The Oncologist, Vol. 10, No. 5, 357-360, May 2005; doi:10.1634/theoncologist.10-5-357
© 2005 AlphaMed Press
Effect of Patient Exclusion Criteria on the Efficacy of Erythropoiesis-Stimulating Agents in Patients with Cancer-Related Anemia
Timothy J. Littlewooda,
Brad Schenkelb,
Martin Lissc
a John Radcliffe Hospital, Headington, Oxford, United Kingdom;
b Johnson and Johnson Pharmaceutical Services, LLC, Raritan, NJ, USA;
c Liberty Mutual Group, Boston, MA, USA (formerly of Policy Analysis Inc., Brookline, MA, USA)
Correspondence: Timothy J. Littlewood, M.D., John Radcliffe Hospital, Headington, OX3 9DU, Oxford, United Kingdom. Telephone: 44-0-1865220331; Fax: 44-0-1865221778; e-mail: tim.littlewood{at}btinternet.com
 |
ABSTRACT
|
|---|
Comparison of the efficacies of erythropoiesis-stimulating agents (ESAs) between different clinical trials is becoming increasingly common, although differences in study design and populations evaluated can have a considerable effect on results. A comparison of two seemingly similar trials of ESAs, one of epoetin alfa and the other of epoetin beta, showed that only 27% of the 115 patients with hematologic malignancies who received epoetin alfa in the epoetin alfa trial met the inclusion criteria for the epoetin beta trial. The mean hemoglobin increase from baseline to week 16 of epoetin alfa therapy in the patients who met these inclusion criteria was 3.3 g/dl. This is substantially higher than the mean hemoglobin increase of 2.2 g/dl from baseline to week 16 of epoetin alfa therapy in the patients who did not meet the epoetin beta study inclusion criteria. These results demonstrate the considerable effects that exclusion criteria can have on trial results and highlight the value of scrutinizing the study design details of clinical trials before comparing outcome data between trials.
Key Words. Erythropoiesis-stimulating agents • Epoetin alfa • Study design • Hemoglobin • Anemia • Cancer
|
INTRODUCTION
|
|---|
Numerous randomized, placebo-controlled trials have established erythropoiesis-stimulating agents (ESAs; epoetin alfa [Eprex®/Epypo®; Ortho-Biotech UK/Janssen-Cilag, High Wycombe, United Kingdom, http://www.orthobiotech.co.uk; Procrit®; Ortho Biotech Products, L.P.; Bridgewater, NJ, http://www.orthobiotech.com], darbepoetin alfa [Aranesp®; Amgen Inc., Thousand Oaks, CA, http://www.amgen.com], epoetin beta [NeoRecormon®; Hoffman-La Roche, Basel, Switzerland, http://www.roche.com]) as an effective treatment option for patients with anemia related to cancer and its treatment [1, 2], including patients with hematologic malignancies [3, 4]. However, as in all therapeutic areas, differences among studies in terms of their methodologies, study designs, and the clinical end points evaluated can considerably influence the study results and conclusions. The effects of these methodologic differences may not be immediately obvious to health care professionals and may hinder their ability to interpret and compare results among studies. For example, comparing the results of clinical trials of various ESAs can be difficult because the trials often differ in methodology in terms of patient populations analyzed, eligibility criteria used (including definition of anemia and baseline hemoglobin [Hb] level), definitions and time points of hematologic response, and length of follow-up (trial duration). Efficacy analyses also have varied in terms of protocol definitions of evaluable patient populations: intent-to-treat (ITT), evaluable for efficacy, and available data subpopulation (sometimes referred to as per-protocol). These inconsistencies have created major challenges for physicians and other health care professionals as they seek to understand, interpret, and apply the results of the various studies to their practices.
