Budget impact analysis of darbepoetin alfa every 3 weeks versus epoetin alfa every week for the treatment of chemotherapy-induced anaemia from a US payer's perspective

Abstract

Objective: This analysis was conducted to compare the direct medical costs of treatment with darbepoetin alfa every 3 weeks (Q3W) and epoetin alfa every week (QW) in patients with chemotherapy-induced anaemia (CIA) from the payer's perspective.

Methods: An analysis was conducted from a US health plan perspective to compare the annual budget impact for CIA with darbepoetin alfa Q3W and epoetin alfa QW over a 16-week treatment period. Dosing regimens were obtained from registration clinical trials.

Results: Mean doses, including dose adjustments, were 375.6 μg Q3W for darbepoetin alfa and 43,187 U QW for epoetin alfa. Costs of medical resources included drug acquisition and administration costs. The base case analysis resulted in a per-patient budget impact of $8,544 and $8,667 for darbepoetin alfa and epoetin alfa, respectively. Per member per month cost was $0.90 for darbepoetin alfa and $0.91 for epoetin alfa, based on an estimate of 2,735 CIA patients in a health plan population of 2.17 million. The analysis was most sensitive to drug dose, treatment period and drug price.

Conclusions: Results suggest that per-patient direct medical costs of CIA treatment, when initiated at labelled starting doses, are comparable for darbepoetin alfa Q3W and epoetin alfa QW.

Keywords::

• darbepoetin alfa
• budget impact
• epoetin alfa
• anaemia

Introduction

Anaemia is a common condition among patients with cancer receiving chemotherapy and can cause fatigue, nausea, dyspnoea, anorexia and cardiac dysfunction1–3. Anaemia-related fatigue is significantly associated with depression, anxiety and loss of productivity4–6. Despite the significant negative impact on patients’ health-related quality of life, typically only two-thirds of patients with chemotherapy-induced anaemia (CIA) receive treatment2,5

In addition to increased cost to the patient, anaemia is also associated with increased costs to the healthcare system7,8. A US administrative claims analysis covering year 2000 data for 2.3 million adult health plan members found an anaemia prevalence of 3.5% among all patients in the health plan and 21.4% among those with cancer, with 86.5% of patients not receiving any treatment for their anaemia8. The average costs for patients with anaemia were more than twice those of their non-anaemic counterparts, even after accounting for age, sex, other co-morbidities and plan type8. The benefits of anaemia treatment are well established and the results of the analysis indicate that not treating anaemia is associated with considerable costs to payers8

Darbepoetin alfa and epoetin alfa are erythropoiesis-stimulating agents (ESAs) that are approved by the US Food and Drug Administration (FDA) and recommended by the National Comprehensive Cancer Network guidelines9 for the treatment of anaemia in patients with non-myeloid malignancies, where the anaemia is due to the effect of concomitantly administered chemotherapy in the oncology setting. The two proteins have the same basic structure but are glycosylated differently. Darbepoetin alfa carries two additional sialic acid residues, which increases the half-life of the modified protein approximately three-fold10–12. The two molecules are available in different doses and both have been shown to be safe and efficacious13–15

The recent approval of darbepoetin alfa for an every 3 weeks (Q3W) regimen in the treatment of CIA has resulted in a need for decision makers to draw on economic comparisons for formulary decision making. Although the most common regimens for darbepoetin alfa and epoetin alfa are every 2 weeks (Q2W) and every week (QW) dosing, respectively16, darbepoetin alfa administered Q3W may be advantageous as it allows for synchronisation of patients’ anaemia treatment with ongoing cancer treatments, which may have implications not only for patients and providers but also for health plans. Synchronisation of treatments may reduce the number of visits and blood tests required for patients, reduce treatment-related opportunity costs for oncology clinics, and potentially reduce costs for health insurers without compromising patient care. Haematological outcomes have been shown to be comparable for ESAs administered synchronously and asynchronously with chemotherapy17,18

This study was designed to estimate the annual budget impact of ESA treatment associated with treating cancer patients receiving chemotherapy with either Aranesp® (darbepoetin alfa)19 Q3W or Procrit®^† (epoetin alfa)20 QW from a US health plan perspective. The analysis used data from randomised registration trials as no head-to-head trials are available comparing these dosing regimens.

