 0.55, 95% CI 0.45-0.67) and VRd (HR 0.74, 95% CI 0.60-0.92), compared to Rd. For OS, I-VRd exhibited the most favorable outcome (HR 0.55, 95% CI 0.36-0.85), followed by D-VRd (HR 0.60, 95% CI 0.38-0.95), D-Rd (HR 0.67, 95% CI 0.55-0.82) and VRd (HR 0.71, 95% CI 0.55-0.91), relative to Rd. Heterogeneity (I 2 ) was 0% across all analyses for ≥CR, PFS, and OS, indicating consistency across included studies. These findings are summarized in the forest plots (Fig. 1 ).

Pairwise comparisons of treatment regimens for ≥CR, PFS, and OS are detailed in the league table (Table 2 ). For ≥CR, D-VRd demonstrated significant superiority compared to D-Rd (OR 2.61, 95% CI 1.26-5.39), VRd (OR 2.73, 95% CI 1.82-4.09). No significant difference was observed between D-VRd and I-VRd. I-VRd demonstrated significant benefit over VRd (OR 1.66, 95% CI 1.10-2.50), but not over D-Rd. For PFS, D-VRd (HR 0.57; 95% CI 0.41-0.79), I-VRd (HR 0.60; 95% CI 0.44-0.82), and D-Rd (HR 0.74; 95% CI 0.55-0.99) were significantly superior to VRd. No statistically significant differences in OS were found among D-VRd, I-VRd, D-Rd, and VRd in direct pairwise comparisons.

Treatment rankings based on P-scores are summarized in Table 3 . For ≥CR, D-VRd was ranked highest (P-score: 0.9876), followed by I-VRd (0.7318), D-Rd (0.4185), VRd (0.3619), and Rd (0.0002). Similarly, for PFS, D-VRd demonstrated the highest probability of being the most effective regimen (P-score: 0.8668), closely followed by I-VRd (0.8114), D-Rd (0.5660), VRd (0.2551), and Rd (0.0008). Regarding OS, I-VRd ranked first (P-score: 0.8328), followed by D-VRd (0.7043), D-Rd (0.5683), VRd (0.4467), and Rd (0.0163).

Reconstructed individual patient data meta-analysis

The pooled Kaplan-Meier survival curves (Fig. 2A ) show estimated 60-month PFS rates of 66.4% for D-VRd, 63.2% for I-VRd, 51.9% for D-Rd, 42.6% for VRd, and 28.6% for Rd. Median PFS was not reached for D-VRd and I-VRd, while median values were 61.9 months for D-Rd, 49 months for VRd, and 32.7 months for Rd. The D-VRd combination showed significant improvement in PFS compared with D-Rd (HR 0.65, 95% CI 0.48-0.87; P = 0.003), VRd (HR 0.51, 95% CI 0.39-0.67; P < 0.0001), and Rd (HR 0.35, 95% CI 0.26-0.45; P < 0.0001). Similarly, I-VRd significantly improved PFS versus D-Rd (HR 0.68; 95% CI 0.52-0.89; P = 0.004), VRd (HR 0.53; 95% CI 0.41-0.67; P < 0.0001), and Rd (HR 0.36; 95% CI 0.28-0.45; P < 0.0001). Additionally, D-Rd demonstrated superior PFS compared to VRd (HR 0.77, 95% CI 0.64-0.93; P = 0.007) and Rd (HR 0.53, 95% CI 0.44-0.63; P < 0.0001).

The 60-month OS rates (Fig. 2B ) were 72.8% for D-VRd, 72.2% for I-VRd, 67.1% for D-Rd, 67.0% for VRd, and 54.8% for Rd. The median OS was not reached for D-VRd and I-VRd, while median values were 90.3 months for D-Rd, 89.4 months for VRd, and 66.6 months for Rd. Although no statistically significant OS differences were observed among D-VRd, I-VRd, D-Rd, and VRd, all demonstrated significant OS benefits compared with Rd: D-VRd (HR 0.56, 95% CI 0.42-0.76; P = 0.0001), I-VRd (HR 0.53, 95% CI 0.41-0.69; P < 0.0001), D-Rd (HR 0.71, 95% CI 0.59-0.86; P = 0.0003), and VRd (HR 0.71, 95% CI 0.61-0.84; P = 0.0001).

