Cost-effectiveness of ranibizumab in the treatment of visual impairment due to diabetic macular edema

Abstract

Objective Ranibizumab, an anti-vascular endothelial growth factor designed for ocular use, has been deemed cost-effective in multiple indications by several Health Technology Assessment bodies. This study assessed the cost-effectiveness of ranibizumab monotherapy or combination therapy (ranibizumab plus laser photocoagulation) compared with laser monotherapy for the treatment of visual impairment due to diabetic macular edema (DME).

Methods A Markov model was developed in which patients moved between health states defined by best-corrected visual acuity (BCVA) intervals and an absorbing ‘death’ state. The population of interest was patients with DME due to type 1 or type 2 diabetes mellitus. Baseline characteristics were based on those of participants in the RESTORE study. Main outputs were costs (in 2013 CA$) and health outcomes (in quality-adjusted life-years [QALYs]) and the incremental cost-effectiveness ratio (ICER) was calculated. This cost-utility analysis was conducted from healthcare system and societal perspectives in Quebec.

Results From a healthcare system perspective, the ICERs for ranibizumab monotherapy and combination therapy vs laser monotherapy were CA$24 494 and CA$36 414 per QALY gained, respectively. The incremental costs per year without legal blindness for ranibizumab monotherapy and combination therapy vs laser monotherapy were CA$15 822 and CA$20 616, respectively. Based on the generally accepted Canadian ICER threshold of CA$50 000 per QALY gained, ranibizumab monotherapy and combination therapy were found to be cost-effective compared with laser monotherapy. From a societal perspective, ranibizumab monotherapy and combination therapy provided greater benefits at lower costs than laser monotherapy (ranibizumab therapy dominated laser therapy).

Conclusions Ranibizumab monotherapy and combination therapy resulted in increased quality-adjusted survival and time without legal blindness and lower costs from a societal perspective compared with laser monotherapy.

Keywords:

• Ranibizumab
• Diabetic retinopathy
• Cost-effectiveness analysis
• Markov process

Introduction

Diabetic macular edema (DME) is an ophthalmological complication of diabetes mellitus. It develops in a sub-set of patients with diabetic retinopathy, which is a leading cause of visual impairment and blindness worldwide¹. DME is characterized by thickening of retinal tissue as a result of fluid leaking from blood capillaries—when this affects the center of the macula, it can lead to visual impairment and, if left untreated, blindness. Approximately 6.8% of patients with diabetes mellitus have DME¹.

Ranibizumab (Lucentis^®, Novartis Pharma AG, Basel, Switzerland) is an anti-vascular endothelial growth factor designed specifically for ocular use. It is approved for the treatment of visual impairment in patients with DME, wet age-related macular degeneration (wAMD), macular edema following retinal vein occlusion (RVO), and myopic choroidal neovascularization. Ranibizumab has been shown to improve best-corrected visual acuity (BCVA) in patients with visual impairment due to DME in a number of randomized controlled trials (RCTs). The pivotal ‘Ranibizumab Monotherapy or Combined with Laser vs Laser Monotherapy for Diabetic Macular Edema (RESTORE)’ phase 3 study compared the clinical benefit and safety of ranibizumab monotherapy and ranibizumab plus laser photocoagulation with that of laser monotherapy in patients with visual impairment due to DME. It consisted of a 12-month, randomized, double-masked control phase followed by a 24-month, open-label extension^2,3. The RESTORE study found that ranibizumab treatment was associated with rapid improvements in BCVA that were sustained to month 36^2,3. Additional RCTs, including the Diabetic Retinopathy Clinical Research Network (DRCR.net) trial, which compared ranibizumab or triamcinolone combined with laser photocoagulation vs laser photocoagulation alone^4,5, support the results of the RESTORE study^4–10. Ranibizumab has been deemed cost-effective in multiple indications by several Health Technology Assessment (HTA) bodies, including the National Institute for Health and Care Excellence in the UK and the Scottish Medicines Consortium. With the inclusion of wider societal costs and benefits becoming more common in economic analysis, additional analyses of the cost-effectiveness of ranibizumab are needed.

The aim of this study was to evaluate the cost-effectiveness of 36 months of ranibizumab monotherapy or combination therapy (ranibizumab plus laser photocoagulation) compared with that of laser monotherapy for the treatment of visual impairment due to DME. Clinical outcomes were based on the RESTORE trial, in which patients were treated with the ranibizumab posology that is currently approved in Canada.