Patients and methods

Methods

A budget impact analysis was developed to evaluate the economic impact of ESA therapy from a US health plan perspective in patients with non-myeloid malignancies and CIA. The analysis focused on the estimation of direct medical costs that are relevant to a third party payer and compared the total direct medical costs of ESA treatments based on drug acquisition costs and drug administration costs (therapeutic injections and separately identifiable physician office visits). The analysis was designed to estimate the budget impact on a health plan in the US and therefore focused on the estimation of direct medical costs that are relevant to a third party payer.

Data sources

ESA dose and administration

This analysis compared a Q3W dose and administration schedule of darbepoetin alfa with a QW dose and administration schedule of epoetin alfa over a 16-week time horizon19–22. The ESA regimens compared in this analysis were taken from registration studies referenced in the package inserts and the published literature, taking into account dose adjustment due to escalation, reduction or withholding19–23. The total dose for each ESA was calculated by multiplying the mean ESA dose per injection by the total number of injections over the time horizon.

The mean dose for darbepoetin alfa was obtained from an active-control registration trial comparing darbepoetin alfa initiated at 500 μg Q3W and darbepoetin alfa 2.25 μg/kg QW21. The mean weekly dose reported by Canon et al21 for the Q3W regimen was 125.2 μg, corresponding to a mean dose of 375.6 μg per Q3W scheduled administration21. The mean dose per scheduled administration for epoetin alfa (43,187 U) was obtained from a registration trial22, as reported by Cremieux et al24. Dosing regimens of darbepoetin alfa and epoetin alfa were subject to dose adjustment based on haemoglobin response.

Time horizon

It was assumed that all patients would receive 15 weeks of ESA therapy and 1 week of observation over the course of 1 year21. The 16-week time horizon was based on randomised clinical trials that have evaluated the use of both darbepoetin alfa and epoetin alfa in CIA patients with similar ESA time horizons15,17,18–21,24–26. Discounting was not incorporated into the analysis because the time horizon in this study was less than 1 year.

Costs

Drug costs were based on the reported 2007 average wholesale price (AWP) from the September 2007 edition of the Drug Topics RED BOOK™27. The AWP of darbepoetin alfa was $5.491/μg and the AWP of epoetin alfa was $15.024/1,000 U. Since the AWP may not necessarily reflect the actual purchase price, and AWP minus 20% is a common basis for reimbursement to payers, AWP minus 20% was used for the base case analysis28

The analysis assumed that all patients require one injection per administration and one physician office visit (level two) per administration. To determine the cost of an office visit, the appropriate procedural codes for each type of visit were identified using Current Procedural Terminology (CPT) codes and were assigned costs for each code based on Medicare payment levels using the 2007 Medicare physician fee schedule29,30. Using Medicare payment levels as a proxy for health plan costs is considered an acceptable costing method in the US29. The cost of a physician office visit for basic evaluation and management (CPT code 99212) was $38.66 and the cost of an injection (CPT code 90772) was $20.0930. The per-visit cost was calculated to be equal to the sum of the physician office visit cost and the injection cost ($58.75). The 99212 office visit code was considered separately identifiable from the 90772 injection code, following recent guidance from the Centers for Medicare and Medicaid Services (CMS)31. Total administration costs were calculated by multiplying the per-visit administration cost by the frequency of visits over the time horizon of the analysis.

Budget impact

To estimate the annual budget impact for a typical health plan, the per-patient budget impact for the 16-week time horizon was determined and multiplied by the estimated number of CIA patients receiving ESA therapy within the plan. Specific assumptions for the following variables were made to estimate the size of the CIA population within a typical health plan: total enrollees; percentage of patients with non-myeloid malignancies; percentage of cancer patients receiving chemotherapy; percentage of chemotherapy patients with anaemia (CIA patients); and percentage of CIA patients receiving ESA therapy.

The total number of enrollees in a typical health plan (2,170,243) was estimated on the basis of the average enrolment for five major US health plans32. To estimate the number of patients in the plan with non-myeloid malignancies (26,043), the prevalence of non-myeloid malignancies in the general US population (1.2%) was multiplied by the total number of enrollees in the plan33. The number of non-myeloid cancer patients was multiplied by the percentage receiving chemotherapy (31.3%) to estimate the number of non-myeloid cancer patients in a typical health plan receiving chemotherapy (8,151)34. The number of CIA patients (4,483) was estimated by multiplying the number of chemotherapy patients by the percentage with anaemia (55%)3. Finally, the number of CIA patients was multiplied by the percentage receiving ESA treatment (61%) to determine the total number of CIA patients receiving ESA therapy (2,735)35. The per member per month budget impact was calculated by dividing the annual budget impact by 12 and dividing the resultant value by the total number of enrollees in the health plan.

