Article Tools
ORIGINAL REPORTS
Hematologic Malignancies
Article Tools
Epoetin Alfa in Patients With Advanced-Stage Hodgkin's Lymphoma: Results of the Randomized Placebo-Controlled GHSG HD15EPO Trial
Abstract
To determine whether epoetin alfa reduces anemia-related fatigue, improves other aspects of health-related patient-recorded outcomes (PROs), reduces the number of RBC transfusions, and has an impact on freedom from treatment failure (FFTF) and overall survival (OS) in patients with advanced-stage Hodgkin's lymphoma (HL).
The prospectively randomized HD15EPO study performed by the German Hodgkin Study Group investigated epoetin alfa administered at doses of 40,000 U weekly during and after chemotherapy (six to eight cycles of bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone [BEACOPP]) in a double-blind, placebo-controlled setting. The study accrued 1,379 patients, of whom 1,328 were assessable for safety, 1,303 were assessable for clinical outcome, and 930 were assessable for PROs.
PROs were not different in patients receiving placebo or epoetin alfa, both after the end of chemotherapy and 6 months thereafter. There was no difference between patients treated with epoetin alfa or placebo with respect to FFTF and OS. There were also no differences in the numbers of deaths, progressions, relapses, and thromboembolic events. The median number of RBC transfusions was reduced from four per patient in the placebo group to two per patient in the epoetin alfa group (P < .001), with 27.4% of patients needing no RBC transfusion in the placebo group compared with 36.7% of patients in the epoetin alfa group (P < .001).
Anemia is one of the clinically relevant problems in the treatment of cancer patients undergoing chemotherapy and has been shown to be a negative prognostic marker in a variety of different malignancies.1 Since their introduction in the late 1980s, erythropoiesis-stimulating agents (ESAs) have been demonstrated to increase the hemoglobin levels in most patients and to reduce the need for RBC transfusions. Subsequently, there was some hope, based on clinical trials, that these drugs might result in better tumor control and possibly improve overall survival (OS).2–5 However, a number of prospectively randomized trials published more recently suggested that ESAs might be associated with poorer outcome in patients with cancer.6–9 The use of ESAs in general is currently under debate and has become a matter of controversy. ESAs were the subject of recent Oncologic Drugs Advisory Committee hearings that attempted to shed some light on the increased ESA-related mortality reported in some clinical studies and possible underlying mechanisms.10 Subsequently, the product labels for ESAs were updated with a boxed warning highlighting potential safety risks.11,12
Anemia is a negative prognostic factor in Hodgkin's lymphoma (HL), both for patients in the first-line setting13 and those who have experienced relapse.14 Because the more intensive bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone (BEACOPP) regimen15 is associated with a higher need for RBC transfusions in patients with advanced-stage HL,16 we investigated the use of epoetin alfa in this setting. The present trial, HD15EPO, was part of the German Hodgkin Study Group (GHSG) HD15 trial in which eight cycles of increased dose BEACOPP (BEACOPPescalated) were compared with six cycles of BEACOPPescalated or eight cycles of baseline BEACOPP administered in 14-day intervals (BEACOPP14). Here, we report the impact of epoetin alfa on patient-recorded outcomes (PROs), RBC transfusions, hemoglobin levels, safety, tumor control, and OS as evaluated in the prospectively randomized HD15EPO trial.
Eligible patients were 18 to 60 years of age with newly diagnosed, histologically confirmed, and previously untreated HL and had Ann Arbor stage IIB with a large mediastinal mass (at least one third of the maximal thoracic diameter) alone or in combination with extranodal lesions, III, or IV disease. Patients had to have an Eastern Cooperative Oncology Group performance status of ≤ 2, to be HIV negative and free of concurrent disease that might hinder treatment, and to give written informed consent. The study protocol was approved by the GHSG panel members and ethics committees for all study centers. Patients were registered and treated in 346 hospitals and practices in Germany, Switzerland, the Netherlands, and Austria (Appendix, online only). Review of biopsy specimen by a review panel of lymphoma pathologists was an obligatory part of the protocol. The HD15 trial was registered as Current Controlled Trial No. ISRCTN32443041.
After clinical staging according to previously reported methods,15 patients in HD15 were randomly assigned to one of the following three chemotherapy regimens: eight cycles of BEACOPPescalated, six cycles of BEACOPPescalated, or eight cycles of BEACOPP14.17 Each cycle was supported with granulocyte colony-stimulating factor, as previously described.15,17 Treatment allocation was stratified by sex, age (< v ≥ 50 years), stage (IIB/IIIA v IIIB/IV), International Prognostic Score (IPS; 0 to 2 v 3 to 7), and hemoglobin level (≤ v > 12 g/dL). Independent from the chemotherapy portion of the trial, patients eligible for HD15EPO were randomly assigned to epoetin alfa or matched placebo, stratified by chemotherapy arm in a double blind setting. Epoetin alfa was administered at doses of 40,000 U subcutaneously once weekly starting in the first week of the first chemotherapy cycle and administered for up to 6 weeks after the last administration of chemotherapy or until the hemoglobin level increased to greater than 13 g/dL (or 12 g/dL in the 6 weeks after chemotherapy). During chemotherapy, injections were resumed if the hemoglobin level decreased to less than 12 g/dL. Study drug and matched placebo were self-administered by the patients after receiving instructions from their physician or study nurse on how to apply, handle, and store the supportive medication. Oral iron(II) substitution with 200 to 300 mg/d was recommended for all patients whose serum ferritin levels were less than 100 ng/mL or with transferritin saturation less than 20%. Patients in either treatment arm received RBC transfusions only when their hemoglobin concentration was less than 8 g/dL or if there was another clear medical indication.
