Cost-effectiveness of lutetium (177Lu) oxodotreotide vs everolimus in gastroenteropancreatic neuroendocrine tumors in Norway and Sweden

Jayne Palmer, Oscar R Leeuwenkamp

Jayne Palmer, Ossian Health Economics and Communications, Basel 4051, Switzerland

Oscar R Leeuwenkamp, Advanced Accelerator Applications, Geneva 1204, Switzerland

Abstract BACKGROUND Gastroenteropancreatic neuroendocrine tumors (GEP-NETs) represent a relatively rare and heterogenous group of tumors. Currently available treatment options for patients with progressive GEP-NETs include lutetium (177Lu) oxodotreotide (177Lu-Dotatate) and everolimus [as well as sunitinib for patients with pancreatic NETs(P-NETs)].AIM To perform a health economic analysis to determine the cost-effectiveness of 177Lu-Dotatate compared with everolimus in patients with unresectable or metastatic midgut-NETs or P-NETs in both Sweden and Norway.METHODS Simulations were performed using a three-state partitioned survival model and analyses were performed separately for patients with midgut-NETs and P-NETs.Clinical input data were sourced from an indirect comparison that utilized survival data from clinical trials of 177Lu-Dotatate and everolimus. The analyses were performed from the healthcare payer perspective over a time horizon of 20 years. For Sweden, future costs and clinical outcomes were discounted at 3% per annum. For Norway, a discount rate of 4% per annum was applied.RESULTS For Sweden, improved survival outcomes and higher lifetime costs with 177Lu-Dotatate resulted in an incremental cost-effectiveness ratio (ICER) of SEK 391194 per quality-adjusted life year (QALY) gained for midgut NETs and SEK 16764 per QALY gained for P-NETs for 177Lu-Dotatate compared with everolimus. For Norway, the corresponding ICERs were NOK 244444 per QALY gained and NOK 106451 per QALY gained, respectively. One-way sensitivity analyses revealed that the results were most sensitive to changes in drug acquisition costs and health state utility values.CONCLUSION In both Sweden and Norway, from a healthcare provider perspective, 177Lu-Dotatate is likely to be considered cost-effective relative to everolimus for the treatment of patients with unresectable or metastatic, progressive midgut-NETs or P-NETs.

Key Words: Cost; Cost-effectiveness; Neuroendocrine tumors; 177Lu-Dotatate; Sweden;Norway

INTRODUCTION

Neuroendocrine tumors (NETs) are a heterogenous group of tumors that may arise from any cell with a neuroendocrine phenotype, although around 70% of NETs emanate from cells in the gastrointestinal tract or pancreas, collectively termed gastroenteropancreatic (GEP)-NETs[1]. Recent estimates relating to the prevalence and incidence of NETs or GEP-NETs specifically in Sweden are lacking, but in Norway in 2010 the overall incidence of all NET types was estimated at 21.3 per 100000 population and the median age at diagnosis was 65 years[2]. Moreover, analysis of temporal trends in incidence has also shown that in Norway over the period 1993–2010 the incidence of NETs increased by 5.1% per year in women and 2.1% per year in men[2], this increase has been attributed, in part, to increased physician and pathologist awareness of NETs, improved diagnostic techniques and improved classification[2,3].With regard to the incidence of GEP-NETs specifically, this has been estimated at 5.25 per 100000 population[4]. For Sweden, this corresponds to an estimated 540 new cases per year (based on a population of 10.28 million[5]) and for Norway the corresponding figure is approximately 280 new cases per year (based on a population of 5.30 million[6]). Based on these estimates, GEP-NETs fulfil the European Medicines Agency (EMA)criteria for orphan disease status[7]. Additionally, the tumor biology and malignant potential of GEP-NETs is highly variable and prognosis is influenced by the location of the primary tumor, Ki-67 (a marker of cell proliferation rate), tumor size and tumor grade, metastases, and disease stage at diagnosis[8]. Indeed, in Norway over 40% of patients already had distant metastases at diagnosis[2]and 5-year survival for patients with pancreatic NETs (P-NETs) with distant disease at diagnosis was just 33%[2].Similarly, 5-year survival rates for patients with small intestinal NETs with distant disease at diagnosis were 63% for tumors classified as low/intermediate aggressiveness[9].