The study investigated both healthcare system and societal perspectives (including productivity costs) in Quebec. A lifetime time horizon was used based on Canadian Agency for Drugs and Technologies in Health (CADTH) guidelines, which state that a lifetime time horizon should be used as a default, and justification should be provided when a shorter time horizon is used¹¹. Other cost-effectiveness analyses of ranibizumab have used a lifetime time horizon^12,13.

Patients and methods

Model structure

A Markov model was developed in 2013 to predict the long-term costs and health outcomes following 3 years of ranibizumab treatment for visual impairment due to DME. In the Markov structure, patients cycled between eight health states, defined by BCVA intervals ranging from 25 letters or fewer to 86–100 letters (Snellen equivalents 20/10 to 20/20) in the treated eye, and a ninth, absorbing, ‘death’ state (Figure 1A). Patients moved between states at the end of each 3-month cycle based on transition probabilities (Supplementary Table 1)¹³.

Figure 1. (A) Markov model structure showing health state transitions. (B) Modeled visual acuity over time, according to treatment. BCVA, best-corrected visual acuity.

Patients

The population of interest was patients with DME due to type 1 or type 2 diabetes mellitus. Baseline characteristics of patients in each treatment arm were based on those of participants in the RESTORE study (Table 1). The mean (standard deviation [SD]) BCVAs were 64.8 (10.1), 63.4 (10.0), and 62.4 (11.1) letters in the ranibizumab monotherapy, combination therapy, and laser monotherapy groups, respectively. The mean age was 63.3 years, and similar in each group; 40.2% of patients were treated in their better-seeing eye (BSE) and 22% of patients had bilateral disease. BCVA at the start of treatment was assumed based on baseline visual acuity scores in the RESTORE study (Supplementary Table 2).

Table 1. Baseline characteristics of the RESTORE study population.

	Ranibizumab monotherapy	Combination therapy	Laser monotherapy
Sample size, n	115	118	110
Mean age, years (SD)	63 (9.3)	64 (8.2)	64 (8.8)
Proportion of patients with type 1 diabetes mellitus	11.2%	12.7%	10.8%
Proportion of patients with type 2 diabetes mellitus	88.8%	86.4%	87.4%
Time since first diagnosis of diabetes mellitus, years (SD)	15.2 (9.9)	14.6 (9.8)	12.9 (9.0)
Mean BCVA, letter score (SD)	64.8 (10.1)	63.4 (10.0)	62.4 (11.1)

BCVA, best-corrected visual acuity; RESTORE, Ranibizumab Monotherapy or Combined with Laser vs Laser Monotherapy for Diabetic Macular Edema; SD, standard deviation.

Withdrawal rates

A proportion of patients were assumed to drop out in each cycle over years 1–3, with treatment-group-specific withdrawal rates determined by data from the RESTORE study. Withdrawal rates were dependent on treatment group, and not on health state. Patients withdrawing from ranibizumab arms were assumed to continue treatment with laser monotherapy. Owing to low withdrawal rates in RESTORE (11.7–12.7%), after withdrawing, patients were assumed to have the same transition probabilities as those continuing treatment, but to only incur the costs associated with laser monotherapy.

Mortality assumptions

All-cause mortality values were obtained using 5-year incremental data from Statistics Canada¹⁴, which were smoothed to generate annual rates for use in the model. Age-specific data on risk of death due to diabetes mellitus were obtained from the Public Health Agency of Canada and applied to the mortality data from Statistics Canada¹⁵. The relative risk of death due to DME was obtained from a US study of patients with type 2 diabetes mellitus and clinically significant macular edema¹⁶.

Adverse events

Adverse events were rare and similar across treatment arms in the RESTORE study; therefore, they were not included in the model.

Utilities

Utility values specific to DME are not available in the literature. Therefore, the utility values used in the model were based on validated data from studies of other ocular diseases (Supplementary Table 1). In this analysis, utility values associated with each BCVA health state were determined separately for patients treated in their BSE and for those treated in their worse-seeing eye (WSE). Quality of life in people with visual impairment is associated primarily with vision in the BSE, and utility values are generally lower in patients whose affected eye is their BSE¹⁷. For patients with bilateral disease, BSE utilities were used. BSE utility values were obtained from a simulation study of visual impairment associated with wAMD, adjusted for age¹⁸. WSE utility values were obtained from a multiple regression analysis of Canadian patients with RVO, under the assumption that the correlation between utility values and BCVA in the WSE of patients with DME is comparable to that in patients with RVO and AMD¹⁹.

