How does darbepoetin compare to epoetin in cancer patients?

Comment by InpharmD Researcher

Studies providing a direct clinical comparison between epoetin and darbepoetin in cancer patients are relatively dated, all having been conducted prior to 2010. Results of these studies generally reflect comparable efficacy and safety outcomes between the two erythropoiesis-stimulating agents. Subsequently, relevant guidelines do not specify a preference for either agent. Selection of agent may be based on clinician preference, cost/availability, and preferred dosing regimen.

darbepoetin vs epoetin oncology cancer malignancy

Background

Across the 2007, 2010, and 2019 American Society of Clinical Oncology (ASCO)/American Society of Hematology (ASH) guidelines, epoetin and darbepoetin are considered equivalent in effectiveness and safety, with no clinical preference for either agent. The 2007 guideline found no clinically significant differences in hematologic response, transfusion rates, or thromboembolic events; evidence was insufficient to establish differences in quality of life, tumor outcomes, survival, or other adverse effects. The 2010 update retained this conclusion because no new comparative studies had emerged, and the 2019 update again found similar efficacy and safety in subgroup analyses; although one analysis suggested greater fatigue improvement with epoetin, the panel considered the finding potentially confounded and did not change its equivalence determination. [1], [2], [3]

Both agents are subject to the same clinical restrictions and class risks, including increased thromboembolism. Discussion in the 2007 guideline permitted use of either agent when hemoglobin was approaching or below 10 g/dL, whereas the 2010 guideline narrowed initiation to hemoglobin below 10 g/dL and emphasized using the lowest hemoglobin level sufficient to avoid transfusion because of evidence of shorter survival and thromboembolic risk. The 2019 guideline more explicitly limited either agent to selected patients with chemotherapy-associated anemia, hemoglobin below 10 g/dL, and noncurative treatment intent, while recommending against their use with curative-intent chemotherapy and in most patients not receiving chemotherapy, except qualifying patients with lower-risk myelodysplastic syndromes. The primary difference between the two medications is dosing regimens, as epoetin may be administered at 150 U/kg three times weekly or 40,000 U weekly, whereas darbepoetin may be administered at 2.25 μg/kg weekly or 500 μg every three weeks. [1], [2], [3]

Background References: [1] Bohlius J, Bohlke K, Castelli R, et al. Management of Cancer-Associated Anemia With Erythropoiesis-Stimulating Agents: ASCO/ASH Clinical Practice Guideline Update. J Clin Oncol. 2019;37(15):1336-1351. doi:10.1200/JCO.18.02142
[2] Rizzo JD, Brouwers M, Hurley P, Seidenfeld J, Somerfield MR, Temin S. American society of clinical oncology/american society of hematology clinical practice guideline update on the use of epoetin and darbepoetin in adult patients with cancer. J Oncol Pract. 2010;6(6):317-320. doi:10.1200/JOP.2010.000132
[3] Rizzo JD, Somerfield MR, Hagerty KL, et al. Use of epoetin and darbepoetin in patients with cancer: 2007 American Society of Clinical Oncology/American Society of Hematology clinical practice guideline update. J Clin Oncol. 2008;26(1):132-149. doi:10.1200/JCO.2007.14.3396
Literature Review

A search of the published medical literature revealed 5 studies investigating the researchable question:

How does darbepoetin compare to epoetin in cancer patients?

Level of evidence

B - One high-quality study or multiple studies with limitations  Read more→



Please see Tables 1-5 for your response.


 

Randomized Comparison of Every-2-Week Darbepoetin Alfa and Weekly Epoetin Alfa for the Treatment of Chemotherapy-Induced Anemia: The 20030125 Study Group Trial
Design

Randomized, open-label, active-controlled, multicenter study

N= 1,220

Objective To systematically compare the efficacy and safety of darbepoetin alfa (DA) and epoetin alfa (EA) using common doses and schedules used in clinical practice
Study Groups

DA 200 µg every two weeks (n= 606)

EA 40,000 units every week (n= 603)

