ARTICLE SUPPLEMENTARY INFORMATION
(Supporting information is provided by the author to support the content
of the parent article)

Transferability of Model-Based Economic Evaluations: The Case of Trastuzumab for the Adjuvant Treatment of HER2-Positive Early Breast Cancer in the Netherlands

Brigitte A.B. Essers, Shanley C. Seferina, Vivianne C.G. Tjan-Heijnen, Johan L. Severens, Annoesjka Novák, Marjolein Pompen, Ulrich H. Oron, Manuela A. Joore

– Value in Health Supplemental Information

Appendix step 2: Adaptations made to the UK model-based cost-effectiveness study

General input

The price level was adapted to 2006. The UK discount rates were replaced by the Dutch discount rates for costs (4%) and effects (1.5%).

Transition probabilities

The UK mortality rates for background mortality were replaced by the 2006 Dutch one-year mortality rates for women aged 50 and older[1] .

Utilities

The UK model applied utilities derived from three different studies [2-4]. Although no Dutch information on health state utilities was available, we preferred to apply utilities from a single and more recent Swedish study ][5]. In the UK model, the estimates for disutilities for local and contralateral recurrence and cardiac side effects were based on a study by Van Hanswijck de Jonge [6], and assumption [7]. In the absence of Dutch estimates for these parameters, we used the estimates from the UK model.

Health care resource use and unit prices

Cardiac related costs: The costs of heart monitoring during treatment with trastuzumab were assumed to consist of five MUGA-scans (at baseline, 3, 6, 9, and 12 months= 5 * € 255 ). This amounts to € 1275 per patient. For the upper (€ 705) and lower value (€ 1394), the number of MUGA scans is varied. Since resource use data for cardiac events were not readily available, we used Dutch data on the costs of illness of women suffering from heart failure [8]. The total costs of curative care for heart failure amounted to € 123,9 million in 2003. The number of newly diagnosed female patients with heart failure in 2003 was 19.200 [9]. Accordingly, the average annual costs of curative care for a female heart failure patient amount to € 6.453 (updated to 2006 € 7.113). It was assumed that the annual costs of a ‘severe cardiac adverse event’ would be twice the average costs of an average female patient with heart failure in the Netherlands. The annual costs of ‘other than severe cardiac adverse events’ and ‘chronic cardiac adverse events’ were assumed to be equal to the average annual costs of a female patient with heart failure.

Direct costs of breast cancer for different health states: Dutch estimates of resource use in the health states was not available, therefore UK estimates of resource use were used. Unit costs of therapies were obtained from the Dutch Pharmaceutical Compass [10]. A few pharmacological therapies were not listed in the Dutch Pharmaceutical Compass, these were left out of the analysis. The unit costs of surgical procedures, radiotherapy, investigations and laboratory tests, hospitalization, and outpatient visits were derived from either the Dutch manual for costing in economic evaluations [11], the standard cost-prices from the Dutch Health Care Insurance Board [12], from the Maastricht University Medical Centre or literature. Since hospice care is still quite uncommon in the Netherlands (the vast majority of patients in the Netherlands die at home, sometimes with home care), these costs were left out of the analyses.

Costs of HER2 test: In order to determine whether a patient qualifies for treatment with trastuzumab, the HER2 status and cardiac function need to be assessed. HER2 status is usually determined with immunohistochemistry (IHC) test. Patients with borderline test results (2+) receive in addition a fluorescence in situ hybridization (FISH) test to confirm the test result. The Dutch unit costs of the IHC test were estimated to amount to € 40, and the FISH test  €125 (personal communication: DAKO Benelux). The proportion of patients receiving the FISH test was based on figures from a Dutch laboratory (Stichting PAMM, 2006). In 2006, this laboratory performed for 620 women a IHC test, with in 91 cases also a FISH test. This indicates that the proportion of women requiring a FISH test in order to confirm the results of the IHC test is 15%. In the Netherlands, women whose tumours were shown HER2 positive also have their cardiac function assessed by means of a MUGA scan (€255). Unfortunately, the proportion of women that is excluded for treatment with trastuzumab because of cardiac dysfunction after having tested HER2 positive is not known. We assumed this proportion to be 10%. Based on these estimates, the average costs of testing amount to € 114 per patient.

Costs of trastuzumab: The total costs of treatment with trastuzumab consist of the costs of administered trastuzumab, vial wastage, and administration. The costs of vial wastage were not included in the UK model. The costs of trastuzumab are based on the Dutch unit price per 150 mg vial of € 627 (UK: £407=€ 570). In the HERA trial, the average number of vials used per cycle was 3.3, and the number of cycles was 18 (one loading dose and 17 maintenance doses). In total, this amounts to € 37.243. However, vials are never used completely which results in vial wastage. In a Belgian study, it was estimated that the amount of trastuzumab remaining in vials was 10.6% of the loading dose and 15,6% of the maintenance doses [13]. Based on these estimates of vial wastage, and an average of 3.3 vials per cycle, the costs associated with vial wastage amount to € 5.708 per patient. The costs of the iv administration amount to € 217 per cycle (UK: € 221). In total, the costs of treatment with trastuzumab amount to € 46.867 (UK: £ 24.420; € 34.188, 1.4 exchange rate) per patient in the adjuvant setting.

Productivity loss costs

In the UK model costs from lost productivity were not included. For the Dutch model, we used the friction cost method. We assumed that for women in the recurrence health state and the metastatic health state, the number of work days lost was equal to the length of the friction period (65 days, taking into account 0.8 correction factor). The cost per working day lost is based on the average of the productivity costs per hour for women aged 45 years and older as reported in Oostenbrink et al.[11], and assuming a work day of 8 hours (updated to 2006: € 308 per day).

References

1. Central Bureau for statistics. www.cbs.nl
2. Tengs TO, Wallace A. One thousand health-related quality-of-life estimates. Med care 2000;38(6):583-637
3. Hayman JA, Hillner BE, Harris JR, Weeks JC. Cost-effectivness of routine radiation therapy following conservative surgery for early-stage breast cancer. J Clin Oncol 1998;16(3):1022-29
4. Carter KJ, Ritchey NP, Castro F, Caccamo LP et al. Treatment of early-stage breast cancer in the elderly: a health-outcome based approach. Med. Decis Making 1998; 18(2):213-19.
5. Lidgren M, Wilking N, Jonsson B, Rehnberg C. Health related Quality of life in different states of breast cancer. Qual Life Res 2007 Aug 16 (6):1073-81.
6. Van Hanswijck de Jonge P et al. Poster 190. Presented at ESMO 2006.
7. Early Breast Cancer Trialists’ Collaborative group (EBCTCG). Lancet 2005;365:1687-1717.
8. Slobbe LCJ, Kommer GJ, Smit JM, Groen J et al. Kosten van ziekten in Nederland 2003. RIVM 2006. www.kostenvanziekten.nl
9. National Institute for Public Health and the Environment.www.rivm.nl
10. Health Care Insurance Board. Pharmacotherapuetic Compass (in Dutch). www.fk.cvz.nl. Amstelveen. 2006.
11. Oostenbrink JB, Bouwmans CAM, Koopmanschap MA, Rutten FFH. Dutch manual for costing research (in Dutch). Health Care Insurance Board.2004
12.  www.cvz.nl. Health Care Insurance Board. Amstelveen.2006.
13.  Federaal Kenniscentrum voor de Gezondheidszorg. Trastuzumab bij vroegtijdige stadia van borstkanker. KCE report 34a. Brussels. 2006.www.kce.fgov.be

Value in Health Article Supplementary Information