Sensitivity analyses

One-way sensitivity analyses were performed on four variables: unit ESA drug cost; mean ESA dose per injection; time horizon; and visit costs (unit cost and frequency of visits). The ESA drug price was varied from the base case value (AWP minus 20%) up to the AWP and down to the average sales price (ASP) plus 6%, in accordance with the 2007 payment allowance limits (fourth quarter) set by CMS, which were $3.08/μg for darbepoetin alfa and $9.60/1,000 U for epoetin alfa36

Mean ESA dose per injection was varied from the minimum to the maximum mean ESA dose per injection, as reported in the literature or calculated from the data presented in the studies15,20–22,24–26,37–39. The minimum mean dose per injection for epoetin alfa was 38,044 U QW, calculated from the data presented in Waltzman et al25, and the maximum mean dose per injection was 46,307 U QW, calculated from the data presented in Witzig et al26. The minimum mean dose per injection for darbepoetin alfa Q3W was varied to 283.8 μg Q3W based on an efficacy study of the Q3W 200-μg regimen38 (representing a 24.4% reduction in dose from the base case), and the maximum mean dose per injection was varied to 467.2 μg Q3W, representing a 24.4% increase from the base case21. The time horizon for receiving ESA therapy was varied between 12 and 24 weeks. For the time horizon sensitivity analysis, it was assumed that the mean dose per injection reported by Canon et al in a 16-week study21 was a valid approximation of the dose required over a 12- or 24-week treatment period (i.e. 11 and 23 weeks of ESA treatment with 1 week of observation, respectively). The cost of a physician visit was varied by adding the cost of a complete blood count panel (CPT 85027 = $9.04) and by removing the cost of an evaluation and management visit (CPT 99212 = $38.66). The frequency of physician office visits was varied by requiring a weekly physician office visit, irrespective of the ESA injection schedule.

Confidence intervals for total costs

Bootstrap and Monte Carlo methods were used to calculate 95% confidence intervals (CIs) for point estimates obtained in the base case analysis. The bootstrapping procedure was done with replication using patient-level data. Monte Carlo simulation was done using @Risk Version 4.5 software add-in to Microsoft Excel 2003. These calculations were based on the distribution of dose, since dose was identified as the single most influential cost driver.

The average per-patient dose per Q3W administration was calculated using patient level data from the study by Canon et al21. A dose distribution for darbepoetin alfa Q3W was obtained by bootstrapping 1,000 iterations of 353 patients with replacement (the sample size for the Q3W darbepoetin alfa arm in the trial by Canon et al21).

Data from five epoetin alfa studies were used to calculate CIs15,22,25,26,40 using two different approaches. The first approach was more conservative and weighted the mean dose for each epoetin alfa trial by sample size. The second approach assumed a uniform dose distribution for epoetin alfa (ranging between the lowest dose of 38,044 U calculated from the trial by Waltzman et al25 to 46,307 U, a calculated dose based on the trial by Witzig et al26). Monte Carlo simulation was applied to the dose distribution of both drugs for 5,000 patients to obtain 95% CIs for total costs.

Results

Budget impact

Based on the 16-week per-patient direct medical costs for darbepoetin alfa Q3W (mean dose per injection of 375.6 μg) and epoetin alfa QW (mean dose per injection of 43,187 U), the per-patient impact was $8,544 and $8,667, respectively. The per member per month costs were $0.90 and $0.91 for darbepoetin alfa and epoetin alfa, respectively. The annual budget impact for CIA patients treated with either 100% darbepoetin alfa Q3W or 100% epoetin alfa QW was $23,367,840 and $23,704,245, respectively.

The per-patient budget impacts of darbepoetin alfa Q3W and epoetin alfa QW were similar, with total 16-week per-patient direct medical costs of $8,544 and $8,667, respectively. The CIs calculated on the basis of an epoetin alfa dose distribution weighted by sample size were $8,229–$8,669 for darbepoetin alfa and $8,087–$8,934 for epoetin alfa; when based on a uniform dose distribution for epoetin alfa (ranging between 38,044 and 46,307 U), the CIs were $8,230–$8,920 for darbepoetin alfa and $7,818–$9,157 for epoetin alfa. Results are summarised in Table 1. Total per-patient estimated drug acquisition costs for darbepoetin alfa and epoetin alfa were $8,250 and $7,786, respectively, using AWP minus 20%. Total per-patient visit costs for darbepoetin alfa and epoetin alfa were $294 and $881, respectively. The lower visit costs for darbepoetin alfa were due to fewer ESA injections over the 16-week time horizon (5 injections for darbepoetin alfa and 15 injections for epoetin alfa) and lower physician office visit costs for darbepoetin alfa.