Treatment costs

Treatment costs were calculated from the perspective of the healthcare system in Quebec using unit costs of treatment, administration, and monitoring, multiplied by resource use (Table 2). It was assumed that all patients were only treated for the first 3 years of the model given the lack of accurate evidence on the clinical effects of both ranibizumab and laser beyond that point in time. The frequency of treatment was based on that in the RESTORE study for all treatment arms in year 1². In years 2 and 3, treatment frequencies for ranibizumab monotherapy and combination therapy were based on the RESTORE extension study, and the number of laser photocoagulation sessions was based on DRCR.net trial data. (In the RESTORE study extension phase, all patients received ranibizumab plus laser photocoagulation, whereas the 36-month DRCR.net trial reported data for laser photocoagulation^3,5,20.) In this model, patients receiving ranibizumab were assumed to be monitored monthly, in line with label recommendations. Based on clinical adviser input, for patients receiving laser monotherapy and for those not receiving treatment, it was assumed that individuals with a BCVA of less than 46 letters would have three monitoring visits per year, while those with a BCVA of 46 letters or higher would have five monitoring visits per year.

Table 2. Treatment costs and resource use.

Costs	Cost (2013 CA$)	Rationale/source
Disaggregated costs
A. Ranibizumab injection	1575	Canada list price^53
B. Ranibizumab administration fee	175	Manuel des médecins spécialistes (code 07331), February 2013^54
C. Ofloxacin 0.3% ophthalmic solution (generic)	3.54	Dose based on clinical adviser input (1 drop/day for 4 days). Cost obtained from  the Liste de médicaments (cost is for a 5 mL bottle)^53
D. Laser photocoagulation	211.00	Manuel des médecins spécialistes (code 07311), February 2013^54
Aggregated costs
Ranibizumab monotherapy	1753.54	A + B + C
Combination therapy	1964.54	A + B + C + D
Laser monotherapy	211.00	D
Resource use	Number of physician visits
	Year 1	Year 2	Year 3	Year 4+
Ranibizumab monotherapy
Treatment visits	7^a	3.9^a	2.9^a	0
Monitoring visits	5	8.1	9.1	3–5
Combination therapy
Treatment visits	7^a	3.5^a	2.5^a	0
Monitoring visits	5	8.5	9.5	3–5^b
Laser monotherapy
Treatment visits	2^a	0.55^c	0.55^c	0
Monitoring visits	3–5^b	3–5^b	3–5^b	3–5^b

a Assumption based on RESTORE trial data.

b Patients were assumed to have five monitoring visits if their BCVA was at least 45 letters, and three monitoring visits if their BCVA was less than 45 letters, based on clinical adviser input.

c Assumption based on DRCR.net trial data.

BCVA, best-corrected visual acuity; DRCR, Diabetic Retinopathy Clinical Research Network; RESTORE, Ranibizumab Monotherapy or Combined with Laser vs Laser Monotherapy for Diabetic Macular Edema.

Based on the ranibizumab indication (for patients with visual impairment due to DME) and label recommendations, it was assumed that patients with no visual impairment (those transitioning into health states with a BCVA of over 75 letters) would cease to receive treatment, and that treatment would only resume if patients transitioned into health states with a BCVA of 75 letters or under.

Costs of visual impairment

The healthcare utilization costs of visual impairment used in the model were derived from an observational study of the burden of illness of wAMD performed at five Canadian retinal clinics, and were broken down according to BCVA (Table 3)²¹. Productivity losses associated with each health state, largely comprising caregiver time to help with daily activities, were calculated using data from the literature^21–45. These costs of lost productivity were added to the direct costs for each health state when a societal perspective was taken. In the absence of data found on the costs associated with the WSE, it was conservatively assumed that patients treated only in their WSE did not incur costs associated with visual impairment or productivity losses.

Table 3. Mean annual healthcare utilization costs associated with visual impairment, according to BCVA.