Inclusion Criteria Patients with nonmyeloid malignancy, ≥ 8 weeks of planned chemotherapy, age ≥ 18 years, and anemia (hemoglobin ≤ 11 g/dL)
Exclusion Criteria History of primary hematologic disorders causing anemia other than nonmyeloid malignancy; unstable/uncontrolled cardiac conditions; clinically significant inflammatory disease; neutralizing antibodies to recombinant human erythropoietin; EA or DA therapy within 4 weeks before randomization
Methods Patients were randomly assigned to DA 200 µg every two weeks or EA 40,000 units every week for up to 16 weeks with identical dose adjustment rules. Efficacy was assessed by the incidence of RBC transfusion. Dose escalation was permitted at week 5 if hemoglobin increase was ≤ 1 g/dL. Study drug was withheld if hemoglobin exceeded 13 g/dL and reinstated at 75% of the previous dose after hemoglobin decreased to ≤ 12 g/dL
Duration 16-week treatment period with 2 weeks of follow-up for adverse events
Outcome Measures

Primary: Incidence of RBC transfusion from week 5 to the end of the treatment phase

Secondary: Transfusion requirements over the entire treatment period, hemoglobin levels

Baseline Characteristics   Darbepoetin Alfa (n= 606) Epoetin Alfa (n= 603)
Female 415 (68%) 381 (63%)
White 506 (83%) 502 (83%)
Age, years  63.2 ± 12.4 63.7 ± 11.6

Primary tumor type

Breast

Gastrointestinal

Gynecologic

Lung

 

131 (22%)

104 (17%)

60 (10%)

156 (26%)

 

126 (21%)

108 (18%)

41 (7%)

164 (27%)

ECOG 0 to 1 518 (85%) 509 (84%)
Baseline hemoglobin, g/dL 10.2 ± 0.9 10.2 ± 0.9
Results   Darbepoetin Alfa (n= 606) Epoetin Alfa (n= 603) p-value
RBC transfusion incidence from week 5 to EOTP 21% 16% <0.05
Units transfused 2.3 (95% CI 2.1 to 2.5) 2.5 (95% CI 2.2 to 2.7) -
Mean hemoglobin change from baseline to EOTP 11.8 g/dL 11.8 g/dL NS
Patients achieving target hemoglobin (11-13 g/dL) 80% 86% NS
Adverse Events No significant differences in safety profiles between DA and EA. Cardiovascular/thromboembolic events reported in 6% of DA patients and 7% of EA patients. No antibodies to erythropoietic compounds detected.
Study Author Conclusions This large, phase III study demonstrates comparable efficacy of DA Q2W and EA QW. Less frequent dosing offers potential benefits for patients, caregivers, and healthcare providers.
Critique The study was well-designed with a large sample size and robust statistical analysis, providing strong evidence for the noninferiority of DA compared to EA. However, the open-label design may introduce bias, and the exclusion of the first month in the primary endpoint analysis could overlook early differences in treatment response. Additionally, the study did not collect long-term survival and tumor progression data, which could be relevant for comprehensive safety evaluation.
Table 1 References:
[4] Glaspy J, Vadhan-Raj S, et al. Randomized comparison of every-2-week darbepoetin alfa and weekly epoetin alfa for the treatment of chemotherapy-induced anemia: the 20030125 Study Group Trial. J Clin Oncol. 2006;24(15):2290-2297. doi:10.1200/JCO.2005.03.8570

 

Randomized Comparison of Epoetin Alfa (40,000 U Weekly) and Darbepoetin Alfa (200 μg Every 2 Weeks) in Anemic Patients with Cancer Receiving Chemotherapy
Design

Randomized, open-label, multicenter trial

N= 358

Objective To directly compare the hemoglobin (Hb) response to epoetin alfa (EPO), 40,000 U once weekly, with that to darbepoetin alfa (DARB), 200 μg every 2 weeks, in anemic patients with cancer receiving chemotherapy
Study Groups

EPO 40,000 U QW (n = 178)

DARB 200 μg Q2W (n = 180)

Inclusion Criteria Adults with solid tumors scheduled to receive chemotherapy for ≥12 weeks and with baseline Hb ≤11 g/dl
Exclusion Criteria Received any erythropoietic agent within 3 months, anemia due to factors other than cancer or chemotherapy, more than two prior chemotherapy regimens, radiation therapy included in treatment plans, history of stem cell or bone marrow transplant, untreated brain metastases, uncontrolled hypertension, poorly controlled seizures, known hypersensitivity to mammalian cell–derived products or human albumin, transfusion of packed red blood cells within 28 days, any unstable medical condition, or a history of uncontrolled cardiac arrhythmias, pulmonary embolism, or thrombosis within the previous 6 months
Methods

Patients were randomized to receive either EPO 40,000 U QW or DARB 200 μg Q2W s.c. for up to 16 weeks. Doses were increased for nonresponders after 4 (EPO) or 6 (DARB) weeks, and reduced for rapid Hb rise or Hb level >13 g/dl.