To demonstrate the effects of clinical trial design on results and the hazards of comparing the results of different clinical trials without a full understanding of the effect of trial design, we conducted a retrospective analysis in which efficacy data from a study of epoetin alfa for anemia correction in patients with cancer were reanalyzed using inclusion criteria from a study of epoetin beta for anemia correction in patients with cancer.
|
METHODS
|
|---|
Full reports of both trials have been published previously [5, 6]. The epoetin alfa study was a randomized, placebo-controlled, multicenter, phase III trial of epoetin alfa in 375 patients with cancer (173 of whom had hematologic malignancies) and anemia [5]. Patients were required to have had no transfusions within 14 days prior to baseline and a baseline Hb level 
10.5 g/dl, or >10.5 g/dl but 12 g/dl after a 
1.5-g/dl decrease during the previous month or chemotherapy cycle. No exclusions were made based on baseline serum erythropoietin levels. Patients were either receiving or scheduled to receive nonplatinum chemotherapy with a minimum cycle duration of 3 weeks. Patients were prospectively stratified according to tumor stratum (solid or hematologic) and Hb level (10.5 g/dl or >10.5 g/dl) and were randomized 2:1 to receive either epoetin alfa initiated at a dose of 150 IU/kg three times weekly (TIW) or a matching volume of placebo. In the epoetin alfa group, the epoetin alfa dose was increased to 300 IU/kg TIW in patients whose Hb levels did not increase by 1 g/dl after 4 weeks of treatment. Epoetin alfa treatment was continued for up to 28 weeks, which included 12 to 24 weeks of chemotherapy plus a 4-week treatment period following the last chemotherapy dose. Transfusions were permitted during the study at the discretion of the physician but were to be avoided in patients with Hb levels >8 g/dl unless clinically indicated.
The more recent epoetin beta study was an open-label, randomized, parallel-group, multicenter, phase III trial that compared the efficacy of epoetin beta given at a dose of 30,000 IU once weekly (QW) (n = 119) with the efficacy of 10,000 IU TIW (n = 122) in patients with low-grade lymphoproliferative malignancies (non-Hodgkin’s lymphoma, multiple myeloma, or chronic lymphocytic leukemia; NeoRecormon Once Weekly [NOW] trial) [6]. Patients enrolled in this study had to have no history of transfusion within 2 months prior to baseline, a baseline Hb level of 9 g/dl to 11 g/dl, and a baseline serum erythropoietin level 100 mU/ml. Patients undergoing chemotherapy had to continue this therapy for at least 4 months, although receipt of chemotherapy during epoetin beta treatment was not a requirement. For patients who required a blood transfusion or experienced an Hb increase of <0.5 g/dl from baseline after 4 weeks of epoetin beta therapy, the epoetin beta dose was doubled (i.e., the dose per injection was increased to 20,000 IU for the TIW group and to 60,000 IU for the QW group [administered as 30,000 IU twice weekly]). Treatment was continued for 16 weeks. Transfusions were to be avoided in patients with Hb levels >8.5 g/dl unless clinically indicated.
In the current retrospective analysis, efficacy results were compared between the subgroups of patients with hematologic malignancies in the epoetin alfa trial who did and did not meet the more strict inclusion/exclusion criteria of the epoetin beta trial. From the ITT population (defined as all randomized patients) of the original epoetin alfa study, the subset of patients with hematologic malignancies was extracted. From this subset, patients were further divided into those who met the inclusion/exclusion criteria of the epoetin beta trial and those who did not. Outcome measures compared between these two subgroups were change in Hb level from baseline to week 16 of epoetin alfa therapy and the percentage of patients transfused. Week 16 was chosen as the end point of treatment since this is the most commonly reported duration of ESA therapy in the medical literature, and it is also a common treatment duration in the community oncology practice setting.
|
RESULTS
|
|---|
One hundred seventy-three of the original 375 patients in the epoetin alfa trial (ITT population) had hematologic malignancies and so were included in this retrospective analysis. Details of the baseline demographics and clinical characteristics of these patients have been reported elsewhere [7]. Of these 173 patients, 115 patients were randomized to the epoetin alfa group. A total of 27% (31/115) met the inclusion criteria of the epoetin beta study, and 73% (84/115) did not. The reasons that these 84 patients did not meet the criteria are shown in Table 1
. The epoetin alfa recipients who met the inclusion criteria of the epoetin beta study had substantially better outcomes in terms of Hb increases and transfusion requirements (Table 2). These patients (who received epoetin alfa at a dose of 150 IU/kg TIW) had a mean increase in Hb level of 3.3 g/dl from baseline to week 16, which was 50% higher than the 2.2-g/dl increase in the cohort that did not meet the entry criteria of the epoetin beta trial. In addition, the percentages of patients requiring transfusions for these two cohorts were 9.7% and 28.6%, respectively, indicating a 66% greater transfusion rate in the patients who did not meet the inclusion criteria of the epoetin beta study.