Healthcare utilization cost	BCVA ≥56 letters		BCVA 36–55 letters		BCVA ≤35 letters
	Cost (2005 CA$)	Proportion of total^21	Cost (2005 CA$)	Proportion of total^21	Cost (2005 CA$)	Proportion of total^21
Direct vision-related medical costs	9205	86.4%	6903	64.7%	4240	35.6%
Direct non-vision-related costs	1450	13.6%	2302	21.6%	2843	23.9%
Direct non-medical-related costs	0	0.0%	1465	13.7%	4824	40.5%
Total cost per patient (values corrected to 2013 CA$)	1665 (1889)	100%	3928 (4454)	100%	7766 (8806)	100.0%

Costs were inflated from 2005 to 2013 prices.

BCVA, best-corrected visual acuity.

Economic analysis

The main outcomes from the model were lifetime costs (in 2013 CA$) and total health outcomes (in quality-adjusted life-years [QALYs]). Costs and health outcomes were discounted at 5% per year in accordance with the recommendation from the CADTH guidelines, and a lifetime time horizon was assumed. The incremental cost-effectiveness ratio (ICER) was calculated as the ratio of the mean incremental cost to the mean number of QALYs gained for ranibizumab monotherapy or combination therapy compared with laser monotherapy. Costs per year without legal blindness were also calculated.

Sensitivity analyses

Deterministic sensitivity analyses were used to test the impact of parameter uncertainty on the results, including the proportion of patients with bilateral DME, the number of injections received, the treatment stopping rule (see Methods section) and the time horizon (Supplementary Table 3). Probabilistic sensitivity analyses using a Monte Carlo simulation were undertaken to assess the joint parameter uncertainty in the model (Supplementary Table 4).

Results

Health outcomes

The model reproduced the sustained improvement in BCVA seen in the RESTORE study with ranibizumab monotherapy and combination therapy compared with laser monotherapy. After treatment, BCVA progressed according to natural decline as described by the Wisconsin Epidemiological Study of Diabetic Retinopathy (WESDR)^46,47. The difference in BCVA observed between patients treated with ranibizumab and those receiving laser monotherapy was maintained over time (Figure 1B). This sustained improvement in BCVA was accompanied by a progressive reduction in the need for re-treatment from year 1 to year 2 to year 3. Compared with laser monotherapy, ranibizumab monotherapy was associated with gains of 0.62 years without legal blindness and 0.40 QALYs. The corresponding gains for combination therapy were 0.56 years without legal blindness and 0.32 QALYs.

Costs

From a healthcare system perspective, the lifetime costs of ranibizumab monotherapy and combination therapy were CA$25 233 and CA$26 854, respectively, and the cost of laser monotherapy was CA$15 383. From a societal perspective, lifetime costs associated with ranibizumab monotherapy and combination therapy were CA$60 200 and CA$68 078, respectively; laser monotherapy was associated with a cost of CA$79 193. The higher cost associated with laser monotherapy is a result of poorer BCVA in this group, which is associated with an increased cost of productivity losses.

Cost-effectiveness

From a healthcare system perspective, the ICERs for ranibizumab monotherapy and combination therapy vs laser monotherapy were CA$24 494 and CA$36 414 per QALY gained, respectively (Table 4). The incremental costs per year gained without legal blindness for ranibizumab monotherapy and combination therapy vs laser monotherapy were CA$15 822 and CA$20 616, respectively. Based on the generally accepted Canadian ICER threshold of CA$50 000 per QALY gained, ranibizumab monotherapy and combination therapy were cost-effective compared with laser monotherapy. From a societal perspective, ranibizumab monotherapy and combination therapy dominated laser monotherapy, providing greater benefits at a lower cost.

Table 4. Cost-effectiveness results for reference case.

