A Cost-Effectiveness Analysis for Avelumab as a First-Line Maintenance Treatment of Advanced Urothelial Carcinoma in the Netherlands

Abstract

Introduction

Advanced or metastatic urothelial carcinoma (UC) is an incurable disease with a high disease burden and a poor prognosis. Avelumab as first-line (1L) maintenance treatment is an innovative therapy option for patients with advanced or metastatic UC that has not progressed after 4–6 cycles of 1L platinum-based chemotherapy. This study aimed to assess the cost-effectiveness of avelumab maintenance treatment plus best supportive care (BSC) versus BSC alone from a Dutch societal perspective.

Methods

A partitioned survival model was developed incorporating JAVELIN Bladder 100 trial data to inform overall and progression-free survival, adverse events incidence, and health-state utilities. Costs for drugs, healthcare resource use, adverse events, and indirect costs were obtained from national databases, the Dutch costing manual, and published literature. Assumptions were validated by clinical experts. An incremental cost-effectiveness ratio (ICER) was determined using lifetime incremental costs and quality-adjusted life years (QALY).

Results

Avelumab 1L maintenance treatment plus BSC was estimated to have €48,186 discounted incremental costs and 0.63 discounted incremental QALYs versus BSC alone, leading to a base-case ICER of €76,450, supported by consistent scenario and sensitivity analyses.

Conclusion

Avelumab 1L maintenance treatment is likely to be a cost-effective treatment in advanced or metastatic urothelial carcinoma in the Netherlands.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40487-025-00345-3.

Key Summary Points

This study aimed to assess the cost-effectiveness of avelumab maintenance treatment plus best supportive care (BSC) versus BSC alone from a Dutch societal perspective.

A partitioned survival model was developed incorporating JAVELIN Bladder 100 trial data to inform overall and progression-free survival, adverse events incidence, and health-state utilities.

Costs for drugs, healthcare resource use, adverse events, and indirect costs were obtained from national databases, the Dutch costing manual, and published literature. Assumptions were validated by clinical experts.

Avelumab 1L maintenance treatment plus BSC was estimated to have €48,186 discounted incremental costs and 0.63 discounted incremental QALYs versus BSC alone, leading to a base-case ICER of €76,450, supported by consistent scenario and sensitivity analyses.

Avelumab 1L maintenance treatment is likely to be a cost-effective treatment in advanced or metastatic urothelial carcinoma in the Netherlands.

Introduction

Bladder cancer is the 10th most common cancer worldwide, with an estimated 613,000 new cases and 220,000 deaths in 2022 [1, 2]. Urothelial carcinoma (UC) is the most common subtype of bladder cancer, accounting for 90% of all bladder cancer cases [1, 2]. In the Netherlands, approximately 6800 people are diagnosed with UC each year [3], and incidence rates are threefold higher for men than for women [3].

In patients with locally advanced or metastatic UC, the cancer has spread to lymph nodes or other distant organs. Patients with metastatic UC who are considered fit for chemotherapy typically are treated with first-line (1L) cisplatin- or carboplatin-based chemotherapy [4, 5]. Those who respond to these regimens are observed until disease progression which typically occurs within approximately 9 months; during this observation period, patients also receive best supportive care (BSC) [6]. After progression, therapy options would include checkpoint inhibitors like pembrolizumab. Median overall survival (OS) with platinum-containing chemotherapy ranges from 11 to 13 months in Dutch clinical practice [7]. Updated guidelines include enfortumab vedotin in combination with pembrolizumab, as well as platinum-based chemotherapy plus gemcitabine in combination with nivolumab.

Avelumab is a human immunoglobin G1λ monoclonal antibody that inhibits the immune checkpoint programmed cell death ligand 1 (PD-L1). Avelumab was investigated in JAVELIN Bladder 100, a phase 3, multicenter, multinational, randomized, open-label, parallel-arm efficacy and safety trial (NCT02603432) [6]. JAVELIN Bladder 100 studied avelumab 1L maintenance plus BSC vs BSC alone as treatment for locally advanced or metastatic UC that had not progressed with 1L platinum-based chemotherapy (four to six cycles of gemcitabine plus cisplatin or gemcitabine plus carboplatin). Patients were treated with avelumab 1L maintenance until disease progression or unacceptable toxicity [8].