An oral supplement of ferrous sulfate, 325 mg, was taken daily if tolerated. Hb, Hct, and blood pressure were monitored weekly. 

QOL was assessed using specific subscales. Safety was assessed by laboratory tests, vital sign measurements, physical examinations, and the incidence and severity of AEs

Duration Up to 16 weeks
Outcome Measures

Primary: Proportion of patients achieving a ≥1-g/dl Hb rise by week 5

Secondary: Proportion of patients achieving at least a 1- or 2-g/dl increase in Hb at week 9 or at the study end, time to response, transfusion requirements

Baseline Characteristics   Epoetin alfa (n = 178) Darbepoetin alfa (n = 180)
Age, years 62.1 ± 11.8 63.4 ± 11.8
Female 109 (61.2%) 119 (66.1%)
Weight, kg 72.8 ± 18.2 74.8 ± 19.6
Hemoglobin, g/dL 10.16 ± 0.749 10.07 ± 0.787
Transferrin saturation, % 20.23 ± 12.81 23.77 ± 19.76

ECOG performance status score

0

1

2

 

57 (32.0%)

95 (53.4%)

26 (14.6%)

 

61 (33.9%)

89 (49.4%)

30 (16.7%)

Chemotherapy type

Platinum based

Nonplatinum based

 

69 (38.8%)

109 (61.2%)

 

76 (42.2%)

104 (57.8%)

Tumor type

Lung

Breast

 

47 (26.4%)

51 (28.7%)

 

47 (26.1%)

38 (21.1%)

Results   Epoetin Alfa (n= 178) Darbepoetin Alfa (n= 180) p-value
Hb increase ≥1 g/dl by week 5 47.0% 32.5% 0.0078
Hb increase ≥2 g/dl by study end 57.7% 41.8% 0.004
Median time to Hb rise ≥1 g/dl, days 35.0 46.0 0.0057
Mean PRBC units transfused per patient 2.5 3.9 0.0334
Adverse Events

The incidences and patterns of adverse events were similar between groups, with most being unrelated to the study treatment. Common adverse events included neutropenia (17% vs. 26%), nausea (33% vs. 27%), and diarrhea (26% vs. 20%).

Clinically significant thrombotic vascular events were reported in 11% of epoetin alfa patients and 9% of darbepoetin alfa patients.

Study Author Conclusions

Epoetin alfa demonstrated an earlier hematologic response compared to darbepoetin alfa in anemic patients with cancer receiving chemotherapy. Transfusion intensity was significantly lower for epoetin alfa, although transfusion frequency was not. Both treatments provided similar improvements in quality of life and had similar safety profiles.

Critique

The study was well-designed as a prospective, randomized trial, providing a direct comparison of epoetin alfa and darbepoetin alfa. However, the open-label design may introduce bias, and the study did not explore long-term outcomes beyond the 16-week treatment period. Additionally, the study focused on patients with solid tumors, which may limit the applicability of the findings to other cancer types.

Table 2 References:
[5] Waltzman R, Croot C, Justice GR, Fesen MR, Charu V, Williams D. Randomized comparison of epoetin alfa (40,000 U weekly) and darbepoetin alfa (200 microg every 2 weeks) in anemic patients with cancer receiving chemotherapy. Oncologist. 2005;10(8):642-650. doi:10.1634/theoncologist.10-8-642

 

Treatment of Chemotherapy-Induced Anemia in Breast Cancer: Results of a Randomized Controlled Trial of Darbepoetin Alfa 200 μg Every 2 Weeks Versus Epoetin Alfa 40,000 U Weekly
Design

Phase II, open-label, randomized, multicenter trial

N= 141

Objective To quantify the efficacy of darbepoetin alfa compared with epoetin alfa in patients with breast cancer and to describe the safety profile of darbepoetin alfa compared with epoetin alfa
Study Groups

Darbepoetin alfa (n= 72)

Epoetin alfa (n= 69)

