Cost-Effectiveness of Initial Versus Delayed Lanreotide for Treatment of Metastatic Enteropancreatic Neuroendocrine Tumors

View More View Less
  • 1 VA Palo Alto Health Care System, Palo Alto, California;
  • 2 Center for Primary Care and Outcomes Research/Center for Health Policy, Department of Medicine, Stanford University School of Medicine, Stanford, California; and
  • 3 Department of Medicine, Stanford University, Stanford, California; and
  • 4 Department of Medical Oncology, Yale School of Medicine, New Haven, Connecticut.

Background: The Controlled Study of Lanreotide Antiproliferative Response in Neuroendocrine Tumors (CLARINET) trial showed prolonged progression-free survival in patients initially treated with lanreotide versus placebo. We evaluated the cost-effectiveness of upfront lanreotide versus active surveillance with lanreotide administered after progression in patients with metastatic enteropancreatic neuroendocrine tumors (NETs), both of which are treatment options recommended in NCCN Clinical Practice Guidelines in Oncology for Neuroendocrine and Adrenal Tumors. Methods: We developed a Markov model calibrated to the CLARINET trial and its extension. We based the active surveillance strategy on the CLARINET placebo arm. We calculated incremental cost-effectiveness ratios (ICERs) in dollars per quality-adjusted life-year (QALY). We modeled lanreotide’s cost at $7,638 per 120 mg (average sales price plus 6%), used published utilities (stable disease, 0.77; progressed disease, 0.61), adopted a healthcare sector perspective and lifetime time horizon, and discounted costs and benefits at 3% annually. We examined sensitivity to survival extrapolation and modeled octreotide long-acting release (LAR) ($6,183 per 30 mg). We conducted one-way, multiway, and probabilistic sensitivity analyses. Results: Upfront lanreotide led to 5.21 QALYs and a cost of $804,600. Active surveillance followed by lanreotide after progression led to 4.84 QALYs and a cost of $590,200, giving an ICER of $578,500/QALY gained. Reducing lanreotide’s price by 95% (to $370) or 85% (to $1,128) per 120 mg would allow upfront lanreotide to reach ICERs of $100,000/QALY or $150,000/QALY. Across a range of survival curve extrapolation scenarios, pricing lanreotide at $370 to $4,000 or $1,130 to $5,600 per 120 mg would reach ICERs of $100,000/QALY or $150,000/QALY, respectively. Our findings were robust to extensive sensitivity analyses. The ICER modeling octreotide LAR is $482,700/QALY gained. Conclusions: At its current price, lanreotide is not cost-effective as initial therapy for patients with metastatic enteropancreatic NETs and should be reserved for postprogression treatment. To be cost-effective as initial therapy, the price of lanreotide would need to be lowered by 48% to 95% or 27% to 86% to reach ICERs of $100,000/QALY or $150,00/QALY, respectively.

Background

Incidence and prevalence of neuroendocrine tumors (NETs) have increased over time. Since the 1980s, incidence has increased by 6.4 times to 7 per 100,000 people in the United States, and 20-year limited-duration prevalence has increased by 8 times to 0.048%.1 Approximately 61% of NETs arise from the gastrointestinal tract, and those that are grades 1 to 2 are relatively indolent, with median overall survival (OS) of approximately 1 decade.1

In asymptomatic patients with low-tumor-burden metastatic gastroenteropancreatic NETs, NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) for Neuroendocrine and Adrenal Tumors2 recommend a somatostatin analog (lanreotide or octreotide) upfront or active surveillance with a somatostatin analog administered after disease progression as first-line therapy options for gastrointestinal NETs and as considerations for pancreatic NETs. Patients in the CLARINET trial (Controlled Study of Lanreotide Antiproliferative Response in Neuroendocrine Tumors) who received lanreotide upfront experienced prolonged progression-free survival (PFS) compared with those who received active surveillance (median, 32.8 vs 18 mo), without a statistically significant difference in OS.3 The CLARINET extension4 followed patients whose disease progressed on placebo and switched to lanreotide, showing a 14-month median PFS. Lanreotide is priced at $7,638 per 28 days or $99,300 per year (January 2020 Medicare Part B average sales price [ASP]–based maximum payment allowance),5 nearly doubling in inflation-adjusted terms over the past decade.6 Because of the indolent nature of NETs, patients often remain on therapy for several years before progression.