View this table:
[in this window]
[in a new window]
|
Table 1. Reasons why patients with hematologic malignancies in the epoetin alfa trial [5] did not meet the eligibility criteria of the epoetin beta trial [6]
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Outcomes for patients with hematologic malignancies in the epoetin alfa trial [5] stratified by whether they met the eligibility criteria of the epoetin beta trial [6]
|
|
|
DISCUSSION
|
|---|
The results of this retrospective analysis highlight the difficulties in comparing data from different studies with noncomparable populations. The application of strict inclusion criteria in a clinical trial creates the potential for a different outcome in the selected population compared with the more general population, and in some cases the difference can be large. In the present example, data were compared between patients with hematologic malignancies and anemia scheduled to receive chemotherapy in the epoetin alfa study who did and did not meet the strict eligibility criteria of the epoetin beta study. However, because of the differences in inclusion criteria between studies, only 27% of patients with hematologic malignancies in the epoetin alfa trial who were scheduled to receive epoetin alfa would have been enrolled in the later epoetin beta trial. Comparing the efficacy outcomes of the epoetin alfa recipients who would and would not have been enrolled in the epoetin beta study revealed large differences for both Hb response and percentage of patients transfused. Patients with hematologic malignancies in the epoetin alfa trial who did not meet the inclusion criteria for the epoetin beta trial experienced a mean increase in Hb level of 2.2 g/dl after 16 weeks of treatment, but those who did meet the criteria experienced a mean increase in Hb level of 3.3 g/dl. This represents a 50% difference in Hb increase between the two patient populations in the epoetin alfa trial. In addition, 9.7% of the patients who met the epoetin beta trial criteria received a transfusion, whereas 28.6% of those who did not meet the criteria received a transfusion (66% difference between groups). These major between-group differences in outcomes suggest that the inclusion criteria used in the epoetin beta trial selected for a patient population with a substantially higher likelihood of response than that observed in the general population.
Assuming comparisons between different trials will continue, it is imperative that data reported from individual studies be closely scrutinized to take into account study design, patient populations, and efficacy variables. Differences between the patient populations in the two studies evaluated here may also have included variations in disease stage, type and intensity of previous and current treatment, and comorbidities. Any or all of these factors might influence the outcome of treatment with an ESA. Bokemeyer et al. [8] have highlighted several aspects of the epoetin beta trial [6] discussed here that may affect the reader’s interpretation of the results and conclusions—that is, that epoetin beta given at a dose of 30,000 IU s.c. QW is as effective as 10,000 IU s.c. TIW. For example, patients were excluded if they had been transfused within 2 months of baseline, had serum erythropoietin levels >100 mU/ml, or had Hb levels outside the range of 9 g/dl to 11 g/dl. This suggests inclusion of patients with only mild to moderate anemia, which is also supported by the mean baseline Hb level >10 g/dl in the study and low transfusion requirements throughout the study in both groups [8]. A previous study in patients with anemia and hematologic malignancies reported that serum erythropoietin level was one of the most important factors predicting response to erythropoietic therapy, as approximately 75% of patients with inappropriately low levels of serum erythropoietin for the degree of anemia responded to treatment [9]. Therefore, the narrow entry criteria in the epoetin beta trial may have contributed to a higher probability of response than would have been achieved with a population more generalizable to standard clinical practice. The current retrospective analysis demonstrates that the decisions made in trial design, including the criteria used for patient inclusion and exclusion, can have a substantial effect on trial results, and readers reviewing and comparing results between any two trials should always bear this in mind.