Table 1.  Base case analysis*: darbepoetin alfa Q3W vs. epoetin alfa QW.

	Darbepoetin alfa Q3W	Epoetin alfa QW
Drug costs ($)	8,250	7,786
Administration costs ($)^†	294	881
Total costs ($)	8,544	8,667

Q3W, every 3 weeks; QW, every week; AWP, average wholesale price.

* Base case assumes that all patients will receive 15 weeks of erythropoiesis-stimulating agent therapy during a 16-week time horizon (5 injections for darbepoetin alfa Q3W and 15 injections for epoetin alfa QW). The base case mean dose per injection for darbepoetin alfa was 375.6 μg (Canon et al21), whilst for epoetin alfa it was 43,187 U (Cremieux et al22). AWP from the Drug Topics RED BOOK 2007 (September)27. AWP minus 20% may not necessarily reflect the actual purchase price but is a common basis for reimbursement by payers. Darbepoetin alfa was estimated at $4.3928/μg and epoetin alfa at $12.0192/1,000 U.

^† Drug Administration Unit Costs ($58.75): base case analysis assumes all patients will require one physician office visit (level 2) per administration ($38.66) and one injection per administration ($20.09).

Sensitivity analyses

Sensitivity analyses showed that the per-patient budget impact over the 16-week time horizon was most sensitive to changes in assumptions of drug dose (mean dose per scheduled injection), time horizon and drug price, and least sensitive to the cost of an office visit (Table 2). For example, when the mean dose per injection for darbepoetin alfa patients decreased from 375.6 to 283.8 μg (as reported in a study by Taylor et al38), the total per-patient incremental budget impact (difference) over the 16-week time horizon increased from $123 to $2,140, favouring darbepoetin alfa. Similarly, when the mean dose per injection for epoetin alfa patients decreased from 43,187 to 38,044 U (as calculated from the data presented in a study by Waltzman et al25), the total per-patient incremental impact (difference) over the 16-week time horizon favoured epoetin alfa (incremental cost of $804). When the duration of the treatment cycle was increased from 16 to 24 weeks, the total per-patient incremental impact increased from $123 to $1,329, favouring darbepoetin alfa. When assumptions about the drug price methodology were changed from AWP minus 20% to ASP plus 6%, the total per-patient incremental impact increased from $123 to $1,014, favouring darbepoetin alfa. For drug administration, when the unit cost of an office visit was changed from $58.75 to $67.79 (due to adding a complete blood count panel at each visit), the total per-patient incremental impact increased from $123 to $214, favouring darbepoetin alfa. Figure 1 shows the impact range of each sensitivity analysis variable on the total incremental budget impact per patient.

Table 2.  One-way sensitivity analyses: total per-patient budget impact of treatment.

	DA Q3W total impact ($)	EA QW total impact ($)
Drug price
Base case: AWP – 20% (DA $4.3928/μg; EA $12.0192/1,000 U)	8,544	8,667
AWP (DA $5.491/μg; EA $15.024/1,000 U)	10,606	10,614
ASP + 6% (DA $3.0846/μg; $9.6015/1,000 U)	6,087	7,101
Mean dose per injection
Base case: DA Q3W mean dose 375.6 μg; EA QW mean dose 43,187 U	8,544	8,667
DA Q3W mean dose 283.8 μg*; EA QW mean dose 43,187 U	6,527	8,667
DA Q3W mean dose 467.2 μg*; EA QW mean dose 43,187 U	10,555	8,667
DA Q3W mean dose 375.6 μg; EA QW mean dose 38,044 U^†	8,544	7,740
DA Q3W mean dose 375.6 μg; EA QW mean dose 46,307 U^‡,$	8,544	9,230
Visit costs
Base case: visit cost = $58.75 (CPT 99212 [$38.66] + CPT 90772 [$20.09])	8,544	8,667
Visit cost + CBC panel = $67.79 ($58.75 + CPT 85027 [$9.04])	8,589	8,803
Visit cost for injection only = $20.09 (CPT 90772 only)	8,350	8,087
Frequency of administration visit
Base case (DA Q3W; EA QW)	8,544	8,667
DA weekly office visit; EA weekly visit	8,930	8,667
Time horizon
Base case (16 weeks)	8,544	8,667
Time horizon 12 weeks	5,126	6,356
Time horizon 24 weeks	11,961	13,290

DA, darbepoetin alfa; EA, epoetin alfa; Q3W, every 3 weeks; QW, every week; AWP, average wholesale price; ASP, average sales price; CPT, current procedural terminology codes; CBC, complete blood count.