As NCCN moves toward incorporating affordability into its evidence base (NCCN Evidence Blocks), it is important to assess the value of recommended therapies.7 This study evaluated the cost-effectiveness of upfront lanreotide versus active surveillance with lanreotide administered after disease progression for patients with grades 1 to 2 nonfunctioning gastroenteropancreatic NETs.

Methods

Study Population

We developed a discrete time semi-Markov model simulating a cohort of patients with advanced, well- or moderately differentiated, nonfunctioning, somatostatin receptor–positive enteropancreatic NETs of grade 1 or 2, 44.1% of whom have tumors of pancreatic origin. Our model followed these patients monthly over their remaining lifetimes.

Treatment Strategies

We evaluated 2 strategies: (1) upfront lanreotide and (2) active surveillance with lanreotide administered after disease progression (“active surveillance”). Outcomes and safety data for the strategies were based on the CLARINET trial and its extension.3,4 Patients began in the “not progressed” state and may remain so, experience disease progression, or die each month (Figure 1). After disease progression, patients began the next line of therapy in the following month. The subsequent therapy sequence was informed by NCCN Guidelines: 177Lu-dotatate, everolimus, followed by sunitinib for patients with pancreatic NETs (Table 1).2

Figure 1.
Figure 1.

Diagram of the Markov model structure. Patients transition from the stable disease/not progressed disease state to progression or death based on transition probabilities derived from the CLARINET trial and the CLARINET extension.3,4 Therapy for progressed disease is outlined in Table 1.

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 18, 9; 10.6004/jnccn.2020.7563

Table 1.

Sequence of Therapies Used in the Model, by Initial Therapya

Table 1.

Model Calibration

We determined rates of progression and preprogression and postprogression survival that produced OS and PFS curves consistent with trial data (Figure 2). We used a previously validated methodology to simulate individual-patient data for the OS and PFS curves.8 To represent the survival curves during the respective trial’s study period and to extrapolate outcomes past the study period, we fit parametric functions to the OS and PFS curves in accordance with the UK’s National Institute for Health and Care Excellence (NICE) recommendations.9 Goodness-of-fit criteria, plausibility of extrapolation, and visual fit were used to choose the parametric survival function in our base case (supplemental eTable 3, available with this article at JNCCN.org). Multiple sensitivity analyses were performed using a wide range of parametric fits (supplemental eTables 5 and 6) to assess the impact of extrapolation assumptions. For patients who underwent active surveillance and received lanreotide after disease progression, subsequent PFS was based on the CLARINET extension.4 We used an optimization-based algorithm programmed using MATLAB R2018b software (MathWorks) to solve for probabilities of progression and preprogression and postprogression mortality. We provide an alternative calibration to examine calibration’s impact on the results in supplemental eAppendix 1.

Figure 2.
Figure 2.

Survival curves. (A) Active surveillance and lanreotide survival curves. Modeled lanreotide and active surveillance OS and PFS curves used in the base case are compared with CLARINET trial curves from Caplin et al.3,4 PFS data for the lanreotide arm have longer follow-up in the extension and thus were used for model calibration. (B) PFS for patients receiving lanreotide after progression while on active surveillance. PFS curves from the CLARINET trial extension4 are plotted with modeled PFS for patients receiving lanreotide after progression while on active surveillance.

Abbreviations: OS, overall survival; PFS, progression-free survival.

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 18, 9; 10.6004/jnccn.2020.7563

Progression rates while receiving subsequent therapies were modeled based on PFS curves from relevant trials (Table 1), whereas mortality rates were informed directly by the CLARINET trial. Sensitivity analyses explored the impact of progression rates on subsequent therapies.

Parametric regression was performed using the flexsurvreg function (flexsurv package version 1.1 in R version 3.3.2; R Core Team, R Foundation for Statistical Computing). We implemented our cost-effectiveness model using TreeAge Pro 2020 release 1.0 software (TreeAge Software). Our methods conform to Consolidated Health Economic Evaluation Reporting Standards10 and the Second Panel on Cost-Effectiveness in Health and Medicine11 (supplemental eTables 16 and 17).

