The Cost- Effectiveness of Enzyme Replacement Therapy (ERT) for Infantile Pompe S Disease

The cost- effectiveness of enzyme replacement therapy (ERT) for Infantile Pompe´s disease: Comparative perspectivefrom a High Income Country (England) and a Middle Income one (Colombia).

Author names and affiliations: Hector E Castro J MD, Msc. Doctor in Medicine and Surgery- Universidad del Rosario (Bogotá Colombia); Msc Occupational Health- Universidad del Valle (Cali Colombia); Msc Health Policy, Planning and Financing- London School of Hygiene and Tropical Medicine/London School of Economics and Political Science, Doctor of Public Health & Policy (Candidate) London School of Hygiene and Tropical Medicine. 43 Kinburn street. London SE16 6DW. United Kingdom. Telephone 4407538616626 ;

ABSTRACT.The objective of this paper is to determine the cost- effectiveness from payers perspective of ERT for Classic infantile form of Pompe´s disease (≤ 6 months of age), usually related to complete deficiency of acid α- glucosidase in two different settings, a High Income Country (HIC) and a Middle Income one (MIC). Classic infantile and late onset forms of Pompe’s disease are very rare and have a combined incidence of 1:40.000 births with different distribution worldwide. Itis unknown precisely how many people suffer fromthis disorder. Methods: literature review of previous clinical studies and historic databases were searched to obtain the natural course of disease in different populations.Bibliographic databases were searched to identify existing cost reimbursed by NHS in England and from third party payers in Colombia. Two Markov processes were constructed to compare; patients movedbetween transitionstates, alive with symptoms and dead. At least 75% survival rate was expected for patients aged ≤ than 6 months under treatment. No long term assessment of health states have been made for populations of patients alive under ERT so far. Most of the parameters were derived from published literature. ERT was assumed to partially restore patient’s health up to a HR- QoL of 0,700 using EQ-5D scale.Results: From the English perspective before running the model the average cost per case per annumwith ERT is £194 K and the ICER per QALY gained was £234.307,7; from the Colombian sidethe average cost per case per annum is £98 K and the ICER per QALY gained was £109.991. One way sensitivity analysis showed that uncertainty about final HR-QoL with ERT, rate of progression of disease and cost related to palliative and supportive inpatient care had the biggest impact on ICER in both models, two way sensitivity analysis showed that if cost of ERT reduces up to 10000 times per dose and a sustained HR-QoLbetween at least 0,750- 0,820is achieved ICERs of £165 K and K£65 could be attainable for England and Colombia respectively. Transactional cost per case in Colombia accounted for an important overload related to logistics and nationalization costs. Conclusions: ERT is more effective that no ERT to treat the disorder, but high levels of uncertainty still remain about rates of survival, progression and quality of life in the long run. ICER showed figures still very high if compared with tacit or explicit international thresholds for decision making, despite it is clear decisions in health are not on the appeal of arbitrary numbers, CE thresholds are still objective national benchmarks related to the affordability criterion of each society. Pompe´s disease treatment is categorized as an orphan drug; hence prices and monopolistic power of manufacturer have a major impact on final CEA results.Further research should be performed to dynamically inform changes incase long run survivals with ERT are achieved in the future.

Pompe´s disease is very rare autosomal recessive disorder,first describedin 1932 by Dutch pathologist J.C. Pompe,it is also known as acid maltase deficiency or glycogen storage disease type II. Disease is characterised by a deficiency of acid α- glucosidase leading to progressive accumulation of glycogen on cardiac, skeletal and respiratory muscle tissue that could result in cardio- respiratory deficiency and death.