Chemotherapy was followed by radiotherapy in patients with positron emission tomography–positive residual lymphoma after the end of chemotherapy. Apart from the preplanned interim analysis of the positron emission tomography question in HD15,18 this trial has not been published.
The Quality of Life Questionnaire for Survivors (QLQ-S),19 compiled by the European Organisation for Research and Treatment of Cancer (EORTC) and the GHSG, was used to assess health-related PRO domains. This instrument contains the EORTC Core Quality of Life Questionnaire C30 (QLQ-C30)20 and the Multidimensional Fatigue Inventory-20 (MFI-20)21 as core modules. The MFI-20 is a 20-item survey designed to differentiate fatigue on five different dimensions including general fatigue, physical fatigue, mental fatigue, reduced motivation, and reduced activity. After random assignment, PRO questionnaires were administered before chemotherapy, at each restaging, and at every follow-up visit (ie, quarterly in the first year after treatment).
The primary outcome of this trial was to compare anemia-related fatigue, as measured with the appropriate three-item EORTC QLQ-C30 scale, between epoetin alfa– and placebo-treated groups directly after chemotherapy and 6 months later. Secondary end points were the remaining PRO scales of the QLQ-S, number of transfusions needed, hemoglobin during and after treatment, safety of treatment, freedom from treatment failure (FFTF), and OS.
The HD15EPO trial was designed to detect a 10-point difference in the fatigue scores both directly after chemotherapy and 6 months later. Sample size calculation using ADDPLAN software (ADDPLAN, Cologne, Germany)22 yielded 726 PRO questionnaires necessary in a group sequential design with Pocock type boundaries. This assumed a standard deviation of 29 points in each arm, a power of 80%, and a two-sided overall significance level of 2.5% each to adjust for the two points in time analyzed. Influence of fatigue was estimated using a linear model with chemotherapy and its interaction with epoetin alfa/placebo as covariates. Sequential median unbiased estimates for differences, exact sequential CIs, and P values23 were calculated based on the least squares means. Sensitivity analyses included adding baseline fatigue as an additional covariate, dropping the interaction term, imputing missing values using last observation carried forward, and restricting analysis to a per-protocol group. Remaining PRO scales were analyzed using linear models with a similar structure.
PRO scores were calculated and adjusted to a 0 to 100 scale according to the rules of the instruments. Changes of more than 10 points were considered relevant. Time windows were 0 to 3 months and 4 to 8 months for the time points directly after chemotherapy and 6 months after chemotherapy. FFTF and OS were determined using Kaplan-Meier estimates. Hazard ratios (HRs) with 95% CIs for epoetin alfa were computed using Cox proportional hazards regression including the chemotherapy arm as a second factor. Contingency tables were analyzed using Fisher's exact test, and ordinal variables were compared using the Mann-Whitney U test.
The analysis set for clinical end points is based on the intention-to-treat (ITT) principle, only excluding nonqualified patients based on objective criteria measured before random assignment and patients without documentation. The primary analysis set for PRO analyses comprised all ITT patients with at least one postchemotherapy PRO questionnaire within the first 8 months after end of chemotherapy. Safety analyses included all patients who received at least one dose of study medication and had at least one postbaseline assessment. Design and analysis of the trial were carried out using SAS version 9.2 (SAS Institute, Cary, NC) and ADDPLAN versions 3.1 and 4.
Between January 2003 and December 2006, a total of 1,379 patients were randomly assigned in HD15EPO. These patients were contributed from a total of 346 participating centers located in Germany, Switzerland, the Netherlands, and Austria. A total of 76 patients (39 in the placebo group and 37 in the epoetin alfa group) were excluded for various reasons (Fig 1). Thus, the ITT population includes 655 (94.4%) of 694 patients in the placebo group and 648 (94.6%) of 685 patients in the epoetin alfa group. At least one postchemotherapy PRO assessment was available from 461 patients in the placebo arm and 469 patients in the epoetin alfa arm. The baseline questionnaire was completed by 94.3% of these patients (94.1% in placebo arm v 94.5% in epoetin alfa arm), the questionnaire after the end of chemotherapy was completed by 90% (90.7% in placebo arm v 89.3% in epoetin alfa arm), and the questionnaire at 6 months after chemotherapy was completed by 61.1% (62.0% in placebo arm v 60.1% in epoetin alfa arm). Safety analyses were based on 1,328 patients (665 in the placebo arm and 663 in the epoetin alfa arm).
Fig 1. Patient flow. (*) Activity index (n = 2), age (n = 1), laboratory (n = 1), and incomplete study treatment (n = 1). HL, Hodgkin's lymphoma; I/E inclusion/exclusion; PRO, patient-reported outcome.