Inclusion Criteria Female patients with anemia (Hb level < 11 g/dL) and histologically confirmed breast cancer, age ≥18 years, anemia induced by chemotherapy, ≥8 planned additional cycles of chemotherapy, Karnofsky performance status ≥50%, and adequate liver and renal function
Exclusion Criteria History of anemia caused by factors unrelated to chemotherapy, uncontrolled hypertension, active bleeding, red blood cell transfusion within 4 weeks of screening, erythropoietic therapy within 2 weeks of randomization, history of red cell aplasia, or known hypersensitivity to recombinant mammalian red cell growth factor product
Methods Patients were randomized to receive darbepoetin alfa 200 μg every 2 weeks or epoetin alfa 40,000 U weekly for ≤16 weeks. Hemoglobin levels, transfusions, and adverse events were assessed every 2 weeks. Dose modifications were made to maintain appropriate Hb levels. Supplemental iron therapy was permitted according to institutional standards
Duration October 2002 to December 2003
Outcome Measures

Primary: Mean change in hemoglobin (Hb) during treatment, mean Hb level at the end of treatment, proportion of patients exhibiting a hematopoietic response

Secondary: Ability to achieve and maintain a target Hb range of 11-13 g/dL, safety profile, patient satisfaction with anemia treatment

Baseline Characteristics   Darbepoetin Alfa (n = 72) Epoetin Alfa (n = 69)
Female 72 (100%) 69 (100%)
Age, years 53.6 ± 11.4 58.4 ± 12.5
White 58 (81%) 57 (83%)

Disease Stage

I/II

III

IV

 

37 (51%)

11 (15%)

22 (31%)

 

26 (38%)

13 (19%)

26 (38%)

Previous Radiation Therapy 9 (13%) 10 (14%)

Previous Chemotherapy

≥ 1 Previous regimen

> 2 Previous regimens

 

23 (32%)

8 (11%)

 

32 (46%)

15 (22%)

Mean Hemoglobin Level, g/dL 10.5 ± 0.8 10.6 ± 0.7
Mean Serum Ferritin, ng/mL 336.3 ± 319.8 290 ± 260.3
Mean Transferrin Saturation, % 27.32 ± 17.11 29.28 ± 21.82
Iron Supplementation 72 (100%) 69 (100%)
Results   Darbepoetin Alfa (n = 72) Epoetin Alfa (n = 69)
Mean change in Hb level from baseline, g/dL 1.9 ± 1.5 1.7 ± 1.4
Hematopoietic response (≥2 g/dL increase or Hb ≥12 g/dL) 88% (95% CI, 80%-95%) 81% (95% CI, 72%-90%)
Proportion of patients receiving transfusion 6% (95% CI, 0-11%) 16% (95% CI, 7%-25%)
Patients achieving target Hb level ≥11 g/dL 93% 90%
Median time to target Hb level, weeks 3 (95% CI, 3-5) 4 (95% CI, 3-5)
No significant differences observed between groups. 
Adverse Events 11 patients in the darbepoetin alfa group (15%) and 16 in the epoetin alfa group (24%) had ≥1 serious adverse event. No thrombotic events were reported for either group during treatment.
Study Author Conclusions Darbepoetin alfa 200 μg every 2 weeks and epoetin alfa 40,000 U weekly achieve similar hematologic and transfusion-based outcomes as well as patient satisfaction with erythropoietic treatment in patients with breast cancer.
Critique The study provides valuable comparative data on the efficacy and safety of darbepoetin alfa and epoetin alfa in breast cancer patients with chemotherapy-induced anemia. However, the lack of formal hypothesis testing limits the ability to draw definitive conclusions about noninferiority. The study's generalizability to other tumor types needs confirmation in larger trials. Despite these limitations, the study's findings are consistent with previous research on erythropoietic agents.
Table 3 References:
[6] Senecal FM, Yee L, Gabrail N, et al. Treatment of chemotherapy-induced anemia in breast cancer: results of a randomized controlled trial of darbepoetin alfa 200 microg every 2 weeks versus epoetin alfa 40,000 U weekly. Clin Breast Cancer. 2005;6(5):446-454. doi:10.3816/cbc.2005.n.050

 

A Randomized Comparison of Every-2-Week Darbepoetin Alfa and Weekly Epoetin Alfa for the Treatment of Chemotherapy-Induced Anemia in Patients With Breast, Lung, or Gynecologic Cancer
Design