Efficacy and Safety Data

Efficacy data for upfront lanreotide, active surveillance, and lanreotide after progression were derived from the CLARINET trial and its extension.3,4 In the base case, we assumed mortality was equal for both treatment arms based on the CLARINET trial (log-rank test comparing OS curves; P=.88). This assumption was relaxed in scenario analyses (supplemental eTables 5 and 6).

Adverse effect (AE) rates were derived from CLARINET and the CLARINET extension (Table 2) and from trials outlined in Table 1 for relapsed disease therapy. Grades 3 and 4 AEs were included.

Table 2.

Model Parameters Used for Base Case Analysis

Table 2.

Costs

Perspective

We adopted a US Medicare healthcare sector perspective, including all healthcare costs borne by Medicare, Medicare supplemental insurance, and out-of-pocket costs. Costs and quality-adjusted life-years (QALYs) were discounted at 3% annually. When necessary, costs were inflated to 2020 US dollars using the Personal Health Care Expenditure Index12,13 when available (through 2017), the Bureau of Economic Analysis personal consumption expenditure price index14 through 2019 per the Second Panel on Cost-Effectiveness in Health and Medicine,11 and the Consumer Price Index15 to inflate to 2020 US dollars.

Drug Costs

Lanreotide prices were based on the Medicare Part B payment allowance limit (ASP + 6%),5 and administration costs were incorporated. 177Lu-dotatate prices were based on the Medicare Outpatient Prospective Payment System reimbursement rates.16 Everolimus and sunitinib were assumed to be paid through Medicare Part D, and drug costs were set to 64% of average wholesale price, based on the recommendations of the US Department of Veterans Affairs Health Economics Resource Center, which are informed by a US Congressional Budget Office report.17,18 For upfront lanreotide, patients received 120 mg every 28 days until progression. Patients treated with the active surveillance strategy received 120 mg of lanreotide every 28 days starting the month after progression. Dosages for subsequent lines of therapy were based on relevant trials (Tables 1 and 2).

AE Costs

Each grade 3 and 4 AE is assumed to result in a hospitalization incurring costs based on Medicare Severity Diagnosis Related Group payments.19

Additional Healthcare Costs

Additional healthcare costs are age-based per data published by Guy et al.20 Additional monitoring costs are based on NCCN recommendations.2

Utilities

Quality of life (QoL) weights for the base case are based on a publication by Swinburn et al,21 who developed clinical vignettes to describe preprogression and postprogression states for patients with NETs and developed utilities through interviews with the general population using the time trade-off method.22 We repeat the analysis using data published by Meng et al,23 who analyzed data collected from patients during the CLARINET trial using the EORTC Quality-of-Life Questionnaire24 and a mapping algorithm to produce EuroQol 5 dimensions–based QoL weights.25 The data from Swinburn et al21 were chosen for the base case because the Meng et al23 data for the progressed state were based on QoL surveys completed only at a visit 1 month after progression (no subsequent QoL data were collected in the CLARINET trial), and therefore they may not be representative of the average QoL experienced while in the progressed state. The QoL weight for the “not progressed” state was similar between Swinburn and Meng data (0.771 vs 0.776, respectively), but the weight for the progressed state was much lower (0.612 vs 0.726, respectively). Using Swinburn data thus resulted in a greater penalty for progression, favoring the lanreotide arm. Results from additional models using the Meng QoL data are presented in supplemental eTable 5. Each AE incurs a utility decrement based on the expected duration of the AE subtracted from the baseline utility (supplemental eTables 7 and 8).

Sensitivity Analyses

One-way sensitivity analyses were performed on key parameters to determine the impact on our results. We performed extensive sensitivity analyses on survival curve extrapolation by producing 28 different models using 2 calibration methods with various combinations of extrapolated survival curves including scenarios in which OS differed between treatment strategies (supplemental eTables 5 and 6). A probabilistic sensitivity analysis with 10,000 Monte Carlo simulations was performed to determine the effect of uncertainty around model parametric estimates.

Results

In the modeled cohort, the projected median PFS in the upfront lanreotide arm and active surveillance arm were 38.5 and 16.5 undiscounted months, respectively. The projected median OS for both arms was 95 undiscounted months. Supplemental eTable 1 lists the percentage of patients reaching the various lines of therapy and time spent in each line of therapy.