	Laser monotherapy	Ranibizumab monotherapy	Combination therapy	Difference
Ranibizumab monotherapy vs laser monotherapy (healthcare system perspective)
Total cost (2013 CA$)	15 383	25 233		9849
Number of years without legal blindness (BCVA >35 letters)	3.44	4.06		0.62
QALYs	7.77	8.17		0.40
Incremental cost per year without legal blindness (2013 CA$)				15 822
ICER per QALY gained (2013 CA$)				24 494
Combination therapy vs laser monotherapy (healthcare system perspective)
Total cost (2013 CA$)	15 383		26 854	11 471
Number of years without legal blindness (BCVA >35 letters)	3.44		4.00	0.56
QALYs	7.77		8.09	0.32
Incremental cost per year without legal blindness (2013 CA$)				20 616
ICER per QALY gained (2013 CA$)				36 414
Ranibizumab monotherapy versus laser monotherapy (societal perspective)
Total cost (2013 CA$)	79 193	60 200		–18 993
Number of years without legal blindness (BCVA >35 letters)	3.44	4.06		0.62
QALYs	7.77	8.17		0.40
Incremental cost per year without legal blindness				Ranibizumab dominates
ICER per QALY gained				Ranibizumab dominates
Combination therapy vs laser monotherapy (societal perspective)
Total cost (2013 CA$)	79 193		68 078	–11 114
Number of years without legal blindness (BCVA >35 letters)	3.44		4.00	0.56
QALYs	7.77		8.09	0.32
Incremental cost per year without legal blindness				Ranibizumab dominates
ICER per QALY gained				Ranibizumab dominates

BCVA, best-corrected visual acuity; ICER, incremental cost-effectiveness ratio; QALY, quality-adjusted life-year.

Sensitivity analyses

Deterministic sensitivity analyses demonstrated that the model is robust. Ranibizumab monotherapy and combination therapy remained cost-effective compared with laser monotherapy when key model assumptions were adjusted. From a healthcare system perspective, the only change to the model parameters that resulted in an ICER of more than CA$50 000 per QALY gained with ranibizumab monotherapy was the removal of the assumption that patients stop treatment if BCVA is higher than 75 letters (the treatment stopping rule). This scenario led to an ICER of CA$72 989 per QALY. The only changes that led to an ICER of more than $50 000 per QALY gained with combination therapy were the removal of the treatment stopping rule (CA$115 357 per QALY) and reduction of the time horizon to 10 years (CA$54 252 per QALY) (Figure 2). From a societal perspective, ranibizumab monotherapy dominated laser monotherapy in all investigated scenarios, and combination therapy dominated laser monotherapy in all scenarios except for the one in which the treatment stopping rule was removed; this scenario led to an ICER of CA$40 873 per QALY gained (data not shown).

Figure 2. Tornado plots for one-way sensitivity analyses of (A) ranibizumab monotherapy vs laser monotherapy, and (B) combination therapy (ranibizumab and laser photocoagulation) vs laser monotherapy (healthcare system perspective). BSE, better seeing eye; CDN BOI, Canadian DME Burden of Illness Study; CNIB, Canadian National Institute for the Blind; DME, diabetic macular edema; RR, relative risk; WSE, worse-seeing eye.

Probabilistic sensitivity analysis demonstrated that ranibizumab monotherapy and combination therapy are likely to be cost-effective compared with laser monotherapy from both healthcare system and societal perspectives. From a healthcare system perspective, the probabilities of ranibizumab monotherapy and combination therapy being cost-effective, given an ICER acceptability threshold of CA$50 000 per QALY, were 74% and 60%, respectively (Figure 3). From a societal perspective, the equivalent probabilities of ranibizumab monotherapy and combination therapy being cost-effective were 87% and 79%, respectively (Figure 4).

Figure 3. Cost-effectiveness acceptability curves for (A) ranibizumab monotherapy vs laser monotherapy, and (B) combination therapy (ranibizumab and laser) vs laser monotherapy (healthcare system perspective). ICER, incremental cost-effectiveness ratio.

Figure 4. Cost-effectiveness acceptability curves for (A) ranibizumab monotherapy vs laser monotherapy, and (B) combination therapy (ranibizumab and laser photocoagulation) vs laser monotherapy (societal perspective). ICER, incremental cost-effectiveness ratio.

Discussion

Ranibizumab monotherapy and combination therapy resulted in increased time without legal blindness and increased quality-adjusted survival compared with laser monotherapy, and in lower costs from a societal perspective. Deterministic sensitivity analyses demonstrated that ranibizumab monotherapy and combination therapy remained cost-effective compared with laser monotherapy when key model assumptions were adjusted.

Ranibizumab monotherapy appears to be more effective and less expensive than combination therapy. The improvement in QALYs gained and years without blindness in the ranibizumab monotherapy group can be explained by the RESTORE trial results, which report that patients receiving ranibizumab monotherapy achieved slightly greater letter gains than those receiving combination therapy^2,3.