Currently available treatment options for patients with unresectable or metastatic GEP-NETs include everolimus, sunitinib (only indicated for patients with welldifferentiated P-NETs with disease progression) and177Lutetium oxodotreotide (177Lu-Dotatate; Lutathera).177Lu-Dotatate was approved for use in Europe in 2017 and is a peptide receptor radionuclide therapy (PRRT) specifically targeting tumors overexpressing somatostatin receptors, particularly of type 2 (SSRT2) on the surface of tumor cells of most NETs[10].177Lu-Dotatate is radiolabeled with the β-emitter177Lu,which has a maximum penetration range of 2.2 mm and binds with high affinity to SSRT2, thereby delivering targeted cytotoxic radiation to tumor cells overexpressing SSRT2 with minimal damage to neighboring healthy tissue[11]. In the phase III NETTER-1 trial, conducted in patients with advanced, progressive, somatostatinreceptor-positive midgut NETs177Lu-Dotatate plus best supportive care (BSC) with octreotide long-acting release (LAR) 30 mg was associated with a statistically significant benefit in both progression-free survival (PFS) and overall survival (OS)compared with octreotide LAR 60 mg[12]. In the primary analysis median [95%confidence interval (CI)] PFS was not yet reached in the177Lu-Dotatate group compared with 8.4 (5.8–9.1) months in the octreotide LAR 60 mg group and the hazard ratio (95%CI) for disease progression or death for177Lu-dotatate-treated patients was 0.21 (0.13–0.33) (P< 0.001)[12]. In parallel with improved survival outcomes, treatment with177Lu-Dotatate also significantly delayed deterioration in quality of life (QoL)relative to octreotide LAR 60 mg[13]. Further, the authors of a 2015 review on QoL in patients with GEP-NETs noted that whilst QoL has been shown to improve in patients treated with PRRT (irrespective of the radionuclide used), it remains relatively unchanged from baseline in patients undergoing treatment with everolimus[14].

In addition to clinical efficacy, novel treatments must also demonstrate costeffectiveness relative to currently available treatments. The acquisition costs of novel treatments, particularly those used in oncology or for orphan indications are often higher than for the current standard of care. As such, healthcare policy makers and payers are increasingly stipulating that for granting reimbursement novel treatments must demonstrate not only better safety and efficacy, but also long-term costeffectiveness relative to the current standard of care. On a European level, spending on cancer drugs has increased substantially in recent years. In 2005, spending on cancer drugs in the EU totaled EUR 7.6 billion and accounted for 12% of total healthcare spending on cancer, by 2014 these figures had increased to EUR 19.1 billion and 23%,respectively[15]. Consequently, to maintain access and equity throughout healthcare systems the acquisition costs associated with novel cancer treatments have been subject to increasing scrutiny from payers and policy makers[16]. In light of the importance of demonstrating the cost-effectiveness and substantiating the budget impact of new interventions the aim of the current analysis was to investigate the longterm cost-effectiveness of177Lu-Dotatate compared with everolimus in populations of patients with unresectable or metastatic, progressive P-NETs and midgut NETs in both Sweden and Norway (analyses were performed based on health economic models utilized in successful reimbursement submissions for177Lu-Dotatate in both Sweden and Norway). In addition, budget impact evaluations for Norway, Sweden and Denmark are presented to reflect the potential influence of the novel treatment options on the health care budget.

MATERIALS AND METHODS

Model structure

Long-term cost-effectiveness analysis was performed using a three-state partitioned survival model developed in Microsoft Excel, wherein the three states were PFS, postprogression survival and death, and analyses were performed separately for patients with midgut NETs and patients with P-NETs. This approach allows for independent modeling of PFS and OS based on the use of parametric survival models to extrapolate clinical data beyond the time frame of clinical studies[17]. A series of parametric survival functions (including exponential, Weibull, Gompertz, log-logistic and lognormal models) were fitted to the survival data obtained from the NETTER-1[12]and ERASMUS studies (AAA, data on file). The most appropriate function was determined based on a combination of visual inspection, Akaike’s Information Criterion and the Bayesian Information Criterion as well as both clinical and biological plausibility. On this basis the Weibull model, was considered to be the most appropriate and best fit(for both PFS and OS) and was utilized in the base case analysis (log-normal and loglogistic functions were excluded on the basis that both have long survival tails considered to be biologically implausible). Further, the Weibull model is flexible as has the advantage that the associated hazard rate is not constant over time in contrast to the exponential survival model.

