a review of economic evaluations of darunavir boosted by low-dose ritonavir in treatment-experienced...

A Review of Economic Evaluationsof Darunavir Boosted by Low-DoseRitonavir in Treatment-ExperiencedPersons Living with HIV InfectionJosephine Mauskopf,1 Lieven Annemans,2 Andrew M. Hill3,4 and Erik Smets5

1 RTI Health Solutions, Research Triangle Park, NC, USA

2 Department of Public Health, Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium

3 Department of Pharmacology, University of Liverpool, Liverpool, UK

4 Tibotec Pharmaceuticals BVBA, Mechelen, Belgium

5 Johnson & Johnson Pharmaceutical Services LLC, Mechelen, Belgium

Abstract Darunavir boosted by low-dose ritonavir (DRV/r), at a daily dose of600/100mg twice a day (bid), has been shown to be superior to alternativehighly active antiretroviral therapy (HAART) regimens for the managementof treatment-experienced, HIV-infected adults in the phase IIb POWER trialsand the phase III TITAN trial.

Economic analyses of different types that have been performed for severalcountries to investigate the cost effectiveness and budgetary impact of DRV/r600/100mg bid for treatment-experienced people living with HIV (PLHIV)based on the clinical data gathered in the POWER and TITAN trialsare reviewed for consistency and their value to different decision-makers isassessed.

Cost-utility analyses for the USA and several European countries indicatethat DRV/r-based HAART is cost effective compared with other standard ofcare protease inhibitor (PI)-based regimens in PLHIV with evidence of PIresistance. For all of these countries, the estimated cost-utility ratio is wellbelow typical benchmark values and these ratios are robust, as demonstratedby one-way sensitivity and variability analyses and multi-way probabilisticsensitivity analyses.

Studies using other metrics including the average 1-year drug costper patient with a plasmaHIV-RNA level less than 50 copies/mL at 48 weeks,the incremental drug cost per additional patient with a plasma HIV-RNAlevel less than 50 copies/mL at 48 weeks, the total (antiretroviral and non-antiretroviral) costs during the first year of treatment, and the total health-care budget impact during the first 5 years of treatment provided furtherevidence of the positive economic outcomes with the use ofDRV/r in treatment-experienced PLHIV.

Different measures of economic outcomes are useful for different types ofdecision-makers and different types of decisions. In general, the resultsof these different types of analyses will be consistent with each other. For

REVIEWARTICLEPharmacoeconomics 2010; 28 Suppl. 1: 1-16

1170-7690/10/0001-0001/$49.95/0

ª 2010 Adis Data Information BV. All rights reserved.

darunavir, the economic analyses reviewed in this paper demonstrate that theuse of DRV/r 600/100mg bid in the management of HIV-infected, treatment-experienced adults who have failed at least one of the other currently avail-able PIs is cost effective and may be cost saving.

1. Introduction

The effectiveness of new drugs for treatingHIV infection is measured using several outcomesmeasures. These include direct outcome measuressuch as life-years gained and quality-adjusted life-years (QALYs) gained, as well as surrogate out-come measures such as viral suppression, plasmaHIV-RNA reduction and CD4 cell count in-crease. These surrogate measures have beenproved in long-term follow-up studies to beassociated with reduced disease progression,[1]

and thus are well accepted as measures of the ef-fectiveness of a new treatment regimen in theabsence of long-term data on the direct outcomesmeasures.

Economic implications are associated with theclinical efficacy of alternative treatments dem-onstrated in clinical trials because HIV-infectedindividuals in the later stages of HIV infection(i.e. those with advanced immune deficiency) arethe most costly to treat. Elsewhere in this sup-plement, several authors report how direct costsof HIV treatment and care vary according tothe degree of HIV-induced immune deficiency(characterized by the CD4 cell count).[2-4] Allthese papers demonstrate that, as the CD4 cellcount decreases, costs increase, with those peopleliving with HIV (PLHIV) with CD4 cell countsbelow 50–100 cells/mm3 (depending on the level ofdetail in the CD4 cell count segmentation) havinghigher costs than thosewith higher CD4 cell counts.

To supplement measures of the disease out-comes of virological and immunological responseassociated with alternative treatment regimensand measures of costs by CD4 cell count range,disease progression models have been developedfor estimating the cost effectiveness and/or bud-get impact of new treatment regimens. These esti-mates are important to decision-makers as inputsto a determination of the value of new treatmentregimens and to help them to plan their HIV/

AIDS healthcare expenditures after the approvalof the new drug. Cost-effectiveness models maypresent estimates of the efficiency of the treat-ment regimen including the new drug comparedwith currently used treatment regimens from asocietal or decision-maker perspective. The cost-effectiveness ratio, estimated comparing the costsand effectiveness with the new treatment regimenwith current standard treatment regimens, is ameasure of the extent to which the new regimen isgood value for money. For chronic diseases, thesemodels generally take a lifetime perspective, al-though in the case of HIV infection in which newtreatments are frequent, a shorter time horizon,such as 5 or 10 years may also be of interest.

Both Markov models and simulation modelshave been developed for estimating the cost ef-fectiveness of alternative treatment regimens forHIV infection.[5-13] The outputs of these modelsare frequently presented from the societal per-spective as the ratio of the incremental lifetimetotal healthcare costs and incremental lifetimeQALYs when comparing the new treatment re-gimen with the standard regimen. Alternatively,cost-effectiveness ratios can be presented withshorter time horizons, with the denominator innatural units and from the decision-maker per-spective rather than the societal perspective. Forexample, the change in 48-week drug cost per in-creased number of people achieving a plasmaHIV-RNA level below detectable at 48 weeks or thenumber of people needing to be treated to achieveone extra person with a plasma HIV-RNA levelbelow detectable at 48 weeks could be estimatedusing data from a 48-week clinical trial, and arealternative measures of the impact of a new treat-ment regimen on both costs and health out-comes that might provide useful information todecision-makers.

Budget impact models are also helpful todecision-makers to help them to plan the impacton their HIV infection healthcare expenditures

2 Mauskopf et al.

ª 2010 Adis Data Information BV. All rights reserved. Pharmacoeconomics 2010; 28 Suppl. 1

when the new drug is approved. Budget impactscan be estimated by adapting the Markov orsimulation models developed for the cost-effectiveness estimates to take a population pre-valence perspective[7,14] or they can be estimatedusing simpler estimation methods,[15,16] compar-ing directly alternative treatment scenarios withand without the new drug on the formulary usingthe data from clinical trials to guide the estimatesof offsetting cost savings as a result of sloweddisease progression.

Darunavir, a novel protease inhibitor (PI) witha high resistance barrier, is currently approvedfor the treatment of HIV infection in treatment-experienced adults who have failed previousantiretroviral therapy (ART). Darunavir boostedby low-dose ritonavir (DRV/r) has been shownin clinical trials to have better efficacy thaninvestigator-selected, currently available PIs aspart of an optimised treatment regimen in thetreatment-experienced population. In particular,the results from DRV/r clinical trials in thetreatment-experienced population indicate thatsignificantly more people in the DRV/r treatmentgroups experienced complete viral suppressionup to 144 weeks than the control groups,[17-24]

and had similar or greater increases in CD4 cellcount than the control group.

In this review paper, we describe a seriesof economic evaluations of DRV/r as part ofa multidrug regimen in people who have pre-viously failed treatment with a PI carried out inseveral different countries and using differentmethods.[12,13,25-27] All the studies use as the basisfor their costs and benefits estimates the efficacyresults of the DRV/r clinical trials in treatment-experienced individuals. Each study, however,estimates a different aspect of the economics ofDRV/r including:� Comprehensive disease progression cost-utility

analysis for a 5-year and remaining lifetimetime horizon;

� Simple individual cost effectiveness and costanalyses during trial time period.