Three concurrent randomized, open-label, multicenter, identical trials

N= 312

Objective To validate the Patient Satisfaction Questionnaire for Anemia (PSQ-An) treatment tool and to compare the efficacies and safety profiles of darbepoetin alfa and epoetin alfa in patients with chemotherapy-induced anemia
Study Groups

Darbepoetin alfa (n= 157)

Epoetin alfa (n= 155)

Inclusion Criteria Patients with a diagnosis of breast cancer, NSCLC (stage IIIb or IV), or gynecologic carcinoma; ≥18 years old; anemic (hemoglobin ≤11 g/dl at screening); planned concurrent chemotherapy for at least 8 additional weeks; KPS score ≥50%; adequate renal and liver function; ability to complete questionnaires
Exclusion Criteria Received an RBC transfusion within 4 weeks of screening or erythropoietic therapy within 2 weeks of randomization; inadequate iron stores; known positive antibody response to any erythropoietic agent; history of pure red cell aplasia, uncontrolled hypertension, or anemia due to a hematologic disorder other than chemotherapy-induced anemia
Methods Patients were randomized 1:1 to receive either darbepoetin alfa at 200 µg Q2W or epoetin alfa at 40,000 U QW for up to 16 weeks. Hemoglobin concentrations were measured at baseline and Q2W. The PSQ-An was administered during treatment and at the end of the study. Doses were adjusted based on hemoglobin levels.
Duration October 2002 to December 2003
Outcome Measures

Hemoglobin- and transfusion-based efficacy endpoints, safety profiles

Baseline Characteristics   Darbepoetin alfa (n= 157) Epoetin alfa (n= 155)
Female 134 (85%) 129 (83%)
Caucasian 132 (84%) 131 (85%)
Age, years 58.7 ± 11.5 61.7 ± 12.1

Tumor type

Breast

NSCLC (IIIb/IV)

Gynecologic

 

72 (46%)

51 (32%)

34 (22%)

 

69 (45%)

51 (33%)

35 (23%)

Results   Darbepoetin alfa (n= 157) Epoetin alfa (n= 155)
Patients achieving target hemoglobin ≥11 g/dl 124 (82%) 128 (86%)
Median time to target hemoglobin (weeks) 5 4
Mean hemoglobin level after achieving target (g/dl) 12.1 12.2
No significant differences observed between groups.
Adverse Events One episode of deep venous thrombosis in each treatment group; one episode of pulmonary embolism in the epoetin alfa group.
Study Author Conclusions Darbepoetin alfa (200 µg Q2W) and epoetin alfa (40,000 U QW) appear to achieve comparable clinical and hematologic outcomes. The PSQ-An enables formal comparisons of the impact of anemia therapies on patients and caregivers in future studies.
Critique The study provided valuable data on the comparability of darbepoetin alfa and epoetin alfa, but was not powered to test noninferiority. The primary focus was on validating the PSQ-An, not on comparing patient satisfaction between treatment groups. Future studies should address these limitations and explore patient satisfaction more comprehensively.
Table 4 References:
[7] Schwartzberg LS, Yee LK, Senecal FM, et al. A randomized comparison of every-2-week darbepoetin alfa and weekly epoetin alfa for the treatment of chemotherapy-induced anemia in patients with breast, lung, or gynecologic cancer. The Oncologist. 2004;9(6):696-707. doi:10.1634/theoncologist.9-6-696

 

Economic Evaluation of Weekly Epoetin Alfa versus Biweekly Darbepoetin Alfa for Chemotherapy-Induced Anaemia
Design

16-week, open-label, multicentre, randomised trial

N= 358

Objective To compare resource use, costs and clinical outcomes between weekly epoetin alfa and biweekly darbepoetin alfa in patients with chemotherapy-induced anaemia using a cost-consequences framework
Study Groups

Epoetin alfa (n= 178)

Darbepoetin alfa (n= 180)

Inclusion Criteria Patients with solid-tumour malignancy, aged ≥18 years, with haemoglobin level ≤11 g/dL, scheduled to receive cyclic chemotherapy for at least 12 weeks
Exclusion Criteria Not specified
Methods Patients were randomised to receive either weekly epoetin alfa 40,000 units or biweekly darbepoetin alfa 200μg. Costs were assigned using US Medicare reimbursement rates and indirect costs based on mean hourly wage. Clinical outcomes included haemoglobin levels and transfusions. Sensitivity analyses evaluated different costing methods and perspectives.
Duration 16 weeks
Outcome Measures