There are twoclinical forms of disease, the classical infantile form usually lethal and related to complete deficiency of α- glucosidase and the milder usually later onset form that may present at any age (sub categorized in non classical infantile, childhood, juvenile and adult) usually related to a partial deficiency of enzyme. 1–4

All combined forms of Pompe’s disease have an incidence of 1:40.000 births with different distribution worldwide5,6. Itis unknown precisely how many people suffer fromit. But it is assumed that no more than 5000–10.000 people worldwide are affected. Although it occurs in people of all races, affects African Americans more often (1/14.000 births) than Caucasians (1/60.000 adults and 1/100.000 children)7. There is an estimated incidence of 0.137 per 10,000 people in the European Union.Consequently Pompe´s disease has been classified as an orphan disease, denomination used for rare conditions (less than de 200.000 people in North America and no more than 5 in 10.000 in Europe).

Feeding problems, failure to gain weight, muscular weakness, motor development retardation, cardiac problems, respiratory difficulty, and airway infections were frequently reported as main symptoms during the first months oflife (mean 1,9- 2,1 months) and most infants die within the first year (92-95%) usually after a short period of inpatient care (median 2,8- 4,0 Months) from cardiorespiratoryInsufficiency8.

Late-onset Pompe’s disease presents predominantly as a slowly progressiveproximal myopathythat affects young and middle aged individuals;symptoms are mostly related to mobility problems and limb girdle weakness, disability varies among patients. Delayed diagnosis of more than 5 years after first symptoms in up to 28% of patientsand respiratory problems are a major cause of death for this type of disease.

Before ERT was introduced, treatment was only palliative and supportive; butsince 2006 a recombinant α-glucosidase under the trade name Myozyme® was approved as the first and only drug to be used in Pompe´s disease replacement therapy. Evidence supports that Alglucosidase alpha significantly prolonged total and ventilator free survival in patients with classic infantile onset of disease. Patients with severe deficiency and early onset of ERT (before 6 months of age) seem to show the most marked benefit9.

Under a ERT scheme of 20mg/kg every two weeks it is expected to increase survival rates for 52 weeks from 73% (95% CI 45, 92)to 89% (95% CI: 68,100) according to some clinical studies 10.Cox proportional hazards regressiondemonstrated that Alglucosidasealfa reduced the riskof death by 99% (hazard ratio: 0.01; 95% CI: 0.00, 0.10; p<0.0001) on patients aged ≤ 6 months and 71% (hazard ratio: 0.29; 95% CI: 0.11, 0.81; p<0.018) for patients aged > 6 months to ≤ 36 months. In addition, Alglucosidasealpha treated patients experienced improvements in cardio-myopathy, growth, weight gain, motordevelopment and functional status. According to the manufacturer at least 76% survival rate is expected for patients aged < than 6 months under treatment. No long term assessment of HR-QoL has been made for populations of patients alive with ERT.

Although potentially all patients treated with Myozyme® (Alglucosidase alpha) could develop adverse events ranging from temporary mild to severe ones. The most common adverse reactions requiring intervention were infusion related (up to 51% of cases during the 2 hours following infusion), most of them mild to moderate; clinical trials and post marketing safety experience with Myozyme® have showed approximately 1% of patients developed anaphylactic shock and/or cardiac arrest during Myozyme® infusion that required life-support measures and approximately another 14% have developed other major allergic reactions.

Methods

Eight different studies were reviewed to establish the natural course of disease (early classic form and the late onset form). Since the focus of this paper is in the early onset classic form, one of the main comparative reviews of historic cohorts came from a Dutch study from 2003 8, were a mortality rate of 92% was observed (109/ n=119) during the first year of life after birth for children with Pompe´s disease and no ERT. Some assumptions have been made in constructing the decision model: Since usually death overcomes after a period of 78 days of inpatient palliative and supportive care, this average length of stay was introduced and multiplied by average cost per diem to calculate the average cost of supportive and palliative care, we assumed that all patients will be treated at some point at ICU before death. Two other studies were referenced to determine the effectiveness rate of ERT that was assumed for the model in a 75% survival rate for each year9.