� Population budget impact estimates for a3-year time horizon.In the review, we compare the results from the

different types of economic analyses for a single

country as well as across countries. The extent towhich consistent estimates are given is noted andthe reasons for differences highlighted. The valueof the different methods used to estimate the im-pact of DRV/r on healthcare costs and health out-comes to different decision-makers is also described.

1.1 Economic Evaluation for Darunavir

When a new drug, such as darunavir, for thetreatment of HIV infection is introduced to themarket, it is critical to understand the impact thatit will have on both healthcare costs and healthconsequences for PLHIV from multiple perspec-tives. Many healthcare decision-makers will re-view the new drug and make decisions about itsapproval for the formulary, its reimbursementand its use in clinical practice. They will also es-timate their budgets with and without the newdrug and determine how to obtain any additionalfunding that might be needed. All of thesedecision-makers need information on the costsand benefits of the new drug as well as the clinicaloutcomes, but they need it in different formatsdepending on their decision needs. For example,regulatory agencies in the UK, Canada andAustralia require a state-of-the art technologyassessment from a societal perspective for eachnew drug before decisions can be made aboutreimbursement.[28-30] These technology assess-ments should follow current good practice guide-lines,[31] and should include extensive sensitivityanalysis so that they can readily understand theuncertainty in the estimates and the value ofobtaining new information. On the other hand,local health authorities or managed care organi-sations in the USA might prefer to have theeconomic information presented from a payerperspective and in a simpler format so that theimpact on their population is clear and the cost-effectiveness and budget impact of the new drugcan be easily understood.

Various types of economic analysis have beenperformed to investigate the cost-effectivenessand individual cost impact and population bud-getary impact of DRV/r 600/100mg twice a day(bid) for treatment-experienced PLHIV, based onthe clinical efficacy data gathered with this dose

Assessing the Economic Profile of Darunavir 3


in the POWER and TITAN trials. These analysesinclude incremental cost-utility analyses, analysesof cost effectiveness compared with no treatment,incremental cost-effectiveness analyses usingHIV-specific efficacy endpoints, individual cost anal-yses and population budget-impact analyses.

1.2 Cost-Utility Analyses of Darunavir withRitonavir at a Dose of 600/100mg Twice a Day

To meet the needs of regulatory agencies for astate-of-the-art cost-effectiveness model, a Mar-kov treatment pathway model was developed toestimate the cost effectiveness of darunavir over arange of time horizons including a lifetime timehorizon.[12] A series of health states were definedusing CD4 cell count ranges, see figure 1. PLHIVwere assumed to enter the model with the char-acteristics of those included in the darunavirclinical trials, and to transition through thehealth states at a rate based on the efficacy ofinitial treatment received in the model as well ason the efficacy of the one switch regimen thatthey were assumed to start following failure ofthe initial regimen and to stay on until death, seefigure 2. The model allowed for disease progres-sion to vary according to the virological responseto treatment at 24 weeks and according to thetreatment regimen. The model estimated the im-pact of uncertainty about the input parametervalues as well as the impact of variability inpractice patterns, population characteristics andmodel assumptions. For the uncertainty analysis,both an extensive one-way analysis was per-formed to illustrate the importance of individualparameter values on the results, as well as aprobabilistic sensitivity analysis (PSA) that dem-onstrated the impact of the results of all thecombined uncertainty for all the parameter val-ues. The model was first developed using US in-put parameter values and assumptions. The inputparameter values were derived either from thedarunavir clinical trials or the published litera-ture on other trials or observational data studiesor from national statistics.

Estimates of the cost-effectiveness of daruna-vir were derived for multiple countries using thesame modelling framework as for the USA esti-

mates, but with country-specific input parametervalues.[25] These countries included Canada, theUK, Sweden, Italy and Belgium. To perform eachcountry-specific adaptation the following stepswere taken. First the model was reviewed and alist of country-specific data items required wasprepared. These items included country-specificvalues for drug costs and other costs in each CD4cell count range, utility values in each health state,switch and other practice patterns, general popu-lation mortality data and demographic assump-tions for the population likely to be treatedwith darunavir. Second, the model structure wasreviewed with local country clinical experts toensure that it was credible and to request their as-sistance in identifying the needed country-specificparameter values. Finally, the country-specificdata were collected and entered into the model toderive the country-specific estimates.

DeathCD4

201−350cells/mm3

CD4101−200cells/mm3

CD451−100

cells/mm3

CD40−50

cells/mm3

CD4>500

cells/mm3

CD4351−500cells/mm3

Fig. 1. Markov model health states.

4 Mauskopf et al.


Two estimates were obtained for each countrybased on different clinical trials as follows:� Incremental cost-utility of DRV/r plus an opti-

mized background regimen (OBR) of nucleo-side reverse transcriptase inhibitors with orwithout enfuvirtide compared with an investi-gator-selected regimen of control PIs plusOBR in highly treatment-experienced PLHIVwho have failed treatment with at least two PIsusing pooled clinical data from the POWER 1and POWER 2 trials.

� Incremental cost-utility of DRV/r plus OBRcompared with lopinavir boosted with ritona-vir (LPV/r) plus OBR in treatment-experiencedPLHIV with a broader degree of previous PIexposure/failure, using data from the subset ofparticipants in the TITAN trial who had atleast one International AIDS Society USA pri-mary PI resistance-associatedmutation (RAM)at baseline.The results of the analysis of DRV/r compared

with an investigator-selected control PI regimenusing the pooled efficacy data from the POWER1 and 2 trials for the USA[12] and the fourEuropean countries[25] over a lifetime horizon arepresented in table I.

In each of these analyses, treatment withDRV/r plus an OBR is estimated to increase thelifetime costs of care due to both the higher costof DRV/r compared with the other PIs and the

incremental costs resulting from the increasedlife expectancy with DRV/r-based therapy com-pared with control PIs. The predicted increase inQALYs is higher than the increase in life-years inall countries with DRV/r-based therapy becausetreatment with DRV/r results in a lower propor-tion of the patient’s remaining lifetime spent inthe lower CD4 cell count ranges in which utility islow.

For all of the countries, the estimated cost-utility ratio is well below typical benchmarkvalues (US$50 000/QALY in the USA,[32] d20 000–d30 000/QALY in the UK,[33] h30 000/QALY incontinental Europe).[34,35] Moreover, these ratiosare robust, as demonstrated by both one-way sen-sitivity and variability analyses and multi-way.[12,25,26] For example, the PSAs estimate thatthe probability of a cost-utility ratio for DRV/rcompared with control PIs of less than US$50000(for the USA), less than d30 000 (for the UK)and less than h30000 (for the three continentalEuropean countries) over a lifetime horizon is95.0%, 93.5% and at least 94%, respectively.

Table II presents the results of the economicanalyses of DRV/r plus an OBR compared toLPV/r plus an OBR using efficacy data fromthe subset of participants in the TITAN trialwho had at least one International AIDS SocietyUSA primary PI RAM at baseline in the samecountries.[13,26]

First regimen Switch regimen

Treatmentfailure

andswitch

Darunavir/r

Death

Tipranavir/r

Tipranavir 500 mg bid+

Ritonavir 200 mg bid+

OBR

Darunavir 600 mg bid+

Ritonavir 100 mg bid+

OBR

Control

Control PIs based onresistance testing

+Ritonavir if chosen

+OBR

Fig. 2. Markov model treatment pathways. NOTE: The OBR was selected based on resistance testing. In the model, the OBR includednucleoside reverse transcriptase inhibitors and the fusion inhibitor enfuvirtide. Treatment failure was defined as either detectable HIV-1-RNAlevel or a decline in CD4 cell count, or both. The switch may occur before or several time periods after the start of the CD4 cell count decline.OBR =optimized background regimen; PIs =protease inhibitors.



