- © 2009 by American Society of Clinical Oncology
Intensive Dose-Dense Compared With Conventionally Scheduled Preoperative Chemotherapy for High-Risk Primary Breast Cancer
- Michael Untch,
- Volker Möbus,
- Walther Kuhn,
- Bernd Rudolph Muck,
- Christoph Thomssen,
- Ingo Bauerfeind,
- Nadia Harbeck,
- Christoph Werner,
- Annette Lebeau,
- Andreas Schneeweiss,
- Stephen Kahlert,
- Franz von Koch,
- Karl Ulrich Petry,
- Diethelm Wallwiener,
- Rolf Kreienberg,
- Ute-Susann Albert,
- Hans-Joachim Lück,
- Axel Hinke,
- Fritz Jänicke and
- Gottfried E. Konecny
- From the Helios Klinikum, Campus Berlin Buch, Berlin; Klinikum Frankfurt Höchst, Frankfurt; Bethesda Krankenhaus, Mönchengladbach; University of Halle (Saale), Halle (Saale); Klinikum Landshut, Landshut; Technical University of Munich; Klinikum Grosshadern, University of Munich, Munich; University of Cologne, Cologne; Frauenklinik, Evangelisches Bethesda Krankenhaus, Duisburg; University of Hamburg-Eppendorf, Hamburg; University of Heidelberg, Heidelberg; Städtisches Klinikum, Wolfsburg; University of Tübingen, Tübingen; University of Ulm, Ulm; University of Marburg, Marburg; Gynäkologisch-Onkologische Praxis, Hannover; Wissenschaftlicher Service Pharma, Langenfeld, Germany; and David Geffen School of Medicine, University of California, Los Angeles, CA.
- Corresponding author: Gottfried E. Konecny, MD, David Geffen School of Medicine, University of California Los Angeles, 2825 Santa Monica Blvd, Suite 200, Santa Monica, CA 90404-2429; e-mail: gkonecny{at}mednet.ucla.edu.
-
Presented in part at the 38th Annual Meeting of the American Society of Clinical Oncology, May 18-21, 2002, Orlando, FL.
Abstract
Purpose To compare preoperative intense dose-dense (IDD) chemotherapy with conventionally scheduled preoperative chemotherapy in high-risk primary breast cancer (BC).
Patients and Methods In this randomized phase III trial a total of 668 eligible primary BC patients stratified for tumors ≥ 3 cm (n = 567) or inflammatory BC (n = 101) were randomly assigned to receive concurrent preoperative epirubicin/paclitaxel every 3 weeks or dose-dense and dose-escalated sequential epirubicin followed by paclitaxel every 2 weeks. All patients received three cycles of cyclophosphamide, methotrexate, and fluorouracil chemotherapy after surgery.
Results IDD treatment significantly improved pathologic complete response rate (18% v 10%; odds ratio [OR] 1.89; P = .008), disease-free survival (DFS; hazard ratio [HR], 0.71; P = .011), and overall survival (OS; HR, 0.83; P = .041) compared to epirubicin/paclitaxel. Patients with inflammatory BC had a particularly poor prognosis and did not appear to benefit from IDD therapy in this trial (DFS HR, 1.10; P = .739; OS HR, 1.25; P = .544). In contrast, patients with noninflammatory BC significantly benefited from IDD treatment (DFS HR, 0.65, P = .005; OS HR, 0.77, P = .013). Treatment effects in multivariate analysis were significant for noninflammatory BC (DFS HR, 0.65, P = .015; OS HR, 0.79, P = .034), but not for all patients (DFS HR, 0.76; P = .088; OS HR, 0.82; P = .059). IDD therapy was associated with significantly more nonhematologic toxicities, anemia, and thrombocytopenia, but with similar neutropenia and infection rates.
Conclusion Our results support the efficacy and short-term safety of IDD as preoperative chemotherapy. IDD was less well tolerated compared to standard treatment, but improved clinical outcomes in patients with noninflammatory high-risk primary BC.