Costs were calculated for inpatient palliative and supportive care at Paediatric Intensive care Unit (PICU) cost, ERT with Myozyme®was set every two weeks in doses of 20mg/Kg (26 per cycle), also non drug costs of ERT infusion and cost of treating complications from ERT application were included. A total probability of 51% for complications from ERT was included, since they require any kind of extra treatment, being severe allergies (including anaphylactic shock and respiratory failure) the most severe ones11, therefore an assumption that any complication form ERT application would account for at least a 10% extra cost of inpatient care. All costs related to the diagnostic phase were not included under the assumption that children will enter the model after being diagnosed and since when disease progressesit is mostly likely treated as inpatient, average ambulatory specialist´s consultations were assumed to be similar for both groups, hence cost related to ambulatory care other than ERT infusion were neglected. Inpatient care costs, ERT cost and costs from treating complications from ERT application for the English perspective were obtained from NHS sources, HRGs, parliament reports and British National Formulary from different years. Non drug costs of ERT application obtained from a Dutch study from 200312. All costs were calculated to present value (according to inflation rates reported by the Office for National Statistics U.K)13-14.For the Colombian perspective, average costs faced by third party payers from the Social Health Insurance Scheme (SHI) were calculated; COOMEVA EPS one of the biggest insurers shared historic data of average inpatient care, ERT, non drug and treating ERT application complications costs.Thenthey were brought to present value according to inflation figures of DANE in Colombia and exchanged from Colombian peso to British Pounds to ease comparison15-16. An extra cost of £4104 per cycle was added to the Colombian model derived from costs of importation and logistics faced by third party payers (£342 per month)All summary costs can be seen in table 1.

Table 1.

Two Markov processes were constructed based on patients moving between transition states alive with symptoms and dead; since the disorder is alwaysprogressive and symptomatic after being diagnosed, no alive asymptomatic transition state was considered. ERT was assumed to partially restore patient’s QoLdue to the limited long term evidence of clinical course with ERT. Motor development impairment and respiratoryinsufficiencyprogression are expected to occur in our model in the long run.

Three scenarios were assumed to classify symptomatic patients alive with Pompe´s disease. UsingEQ-5D from physician´s perspective, the worse off state scenario would be health state 22322 with a utility value of 0,189, a middle state would be 22221 with a utility value of 0,587 and the better off state would be 21211 with a value of 0,81417.Since there is a 50% chance due to uncertainty to move from the better off or middle state to the worse off state, either with or without ERT; we combined these states into two main health states (0,189*0,5)+(0,587*0,5) = 0,388 and (0,587*0,5)+(0,814*0,5) = 0,70017, and under the assumption that ERT favours a higher HR-QoL we assigned the higher combined value to being alive with ERT and the lower combined one to being alive without ERT. Most of the parameters were derived from the published literature. Each cycle was considered yearly based and the model was run for 20 cycles since some expert´s opinion state that age as an option for life expectancy9. Assuming a marked deviation from normal weight gain curves for boys and girls with Pompe´s disease due to chronic illness, 5% Percentile for weight was used as reference framework for calculating total dose of ERT per case 18.Discounting rates were 5% for both costs and effects in each cycle.Effectiveness was calculated in QALYs gained and cost in GBP year 2010. Markov processes were run using TreeAge® pro 2009.

The Markov Model

Results

England´s perspective after running the model showed decision branch No ERT resulted in a cost of £149.178 per 0,16 QALYs gained and branch ERT showed £1.337.118per 5,23 QALYs gained, the incremental cost- effectiveness ratio ICER per QALY was £234.307,7. (Table 2a.)

From the Colombian perspective branch No ERT resulted in a cost of £49.676 per 0,16QALYs gained and branch ERT showed £607.329 per 5,23 QALYs gained, the incremental cost- effectiveness ratio ICER per QALY was £109.991(Table 2b.)

A number of one-way and two way sensitivity analyses are presented to reflect alternative structural assumptions.

Sensitivity analysis 1:ERT has different impact on Health State:

It is likely that ERT does not resolve all disease symptoms and patients remain at least in a mild disease state. In the present model all patients are assumed to live a QoL valued in 0,700 under ERT treatment, some limited literature shows that cardiac hypertrophy may resolve after ERT but other complications like inability to achieve normal motor development, feeding problems and mild respiratory symptomsmight persist or even progress with an expected impact on the HR-QoL value. If glycogen deposits continue to exist even with ERT HR-QoL could be much lower than expected.