With the exception of the efficacy-related in-puts, which were directly derived from this sub-population of TITAN study participants, mostinput parameter values and model assumptionsused in the POWER-based models were kept thesame in this adaptation. Switching to the tipra-navir boosted with ritonavir-based follow-up re-gimen was, however, assumed to occur sooner inPLHIV with a less than 1 log10 decrease in theHIV-RNA level from baseline at 24 weeks andbefore the CD4 cell count begins to decline forall virological response groups. This was doneto reflect the likely more proactive attitude totreatment modification in a routine treatmentsetting in TITAN-like PLHIV who generally hada lesser degree of previous treatment exposure,less advanced disease characteristics and moreadditional treatment options than the POWERPLHIV.[1,36-38]

In comparison with the results of the POWER-based models, DRV/r-based HAART is esti-mated to be associated with lower incrementalcosts and lower increases in quality-adjusted sur-vival when compared with LPV/r-based therapyin the modelled TITAN subgroup over a lifetimehorizon. The cost-utility ratios for DRV/r in this

population in each of these countries are, never-theless, similar (or even somewhat lower) to thosepresented in Table I for the POWER-based anal-yses and well below the country or region-specificbenchmark values. Moreover, as in the POWER-based analyses, sensitivity and variability analysesconfirmed the robustness of these conclusions,although the outcome of the TITAN modelwas more sensitive than the POWER models tothe estimated changes in CD4 cell count, theduration of efficacy for those achieving a less than1 log10 decrease in the plasma HIV-RNA level at24 weeks and, in the UK and Italy, the time hor-izon. Reflecting this, the PSAs generally estimatedsomewhat lower probabilities (75.4% for the USAand at least 67% for the European countries) ofhaving a cost-utility ratio below US$50 000 orh30 000, respectively.[13,26]

It should be noted that, in comparison withthe PLHIV included in the POWER trials, theTITAN subpopulation modelled in this analysishad less advanced HIV disease and a more lim-ited degree of previous treatment experience(only 60% of them were three-class experiencedin contrast to the POWER participants, whowere all three-class experienced at baseline).[19,21]

Table I. Cost-utility analysis in five countries: POWER 1 and 2 trial data

USAa UKb Italya Swedena Belgiumc

Lifetime total cost

DRV/r US$565 358 £181492 h223200 h323113 h310258

Control US$527 287 £162019 h203680 h311188 h292673

Difference US$38 071 £19473 h19521 h11925 h17585

Life-years

DRV/r 11.12 8.99 9.67 9.54 10.87

Control 9.86 7.64 8.53 8.43 9.48

Difference 1.25 1.35 1.15 1.11 1.39

QALYs

DRV/r 10.03 8.04 8.69 8.57 9.74

Control 8.76 6.67 7.52 7.43 8.34

Difference 1.27 1.37 1.17 1.14 1.40

Cost/QALY gained US$30 046 (h19229) £14 187 (h18 052) h16668 h10442 h12585

Exchange rates, 2008 April 6: US$1 = h0.64; £1 = h1.27.

a Discount rate of 3% for costs and benefits.

b Discount rate of 3.5% for costs and benefits.

c Discount rate of 3% for costs and 1.5% for benefits.

DRV/r =darunavir boosted by low-dose ritonavir; POWER =Performance Of TMC114/r When evaluated in treatment-Experienced patients

with protein inhibitor Resistance; QALYs = quality-adjusted life-years.

6 Mauskopf et al.


More importantly, whereas all POWER partici-pants had received (and failed) at least two PI-containing HAART regimens before enrolment,more than 40% of the PI-resistant TITAN sub-group modelled in this analysis had only receivedup to one PI before enrolment in TITAN.

1.3 Cost-Effectiveness Analyses of Darunavirwith Ritonavir at a Dose of 600/100mgTwice a Day

In addition to using a measure of the incre-mental cost per QALY, the cost effectiveness ofHIV antiretroviral drugs can also be estimated asthe incremental cost per additional unit of a clin-ically relevant outcome. As, in HIV infection, thecurrent goal of HAART is the sustained suppres-sion of plasma HIV-1-RNA levels to less than50 copies/mL, this is a clinically relevant outcomemeasure. The incremental cost per additional per-son achieving sustained virological suppressionmay be a more useful measure for decision-makers who need economic estimates to helpthem make choices between drugs for a specificcondition and for whom clinically relevant end-

points important for making decisions about themanagement of a condition are more under-standable measures than QALYs.

Hill et al.[27] have calculated the mean annualantiretroviral drug cost per patient with an un-detectable plasma HIV-RNA level for HAARTregimens containing DRV/r 600/100mg bidcompared with HAART regimens containingeither investigator-selected control PIs or LPV/r400/100mg bid, using the 48-week results of thepooled POWER 1 and 2 trials across 11 differentEuropean healthcare settings, theUSAandCanada.These analyses for the different countries used dif-ferent drug regimen costs depending on the prices ofthe drugs in each country.

In the European analysis, themean annual drugcost per patient with a plasma HIV-RNA levelless than 50 copies/mL at week 48 was 75% (range73–76%) lower for the DRV/r-based therapy com-pared with control PI-containing HAART. Com-parable results were seen in the USA (78% lower)and Canada (76% lower).

Analyses of the mean annual drug cost per pa-tient achieving a plasma HIV-RNA level less than50 copies/mL may be useful for decision-makers

Table II. Cost-utility analysis in five countries: TITAN subpopulation with one or more International AIDS Society – USA primary protease

inhibitor mutation at baseline

USAa UKb Italya Swedena Belgiumc

Lifetime total cost

DRV/r US$593 024 h289 143 h227901 h322 881 h294822

LPV/r US$581 666 h279 134 h217199 h318 437 h288363

Difference US$11 358 h10 009 h10702 h4444 h6459

Life-years

DRV/r 13.496 11.232 11.889 11.660 13.819

LPV/r 13.035 10.718 11.316 11.112 13.068

Difference 0.461 0.513 0.572 0.548 0.751

QALYs

DRV/r 12.512 10.396 11.000 10.790 12.775

LPV/r 12.020 9.846 10.39 10.206 11.990

Difference 0.493 0.550 20.608 0.584 0.785

Cost/QALY gained US$23 057 (h14 756) h18 213 h17592 h7605 h7990

Exchange rate, 2008 Apr 6: US$1 = h0.64.

a Discount rate of 3% for costs and benefits.

b Discount rate of 3.5% for costs and benefits.

c Discount rate of 3% for costs and 1.5% for benefits.

DRV/r =darunavir/ritonavir; LPV/r = lopinavir/ritonavir; QALYs = quality-adjusted life-years; TITAN =TMC114/r In Treatment-experienced

pAtients Naive to lopinavir.



who manage fixed medication-related budgetssuch as the US AIDS Drug Assistance Program,since they may help these payers identify regimensthat provide virological suppression to the largestnumber of PLHIV given their finite financial re-sources. The results derived from the POWERtrial clinical efficacy data imply that with a fixedannual budget of US$10 million for antiretroviralagents for themanagement of treatment-experiencedindividuals with a profile similar to the POWERpatient population, it would be possible to achieve aplasmaHIV-RNA level of less than 50copies/mL in128 PLHIV with DRV/r-based regimens comparedwith only 29 PLHIV with control PI-based regi-mens. Similar ratios for other antiretroviral drugshave been computed.[39,40]

The ratio of mean annual drug cost and virolo-gical suppression rates, however, only enables thecomparison of an ART to no therapy. As such,and while they may provide relevant informationto some HIV (drug) budget holders about the re-turn on investment for alternative ART options,these measures are only indirectly related to thestandard measures of cost effectiveness.