Demographics, clinical characteristics, and reasons for exclusion from the various analyses were similar between the two groups (Table 1). The median age of patients was 34 years. There were more males than females (62% v 38%, respectively), and most patients were in clinical stages III and IV (51.2% and 34.4%, respectively). The most frequent disease-related characteristics were ≥ three lymph node areas involved (85.1%), high erythrocyte sedimentation rate (64.9%), large mediastinal mass (27.5%), and extranodal involvement (18%). The risk factors of the IPS13 were also equally distributed between groups, with a total of 32.3% of patients having an IPS of 0 to 1, 52.2% having an IPS of 2 to 3, and 15.6% having an IPS of 4 to 7.
|
| Characteristic | Placebo (n = 655) | Epoetin Alfa (n = 648) | ||
|---|---|---|---|---|
| No. of Patients | % | No. of Patients | % | |
| Age, years | ||||
| Median | 34 | 34 | ||
| Range | 18-60 | 18-60 | ||
| Sex | ||||
| Female | 249 | 38.0 | 246 | 38.0 |
| Male | 406 | 62.0 | 402 | 62.0 |
| Ann Arbor stage | ||||
| IIB | 102 | 15.6 | 86 | 13.3 |
| IIIA | 162 | 24.8 | 167 | 25.8 |
| IIIB | 169 | 25.9 | 168 | 25.9 |
| IVA | 68 | 10.4 | 64 | 9.9 |
| IVB | 152 | 23.3 | 163 | 25.2 |
| ECOG performance status | ||||
| 0 | 350 | 53.6 | 353 | 54.5 |
| 1 | 269 | 41.2 | 272 | 42.0 |
| 2 | 34 | 5.2 | 23 | 3.5 |
| GHSG risk factors | ||||
| Large mediastinal mass | 189 | 29.1 | 166 | 25.8 |
| Extranodal involvement | 115 | 17.7 | 118 | 18.4 |
| ≥ 3 lymph node areas involved | 557 | 85.3 | 549 | 85.0 |
| High ESR | 411 | 64.8 | 401 | 64.9 |
| IPS | ||||
| 0-1 | 189 | 31.7 | 192 | 32.9 |
| 2-3 | 318 | 53.3 | 298 | 51.0 |
| 4-7 | 90 | 15.1 | 94 | 16.1 |
| Stratification factors | ||||
| Age ≥ 50 years | 98 | 15.0 | 87 | 13.4 |
| Stage IIIB/IV | 390 | 59.6 | 395 | 61.0 |
| Hemoglobin > 12 g/dL | 362 | 55.7 | 366 | 56.8 |
| Reference histology | ||||
| Nodular sclerosis cHL | 342 | 57.4 | 307 | 54.0 |
| Mixed cellularity cHL | 131 | 22.0 | 168 | 29.5 |
| Lymphocyte-rich cHL | 18 | 3.0 | 13 | 2.3 |
| Lymphocyte-depleted cHL | 3 | 0.5 | 3 | 0.5 |
| cHL, not further specified | 69 | 11.6 | 58 | 10.2 |
| NLPHL | 31 | 5.2 | 18 | 3.2 |
| CT arm | ||||
| BEACOPPescalated, 8 cycles | 216 | 33.0 | 213 | 32.9 |
| BEACOPPescalated, 6 cycles | 221 | 33.7 | 220 | 34.0 |
| BEACOPP14, 8 cycles | 218 | 33.3 | 215 | 33.2 |
Abbreviations: ECOG, Eastern Cooperative Oncology Group; GHSG, German Hodgkin Study Group; ESR, erythrocyte sedimentation rate; IPS, International Prognostic Score; cHL, classical Hodgkin's lymphoma; NLPHL, nodular lymphocyte-predominant Hodgkin's lymphoma; CT, chemotherapy; BEACOPP, bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone; BEACOPP14, baseline bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone administered in 14-day intervals.
The most frequent histology confirmed by expert reference pathology was nodular sclerosis classical HL (55.7%), followed by mixed cellularity (25.7%). Of all patients, 32.9% received eight cycles of BEACOPPescalated, 33.8% received six cycles BEACOPPescalated, and 33.2% received eight cycles of BEACOPP14; additional radiotherapy to residual masses after chemotherapy was given in 8.7% of all patients. The subgroup with PRO information is representative, with slightly more women (39.6%, P = .07) and more patients reporting a large mediastinal mass (29.1%, P = .04).
In this concluding sequential analysis, epoetin alfa had no significant impact on fatigue, the primary end point of this trial. Fatigue levels were elevated before treatment and relevantly increased by more than 10 points after chemotherapy. Six months later, fatigue levels recovered, decreasing below baseline values, but they were still clearly increased compared with the general German population.24 The model-based sequential differences between placebo and epoetin alfa were −0.8 points after the end of chemotherapy (97.5% CI, −4.9 to 3.2; P = .64) and 2.7 points 6 months after chemotherapy (97.5% CI, −2.3 to 7.6; P = .22). Sensitivity analyses with variants of the linear model, imputed missing values, or using the per-protocol data set yielded similar results (data not shown). Mean values for each measurement are listed in Table 2.
|
| Scale | Baseline Questionnaire | After End of Chemotherapy Questionnaire | 6 Months After End of Chemotherapy Questionnaire | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Placebo (n = 434) | Epoetin Alfa (n = 443) | Placebo (n = 418) | Epoetin Alfa (n = 419) | Placebo (n = 286) | Epoetin Alfa (n = 282) | |||||||
| Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | |
| EORTC QLQ-C30, fatigue | 49.5 | 31.2 | 49.0 | 30.5 | 61.1 | 26.5 | 60.2 | 26.7 | 32.8 | 26.9 | 35.4 | 25.1 |
| MFI-20, general fatigue | 50.7 | 29.7 | 50.5 | 29.5 | 59.3 | 26.3 | 58.3 | 28.1 | 37.1 | 28.3 | 39.6 | 26.9 |
| MFI-20, physical feeling of fatigue | 48.2 | 31.3 | 47.7 | 31.4 | 62.6 | 28.8 | 61.7 | 29.7 | 32.8 | 28.8 | 34.3 | 27.7 |
| MFI-20, reduced activity | 47.8 | 30.4 | 49.7 | 30.7 | 56.8 | 29.8 | 56.8 | 29.5 | 30.3 | 28.4 | 33.6 | 28.4 |
| MFI-20, reduced motivation | 28.3 | 23.4 | 27.4 | 22.8 | 28.2 | 23.5 | 26.1 | 21.7 | 18.2 | 21.2 | 19.3 | 20.6 |
| MFI-20, mental fatigue | 31.8 | 25.4 | 32.2 | 28.0 | 39.0 | 27.9 | 37.6 | 29.3 | 28.2 | 27.8 | 29.4 | 28.9 |
Abbreviations: SD, standard deviation; EORTC QLQ-C30, European Organisation for Research and Treatment of Cancer Core Quality of Life Questionnaire C30; MFI-20, Multidimensional Fatigue Inventory-20.