We conducted an additional age/intent‑restricted OS analysis (Supplementary Fig. 3A ) excluding patients aged <65 years from the SWOG S0777 trial and those who deferred transplant in the CEPHEUS trial, to better reflect the TI-NDMM population. In this analysis, both D-VRd and I-VRd demonstrated a statistically significant OS benefit compared to VRd (HR 0.60; 95% CI, 0.41–0.88; P = 0.005, and HR 0.67; 95% CI, 0.50–0.88; P = 0.004, respectively). When compared with D-Rd, both regimens showed a trend toward improved OS; however, these differences did not reach statistical significance (HR 0.69; 95% CI, 0.47–1.03; P = 0.06, and HR 0.76; 95% CI, 0.57–1.02; P = 0.06, respectively).

Quadruplet (D-VRd and I-VRd) and triplet (D-Rd and VRd) regimens were analyzed as groups strategies (Fig. 3 ). The median PFS and OS were not reached for the quadruplet group, whereas the triplet group demonstrated median PFS of 54.3 months and median OS of 90.4 months. At 60 months, quadruplets showed significantly higher PFS (64.7% vs. 46.3%) and OS (72.5% vs. 67.1%) compared to triplets. In pooled analyses, quadruplet regimens significantly improved PFS (HR 0.57; 95% CI 0.47–0.69; P < 0.0001) and OS (HR 0.78; 95% CI 0.63–0.96; P = 0.02) relative to triplet therapy. The OS benefit remained evident in the analysis excluding patients aged <65 years from SWOG S0777 and those who deferred transplant in CEPHEUS (73.5% vs. 65.4%; HR 0.69; 95% CI 0.55–0.87; P = 0.001) (Supplementary Fig. 3B ).

Quadruplet regimens were also individually compared with each triplet regimen (D-Rd and VRd). At 60 months, quadruplet therapy demonstrated significantly superior PFS and OS compared to VRd (PFS: 64.7% vs. 42.6%; HR 0.52; 95% CI, 0.43–0.63; P < 0.0001; OS: 72.5% vs. 67.0%; HR 0.77; 95% CI, 0.62–0.97; P = 0.02) (Supplementary Fig. 4 ). Similarly, when compared to D-Rd, quadruplet regimens were associated with significantly improved PFS (64.7% vs. 51.9%; HR 0.66; 95% CI: 0.53–0.83; P = 0.0003) and OS (72.5% vs. 67.1%; HR 0.77; 95% CI: 0.60–0.98; P = 0.04) (Supplementary Fig. 5 ).

To assess the robustness of the OS findings, we performed a non-parametric bootstrap analysis with 1000 resamples from the reconstructed individual patient data. For the comparison between quadruplet and triplet regimens, the bootstrapped median HR for OS was 0.64 (95% CI: 0.62–0.83), consistent with the primary estimate. These results support the reproducibility and stability of the observed OS benefit associated with quadruplet therapy.

Sensitivity analyses using the leave-one-out method were performed to evaluate the influence of individual trials on the OS estimates. Excluding the SWOG S0777 trial had minimal impact on the pooled results, with a HR of 0.77 (95% CI, 0.62–0.95; P = 0.01) for OS when comparing quadruplet versus triplet regimens (Supplementary Fig. 6 ). A complementary analysis excluding both the SWOG S0777 and MAIA trials yielded consistent findings, with an HR of 0.74 (95% CI, 0.58–0.95; P = 0.02) for OS (Supplementary Fig. 7 ).

Quality assessment

The methodological quality of the included studies was systematically assessed using the Cochrane Risk-of-Bias Tool for randomized trials (RoB 2). All four RCT (SWOG S0777, MAIA, CEPHEUS, and IMROZ) demonstrated a low risk of bias in the majority of the evaluated criteria (Supplementary Fig. 8 ). Additionally, the network geometry of our meta-analysis did not include closed loops of evidence, precluding a statistical assessment of inconsistency and small-study effects (Supplementary Fig. 2 ).