Standard measures of cost effectiveness re-quire estimates of the ratio of the incrementalcost to the incremental clinical outcomes of a newintervention relative to an existing intervention.In cardiovascular and respiratory diseases, forexample, authors have previously reported thecost per cardiovascular event avoided,[41] whereasthe cost per additional pain-free day has beencalculated in the management of pain,[42] andthe cost per additional symptom-free day is acommon cost-effectiveness measure for asthmadrugs.[43]

A similar approach has been used to estimatethe incremental drug cost per additional patientwith an undetectable plasma HIV-RNA levelat 48 weeks for HAART regimens containingDRV/r 600/100mg bid compared with HAARTregimens containing either investigator-selectedcontrol PIs or LPV/r 400/100mg bid, using the48-week efficacy results of the pooled POWER 1and 2 trials.[19] The results of this analysis areshown in table III for several European countries,the USA and Canada for the overall populationin POWER.[27]

Whether or not DRV/r is cost effective relativeto the comparators using a measure of incremen-tal drug costs per additional patient with an un-detectable plasma HIV-RNA level at 48 weekshas to be left to the judgement of each individualdecision-maker, because aswithmany other disease-specific cost-effectiveness ratios, a benchmark valuefor ‘good’ value does not exist for this type ofratio. These results can, however, be compared withestimates of the incremental cost per additional pa-tient with an undetectable plasma HIV-RNA levelderived from the comparisons of different medica-tions from other controlled trials evaluating similartreatment populations as reported by Hill et al.[27]

to place these estimates in perspective. Therefore,for the TORO (T-20 vs Optimized Regimen Only) 1and 2 trials the incremental cost per additional pa-tient with virological suppression was d137740 forenfuvirtide versus PI boosted with ritonavir (PI/r)and for the RESIST (Randomized Evaluation ofStrategic Intervention in multidrug-resiStant pa-tients with Tipranavir) 1 and 2 trials the incrementalcost per additional patient with virological suppres-sion was d32176 for tipranavir versus PI/r. These

Table III. Incremental antiretroviral therapy cost per additional pa-

tient achieving viral suppression: pooled POWER 1 and 2 population

Country ART cost per additional patient with

viral suppression

Pooled POWER 1 and 2 population

Austria (h) 7308

Belgium (h) 10 686

France (h) 6537

Germany (h) 8694

Italy (h) 9691

The Netherlands (h) 8823

Poland (PLN) 48 163 (h13967)

Spain (h) 9446

Sweden (SEK) 70 069 (h7708)

Switzerland (CHF) 13 426 (h8458)

UK (£) 4663 (h5922)

USA (US$) 823 (h527)

Canada (Can$) 8311 (h3092)

Exchange rates, 2008 Apr 6: US$1 = h0.64; £1= h1.27; 1CHF =h0.63; 1SEK= h0.11; 1PLN = h0.29; Can$1 = h0.372.

ART = antiretroviral therapy; POWER =Performance Of TMC114/rWhen evaluated in treatment-Experienced patients with protease

inhibitor Resistance.

8 Mauskopf et al.


values are both much higher than the comparablevalue for the POWER 1 and 2 trials for the UK(d4663).

Incremental cost-effectiveness analyses usingthe cost per patient with an undetectable plasmaHIV-RNA level as the endpoint could also beused to determine whether a DRV/r-containingregimen lies on the efficiency frontier for the HIV-positive, treatment-experienced population, as de-fined in the recent German Institute for Qualityand Efficiency in Health Care guidelines.[44] Theproportion of those treated who achieve plasmaHIV-RNA levels less than 50 copies/mL at 1 yearis a well-accepted clinical measure for the relativebenefits of different treatment regimens in HIVinfection and would form the vertical axis. Esti-mates of the net costs with a DRV/r regimencompared with those with other regimens overa lifetime horizon – or, if warranted by the lim-itations of available data or the rapidly evolvingtreatment practices in HIV care, shorter, clinicallyrelevant time horizons (e.g. 1–5 years) – couldbe plotted on the horizontal axis. The results ofsuch an approach were recently presented by sev-eral authors in both treatment-experienced andtreatment-naive, HIV-infected individuals.[45-48]

1.3.1 Individual Cost Analyses of Darunavir withRitonavir at a Dose of 600/100mg Twice a Day

Cost analyses that provide information on theimpact of the new drug on total per-patient costs(drug costs plus other healthcare costs) over aclinically relevant time horizon provide a simpleand straightforward illustration of the impact ofa new drug on costs to budget holders anddecision-makers in addition to the cost-utility orcost-effectiveness analyses. Therefore, estimateswere also calculated of the direct costs of care perpatient for the first year after starting treatmentwith DRV/r-based HAART compared with com-bination therapy containing the investigator-selected control PIs used in the pooled POWER 1and 2 trials in PLHIV with a profile similar to thestudy population included in these trials.[49]

Efficacy data from the POWER trials wereused to estimate the percentage of time during thefirst year each patient in each treatment arm wasin each of a number of clinically relevant CD4 cell

count strata. The weighted, annual cost of careper patient, excluding antiretroviral medicationcosts, was then computed for both arms usingthese percentages and the estimates of the costs ofcare in each of these CD4 cell count strata, de-rived from the cost of care studies in each of thecountries included in the analyses (the USA,Sweden, Belgium, Italy and the UK). These costswere then added to the observed annual, meanantiretroviral drug cost per patient in each arm,yielding the total per-patient costs of care for thefirst year of treatment of each therapy.

The results of this analysis, shown in table IV,indicate that the expected non-antiretroviral drugcosts are lower during the first year of therapywithDRV/r-basedHAART comparedwith controlPI-containing combination therapies in all countries.Depending on the healthcare setting considered,these cost reductions either completely or partly off-set the increase in antiretroviral-related costs asso-ciated with the use of DRV/r instead of control PIs,resulting in lower anticipated total (antiretroviraldrug plus non-antiretroviral drug costs) costs per pa-tient year in the USA and Sweden (-7% and -8%,respectively), comparable costs in the UK andBelgium, and moderately higher costs of care (+5%)in Italy for DRV/r-based HAART compared withcontrol PI-containing regimens.[49]

A key assumption for this calculation was thatparticipants who dropped out of the study con-tinued with their original drug regimen and theirCD4 cell counts stayed at the drop-out level forthe rest of the year. Although this approach canbe justified based on the standard practices usedin the analysis of HIV clinical trials and the evo-lution of the CD4 cell count observed in highlypre-treated PLHIV failing therapy,[50] the resultsabove do not consider how premature discontin-uation of the randomized treatment and the useof salvage therapies impacted medication costs,nor their impact on non-antiretroviral-related costsof therapy. In addition, the conclusions about totalcosts only apply to the first year of therapy, al-though recently published 144-week efficacy resultsof the pooled POWER trials indicate that the CD4cell count increase from baseline observed withDRV/r-based HAART is maintained over the longterm.[17,20]



1.3.2 Population Budget Impact of Darunavir withRitonavir at a Dose of 600/100mg Twice a Day

Finally, population budget-impact analysesusing drug or total healthcare costs for the trea-ted population in a 1-year or multi-year time pe-riod as the summary measure are important fordecision-makers concerned about the afford-ability of new drugs for the population and whoneed estimates of the amount of new resourcesand funding needed (if any) if they provide re-imbursement for the new drug.

In a budget impact analysis based on the in-troduction of DRV/r to the French healthcareformulary, the budget impact was assumed todepend on four primary factors: the cost of aDRV/r regimen compared with currently used PIregimens; the effectiveness of DRV/r in increas-ing CD4 cell counts; the total number of highlytreatment-experienced PLHIV eligible for treat-ment with DRV/r covered under the Frenchsocial security system; and the proportion of eli-gible people who would be prescribed a DRV/rregimen rather than one of the current PI regi-mens (see figure 3).[16] Efficacy data were taken

from the POWER 1 and 2 clinical trials and thenumber of people eligible for treatment withDRV/r was estimated from French epidemiolo-gical data. The proportion of eligible people whowere treated with darunavir was assumed to in-crease over the first 3 years after its introductionto the French formulary from 20% in the firstyear to 70% in the third year.