INTRODUCTION
Preoperative chemotherapy is the initial choice of treatment for patients with locally advanced tumors.1 Interest has developed in extending this approach to patients with less advanced operable breast cancer (BC).2–4 Increased dose-intensity has been predicted to improve the effectiveness of adjuvant chemotherapy in BC.5 Dose-intensity can be increased by escalating the single dose per cycle6 or the total dose of cytotoxic drugs.7 The dose-dense approach increases dose-intensity (drug delivery over time) by reducing the intertreatment interval for chemotherapy delivery.8–10 Hudis et al8 reported a promising 4-year disease-free survival (DFS) rate of 78% and an acceptable toxicity profile in women with high-risk node-positive BC who were treated in an adjuvant phase II study with both sequential dose-dense and dose-escalated paclitaxel, doxorubicin, and cyclophosphamide. Motivated by these results we intended to compare a sequential epirubicin paclitaxel dose-dense and dose-escalated regimen with a conventionally scheduled epirubicin and paclitaxel containing regimen in women with high-risk primary BC. Furthermore, we intended to investigate this therapy concept in the preoperative setting that would allow us to compare the objective pathologic complete response (pCR) rate as well as DFS and overall survival (OS). Because intensifying cyclophosphamide two or four times that given in standard clinical practice did not substantively improve outcome in reports of National Surgical Adjuvant Breast and Bowel Project (NSABP) B-22 and NSABP B-25, we elected to give cyclophosphamide at a standard dose as combination chemotherapy with methotrexate and fluorouracil after surgery.11 The Southwest Oncology Group/Intergroup study 9623 was initiated to determine whether the intensive dose-dense (IDD) approach to adjuvant chemotherapy would be superior to near standard doses of adjuvant chemotherapy followed by high-dose chemotherapy with autologous hematopoietic progenitor cell transplantation. However, the trial was stopped due to slow accrual after the release of data from transplantation trials in BC showing no significant benefit from this approach.12 iImportantly, to our knowledge, our study is the first randomized comparison of an IDD regimen including both an anthracycline and a taxane with a concurrent conventionally scheduled regimen in patients with primary BC. Initial results have been reported in abstract format.13 This article is the first complete reporting of the findings of this study.
PATIENTS AND METHODS
Patient Eligibility and Entry Procedures
Women with noninflammatory BC with a large primary tumor (≥ 3 cm) or inflammatory BC diagnosed by core biopsy were eligible (Fig 1). Tumor size was assessed by mammogram, ultrasound, or magnetic resonance imaging examination at study initiation. At each participating institution, the study was approved by the local institutional review board. Patients were required to give written consent and underwent chest x-ray, bone scan, abdomen ultrasound, or computed tomography scan of the abdomen before study inclusion. Adequate organ function, bone marrow reserve, and a normal left ventricular ejection fraction by echocardiography or multiple-gated acquisition scan were required.
CONSORT diagram.
Treatment
IDD treatment consisted of preoperative sequential administration of each three cycles of epirubicin 150 mg/m2 and paclitaxel 250 mg/m2 (total six cycles over 12 weeks), with cumulative epirubicin and paclitaxel doses of 450 mg/m2 and 750 mg/m2, respectively. Patients received filgrastim 5 μg/kg subcutaneously on days 3 through 10 of each cycle (Fig 2A). Conventionally scheduled chemotherapy consisted of four cycles of epirubicin 90 mg/m2 and paclitaxel 175 mg/m2 in combination repeated every 3 weeks (total four cycles over 12 weeks), with cumulative epirubicin and paclitaxel doses of 360 mg/m2 and 700 mg/m2, respectively. After the completion of preoperative chemotherapy, patients underwent surgical tumor removal with either lumpectomy plus axillary node dissection or modified radical mastectomy. After surgery all patients received three cycles of cyclophosphamide 500 mg/m2, methotrexate 40 mg/m2, and fluorouracil 600 mg/m2 given on days 1 and 8 of each cycle every 4 weeks (CMF). In all patients with a positive hormone receptor status, tamoxifen (20 mg/d for 5 years) was to be initiated after completion of chemotherapy. Patients undergoing lumpectomy were to receive radiotherapy to the breast after completion of chemotherapy. Radiotherapy of the chest wall and supraclavicular regional lymph node area was given when indicated.
(A) Preoperative chemotherapy schema. (B) Complete pathologic tumor response (pCR) rates at time of surgery by treatment arm (intense dose-dense chemotherapy, n = 310; conventionally scheduled chemotherapy, n = 322). Kaplan-Meier (C) disease-free survival curves by treatment categories, (D) overall survival curves by treatment categories, (E) disease-free survival curves by stratum and treatment categories, and (F) overall survival curves by stratum and treatment categories.
Evaluation of Therapy Response
The assessment of tumor response was based on the pCR rate. At surgery, no invasive cancer in the breast was considered to be the primary definition for a pCR. Furthermore, absence of both invasive and in situ disease in the breast and nodes was also compared between treatment arms.