In the analysis a cycle length of 4 wk was used and half-cycle correction was applied(note that177Lu-Dotatate is administered once every 8 wk for a total of four courses and is therefore integrated into the model every second cycle).

Clinical input data

For177Lu-Dotatate-treated patients, survival data were sourced from the NETTER-1 trial for patients with midgut-NETs and from the ERASMUS study for patients with PNETs. The analysis for Norway was performed prior to that for Sweden, consequently for the Swedish analysis updatedpost-hocsurvival data were used. For the Norwegian analysis, survival data from the primary analysis were used aspost-hocsurvival data were not yet available at the time the Norwegian analysis was performed. Baseline cohort characteristics for the simulated patient cohort were based on the NETTER-1 trial and for patients treated with everolimus survival data were sourced from the RADIANT-4[18]trial in patients with gastrointestinal-NETs (GI-NETs) and the RADIANT-3[19]trial in patients with P-NETs. For the survival modeling approach, a hazard ratio was applied to the baseline risk curve to compare each active treatment to a control so that177Lu-Dotatate and everolimus could be compared indirectlyviaa common comparator assuming constant proportionality.

For both treatments, missed doses or dose modifications due to toxicity were accounted for using relative dose intensity (RDI) and missed doses were also taken into account in terms of the acquisition costs for each treatment. For everolimus, an RDI of 79.4% was assumed based on data from the RADIANT-4 trial[18]. For177Lu-Dotatate, RDI refers to the amount of the drug that is actually administered relative to the planned number of four doses, the RDI may be below 100% if patients either miss a dose or have the dose modified due to toxicity or an adverse event. A complete course of177Lu-Dotatate consists of 4 doses, but based on the findings of the ERASMUS study an RDI of 84.4% was assumed, which corresponds to a mean of 3.4 doses per patient.

Only grade 3-4 adverse events (AEs) were conservatively included in the analysis.For177Lu-Dotatate-treated patients the frequency of AEs was sourced from the NETTER-1 trial. For everolimus-treated patients the frequency of AEs was sourced from the RADIANT-4 study[18]for midgut-NETs and the RADIANT-3 study[19]for patients with P-NETs.

Resource utilization and costs

Drug acquisition costs and resource utilization costs associated with drug administration specific to each setting (e.g.,for177Lu-Dotatate specialist physician,biochemist and nurse as well as general inpatient and outpatient costs) were accounted in the analysis. For Sweden, drug acquisition costs were sourced from FASS(Pharmaceutical specialties in Sweden)[20]and for Norway drug acquisition costs were sourced from the Norwegian Medicines Agency[21](Table 1). The costs of routine monitoring in each setting (e.g., CT scans, blood testsetc.) were also captured in the analysis (resource use was country-specific owing to differences between health care systems in terms of monitoring frequency). Applying a conservative assumption, costs associated with palliative care were not captured in the analysis.

The cost of AEs was calculated as a single cost based on a weighted probability of AEs in each treatment arm. For Sweden, costs for individual AEs were not identified,therefore Norwegian AE costs were used as a proxy and converted to SEK, which included a conversion factor that corrected for the relative cost level in specialist care and the average exchange rate in the previous 12 mo. For Norway, AE costs were calculated based on DRG codes. All costs are presented in 2017 values.

Health state utilities

In the base case analysis, for Sweden, the health state utility value for the PFS state for patients with midgut-NETs was sourced from a real-world study in patients with GINETs conducted in the United Kingdom (AAA, data on file; Table 2). For Norway, the health state utility value for the PFS state in midgut-NET patients was sourced from NETTER-1 study. For both Sweden and Norway, utility values for all other health states were sourced from the real-world ERASMUS study conducted in the Netherlands. In both the NETTER-1 and ERASMUS studies, QoL was assessed using the European Organization for Research and Treatment of Cancer QLQ-C30 questionnaire[22]and subsequently mapped to the EQ-5D using the algorithm developed by Longworthet al[23], 2014).