Sensitivity analysis 1 included utility values ranging from 0,100 to 0,900. Under these circumstances the ICERunder ERT substantially decreases the more optimistic the assumption is about HR-QoL. From the English perspective at 0,900 the ICER is near to £180 K and slopes up near to £2015 K when value is 0,100. From the Colombian perspective at 0,900the ICER is near to £85 K and slopes up to near to £946 K when value is 0,100 (See graphs 1a and 1b)

Graph1a. English perspective ICER with various health states under ERT.

Graph1b. Colombian Perspective ICER with various health states under ERT..

Sensitivity analysis 2: Pompe´s disease is less aggressive/Better Supportive care

Natural course of disease from different sources accounts for mortality rates per year ranging from 92%- 99% in the classic infantile form without ERT and are estimated to be around 25% under ERT. Although the focus of this study is on infantile form (usually related to total deficiency or very low activity of Acid α- glucosidase), what if mild or late onset forms of disease can be seen with lower rates of mortality and progression also PICUpotential technological innovation could in the future increase survival rates for patients with severe Pompe´s disease under supportive treatment.

This uncertainty in estimates of mortality rates have been incorporated in probabilistic sensitivity analysis lowering hypothetical probability of deathin a rage from 0 -10% to reflect this alternative assumption.

In the English scenario if probability to die from Pompe´s disease is 5% with or without ERT the ICER is 0, but if probability to die raises to 6% ICER for ERT goes up to near £143 K; from the Colombian perspective if probability to die is 4% with or without ERT the ICER is 0, but if probability to die raises to 5% ICER goes up to near £915 K sloping down to near £207K when probability to die is 10%. (See graphs 2a and 2b).

Graphs 2a and 2b. English and Colombian perspectives:ICER if disease were less aggressive

Sensitivity Analysis 3: Treating progression of disease is less expensive

In our model progression of disease accounted for at least 78 days of PICU before death, and total costs of supportivecare were calculated multiplying average length of stay by average cost per diem (PICU). But if effectiveness of ERT reduces either the frequency, average inpatient length of stay or the level of health care required; then it would mean a reduction in total cost of supportive and palliative care, then ICER could be ostensibly different. The effect of altering total cost of supportive treatment resulted in anICER per QALY of £45 K when cost of supportive treatment was 0, an ICER of £92 K when cost was £37 K and so on, up to an ICER of £234K when cost of supportive treatment was £148 K in the English model. An ICER per QALY of £47K when cost of supportive treatment was 0, an ICER of £62,7 when cost was 12K and so on, up to ICER £110 K when cost of supportive treatment was £39 K in the Colombian model (See graphs 3a and 3b)

Graphs 3a and 3b. ICER under different ranges cost of supportive care English and Colombian perspective

Graphs 4a and 4b. ICER under different cost of ERT (20 mg/Kg dose) English and Colombian perspective

Graph 5. ICER under different cost of impotation and nationalization, Colombian Perspective

Graphs 6a and 6b.Two way sensitivity analysis Cost of ERT per Kg and Utility of alive with ERT (WTP £165K England and £65K Colombia).

Analysis 4: Changes in drug cost

ERT is a relatively costly treatment. The base-caseanalysis estimates an annual patient cost per Kg of£3833 for England and £4166.5 for Colombia. In sensitivity analysis 4, the extent towhich the unit cost of ERT drives the cost-effectivenessmodel is identified. The cost per Kgis varied between £0 and the final estimates included in each country and the resulting ICERs are plotted in graphs 4a and 4b. Results showed a mild impact over ICER ranging from £ 200 K at cost per Kg of £0 up to £ 234 K at actual cost in England and from £ 73 K at cost per Kg of £0 up to £ 110 K at actual cost in Colombia.