Statistical Methods
The primary end point of the study was the pCR rate. Secondary end points included DFS, OS, and toxicity analysis. A sample size of 646 assessable patients was required to achieve 90% power to significantly detect a doubling of the pCR rate from an anticipated 9% in the standard arm to 18% by dose intensification, on a type I error level of 5% (two sided). Patients were stratified according to having either noninflammatory BC with a large primary tumor (≥ 3 cm) or inflammatory BC and randomly assigned within each stratum. DFS was defined as the time from registration to first documentation of relapse or death due to any cause. OS was the time from the date of registration to the date of death due to any cause. For univariate analysis the log-rank test and Cox proportional hazard model were used to make group comparisons, and Kaplan-Meier curves were utilized to quantify the values of DFS and OS. The pCR rates were compared using the Fisher's exact test. Toxicities were compared using the Wilcoxon rank sum, χ2, or Fisher's exact test. For multivariate analysis, logistic regression was applied in case of pCR rates and the Cox proportional hazard model for survival end points, implementing a backward selection process with a cutoff level of P < .1. Adjustments were made for age, hormone receptor status, tumor size, grade, histology (noninflammatory v inflammatory), and clinical nodal status. However, only univariately significant factors were included in the respective regression models. All P values are two sided. A subset analysis according to HER2, TOP2A, and hormone receptor status will be submitted as a separate article.
RESULTS
Patient Characteristics
Between January 1998 and May 2002 a total of 671 patients with either noninflammatory BC with large primary tumors (≥ 3 cm) or histologically proven inflammatory BC were randomly assigned to a treatment arm. Three patients were later found to be ineligible due to metastatic disease at randomization. The 668 patients who fulfilled the selection criteria provide the basis for this report (Fig 1; Table 1).
Characteristics of Patients and the Tumors at Baseline
Treatment Administration and Dose Intensity
The proportion of women who received all doses of preoperative chemotherapy was 86% in the group receiving IDD chemotherapy and 95% in the control arm. Data were available to calculate dose intensity for 658 patients. The mean (with or without standard deviation [SD]) delivered dose intensity for epirubicin was 70.8 ± 8.2 and 29.6 ± 2.6 mg/m2/week for each treatment arm, respectively (planned dose intensity 75 and 30 mg/m2/week). The mean delivered dose intensity for paclitaxel was 119.9 ± 12.7 and 57.3 ± 4.7 mg/m2/week for each treatment arm, respectively (planned dose intensity 125 and 58 mg/m2/week). Thus, patients receiving IDD chemotherapy had an actual 139% and 109% increase of the epirubicin and paclitaxel dose intensity (mg/m2/week), respectively, compared to patients treated with conventionally scheduled chemotherapy. Nevertheless, patients in the IDD group only had an actual modest 24% increase in the cumulative total epirubicin dose (436.7 ± 49 v 349.7 ± 50.8 mg/m2) and a small 3% increase in the cumulative paclitaxel dose (701.1 ± 123.9 v 679.4 ± 96.7 mg/m2), compared with the control group. Overall, 1,844 courses of IDD chemotherapy were administered and we observed 93 dose reductions (5%; 44 for hematologic, 36 for nonhematologic, five for both, and eight for other reasons) and 305 treatment delays (17%; 77 for hematologic, 63 nonhematologic, nine for both, and 145 for other reasons). In comparison, 1,286 courses of conventionally scheduled preoperative therapy were administered and we observed 9 dose reductions (1%; six for hematologic, one for nonhematologic, and two for other reasons), and 127 treatment delays (10%; 11 for hematologic, 17 nonhematologic, and 89 for other reasons). Of the 69 patients who discontinued the IDD treatment before receiving all planned doses of preoperative chemotherapy, 35 patients (51%) stopped because of adverse events. Of the 43 patients who discontinued conventionally scheduled chemotherapy 10 patients (23%) stopped early because of adverse events. Overall, 1,675 courses of CMF chemotherapy were documented. We observed dose reductions in 29 patients (2%) and treatment delays in 155 (15%) of the CMF cycles. There were no significant differences between both treatment arms regarding dose reduction or treatment delays of CMF chemotherapy.
Efficacy
After preoperative chemotherapy, the pCR rate was statistically significantly higher in the group receiving IDD chemotherapy compared to the control group (18% v 10%; P = .008). This difference retained statistical significance in multivariate analysis when accounting for grade and hormone receptor status (Table 2). Figure 2B shows the pCR rates as defined in materials and methods. The IDD chemotherapy increased the proportion of patients with negative axillary nodes from 43% to 50%, although this difference did not reach statistical significance (P = .120). Overall, breast conserving surgery (BCS) was possible in 369 (58%) of 632 patients. In contrast, the proposed rate of BCS assessed at study entry was estimated to be 25% (Table 1). The overall rate of postoperative wound infections was 5%, postoperative antibiotic treatment was 19%, and secondary resections was 11%, and no differences were seen between treatment groups. IDD chemotherapy was initially associated with a higher rate of BCS compared to the control group (63% v 54%; P = .044); however, 25 patients underwent secondary mastectomies within the first 3 months in the group receiving IDD chemotherapy and 15 in the control group, which reduced the difference in the rate of BCS to 55% versus 50%, respectively (P = .260).
