Mirvetuximab soravtansine in platinum-resistant recurrent ovarian cancer with high folate receptor-alpha expression: a cost-effectiveness analysis

Abstract

Objective

Mirvetuximab soravtansine (MIRV), a new antibody-drug conjugate, versus the investigator’s choice of chemotherapy (IC) was the first treatment to demonstrate benefits for progression-free and overall survival in platinum-resistant recurrent ovarian cancer (PROC) with high folate receptor-alpha (high-FRα) expression. Efficacy, safety, and economic effectiveness make MIRV the new standard of care for these patients.

Methods

Based on patients and clinical parameters from MIRASOL (GOG 3045/ENGOT-ov55) phase III randomized controlled trials, the Markov model with a 20-year time horizon was established to evaluate the cost and efficacy of MIRV and IC for PROC with high-FRα expression, considering the bevacizumab-pretreated situation from the American healthcare system. Total cost, life-years (LYs), quality-adjusted life-years (QALYs), incremental cost-effectiveness ratio (ICER), and incremental net health benefits were the main outcome indicators and compared with willingness-to-pay threshold of $100,000/QALY. Sensitivity and scenario analyses were conducted.

Results

Compared with the IC, MIRV was associated with incremental costs of $538,251, $575,674, and $188,248 with the corresponding QALYs (LYs) increased by 0.90 (1.55), 1.09 (1.88), and 0.53 (0.79), leading to ICERs of $596,189/QALY ($347,995/LY), $530,061/QALY ($306,894/LY), and $1,011,310/QALY ($680,025/LY) in the overall, bevacizumab-naïve, and bevacizumab-pretreated patients, respectively. When MIRV is reduced by more than 75%, it may be a cost-effective treatment.

Conclusion

At the current price, MIRV for PROC with high-FRα expression is not the cost-effective strategy in the US. However, its treatment has higher health benefits in bevacizumab-naïve patients, which is likely to be an alternative.

Synopsis

Mirvetuximab soravtansine as a treatment for platinum-resistant recurrent ovarian cancer with high folate receptor-alpha is not the cost-effective strategy in the USA, providing objective data reference for clinicians’ clinical and national health insurance decision-making.

Graphical Abstract

INTRODUCTION

Ovarian cancer (OC) is the deadliest and second most prevalent cancer of the female reproductive tract, with 19,710 diagnoses and 13,270 deaths forecast in the USA in 2023 alone [1]. Of these, approximately 90% are diagnosed as epithelial ovarian carcinoma cases including high-grade epithelial ovarian, fallopian tube, or primary peritoneal cancers [2,3]. An estimated 70% of OC cases are diagnosed when the disease is already advanced, and these patients commonly experience recurrence following platinum-based chemotherapy, exhibiting a poor 30% 5-year survival rate [3]. Platinum-resistant OC (PROC) develops in almost all patients who experience such recurrence, as defined by tumor recurrence fewer than 6 months following the most recent platinum treatment [4]. At present, the primary therapeutic strategies approached for PROC patient treatment consist of non-platinum monotherapy or combination bevacizumab treatment, although patients administered these regimens exhibit a median progression-free survival (PFS) or just 3.4–6.7 months and a median overall survival (OS) of just 10.9–13.3 months [5,6,7]. Systemic treatment also tends to be inefficient and associated with substantial toxicity such that there remains a pressing need for alternative therapeutic options for platinum-resistant cancer patients.