All reconstructed survival data met the predefined accuracy thresholds, demonstrating a high degree of concordance between the reconstructed and original KM curves (Supplementary Fig. 9 ). The numbers at risk at all reported time points were individually verified for each curve and matched exactly with those in the original publications. This strong level of agreement supports the reliability of the survival estimates used in our subsequent analyses.

Discussion

This systematic review and meta-analysis presents the first comprehensive comparison of anti-CD38 mAb-based quadruplet versus standard triplet therapies in TI-NDMM. Synthesizing data from 2038 patients across four RCT, our findings provide robust evidence supporting the superiority of quadruplet regimens incorporating daratumumab or isatuximab, with significant improvements in both PFS and OS.

Both D-VRd and I-VRd were associated with substantial gains in PFS compared to triplet regimens (D-Rd and VRd), with estimated 60-month PFS rates exceeding 60%. In direct comparisons, D-VRd and I-VRd significantly reduced the risk of progression or death by 35% and 32% compared with D-Rd, and by 49% and 47% compared with VRd, respectively. These findings highlight the enhanced and durable disease control achieved with quadruplet regimens in a population traditionally considered challenging to treat [ 27 ].

Notably, this benefit extended to OS. The median OS was not reached in the quadruplet arms, and it was approximately 90 months among patients receiving the triplet therapy. At 60 months, the estimated OS rate was 72.5% for patients treated with quadruplet regimens, compared to 66.7% for those receiving triplets. While direct OS comparisons between individual regimens did not reach statistical significance, this was likely due to the limited number of events and the requirement for a longer follow-up. Nevertheless, the observed trend consistently favored quadruplet-based therapy.

To enhance statistical power and clinical interpretability, quadruplet (D-VRd and I-VRd) and triplet (D-Rd and VRd) regimens were analyzed as grouped strategies. The pooled analysis reinforced the clinical value of quadruplet therapy, demonstrating a 43% reduction in the risk of progression or death (HR 0.57; 95% CI, 0.47–0.69; P < 0.0001) and a 22% reduction in the risk of death (HR 0.78; 95% CI, 0.63–0.96; P = 0.02) compared with triplet regimens. These findings are particularly compelling, given the advanced age, frequent comorbidities, and potential frailty of the TI-NDMM population, factors traditionally associated with poorer outcomes and a preference for less intensive therapeutic approaches [ 28 ]. Our results challenge this convention, demonstrating that intensive yet tolerable frontline approaches significantly extend survival, even among patients typically considered unsuitable for aggressive therapies.

Given that the triplet regimens consist of lenalidomide-based combinations with either daratumumab without a PI (D-Rd) or a PI without daratumumab (VRd), we conducted separate comparisons of quadruplet regimens versus each triplet. In both comparisons, quadruplets demonstrated marked improvements in PFS and OS, reinforcing the therapeutic synergy of combining all four classes to enhance disease control and prolong survival [ 29 ].

Demonstrating an OS benefit in the frontline setting is increasingly difficult due to prolonged follow-up needs and the availability of effective salvage options, including second-line anti-CD38 combinations and T-cell redirecting immunotherapies [ 30 ]. Nevertheless, our results suggest that earlier introduction of anti-CD38 mAb within a quadruplet regimen confers meaningful and durable survival benefits. Future longitudinal studies will be essential to define optimal sequencing strategies and to assess whether an optimized frontline regimen alone can achieve sustained disease control.

To assess the reliability of our findings, we performed both a leave-one-out sensitivity analysis and a bootstrapping procedure. OS estimates remained consistent when either the SWOG S0777 or MAIA trial was individually excluded, suggesting that no single study unduly influenced the pooled effect. Furthermore, bootstrapping produced HR estimates that were highly consistent with our primary analysis, reinforcing the stability and reproducibility of the observed survival benefit. Although we acknowledge the approximate nature of reconstructed data, the strong agreement of bootstrapped results further supports the credibility of our conclusions.

