^{1}

^{†}

^{2}

^{†}

^{2}

^{3}

^{1}

This article was submitted to Pharmaceutical Medicine and Outcomes Research, a section of the journal Frontiers in Pharmacology

These authors have contributed equally to this work.

This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

Multiple myeloma (MM) is the second most common hematological malignancy in the world after non-Hodgkin’s lymphoma (

Bortezomib, melphalan, and prednisone (VMP) is one of the standard treatments for transplant-ineligible newly diagnosed multiple myeloma (TNE NDMM) (

D-VMP conferred patients with substantial improvement in health outcomes, but its cost was much higher than VMP. Therefore, we did this study to evaluate the cost-effectiveness of D-VMP and VMP to treat TNE NDMM patients from the perspective of US payers.

We developed a decision-analytic Markov model which included three mutually exclusive health states: PFS state, progression disease and death. In our study, all patients (350 patients in d-VMP group and 356 patients in VMP group) were simulated from the PFS state, the patient may then progress or die. As the progression state, according to the update analysis of ALCYONE, both groups could receive subsequent line(s) of active treatment or best supportive care until death (

Markov state transition model.

We considered only direct medical care costs, including drug, evaluation and management, treatment costs of adverse events (AEs). The drug unit cost was obtained from Centers for Medicare and Medicaid Services (CMS), costs of AEs were used the data from published literature (

As for evaluation and management, we consulted clinicians about the specific use duration and usage of these drugs in clinical application, and calculated the management costs of the corresponding drugs and laboratory tests according to the calculation formula in the Medicare Fee-for-Service Payment and of CMS (

Key model parameters.

Parameter | Base case | 95% CI or range | PSA Distribution | References |
---|---|---|---|---|

Drug cost/$ | ||||

Daratumumab 10 mg | 56.063 | ±20% | Gamma | CMS |

Bortezomib 0.1 mg | 24.167 | ±20% | Gamma | CMS |

Prednisone 1 mg | 0.012 | ±20% | Gamma | CMS |

Dexamethasone 0.25 mg | 0.103 | ±20% | Gamma | CMS |

Paracetamol 650 mg | 0.037 | ±20% | Gamma | REDBOOK |

Diphenhydramine 50 mg | 0.017 | ±20% | Gamma | REDBOOK |

Melphalan 2 mg | 9.159 | ±20% | Gamma | REDBOOK |

Lenalidomide 25 mg | 763.010 | ±20% | Gamma | REDBOOK |

Carfilzomib 1 mg | 38.972 | ±20% | Gamma | CMS |

PD cost for D-VMP | 17,766.986 | ±20% | Gamma | Estimated |

PD cost for VMP | 16,440.419 | ±20% | Gamma | Estimated |

Administration/$ | ||||

D-VMP/every infusion | 349.821 | ±20% | Gamma | Estimated |

VMP/every injection | 49.928 | ±20% | Gamma | Estimated |

Test or Monitoring/$ | ||||

Laboratory tests | 36.145 | ±20% | Gamma | Estimated |

Monitoring for PD | 481.695 | ±20% | Gamma | Usmani SZ et al. |

AE related/$ | ||||

AE cost of D-VMP/per cycle | 240.318 | ±20% | Gamma | Estimated |

AE cost of VMP/per cycle | 621.084 | ±20% | Gamma | Estimated |

Survival & Utilities | ||||

HR of PFS | 0.42 | 0.34–0.51 | Beta | ALCYONE |

HR of OS | 0.60 | 0.46–0.80 | Beta | ALCYONE |

PFS for D-VMP | 0.685 | ±20% | Beta | Estimated |

PFS for VMP | 0.627 | ±20% | Beta | Hatswell AJ et al. |

PD for D-VMP | 0.59 | ±20% | Beta | Usmani SZ et al. |

PD for VMP | 0.59 | ±20% | Beta | Usmani SZ et al. |

Others | ||||

Body weight | 79.0 | 49.8–199.0 | Gamma | Gong CL et al. |

Body surface area | 1.91 | 1.40–2.53 | Gamma | Gong CL et al. |

Discount factor | 3% | 0%–5% | — | Wan X et al. |

PD, progression disease; D-VMP, daratumumab plus bortezomib, melphalan and prednisone; VMP, bortezomib, melphalan and prednisone; AE, Adverse event; HR, hazard ratio; PFS, progression-free survival; OS, overall survival; CI, confidence interval; PSA, probabilistic sensitivity analysis; CMS, Centers for Medicare and Medicaid Services.

We used the approach described by Hoyle and Henley (^{γ}). Then calculated the transition probability of D-VMP regimen according to HR (D-VMP vs. VMP). We estimated the probability of death in each age background based on the 2017 US life Table (

Parametric survival distributions.