Primary: Proportion of patients experiencing a ≥1 g/dL rise in hemoglobin by week 5 (i.e., hemoglobin response)

Secondary: Hematological response (≥2 g/dL rise by week 9), AUC for change in hemoglobin, mean hemoglobin levels at weeks 5, 9, 13, and 17, proportion of patients with hemoglobin ≥11 g/dL by weeks 5 and 9

Baseline Characteristics   Epoetin alfa (n = 175) Darbepoetin alfa (n = 177)
Female 107 (61.1%) 118 (66.7%)
Age, years (mean, SD) 62.0 (11.7) 63.3 (11.8)

Type of cancer

Breast

Lung

Other

 

38 (21.7%)

45 (25.7%)

92 (52.6%)

 

50 (28.2%)

46 (26.0%)

81 (45.8%)

Baseline haemoglobin level, g/dL (mean, SD) 10.1 (0.76) 10.0 (0.82)

Baseline ECOG status

Fully active

Strenuous activity limited

Self-care but no work

 

57 (32.6%)

92 (52.6%)

26 (14.9%)

 

61 (34.5%)

87 (49.2%)

29 (16.4%)

Platinum chemotherapy 68 (38.9%) 75 (42.4%)
Results   Epoetin alfa (n = 175) Darbepoetin alfa (n = 177) p-value or difference (95% CI)
Within-trial resource used
Patients hospitalized 43 (24.6%) 39 (22.0%) 0.57
Hospitalizations 0.34 (0.72%) 0.27 (0.55%) 0.72
Inpatient days 2.61 ± 6.00 1.58 ± 4.10 1.03 (0.074 to 2.27)
Blood transfusions 0.26 ± 0.65 0.45 ± 1.08 -0.19 (-0.37 to -0.009)
Units of blood transfused 0.46 ± 1.19 0.88 ± 2.11 -0.42 (-0.77 to -0.08)
Days of unplanned radiation therapy 0.31 ± 2.97 0.71 ± 4.38 -0.40 (-1.24 to 0.29)
Response metrics
Hemoglobin response 78 (44.6%) 60 (33.9%) 10.7 (1.7 to 20.6)
Hematological response 78 (44.6%) 48 (27.1%) 17.5 (8.0 to 26.7)
AUC for change in hemoglobin, g/dL 11.1 ± 13.5 6.0 ± 13.8 5.1 (2.4 to 7.7)

Hemoglobin level, g/dL

Week 5

Week 9 

Week 13

Week 17

 

10.9 ± 1.5

11.2 ± 1.5

11.4 ± 1.4

11.4 ± 1.5

 

10.4 ± 1.4

10.6 ± 1.7

10.8 ± 1.7

10.9 ± 1.8

 

0.5 (0.3 to 0.9)

0.6 (0.3 to 1.0)

0.6 (0.3 to 1.0)

0.5 (0.2 to 0.9)

Patients with hemoglobin ≥ 11 g/dL

By week 5

By week 9

 

98 (56%)

125 (71.4%)

 

79 (44.6%)

105 (59.3%)

 

11.4 (0.4 to 21.0)

12.1 (1.3 to 21.3)

Adverse Events N/A
Study Author Conclusions Most clinical outcome measures suggested greater improvement with epoetin alfa relative to darbepoetin alfa, but most costs for both agents appeared similar. Decision makers must evaluate the differences in costs and efficacy measures that are most relevant from their perspectives.
Critique The study's strengths include its randomised design and detailed cost analysis. However, the open-label design may introduce bias, and the short follow-up period limits the assessment of long-term outcomes. Additionally, the study did not account for potential differences in patient management practices across sites, which could affect resource use and costs.
Table 5 References:
[8] Reed SD, Radeva JI, Daniel DB, et al. Economic evaluation of weekly epoetin alfa versus biweekly darbepoetin alfa for chemotherapy-induced anaemia: evidence from a 16-week randomised trial. Pharmacoeconomics. 2006;24(5):479-494. doi:10.2165/00019053-200624050-00006