In the primary analysis (data cutoff 19 October 2021) of JAVELIN Bladder 100, avelumab plus BSC demonstrated a statistically significant and clinically meaningful benefit in OS (stratified hazard ratio [HR] 0.69; 95% confidence interval [CI] 0.56, 0.86; p = 0.001) and progression-free survival (PFS; stratified HR 0.62; 95% CI 0.52, 0.75; p < 0.0001) compared to BSC alone [6]. Long-term follow-up (> 2 years in all patients) data have been published and are in line with the findings of the primary analysis [9]. On the basis of the results from JAVELIN Bladder 100, avelumab received approval from the US Food and Drug Administration in June 2020 and from the European Commission (EC) in January 2021 [10]. Following a positive recommendation from medical oncologists and the Dutch Health Care Institute (ZIN), avelumab was reimbursed in the Netherlands in 2021 [11, 12]. Enfortumab vedotin in combination with pembrolizumab and platinum-based chemotherapy plus gemcitabine in combination with nivolumab are not reimbursed in the Netherlands currently.

By comparing the incremental costs and health outcomes of novel therapies against existing standards of care, cost-effectiveness analyses help policy-makers allocate limited healthcare budgets [13, 14]. Cost-effectiveness is one of the criteria weighed by the Dutch Health Care Institute for reimbursement decision-making. A cost-effectiveness model was developed to support the reimbursement discussions in the Netherlands for avelumab 1L maintenance plus BSC versus BSC alone. This included subsequent treatments such as checkpoint inhibitors.

The objective of the current study was to evaluate the cost-effectiveness of avelumab plus BSC compared with BSC alone at a willingness-to-pay (WTP) threshold of €80,000 per quality-adjusted life year (QALY) using a Dutch societal perspective. 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.

Methods

Population and Interventions

The model estimated the cost-effectiveness of avelumab plus BSC compared with BSC alone. Prior to the introduction of avelumab, BSC after 1L platinum-based chemotherapy was considered standard clinical practice in the Netherlands. The model evaluated patients with locally advanced or metastatic UC who have not progressed after four to six cycles of 1L platinum-based chemotherapy. This population is consistent with the patients enrolled in the JAVELIN Bladder 100 trial [6]. Patient baseline characteristics in the model such as age, sex, body weight, body surface area, and glomerular filtration rate were obtained from the JAVELIN Bladder 100 trial [6].

Model Framework

A survival partition model in Microsoft Excel (version 2502) was developed with three primary health states: progression free, post progression (PP), and death. Extrapolated PFS and OS curves derived from the JAVELIN Bladder 100 trial [6] were used to calculate the proportion of patients in each health state over time; the model did not require explicit transition probabilities between health states.

The model estimates the probability of being in each of the three health states according to a 7-day cycle. PFS curve estimates provided the probability of a patient being in the progression-free health state, and the estimates from the OS distribution represented the patient’s likelihood of survival at each time point. The difference between the OS and PFS curves yielded the probability of the patient being in the post-progression health state. See Online Resource 1 for an overview of the model structure. A lifetime time horizon of 20 years was set and discount rates of 4% for costs and 1.5% for effects were applied, which aligned with Dutch guidelines [14].

Clinical Inputs

Survival extrapolations were fit to the OS and PFS (per blinded independent central review) Kaplan–Meier (KM) data from the JAVELIN Bladder 100 trial to determine lifelong OS and PFS probabilities [6]. The extrapolation was conducted according to National Institute for Health and Care Excellence (NICE) guidelines [15, 16]. Individual fits with the log-normal distribution were applied for OS. For PFS, a piecewise model was applied with a split at 8 weeks. The Gompertz distribution was chosen for both arms for the 0- to 8-week period and the log-normal distribution was selected for both arms for the period > 8 weeks. The OS and PFS extrapolations are presented in Figs. 1 and 2.

Fig. 1.

Overall survival

Fig. 2.

Progression-free survival

Other clinical inputs such as health-state utilities (mapped to the Dutch setting [17], see Table 1), treatment duration, and incidence of grade ≥ 3 adverse events were also derived from JAVELIN Bladder 100 data. Treatment duration was modelled through extrapolation of the time to treatment discontinuation endpoint in the JAVELIN Bladder 100 trial [6]. Avelumab treatment costs stopped at 2 years, in line with Dutch clinical practice for immunotherapies based on clinical expert input [18].

Table 1.