The combination of modeled survival and progression benefits led to 5.21 discounted QALYs (8.98 undiscounted years; 7.50 discounted years) at $804,600 for the upfront lanreotide arm and 4.84 QALYs (8.98 undiscounted years; 7.50 discounted years) at $590,200 for the active surveillance arm (Table 3). The upfront lanreotide arm cost $578,500/QALY gained compared with the active surveillance arm. If the price of lanreotide were decreased to match that of octreotide long-acting release (LAR), upfront lanreotide would cost $482,700/QALY gained.

Table 3.

Cost-Effectiveness Analysis Results From Base Case

Table 3.

Sensitivity Analysis

Because of the uncertainty in extrapolating survival curves beyond the study period, we performed extensive analyses using various combinations of parametric forms to model PFS and OS of the patients undergoing active surveillance, those receiving upfront lanreotide, and those who cross over to lanreotide after progression on active surveillance. We also provide results for an alternative method of calibration (supplemental eAppendix 1). The incremental cost-effectiveness ratios (ICERs) range from $373,000 to $581,600 per QALY gained for the models assuming equal OS between the 2 treatment arms. When upfront lanreotide was assumed to have improved OS compared with active surveillance, the ICERs ranged from $214,000 to $458,300 per QALY gained.

Based on our analyses, the ICER is most sensitive to the price of lanreotide. Substantial decreases in the price of lanreotide would improve its economic value compared with active surveillance followed by lanreotide. Lowering the price of lanreotide by 96% to $370 per 120 mg and by 85% to $1,130 per 120 mg would allow the upfront lanreotide strategy to meet willingness-to-pay (WTP) thresholds of $100,000/QALY gained and $150,000/QALY gained, respectively. This price is lower than publicly available prices for federal pharmaceutical purchasers (Figure 3A).26 Across the extrapolation scenarios, the price of lanreotide would have to be lowered to $370 to $4,000 or $1,130 to $5,600 per 120 mg to meet a WTP threshold of $100,000/QALY gained or $150,000/QALY gained, respectively. Across all extrapolation scenarios in which OS is assumed to be equivalent between strategies, lanreotide’s cost would need to be decreased to <$3,200 or <$4,000 per 120 mg to cost less than $100,000/QALY gained or $150,000/QALY gained, respectively; these results are derived from models that produce results with upper bounds on cost of lanreotide to meet these WTP thresholds (see “Model Calibration Methods” in supplemental eAppendix 1).

Figure 3.
Figure 3.

Sensitivity analyses. (A) Sensitivity to cost of 120 mg of lanreotide. The base case ($7,638) is plotted together with other benchmarks, such as the FSS and Big 4 prices. The current cost to Medicare of 30 mg of octreotide LAR ($6,138) is also highlighted. At costs above the WTP threshold, active surveillance is the preferred strategy. (B) One-way sensitivity analysis of various parameters. The ICER is sensitive to changes in the utilities for the progressed and not progressed states; however, over the 95% CIs for the utilities used in the analysis, those elicited by Swinburn et al,21 the ICER remains above $350,000/QALY gained. (C) PSA results (cost-effectiveness acceptability curve). In the base case, upfront lanreotide is the preferred strategy in 0.4% of iterations for a $100,000/QALY WTP threshold and 3.9% of iterations for a $150,000/QALY WTP threshold. (D) Modeled ICERs over time for lanreotide and octreotide LAR. The ICER increases since 2012 are largely caused by increases in the costs of lanreotide and octreotide LAR, because the cost of lanreotide increased from $3,670 per 120 mg in January 2012 ($4,470 in 2020 US dollars) to $7,638 per 120 mg in January 2020, and the cost of octreotide LAR increased from $3,650 per 30 mg in January 2012 ($4,100 in 2020 US dollars) to $6,200 per 30 mg in January 2020 (ASP plus 6%). For this analysis, 177Lu-dotatate is included as a line of therapy starting in 2018 and excluded before then. When necessary, costs were inflated to 2020 US dollars.

Abbreviations: ASP, average sales price; Big 4 price, price available to 4 largest federal purchasers of pharmaceuticals: US Department of Veterans Affairs, US Department of Defense, US Coast Guard, and US Public Health Service; HR, hazard ratio; FSS, Federal Supply Schedule; ICER, incremental cost-effectiveness ratio; LAR, long-acting release; PSA, probabilistic sensitivity analysis; QALY, quality-adjusted life-year; WTP, willingness to pay.