In the MFI-20, the largest but nonsignificant difference estimate was −3.4 points (95% CI, −8.1 to 1.3; P = .16) for the reduced activity scale 6 months after end of chemotherapy. This dimension, general fatigue, and physical feeling of fatigue were more severe before treatment and increased during chemotherapy without any difference between epoetin alfa and placebo. In contrast, reduced motivation and mental fatigue did not much vary with time. Within 6 months after chemotherapy, patients recovered to less than baseline (Table 2). In addition, there were also no differences between the two arms with respect to the remaining seven functional scales, two symptom scales, and 18 single-item scales that form the QLQ-S (data not shown).
The outcomes of the 1,303 patients of the full analysis set are listed in Table 3. There were no differences between patients receiving placebo or epoetin alfa in terms of complete response/unconfirmed complete response (93.7% v 93.5%, respectively), progression or relapse (7.8% v 8.3%, respectively), treatment-related mortality (1.7% v 0.8%, respectively), and overall mortality (5.5% v 4.2%, respectively).
|
| Outcome and Failure Event | Placebo (n = 655) | Epoetin Alfa (n = 648) | ||
|---|---|---|---|---|
| No. of Patients | % | No. of Patients | % | |
| Final treatment outcome | ||||
| CR/CRu | 614 | 93.7 | 606 | 93.5 |
| PR | 11 | 1.7 | 8 | 1.2 |
| NC | 0 | 0.0 | 2 | 0.3 |
| PD | 10 | 1.5 | 19 | 2.9 |
| Not staged | 20 | 3.1 | 13 | 2.0 |
| Failure event | ||||
| Progression/relapse | 51 | 7.8 | 54 | 8.3 |
| Death | 36 | 5.5 | 27 | 4.2 |
| Cause of death | ||||
| HL | 9 | 1.4 | 13 | 2.0 |
| Toxicity | 11 | 1.7 | 5 | 0.8 |
| Secondary neoplasia | 5 | 0.8 | 5 | 0.8 |
| Other | 11 | 1.7 | 4 | 0.6 |
Abbreviations: CR, complete remission; CRu, unconfirmed complete remission; PR, partial remission; NC, no change; PD, progressive disease; HL, Hodgkin's lymphoma.
The time-to-event outcome measures are shown in Figure 2. After a median observation time of 3 years, there were no differences in terms of FFTF (HR = 0.87; 95% CI, 0.63 to 1.20; Fig 2A) or OS (HR = 0.74; 95% CI, 0.45 to 1.22; Fig 2B), each adjusted for chemotherapy arms. The total rate of thromboembolic events reported was 10.1%, with an odds ratio of 1.2 (95% CI, 0.7 to 1.9). A total of 138 serious adverse events were reported in 79 (5.9%) of 1,328 patients, with no difference between arms (odds ratio = 1.0; 95% CI, 0.6 to 1.6). The most frequently encountered adverse events were infections and infestations (n = 58); respiratory, thoracic, and mediastinal disorders (n = 19); and vascular disorders (n = 11), with 76% of events considered unrelated to treatment.
Fig 2. Time-to-event outcome measures: (A) freedom from treatment failure and (B) overall survival. Hazard ratio (HR) was adjusted for chemotherapy arm.
The number of RBC transfusions administered ranged from zero to 44 in the placebo group and zero to 35 in the epoetin alfa group. The median number of transfusions in the placebo arm was four compared with two in the epoetin alfa arm (P < .001). This difference is reflected in the fact that 27.4% of patients in the placebo group needed no RBC transfusion compared with 36.7% of patients in the epoetin alfa group. Detailed numbers of transfusions given per patient are listed in Table 4. The hemoglobin levels at the start of each cycle were higher in patients receiving epoetin alfa compared with the placebo group. Although levels decreased in both groups over time, this effect was more pronounced in the placebo group during the first cycle, and the difference between groups increased over time. However, even in the epoetin alfa group, 75% of patients reported a hemoglobin level less than 12 g/dL by the third cycle (Fig 3).
|
| RBC Transfusion | Placebo (n = 643) | Epoetin Alfa (n = 640) | ||
|---|---|---|---|---|
| No. of Patients | % | No. of Patients | % | |
| RBC transfusions | ||||
| No | 176 | 27.4 | 235 | 36.7 |
| Yes | 467 | 72.6 | 405 | 63.3 |
| No. of treatment courses with RBC transfusions | ||||
| 1-2 | 218 | 33.9 | 247 | 38.6 |
| 3-4 | 155 | 24.1 | 115 | 18.0 |
| 5-8 | 94 | 14.6 | 43 | 6.7 |
| No. of RBC transfusions | ||||
| Median | 4 | 2 | ||
| Range | 0-44 | 0-35 | ||
| 1-2 | 100 | 15.6 | 127 | 19.8 |
| 3-5 | 91 | 14.2 | 109 | 17.0 |
| 6-10 | 187 | 29.1 | 116 | 18.1 |
| 11-15 | 56 | 8.7 | 31 | 4.8 |
| > 15 | 33 | 5.1 | 21 | 3.3 |
Fig 3. Hemoglobin (Hb) levels over time. Measurements taken prior to start of cycle. CT, chemotherapy.
