MODELLING FOR RARE DISEASES- CASE OF CANAKINUMAB IN PERIODIC FEVER SYNDROMES IN FRANCE
Author(s)
Duteil E1, Cawston H2, Pibouleau L2, Bregman C2, Mahieu N1, Cariou C1, Sion M3, Duco J4
1Novartis Pharma, Rueil-Malmaison, France, 2Amaris, London, UK, 3Novartis Pharma SAS, Rueil-Malmaison, France, 4Novartis Pharma, Rueil Malmaison, France
OBJECTIVES: Canakinumab, an interleukin-1 inhibitor is indicated in periodic fever syndromes (PFS), including familial Mediterranean fever resistant to colchicine, mevalonate kinase deficiency, tumour necrosis factor receptor associated periodic syndrome, in adults, adolescents and children aged 2 years and older. Through the example of canakinumab in France, the objective was to evaluate the challenges related to modelling rare diseases. METHODS: No model in PFS was identified in the literature. Clinical inputs regarding the progression of the disease and its management were obtained through clinicians’ interviews and questionnaires. Among complications associated with amyloidosis, only renal complications were included due to lack of data. Transition probabilities for amyloidosis complications were obtained from a UK study including PFS patients. While efficacy and utility data came from the phase III CLUSTER study (no stratification by age group was possible due to low sample sizes), safety data were obtained from the Beta-Confident registry, and long-term persistence data were obtained from the French ENVOL observational study. All-cause, amyloidosis and end stage renal disease (ESRD) related-mortality were accounted for. RESULTS: The model combined a 16-week decision tree evaluating response and dosage patterns with a Markov model estimating the lifetime consequences of the disease. Health states were: response on canakinumab by dosage, spontaneous response, non-response, amyloidosis, ESRD and death. Three age-group cohorts were used to better account for the disease management in children. Canakinumab was associated with a higher number of life years and QALYs, although the utility data was uncertain. Canakinumab had a lower number of flares, which also resulted into a reduction in caregiver time. CONCLUSIONS: Modelling rare diseases is associated with high uncertainty, linked to low sample sizes, few long-term real world data, and requires relying on clinician inputs. Therefore, budget impact analysis is useful to discuss the financial resources Health Insurance will have to support.
Conference/Value in Health Info
2017-11, ISPOR Europe 2017, Glasgow, Scotland
Value in Health, Vol. 20, No. 9 (October 2017)
Code
PSY70
Topic
Economic Evaluation
Topic Subcategory
Cost-comparison, Effectiveness, Utility, Benefit Analysis
Disease
Pediatrics, Rare and Orphan Diseases