Parametric model | PFS | OS | ||
---|---|---|---|---|

AIC | BIC | AIC | BIC | |

Weibull | 2329.102 | 2336.858 | 1466.894 | 1474.638 |

Lognormal | 2381.626 | 2389.382 | 1501.143 | 1508.888 |

Loglogistic | 2355.820 | 2363.576 | 1478.216 | 1485.960 |

Logistic | 2423.814 | 2431.570 | 1513.712 | 1521.457 |

PFS, progression-free survival; OS, overall survival; AIC, Akaike information criterion; BIC, Bayesian information criterion.

The quality of life (QOL) is often referred to as utility. It is usually calculated based on the results obtained from EuroQol five dimensions questionnaire (EQ-5D) and Treatment of Cancer Quality of Life Questionnaire C30 (EORTC QLQ-C30). For PFS, Hatswell AJ et al. (

In order to evaluate the stability of the model and solve the uncertainty of the input parameters, it is generally necessary to carry out one-way sensitivity analysis and probabilistic sensitivity analysis (PSA).

For one-way sensitivity analysis, the lower limit and upper limit used between bounds of their 95% CIs to determine the impact of key parameters on the model. If there no 95% CI values, it will be changed according to the published literature or varied by ±20% of the parameters.

For PSA, we conducted 1,000 replicated Monte Carlo simulations to test the effect of changing all parameters of the model simultaneously on our outcomes. Statistical distribution sampling based on parameter characteristics. This result was used to plot the cost-effectiveness acceptability curves and scatter plot (

The model simulated the disease development of patient’s life time for D-VMP and VMP regimens. For TNE NDMM patients, adding daratumumab to VMP provided an additional 3.00 LYs. Accounting for QOL, patients in D-VMP group gained 6.40 QALYs; this was 2.03 QALYs more than for patients in VMP group. Compared with the VMP strategy, the mean incremental costs of the D-VMP were $788,541; the incremental cost per QALY gained was $388,364; the incremental cost per LY gained was $262,526 (

Base case results.

Regimen | LYs | QALYs | Cost | ICER | |
---|---|---|---|---|---|

per LY | per QALY | ||||

D-VMP | 10.300 | 6.404 | $1,927,635 | $262,526 | $388,364 |

VMP | 7.296 | 4.373 | $1,139,094 |

D-VMP, daratumumab plus bortezomib, melphalan and prednisone; VMP, bortezomib, melphalan and prednisone; LY, life-year; QALY, quality-adjusted life-year; ICER, incremental cost-effectiveness ratio.

The results of one-way sensitivity analysis were showed in

The tornado diagrams of one-way sensitivity analysis. D-VMP: daratumumab plus bortezomib, melphalan and prednisone; VMP: bortezomib, melphalan and prednisone; PD: progression disease; HR: hazard ratio; PFS: progression-free survival; OS: overall survival. ICER: incremental cost-effectiveness ratio; QALY: quality-adjusted life-year.

The results of PSA with 1,000 replicated Monte Carlo simulations were showed in

Cost-effectiveness acceptability curves of D-VMP vs. VMP. QALY: quality-adjusted life-year; D-VMP: daratumumab plus bortezomib, melphalan and prednisone; VMP: bortezomib, melphalan and prednisone.

Scatter plot of probabilistic sensitivity analysis. ICER: incremental cost-effectiveness ratio; WTP: willingness to pay.

Although the D-VMP regimen had better efficacy than VMP in TNE NDMM, it was costly. We conducted a long-term economic evaluation of adding daratumumab to VMP as first-line therapy in TNE NDMM. The base case analysis results showed that the LYs and QALYs gained of D-VMP vs. VMP were 3.00 and 2.03, respectively. The results of cost-effectiveness were estimated at $262,526 per LY and $388,364 per QALY. Sensitivity analysis showed within the range of parameters changed, the ICERs remained > $150,000. All above indicated that D-VMP regimen was not cost-effectiveness.

This study was the first cost-effectiveness analysis of daratumumab-based regimens for TNE NDMM. Therefore, it cannot be compared with other results of similar studies. There were only a few cost-effectiveness studies on daratumumab-based treatments of RRMM. Zeng et al. (

There are some limitations in our study. First, ALCYONE (

From the perspective of US payers and based on $150,000 as a willingness to pay threshold, adding daratumumab to bortezomib, melphalan and prednisone for untreated multiple myeloma was estimated not to be cost-effective.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.

YC and LZ led the development of this cost-effectiveness analysis, and contributed to writing of the draft manuscript. WC and TZ contributed to the study design and methodology. All authors contributed to interpretation of the results and approved the final version.

Supported in part by Sanming Project of Medicine in Shenzhen (SZSM201801060).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.