To our knowledge, HD15EPO is the largest randomized trial to date in which an ESA was used to prevent anemia in cancer patients undergoing chemotherapy. The major findings to emerge from this study are as follows. With a total of 1,379 patients with advanced-stage HL receiving BEACOPP chemotherapy randomly assigned between placebo and epoetin alfa 40,000 U weekly, this trial shows no difference in terms of fatigue or other aspects of health-related PROs as measured by a number of different scales and items. There was no difference in terms of FFTF, OS, progressive disease, relapse, thromboembolic events, serious adverse events, and overall mortality between the two groups.
More patients receiving epoetin alfa did not need any RBC transfusion during the course of chemotherapy compared with the placebo group (36.7% v 27.4%, respectively). In addition, the hemoglobin levels in the epoetin alfa group were higher than in the placebo group, and the number of transfusions was significantly reduced (median, two v four transfusions, respectively).
Currently, conflicting data exist on the impact of ESAs on PROs, with some studies suggesting improved PROs and others showing no effect.25,26 One of the problems in these trials is the lack of standardization and the plethora of questionnaires being used, including nonverified instruments. In contrast, the questionnaires and instruments used in the present study had been shown to be relevant and reproducible before being applied in this trial. We found no improved PROs in patients with advanced-stage HL undergoing intensive treatment with different BEACOPP variants, both after chemotherapy and with longer follow-up. It is possible that the chemotherapy administered in HD15EPO was too toxic, overruling any measurable PRO improvements related to treatment with epoetin alfa. Another important aspect is the influence of RBC transfusions, which were more frequent in the placebo group, likely diluting anemia-induced negative PRO effects. Unfortunately, we could not assess whether iron substitution was adequate because of incomplete documentation. Thus, a possible influence of iron substitution on the hemoglobin levels could not be investigated.
The clinical outcome of HD15EPO contrasts with some other recent prospectively randomized cancer trials that reported increased mortality in patients receiving ESAs.6–9 A recent systematic review based on individual patient data also gave a safety signal, in that the overall group of 13,933 patients treated in 53 randomized clinical trials had an increased risk of mortality (HR = 1.17; 95% CI, 1.06 to 1.30) and poorer OS (HR = 1.06; 95% CI, 1.0 to 1.12) during ESA treatment, although this was not significant in patients receiving chemotherapy only.27 Whether endogenous or exogenous erythropoietins stimulate proliferation of cancer cells expressing erythropoietin receptors is an ongoing debate28,29; however, little hard data support the hypothesis that ESA-induced tumor stimulation might explain the higher mortality associated with ESA treatment. A more likely explanation of increased mortality is the significant risk of thromboembolic events associated with ESA treatment, both in cancer patients (HR = 1.67; 95% CI, 1.35 to 2.06)30 and in patients undergoing hemodialysis.31 In line with these findings, the individual patient data meta-analysis identified a history of prior thrombosis as being associated with a lower risk of new thromboembolic events.27 Prophylactic anticoagulation during cancer treatment of patients with previous thromboembolic events might have protected them against the thrombogenic effects of ESA treatment. This is supported by results from a randomized trial in critically ill patients showing that patients receiving prophylaxis with heparin did not have a higher thromboembolic risk when receiving ESAs.32 These findings might in part explain why there was no increase in thromboembolic events in HD15EPO, because high-risk patients received prophylaxis and the average age of patients in this trial was 34 years. Another factor that might have prevented an increased treatment-related mortality in the present trial could be the fact that the chemotherapy given (BEACOPP) is more aggressive than what is being used in most patients with solid tumors. Overall, more than 75% of patients reported a hemoglobin level less than 12 g/dL by the third cycle.
Because there were no outcome differences, the data from this study question the current American Society of Clinical Oncology/American Society of Hematology recommendations on the use of ESAs in cancer patients.10 The most recent label changes recommend ESAs be used in cancer patients undergoing chemotherapy with no curative treatment option.11,12 In addition, the hemoglobin levels should range from 10 to 12 g/dL. On the basis of our trial, ESAs seem to be safe in younger patients undergoing more aggressive chemotherapy even if they have lower hemoglobin levels. In this respect, it is of interest that the individual patient data meta-analysis also failed to detect differences in mortality in patients with different hemoglobin concentrations.27
Although, to our knowledge, this is the largest prospectively randomized placebo controlled trial on ESAs in cancer patients, it has to be kept in mind that its design predates the current safety discussion. Thus, issues relevant to safety were only analyzed as secondary end points. In particular, the study was not powered to detect differences in survival or FFTF between patients receiving epoetin alfa and placebo. Taken together, the present trial indicates that epoetin alfa does not improve PROs of patients with advanced-stage HL receiving BEACOPP chemotherapy, reduces the number of RBCs needed during treatment, and seems to be safe in this setting.
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
Clinical trial information can be found for the following: ISRCTN32443041.
The author(s) indicated no potential conflicts of interest.