In summary, a closer inspection of the methods and results of studies evaluating ESAs is necessary to adequately compare outcomes, especially if one is attempting to compare the efficacy of different agents or regimens. In the absence of data from direct comparative studies, it will always be necessary for health care professionals to make treatment choices based upon the available data, but the designs and limitations of the studies must be thoroughly understood and taken into account when making a treatment choice.
|
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
|
|---|
Dr. Littlewood has received honoraria from OrthoBiotech, Roche, and Amgen for attending Expert Advisory Panels and giving lectures. Mr. Schenkel is an employee and stock owner of Johnson and Johnson.
|
ACKNOWLEDGMENT
|
|---|
This research was funded by a research grant from Johnson and Johnson Pharmaceutical Research and Development, L.L.C., Raritan, NJ.
|
REFERENCES
|
|---|
- Rizzo JD, Lichtin AE, Woolf SH et al. Use of epoetin in patients with cancer: evidence-based clinical practice guidelines of the American Society of Clinical Oncology and the American Society of Hematology. J Clin Oncol 2002;20:4083–4107.[Abstract/Free Full Text]
- National Comprehensive Cancer Network. Clinical Practice Guidelines in Oncology—Cancer and Treatment-Related Anemia, Version 2.2004. Jenkintown, PA: National Comprehensive Cancer Network, 2004. Available at http://www.nccn.org/physician_gls/f_guidelines.html. Accessed February 11, 2005.
- Dammacco F, Castoldi G, Rödjer S. Efficacy of epoetin alfa in the treatment of anaemia of multiple myeloma. Br J Haematol 2001;113:172–179.[CrossRef][Medline]
- Österborg A, Brandberg Y, Molostova V et al. Randomized, double-blind, placebo-controlled trial of recombinant human erythropoietin, epoetin beta, in hematologic malignancies. J Clin Oncol 2002;20:2486–2494.[Abstract/Free Full Text]
- Littlewood TJ, Bajetta E, Nortier JW et al. Effects of epoetin alfa on hematologic parameters and quality of life in cancer patients receiving nonplatinum chemotherapy: results of a randomized, double-blind, placebo-controlled trial. J Clin Oncol 2001;19:2865–2874.[Abstract/Free Full Text]
- Cazzola M, Beguin Y, Kloczko J et al. Once-weekly epoetin beta is highly effective in treating anaemic patients with lymphoproliferative malignancy and defective endogenous erythropoietin production. Br J Haematol 2003;122:386–393.[CrossRef][Medline]
- Littlewood TJ, Nortier J, Rapoport B et al. Epoetin alfa corrects anemia and improves quality of life in patients with hematologic malignancies receiving non-platinum chemotherapy. Hematol Oncol 2003;21:169–180.[CrossRef][Medline]
- Bokemeyer C, Littlewood T, Aapro M. Is once-weekly epoetin beta ‘highly effective’ in treating anaemia in patients with lymphoproliferative malignancy? Br J Haematol 2004;125:98–99.[Medline]
- Cazzola M, Messinger D, Battistel V et al. Recombinant human erythropoietin in the anemia associated with multiple myeloma or non-Hodgkin’s lymphoma: dose finding and identification of predictors of response. Blood 1995;86:4446–4453.[Abstract/Free Full Text]
Received September 9, 2004;
accepted for publication March 11, 2005.
This article has been cited by other articles:
|
|
|
|
 
B. Coiffier
Response to "Effect of Patient Exclusion Criteria on the Efficacy of Erythropoiesis-Stimulating Agents in Patients with Cancer-Related Anemia"
Oncologist,
October 1, 2005;
10(9):
762 - 763.
[Full Text]
[PDF]
|
|
|
|
|
|
|
T. J. Littlewood, B. Schenkel, and M. Liss
Response to a Letter to the Editor from Coiffier Regarding "Effect of Patient Exclusion Criteria on the Efficacy of Erythropoiesis-Stimulating Agents in Patients with Cancer-Related Anemia"
Oncologist,
October 1, 2005;
10(9):
764 - 765.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