Biomarker-based patient selection has been found to offer value in patients with platinum-sensitive recurrent OC, enabling the individualized selection of poly ADP-ribose polymerase (PARP) inhibitors, for example, based on patient BRCA or homologous recombination deficiency status [8]. Strikingly, many epithelial tumors exhibit the overexpression of the membrane protein folate receptor-alpha (FRα) responsible for folate binding and internalization, as defined by the positive immunohistochemical (IHC) staining of ≥75% of viable tumor cells, with this being particularly true in serous endometrial cancer and high-grade serous OC tumors [9,10,11,12]. Mirvetuximab soravtansine (MIRV; Elahere™; ImmunoGen Inc., Waltham, MA, USA) is an antibody-drug conjugate (ADC) that consists of an FRα-specific antibody, a cleavable linker domain, and a payload consisting of the potent tubulin-binding maytansinoid DM4 [13,14]. MIRV engages a unique antitumor mechanism of action termed a ‘bystander effect’ [14,15,16]. In the phase III MIRASOL (NCT04209855) randomized controlled trial (RCT), MIRV was associated with the significant prolongation of the OS (overall patients: median OS, 16.46 vs. 12.75 months; hazard ratio, 0.67; 95% confidence interval, 0.50–0.89; p=0.046; Bevacizumab-naïve patients: median OS, 20.2 vs. 14.4 months; 0.51; 0.31–0.86; 0.0099; Bevacizumab-pretreated patients: median OS, 15.4 vs. 10.9 months; 0.74; 0.54–1.04; 0.0789) and PFS (overall patients: median PFS, 5.62 vs. 3.98 months; 0.65; 0.52–0.81; <0.0001; Bevacizumab-naïve patients: median PFS, 7.0 vs. 5.6 months; 0.66; 0.46–0.94; 0.021; Bevacizumab-pretreated patients: median PFS, 4.4 vs. 3.0 months; 0.64; 0.49–0.84; 0.0011) in PROC patients with high folate receptor-alpha (high-FRα) expression compared to investigator’s choice of chemotherapy (IC) [12,17]. The survival benefits associated with MIRV treatment offer new opportunities to improve OC patient outcomes such that it has received approval from the US Food and Drug Administration (FDA) and is included in the National Comprehensive Cancer Network (NCCN) guidelines as an option to treat FRα-positive, PROC patients that have been administered 1–3 prior systemic treatment regimens [5,13].

While MIRV treatment offers clear clinical advantages, this pioneering therapy is associated with inevitably increased healthcare costs and no economic analyses to date have directly evaluated the relative benefits of employing MIRV as a treatment for FRα-high PROC patients. There is a growing need to rigorously assess the cost-effectiveness of innovative treatment regimens to guide clinical decision-making about the selection of limited potential patient populations, enable the appropriate pricing of these novel drugs by healthcare policymakers, and improve patient acceptance of these emerging treatment strategies. As such, this study was developed to assess the cost-effectiveness of MIRV as an alternative to IC when used to treat PROC with high-FRα expression from the perspective of US healthcare payers.

MATERIALS AND METHODS

The safety and efficacy data used to guide the present analysis were derived from the phase III MIRASOL (GOG 3045/ENGOT-ov55) RCT, which had a data cut-off date of March 6, 2023, and additional detailed data were obtained from the American Society of Clinical Oncology (ASCO) meeting abstract and ImmunoGen Inc. [12,17]. The present analysis was conducted as per the Consolidated Health Economic Evaluation Reporting Standards 2022 (CHEERS 2022) guideline for economic evaluations (Table S1) [18].

1. Population and treatment details

The enrolled patient population for this analysis consisted of FRα-high PROC patients with 1–3 prior treatments, including patients who had undergone prior bevacizumab and PARP inhibitor treatment, and patients harboring BRCA mutations [12,17]. In total, 453 PROC patients were randomly assigned to the MIRV (n=227, 50%) or IC (paclitaxel [n=92, 41%], pegylated liposomal doxorubicin [PLD; n=81, 36%], or topotecan [n=53, 23%]) groups. Patients were also classified based on whether or not they had undergone prior bevacizumab administration, including bevacizumab-naïve (n=172, 38%) and bevacizumab-pretreated (n=281, 62%) patients [12,17]. All patients were administered MIRV (6 mg/kg every 3 weeks) or IC (paclitaxel [80 mg/m² every 3 weeks], PLD [50 mg/m² every 4 weeks], or topotecan [1.5 mg/m²/d on days 1–5 every 3 weeks]) based on their group assignments, with the regimens administered to these patients being formulated based on corresponding RCTs and associated guidelines (Table S2) [5,12,17]. When calculating chemotherapy doses for this study, these patients were all assumed to be 63-year-old females with a height of 170 cm, a weight of 70 kg, and a body surface area of 1.84 m² (Table 1) [19]. When progressive disease (PD) arises in 14% and 24% of patients in the MIRV and IC groups, respectively, these patients are administered best supportive care (BSC) (Table 1) [12,17]. Terminal care was assumed to be administered to all patients before death [5].

