Losartan: a pharmacoeconomic review

Summary

The angiotensin II antagonist losartan has been clinically studied in several patient populations including type 2 diabetes mellitus (T2DM) with nephropathy, hypertension with left ventricular hypertrophy, and the elderly with heart failure. The aim of this paper is to provide a review of the health economic evaluations based on the clinical trial results of losartan.

In patients with T2DM and nephropathy, losartan was shown to be cost saving in 14 countries and the EU. Net cost savings per patient, after factoring in the drug cost of losartan, ranged from $56 and $5,149 over a 3.5 year time horizon. For the two countries with published lifetime projections, the US and Mexico, net cost savings per patient were $24,632 and $2,223, respectively.

In patients with hypertension and left ventricular hypertrophy, losartan as compared with atenolol was found to be cost effective in five countries and cost saving in one country. The incremental cost-effectiveness ratios ranged from $1,274 to $5,764 per quality-adjusted life year gained for four of the countries and was $1,083 per life-year gained in one country. The other country evaluation reported a $20 net cost saving per patient.

Pharmacoeconomic evaluations in other patient populations comparing losartan with the angiotensin-converting enzyme inhibitor captopril generally did not demonstrate differences in health economic outcomes.

Keywords: :

• cost
• losartan
• diabetic nephropathy
• end-stage renal disease
• left ventricular hypertrophy

Introduction

Cardiovascular disease imposes a major economic burden on society worldwide. In the US, stroke, hypertension, heart failure and other cardiovascular diseases cost $149 billion annually, comprising approximately 17% of all medical expenditures and nearly 30% of Medicare costs1. In the EU, cardiovascular disease is the main source of morbidity and mortality, accounting for $234 billion annually when productivity and informal care costs are included2.

Patients with type 2 diabetes mellitus (T2DM) and hypertension have an increased risk of cardiovascular morbidity and mortality as compared with those with hypertension alone. These patients are also at increased risk of nephropathy and consequently end-stage renal disease (ESRD). There has been a marked rise in the prevalence of T2DM and resultant nephropathy, particularly in Europe, the US and Japan. A study using data from the European Renal Association-European Dialysis and Transplant Association (ERA-EDTA) Registry found that the number of persons with T2DM and nephropathy entering ESRD has been increasing 11.9% annually3. This increased incidence and prevalence as well as the high cost of renal replacement therapy has resulted in a major burden on healthcare budgets. In 2000, 379,000 persons in the US suffered from ESRD and the Medicare ESRD programme incurred $12.4 billion in costs. This cost is expected to increase to $28 billion a year by 20103. In the UK, the total annual cost of managing type 2 diabetic nephropathy to the National Health Service was $993 million3.

Hypertension alone and in combination with left ventricular hypertrophy (LVH) also results in a significant cardiovascular burden, particularly with respect to increased risk of stroke. Stroke occurs in 30.9 million individuals worldwide and is responsible for >$49 billion in costs in the US (2002)4. The average lifetime cost of stroke in Norway exceeds $100,0005. Much of this cost is due to severe long-term physical and mental disability manifested in physical handicap, cognitive dysfunction and depression4. The burden is expected to increase over time as the population ages, even if the incidence of stroke remains the same. In Germany, the number of newly diagnosed first strokes over a 5-year period is expected to rise from 756,000 (2006–2010) to 955,000 (2021–2025)6. During this 20-year period (2006–2025), the net present value of strokes in Germany is estimates to be €108.6 billion.

In patients with hypertension, antihypertensive therapy has been shown to be effective in reducing the incidence of stroke and, in patients with T2DM, in slowing the progression of nephropathy5. More recently, clinical trials in these patient groups have demonstrated benefits beyond those attributable to blood pressure control alone.