Conception and design: Andreas Engert, Peter Borchmann, Michael Fuchs
Collection and assembly of data: Andreas Josting, Matthias Villalobos, Andreas Lohri, Martin Sökler, Josée Zijlstra, Isrid Sturm, Max S. Topp, Andreas Rank, Thorsten Zenz, Martin Vogelhuber, Lucia Nogova, Hans-Henning Flechtner
Data analysis and interpretation: Andreas Engert, Heinz Haverkamp, Michael Fuchs
Manuscript writing: Andreas Engert, Andreas Josting, Heinz Haverkamp, Peter Borchmann, Michael Fuchs
Final approval of manuscript: Andreas Engert, Andreas Josting, Heinz Haverkamp, Matthias Villalobos, Andreas Lohri, Martin Sökler, Josée Zijlstra, Isrid Sturm, Max S. Topp, Andreas Rank, Thorsten Zenz, Martin Vogelhuber, Lucia Nogova, Peter Borchmann, Michael Fuchs, Hans-Henning Flechtner, Volker Diehl
| 1. | JE Groopman, LM Itri : Chemotherapy-induced anemia in adults: Incidence and treatment J Natl Cancer Inst 91: 1616– 1634,1999 Crossref, Medline, Google Scholar |
| 2. | S Cascinu, A Fedeli, E Del Ferro , etal : Recombinant human erythropoietin treatment in cisplatin-associated anemia: A randomized, double-blind trial with placebo J Clin Oncol 12: 1058– 1062,1994 Link, Google Scholar |
| 3. | M Cazzola, D Messinger, V Battistel , etal : Recombinant human erythropoietin in the anemia associated with multiple myeloma or non-Hodgkin's lymphoma: Dose finding and identification of predictors of response Blood 86: 4446– 4453,1995 Crossref, Medline, Google Scholar |
| 4. | A Osterborg, MA Boogaerts, R Cimino , etal : Recombinant human erythropoietin in transfusion-dependent anemic patients with multiple myeloma and non-Hodgkin lymphoma: A randomized multicenter study Blood 87: 2675– 2682,1996 Crossref, Medline, Google Scholar |
| 5. | TJ Littlewood, E Bajetta, JW Nortier , etal : 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 19: 2865– 2874,2001 Link, Google Scholar |
| 6. | M Henke, R Laszig, C Ruebe , etal : Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy: Randomized, double-blind, placebo-controlled trial Lancet 362: 1255– 1260,2003 Crossref, Medline, Google Scholar |
| 7. | B Leyland-Jones : Breast cancer trial with erythropoietin terminated unexpectedly Lancet Oncol 4: 459– 460,2003 Crossref, Medline, Google Scholar |
| 8. | RE Smith, NS Tchekmedyian, D Chan , etal : A dose- and schedule-finding study of darbepoetin alpha for the treatment of chronic anaemia of cancer Br J Cancer 88: 1851– 1858,2003 Crossref, Medline, Google Scholar |
| 9. | J Overgaard, C Hoff, A Sand Hansen , etal : Randomized study of the importance of novel erythropoiesis stimulating protein (Aranesp) for the effect of radiotherapy in patients primary squamous cell carcinoma of the head and neck: The Danish Head and Neck Cancer Group DAHANCA 10 randomized trial Eur J Cancer 5: 6LB,2007 suppl abstr Google Scholar |
| 10. | V Brower : ESAs further restricted, but debate continues J Natl Cancer Inst 100: 1344– 1351,2008 Crossref, Medline, Google Scholar |
| 11. | Aranesp (darbepoetin alfa) package insert 2008 Amgen Thousand Oaks, CA Amgen Google Scholar |
| 12. | L.P. PROCRIT (epoetin alfa) package insert 2008 Ortho Biotech Products Raritan, NJ Ortho Biotech Google Scholar |
| 13. | D Hasenclever, V Diehl : A prognostic score for advanced Hodgkin's disease: International Prognostic Factors Project on Advanced Hodgkin's Disease N Engl J Med 339: 1506– 1514,1998 Crossref, Medline, Google Scholar |
| 14. | A Josting, J Franklin, M May , etal : New prognostic score based on treatment outcome of patients with relapsed Hodgkin's lymphoma registered in the database of the German Hodgkin Lymphoma Study Group J Clin Oncol 20: 221– 230,2002 Link, Google Scholar |
| 15. | V Diehl, J Franklin, M Pfreundschuh , etal : Standard and increased-dose BEACOPP chemotherapy compared with COPP-ABVD for advanced Hodgkin's disease N Engl J Med 348: 2386– 2395,2003 Crossref, Medline, Google Scholar |
| 16. | C Engel, M Loeffler, S Schmitz , etal : Acute hematologic toxicity and practicability of dose-intensified BEACOPP chemotherapy for advanced stage Hodgkin's disease Ann Oncol 11: 1105– 1114,2000 Crossref, Medline, Google Scholar |
| 17. | M Sieber, H Bredenfeld, A Josting , etal : Fourteen-day variant of the bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone regimen in advanced-stage Hodgkin's lymphoma: Results of a pilot study of the German Hodgkin's Lymphoma Study Group J Clin Oncol 21: 1734– 1739,2003 Link, Google Scholar |