Three previous studies have investigated the cost-effectiveness of ranibizumab compared with laser monotherapy in the treatment of DME. Dewan et al.⁴⁸ estimated an incremental cost of $5943 (2010 USD) per letter gained in visual acuity for ranibizumab combination therapy vs laser monotherapy over a time horizon of 2 years. As discussed, Mitchell et al.¹³ presented an earlier version of the model reported in this study, in which outcomes were simulated from a UK healthcare system perspective using a 15-year time horizon based on available data. The ICERs for ranibizumab monotherapy and combination therapy vs laser monotherapy were £24 028 and £36 106 per QALY gained, respectively (2010 GBP). The final study by Stein et al.⁴⁹ found that the ICER for ranibizumab combination therapy vs laser monotherapy was $89 903 per QALY gained (2013 USD) over a 25-year time horizon, which suggests that it would be cost-effective at a willingness-to-pay threshold of USD $100 000, if patients receive fewer than 0.45 injections annually after year 2. The higher cost of ranibizumab combination therapy reported by Stein et al.⁴⁹ can be explained by the fact that their model adopted a US perspective; therefore, the costs are likely to be substantially different.

The present analysis builds on previous models that included clinical trial data of only 12 months¹³ by including clinical trial data up to 36 months, taking into account patients treated in their WSE, their BSE or both eyes, and using a lifetime time horizon. The inclusion of wider societal costs and benefits is becoming more common in economic analysis, and the present study is the first analysis of ranibizumab for the treatment of visual impairment due to DME to take into account productivity losses. Patient baseline data and input variables related to treatment frequency and efficacy were based on individual patient records taken from large clinical trials, and are likely to be broadly applicable to clinical practice. One limitation of this and previous studies is the lack of utility value estimates specific to DME. However, our base-case analysis used utility values derived from studies of other conditions associated with a loss of vision, which are likely to be relevant to the DME population and, in addition, the DME utility values are consistent across the treatment arms. This was more appropriate than using EQ-5D utility scores, which have been shown to be extremely insensitive to changes in visual acuity^50,51.

A potential limitation of this study is the choice of comparator group. Laser photocoagulation was selected as the comparator because, at the time of the study, it was the only licensed treatment for DME in Canada other than ranibizumab. Aflibercept and dexamethasone intravitreal implants were not reimbursed in Canada at the time this analysis was conducted and, therefore, were excluded from the analysis. The ranibizumab dose used in the current analysis was 0.5 mg pro re nata based on the RESTORE study². A more recent study, the DRCR.net trial, has compared the efficacy of ranibizumab, aflibercept, and bevacizumab⁵². Results show that, in patients with initial visual acuity 20/40 or better, there were no differences in efficacy among the three treatments, whereas in those with visual acuity 20/50 or worse, mean improvement was greatest with aflibercept and not significantly different between ranibizumab and bevacizumab. However, the DRCR.net trial⁵² used a lower dose of ranibizumab (0.3 mg) using the pro re nata posology than either the RESTORE study² or a previous DRCR.net trial⁴, both of which showed greater efficacy of ranibizumab 0.5 mg (either as monotherapy or combined with laser therapy) than with laser therapy alone. Comparison with these other therapeutic options would be a useful direction for future research.

Conclusions

The analysis confirmed that ranibizumab as monotherapy or in combination with laser photocoagulation for the treatment of DME with visual impairment is associated with increased time without legal blindness and increased quality-adjusted survival compared with laser monotherapy. The results demonstrate the continued cost-effectiveness of ranibizumab compared with laser monotherapy over 36 months. When a healthcare system perspective was taken, ranibizumab monotherapy and combination therapy were cost-effective compared with laser monotherapy and, when productivity losses were taken into account, they both dominated laser monotherapy. These results are robust and the ICERs are below commonly accepted cost-effectiveness thresholds in Canada.

Transparency

Declaration of funding

This study was funded by Novartis Pharma AG.

Declaration of financial and other relationships

JH is an employee of Optum, Burlington, ON, Canada. MB is an employee of Novartis Pharmaceuticals Canada Inc., Dorval, QC, Canada. AF is an employee of Novartis Pharma AG, Basel, Switzerland. JME peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgments

The authors would like to thank Megan Pickering, Novartis Pharmaceuticals Canada Inc., for her contribution to the study. Editorial support for the preparation and revision of drafts under the direction of the authors was provided by Sophie Shina, Oxford PharmaGenesis, and funded by Novartis Pharma AG.