The Reduction of Endpoints in Non-insulin Dependent Diabetes with the Angiotensin II Antagonist Losartan (RENAAL) study demonstrated that, in hypertensive patients with T2DM and nephropathy, treatment with losartan plus conventional antihypertensive therapy (CT) reduced the relative risk of ESRD by 29% vs. placebo plus CT, for a similar reduction in blood pressure7. The Losartan Intervention for Endpoint Reduction in Hypertension (LIFE) study demonstrated a 25% reduction in the relative risk of stroke compared with an atenolol-based regimen in hypertensive patients with LVH, also for a similar reduction in blood pressure8. Studies in other patient populations generally did not show an efficacy benefit of losartan when compared with the angiotensin-converting enzyme (ACE)-inhibitor captopril.

In this review, the methods and results of the published pharmacoeconomic studies of losartan are summarised and discussed.

Methods

Pharmacoeconomic evaluations were identified by searching PubMed (indexed for MEDLINE) using the keywords losartan, cost, economics and outcomes. Additional published and in-press studies were identified by the authors of the manuscripts. For patients with T2DM and nephropathy, 14 within-trial country-specific evaluations were identified based on six publications and two within-trial region-specific evaluations. Two lifetime evaluations were identified and are included in this review. Another country-specific evaluation, which combined aspects of a within-trial and beyond-trial analysis, is also included. All evaluations were based on the RENAAL clinical trial and several evaluations represent recalculations of the same model with country-specific cost estimates.

For patients with hypertension and LVH, there were six country-specific pharmacoeconomic evaluations based on the LIFE clinical trial. Four of these analyses were based on a method that combines aspects of both a within-trial and beyond-trial evaluation. Another study was a within-trial evaluation and one study was a lifetime evaluation from a societal perspective.

Two economic/outcomes studies were identified comparing losartan to the ACE-inhibitor captopril, both in elderly patients with heart failure.

All costs were converted to US$ using July 2007 exchange rates.

Results

Losartan in patients with type 2 diabetes mellitus and nephropathy

Table 1 displays the essential features and results of the within-trial pharmacoeconomic evaluations of losartan in patients with T2DM and nephropathy. All of these studies were based on results from the RENAAL clinical trial and took the perspective of a healthcare system responsible for all direct medical costs. Of these studies, 13 (81%) reported cost savings per patient over 3.5 years, the pre-specified time frame for the economic evaluation for the US7. The costs in all but one of these evaluations were based on the number of days the average patient had ESRD in the RENAAL trial, the daily cost of ESRD in the respective country or region (the EU analysis used ESRD costs from Germany), the average usage by dose of losartan, and the local cost of losartan drug therapy. The number of days the average patient had ESRD by treatment group was computed by taking the difference in the area under the Kaplan-Meier curve for all-cause mortality (per patient lifetime in days) and the area under the Kaplan-Meier curve for time to ESRD or all-cause mortality (per patient ESRD-free days) as shown in Figure 1.

Figure 1. Days with ESRD and ESRD-free days per patient in the RENAAL study.

Table 1. Within-trial pharmacoeconomic evaluations of losartan in patients with type 2 diabetes mellitus and nephropathy based on RENAAL.