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 18, 9; 10.6004/jnccn.2020.7563

In one-way sensitivity analyses, varying utilities for health states, costs of subsequent therapies, and costs after the modeled lines of therapy did not alter our conclusions (Figure 3B). Across these analyses, the upfront lanreotide arm cost >$350,000/QALY gained compared with the active surveillance arm.

Using Meng et al23 utilities, QoL in the progressed state is modeled to be better (0.726 vs 0.612 from Swinburn); thus, progression is less negatively impactful. Assuming utilities from Meng et al, the economic value of upfront lanreotide is substantially worse ($1,837,000/QALY) than the base case.

The cost of lanreotide has increased over time, from $3,970 per 120 mg in January 2012 ($4,470 in 2020 US dollars) to $7,638 per 120 mg in January 2020.27 Figure 3C shows the modeled increase in ICER over this time period, from $336,100/QALY gained in 2012 to $578,600/QALY gained in 2020 (in 2020 US dollars).

Although the main results of the analysis are not sensitive to the costs of 177Lu-dotatate, everolimus, and sunitinib, their costs do impact the price at which lanreotide becomes cost-effective. For our base case, with lower costs of 177Lu-dotatate, sunitinib, and everolimus, the price of lanreotide at which upfront lanreotide becomes cost-effective is higher. For our base case analysis, if 177Lu-dotatate, everolimus, and sunitinib costs are decreased by 75%, lanreotide is cost-effective at $880 and $1,640 per 120 mg for the $100,000/QALY gained and $150,000/QALY gained WTP thresholds, respectively (supplemental eTable 13). In a probabilistic sensitivity analysis with 10,000 samples, upfront lanreotide was cost-effective at a WTP of $100,000/QALY gained and $150,000/QALY gained in 0.4% and 3.9% of samples, respectively (Figure 3C).

Discussion

Cancer is among the costliest chronic medical conditions in the United States,28 and costs have continued to increase, with a projected increase of 40% from 2010 to 2020.29 Drug costs paid through Medicare Part B have increased from $13.4 billion in 2005 to $29.1 billion in 2016, reflecting an annualized increase of 7.4%,30 outpacing gross domestic product growth (3.7% annualized over this same period).31 As of 2014, cancer drug costs represented 42% of Medicare Part B drug costs.32 Because of concerns related to the economic impact of the increasing cost of cancer care, there has been recent emphasis on quantifying cost and value.7,33,34 NETs are increasing in incidence and prevalence, with a higher prevalence than esophageal cancer and pancreatic adenocarcinoma combined,35 representing a bigger public health problem than previously recognized. In addition, patients with NETs have long life expectancy compared with those with other cancer types, with an estimated median survival of 9.3 years.35 Shen et al36 previously reported relatively high 5-year costs of care for patients with NETs, thought to be due largely to high costs in the continuing phase of care and longer survival compared with other cancers. Recognition of this illustrates that it is important to study the value of therapy for NETs and all chronic cancers.

Our study shows that, among patients with grades 1 and 2 nonfunctioning gastroenteropancreatic NETs, upfront lanreotide treatment likely results in QoL gains compared with active surveillance followed by lanreotide after progression. However, at current prices, it would not be cost-effective at conventional WTP thresholds. For our base case analysis, the ICER is $578,500/QALY gained.

Based on the CLARINET trial, lanreotide was approved by the FDA in 2014 for the treatment of patients with unresectable, well or moderately differentiated, locally advanced or metastatic gastroenteropancreatic NETs to prolong PFS.3 Patients whose disease progressed on placebo and who crossed over to lanreotide achieved stabilization of their disease and experienced a median of 14 months of PFS in the CLARINET extension.4 Although lanreotide and octreotide LAR have never been directly compared for this indication, both are given equal recommendation in the NCCN Guidelines.2 NCCN Guidelines recommend a somatostatin analogue (lanreotide or octreotide) up front or active surveillance with somatostatin analogue upon progression as a first-line therapy option for gastrointestinal NETs and as a consideration for pancreatic NETs.2