Disutilities associated with treatment-related AEs were sourced from publishedliterature[24-27](Table 3). QoL decrements associated with palliative/end of life care were not included in the analysis.

Table 1 Drug acquisition costs for gastrointestinal-neuroendocrine tumors and pancreatic-neuroendocrine tumor

Table 2 Utilities used in the base case analysis

Time horizon, discount rate and perspective

For both Sweden and Norway, the analysis was performed from the perspective of the healthcare payer and the time horizon was 20 years reflecting a life-time analysis given the average age of NET patients at time of diagnosis. In line with national recommendations, for Sweden both future costs and clinical outcomes were discounted at a rate of 3%per annum, and for Norway a discount rate of 4%per annumwas applied for both costs and outcomes.

Table 3 Disutilities for grade 3 and 4 adverse events

Sensitivity analysis

A series of one-way sensitivity analyses were performed to determine key drivers of outcomes. For Sweden, the impact of changing key input values by ± 20% was explored. For Norway key input parameters were varied by ± 30%. Sensitivity analyses were performed around drug acquisition costs, health state utility values,efficacy, resource use and the overall disutility associated with AEs.

Budget impact analysis

Budget impact analyses were performed to estimate the budget impact of introducing177Lu-Dotatate into clinical practice in Sweden and Norway. For both countries, the treatment eligible GEP-NET population was based on national population figures,incidence estimates published by ?berget al[4](2012) and it was assumed that 40% of patients had metastasis at baseline[28]. Market share projections for each country were based on estimates from the manufacturer of177Lu-Dotatate (AAA, data on file). In both Sweden and Norway, it was assumed that177Lu-Dotatate would replace everolimus as a treatment option. Budget impact analyses were performed over a time horizon of 5 years, in line with good practice[29]and national guidance[30]and costs were not discounted.

RESULTS

Cost-effectiveness of 177Lu-Dotatate in Sweden

In patients with midgut NETs, treatment with177Lu-Dotatate was associated with a projected gain in quality-adjusted life expectancy of 0.48 quality-adjusted life years(QALYs) relative to everolimus (3.45 QALYs for177Lu-Dotatatevs2.97 QALYs for everolimus), which was driven by longer PFS and OS with177Lu-Dotatate (Table 4).Total lifetime costs were SEK 189818 higher with177Lu-Dotatate compared with everolimus (SEK 1,170,731vsSEK 980913), which resulted in an incremental cost effectiveness ratio (ICER) of SEK 391194 (EUR 36154, September 2019 exchange rates)per QALY gained.

For patients with P-NETs, the difference in projected quality-adjusted life expectancy between the two treatments was more pronounced. Treatment with177Lu-Dotatate resulted in a mean quality-adjusted life expectancy of 4.90 QALYs compared with 3.55 QALYs for patients treated with everolimus (difference 1.35 QALYs)(Table 4). Total mean lifetime costs were also SEK 22498 higher with177Lu-Dotatate(SEK 1398533vsSEK 1376034), resulting in an ICER of SEK 16764 (EUR 1549) per QALY gained.

One-way sensitivity analyses showed that for patients with midgut NETs, the findings of the analysis were most sensitive to changes in drug acquisition costs for both177Lu-Dotatate and everolimus as well as assumptions around health state utility values for the PFS and PPS states (Figure 1). Similarly, for P-NET patients the ICER was most sensitive to changes in drug acquisition costs and changes in the total weighted disutility value for AEs experienced by everolimus-treated patients.

Cost-effectiveness of 177Lu-Dotatate in Norway

For patients with midgut-NETs, treatment with177Lu-Dotatate led to a gain in qualityadjusted life expectancy of 1.36 QALYs relative to the use of everolimus (3.15 QALYs for177Lu-Dotatatevs1.79 QALYs for everolimus). Total mean lifetime costs were NOK 331995 higher with177Lu-Dotatate compared with everolimus (NOK 985062vsNOK 653066), which was primarily driven by the higher acquisition costs associated with177Lu-Dotatate (Table 4). The incremental gain in quality-adjusted life expectancy combined with higher lifetime costs resulted in an ICER of NOK 244444 (EUR 24416;September 2019 exchange rates) per QALY gained for177Lu-Dotatatevseverolimus.