Differences in age distribution and frailty status across the included trials represent important sources of potential confounding when interpreting survival outcomes. Notably, the quadruplet trials (CEPHEUS and IMROZ) capped patient age at 80 years, whereas the MAIA trial enrolled a broader population, including patients well into their 80 s and beyond [ 9 , 10 , 22 ]. This discrepancy may affect treatment tolerance, increase the impact of competing risks of death, and influence long-term survival estimates. Moreover, frailty—an independent and clinically relevant prognostic factor in TI-NDMM—is not consistently assessed or reported across studies and may significantly affect treatment outcomes. As such, comparisons of OS across regimens should be interpreted with caution, particularly in very elderly or frail patients who may be underrepresented in some trials.

To further mitigate the confounding effects of age and fitness, we conducted an additional age- and intent-restricted analysis designed to more accurately represent a true TI-NDMM population. Not all participants in the SWOG S0777 and CEPHEUS trials were truly TI-NDMM [ 6 , 10 ]. Both trials enrolled patients for whom immediate autologous stem cell transplant was not intended, resulting in a heterogeneous population that included TI patients as well as TE patients who declined or deferred transplant. In this restricted analysis, we excluded patients aged <65 years from the SWOG S0777 trial and those who deferred transplant in the CEPHEUS trial, as these individuals may represent a younger, fitter, and potentially TE subset. By removing these patients, we aimed to minimize bias introduced by differences in baseline fitness. Importantly, the OS benefit of quadruplet regimens remained evident even after these exclusions, reinforcing the robustness of our findings. In this refined population, we also observed a sustained OS advantage for both D-VRd and I-VRd compared to VRd.

Beyond the comparison between quadruplet and triplet strategies, our analysis also examined PFS within triplet regimens and confirmed the superiority of D-Rd over VRd, demonstrating a 23% reduction in the risk of progression or death. This finding is consistent with prior indirect comparisons and real-world data, further supporting D-Rd as the preferred triplet regimen in the TI setting [ 31 – 33 ].

In the treatment-ranking probabilities derived from the NMA, anti-CD38 mAb-based quadruplet regimens (D-VRd and I-VRd) consistently demonstrated the highest likelihood of being the most effective options across all evaluated outcomes, including ≥CR, PFS, and OS. The anti-CD38 triplet regimen (D-Rd) ranked next in efficacy, outperforming regimens lacking anti-CD38 mAb (VRd an Rd). The consistency of these rankings across multiple clinically relevant endpoints reinforces the robustness of the evidence supporting anti-CD38 mAb-based quadruplets as the preferred frontline strategy for TI-NDMM. When quadruplet therapy is not feasible, anti-CD38 mAb-based triplets represent a strong alternative.

While several previous meta-analyses have evaluated frontline treatments for TI-NDMM, none have incorporated the recently introduced quadruplet regimens combining anti-CD38 mAbs with a PI and IMiD [ 34 – 39 ]. To our knowledge, this is the first analysis to address this evidence gap in the transplant-ineligible population. Moreover, our study provides a detailed and comparative evaluation of OS, a key endpoint underreported in earlier analyses. Similar findings were observed in transplant-eligible patients [ 40 – 42 ], further supporting the efficacy of quadruplet regimens across multiple myeloma subgroups and reinforcing the adoption of more intensive, yet tolerable, frontline therapies.

Although not included in this meta-analysis due to its melphalan-based backbone, the ALCYONE study was the first phase III trial to evaluate a quadruplet regimen in TI-NDMM, showing that D-VMP significantly improved outcomes over VMP alone [ 12 , 43 ]. However, subsequent regimens like D-Rd have demonstrated superior efficacy in both PFS and OS [ 9 , 23 – 25 ]. Network meta-analyses have also consistently ranked D-Rd higher than D-VMP for both PFS and OS outcomes [ 34 ]. While melphalan-based regimens have declined in contemporary practice [ 2 , 44 ], D-VMP remains a relevant option in settings with limited access to lenalidomide or newer quadruplets.