The recently adopted NCCN Evidence Blocks provide guidance on the “affordability” (projected overall cost) of a given treatment course; they do not directly evaluate its value (whether its benefits justify the costs).7 Affordability ratings are based on expert panel members’ knowledge of costs associated with the treatment regimen (efficacy and safety are evaluated similarly as separate dimensions), and Carlson and Jonasch7 indicated that the methodology to determine affordability is still a work in progress. Current NCCN Evidence Blocks designate lanreotide and octreotide as “expensive” treatment options, assigning a 2 on a scale from 1 through 5 (with 1 being least affordable).2 Our study is the first published analysis to directly address the value of upfront lanreotide compared with active surveillance followed by lanreotide. The base case and extensive sensitivity analyses show that although upfront lanreotide may provide an increase in QALYs over active surveillance followed by lanreotide upon progression, at its current Medicare price, its value is poor for this indication—it costs approximately 3 to 5 times the often-cited WTP thresholds for the United States.37,38 Upfront lanreotide would be cost-effective at conventional WTP thresholds with considerable price reductions (96% and 85% decreases to meet <$100,000/QALY gained and <$150,000/QALY gained thresholds, respectively).

It has been noted that postlaunch prices of individual injectable cancer drugs have increased substantially over time.39 Lanreotide has experienced significant price increases in recent years. We examined the value impact of this price increase on the ICER from 2012 through 2020 (Figure 3C). Although it never would have met commonly cited WTP thresholds during this period, the increase in cost has made upfront lanreotide an increasingly poor value.

Our study has some important limitations. We limited our analysis to examine only strategies that incorporate lanreotide as either first-line or second-line therapy, and therefore, we did not evaluate strategies that do not consider lanreotide for therapy in this population; the ICERs and prices at which lanreotide meets the reported WTP thresholds must be interpreted within this context. We believe this approach is justified because these are the 2 strategies recommended by the NCCN Guidelines for this population. Extrapolation beyond the study period for which reported data exist is a challenge for all cost-effectiveness analyses. To address this, we present the results of multiple models that have varying assumptions about extrapolation of PFS and OS. The overall results are not sensitive to the extrapolation assumptions; however, the cost at which lanreotide becomes cost-effective does vary. Other potential limitations result from the use of the placebo arm to model active surveillance. The QoL measurements used (Swinburn et al21 and Meng et al23) do not incorporate a decrement associated with anxiety due to lack of treatment, which may result in a slight overestimate of utilities for the preprogression state in the active surveillance arm of our analysis. However, unrealistically large decreases in the preprogression utility to far below the utility of the progressed state would be needed to alter the conclusion of our analysis (supplemental eFigure 9). Other potential placebo effects, outlined by Hawkins and Scott,40 for which we are unable to account may either increase or decrease the ICER. However, use of the placebo arm for the active surveillance comparator agrees with the manufacturer’s health technology assessment submitted to Australia’s Pharmaceutical Advisory Committee.41 In addition, we did not model the expiration of patents for lanreotide. Lanreotide currently has active patents and marketing exclusivity in the United States.42 We adopted the position of the Institute for Clinical and Economic Review43 that it is difficult to predict future prices of drugs when coming off patent because of unpredictable price trajectory before patent expiration and other factors. We present the ICER if lanreotide were priced similarly to octreotide LAR, which has no active patent protections (see Figure 3A)44; at this price, it would also not be cost-effective at a WTP threshold of $100,000 or $150,000 per QALY gained.

Conclusions

At its current price, our analysis indicates that lanreotide is not cost-effective as initial therapy for patients with metastatic enteropancreatic NETs and that it should be reserved for postprogression treatment. Treatments that offer appreciable clinical benefit but that are not deemed cost-effective put practicing oncologists in a difficult position even if guidelines recommend against them. The importance of further innovation in pricing for value is highlighted by such situations. We do not recommend that this study be used in isolation to make treatment recommendations for patients with NETs; instead, we advise that it be viewed in conjunction with other available evidence to inform updated practice guidelines going forward and as part of the discussion in working toward high-value care and an economically sustainable healthcare system.

References

If the inline PDF is not rendering correctly, you can download the PDF file here.

Submitted October 31, 2019; accepted for publication March 23, 2020.