For patients with P-NETs, the incremental gain in quality-adjusted life expectancy associated with the use of177Lu-Dotatate compared with everolimus was 2.10 QALYs(5.31 QALYs for177Lu-Dotatatevs3.21 QALYs for everolimus) (Table 4), which was driven by the notably longer PFS and OS projected for patients treated with177Lu-Dotatate. Lifetime costs were NOK 223487 higher with177Lu-Dotatate than with everolimus (NOK 1280254vsNOK 223487). This led to an ICER of NOK 106451(EUR 10633) per QALY gained.

One-way sensitivity analyses revealed that for patients with midgut-NETs costeffectiveness was most sensitive to changes in drug acquisition costs for both177Lu-Dotatate and everolimus and also to the utility value assigned to the PFS health state(Figure 2). Similarly, for P-NET patients results were most sensitive to changes in drug acquisition costs and changes in assumptions around efficacy.

Budget impact of 177Lu-Dotatate

In Sweden it was assumed that the market share of177Lu-Dotatate for the treatment of patients with GEP-NETs would increase from 30% in Year 1 to 90% in Years 4 and 5(Table 5). In Year 1, the total budget impact associated with the introduction of177Lu-Dotatate was SEK 17.3 million (EUR 1.6 million), which increased to SEK 52.2 million(EUR 4.8 million) in Years 4 and 5 (Table 5). Over a 5-year period the total budget impact associated with the introduction of177Lu-Dotatate was SEK 202 million (EUR 19 million). In Norway, the market share of177Lu-Dotatate was assumed to remain at 50%over the 5-year time horizon of the analysis and the annual budget impact associated with the introduction of177Lu-Dotatate was NOK 23.5 million (EUR 2.4 million). Over a 5-year time horizon the total budget impact was estimated at NOK 117 million(EUR 11.8 million).

An additional budget impact analysis was also performed for Denmark. In Denmark, it was assumed that177Lu-Dotatate would wholly replace lanreotide as a treatment option for GEP-NETs (and that 60 GEP-NET patients per year would be treated with177Lu-Dotatate) and the annual budget impact was estimated at DKK 27.2 million (EUR 3.6 million), which led to a total budget impact over 5 years of DKK 136 million (EUR 18.2 million).

DISCUSSION

Overall, the findings of long-term cost-effectiveness analyses suggest that in both Sweden and Norway,177Lu-Dotatate is associated with substantial gains in quality-adjusted life expectancy and higher lifetime costs in patients with unresectable/metastatic midgut-NETs or P-NETs. In particular, in both countries177Lu-Dotatate was most cost-effective in patients with P-NETs and was associated with an ICER below EUR 11000 per QALY gained relative to everolimus in both settings. There is no official willingness-to-pay (WTP) threshold in Sweden, in Norway an unofficial threshold of NOK 600000 (EUR 59939) per QALY gained is often cited[31]and in high income European countries a WTP threshold of EUR 50000 per QALY gained is also frequently cited. Consequently, applying these commonly applied thresholds,177Lu-Dotatate is likely to be considered cost-effective for the treatment of patients with unresectable or metastatic midgut NETs or P-NETs in both Sweden and Norway.

Table 4 Summary of base case results

Although177Lu-Dotatate was shown to be cost-effective in both midgut-NETs and PNETs the findings in patients with P-NETs may be particularly salient to healthcare payers as historically 5-year survival rates in patients with P-NET have been low relative to patients with GI-NETs[32]. In Sweden, total direct medical costs are also substantially higher for patients with P-NETs compared with patients with small intestine NETs. In 2013, annual per patient direct medical costs were EUR 24800 for patients with small intestine-NETs compared with EUR 37300 for patients with PNETs[33]. Further, overall 54% of direct costs for patients with GEP-NETs were attributable to cancer drugs and on a national level, EUR 5.1 million was spent on pharmacologic agents for the treatment of patients with GEP-NETs[33]. In terms of budget impact, in Sweden and Denmark overall healthcare spending on cancer in 2014 was EUR 2.7 billion and EUR 1.2 billion, respectively (similar data for Norway were not available)[15]. As such, the budget impact associated with the introduction of177Lu-Dotatate is likely to have only a minimal impact on overall spending on pharmacologic agents used in cancer treatment.