Our analysis adhered to rigorous methodological standards, including strict inclusion criteria and advanced analytical techniques. Although the BENEFIT trial represents an important addition to the therapeutic landscape for TI-NDMM, it was not included in our meta-analysis because HR for OS and PFS were not available at the time of analysis [ 45 ]. As our meta-analysis required time-to-event outcomes, inclusion of trials lacking these endpoints was not methodologically feasible. As these data become available, future analyses may incorporate its findings to further refine the evolving comparative landscape.

A major strength of our study lies in its dual methodological approach, combining frequentist NMA with RIPDMA. By integrating both aggregate and reconstructed time-to-event data, we offer a comprehensive and reliable evaluation of survival outcomes [ 46 ]. The methodological consistency and inclusion of trials with mature follow-up enhance the validity and clinical applicability of our findings.

Several limitations should be acknowledged. First, the indirect nature of the comparisons, due to the absence of direct head-to-head RCT comparing quadruplet regimens, introduces inherent uncertainty. Second, the SWOG S0777 trial enrolled patients without a definitive intent for autologous stem cell transplantation, which may limit its applicability to strictly transplant-ineligible populations. Third, the limited reporting of postprotocol therapies across studies raises concerns about treatment imbalances, particularly in control arms, that may have influenced the observed OS outcomes. Fourth, although the reconstruction of individual patient data is a validated and reproducible technique, it remains an approximation and cannot fully substitute for access to original patient-level datasets. Fifth, patient age and frailty index distributions differed across trials and may have impacted the observed OS benefit of quadruplet regimens; thus, OS estimates should be interpreted with caution in very elderly or frail subgroups. Finally, despite minimal statistical heterogeneity, subtle variations in patient characteristics, treatment durations, frailty assessments, therapeutic sequencing, and follow-up protocols may affect comparability and limit generalizability.

Despite these limitations, our findings have clear clinical implications. Anti-CD38 mAb-based quadruplet regimens should be strongly considered as a new standard of care for fit and intermediate-fitness TI-NDMM patients. Both daratumumab and isatuximab demonstrated comparable efficacy, offering flexibility in regimen selection. The observed OS benefit with quadruplets supports a more intensive frontline strategy aimed at maximizing long-term disease control.

In conclusion, our meta-analysis consolidates compelling evidence that frontline quadruplet regimens incorporating anti-CD38 mAbs significantly improve survival outcomes in TI-NDMM, substantially reducing the risk of disease progression and death.

Supplementary information

Supplemental Material

Acknowledgements

This study was presented at the 67th ASH Annual Meeting and Exposition, December 6-9, 2025, Orlando, Florida, USA. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author contributions

JTDSF designed the research and analyzed the data. JTDSF and LOC were responsible for data acquisition and interpretation. JTDSF drafted the manuscript, while LOC, EQC, VH, and AM contributed to its revision. All authors approved the final manuscript.

Data availability

This meta-analysis was conducted using data sourced from previously published studies, all of which are publicly accessible. The authors did not have access to individual patient-level data from the included studies. All pertinent datasets generated and analyzed during this study are provided within the article and its Supplementary Materials. For further inquiries or requests for additional data, please contact the corresponding author.

Competing interests

JTDSF reported receiving honoraria from AbbVie, AstraZeneca, BeOne, BMS, Johnson & Johnson, Novartis, and Sanofi. LOC reported receiving honoraria from Johnson & Johnson. EQC reported receiving research funding from Johnson & Johnson and honoraria from Amgen, BMS, Johnson & Johnson, Sanofi, GSK, and Takeda. VH reported receiving honoraria from BMS, GSK, Johnson & Johnson, Regeneron, Sanofi, Takeda, Amgen, and AbbVie. AM reported receiving honoraria from Johnson & Johnson, BMS, Sanofi, Takeda, Pfizer, and Amgen. No other disclosures were reported.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

These authors contributed equally: Vania Hungria, Angelo Maiolino.

Supplementary information

The online version contains supplementary material available at 10.1038/s41408-025-01413-7.

References

1. Malard F, Neri P, Bahlis NJ, Terpos E, Moukalled N, Hungria VTM, et al. Multiple myeloma. Nat Rev Dis Prim. 2024;10:45.

