NDT Advance Access originally published online on September 4, 2007
Nephrology Dialysis Transplantation 2007 22(11):3355-3357; doi:10.1093/ndt/gfm434
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Cost–utility analysis of cinacalcet in addition to standard of care in the UK
Correspondence and offprint requests to: Email: valentine{at}thecenter.chSir,
We would like to comment on a manuscript describing the cost-effectiveness of cinacalcet plus standard care vs standard care alone for patients with hyperparathyroidism secondary to end-stage renal disease (ESRD) in the UK setting [1]. In our opinion, the value of the conclusions drawn is limited, due to several serious shortcomings and biases in the analysis.
Garside et al. [1] described a health economic model developed to support the independent evaluation of cinacalcet for the National Institute of Health and Clinical Excellence (NICE), the agency responsible for making reimbursement decisions about medical technologies in England and Wales. We do not believe that the economic model described and used by Garside et al. adequately represents the economic impact of cinacalcet on the treatment of hyperparathyroidism secondary to ESRD [2].
Mixed results on the relationship between mineral metabolism and death in patients on haemodialysis have been reported, but these trials did not monitor patients over a specific period of time [3–7]. More recent data indicate that the risk of death is increased when mineral metabolism [i.e. levels of phosphorus, calcium, parathyroid hormone (PTH)] is uncontrolled over a prolonged period of time [8]. This prospective study by Melamed et al. [8] reported that patients with intact PTH levels >308 pg/ml had a 23% increased risk of death compared with those with iPTH 160–308 pg/ml. In contrast, intact PTH <76 pg/ml was associated with a lower mortality risk compared with levels of 160–310 pg/ml. The economic analysis by Garside et al. relied on data from one epidemiological study that reported a much lower increase in the risk of death associated with mineral metabolism [4]. Furthermore, the assumptions in their economic analysis failed to account for the time dependency between PTH levels and death. The authors state that the relative risk of mortality for patients with ‘uncontrolled’ and ‘very uncontrolled’ PTH level was one of the primary points of sensitivity within the model, although they were not explicit regarding the ranges for this variable and the quantitative impact on the results.
In their economic analysis, Garside et al. have assumed that the average daily dose of cinacalcet in routine use is 94 mg/day. This average dose was derived from clinical trials that were developed before the KDOQITM guidelines were published and the dosing strategy attempted to achieve plasma PTH levels within a range modestly lower than the recommended KDOQITM guidelines that were published subsequently [9,10]. Recent studies have indicated that the median dose of cinacalcet used in clinical practice is 60 mg/day [11]. Indeed, based on this evidence, the World Health Organization now defines a daily dose of 60 mg cinacalcet, which is much lower than the dose used (94 mg/day) in the economic analysis [12]. Garside et al. conclude that cinacalcet treatment would be considered cost-effective if drug costs are considerably reduced; this conclusion is flawed since the appropriate sensitivity analysis based on the recently recommended doses was not performed.
The details of the model are very well documented within the manuscript [1]. However, several key factors used in the analysis are based upon assumptions and inappropriate guesswork. One parameter is the initial distribution of patients within three PTH strata at baseline. The clinical trial results did not specify the proportion of patients who had ‘uncontrolled’ or ‘very uncontrolled’ PTH values, therefore, the stratification estimates were derived from the UK Renal Registry. The authors also state that all patients are entered into the model with the same condition, although a 3-month titration period is accounted for, in which patients were distributed into the three PTH strata. For example, an assumption was made that 40% of patients had controlled PTH levels. However, the trial data indicated that 56% of patients in the pooled data set had truly controlled PTH levels. Thus, the authors’ assumption led to an over estimation of cost-effectiveness ratio. Randomized controlled trials are widely accepted as providing the best evidence for estimates of efficacy, so it is not clear why the authors made this assumption rather than use published evidence from such trials. In our opinion, the analysis would be a more balanced assessment of cinacalcet cost-effectiveness if the baseline proportions stratified in each treatment arm without any intervention were identical, rather than biasing against the cinacalcet treatment arm from the outset.
In the economic model [1], distributions were applied to various parameters to account for uncertainty. Results were calculated using probabilistic sensitivity analysis (second-order Monte Carlo simulation) as recommended by NICE [13], as well as using a deterministic approach in the base case. It is noteworthy, however, that in the base case, Garside et al. assumed that cinacalcet-treated patients did not experience a deterioration of PTH control as part of the base-case assumption within this analysis. However, when the results were calculated probabilistically, PTH control in cinacalcet-treated patients was allowed to deteriorate according to expert opinion, based on a distribution which did not include the probability used in the base-case analysis [14].
It is also unclear how the model by Garside et al. accounted for the withdrawal of patients from each treatment arm, and what impact this had on effectiveness and treatment costs over time. The authors do state that the difference in withdrawal rates (15% from the cinacalcet arm; 8% from the standard care arm) was accounted for in the model. However, 8% of cinacalcet patients terminated treatment after the initial titration period. The manuscript does not clarify whether these patients assumed the same transition probabilities as those treated with standard of care, or if they were excluded from the model entirely. Both of these evaluations would have implications on reported outcomes; although the actual impact of either assumption on the final outcome is unclear.