Table 1. Clinical and health parameters.

Variables			Baseline value (range)	Reference	Distribution
Clinical parameters
	Weibull survival model for OS
		IC	Scale=0.018380, Shape=1.428240	[12,17]	NA
		MIRV	Scale=0.021565, Shape=1.205684	[12,17]	NA
	Weibull survival model for PFS
		IC	Scale=0.182480, Shape=1.075620	[12,17]	NA
		MIRV	Scale=0.121220, Shape=1.075170	[12,17]	NA
	Rate of post-discontinuation therapy
		IC	0.270 (0.216–0.324)	[12,17]	Beta
		MIRV	0.140 (0.112–0.168)	[12,17]	Beta
	Risk for main AEs in IC group
		Neutropenia	0.570 (0.456–0.684)	[12,17]	Beta
		Anemia	0.420 (0.336–0.504)	[12,17]	Beta
		Thrombocytopenia	0.270 (0.216–0.324)	[12,17]	Beta
		Nausea	0.050 (0.040–0.060)	[12,17]	Beta
	Risk for main AEs in MIRV group
		Keratopathy	0.090 (0.072–0.108)	[12,17]	Beta
		Blurred vision	0.080 (0.064–0.096)	[12,17]	Beta
	Body weight, Kg		70 (35–105)	[19]	Uniform
	Body surface area, Meters2		1.840 (1.472–2.208)	[19]	Uniform
Health parameters
	Utility and disutility
		Utility of PFS	0.750 (0.600–0.900)	[20]	Beta
		Utility of PD	0.500 (0.400–0.600)	[20]	Beta
		Disutility of anemia	0.073 (0.058–0.088)	[21,22]	Beta
		Disutility of neutropenia	0.090 (0.072–0.108)	[21,22]	Beta
		Disutility of nausea	0.048 (0.038–0.058)	[22]	Beta
		Disutility of thrombocytopenia	0.020 (0.016–0.024)	[23]	Beta
		Disutility of blurred vision	Unreported	NA	NA
		Disutility of keratopathy	Unreported	NA	NA
	Discount rate		0.03 (0–0.05)	[24,25]	Uniform

AE, adverse event; IC, investigator’s choice of chemotherapy; MIRV, mirvetuximab soravtansine; NA, not applicable; OS, overall survival; PD, progressive disease; PFS, progression-free survival.

2. Construction and summarize model

To gauge the cost-effectiveness of MIRV relative to IC, the TreeAge Pro program (version 2021) was used to develop a Markov model with three mutually exclusive health states (PFS, PD, and death) to simulate the disease course in FRα-high PROC patients. All patients were assumed to be in the PFS state when entering the trial, with conversion to either PD status or death throughout the study (Fig. S1). This model terminates at a lifetime horizon of ~20 years, after which all patients were expected to be deceased. Each model cycle was 1 year in length, and an annual discounting rate of 3% was applied to costs and survival simulations for all treatment regimens [24,25].

Survival data for this model were extracted with the GetData Graph Digitizer program (v 2.26) from OS and PFS Kaplan-Meier curves in the original RCT. Transition probabilities between health states, the probability of death, and long-term survival data fitting were assessed based on the extracted survival data. Five different survival models were evaluated to assess the fit of these time-event data, including the Weibull, Log-logistic, Log-normal, Exponential, and Gompertz distributions [26]. Based on visual examination and the Akaike information criterion and Bayesian information criterion, the Weibull distribution was ultimately determined to be the most appropriate extrapolated distribution using MATLAB (v R2020a) (Fig. S2 and Table S3). R Studio (v 4.2.2) was used to extract key parameters including shape (λ) and scale (γ) (Table 1), and the transition probability over time was computed as follows:

(1-exp {λ(t-u)^γ-λt^γ} (λ>0, γ>0)

Where u and t were the Markov period and the current model period, respectively [21].