2. Fonseca R, Rossi A, Medhekar R, Voelker J, Homan T, Wilcock J, et al. Clinical consensus on treatments for transplant-ineligible newly diagnosed multiple myeloma: double-blinded Delphi panel. Future Oncol. 2024;20:1645–56.

3. Dimopoulos MA, Moreau P, Terpos E, Mateos MV, Zweegman S, Cook G, et al. Multiple myeloma: EHA-ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2021;32:309–22.

4. San Miguel JF, Schlag R, Khuageva NK, Dimopoulos MA, Shpilberg O, Kropff M, et al. Bortezomib plus melphalan and prednisone for initial treatment of multiple myeloma. N Engl J Med. 2008;359:906–17.

5. Benboubker L, Dimopoulos MA, Dispenzieri A, Catalano J, Belch AR, Cavo M, et al. Lenalidomide and dexamethasone in transplant-ineligible patients with myeloma. N Engl J Med. 2014;371:906–17.

6. Durie BGM, Hoering A, Abidi MH, Rajkumar SV, Epstein J, Kahanic SP, et al. Bortezomib with lenalidomide and dexamethasone versus lenalidomide and dexamethasone alone in patients with newly diagnosed myeloma without intent for immediate autologous stem-cell transplant (SWOG S0777): a randomised, open-label, phase 3 trial. Lancet. 2017;389:519–27.

7. van de Donk NWCJ, Moreau P, Plesner T, Palumbo A, Gay F, Laubach JP, et al. Clinical efficacy and management of monoclonal antibodies targeting CD38 and SLAMF7 in multiple myeloma. Blood. 2016;127:681–95.

8. Rajkumar SV. Multiple myeloma: 2024 update on diagnosis, risk-stratification, and management. Am J Hematol. 2024;99:1802–24.

9. Facon T, Kumar S, Plesner T, Orlowski RZ, Moreau P, Bahlis N, et al. Daratumumab plus lenalidomide and dexamethasone for untreated myeloma. N Engl J Med. 2019;380:2104–15.

10. Usmani SZ, Facon T, Hungria V, Bahlis NJ, Venner CP, Braunstein M, et al. Daratumumab plus bortezomib, lenalidomide and dexamethasone for transplant-ineligible or transplant-deferred newly diagnosed multiple myeloma: the randomized phase 3 CEPHEUS trial. Nat Med. 2025;31:1195–202.

11. Facon T, Dimopoulos M-A, Leleu XP, Beksac M, Pour L, Hájek R, et al. Isatuximab, bortezomib, lenalidomide, and dexamethasone for multiple myeloma. N Engl J Med. 2024;391:1597–609.

12. Mateos M-V, Dimopoulos MA, Cavo M, Suzuki K, Jakubowiak A, Knop S, et al. Daratumumab plus bortezomib, melphalan, and prednisone for untreated myeloma. N Engl J Med. 2018;378:518–28.

13. Higgins J, Thomas J, Chandler J, Cumpston M, Li T, Page M, et al. Cochrane handbook for systematic reviews of interventions version 6.4. Cochrane; 2023. www.training.cochrane.org/handbook .

14. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71.

15. Durie BGM, Harousseau J-L, Miguel JS, Bladé J, Barlogie B, Anderson K, et al. International uniform response criteria for multiple myeloma. Leukemia. 2006;20:1467–73.

16. Sterne JAC, Savović J, Page MJ, Elbers RG, Blencowe NS, Boutron I, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019;366:l4898.

17. Tierney JF, Stewart LA, Ghersi D, Burdett S, Sydes MR. Practical methods for incorporating summary time-to-event data into meta-analysis. Trials. 2007;8:16.

18. Rücker G, Schwarzer G. Ranking treatments in frequentist network meta-analysis works without resampling methods. BMC Med Res Methodol. 2015;15:58.

19. Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327:557–60.

20. Liu N, Zhou Y, Lee JJ. IPDfromKM: reconstruct individual patient data from published Kaplan-Meier survival curves. BMC Med Res Methodol. 2021;21:111.

21. Messori A. Synthetizing published evidence on survival by reconstruction of patient-level data and generation of a multi-trial Kaplan-Meier curve. Cureus. 2021;13:e19422.

