Limits...
Cost-effectiveness of tipranavir versus comparator protease inhibitor regimens in HIV infected patients previously exposed to antiretroviral therapy in the Netherlands.

Hubben GA, Bos JM, Veltman-Starkenburg CA, Stegmeijer S, Finnern HW, Kappelhoff BS, Simpson KN, Tramarin A, Postma MJ - Cost Eff Resour Alloc (2007)

Bottom Line: We compared the projected lifetime costs and effects of two theoretical groups of 1000 patients, one receiving a standard of care regimen with TPV/r as a component and the other receiving a standard of care regimen with CPI/r.The accumulated discounted effect is 7.43 life years or 6.31 quality adjusted life years (QALYs) per patient receiving TPV/r, compared to 6.91 life years or 5.80 QALYs per patient receiving CPI/r.The corresponding incremental cost effectiveness ratios (iCERs) are euro41,600 per LYG and euro42,500 per QALY.

View Article: PubMed Central - HTML - PubMed

Affiliation: University Center for Pharmacy, University of Groningen, The Netherlands. g.hubben@rug.nl

ABSTRACT

Background: This study compares the costs and effects of a regimen with ritonavir-boosted tipranavir (TPV/r) to a physician-selected genotypically-defined standard-of-care comparator protease inhibitor regimen boosted with ritonavir (CPI/r) in HIV infected patients that were previously exposed to antiretroviral therapy in the Netherlands.

Methods: We compared the projected lifetime costs and effects of two theoretical groups of 1000 patients, one receiving a standard of care regimen with TPV/r as a component and the other receiving a standard of care regimen with CPI/r. A 3-stage Markov model was formulated to represent three different consecutive HAART regimens. The model uses 12 health states based on viral load and CD4+ count to simulate disease progression. The transition probabilities for the Markov model were derived from a United States cohort of treatment experienced HIV patients. Furthermore, the study design was based on 48-week data from the RESIST-2 clinical trial and local Dutch costing data. Cost and health effects were discounted at 4% and 1.5% respectively according to the Dutch guideline. The analysis was conducted from the Dutch healthcare perspective using 2006 unit cost prices.

Results: Our model projects an accumulated discounted cost to the Dutch healthcare system per patient receiving the TPV/r regimen of euro167,200 compared to euro145,400 for the CPI/r regimen. This results in an incremental cost of euro21,800 per patient. The accumulated discounted effect is 7.43 life years or 6.31 quality adjusted life years (QALYs) per patient receiving TPV/r, compared to 6.91 life years or 5.80 QALYs per patient receiving CPI/r. This translates into an incremental effect of TPV/r over CPI/r of 0.52 life years gained (LYG) or 0.51 QALYs gained. The corresponding incremental cost effectiveness ratios (iCERs) are euro41,600 per LYG and euro42,500 per QALY.

Conclusion: We estimated the iCER for TPV/r compared to CPI/r at approximately euro40,000 in treatment experienced HIV-1 infected patients in the Netherlands. This ratio may well be in range of what is acceptable and warrants reimbursement for new drug treatments in the Netherlands, in particular in therapeutic areas as end-stage oncology and HIV and other last-resort health-care interventions.

No MeSH data available.


Related in: MedlinePlus

Diagram of the Markov model. The model consists of three stages representing three different consecutive therapy regimens. All patients start in stage 1 and are assigned an initial health state based on the population of RESIST-2 at the start of the trial. During the first four cycles of the model patients' transitions to other health states and the death stage are identical to those observed in the RESIST-2 trial. After this trial-period, the patients' health state transitions are controlled by transition matrices based on observational data. The main trend over time is towards a less favorable health state indicated by the larger arrow pointing to the right. When patients reach health state 9–12 (treatment failure), they are moved to stage 2 through intermediate period 1. Patients remain in this intermediate period for 1 cycle where they transit to a more favorable health state controlled by improvement matrix 1, represented by the arrow pointing to the left. Patients move from stage 2 to 3 following the same pattern. TPV/r: tipranavir with ritonavir. CPI/r: comparator protease inhibitor with ritonavir. HAART: highly active antiretroviral therapy.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2211743&req=5