The results of the budget impact analysis areshown in table V and indicate that adding DRV/rto the French formulary would probably result inhigher antiretroviral drug treatment costs, butlower total healthcare costs for treating HIV in-fection over the first 3 years when compared withthe use of other PI-containing regimens. This isbecause the increase in antiretroviral drug treat-ment costs is expected to be more than fully offsetby the reduction in non-antiretroviral drug-related direct costs as a result of the significantimmune reconstitution of the target populationwhen treated with DRV/r-basedHAART and theassociated increases in CD4 cell count levelsreducing the risk of developing costly AIDS-defining and non-AIDS-defining morbidity.[51-53]

Over longer time horizons, the annual healthcarecosts associated with the use of DRV/r in thetreated population are likely to increase as a re-sult of the anticipated favourable impact ofDRV/r on HIV-related mortality rates increasingthe number of PLHIV in the treated population.

It should be noted that the conclusions onthe net impact of darunavir-containing HAARTregimens on the overall costs of care derived fromthis budget impact analysis only apply to highlytreatment-experienced PLHIV with a profile sim-ilar to the POWER population. In the TITANtrial, which enrolled less treatment-experiencedPLHIV, the superior virological responses ob-served with HAART regimens containing DRV/r600/100mg bid, when compared with LPV/r,were not associated with a higher CD4 cell countincrease from baseline for the DRV/r treatmentafter 48 weeks of therapy.[21,24] This discordancebetween the virological and immunological resultsimplies that the impact of DRV/r 600/100mgbid on the non-antiretroviral drug-related costsof care observed in the analysis of the POWER 1and 2 data will not be seen in less treatment-

Table IV. First-year costs with darunavir with ritonavir compared

with investigator-selected protease inhibitors: results from the

POWER 1 and 2 trials

Country ART costs Non-ART costs Total costs

USA

DRV/r $US31 252 $US18 974 $US50 226

Control $US30 825 $US23 217 $US54 042

UK

DRV/r £18 429 £10 698 £29 127

Control £16 797 £12 785 £29 582

Sweden

DRV/r SEK246 935 SEK87199 SEK334134

Control SEK222 411 SEK140090 SEK362501

Belgium

DRV/r h28 205 h10 529 h38 734

Control h24 465 h14 263 h38 728

Italy

DRV/r h21 292 h4336 h25 628

Control h17 905 h6521 h24 426

ART = antiretroviral therapy; DRV/r =darunavir/ritonavir; PI = pro-tease inhibitor; POWER =Performance Of TMC114/r When eval-

uated in treatment-Experienced patients with protease inhibitor

Resistance.

10 Mauskopf et al.


experienced PLHIV, at least not in their first yearof therapy.

2. Discussion

The economic data summarized in this sup-plement confirm that DRV/r 600/100mg bid notonly has major clinical benefits, but also has a

favourable economic profile for the managementof treatment-experienced PLHIV, in particularthose PLHIV who have failed PI-based HAART.A variety of methodologies and techniques wereapplied in this supplement to assess the economicoutcomes of DRV/r 600/100mg bid in treatment-experienced PLHIV and included: both combin-ed healthcare costs (antiretroviral drug and other

ART mix inscenario without DRV/r

ART mix in scenariowhere DRV/r is increasingly usedStudy sample

(DRV/r target population)

Resources consumed perCD4 cell count range

Estimated budget impact ofDRV/r over the model time horizon

Costing• ART costs• Other treatment costs• Hospitalization costs

CD4 cell countdistribution

CD4 cell countdistribution

Overall costs ofscenario without DRV/rduring year 1, 2 and 3

Overall costs ofscenario with DRV/r

during year 1, 2 and 3

Fig. 3. Model schematic for budget impact model. ART = antiretroviral therapy; DRV/r = darunavir/ritonavir.

Table V. Results of the budget impact model: total healthcare expenditure associated with following the current treatment strategy and

following the darunavir treatment strategy over 3 years

Darunavir not available (1) Base-case

analysis

Year 1 Year 2 Year 3 3-Year period

ART costs 69 749 645 75020 612 81853 708 91212 618 248 086 938

Other HIV-related drugs 18 806 546 19674 933 20550 904 20952 827 61 178 665

Hospitalizations and follow-up costs 24 766 533 25909 118 27051 580 27565 984 80 526 682

Total costs 113 322 724 120604 663 129456 192 139731 429 389 792 285

Darunavir available (2)

ART costs 69 749 645 76486 786 84349 510 92890 739 253 727 035

Other HIV-related drugs 18 806 546 18579 948 17976 725 17263 692 53 820 365

Hospitalizations and follow-up costs 24 766 533 24453 889 23654 983 22705 269 70 814 141

Total costs 113 322 724 119520 623 125981 218 132859 700 378 361 542

Budget impact (1)–(2)

ART costs 1 466 173.55 2 495 801.63 1 678 121.47 5 640 096.64

Other HIV-related drugs -1094 985.29 -2574 179.03 -3689 135.22 -7 358 299.54

Hospitalizations and follow-up costs -1455 228.53 -3396 596.75 -4860 715.06 -9 712 540.33

Total costs -1084 040.27 -3474 974.15 -6871 728.81 -11 430 743.23ART = antiretroviral therapy; DRV/r = darunavir/ritonavir.



healthcare costs) as well as antiretroviral drug-only expenditures; and generic health outcomes(QALY) as well as disease-specific endpoints(plasma HIV-RNA level to <50 copies/mL). Inaddition, both short-term (1 year) and lifetimehorizons were considered. In performing such arange of economic evaluations, the researcherscontributing to this supplement have attemptedto present a comprehensive picture of the eco-nomic profile of DRV/r in treatment-experiencedPLHIV.

What is apparent from all the analyses pre-sented in this review paper is that, although theresults are consistent with each other, they varydepending on the time horizon taken. In particular,the analyses indicate that, in some patient sub-groups when compared with a mix of investigator-selected PIs, DRV/r may result in overall costsavings in the short run, with any additional anti-retroviral drug costs more than offset by non-antiretroviral cost savings. When taking a lifetimeperspective in these same patient subgroups andwith the same comparators, the use of DRV/r mayresult in increases in lifetime total treatment costsbecause of the increased life expectancy and, thus,increased duration of treatment required. Previousstudies have shown that the lifetime cost of treatingan HIV-infected individual has increased fromapproximately US$100 000 with a life expectancyof approximately 1–2 years after enteringHIV carein 1985 to US$618 900 undiscounted (US$385 200discounted at 3%) with a life expectancy of 24.2 yearsafter entering HIV care in 2004,[54] so such increasesin costs with increases in life expectancy are to be ex-pected. The increases in life expectancy and QALYsestimated in our analysis when treating with DRV/rprovide value to society and the individual throughproductivity and quality of life gains andmay justifythese increased costs. Benchmark values applied tocost per QALY ratios, which measure the cost perunit gain in health outcomes, are frequently used toassess the efficiency of a new intervention that in-creases life expectancy and QALYs. Our results alsoshow that taking a lifetime perspective, DRV/r iscost effective compared with LPV/r (h7605–18213/QALY gained) and compared with investigator-selected PIs (h10442–19229/QALY gained) usingcommonly accepted benchmarks for quality-adjusted

cost effectiveness.[32,33] These results are robust evenwhen tested in extensive sensitivity and variabilityanalyses.[12,13,25,26]

The comprehensive approach to economicevaluation for DRV/r taken in this supplementreflects our understanding that in the currentera healthcare funding – even for high-profile dis-eases such as HIV – is coming under increasedpressure and lifetime costs of treating HIV areincreasing.[3,55-57] A highly diverse group ofdecision-makers are involved in discussions aboutantiretroviral drug pricing, reimbursement andmarket access, and therefore need to understandthe economic impacts of a new antiretroviral drugtomake sound decisions. These economic impactsinclude the changes in medical care costs associ-ated with treatment with darunavir as well as thechanges in a surrogate clinical endpoint of virol-ogical suppression. Virological suppression hasbeen shown to be associated with increased lifeexpectancy,[1] attributable to increased time inhigher CD4 cell count ranges, resulting in anincrease in QALYs and other benefits such asincreased productivity not explicitly estimated inthe reviewed analyses. These decision-makersinclude large institutional payers such as the UKNational Health Service, public insurers, regio-nal authorities, managed care organizations andother private health insurers who manage budgetsand expenditures spanning multiple disease areas,organizations with a HIV-specific focus such asthe US AIDS Drug Assistance Program, hospitalbudget holders and pharmacists, HIV care providersand, last but not least, PLHIV and their repre-sentatives. All these parties have their own know-ledge, perspective, requirements and relevant metricswhen it comes to assessing the economic profile andimpact of antiretroviral therapeutic options.