Health-state utilities

	Health-state utility (95% CI)
Progression free and > 30 days until death	0.7890 (0.7770, 0.8010)
Progression and > 30 days until death	0.7210 (0.1080, 0.7350)
Progression free and ≤ 30 days until death	0.5670 (0.5220, 0.6110)
Progression and ≤ 30 days until death	0.4990 (0.4550, 0.5420)
Death	0

CI confidence interval

Cost Inputs

A Dutch societal perspective was selected, and therefore both direct and indirect costs were considered in the cost-effectiveness analysis [14]. The modelled inputs included direct costs associated with drug acquisition, drug administration, healthcare resource use, subsequent treatments, end-of-life care, adverse events, and informal care. Additional healthcare costs in life years (LY) due to unrelated diseases were considered in a scenario analysis using the Practical Application to Include Future Disease Costs tool [19]. Productivity losses were not considered in the analysis since the median age of patients in the JAVELIN Bladder 100 trial (avelumab plus BSC, 68 years; BSC alone, 69 years) [6] was above the Dutch retirement age. Table 2 gives an overview of the cost inputs in the cost-effectiveness model and Table 3 presents the subsequent treatment costs per treatment arm based on the distribution in the JAVELIN Bladder 100 trial. A higher proportion of patients in the BSC arm received checkpoint inhibitors (mostly pembrolizumab), leading to higher costs.

Table 2.

Cost inputs

	Drug cost per infusion, €	Administration cost per infusion, €	Dosing schedule	Source
Drug costs
Avelumab 800 mg (IV)	2605.36	192.93	2 infusions per 4 weeks	Z-index, Franken et al. 2018 [28], avelumab summary of product characteristics [8]
BSC	0	0	NA	Assumption

	Unit cost, €	PFS frequency	PD frequency	Source
Resource costs
Oncologist visit	305.47	1 per month	1 per month	ZIN costing manual
Home-care	70.85		1 per month	ZIN costing manual
GP home visit	61.08		1 per month	ZIN costing manual
Terminal care costs	6545.71	Upon death		NZA DBC zorgproducten
Informal care costs	17.10	6 h	10 h	ZIN costing manual
Travel expenses	6.91	1 per month	1 per month	ZIN costing manual

BSC best supportive care, DBC drug treatment combination, GP general practitioner, IV intravenous, NA not applicable, NZA Nederlandse zorgautoriteit, OS overall survival, PD progressive disease, PFS progression-free survival, ZIN Zorginstituut Nederland

Table 3.

Subsequent therapy costs

	Avelumab + BSC	BSC alone
Total subsequent treatment cost on progression	€10,239.40	€35,848.88

BSC best supportive care

Source: Z-index and JAVELIN Bladder 100 study

Model Outputs

The time in each health state was estimated on the basis of the OS and PFS extrapolations from the JAVELIN Bladder 100 trial [6]. Both costs and utilities were assigned to each health state. LYs, QALYs, and total costs per treatment arm were estimated for both the avelumab plus BSC and BSC alone arms. The incremental costs and QALYs were then used to estimate the incremental cost-effectiveness ratio (ICER).

Several scenario analyses, univariate sensitivity analyses, and a probabilistic sensitivity analysis were conducted to test the robustness of the model and address any uncertainty in the estimates. The probabilistic analysis was performed with 5000 iterations.

Results

Base-Case Analyses

The results of the base-case analysis are presented in Tables 4 and 5. The discounted incremental costs of avelumab plus BCS vs BSC alone are estimated to be €48,186. The main driver of incremental costs was the difference in drug acquisition costs, where avelumab plus BSC was associated with an increase in costs in the maintenance phase but a reduction in costs of subsequent treatment. As a result of the improved OS and PFS, avelumab plus BSC led to an increase in LYs and QALYs vs BSC alone. The discounted incremental LYs and QALYs are estimated to be 0.80 and 0.63, respectively, leading to an ICER of €76,450 per QALY. The WTP threshold for avelumab plus BSC was estimated to be €80,000 per QALY gained based on a proportional shortfall of 0.88 [20].

Table 4.

Results deterministic analysis

	Avelumab + BSC	BSC	Difference
Discounted (costs 4.0%, effects 1.5%)
Total LYs	3.52	2.72	0.80
Total QALYs	2.62	1.99	0.63
Total costs, €	110,734	62,548	48,186
ICER, €/LYs	60,113
ICER, €/QALYs	76,450
Undiscounted (0% costs, 0% effects)
Total LYs	3.75	2.88	0.88
Total QALYs	2.80	2.11	0.69
Total costs, €	118,628	67,822	50,806
ICER, €/LYs	57,934
ICER, €/QALYs,	73,788

BSC best supportive care, ICER incremental cost-effectiveness ratio, LY life year, QALY quality-adjusted life year

Table 5.