Figure 1: Diagram of the Markov model. The model consists of three stages representing three different consecutive therapy regimens. All patients start in stage 1 and are assigned an initial health state based on the population of RESIST-2 at the start of the trial. During the first four cycles of the model patients' transitions to other health states and the death stage are identical to those observed in the RESIST-2 trial. After this trial-period, the patients' health state transitions are controlled by transition matrices based on observational data. The main trend over time is towards a less favorable health state indicated by the larger arrow pointing to the right. When patients reach health state 9–12 (treatment failure), they are moved to stage 2 through intermediate period 1. Patients remain in this intermediate period for 1 cycle where they transit to a more favorable health state controlled by improvement matrix 1, represented by the arrow pointing to the left. Patients move from stage 2 to 3 following the same pattern. TPV/r: tipranavir with ritonavir. CPI/r: comparator protease inhibitor with ritonavir. HAART: highly active antiretroviral therapy.

Mentions: To assess the life-time costs and effects of a TPV/r based regimen compared to CPI/r, we used two theoretical groups of 1000 HIV-1 infected patients in the Netherlands. One group was assumed to receive a regimen containing TPV/r and the other a CPI/r-containing regimen. All patients had previously been exposed to antiretroviral regimens. A 3-stage Markov model was formulated to simulate the costs and effects during the lifetime of this treatment experienced group of patients. A similar model was previously applied to evaluate the cost-effectiveness of lopinavir/ritonavir versus nelfinavir [16]. A diagram of the model is shown in figure 1. The three stages represent three different consecutive HAART regimens. Patients transit to a next stage if their HAART regimen fails. Treatment failure is determined on the patients' health state. The model uses 12 health states based on viral load and CD4+ count to simulate disease progression (table 1). We use a Markov cycle of three months, corresponding to the average time between patients' visits and the associated analysis of markers for disease progression (CD4+ and viral load parameters). The model runs until 90% of the patients have died ("have entered the absorbing death stage in the Markov model"). The analysis was conducted from a Dutch healthcare perspective, implying that only direct medical costs were taken into account and indirect costs due to production losses were excluded. The analysis meets the most recent Dutch guidelines for pharmacoeconomic research, implying that both costs and effects were discounted at 4% and 1.5%, respectively [17]. All costs were measured in Euros using 2006 price levels.


Cost-effectiveness of tipranavir versus comparator protease inhibitor regimens in HIV infected patients previously exposed to antiretroviral therapy in the Netherlands.

Hubben GA, Bos JM, Veltman-Starkenburg CA, Stegmeijer S, Finnern HW, Kappelhoff BS, Simpson KN, Tramarin A, Postma MJ - Cost Eff Resour Alloc (2007)

Diagram of the Markov model. The model consists of three stages representing three different consecutive therapy regimens. All patients start in stage 1 and are assigned an initial health state based on the population of RESIST-2 at the start of the trial. During the first four cycles of the model patients' transitions to other health states and the death stage are identical to those observed in the RESIST-2 trial. After this trial-period, the patients' health state transitions are controlled by transition matrices based on observational data. The main trend over time is towards a less favorable health state indicated by the larger arrow pointing to the right. When patients reach health state 9–12 (treatment failure), they are moved to stage 2 through intermediate period 1. Patients remain in this intermediate period for 1 cycle where they transit to a more favorable health state controlled by improvement matrix 1, represented by the arrow pointing to the left. Patients move from stage 2 to 3 following the same pattern. TPV/r: tipranavir with ritonavir. CPI/r: comparator protease inhibitor with ritonavir. HAART: highly active antiretroviral therapy.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2211743&req=5