Country/region	Methods/metrics	Cost savings per patient over 3.5 years	Other results	References
US	Days with ESRD; US costs of ESRD; losartan drug use/cost	$3,522*	Cost savings increased to $5,298 over 4 years*	Herman WH et al ^7
Canada	Days with ESRD; Canada costs of ESRD; losartan drug use/cost	$3,480*	Cost savings increased to $5,156 over 4 years*	Burgess ED et al^8
Latin America	Competing risks; estimates of number of persons with T2DM and nephropathy	Not reported	Reduction of 36,238 cases of ESRD over 3.5 years	Gerth WC et al^13
France	Days with ESRD; France costs of ESRD; losartan drug use/cost	Not reported	Cost savings were $7,948 over 4 years*	Souchet T et al^12
Switzerland	Decision analytic model using published sources	$3,373*	–	Szucs TD et al^9
Denmark	Days with ESRD; Denmark costs of ESRD; losartan drug use/cost	$3,552*	Cost savings increased to $5,611 over 4 years*	Jönsson L et al^10
Finland	Days with ESRD; Finland costs of ESRD; losartan drug use/cost	$4,795*	Cost savings increased to $7,353 over 4 years*	Jönsson L et al^10
Norway	Days with ESRD; Norway costs of ESRD; losartan drug use/cost	$5,149*	Cost savings increased to $7,796 over 4 years*	Jönsson L et al^10
Sweden	Days with ESRD; Sweden costs of ESRD; losartan drug use/cost	$5,126*	Cost savings increased to $7,696 over 4 years*	Jönsson L et al^10
EU	Days with ESRD; estimated at risk population in the EU; Germany ESRD costs	Not reported	Reduction of 44,100 cases of ESRD over 3.5 years	Gerth WC et al^14
Hong Kong	Days with ESRD for Asian subgroup; Hong Kong costs of dialysis and losartan	$515	–	Seng WK et al^11
Japan	Days with ESRD for Asian subgroup; Japan costs of dialysis and losartan	$505	–	Seng WK et al^11
Korea	Days with ESRD for Asian subgroup; Korea costs of dialysis and losartan	$55	–	Seng WK et al^11
Malaysia	Days with ESRD for Asian subgroup; Malaysia costs of dialysis and losartan	$255	–	Seng WK et al ^11
Singapore	Days with ESRD for Singapore subgroup; Hong Kong costs of dialysis and losartan	$202	–	Seng WK et al^11
Taiwan	Days with ESRD for Asian subgroup; Taiwan costs of dialysis and losartan	$56	–	Seng WK et al^11

All costs converted to US$. RENAAL, Reduction of Endpoints in Non-insulin Dependent Diabetes with the Angiotensin II Antagonist Losartan; ESRD, end-stage renal disease; T2DM, type 2 diabetes mellitus. * p<0.05.

Over 3.5 years, the average losartan patient experienced 33.6 fewer days with ESRD than the average placebo patient (109.7 - 76.1 = 33.6)7. For the Asian countries, days with ESRD were calculated based on the Asian subgroup of patients in RENAAL. The reduction in days with ESRD resulted in ESRD-related cost savings for losartan. The cost savings per patient shown in Table 1 include the additional cost of losartan drug therapy for the losartan group.

The cost savings per patient over 3.5 years for the losartan group ranged between $3,373 and $5,149 for the US7, Canada8, Switzerland9 and the Nordic countries (Denmark, Finland, Norway and Sweden)10. Cost savings ranged between $55 and $515 for the Asian countries (Hong Kong, Japan, Korea, Malaysia, Singapore and Taiwan)11. The lower cost savings for the Asian countries as compared with the other countries is primarily due to a lower cost of renal replacement therapy.

The country evaluations other than those for the Asian countries also reported increased cost savings for losartan over 4 years, ranging from $5,156 to $7,948^7,8,10,12 per patient, as compared with the 3.5-year time point. These increased cost savings stem from the greater number of ESRD days saved for losartan; 46.9 days over 4 years vs. 33.6 days over 3.5 years5. The regional evaluations for Latin America and the EU focused on public health implications and did not report cost savings per patient. However, these evaluations did estimate reductions of 36,238 cases of ESRD and 44,100 cases of ESRD over 3.5 years for Latin America¹³ and the EU14, respectively.

Table 2 shows the results of the beyond-trial pharmacoeconomic evaluations of losartan in T2DM and nephropathy. The evaluations for the US and Mexico were based on a competing risks method to estimate the lifetime cumulative incidence of ESRD3. This method accounts for the possibility that a patient may die prior to experiencing ESRD. The cumulative incidence of ESRD was combined with an estimate of the lifetime discounted cost of ESRD in each country to project lifetime ESRD-related costs. Lifetime discounted drug costs of losartan and other non-ESRD-related costs were also included to deliver projected net costs. The lower lifetime cost savings reported for Mexico, $2,223 per patient15, as compared with those for the US, $24,632 per patient^3,15, are primarily explained by the much lower cost of renal replacement therapy in Mexico.