For example, health technology appraisal bodies,such as the Scottish Medicines Consortium in theUK, the Canadian Agency for Drugs and Technol-ogies in Health or the Australian PharmaceuticalBenefits Advisory Committee, are concerned withall health conditions, and are interested in cost-utility analyses for the complete disease duration.On the other hand, some healthcare budget holders,such as the French sickness fund, might be moreinterested in an analysis of the likely budget impact

12 Mauskopf et al.


for their covered population over the next 1–3 years.Finally, HIV care providers and decision-makersconsidering placing a new antiretroviral drug in theformulary or setting up treatment guidelines forthose with HIV infection might find the results ofcost-effectiveness analyses using clinical endpointsmore helpful for comparing the economic profileof antiretroviral drugs and determining the effi-ciency frontier of ART in HIV-infected patient(sub)populations, as defined in the recent GermanInstitute for Quality and Efficiency in Health Careguidelines.[44]

One limitation of the economic analyses thatincluded the costs of non-antiretroviral drug-related care was the paucity of recent publiclyavailable information on the cost of treating andmanaging HIV at various levels of HIV-inducedimmune deficiency in many countries. As Levyet al.[3] point out elsewhere in this supplement,knowledge of the (direct and indirect) economicimpact of HIV and its current treatment is ur-gently required now by healthcare decision-makers. Although the interest in, quantity andquality of costing studies have improved in theHAART era,[58,59] individual HIV care providers,administrators, national and international scien-tific steering committees of HIV-infected cohortscan play a crucial role in further promoting thecollection of high quality and detailed data on allcomponents of resource use and costs of HIVcare to help to inform the economic analyses ofnew antiretroviral drugs and other interventions.Ultimately, the ideal strategy would be an inter-national, collaborative study with a commondesign, detailed and standardized clinical and eco-nomic data elements and methods of reporting.The success of such an approach has been shownrecently by the Data Collection on Adverse Eventsof Anti-HIV Drugs cohort. This internationaleffort of multiple HIV cohorts across the globe,driven by a partnership between regulatory au-thorities, cohort leaders and antiretroviral man-ufacturers, has provided valuable information onthe long-term tolerability of ART in HIV-infectedadults and has influenced clinical decision-makingand policies.[60,61] While the costs for such an effortwould be relatively high, enormous value wouldaccrue to societies and governments worldwide, in

particular if its commitment to performing highquality research would be linked to an open atti-tude towards data sharing with those interested instudying the economics of the prevention andmanagement of HIV.

Another limitation of the economic analysessummarized in this review paper is the use ofclinical efficacy data derived from randomized,controlled clinical trials with their associated ex-clusion criteria. The efficacy of the different drugregimens in the clinical trials might thus not befully representative of the effectiveness of theseregimens in a clinic setting where a more hetero-geneous population would be treated. As a resultof the seriousness of HIV infection, however, thedarunavir clinical trials all included active stand-ard of care comparators and these head-to-headdata were used for the economic evaluations ofDRV/r. The cost-effectiveness estimates present-ed in this paper are thus all compared with stan-dard of care regimens using direct head-to-headcomparison data, which are more valuable foruse in economic evaluations than the data fromplacebo-controlled trials.[19,21]

Finally, it should be stressed that the resultsand conclusions contained in this supplement re-flect the currently available clinical data on the useofDRV/r 600/100mgbid in treatment-experiencedPLHIV. The phase III darunavir developmentprogramme has also evaluated a once-daily dose ofDRV/r 800/100mg in treatment-naive[62] as well astreatment-experienced PLHIV without darunavir-specific RAMs (TMC114-TiDP31-C229; clinicaltrials.gov refNCT00524368), who represented 82%of the overall population in the TITAN trial.[63]

This lower dose of DRV/r has a number of attri-butes (lower pill burden, once-daily dosing fre-quency) that have been shown to impact adherenceand long-term treatment outcomes positively.[64-66]

Future studies will assess its economic profile in themanagement of HIV infection in treatment-naiveand treatment-experienced PLHIV.

3. Conclusions

Darunavir has been shown in the POWER andTITAN clinical trials to improve the clinical out-comes of treatment-experienced, HIV-infected



adults significantly, even those with extensiveprevious treatment exposure.

Given the increasing pressure on HIV budgetsand the wide variety of healthcare decision-makers involved in the management of the fund-ing of HIV care, it is desirable to demonstrate theeconomic profile and impact of new antiretro-viral drugs and other interventions in HIV careusing metrics and endpoints that are in line withthe product’s clinical attributes and relevant tothe different healthcare decision-makers.

Using such a comprehensive approach to eco-nomic evaluation, DRV/r 600/100mg bid has beenshown to have a favourable economic profile for themanagement ofHIV-infected, treatment-experiencedadults, in particular those who have failed at leastone of the other currently available PIs.

Further efforts are needed to better quantifythe costs of the management of HIV infectionand its determinants in different countries inorder tomeet the increasing demand for economicjustification of clinical decisions in the manage-ment of HIV-infected individuals.

Acknowledgements

The authors wish to thank Tony Vangeneugden, Ben VanBaelen, Els De Paepe, Alain Smits, Eric Lefebvre, SabrinaSpinosa-Guzman, Frank Tomaka, Frederic Godderis, Piet DeDoncker, Martine De Pauw and the rest of the darunavirstudy team at Tibotec Pharmaceuticals, Mechelen, Belgium,for their contributions in analyzing and generating the clinicaltrial data that supported the above-mentioned economicanalyses. Special thanks also go to all the co-authors ofthe papers included in the Darunavir PharmacoEconomicsSupplement. The authors also acknowledge CatherineElliott (medical writer, Gardiner-Caldwell Communications,Macclesfield, UK) for her editorial support. Last but not least,the authors wish to thank all the study investigators, thePLHIV and their families for their (past, current and future)participation and support in the (further) clinical developmentof darunavir. This project was financially supported byJohnson & Johnson Pharmaceutical Services.

JM has received grant support from Janssen Cilag, themanufacturer of darunavir, to assist with the preparation ofthis manuscript. JM was not restricted by Janssen Cilag in herinterpretation of the individual papers on which the reviewwas based. LA has received consulting fees from Johnson &Johnson. AH has received consultancy payments fromTibotec to work on the health economics of darunavir. ES isan employee of Johnson & Johnson Pharmaceutical Services,Beerse, Belgium, and owns stock options and shares in thiscompany.