Cost breakdown avelumab + BSC vs. BSC alone (discounted 4%)

Costs, €	Avelumab + BSC	BSC	Incremental
Drug acquisition (first-line)	57,364	0	57,364
Drug administration (first-line)	4226	0	4226
Adverse event management	239	153	86
Medical resource use	14,474	11,931	2544
Subsequent treatment	4983	24,703	− 19,720
End of life	5735	5955	− 220
Informal care giving	23,293	19,597	3696
Travel	419	208	210
Total	110,734	62,548	48,186

BSC best supportive care

Sensitivity Analyses

Figure 3 shows a tornado diagram of the univariate sensitivity analysis results, where the upper and lower estimates of the most influential parameters are plotted. The ICERs range from approximately €65 to €90,000, indicating a medium impact of extreme values. The parameters with the greatest influence on the ICER were post-progression informal care costs, discount rate for benefits, and parameters relating to subsequent treatment costs. Informal care costs have a particularly high impact as they are varied for the arms separately, instead of per health state. The upper limit for discounting benefits (5.0%) gives the highest value (€91,634) because incremental long-term survival is an important driver of cost-effectiveness. Residual uncertainty was explored through several sensitivity analyses in which parametric distributions were varied; results were reasonably consistent with those of the base-case analysis (Table 6). The most influential scenarios were a time horizon of 10 years (€95,307) and applying trial relative dose intensity to avelumab (€62,434), demonstrating sensitivity of the ICER to long-term extrapolation of OS and to avelumab costs.

Fig. 3.

Tornado diagram of univariate sensitivity analysis results

Table 6.

Scenario analysis results

Scenario	Incremental QALYs (discounted)	Incremental costs (discounted), €	ICER, €/QALY
Healthcare perspective	0.63	44,298	70,282
Time horizon of 10 years	0.49	46,761	95,307
Combined OS fit: log-normal (best statistical fit)	0.60	48,556	80,831
OS individual fits (best statistical fits)  Avelumab + BSC: log-normal  BSC alone: log-normal	0.66	49,339	74,908
PFS with avelumab + BSC: generalized gamma (second statistical fit)	0.63	48,211	76,529
PFS with BSC alone: log-logistic (second statistical fit)	0.63	48,243	76,642
TTD: treat to progression (including 2-year treatment stop)	0.63	44,746	70,992
Relative dosing intensity from clinical trial applied to avelumab costs	0.63	39,352	62,434

BSC best supportive care, ICER incremental cost-effectiveness ratio, OS overall survival, PFS progression-free survival, QALY quality-adjusted life year, TTD time to discontinuation

Probabilistic sensitivity results are shown in Fig. 4. The curves show the probability that avelumab plus BSC is cost-effective across increasing WTP thresholds. On the basis of the probabilistic analysis, which explored uncertainty around the included parameters, there was a 55% chance that avelumab plus BSC is cost-effective vs BSC alone at a WTP of €80,000 per QALY gained as can be seen in the cost-effectiveness acceptability curves.

Fig. 4.

Cost-effectiveness acceptability curves

Discussion

On the basis of an ICER of €76,450 per QALY and a 55% probability of being below €80,000 per QALY, this study shows that implementation of avelumab is likely to be a cost-effective innovation for Dutch clinical practice, although there is uncertainty. Sensitivity analyses showed the certainty around the results; univariable sensitivity analyses and scenario analyses generally fell below the €80,000 per QALY threshold for cost-effectiveness. Consequently, these analyses can promote sustainable healthcare decisions in the Netherlands by assessing clinical innovations at the Dutch cost-effectiveness threshold.

The results of the present analysis are consistent with those of other published cost-effectiveness analyses of avelumab as a 1L maintenance treatment in advanced or metastatic UC. In Critchlow et al. 2023, avelumab was found to be a cost-effective treatment in Scotland [21]. As a result of differences in perspective and costs, ICERs cannot be compared between studies; however, the approach for extrapolating survival and the results on incremental QALYs (0.63 discounted QALYs) were very similar to those in the Dutch analysis. Similarly, for Taiwan, Su et al., 2023, concluded that avelumab was cost-effective, based on 0.61 discounted incremental QALYs from a Taiwanese payer perspective [22]. In three other published analyses, patient-level data from the JAVELIN Bladder 100 study were available, and the authors came to slightly different conclusions from a US and Chinese perspective [23–25]. However, conclusions from these studies were similar: maintenance avelumab 1L maintenance plus BSC in metastatic UC is more effective but also more costly than treatment with BSC alone.