Table 2. Beyond-trial* pharmacoeconomic evaluations of losartan in patients with type 2 diabetes mellitus and nephropathy.

Country/region	Methods/metrics	Cost savings per patient	Other results	References
US	Competing risks; lifetime costs of ESRD, losartan, and others; life expectancies with and without ESRD; 3% discounting of costs and outcomes	$24,632 lifetime cost savings per patient	Lifetime NNT to prevent one case of ESRD was 6; 0.99 life-years gained per patient (0.70 discounted)	Carides GW et al ^3
Mexico	Competing risks; lifetime costs of ESRD, losartan, and others; life expectancies with and without ESRD; 3% discounting of costs and outcomes	$2,223 lifetime cost savings per patient	Lifetime NNT to prevent one case of ESRD was 6; 0.70 discounted life years gained per patient	Arredondo A et al ^15
UK	Within-trial incidence of ESRD; lifetime costs of ESRD; within-trial costs of losartan; 3.5% discounting of costs and effects	$13,428	0.44 discounted life years gained per patient	Vora J et al ^16

All costs converted to US$. ESRD, end-stage renal disease; NNT, number needed to treat.

* The economic evaluation for the UK combined elements of a within-trial and beyond-trial analysis.

The economic evaluation for the UK employed a method that combines aspects of both a within-trial and beyond-trial evaluation. The rationale behind this approach is to account for all the beyond-trial downstream costs of ESRD for patients experiencing ESRD within the trial, while not making any projections regarding the incidence of ESRD beyond the trial. The within-trial cumulative incidence of ESRD was estimated using the competing risks approach as above but without the lifetime projection. This ESRD incidence was combined with an estimate of lifetime ESRD costs in the UK. Within-trial losartan costs were included to arrive at an estimate of net costs16.

Losartan in patients with hypertension and left ventricular hypertrophy

Table 3 displays the essential features and results of the pharmacoeconomic evaluations of losartan in patients with hypertension and LVH. Six published and in-press economic evaluations were identified comparing losartan with the beta-blocker atenolol. Unless otherwise noted, all of these studies were based on results from the LIFE clinical trial and took the perspective of a healthcare system responsible for all direct medical costs. Four of these evaluations (Sweden, Denmark, Netherlands and the UK) employed a method similar to that used for the RENAAL trial UK evaluation discussed in the previous section and were based on patient-level LIFE data. The cumulative incidence of stroke during the LIFE trial for each treatment group was estimated by the competing risks method to account for the possibility of death prior to stroke17. The cumulative incidence of stroke was combined with the respective country's estimate of lifetime discounted stroke-related costs. Within-trial losartan and atenolol drug costs were included in the estimate of net costs for each treatment group.

Table 3. Pharmacoeconomic evaluations of losartan vs. atenolol in patients with hypertension and left ventricular hypertrophy based on LIFE*.