A number of additional assumptions applied in the present analysis could also have a significant impact on projected outcomes. For example, the authors assume that the multiplier for the increased risk of fracture following parathyroidectomy surgery is 1 (i.e. no increased risk). This assumption is not valid, since the risk of fracture prior to surgery was 1.94, due to the long-term effects on bone decalcification. In our opinion, the base-case analysis would have been better applied if the base-case value was 1.94 rather than 1. Using the inappropriate value of 1 also biases against cinacalcet. The present modelling analysis is also based on the assumption that there is no difference in the health state utilities between the ‘uncontrolled’ and ‘controlled’ PTH health states. Given the impact of increased PTH levels on bone pain and pruritis, it would appear reasonable that a utility value between the health state utility for controlled PTH levels and that for ‘very uncontrolled’ levels should be applied to the uncontrolled strata of the model.
Some of the issues discussed above were addressed in the model by using a one-way sensitivity analysis. Furthermore, when each parameter was varied individually within the model, the impact on the final outcome was not large. However, we believe that if more reasonable (and less biased) assumptions were used simultaneously, the incremental cost-effectiveness ratio would very likely be below the current NICE threshold. In our opinion, a new base-case analysis is required, using the following assumptions:
- Adjusted distribution of patients within the three strata at baseline.
- Adjusted real-life dose of cinacalcet.
- Adjusted risk of death to account for previous PTH levels.
- Revised estimate of the fracture risk following parathyroidectomy procedure.
- Inclusion of a disutility adjustment for patients with ‘uncontrolled’ PTH levels.
In conclusion, we believe that there are a number of shortcomings and biases against the cinacalcet treatment arm in the base-case analysis used in the economic model by Garside et al. In addition, certain key parameters have been derived from opinion rather than based on clinical evidence. In our opinion, the combination of these weaknesses and biases mean that the results of this analysis are potentially misleading. Furthermore, their conclusions are in contrast to those of NICE.
Conflict of interest statement: At the time of writing, the authors were employees of IMS Health, which has received funding from Amgen Inc.
IMS Health
Allschwil
Switzerland
References
- Garside R, Pitt M, Anderson R, Mealing S, D'Souza R, Stein K. The cost-utility of cinacalcet in addition to standard care compared to standard care alone for secondary hyperparathyroidism in end-stage renal disease: a UK perspective. Nephrol Dial Transplant (2007) 22:1428–1436.
[Abstract/Free Full Text] - National Institute for Health and Clinical Excellence. Cinacalcet for treatment of secondary hyperparathyroidism in patients with end-stage renal disease on maintenance dialysis therapy (Technology appraisal guidance 117). (2007) Available from: http://www.nice.org.uk/guidance/TA117.
- Block GA, Hulbert-Shearon TE, Levin NW, Port FK. Association of serum phosphorus and calcium x phosphate product with mortality risk in chronic hemodialysis patients: a national study. Am J Kidney Dis (1998) 31:607–617.[Web of Science][Medline]
- Block GA, Klassen PS, Lazarus JM, Ofsthun N, Lowrie EG, Chertow GM. Mineral metabolism, mortality, and morbidity in maintenance hemodialysis. J Am Soc Nephrol (2004) 15:2208–2218.
[Abstract/Free Full Text] - Ganesh SK, Stack AG, Levin NW, Hulbert-Shearon T, Port FK. Association of elevated serum PO(4), Ca x PO(4) product, and parathyroid hormone with cardiac mortality risk in chronic hemodialysis patients. J Am Soc Nephrol (2001) 12:2131–2138.
[Abstract/Free Full Text] - Slinin Y, Foley RN, Collins AJ. Calcium, phosphorus, parathyroid hormone, and cardiovascular disease in hemodialysis patients: the USRDS waves 1, 3, and 4 study. J Am Soc Nephrol (2005) 16:1788–1793.
[Abstract/Free Full Text] - Teng M, Wolf M, Ofsthun MN, et al. Activated injectable vitamin D and hemodialysis survival: a historical cohort study. J Am Soc Nephrol (2005) 16:1115–1125.
[Abstract/Free Full Text] - Melamed ML, Eustace JA, Plantinga L, et al. Changes in serum calcium, phosphate, and PTH and the risk of death in incident dialysis patients: a longitudinal study. Kidney Int (2006) 70:351–357.[CrossRef][Web of Science][Medline]
- Cunningham J, Danese M, Olson K, Klassen P, Chertow GM. Effects of the calcimimetic cinacalcet HCl on cardiovascular disease, fracture, and health-related quality of life in secondary hyperparathyroidism. Kidney Int (2005) 68:1793–1800.[CrossRef][Web of Science][Medline]
- K/DOQI clinical practice guidelines for bone metabolism and disease in chronic kidney disease. Am J Kidney Dis (2003) 42(Suppl 3):S1–S201.[Medline]
- Ureña P, Neyer U, Jacobson SH, et al. Cinacalcet (Mimpara®/Sensipar®) brought more patients to within NKF/KDOQITM recommended targets for bone and mineral metabolism in ‘real-world’ clinical practice: The ECHO pan-European study. World Congress of Nephrology Abstract Book (2007) 69. (Abstract S-PO-0079).
- World Health Organization Collaborating Centre for Drug Statistics Methodology: Oslo. The anatomical therapeutic chemical classification system with defined daily doses (ATC/DDD). (2003).
- National Institute for Clinical Excellence. Technical guidance for manufacturers and sponsors on making a submission to a technology appraisal. (2001) Available from: http://www.nice.org.uk/page.aspx?o=16183.
- Phillips B, Ball C, Sackett DL, et al. Evidence-based on-call: levels of evidence. In: Evidence-based on-call 2007. Available from: http://www.eboncall.org/.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||