This cost-effectiveness analysis should be considered with the following limitations. The survival data from the JAVELIN Bladder 100 trial were immature, needing the use of parametric survival extrapolations to estimate the long-term survival of avelumab plus BSC vs BSC alone, which causes uncertainty. To mitigate this uncertainty, extrapolation methods followed NICE guidance [15, 16], and the results were validated by Dutch clinical experts. Additionally, parameter uncertainty was explored through probabilistic sensitivity and scenario analyses showing relatively consistent results with the deterministic base case.

Another limitation of the analysis is the underlying assumption of generalizability of the JAVELIN Bladder 100 trial to Dutch clinical practice. Generalizability and external validity are a general limitation of international clinical trials [26]. An example of potential differences between the JAVELIN Bladder 100 and Dutch clinical practice is the international variation in post-progression standard of care. Although 70% of those patients who went on to 2L treatment in the BSC arm received a checkpoint inhibitor (anti-PD-(L)1) as subsequent treatment, this percentage might be higher in Dutch clinical practice. Because differences in subsequent treatments would affect both incremental costs and effects, it is not possible to determine the exact impact on the ICER. The cost-effectiveness of checkpoint inhibitors in second-line UC was not assessed in the Dutch setting, complicating estimating the impact on the ICER. More expensive standard of care might decrease the ICER because of lower incremental costs, but an improved post-progression survival in the BSC arm would increase the ICER. However, a recent analysis after the introduction of avelumab shows that this variation is likely limited. Of patients diagnosed between 2018 and 2021 in the Netherlands, 34.1% (159/466 patients) of all patients starting 1L platinum-based chemotherapy received 2L checkpoint inhibitors [27]. This translates to 73.9% receiving a 2L checkpoint inhibitor of all those patients who went on to receive any 2L therapy (159/216 patients) accounting for the fact that, comparable to the JAVELIN Bladder 100 study, many patients did not receive any subsequent treatment [27].

The analysis presented in this study can provide a point of reference for future cost-effectiveness analyses of avelumab or new innovations in advanced or metastatic UC in the Netherlands. Prospective single-arm multicenter nationwide studies assessing the efficacy of avelumab in advanced or metastatic UC at a shorter treatment duration (e.g., Ave-short study) are ongoing in the Netherlands as there is an increasing focus on the efficient use of resources in oncology treatments. Additionally, there have been further advancements in the treatment of patients with previously treated metastatic UC, with the approval of enfortumab vedotin monotherapy by the EC, and with other novel therapies for the treatment of patients with previously untreated metastatic UC, after the start of the JAVELIN Bladder 100 study.

Conclusion

This analysis demonstrated that maintenance therapy with avelumab plus BSC may be a cost-effective option for patients with advanced or metastatic UC at a WTP threshold of €80,000 per QALY gained. On the basis of probabilistic sensitivity analysis, avelumab had a 55% chance of being cost-effective at this WTP threshold. This analysis can act as a reference for cost-effectiveness analyses for new innovations in advanced or metastatic UC in the Netherlands.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

All authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this manuscript, take responsibility for the integrity of the work as a whole, and have given their approval for this version to be published. David Smalbrugge, Tim Walsteijn, Jeantine de Feijter, Britt Suelmann, Mairead Kearney, Agnes Benedict, Venediktos Kapetanakis, and Sophie van Beekhuizen contributed to the study design, writing, review, and approval of this manuscript.

Funding

This study was sponsored and funded by Merck B.V., Schiphol-Rijk, Netherlands, an affiliate of Merck KGaA (CrossRef Funder ID: 10.13039/100009945).

Data Availability

The authors declare that all the data supporting the findings of this study are available within the article and the Supplementary Material.

Declarations

Conflict of Interest

David Smalbrugge and Sophie van Beekhuizen are full-time employees of Cytel. Tim Walsteijn is a full-time employee of Merck B.V., Schiphol-Rijk, Netherlands, an affiliate of Merck KGaA. Mairead Kearney is a full-time employee of Merck Healthcare KGaA, Darmstadt, Germany. Agnes Benedict and Venediktos Kapetanakis are full-time employees of Evidera. JdF has participated in advisory boards for Janssen. Britt Suelmann has participated in medical advisory boards for MSD, Merck, BMS, IPSEN, Janssen, AstraZeneca, Pfizer and has been a study sponsor of Pfizer, Janssen, Astellas and AstraZeneca.

Ethical Approval

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.