| 18. | C Kobe, M Dietlein, J Franklin , etal : Positron emission tomography has a high negative predictive value for progression or early relapse for patients with residual disease after first-line chemotherapy in advanced-stage Hodgkin lymphoma Blood 112: 3989– 3994,2008 Crossref, Medline, Google Scholar |
| 19. | H Flechtner, JU Rueffer, M Henry-Amar , etal : Quality of life assessment in Hodgkin's disease: A new comprehensive approach—First experiences from the EORTC/GELA and GHSG trials Ann Oncol 9: 147– 154,1998 suppl 5 Crossref, Medline, Google Scholar |
| 20. | NK Aaronson, S Ahmedzai, B Bergman , etal : The European Organisation for Research and Treatment of Cancer QLQ-C30: A quality-of-life instrument for use in international clinical trials in oncology J Natl Cancer Inst 85: 365– 376,1993 Crossref, Medline, Google Scholar |
| 21. | EMA Smets, B Garsson, B Bonke , etal : The Multidimensional Fatigue Inventory (MFI): Psychometric qualities of an instrument to assess fatigue J Psychosom Res 39: 315– 325,1995 Crossref, Medline, Google Scholar |
| 22. | G Wassmer, R Eisebitt : ADDPLAN Adaptive Designs–Plans and Analyses User's Guide, Release 4 2007 Cologne, Germany ADDPLAN Google Scholar |
| 23. | AA Tsiatis, GL Rosner, C Metha : Exact confidence intervals following a group sequential test Biometrics 40: 797– 803,1984 Crossref, Medline, Google Scholar |
| 24. | R Schwarz, A Hinz : Reference data for the quality of life questionnaire EORTC QLQ-C30 in the general German population Eur J Cancer 37: 1345– 1351,2001 Crossref, Medline, Google Scholar |
| 25. | M Boogaerts, B Coiffier, C Kainz , etal : Impact of epoetin beta on quality of life in patients with malignant disease Br J Cancer 88: 988– 995,2003 Crossref, Medline, Google Scholar |
| 26. | M Razzouk, J Hord, M Hockenberry , etal : Double-blind, placebo-controlled study of quality of life, hematologic end points, and safety of weekly epoetin alfa in children with cancer receiving myelosuppressive chemotherapy J Clin Oncol 24: 3583– 3589,2006 Link, Google Scholar |
| 27. | J Bohlius, K Schmidlin, C Brillant , etal : Recombinant human erythropoiesis-stimulating agents and mortality in patients with cancer: A meta-analysis of randomized trials Lancet 373: 1532– 1542,2009 Crossref, Medline, Google Scholar |
| 28. | AM Sinclair, N Rogers, L Busse , etal : Erythropoietin receptor transcription is neither elevated nor predictive of surface expression in human tumour cells Br J Cancer 98: 1059– 1067,2008 Crossref, Medline, Google Scholar |
| 29. | MO Arcasoy : Erythropoiesis-stimulating agent use in cancer: Preclinical and clinical perspectives Clin Cancer Res 14: 4685– 4690,2008 Crossref, Medline, Google Scholar |
| 30. | J Bohlius, J Wilson, J Seidenfeld , etal : Recombinant human erythropoietins and cancer patients: Updated metaanalysis of 57 studies including 9353 patients J Natl Cancer Inst 98: 708– 714,2006 Crossref, Medline, Google Scholar |
| 31. | A Besarab, WK Bolton, JK Browne , etal : The effects of normal as compared with low hematocrit values in patients with cardiac disease who are receiving hemodialysis and epoetin N Engl J Med 339: 584– 590,1998 Crossref, Medline, Google Scholar |
| 32. | HL Corwin, A Gettinger, TC Fabian , etal : Efficacy and safety of epoetin alfa in critically ill patients N Engl J Med 357: 965– 976,2007 Crossref, Medline, Google Scholar |
Acknowledgment
Supported by the Deutsche Krebshilfe, the Swiss Federal Government, and Jannsen-Cilag GmbH, a division of Johnson & Johnson. Presented in part at the 50th Annual Meeting of the American Society of Hematology, December 6-9, 2008, San Francisco, CA.
Recruiting centers (in Germany, unless indicated otherwise) recruiting more than three patients onto the HD15EPO trial were as follows: Universitätsklinikum Heidelberg, Heidelberg; Swiss Group for Clinical Cancer Research Koordinationszentrum, Bern, Switzerland; Eberhard-Karls-Universität, Tübingen; Free University of Amsterdam, Amsterdam, the Netherlands; Campus Virchow Klinikum, Medizinische Klinik, Berlin; Universitätsklinik Köln, Köln; Universitätsklinik u. Poliklinik Würzburg, Würzburg; Klinikum Großhadern, München; Universitätsklinik Regensburg, Regensburg; Universitätsklinikum Ulm, Ulm; Med. Universitätsklinik Homburg, Homburg; Universitätsklinik Mainz, III. Med. Klinik, Mainz; Zentralklinikum Augsburg, Augsburg; Asklepios Klinik St. Georg, Onkologisches Zentrum, Hamburg; Universitätsklinik Schleswig-Holstein, II. Medizinische Klinik, Kiel; Charité Campus Mitte, Zentrum für Innere Medizin, Berlin; Krankenhaus München