Country	Methods/metrics	Incremental cost-effectiveness ratio	Other results	References
Sweden	Within-trial incidence of stroke; lifetime costs of stroke; within-trial costs of losartan and atenolol; 3% discounting of costs and effects	$5,764 per QALY gained for losartan vs. atenolol	Losartan reduced cumulative incidence of stroke from 6.5 to 4.9% after 5.5 years; losartan resulted in 0.092 LYG and 0.069 QALYs gained after 5.5 years	Jönsson B et al 17
Denmark	Within-trial incidence of stroke; lifetime costs of stroke; within-trial costs of losartan and atenolol; 3% discounting of costs and effects	$3,634 per QALY gained for losartan vs. atenolol when only the cost of the healthcare sector is included	$13,403 per QALY gained for losartan vs. atenolol if all costs to society are included	Keiding H et al^18
The Netherlands	Within-trial incidence of stroke; lifetime costs of stroke; within-trial costs of losartan and atenolol; 4% discounting of costs and effects	$1,083 per LYG for losartan vs. atenolol	Losartan resulted in 0.059 LYG	Boersma C et al^20
UK	Within-trial incidence of stroke; lifetime costs of stroke; within-trial costs of losartan and atenolol; 3.5% discounting of costs and effects	$4,327 per QALY gained for losartan vs. atenolol	$23,063 per QALY gained when stroke costs beyond the first 5 years were excluded	McInnes G et al^19
Switzerland	Decision analysis framework using published data	–	$20 net cost saving per patient for losartan vs. atenolol over the mean trial follow-up period of 4.8 years	Szucs TD et al^22
Canada	Markov state transition model based on published data; 3% discounting of costs and effects	$1,274 per QALY gained for losartan over the lifetime from a societal perspective	–	Anis AH et al^21

All costs converted to US$.

LIFE, Losartan Intervention for Endpoint Reduction in Hypertension; QALY, quality-adjusted life year; LYG, life years gained.

* The time horizon for the within-trial evaluations was 5.5 years.

Three of these four economic evaluations reported incremental cost-effectiveness ratios (ICERs) ranging between $3,634 and $5,764 per QALY gained for losartan as compared with atenolol171819 while the evaluation for The Netherlands reported an ICER of $1,083 per life year gained (LYG)20. For Denmark, the incremental cost per QALY gained was $13,403 when all costs to society were included. The evaluation for Canada was based on a Markov state transition model, which was based on published data, and reported an ICER of $1,274 per QALY gained over a lifetime horizon from a societal perspective21. The Switzerland analysis, which was based on a decision analysis framework using published data, reported net cost savings of $20 per patient for losartan over the within-trial period of LIFE22.

Losartan in elderly patients with heart failure

The Evaluation of Losartan in the Elderly (ELITE) study was a multinational, double-blind, randomised 48-week study comparing the safety and efficacy of losartan to captopril in ACE inhibitor-naive patients ≥65 years of age with symptomatic heart failure23. This study was followed up with the larger ELITE II trial, which enrolled a total of 3,152 patients, aged 60 years or older, with New York Heart Association (NYHA) classes II to IV heart failure and ejection fraction ≤40%24. A cost-effectiveness analysis based on the results of ELITE reported a lifetime ICER of $4,047 per LYG for losartan as compared with captopril. However, an analysis of heart failure-related outcomes based on the ELITE II study did not show any differences between treatment groups24.

Discussion

ESRD imposes a major burden on society. The numerous pharmacoeconomic evaluations of losartan in patients with T2DM and nephropathy published from 2002 to 2006 have consistently demonstrated cost savings over the within-trial period of RENAAL. Most of these analyses were rooted in the reduction in days with ESRD for the losartan (plus CT) vs. the placebo (plus CT) group over the US evaluation pre-specified time horizon of 3.5 years7. In every country evaluation, the reduction in ESRD-related costs offsets the costs of study medication by a multiple. For example, in the US, $3.17 in ESRD-related cost savings would be realised for every dollar spent treating patients with T2DM and nephropathy7.

These results involve major public health implications. In particular, in the EU, there are an estimated 700,000 T2DM patients with proteinuria (urine albumin/creatinine ≥300 mg/g). Adding losartan to the treatment regimen of these patients would lead to an estimated reduction of 44,100 cases of ESRD, 64,400 fewer person years with ESRD, and reduce ESRD-related costs by $3.6 billion over 3.5 years14.

From a societal perspective, the savings underestimate the full costs by excluding patient out-of-pocket costs and productivity losses.