Previous presentation: This study was presented as an abstract and poster at the North American Neuroendocrine Tumor Society (NANETS) Annual Multidisciplinary NET Disease Symposium; October 6, 2018; Seattle, Washington. Abstract 114.

Author contributions: Study concept and design: All authors. Data collection and assembly: Barnes, Lin, Gupta, Kunz. Data analysis and interpretation: Barnes, Lin, Owens, Goldhaber-Fiebert, Kunz. Manuscript preparation: All authors.

Disclosures: Dr. Kunz has disclosed that she receives grant/research support from Advanced Accelerator Applications, Ipsen, Lexicon, Xencor, and B.R.A.H.M.S., and that she is a scientific advisor for Advanced Accelerator Applications. The remaining authors have disclosed that they have not received any financial consideration from any person or organization to support the preparation, analysis, results, or discussion of this article.

Funding: Drs. Barnes and Lin were supported by funding from the Office of Academic Affiliations, Department of Veterans Affairs, Advanced Fellowship in HSR&D. Dr. Owens was supported by the Department of Veterans Affairs. All views expressed in this article are those of the authors and do not necessarily reflect the views of the Department of Veterans Affairs.

Correspondence: James I. Barnes, MD, MS, Center for Primary Care and Outcomes Research/Center for Health Policy, Department of Medicine, Stanford University School of Medicine, 117 Encina Commons, Stanford, CA 94305. Email: jibarnes@stanfordalumni.org
  • View in gallery

    Diagram of the Markov model structure. Patients transition from the stable disease/not progressed disease state to progression or death based on transition probabilities derived from the CLARINET trial and the CLARINET extension.3,4 Therapy for progressed disease is outlined in Table 1.

  • View in gallery

    Survival curves. (A) Active surveillance and lanreotide survival curves. Modeled lanreotide and active surveillance OS and PFS curves used in the base case are compared with CLARINET trial curves from Caplin et al.3,4 PFS data for the lanreotide arm have longer follow-up in the extension and thus were used for model calibration. (B) PFS for patients receiving lanreotide after progression while on active surveillance. PFS curves from the CLARINET trial extension4 are plotted with modeled PFS for patients receiving lanreotide after progression while on active surveillance.

    Abbreviations: OS, overall survival; PFS, progression-free survival.

  • View in gallery

    Sensitivity analyses. (A) Sensitivity to cost of 120 mg of lanreotide. The base case ($7,638) is plotted together with other benchmarks, such as the FSS and Big 4 prices. The current cost to Medicare of 30 mg of octreotide LAR ($6,138) is also highlighted. At costs above the WTP threshold, active surveillance is the preferred strategy. (B) One-way sensitivity analysis of various parameters. The ICER is sensitive to changes in the utilities for the progressed and not progressed states; however, over the 95% CIs for the utilities used in the analysis, those elicited by Swinburn et al,21 the ICER remains above $350,000/QALY gained. (C) PSA results (cost-effectiveness acceptability curve). In the base case, upfront lanreotide is the preferred strategy in 0.4% of iterations for a $100,000/QALY WTP threshold and 3.9% of iterations for a $150,000/QALY WTP threshold. (D) Modeled ICERs over time for lanreotide and octreotide LAR. The ICER increases since 2012 are largely caused by increases in the costs of lanreotide and octreotide LAR, because the cost of lanreotide increased from $3,670 per 120 mg in January 2012 ($4,470 in 2020 US dollars) to $7,638 per 120 mg in January 2020, and the cost of octreotide LAR increased from $3,650 per 30 mg in January 2012 ($4,100 in 2020 US dollars) to $6,200 per 30 mg in January 2020 (ASP plus 6%). For this analysis, 177Lu-dotatate is included as a line of therapy starting in 2018 and excluded before then. When necessary, costs were inflated to 2020 US dollars.

    Abbreviations: ASP, average sales price; Big 4 price, price available to 4 largest federal purchasers of pharmaceuticals: US Department of Veterans Affairs, US Department of Defense, US Coast Guard, and US Public Health Service; HR, hazard ratio; FSS, Federal Supply Schedule; ICER, incremental cost-effectiveness ratio; LAR, long-acting release; PSA, probabilistic sensitivity analysis; QALY, quality-adjusted life-year; WTP, willingness to pay.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 616 616 75
PDF Downloads 181 181 37
EPUB Downloads 0 0 0