3. Utility and cost estimates

The health states in the developed Markov model were assigned health utility values based on the stage of disease progression. As the MIRASOL (GOG 3045/ENGOT-ov55) RCT did not include health utility data, quality-of-life (QoL)-related data could not be directly retrieved. Instead, average utility values of 0.75 and 0.50 were assigned to PFS and PD, respectively, as in a prior report (Table 1) [20]. Grade 3 or higher adverse events (AEs) affecting > 5% of patients were assigned a disutility value to correct the average utility value of PFS (Table 1) [21,22,23].

As this study was conducted from the perspective of the healthcare sector, only the following direct estimated costs were taken into consideration: costs of drugs, laboratory testing, intravenous administration, tumor imaging, IHC staining, severe AE management, BSC, and terminal care (Table 2). Drug costs were obtained from the Centers for Medicare & Medicaid Services [27], while all other costs were from prior publications [22,24,26,28,29,30,31]. Costs were adjusted to 2023 US dollars based on the consumer price index [32].

Table 2. Cost estimates.

Parameters		Baseline value (range)	Reference	Distribution
Drug cost, $/cycle
	MIRV	110,547 (88,438–132,656)	[27]	Gamma
	Paclitaxel	72 (58–86)	[27]	Gamma
	PLD	69 (55–83)	[27]	Gamma
	Topotecan	276 (221–331)	[27]	Gamma
Cost of AEs
	IC	27,836 (22,269–33,403)	[22,26]	Gamma
	MIRV	984 (787–1,181)	[28]	Gamma
Terminal care per patient		56,210 (44,968–67,452)	[20]	Gamma
Administration per cycle		167 (134–200)	[20]	Gamma
Immunohistochemical tests per patient		122 (98–146)	[29]	Gamma
Laboratory and tumor imaging per cycle		482 (386–578)	[30]	Gamma
Best supportive care per cycle		18,846 (15,077–22,615)	[31]	Gamma

AE, adverse event; IC, investigator’s choice of chemotherapy; MIRV, mirvetuximab soravtansine; PLD, pegylated liposomal doxorubicin.

4. Cost-effectiveness analyses

Primary study outcomes included total costs, life years (LYs), quality-adjusted LYs (QALYs), incremental cost-effectiveness ratio (ICER) values, and incremental net health benefit (INHB). ICERs were computed based on the relative cost-effectiveness of the two compared strategies and were expressed as the incremental cost per QALY obtained. These values were compared to a willingness-to-pay (WTP) threshold of $100,000/QALY [24,25]. A treatment was considered to not be cost-effective as per World Health Organization (WHO) guidelines if the ICER was greater than the WTP threshold. INHB was calculated as follows:

INHB=(u_EMIRV-u_EIC)-(u_CMIRV-u_CIC)/WTP

where u_E and u_C respectively correspond to the efficacy and costs associated with MIRV or IC [33]. Given that novel therapeutic regimens have high costs, the potential cost-effectiveness of MIRV was also assessed at 100%, 70%, 50%, 30%, and 10% of current prices.

5. Sensitivity analyses

One-way sensitivity analysis and probabilistic sensitivity analysis (PSA) approaches were used to gauge the robustness of study results and to identify those variables that significantly impacted the results of this analysis among 24 different input categories in the overall, bevacizumab-naïve, and bevacizumab-pretreated patient groups, with results being presented in the form of a tornado diagram [24]. All parameters were varied within ±20% of base values (Tables 1 and 2). PSA approaches were conducted via Monte Carlo simulations with 10,000 iterations based on appropriate variable ranges and parameter distributions, sampling each model input randomly from across its probability distribution for each simulation [24,26]. Results were presented in the form of cost-effectiveness scatter plots and acceptability curves [24,26].

6. Ethics approval and consent to participate

This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors, thus it does not require the approval of the independent ethics committee.