The issue as to whether these cost savings would be sustained over the long term was raised in several of these articles. It was noted, for example, that losartan treatment could bring about a delay in the onset of renal failure rather than prevention for some patients17. The beyond-trial economic evaluations have helped to address this question by projecting the cumulative incidence of ESRD while accounting for the competing risk of non-ESRD death. Given the high-risk nature of this population, the delay in the onset of ESRD makes it more likely that patients will die of other causes, such as cardiovascular diseases, before ever requiring renal replacement therapy3.

These beyond-trial economic evaluations for patients with T2DM also consistently showed cost savings for the losartan treatment group, again with some differences, primarily due to differences in ESRD costs. Disparities across countries are also the result of differences in practice patterns (e.g. the proportionate breakdown between haemodialysis and peritoneal dialysis). It is noteworthy that, in addition to the cost savings, the beyond-trial evaluations reported gains in life years for losartan based on the reduction in the incidence of ESRD3. The analyses for the US and Mexico also reported a lifetime number needed to treat to prevent one case of ESRD of 6.

A limitation of the lifetime evaluations for the US and Mexico is the use of the 4 years of within-trial data to project outcomes and costs beyond the trial3. This concern was addressed by sensitivity analyses, which consistently showed persistent, although in some cases reduced, cost savings. Another possible criticism is the exclusion of the benefits of losartan in reducing the incidence of first hospitalisation for heart failure (89 patients in the losartan group [11.9%] vs. 127 in the placebo group [16.7%]; relative risk reduction 32%; p=0.005). Inclusion of these benefits would have resulted in increased cost savings for losartan, but it was decided to exclude them given that this was a secondary end point.

Stroke also imposes a great onus on society. All the published pharmacoeconomic evaluations of losartan from 2004 to 2006, based on the LIFE study and primarily on the stroke reduction for losartan vs. atenolol, have established the clinical and economic value of losartan in patients with hypertension and LVH. The estimated incremental costs per QALY or LYG are all within commonly accepted benchmark values for cost effectiveness. The reduction in stroke-related costs substantially offsets the higher losartan (vs. atenolol) study medication cost. For example, in Sweden, 80% of the incremental study cost is offset by the reduced stroke-related cost. The LIFE study also showed a reduction in the onset of new diabetes for the losartan treatment group17. Inclusion of the increased costs associated with diabetes in the economic evaluations would have resulted in additional cost reductions for the losartan group as well as possible gains in life years and QALYs.

A potential criticism of the Sweden, Denmark, Netherlands and the UK evaluations is that the stroke incidence on which the estimated reduction in stroke-related costs and gains in life expectancy were based is limited to the 5.5-year within-trial period17. Lifetime projections would provide additional understanding regarding the long-term implications of losartan vs. atenolol treatment in patients with hypertension and LVH. However, these projections would also involve additional assumptions concerning the cumulative incidence of stroke and use of study medication beyond the LIFE study.

The pharmacoeconomic evaluations of losartan in elderly patients with heart failure have not consistently demonstrated a benefit of losartan over the ACE-inhibitor captopril.

In general, one should be careful when making international comparisons from economic evaluations. The transferability, or applicability, of results from one country to another has been widely debated25. Differences can arise from a variety of sources, including those that are country specific (such as differences in practice patterns and the cost of care) as well as those that are country independent (such as differences in the modelling approach). Nevertheless, there is often a need to at least interpret economic evaluations from an international level. In this review, costs were converted to US$ using July 2007 exchange rates to facilitate this interpretation. Although other methods (such as purchasing price parity) can be used to convert national currencies to US$, the general observations noted in this review should hold.

In conclusion, losartan reduces ESRD and consistently results in cost savings in patients with T2DM and nephropathy, for similar reductions in blood pressure. In patients with hypertension and LVH, losartan as compared with atenolol reduces the cumulative incidence of stroke and has been shown to be cost effective across all the published pharmacoeconomic evaluations based on LIFE.

Acknowledgement

Declaration of interest: Funding for this study was provided by Merck & Co.

Notes