Marko Obradovic

Cost-effectiveness of UGT1A1 genotyping in second-line, high-dose, once every 3 weeks irinotecan monotherapy treatment of colorectal cancer.

AIM The aim of the present study was to evaluate the cost-effectiveness of UGT1A1 genotyping in second-line, high-dose, once every 3 weeks irinotecan monotherapy treatment of colorectal cancer. METHODS Standard therapy was compared with alternative strategies based on UGT1A1 genotyping from the US healthcare payer perspective. Two alternative strategies (dose reduction and prophylactic use of G-CSF with prior genotyping) and standard therapy were evaluated in a decision analysis, whereas alternative regimens were considered in discussion. The effectiveness outcome was severe neutropenia occurrence and number of life-years gained. RESULTS & CONCLUSION Genotyping in combination with a subsequent reduction of initial irinotecan dose for UGT1A1 7/7 genotype patients was cost-saving for the population of African and Caucasian origin. By contrast, UGT1A1 genotyping was not cost effective for the population of Asian ancestry. Furthermore, the prophylactic use of G-CSFs in UGT1A1 7/7 genotype patients was not cost effective in any population group. Finally, the application of a 3-weekly high-dose treatment regimen with a 20% reduced dosage compared with the low-dose weekly irinotecan regimen in patients with UGT1A1 7/7 genotype was less expensive and is more convenient for the patient.

Cost-effectiveness of antipsychotics for outpatients with chronic schizophrenia.

Aims:  The aim of the present analysis was to evaluate the cost‐effectiveness of alternative treatments for outpatients with chronic schizophrenia from the healthcare payers perspective.

Cost-effectiveness analysis of HPV vaccination alongside cervical cancer screening programme in Slovenia.

BACKGROUND The objective of the present study is to evaluate the cost-effectiveness of human papillomavirus (HPV) vaccination alongside cervical cancer screening programme in Slovenia. METHODS A previously published Markov model representing natural history of HPV infection was adapted to Slovenian context. The model followed a cohort of 12-year-old girls to 85-year-old women. Two strategies were compared: HPV vaccination alongside conventional cytological screening versus screening alone. Analysis was performed from the health care payer perspective. RESULTS Vaccination with screening compared with screening alone was associated with an incremental cost-effectiveness ratio (ICER) of 23,178 EUR per quality adjusted life-year (QALY) gained and 54,536 EUR per life-year gained (LYG) at the discounting rate of 5%. Sensitivity analyses demonstrated that the ICER was most sensitive to the need for booster dose and to different values of discount rates. In case the booster dose was assumed 10 years after initial vaccination, the ICER value was increased to 58,690 EUR per QALY. On the other hand, using lower values of discount rates than the base case 5% significantly reduced the ICER value. CONCLUSION According to the cost-effectiveness thresholds of 30,000 EUR per QALY which was adopted by the Health Council in Slovenia, HPV vaccination alongside screening programme can be regarded as cost-effective. However, cost-effectiveness of HPV vaccination would become questionable in case a booster dose was needed to provide lifetime protection.

Market uptake of biologic and small-molecule—Targeted oncology drugs in Europe

OBJECTIVE The aim of this study was to investigate the market uptake of biologic and small-molecule-targeted oncology drugs in Europe. METHODS Targeted oncology drugs that were used in one of the selected European countries before the end of 2007 were eligible for inclusion in the analysis. The following European countries were included: Austria, Croatia, France, Germany, Hungary, Italy, Slovenia, and the United Kingdom. Monetary market uptake of targeted oncology drugs was assessed by using sales data (in euros) obtained from 2 large data- bases for the period 1997-2007. Market uptake was assessed in terms of expenditures for specific drugs in euros per capita and in market shares. RESULTS The monetary market uptake of targeted oncology drugs had an exponential growth from 1997 to 2007 in all comparison countries and reached 40% of the total oncology drug market in 2007. Although the various European countries allocate substantially different amounts of resources per capita for oncology drugs, the share of expenditures attributed to targeted oncology drugs did not differ substantially among the countries. Biologic molecules were used in clinical practice before the small-molecule-targeted oncology drugs. Targeted oncology drugs that were introduced first to clinical practice in most of the comparison countries (ie, rituximab, trastuzumab, imatinib mesylate) maintained the leading positions on the market throughout the period of the analysis. In 2007, approximately 25% of all expenditures for oncology drugs were attributed to biologic oncology drugs, and approximately 15% were spent on small-molecule-targeted oncology drugs. CONCLUSIONS Expenditures on targeted oncology drugs have been increasing exponentially in Europe throughout the past decade and have reached a 40% share of the oncology drug market. As of 2007, the market share of biologic oncology drugs was higher than the market share of small-molecule-targeted oncology drugs.

Response to the comments on ‘Cost‐effectiveness of antipsychotics for outpatients with chronic schizophrenia’ by Steve S. Koh and Jonathan M. Meyer

To the Editor: Cost-effectiveness analysis (CEA) and decision analysis are increasingly employed in healthcare systems to determine optimal resource allocation. Obradovic and colleagues’ use of CEA to examine relative cost-effectiveness between antipsychotic treatments (‘‘Costeffectiveness of antipsychotics for outpatients with chronic schizophrenia’’ December 2007) illustrates the potential value of such analyses, but also the pitfalls. The proposed decision analysis model examines one-year schizophrenia remission rates as the primary outcome (ie. terminal nodes), based on one decision node (antipsychotic choice), and two main chance nodes dependent on differential rates of medication compliance and relapse with hospitalization. Unfortunately, the model ignores the preferential value a patient may place onto each terminal result from an antipsychotic use. The authors define cost effectiveness assuming the most preferential outcome of antipsychotic choice is avoidance of hospitalization, as opposed to avoidance of adverse effects; however, for some patients a decreased remission rate does not necessarily outweigh adverse effect considerations. The true assessment of outcome values requires a detailed analysis of expressed preference, information not readily obtainable in most antipsychotic clinical trials. Another important issue is that, for CEA and decision analysis, the quality of input data determines the overall strength of the analysis result. This point is particularly important in decision analysis because different probabilities can significantly affect the outcome measures. For example, the main studies used for compliance and hospitalization rates (CATIE and EFESO) have fundamental differences in design, patient populations, and reported outcomes. By combining these diverse studies, the authors introduce uncertainties in the probability values of each chance node, a source of error that becomes magnified at the terminal values. Additionally, if we accept that the individual costs cited by the authors are correct and represent exhaustive cost categories associated with respective antipsychotic use, it is incorrect to assume that the cost involved in choosing a particular antipsychotic is a summation of all possible costs over one year horizon. Weight gain in particular is problematic when assessed just on a cost of treatment basis, because treatment emergent weight gain has a direct impact on medication adherence and relapse risk. Moreover, since the terminal node is that of relapse rate and hospitalization, by including hospital cost into sum cost of antipsychotic use, the study paradoxically use the probability rate of hospitalization twice in its analysis. For these reasons, a strong argument can be made for confining the cost assessment to the known cost of medication administration only. In clinical practice, where outcomes are often unpredictable, decision analysis may prove to be burdensome, difficult to perform, and at worst, lead to poor decisions for patient care. We hope that further clinical research will provide more refined data on antipsychotic outcomes to inform future decision analysis and CEA. These analytic methods have significant promise to provide guidance for healthcare planners and clinicians caring for patients with schizophrenia; however, given the limitations of such analyses, the implementation of decision analysis in clinical practice guidelines must be exhaustively evaluated based on initial assumptions and the data sources used.