RESULTS

1. Base-case results

In the overall patients, the incremental cost and LYs of MIRV were $538,251 and 1.55 LYs (18.60 months) compared to IC, respectively, with an incremental QALYs of 0.90 when QoL was included; In the bevacizumab-naïve patients, the incremental cost and LYs of MIRV were $575,674 and 1.88 LYs (22.56 months) compared to IC, respectively, with an incremental QALYs of 1.09 if QoL was taken into account; In the bevacizumab-pretreated patients, the incremental cost and LYs of MIRV were $188,248 and 0.79 LYs (9.48 months) compared to IC, respectively, with an incremental QALYs of 0.53 if QoL was calculated. The corresponding ICERs were $596,189/QALY, $530,061/QALY, and $1,011,310/QALY, respectively (Table 3). All three ICERs exceeded the WTP threshold of $100,000/QALY, suggesting that the MIRV for PORC with high-FRα expression is not a cost-effective strategy in the US.

Table 3. Cost-effectiveness results.

Cost of MIRV			Incremental cost, $*	Incremental benefits*		ICER*		Comments*	INHB, QALY*
				LYs	QALYs	$/LY	$/QALY
Overall patients
	Full cost (baseline results)		538,251	1.55	0.90	347,995	596,189	Not cost-effective	−4.48
		70% cost	339,703	1.55	0.90	219,163	376,264	Not cost-effective	−2.50
		50% cost	207,338	1.55	0.90	133,767	229,653	Not cost-effective	−1.17
		30% cost	74,973	1.55	0.90	48,370	83,042	Cost-effective	0.15
		10% cost	−57,392	1.55	0.90	Dominant†	Dominant†	Cost-effective	1.47
Bevacizumab-Naïve patients
	Full cost (baseline results)		575,674	1.88	1.09	306,894	530,061	Not cost-effective	−4.67
		70% cost	367,216	1.88	1.09	195,328	338,120	Not cost-effective	−2.58
		50% cost	228,244	1.88	1.09	121,406	210,159	Not cost-effective	−1.19
		30% cost	89,272	1.88	1.09	47,485	82,198	Cost-effective	0.20
		10% cost	−49,700	1.88	1.09	Dominant†	Dominant†	Cost-effective	1.59
Bevacizumab-pretreated patients
	Full cost (baseline results)		188,248	0.79	0.53	680,025	1,011,310	Not cost-effective	−4.87
		70% cost	337,286	0.79	0.53	426,944	631,960	Not cost-effective	−2.84
		50% cost	202,310	0.79	0.53	256,089	379,060	Not cost-effective	−1.49
		30% cost	67,334	0.79	0.53	85,233	126,160	Not cost-effective	−0.14
		10% cost	−67,643	0.79	0.53	Dominant†	Dominant†	Cost-effective	1.21

IC, investigator’s choice of chemotherapy; ICER, incremental cost-effectiveness ratio; INHB, incremental net health benefits; LY, life-year; MIRV, mirvetuximab soravtansine; QALY, quality-adjusted life-year.

^*MIRV versus the IC; ^†Dominant, MIRV showed higher effectiveness and lower cost, as compared with the IC.

2. Sensitivity analyses

As shown in the established tornado diagram (Fig. 1), body weight, MIRV prices, and utility factors were the primary factors that significantly impacted the ICER values when comparing MIRV and IC regimens. ICER values for one-way sensitivity values ranged from $229,653/QALY–$962,708/QALY. The cost of IC, drug administration, and IHC staining had a relatively minimal impact on ICER values. Cost-effectiveness scatter plots indicated that the MIRV regimen exhibited an almost 0% chance of being cost-effective at a US WTP threshold of $100,000/QALY (Fig. S3). The established acceptability curve indicated that the odds of MIRV being cost-effective were strongly correlated with changes in WTP thresholds. At respective WTP thresholds of $580,000/QALY, $530,000/QALY, and $1,030,000/QALY, MIRV exhibited a 50% chance of being cost-effective in the overall, bevacizumab-naïve, and bevacizumab-pretreated patient populations (Fig. 2). On the whole, the results of this sensitivity analysis support the robustness of the established model.

Fig. 1. The One-way Sensitivity Analyses for the MIRV Compared to the IC in Patients with PROC with high-FRα expression.