References1. US Department of Health and Human Services, Panel on

Antiretroviral Guidelines for Adults and Adolescents(USDHHS). Guidelines for the use of antiretroviral agentsin HIV-1-infected adults and adolescents; 2008 Jan 29[online]. Available from URL: http://AIDSinfo.nih.gov[Accessed 2009 Dec 10]

2. Ghatnekar O, Hjortsberg C, Gisslen M, et al. Medicalresource utilization and cost of HIV-related care in thehighly active antiretroviral therapy era at a university clinicin Sweden. PharmacoEconomics 2010; 28 Suppl. 1: 49-57

3. Levy A, Johnston K, Annemans L, et al. The impact ofdisease stage on direct medical costs of HIVmanagement: areview of the international literature. PharmacoEconomics2010; 28 Suppl. 1: 35-47

4. Colin X, Lafuma A, Costagliola D, et al. Modelling thebudget impact of darunavir in the treatment of highlytreatment-experienced, HIV-infected adults in France.PharmacoEconomics 2010; 28 Suppl. 1: 183-97

5. Mauskopf JA, KitahataM, Kauf T, et al. HIV antiretroviraltreatment: early versus later. J Acquir Immune Defic Syndr2005; 39: 562-9

6. Freedberg KA, Losina E, Weinstein MC, et al. The costeffectiveness of combination antiretroviral therapy forHIV disease. N Engl J Med 2001; 344: 824-31

7. Schackman BR, Freedberg KA, Weinstein MC, et al. Cost-effectiveness implications of the timing of antiretroviraltherapy in HIV-infected adults. Arch Intern Med 2002;162: 2478-86

8. Sax PE, Losina E,WeinsteinMC, et al. Cost-effectiveness ofenfuvirtide in treatment-experienced patients with advancedHIV disease. J Acquir Immune Defic Syndr 2005; 39: 69-77

9. Hornberger J, Green J, Wintfeld N, et al. Cost-effectivenessof enfuvirtide for treatment-experienced patients with HIVin Italy. HIV Clin Trials 2005; 6: 92-102

10. Hornberger J, Kilby JM, Wintfeld N, et al. Cost-effectiveness of enfuvirtide in HIV therapy for treatment-experienced patients in the United States. AIDS Res HumRetroviruses 2006; 22: 240-7

11. Simpson KN, Chumney ECG, Hicks CB, et al. Cost effec-tiveness of tipranavir in treatment-experienced HIVpatients in the USA [poster PIN10]. 11th Annual Interna-tional Meeting of the International Society for Pharmaco-economics and Outcomes Research; 2006 May 20–24;Philadelphia, Pennsylvania, USA

12. Mauskopf J, Brogan A, Martin S, et al. Cost effectiveness ofdarunavir/ritonavir in highly treatment-experienced, HIV-1-infected adults in the USA. PharmacoEconomics 2010;28 Suppl. 1: 83-105

13. Brogan A,Mauskopf J, Talbird S, et al. US cost effectivenessof darunavir/ritonavir 600/100mg bid in treatment-experienced, HIV-infected adults with evidence of proteaseinhibitor resistance included in the TITAN trial. Pharmaco-Economics 2010; 28 Suppl. 1: 129-46

14. Mauskopf J. Meeting the NICE requirements: a Markovmodel approach. Value Health 2000; 3: 287-93

15. Mauskopf JA, Tolson JM, Simpson KN, et al. Impact ofzidovudine-based triple combination therapy on an AIDSdrug assistance program. J Acquir Immune Defic Syndr2000; 23: 302-13

14 Mauskopf et al.


16. Colin X, Lafuma A, Costagliola D, et al. The cost ofmanaging HIV infection in highly treatment-experienced,HIV-infected adults in France. PharmacoEconomics 2010;28 Suppl. 1: 59-68

17. Pozniak A, Jayaweera D, Hoy J, et al. Efficacy of darunavir/ritonavir in treatment-experienced HIV-1-infected patientsat 96 weeks in the POWER 1 and 2 trials [abstract P7.2/0].11th European AIDS Conference; 2007 October 24–27;Madrid, Spain

18. Molina JM, Cohen C, Katlama C, et al. Safety and efficacyof darunavir (TMC114) with low-dose ritonavir in treatment-experienced patients: 24 week results of POWER 3.J Acquir Immune Defic Syndr 2007; 46: 24-31

19. Clotet B, Bellos N, Molina JM, et al. Efficacy and safety ofdarunavir–ritonavir at week 48 in treatment-experiencedpatients with HIV-1 infection in POWER 1 and 2: a pooledsubgroup analysis of data from two randomised trials.Lancet 2007; 369 (9568): 1169-7

20. Katlama C, Bellos N, Grinsztejn B, et al. POWER 1 and 2:combined final 144-week efficacy and safety results fordarunavir/ritonavir 600/100mg bid in treatment-experiencedHIV patients [abstract]. 9th International Congress on DrugTherapy in HIV Infection; 2008 Nov 9–13; Glasgow, UK

21. Madruga JV, Berger D, McMurchie M, et al. Efficacyand safety of darunavir/ritonavir compared with that oflopinavir/ritonavir at 48 weeks in treatment-experienced,HIV infected patients in TITAN: a randomized controlledPhase III trial. Lancet 2007; 370: 49-58

22. Pozniak A, Arasteh K, Molina JM, et al. POWER 3analysis: 144-week efficacy and safety results for darunavir/ritonavir 600/100mg bid in treatment-experienced patients[abstract]. 9th International Congress on Drug Therapy inHIV Infection; 2008 Nov 9–13; Glasgow, UK

23. Arasteh K, Grinsztejn B, de Bethune MP, et al. Efficacyanalysis of darunavir/r in treatment-experienced POWER3 patients at Week 96 [abstract P7.2/05]. 11th EuropeanAIDS Conference; 2007 Oct 24–27; Madrid, Spain

24. Banhegyi D, Katlama C, da Cunha C, et al. Phase IIITITAN week 96 final analysis: efficacy/safety of darunavir/r(DRV/r) vs. lopinavir/r (LPV/r) in LPV-naıve, treatmentexperienced patients [abstract]. 9th International Congresson Drug Therapy in HIV Infection; 2008 Nov 9–13,Glasgow, UK

25. Moeremans K, Annemans L, Lothgren M, et al. Costeffectiveness of darunavir/ritonavir 600/100mg bid inprotease inhibitor-experienced, HIV-1-infected adults inBelgium, Italy, Sweden and the UK. PharmacoEconomics2010; 28 Suppl. 1: 107-28

26. Moeremans K, Hemmett L, Hjelmgren J, et al. Cost effective-ness of darunavir/ritonavir 600/100mg bid in treatment-experienced, lopinavir-naive, protease inhibitor-resistant,HIV-infected adults in Belgium, Italy, Sweden and the UK.PharmacoEconomics 2010; 28 Suppl. 1: 147-67

27. Hill AM, Clotet B, Johnson M, et al. Costs to achieve un-detectable HIV RNA with darunavir-containing highlyactive antiretroviral therapy in highly pretreated patients:the POWER experience. PharmacoEconomics 2010; 28Suppl. 1: 69-81

28. Canadian Agency for Drugs and Technologies in Health(CADTH). CADTH guidelines for the evaluation of healthtechnologies: Canada, 3rd ed. 2006. Available at: http://

www.cadth.ca/media/pdf/186_EconomicGuidelines_e.pdf[Accessed 2010 Jun 30]

29. National Institute for Health and Clinical Excellence(NICE). Single technology appraisal specification for man-ufacturer/sponsor submission of evidence. 2009. Availableat: http://www.nice.org.uk/media/E29/DF/SpecificationForManufacturerSponsorSubmissionofEvidenceDraft.pdf[Accessed 2010 Jun 30]

30. Pharmaceutical Benefits Advisory Committee (PBAC). Guide-lines for preparing submissions to the Pharmaceutical BenefitsAdvisory Committee, version 4.2. 2007. Available at: http://www.health.gov.au/internet/main/publishing.nsf/Content/pbacguidelines-index [Accessed 2010 Jun 30]

31. National Institute for Health and Clinical Excellence(NICE). Guide to the methods of technology appraisal;2008 June [online]. Available from URL: http://www.nice.org.uk/media/B52/A7/TAMethodsGuideUpdatedJune2008.pdf [Accessed 2009 Dec 10]

32. Hirth RA, Chernew ME, Miller E, et al. Willingness to payfor a quality-adjusted life year: in search of a standard.Med Decis Making 2000; 20: 332-42

33. Towse A. What is NICE’s threshold? An external view.Chapter 2. In: Devlin N, Towse A, editors. Cost effective-ness thresholds: economic and ethical issues. London:King’s Fund/Office for Health Economics, 2002