Figure 1: Diagram of the Markov model. The model consists of three stages representing three different consecutive therapy regimens. All patients start in stage 1 and are assigned an initial health state based on the population of RESIST-2 at the start of the trial. During the first four cycles of the model patients' transitions to other health states and the death stage are identical to those observed in the RESIST-2 trial. After this trial-period, the patients' health state transitions are controlled by transition matrices based on observational data. The main trend over time is towards a less favorable health state indicated by the larger arrow pointing to the right. When patients reach health state 9–12 (treatment failure), they are moved to stage 2 through intermediate period 1. Patients remain in this intermediate period for 1 cycle where they transit to a more favorable health state controlled by improvement matrix 1, represented by the arrow pointing to the left. Patients move from stage 2 to 3 following the same pattern. TPV/r: tipranavir with ritonavir. CPI/r: comparator protease inhibitor with ritonavir. HAART: highly active antiretroviral therapy.
Mentions: To assess the life-time costs and effects of a TPV/r based regimen compared to CPI/r, we used two theoretical groups of 1000 HIV-1 infected patients in the Netherlands. One group was assumed to receive a regimen containing TPV/r and the other a CPI/r-containing regimen. All patients had previously been exposed to antiretroviral regimens. A 3-stage Markov model was formulated to simulate the costs and effects during the lifetime of this treatment experienced group of patients. A similar model was previously applied to evaluate the cost-effectiveness of lopinavir/ritonavir versus nelfinavir [16]. A diagram of the model is shown in figure 1. The three stages represent three different consecutive HAART regimens. Patients transit to a next stage if their HAART regimen fails. Treatment failure is determined on the patients' health state. The model uses 12 health states based on viral load and CD4+ count to simulate disease progression (table 1). We use a Markov cycle of three months, corresponding to the average time between patients' visits and the associated analysis of markers for disease progression (CD4+ and viral load parameters). The model runs until 90% of the patients have died ("have entered the absorbing death stage in the Markov model"). The analysis was conducted from a Dutch healthcare perspective, implying that only direct medical costs were taken into account and indirect costs due to production losses were excluded. The analysis meets the most recent Dutch guidelines for pharmacoeconomic research, implying that both costs and effects were discounted at 4% and 1.5%, respectively [17]. All costs were measured in Euros using 2006 price levels.

Bottom Line: We compared the projected lifetime costs and effects of two theoretical groups of 1000 patients, one receiving a standard of care regimen with TPV/r as a component and the other receiving a standard of care regimen with CPI/r.The accumulated discounted effect is 7.43 life years or 6.31 quality adjusted life years (QALYs) per patient receiving TPV/r, compared to 6.91 life years or 5.80 QALYs per patient receiving CPI/r.The corresponding incremental cost effectiveness ratios (iCERs) are euro41,600 per LYG and euro42,500 per QALY.

View Article: PubMed Central - HTML - PubMed

Affiliation: University Center for Pharmacy, University of Groningen, The Netherlands. g.hubben@rug.nl

ABSTRACT

Background: This study compares the costs and effects of a regimen with ritonavir-boosted tipranavir (TPV/r) to a physician-selected genotypically-defined standard-of-care comparator protease inhibitor regimen boosted with ritonavir (CPI/r) in HIV infected patients that were previously exposed to antiretroviral therapy in the Netherlands.

Methods: We compared the projected lifetime costs and effects of two theoretical groups of 1000 patients, one receiving a standard of care regimen with TPV/r as a component and the other receiving a standard of care regimen with CPI/r. A 3-stage Markov model was formulated to represent three different consecutive HAART regimens. The model uses 12 health states based on viral load and CD4+ count to simulate disease progression. The transition probabilities for the Markov model were derived from a United States cohort of treatment experienced HIV patients. Furthermore, the study design was based on 48-week data from the RESIST-2 clinical trial and local Dutch costing data. Cost and health effects were discounted at 4% and 1.5% respectively according to the Dutch guideline. The analysis was conducted from the Dutch healthcare perspective using 2006 unit cost prices.

Results: Our model projects an accumulated discounted cost to the Dutch healthcare system per patient receiving the TPV/r regimen of euro167,200 compared to euro145,400 for the CPI/r regimen. This results in an incremental cost of euro21,800 per patient. The accumulated discounted effect is 7.43 life years or 6.31 quality adjusted life years (QALYs) per patient receiving TPV/r, compared to 6.91 life years or 5.80 QALYs per patient receiving CPI/r. This translates into an incremental effect of TPV/r over CPI/r of 0.52 life years gained (LYG) or 0.51 QALYs gained. The corresponding incremental cost effectiveness ratios (iCERs) are euro41,600 per LYG and euro42,500 per QALY.

Conclusion: We estimated the iCER for TPV/r compared to CPI/r at approximately euro40,000 in treatment experienced HIV-1 infected patients in the Netherlands. This ratio may well be in range of what is acceptable and warrants reimbursement for new drug treatments in the Netherlands, in particular in therapeutic areas as end-stage oncology and HIV and other last-resort health-care interventions.

No MeSH data available.


Related in: MedlinePlus