AE, adverse event; high-FRα, high folate receptor-alpha; IC, investigator’s choice of chemotherapy; ICER, incremental cost-effectiveness ratio; MIRV, mirvetuximab soravtansine; PD, progressive disease; PFS, progression-free survival; PROC, platinum-resistant recurrent ovarian cancer; QALY, quality-adjusted life-year.

Fig. 2. The cost-effectiveness acceptability curves the MIRV compared to the IC in the overall patients (A), the bevacizumab-naïve patients, and the bevacizumab-pretreated patients (C), respectively.

IC, investigator’s choice of chemotherapy; MIRV, mirvetuximab soravtansine; QALY, quality-adjusted life-year.

When the cost of MIRV was reduced by 70%, the ICERs were $83,042/QALY, $82,198/QALY, and $126,160/QALY in the overall, bevacizumab-naïve patients, and bevacizumab-pretreated patients, respectively (Table 3). A 73% reduction in the cost of MIRV was associated with an ICER of $88,225/QALY in bevacizumab-pretreated patients. As such, MIRV would represent a cost-effective treatment option for these patient populations with a 70%–73% cost reduction.

DISCUSSION

OC is the most deadly gynecological malignancy. Three weeks of platinum-based chemotherapy was the primary treatment for advanced OC [5,34]. However, most patients develop platinum-based resistance, so more treatment options are needed, such as PARP inhibitors were recently found to have some efficacy for PROC but limited [8,35]. ADCs and other novel antitumor drugs have recently provided innovative benefits to patients with gynecologic malignancies, but they have also imposed substantial economic burdens on these patients, their families, and the healthcare system. In one report, an estimated 48.5% of patients bearing gynecological tumors were considered to face a high degree of financial toxicity, with this being particularly true for OC patients [36]. Pharmacoeconomic analyses have emerged as an invaluable basis for negotiating drug pricing at the national medical insurance level. Economic evaluations of drugs with marked clinical benefits can aid decision-making related to medical insurance payments and rational drug use. To date, however, cost-effectiveness analyses of ADCs have been restricted to lymphoma and breast cancer patients [29,37,38]. These studies have demonstrated that the economic outcomes associated with ADCs differ across patients with different cancer types. No reports to date, however, have documented the cost-effectiveness of ADCs when used to treat patients with OC. As such, this study based on the MIRASOL trial was developed as the first economic evaluation of MIRV therapy for PROC with high-FRα expression from an American payer perspective.

Based on the Markov model developed herein, MIRV was not found to be a cost-effective alternative to IC in this patient population, yielding an additional 0.90 QALYs at an incremental cost of $538,251 for an ICER of $596,189/QALY. The higher costs associated with MIRV treatment were primarily drug and BSC costs, which were accompanied by prolonged survival and price discussions. In one-way sensitivity analyses, body weight had the greatest impact on model results such that MIRV may be a cost-effective treatment option for patients weighing <22 kg at the $100,000/QALY threshold. This raises ethical concerns, however, regarding the higher rates charged to normal-weight and obese patients for access to life-sustaining treatment. MIRV costs were also based on the number of vials used rather than the actual dose administered to patients in a real-world setting. MIRV costs may thus be reduced based on payments on a per-patient or per-treatment cycle basis, or by arranging for multiple patients to share vials on the same day. The price of MIRV also had a strong impact on model outcomes, and if the cost of MIRV were reduced to under $33,164 per cycle ($19.74 per mg), then it may be a cost-effective therapeutic option. To adapt this innovative treatment strategy to meet international guidelines while maintaining safety concerns, efforts to adjust treatment prices and management programs offer the greatest opportunity to improve the economic prospects of MIRV treatment, expanding its potential for clinical implementation.