34. Sacristan JA, Oliva J, Del Llano J, et al. +Que es una tec-nologıa sanitaria eficiente en Espana? Gac Sanit 2002; 16:334-43

35. Chapman RH, Stone PW, Sandberg EA, et al. A compre-hensive league table of cost-utility ratios and a sub-table of‘‘panel-worthy’’ studies.MedDecisMaking 2000; 20: 451-67

36. BHIVA Writing Committee. British HIV Association(BHIVA). Guidelines for the treatment of HIV-infectedadults with antiretroviral therapy (2006). HIV Medicine2006; 7: 487-503

37. European AIDS Clinical Society (EACS). 2009 Guidelinesfor the clinical management and treatment of HIV-infected adults in Europe; [online]. Available from URL:http://www.europeanaidsclinicalsociety.org/guidelinespdf/EACS-EuroGuidelines2009FullVersion.pdf [Accessed 2009Dec 10]

38. International AIDS Society USA (IAS–USA). Treatmentfor adult HIV infection: 2006 recommendations of theInternational AIDS Society–USA panel. JAMA 2006; 296:827-43

39. Basu RP, Grimes RM, Helmy A. Cost to achieve an un-detectable viral load using recommended antiretroviral re-gimens. HIV Clin Trials 2006; 7: 309-18

40. Ruof J, Dusek A, DeSpirito M, et al. Cost-efficacy com-parison among three antiretroviral regimens in HIV-1 in-fected, treatment-experienced patients. Clin Drug Invest2007; 27: 469-79

41. Lamy A, Jonsson B, Weintraub WS, et al. The cost-effec-tiveness of the use of clopidogrel in acute coronary syn-dromes in five countries based upon the CURE study. EurJ Cardiovasc Prev Rehabil 2004; 11: 460-5

42. Kumar S, Williams AC, Sandy JR. How do we evaluate theeconomics of health care? Eur J Orthod 2006; 28: 513-19



43. Shih YC, Mauskopf J, Borker R. A cost-effectiveness anal-ysis of first-line controller therapies for persistent asthma.PharmacoEconomics 2007; 25: 577-90

44. IQWiG (Instutut fur Qualitat und Wirtschaftlichkeit imGesundheitswesen-Institute for Quality and Efficiency inHealth Care). Methods for assessment of the relation ofbenefits to costs in the German statutory health care sys-tem. Version 1, 24th January, 2008

45. Annemans L, Hill A, Smets E, et al. Cost-effectiveness ofcurrent treatment options in treatment-resistant HIV/AIDSpatients in the German setting [abstract PE19.5/5]. 12thEuropean AIDS Conference; 2009 Nov 11–14, Cologne,Germany

46. BroganA, Soorapanth S, Donatz V, et al. Comparative cost-efficacy analysis of darunavir/r and other ritonavir-boostedprotease inhibitors for first-line treatment of HIV-1 infec-tion in Germany [abstract PIN48]. 12th Annual Meeting ofthe International Society for Pharmacoeconomics andOutcomes Research; 2009 October 24–27; Paris, France

47. Soorapanth S, Brogan A, Mrus J, et al. Comparative cost-efficacy analysis of darunavir/r for first-line treatment ofHIVinfection in theUnited States. Value inHealth 2009; 12: A112

48. Smets E, Stoll M, Hill A, et al. Comparative cost-efficacyanalysis of darunavir/r and other ritonavir-boosted pro-tease inhibitors for first-line treatment of HIV-1 infectionin Germany [abstract PE19.5/3]. 12th European AIDSConference; 2009 Nov 11–14, Cologne, Germany

49. Hill AM, Gebo K, Hemmett L, et al. Predicting direct costsof HIV care during the first year of darunavir-based highlyactive antiretroviral therapy using CD4 cell counts: evi-dence fromPOWER. PharmacoEconomics 2010; 28 Suppl. 1:169-81

50. Ledergerber B, Lundgren JD, Walker AS, et al. Predictors oftrend in CD4-positive T-cell count and mortality amongHIV-1-infected individualswith virological failure to all threeantiretroviral-drug classes. Lancet 2004; 364 (9428): 51-62

51. Phillips AN, Lundgren JD. The CD4 lymphocyte count andrisk of clinical progression: the T cell in HIV infection anddisease. Curr Opin HIV AIDS 2006; 1: 43-9

52. Holkmann Olsen C, Gatell J, Ledergerber B, et al. Risk ofAIDS and death at given HIV-RNA and CD4 cell counts,in relation to specific antiretroviral drugs in the regimen.AIDS 2005; 19: 319-30

53. Weber R, Friis-Moller N, Sabin C, et al. HIV and non-HIV-related deaths and their relationship to immunodeficiency:the D:A:D study [abstract 595]. 12th Conference on Ret-roviruses and Opportunistic Infections; 2005 Feb 22–25;Boston, MA, USA

54. Schackman BR, Gebo KA, Walensky RP, et al. The lifetimecost of current human immunodeficiency virus care in theUnited States. Med Care 2006; 44: 990-7

55. Health Protection Agency (HPA) Center for Infections.A complex picture. HIV and other sexually transmittedinfections in the United Kingdom; 2006 [online]. Availablefrom URL: http://www.hpa.org.uk/publications/2006/hiv_sti_2006/ default.htm [Accessed 2009 Dec 10]

56. England R. Are we spending too much on HIV? BMJ 2007;334: 344

57. de Lay P, Greener R, Izazola JA. Are we spending too muchon HIV? BMJ 2007; 334: 345

58. Harling G, Wood R, Beck EJ. Efficiency of interventions inHIV infection, 1994–2004. Dis Manage Health Outcomes2005; 13: 371-94

59. BeckEJ,MinersAH, TolleyK. The cost ofHIV treatment andcare. A global review. PharmacoEconomics 2001; 19: 13-39

60. Friis-Moller N, Reiss P, Sabin CA, et al. Class of antiretro-viral drugs and the risk of myocardial infarction. New EnglJ Med 2007; 356: 1723-35

61. Sabin CA, Worm SW, Weber R, et al. Use of nucleoside re-verse transcriptase inhibitors and risk of myocardial infarc-tion in HIV-infected patients enrolled in the D:A:D study:a multi-cohort collaboration. Lancet 2008; 371: 1417-26

62. DeJesus E, Ortiz R, Khanlou H, et al. Efficacy and safetyof darunavir/ritonavir versus lopinavir/ritonavir in ARVtreatment-naıve HIV-1-infected patients at week 48:ARTEMIS [abstract H-718b]. 47th Annual InterscienceConference on Antimicrobial Agents and Chemotherapy;2007 Sept 17–20; Chicago, IL, USA

63. Berger DS, Northland R, Scribner A, et al. Effect of baselinefactors on virological response to darunavir/r andlopinavir/r at week 48 in TITAN [abstract P7.3/27]. 11thEuropean AIDS Conference; 2007 Oct 24–27; Madrid,Spain

64. Ammassari A, Trotta MP, Murri R, et al. Correlates andpredictors of adherence to highly active antiretroviraltherapy: overview of published literature. J Acquir Im-mune Defic Syndr 2002; 31 Suppl.: S123-7

65. Maggiolo F, Ripamonti D, Gregis G, et al. Once-a-daytherapy for HIV infection: a controlled, randomized studyin antiretroviral-naıve, HIV-1-infected patients. AntiviralTher 2003; 8: 339-46

66. Negredo E, Bonjoch A, Clotet B. Benefits and concerns ofsimplification strategies in HIV-infected patients. J Anti-microb Chemother 2006; 58: 235-42

Correspondence: Dr Josephine Mauskopf, RTI Health Solu-tions, Research Triangle Park, 3040 Cornwallis Road, PostOffice Box 12194, Research Triangle Park, NC 27709-2194,USA.E-mail: [email protected]

16 Mauskopf et al.


a review of economic evaluations of darunavir boosted by low-dose ritonavir in treatment-experienced...

Documents