* The minimum mean dose per injection for darbepoetin alfa Q3W was 283.8 μg Q3W, based on an efficacy study of the Q3W 200 μg regimen38 (representing a 24.4% reduction in dose from the base case), and the maximum mean dose per injection was varied to 467.2 μg Q3W, representing a 24.4% increase from the base case21.

^† Mean dose per injection was calculated based on values reported in Waltzman et al25 using the following equation: mean weekly dose = mean cumulative dose/mean duration of treatment.

^‡ Mean dose per injection was calculated based on values reported in Witzig et al26. A weighted average dose per injection was calculated based on dose (i.e. starting, escalated and reduced dose), number of injections administered and the number of patients on each regimen.

^$ It was assumed that during a 16-week duration of treatment, a patient on a QW regimen will receive 15 injections.

Figure 1.  Sensitivity analyses incremental per-patient budget impact: darbepoetin alfa every 3 weeks compared with epoetin alfa every week.

Discussion

This analysis suggests that the budget impact of darbepoetin alfa Q3W and epoetin alfa QW are comparable in a health plan population with non-myeloid malignancies and CIA. The per-patient budget impact of CIA treatment for 16 weeks, using AWP minus 20% to estimate drug price, was found to be $8,544 (95% CI $8,229–$8,669) for darbepoetin alfa Q3W and $8,667 (95% CI $8,087–$8,934) for epoetin alfa QW, based on simulations using a dose distribution for epoetin alfa weighted by sample size. These results remained consistent when a uniform dose distribution for epoetin alfa (ranging between 38,044 to 46,307 U) was used to calculate CIs.

One-way sensitivity analyses showed that the results were most sensitive to changes in drug dose, time horizon and drug price, least sensitive to the cost of an office visit and moderately sensitive to the frequency of physician office visits. These findings are consistent with results from other clinical trial data and analyses based on claims data41–43. However, they are not consistent with the results of a meta-analysis estimating cost effectiveness carried out by Rosberg et al, which employed a dose-conversion ratio as an outcome measure37. It is important to note that a dose-conversion ratio has uncertain clinical utility and is not considered an acceptable endpoint in clinical trials by the FDA.

In this analysis, all patients were assumed to require only one physician office visit (level two) and one injection per administration, with the number of visits equal to the number of administrations. However, the frequency of office visits may vary in clinical practice, with more frequent office visits for darbepoetin alfa than one visit every 3 weeks. Changing the frequency of office visits for darbepoetin alfa Q3W to one physician office visit every week, irrespective of ESA administration schedule, resulted in an increase of $386 for darbepoetin alfa administration costs; overall costs for darbepoetin alfa and epoetin alfa were $8,930 and $8,667, respectively, favouring epoetin alfa by a small margin.

These findings are consistent with an economic evaluation by Reed et al44 of an open-label randomised trial25 where patients received either darbepoetin alfa Q2W using a 200-μg injection or epoetin alfa QW. Reed et al found that over a duration of 16 weeks there were no significant differences in total costs, although total costs were higher by $900 for patients receiving epoetin alfa than for patients receiving darbepoetin alfa44. The evaluation also found that the single largest component of total costs was the cost of ESAs and their administration. When a sensitivity analysis was conducted on the method of reimbursement, four out of five methods used resulted in lower drug cost point estimates for darbepoetin alfa than for epoetin alfa. In addition, when analytical perspectives were varied, the costs of darbepoetin alfa Q2W and epoetin QW were not significantly different; however, epoetin alfa tended to be more costly. Six different methods were used to assign costs to study medications and injection time administration. Epoetin alfa was found to be more expensive in all cases except when costs were based on the Veteran's Administration supply schedule44. The benefits of extended dosing with darbepoetin alfa (e.g. less frequent physician office visits) may not have been fully captured, as patients were required to return to the clinic every week for blood tests as part of the clinical trial44. A more complete accounting of the benefits of the darbepoetin alfa extended dosing regimen may further increase the estimated cost difference between the two regimens.