22. Durie BGM, Hoering A, Sexton R, Abidi MH, Epstein J, Rajkumar SV, et al. Longer term follow-up of the randomized phase III trial SWOG S0777: bortezomib, lenalidomide and dexamethasone vs. lenalidomide and dexamethasone in patients (Pts) with previously untreated multiple myeloma without an intent for immediate autologous stem cell transplant (ASCT). Blood Cancer J. 2020;10:53.

23. Facon T, Kumar SK, Plesner T, Orlowski RZ, Moreau P, Bahlis N, et al. Daratumumab, lenalidomide, and dexamethasone versus lenalidomide and dexamethasone alone in newly diagnosed multiple myeloma (MAIA): overall survival results from a randomised, open-label, phase 3 trial. Lancet Oncol. 2021;22:1582–96.

24. Facon T, Kumar SK, Orlowski RZ, Bahlis N, Moreau P, Goldschmidt H, et al. Final survival analysis of daratumumab plus lenalidomide and dexamethasone versus lenalidomide and dexamethasone in transplant-ineligible patients with newly diagnosed multiple myeloma: MAIA study. HemaSphere. 2024;8:1723–5.

25. Facon T, Moreau P, Weisel K, Goldschmidt H, Usmani SZ, Chari A, et al. Daratumumab/lenalidomide/dexamethasone in transplant-ineligible newly diagnosed myeloma: MAIA long-term outcomes. Leukemia. 2025;39:942–50.

26. Facon T, Zweegman S, Hungria V, Bahlis NJ, Venner CP, Braunstein MJ, et al. Daratumumab plus bortezomib, lenalidomide, and dexamethasone (DVRd) in patients with newly diagnosed multiple myeloma (NDMM): subgroup analysis of transplant-ineligible (TIE) patients in the phase 3 CEPHEUS study. JCO. 2025;43:7516.

27. Grant SJ, Mian HS, Giri S, Boutin M, Dottorini L, Neuendorff NR, et al. Transplant-ineligible newly diagnosed multiple myeloma: current and future approaches to clinical care: a Young International Society of Geriatric Oncology Review Paper. J Geriatr Oncol. 2021;12:499–507.

28. Mian H, McCurdy A, Giri S, Grant S, Rochwerg B, Winks E, et al. The prevalence and outcomes of frail older adults in clinical trials in multiple myeloma: a systematic review. Blood Cancer J. 2023;13:6.

29. Kostopoulos IV, Fotiou D, Gavriatopoulou M, Rousakis P, Ntanasis-Stathopoulos I, Panteli C, et al. Efficacy and immune modulation associated with the addition of IMiDs to Daratumumab backbone in multiple myeloma patients refractory to both drug classes: resetting synergistic activity. Blood Cancer J. 2024;14:26.

30. van de Donk NWCJ. Sequencing multiple myeloma therapies with and after antibody therapies. Hematology Am Soc Hematol Educ Program. 2020;2020:248–58.

31. Durie BGM, Kumar SK, Ammann EM, Fu AZ, Kaila S, Lam A, et al. Adjusted indirect treatment comparison of progression-free survival with D-Rd and VRd based on MAIA and SWOG S0777 individual patient-level data. Adv Ther. 2024;41:1923–37.

32. Durie BGM, Kumar SK, Usmani SZ, Nonyane BAS, Ammann EM, Lam A, et al. Daratumumab-lenalidomide-dexamethasone vs standard-of-care regimens: efficacy in transplant-ineligible untreated myeloma. Am J Hematol. 2020;95:1486–94.

33. Gordan LN, Tan CR, Vescio R, Ye JC, Schinke C, Medhekar R, et al. Progression-free survival of daratumumab versus bortezomib triplet combination with lenalidomide and dexamethasone in transplant ineligible patients with newly diagnosed multiple myeloma: TAURUS chart review study. Clin Lymphoma Myeloma Leuk. 2024;24:55–63.

34. Facon T, San-Miguel J, Dimopoulos MA, Mateos M-V, Cavo M, van Beekhuizen S, et al. Treatment regimens for transplant-ineligible patients with newly diagnosed multiple myeloma: a systematic literature review and network meta-analysis. Adv Ther. 2022;39:1976–92.