The findings of the cost-effectiveness analyses presented here align with previously published findings from the National Institute for Health and Care Excellence(NICE)[34]and the Scottish Medicines Consortium[35]in the United Kingdom as well as findings published by the Tandv?rds-och l?kemedelsf?rm?nsverket (TLV; Dental and Pharmaceutical Benefits Agency) in Sweden. In particular, the NICE assessment reported that for patients with GI-NETs, when compared with everolimus or bestsupportive care177Lu-Dotatate was considered to be cost-effective and associated with an ICER below the threshold of GBP 30000 per QALY gained. ICERs were also below GBP 30000 per QALY gainedvseverolimus, sunitinib or best supportive care in patients with P-NETs, with the NICE appraisal also noting that for P-NETs177Lu-Dotatate met the criteria for end of life care of extending life expectancy by at least 3 mo[34]. Additionally, the TLV conducted cost-minimization rather than costeffectiveness analysis and reported that whilst177Lu-Dotatate was associated with higher costs than everolimus in patients with GI-NETs it was associated with lower costs than both everolimus and sunitinib in patients with P-NETs[36].

Table 5 Summary findings of budget impact analysis

A key limitation of the current analysis is the use of clinical input data obtainedviaan indirect comparison. Although everolimus was considered to be the most relevant comparator for177Lu-Dotatate, to date, there are no head-to-head trials that directly compare the efficacy and safety of177Lu-Dotatate with everolimus in patients with midgut NETs or P-NETs. Consequently, this necessitated the use of data sourced from indirect comparisons, which is inherently associated with a degree of uncertainty. The findings of the indirect comparison used here align with those of a recent systematic review and meta-analysis of177Lu-Dotatatevseverolimus in advanced P-NETs,wherein177Lu-Dotatate was associated with a significantly higher objective response rate, disease control rate and PFS relative to everolimus[37]. Additionally, although the analysis included the incidence of grade 3–4 AEs it did not account for long-term persistent hematologic dysfunction that has recently been reported in a small proportion of GEP-NET patients undergoing treatment with177Lu-Dotatate. A recent real-world analysis reported that 4% of177Lu-Dotatate-treated GEP-NET patients experienced persistent hematologic dysfunction, occurring at a median of 41 mo posttreatment[38], the costs and clinical implications of which should be considered in future analyses.

Figure 1 Summary findings of one-way sensitivity analyses for Sweden. A: One-way sensitivity analyses in midgut neuroendocrine tumors; B: One-way sensitivity analyses in pancreatic neuroendocrine tumors. QALY: Quality-adjusted life year; PPS: Post-progression survival; PFS: Progression-free survival; GI-NETs:Gastrointestinal-neuroendocrine tumors; LAR: Long-acting release; CT: Computed tomography; MRI: Magnetic resonance imaging; P-NET: Pancreaticneuroendocrine tumor; AE: Adverse event.

A further limitation in the analysis in patients with P-NETs is that only everolimus was included as a relevant comparator, based on clinical practice utilization patterns and input from local clinicians. In Europe, the tyrosine kinase inhibitor sunitinib is indicated for use in the treatment of patients with unresectable or metastatic, well differentiated P-NETs experiencing disease progression[39]. However, as with everolimus, head-to-head clinical data comparing177Lu-Dotatate with sunitinib are currently lacking. Evidence from two different indirect comparisons, reported that sunitinib and everolimus were associated with similar PFS and OS in P-NET[40,41]. Drug acquisition costs for everolimus and sunitinib were also similar in both Sweden and Norway. It is therefore feasible that given the similar efficacy profile and costs the costeffectiveness profile of177Lu-Dotatatevssunitinib would be similar to that of177Lu-Dotatatevseverolimus.

Additionally, although Nordic guidelines[3]relating the treatment of patients with GEP-NETs were published prior to the approval of177Lu-Dotatate, European guidelines advocate the use of PRRT as a treatment option for patients with metastatic midgut-NETs or P-NETs[28]and the findings of the NETTER-1 and ERAMUS trials lend further credence to the clinical evidence base for these recommendations.