Schwabing, 1. Medizinische Abteilung, München; St. Bernward-Krankenhaus, Hildesheim; HELIOS Klinikum Berlin-Buch, Robert-Rössle-Klinik, Berlin; Sana Kliniken Lübeck GmbH, Krankenhaus Süd, Lübeck; Univ. Krankenhaus Eppendorf, Onkolog. Zentrum, Hamburg; Georg-August-Universität, Göttingen; Universitätsklinik Essen, Essen; Dr. Horst-Schmidt-Kliniken Wiesbaden, Wiesbaden; Dr. v. Grünhagen, Praxis Innere Medizin, Cottbus; Klinikum Bremen Mitte gGmbh, Bremen; Klinikum Nürnberg, Nürnberg; Klinikum St. Georg gGmbH, Leipzig; St. Marien-Krankenhaus, Siegen; Universitätsklinik Münster, Münster; Gemeinschaftspraxis am Städt. Krankenhaus, Gütersloh; Klinikum Ernst von Bergmann, Medizinische Klinik, Potsdam; Krankenhaus der Barmherzigen Brüder, Trier; Landeskliniken Salzburg, Salzburg, Austria; Städtisches Klinikum Braunschweig, Braunschweig; Städtisches Klinikum Karlsruhe, Karlsruhe; Ernst-Moritz-Arndt-Universität, Greifswald; Kliniken Maria Hilf, Medizinische Klinik (St. Franziskus), Mönchengladbach; Klinikum Krefeld, Krefeld; Martin-Luther-Universität Halle-Wittenberg, Halle; Universitätsklinikum Mannheim GmbH, Mannheim; Wilhelm-Anton-Hospital Goch, Goch; Dr. Schliesser, Praxis für Hämatologie, Gießen; Evang. Diakoniekrankenhaus, Bremen; Katharinenhospital Stuttgart, Stuttgart; Klinikum der Philipps-Universität, Marburg; Klinikum Oldenburg, Oldenburg; Med. Fakultät der RWTH Aachen, Aachen; Otto v. Guericke Universität Magdeburg, Magdeburg; Robert-Bosch-Krankenhaus, Stuttgart; Städt. Kliniken Darmstadt, Elisabethen Stift, Darmstadt; Universität Rostock, Rostock; Universitätsklinik Erlangen, Erlangen; Universitätsklinikum Benjamin Franklin, Freie Universität Berlin, Berlin; Vivantes Medizinisches Versorgungszentrum Neukölln, Berlin; Asklepios Klinik Altona, Onkologisches Zentrum, Hamburg; Dr. Burkhard, Onkologische Schwerpunktpraxis, Worms; Dr. Marschner, Onkologische Schwerpunktpraxis, Freiburg; Dr. Schäfer/Dr. Just/Görner, Gemeinschaftspraxis Innere Medizin, Bielefeld; Kath. Krankenhaus Hagen, St. Marien Hospital, Hagen; Klinikum der Stadt Ludwigshafen, Ludwigshafen; Klinikum Kempten/Oberallgäu, Kempten/Allgäu; Klinikum Minden, Minden; Krankenhaus Düren, Düren; Krankenhaus Nordwest, Frankfurt; Medizinische Poliklinik, der Universität Bonn, Bonn; Ortenau Klinikum, Offenburg; Prof. Dr. Tesch, Onkologische Gemeinschaftspraxis, Frankfurt; St. Antonius Hospital, Eschweiler; Städt. Klinikum, Elisabethen Stift, Dessau; Westpfalz Klinikum GmbH, Kaiserslautern; Asklepios Harzkliniken Goslar, Goslar; Bürgerhospital Stuttgart, Dr. Demandt, Stuttgart; Diakonie-Klinikum, Dr. Demandt, Stuttgart; Dietrich-Bonhoeffer-Klinikum, Neubrandenburg; Dr. Marquard, Onkologische Schwerpunktpraxis, Celle; Dr. Heine/Dr. Haessner, Gemeinschaftspraxis, Wolfsburg; Evangelisches Krankenhaus, Essen-Süd, Essen; Kliniken der Landeshauptstadt Düsseldorf, Krankenhaus Benrath, Düsseldorf; Kliniken Dortmund gGmbH, Medizinische Klinik Nord, Dortmund; Klinikum “Rechts der Isar,” München; Klinikum Fulda, Fulda; SLK Kliniken Heilbronn GmbH, Klinikum am Gesundbrunnen, Heilbronn; Universitätsklinik Graz, Graz, Austria; Caritasklinik St. Theresia, Saarbrücken; Dr. Eckart, Praxisgemeinschaft, Erlangen; Dr. Wehmeyer/Dr. Kratz-Albers/Dr. Lerchenmüller, Praxis Innere Medizin, Münster; Helios Kliniken Schwerin, Zentrum für Innere Medizin, Schwerin; HELIOS Klinikum Bad Saarow, Robert-Rössle-Klinik, Bad Saarow; Justus-Liebig-Universität Gießen, Gießen; Klinikum Bamberg, Bamberg; Klinikum Garmisch-Partenkirchen, Garmisch-Partenkirchen; Klinikum Idar-Oberstein GmbH, Idar-Oberstein; Klinikum Lippe-Lemgo, Lemgo; Klinikum Rosenheim, Städtisches Krankenhaus, Rosenheim; Klinikum Schwäbisch Gmünd, Mutlangen; Klinikum St. Marien/Med. Klinik II, Amberg; Klinikum Stadt Hanau, Hanau; Klinikum Traunstein, Traunstein; Krankenhaus Ludmillenstift, Meppen; Marienhospital, Stuttgart; Med. Hochschule Hannover, Hannover; Pius-Hospital, Oldenburg; Schwarzwald-Baar-Klinikum VS GmbH, Med. Klinik Villingen, Villingen-Schwenningen; St. Johannes Hospital, Duisburg; St. Vincentius-Kliniken gAG, Karlsruhe; Städt. Klinikum Magdeburg, Magdeburg; Therapiebereich Siloah, Med. Klinik III, Hannover; Universitätsklinik Düsseldorf, Düsseldorf; Universitätsklinik Frankfurt, Frankfurt; Universitätsklinik Leipzig, Leipzig; Universitätsklinikum Schleswig-Holstein, Campus Lübeck, Lübeck; Brüderkrankenhaus St. Josef, Paderborn. Forty-three centers recruited three patients, 83 centers recruited two patients, and 108 centers recruited one patient.