Similarly, the budget impact analysis results for this study may not fully capture the economic benefits of Q3W dosing. In the analysis, costs for complete blood counts were assumed to be similar between the two products, a conservative assumption as an extended darbepoetin alfa dosing regimen may potentially reduce the need for weekly complete blood counts. In addition, extended dosing regimens allow for synchronisation with chemotherapy treatment; this is likely to reduce the number of required clinic visits, which may also be of economic benefit to payers, patients and caregivers, their employers and healthcare providers45,46. Benefits may also include less time spent by the patient and provider for drug administration and treatment; the average patient receiving any epoetin alfa regimen requires 17.6 hours more treatment time to complete a course of treatment than patients receiving any darbepoetin alfa regimen, as reported by Houts et al47. Indirect benefits, such as member satisfaction with extended dosing, are also not captured. One study reported 72% of patients receiving epoetin alfa QW preferred to switch to a less frequent dosing regimen48

To date, few economic comparisons of darbepoetin alfa and epoetin alfa have been conducted and, to the authors’ knowledge, only one study has been based on a head-to-head trial44. Economic comparisons that are available have a few notable limitations, such as not fully accounting for factors that should be included in the analysis based on the analytical perspective. These factors include the omission of indirect costs, such as the costs associated with treatment-related travel, in an analysis from the societal perspective37, relying on meta-analyses based on incomplete evidence49, and not addressing the effect of differences in treatment regimens50. Therefore, there is an ongoing need for direct head-to-head comparisons to include a comprehensive analysis of budget impact of ESA treatments.

Finally, it is important to note the recent safety concerns regarding the use of ESAs. Revisions to labelling include changes to the US FDA label for ESAs (November 2007) and a recent announcement from the European Agency for the Evaluation of Medicinal Products (October 2007) regarding upcoming changes to product information on the ESA labelling in Europe. The availability of this new information is intended to provide updated information on the risks and benefits of ESAs and to facilitate a full benefit–risk discussion between the prescribing physician and patient. In July 2007, a National Coverage Decision, published by CMS51, restricted coverage of ESAs in Medicare patients. Given that any potential actions will likely affect the entire class of ESAs, the authors anticipate that future changes will impact both darbepoetin alfa and epoetin alfa similarly.

Limitations

There are several limitations to this study that warrant further discussion. The present analysis was carried out from a US health plan perspective, with budget impact based on drug acquisition and drug administration costs. Indirect costs (e.g. lost productivity and travel costs) were not included because the analysis was based on a comparison of direct medical costs from the perspective of a third party payer48. This analysis did not include medical costs related to the management of treatment-related or non-treatment-related adverse events, blood transfusions, laboratory tests, iron supplementation, potential improvement in long-term clinical outcomes or hospitalisations. However, this is not expected to affect the results of the budget impact analysis because these cost components may be impacted similarly by both medications. Additional research is needed to quantify the impact of these particular variables on overall costs. As there are no head-to-head clinical trials comparing these dosing regimens, mean doses were taken from registration trials for both products. In addition, the annual budget impact was estimated based on the assumption that a 16-week time horizon, in which all patients received 15 weeks of treatment and 1 week of observation, reflects the drug acquisition and drug administration costs to the health plan on an annual basis. This assumption is based on the 15-week treatment period in the Canon et al21 study. Base case parameter estimates were varied in the sensitivity analyses and the results were found to be plausibly sensitive to a few key variables.

Conclusion

The per-patient budget impact of CIA treatment for 16 weeks was comparable from a US health plan perspective for darbepoetin alfa administered Q3W and epoetin alfa administered QW. The potential advantages to patients and caregivers of extended dosing regimens of darbepoetin alfa should be further investigated. Comparative trials between QW and Q3W dosing are needed to gain further insight on dosing equivalence. In addition, as new and more efficient ESA dosing regimens are adopted by oncology clinics, their potential impact on the cost of treatment should be calculated.

Acknowledgements

Declaration of interest: Research support was provided by Amgen Inc. The authors would like to thank Niall Harrison and Kirsten Bunch of Gardiner-Caldwell London (Maidenhead, UK) for writing and editorial assistance; Arie Barlev at Amgen Inc. for providing expertise on economic analysis techniques; Peter J Neumann of the Centre for the Evaluation of Value and Risk in Health, Tufts-New England Medical Centre, for comments on an earlier draft; and Jason Scharf of Covance Inc. (San Diego, CA) for economic research and analytical assistance.

Notes

* Aranesp® is a registered trademark of Amgen, Inc.

† Procrit® is a registered trademark of Ortho Biotech Products, LP.