35. Tang W, Zhang L, Zheng Y, Pan L, Niu T. Daratumumab-based immunotherapy vs. lenalidomide, bortezomib and dexamethasone in transplant-ineligible newly diagnosed multiple myeloma: a systemic review. Front Oncol. 2024;14:1286029.

36. Kiss S, Gede N, Soós A, Hegyi P, Nagy B, Imrei M, et al. Efficacy of first-line treatment options in transplant-ineligible multiple myeloma: a network meta-analysis. Crit Rev Oncol Hematol. 2021;168:103504.

37. Giri S, Aryal MR, Yu H, Grimshaw A, Pathak R, Huntington SP, et al. Efficacy and safety of frontline regimens for older transplant-ineligible patients with multiple myeloma: a systematic review and meta-analysis. J Geriatr Oncol. 2020;11:1285–92.

38. Gil-Sierra MD, Gimeno-Ballester V, Fenix-Caballero S, Alegre-Del Rey EJ. Network meta-analysis of first-line treatments in transplant-ineligible multiple myeloma patients. Eur J Haematol. 2020;105:56–65.

39. Ramasamy K, Dhanasiri S, Thom H, Buchanan V, Robinson S, D’Souza VK, et al. Relative efficacy of treatment options in transplant-ineligible newly diagnosed multiple myeloma: results from a systematic literature review and network meta-analysis. Leuk Lymphoma. 2020;61:668–79.

40. Souto Filho JTD, Cantadori LO, Crusoe de EQ, Hungria V, Maiolino A. Daratumumab-based quadruplet versus triplet induction regimens in transplant-eligible newly diagnosed multiple myeloma: a systematic review and meta-analysis. Blood Cancer J. 2025;15:37.

41. Ebraheem MS, Chakraborty R, Rochwerg B, Visram A, Mohyuddin GR, Venner CP, et al. Quadruplet regimens for patients with newly diagnosed multiple myeloma: a systematic review and meta-analysis. Blood Adv. 2024;8:5993–6002.

42. Paul B, Anwer F, Raza S, Mammadzadeh A, Khasawneh B, Shatnawi S, et al. Comparative meta-analysis of triplet vs. quadruplet induction regimens in newly diagnosed, treatment naïve, multiple myeloma. Cancers. 2024;16:2938.

43. Mateos M-V, San-Miguel J, Cavo M, Suzuki K, Jakubowiak A, Knop S, et al. Bortezomib, melphalan, and prednisone with or without daratumumab in transplant-ineligible patients with newly diagnosed multiple myeloma (ALCYONE): final analysis of an open-label, randomised, multicentre, phase 3 trial. Lancet Oncol. 2025;26:S1470-2045(25)00018-X.

44. Côté J, Kotb R, Bergstrom DJ, LeBlanc R, Mian HS, Othman I, et al. First line treatment of newly diagnosed transplant ineligible multiple myeloma: recommendations from the Canadian Myeloma Research Group Consensus Guideline Consortium. Clin Lymphoma Myeloma Leuk. 2023;23:340–54.

45. Leleu X, Hulin C, Lambert J, Bobin A, Perrot A, Karlin L, et al. Isatuximab, lenalidomide, dexamethasone and bortezomib in transplant-ineligible multiple myeloma: the randomized phase 3 BENEFIT trial. Nat Med. 2024;30:2235–41.

46. Messori A, Fadda V, Romeo MR, Veneziano S, Trippoli S. A comparison of statistical analysis between “Real” patients reported in Kaplan-Meier curves and “Reconstructed” patients estimated through the IPDfromKM method: analysis of eight trials evaluating catheter ablation of ventricular tachycardia. Cureus. 2023;15:e47891.

Associated Data

Supplementary Materials

Supplemental Material

Data Availability Statement

This meta-analysis was conducted using data sourced from previously published studies, all of which are publicly accessible. The authors did not have access to individual patient-level data from the included studies. All pertinent datasets generated and analyzed during this study are provided within the article and its Supplementary Materials. For further inquiries or requests for additional data, please contact the corresponding author.