Bevacizumab is a monoclonal antibody specific for vascular endothelial growth factor A (anti-VEGF-A) that has been shown to contribute to significant improvements in PROC patient survival, although many patients who underwent such treatment ultimately experience eventual relapse [7,39]. Here, a subgroup analysis focused on prior bevacizumab acceptance revealed that the ICERs of MIRV in bevacizumab-naïve and bevacizumab-pretreated patients were $530,061/QALY (1.09 QALYs), and $1,011,310/QALY (0.79 QALYs), respectively, relative to IC, indicating that MIRV was not cost-effective in either group, but that it exhibited greater efficacy in bevacizumab-naïve patients. This aligns well with the results of a prior phase II RCT conducted by Lee et al., [40] who observed a partial response in 26% of patients who had not undergone prior bevacizumab treatment but no response in bevacizumab-treated patients. It is thus vital that the most effective treatment regimen for a given patient subpopulation be selected in light of both clinical trial data and associated cost-benefit analyses. Decision models should take both drug efficacy and clinical resource availability into account when seeking to improve outcomes for particular groups of cancer patients.

There are many strengths to this analysis that should be highlighted. For one, this is the first cost-effectiveness analysis to our knowledge directly comparing MIRV and IC as treatments for FRα-high PROC patients based on the most recent evidence available. ADCs are emerging as increasingly promising tools that may enable the more effective treatment of a range of solid tumor types. MIRV has received FDA approval and inclusion in international guidelines for FRα-high PROC patient treatment, yet insight into the economic viability of this novel regimen remains limited. Secondly, the clinical and economic outcomes of PROC patients were successfully simulated in this study with a Markov model. Based on an evaluation of five different survival models, the Weibull distribution was found to most effectively fit the available survival data while providing the possibility of more flexible linear extrapolation, enhancing the accuracy of survival results and allowing other research groups to reproduce the present findings. Third, patients were stratified based on whether or not they had undergone prior bevacizumab treatment in order to gauge the implications of prior care on the cost-effectiveness of MIRV. These subgroup-specific economic analyses may better guide patient care in the future. Lastly, owing to differences in the national and medical conditions in different countries, this analysis was specifically conducted from the perspective of the American medical system, offering valuable data that can be leveraged by patients, medical professionals, and the government. The effects of future reductions in MIRV pricing on study results were also assessed, providing a foundation for multilateral drug price negotiations about MIRV availability in the US-related markets.

This study is subject to certain limitations. For one, all survival data were derived from the MIRASOL trial, which was a multi-center trial enrolling patients from throughout the globe. In contrast, the present cost-effectiveness analysis was specific to the US population, and these results may thus be impacted by differences in participant ethnicity. However, US patients did comprise the majority of patients in the original RCT such that the present analyses still exhibit good validity. Secondly, survival beyond the follow-up period was inferred for this analysis, and these results are subject to less certainty than those occurring during the follow-up period. While this has the potential to reduce the overall robustness of this study, a wide range of variables were taken into consideration when conducting sensitivity analyses, and the project results were predicted with appropriate modeling techniques. Third, only, AEs of grade 3 or higher were taken into consideration in the established Markov model, potentially resulting in the underestimation of the costs associated with AE management. Even so, the costs associated with the management of milder AEs tend to be limited, and sensitivity analyses indicated that the effect of these costs on study conclusions was fairly limited. Lastly, the MIRASOL trial did not report on patient QoL such that direct health-related utility data could not be extracted, necessitating the extraction of these data from prior studies. However, the sensitivity analyses suggested that altering these values failed to significantly alter the study conclusions.

The economic evaluation was herein performed to assess the cost-effectiveness of MIRV as a component of the treatment regimen for FRα-high PROC patients. While MIRV is associated with clear clinical benefits, it is also a costly treatment. The present data suggest that at the selected WTP threshold, MIRV would not be regarded as a cost-effective option for patient treatment at an acceptable price. Efforts to reduce the cost of MIRV treatment thus represent a potentially effective means of improving clinical access to this innovative therapy. These results can inform healthcare decision-makers and medical reimbursement policymakers when evaluating the relative advantages of particular regimens in specific patient populations.

SUPPLEMENTARY MATERIALS

Table S1

The CHEERS 2022 checklist

Table S2

Details of drug and unit costs

Table S3

Summary of statistical goodness-of-fit of Kaplan-Meier curve

Fig. S1

Model structure.

Fig. S2

Kaplan-Meier curve fitting and extrapolation.

Fig. S3

Probability sensitivity analysis scatter plot.