For consistency, the analyses in both Sweden in Norway were performed from the healthcare payer perspective. In Sweden reimbursement decisions are based on analyses performed from a societal perspective, which takes into account indirect costs associated with lost productivity. However, the mean age of patients in the simulated cohort was 63.7 years. In Sweden, the minimum retirement age is 61 years and the mean age at retirement was 64.5 years in 2016[42]. As such is it unlikely that the inclusion of indirect costs associated with lost productivity in this population would have a notable impact on the conclusions of the analysis. Further, both PFS and OS were projected to be longer for177Lu-Dotatate patients, suggesting that the inclusion of indirect costs would likely further reduce the ICER for177Lu-Dotatatevseverolimus and as such the analysis was conservative.

Figure 2 Summary findings of one-way sensitivity analyses for Norway. A: One-way sensitivity analyses in midgut neuroendocrine tumors; B: One-way sensitivity analyses in pancreatic neuroendocrine tumors. QALY: Quality-adjusted life year; PPS: Post-progression survival; PFS: Progression-free survival; GI-NETs:Gastrointestinal-neuroendocrine tumors; LAR: Long-acting release; CT: Computed tomography; MRI: Magnetic resonance imaging; P-NET: Pancreaticneuroendocrine tumor; AE: Adverse event.

CONCLUSION

Overall, the findings of life-long cost-effectiveness analyses for both Sweden and Norway indicate that owing to improved survival outcomes177Lu-Dotate is likely to be cost-effective and associated with minimal budget impact relative to everolimus for the treatment of patients with unresectable or metastatic progressive midgut-NETs or P-NETs.

ARTICLE HIGHLIGHTS

Research background

177Lu oxodotreotide (177Lu-Dotatate) is a peptide receptor radionuclide therapy that was approved for use in Europe in 2017. Findings from the phase III NETTER-1 trial,conducted in patients with advanced, progressive, somatostatin-receptor-positive midgut neuroendocrine tumors (NETs) showed that the use of177Lu-Dotatate in combination with best supportive care with octreotide long-acting release (LAR) 30 mg resulted in a significant benefit in terms of both progression-free survival (PFS) and overall survival (OS) compared with octreotide LAR 60 mg.

Research motivation

Health systems frequently have to operate within finite budgetary constraints;therefore, it is important to establish not only the efficacy but also the long-term cost effectiveness of new treatments as they are introduced into the healthcare system.

Research objectives

The objective of the analysis was to establish the long-term cost-effectiveness of177Lu-Dotatate compared with everolimus in patients with unresectable or metastatic midgut-NETs or pancreatic-NETs (P-NETs) in both Sweden and Norway.

Research methods

Cost-effectiveness analysis was performed using a three-state partitioned survival model, which allows for the independent modeling of PFS and OS and the extrapolation of clinical data beyond the time frame of a clinical trial. In the absence of head-to-head trials clinical input data used in the model for the P-NET analysis were sourced from an indirect treatment comparison using input data from trials of177Lu-Dotatate and everolimus.

Research results

In both Sweden and Norway, treatment with177Lu-Dotatate was associated with an incremental gain in quality-adjusted life expectancy relative to everolimus for both midgut-NETs and P-NETs, which was driven by projected improvements in PFS and OS associated with the use of177Lu-Dotatate. For Sweden, improved survival outcomes and higher lifetime costs with177Lu-Dotatate resulted in an incremental costeffectiveness ratio (ICER) of SEK 391194 per quality-adjusted life year (QALY) gained for midgut NETs and SEK 16764 per QALY gained for P-NETs for177Lu-Dotatate compared with everolimus. For Norway the corresponding ICERs were NOK 244444 per QALY gained for midgut NETs and NOK 106,451 per QALY gained for P-NETs.

Research conclusions

In both Sweden and Norway,177Lu-Dotatate is likely to be considered cost-effective relative to everolimus for the treatment of patients with unresectable or metastatic,progressive midgut-NETs or P-NETs when considered from the perspective of the healthcare provider.

Research perspectives

Based on clinical data derived through indirect comparison177Lu-Dotatate is likely to be a cost-effective treatment option relative to everolimus for patients with midgut NETs or P-NETs based in Sweden or Norway.