Associations of Prognostic Factors and Treatment With the Pathologic Response Rate
At a median follow-up of 55 months, 215 recurrences had been recorded (92 in the IDD group and 123 in the control group). DFS was significantly longer for the group receiving IDD chemotherapy compared to the control group (HR, 0.71; 95% CI, 0.54 to 0.92; P = .011). The estimated DFS rates for the IDD and the conventional schedules were 76% versus 68% at 3 years and 70% versus 59% at 5 years, respectively (Fig 2C). By the same median follow-up, 123 OS events had been recorded (52 in the IDD group and 71 in the control group). OS was significantly longer in the group receiving IDD chemotherapy compared to the control group (HR, 0.83; 95% CI, 0.69 to 0.99; P = .041). The estimated OS rates for the IDD and control arms were 90% versus 85% at 3 years and 83% versus 77% at 5 years, respectively (Fig 2D). After adjusting for preoperatively assessed nodal status, tumor size, histology (noninflammatory v inflammatory), and hormone receptor status the IDD effect did not reach statistical significance for DFS (HR, 0.76; 95% CI, 0.55 to 1.04; P = .088; Table 3) and borderline statistical significance for OS (HR, 0.82; 95% CI, 0.66 to 1.01; P = .059). In this phase III, multicenter, prospective trial, patients were stratified before randomization according to noninflammatory (≥ 3 cm; n = 567) versus inflammatory BC (n = 101). In this preplanned subset analysis, patients with inflammatory BC had a poor prognosis and did not appear to benefit from IDD therapy (DFS HR, 1.10; 95% CI, 0.62 to 1.94; P = .739; OS HR, 1.25; 95% CI, 0.61 to 2.60; P = .544; Figs 2E, 2F). In contrast, patients with noninflammatory BC significantly benefited from IDD treatment in unadjusted analysis (DFS HR, 0.65; 95% CI, 0.48 to 0.88; P = .005; OS HR, 0.77; 95% CI, 0.63 to 0.95; P = .013) and when adjusted for size, nodal status, and hormone receptor status (DFS HR, 0.65; 95% CI, 0.46 to 0.92; P = .015; OS HR, 0.79; 95% CI, 0.61 to 0.94; P = .034; Table 4).
Association of Prognostic Factors and Treatment With DFS and OS in Univariate and Multivariate Analysis for All Patients (large primary tumor size or inflammatory breast cancer)
Association of Prognostic Factors and Treatment With DFS and OS in Univariate and Multivariate Analysis for Patients With Noninflammatory Breast Cancer and Large Primary Tumor Size
Safety
Detailed information on toxicity is reported for the preoperative chemotherapy (Table 5). WHO grade 1 or worse nonhematologic toxic effects were more common with the IDD regimen than with the conventionally scheduled regimen (stomatitis 71% v 51%, P < .001; nausea and vomiting 80% v 70%, P < .001; fever 25% v 12%, P < .001; allergic reactions, 12% v 7%, P = .041; pain 79% v 56%, P < .001; skin toxicity 42% v 19%, P < .001; infection 40% v 27%, P < .001; and neurosensory toxicity 81% v 55%, P < .001). Grade 1 or worse anemia and thrombocytopenia were more common following the IDD regimen (84% v 44% and 33% v 6%, respectively, both P < .001). However, grade 1 or worse neutropenia was similar between both treatment regimens (72% v 84%, P = .571). The rate of grade 3 and 4 neutropenia (54% v 59%, P = .340) or infections (5% v 4%, P = .580) were similar between treatment regimens. However, grade 3 and 4 thrombocytopenia was more common with the IDD regimen (5% v 1%, P = .004). Grade 3 and 4 stomatitis (11% v 2%, P < .001) and nausea or vomiting (9% v 5%, P = .030) were also more common in the IDD group. Transfusions became necessary in 32 (10%) of 328 patients receiving IDD chemotherapy compared to none with the conventionally scheduled regimen. Severe post chemotherapy sensory neurotoxicity (grade 3 and 4) was more frequent in the group receiving IDD chemotherapy (7% v 1%, P < .001). No instances of acute myelogeneous leukemia or myelodysplastic syndrome were reported. Grade 3 congestive heart failure was observed in one patient treated with IDD chemotherapy and in two patients treated with conventionally scheduled chemotherapy.
Toxicity per Patient by WHO Grade
DISCUSSION
In this randomized trial of preoperative chemotherapy in women with operable high-risk primary BC, IDD chemotherapy significantly improved the pCR rate, DFS, and OS in patients with noninflammatory BC with large primary tumors (≥ 3 cm) when compared to patients receiving conventionally scheduled preoperative chemotherapy. Patients were stratified for having either noninflammatory or inflammatory BC. Although the study was underpowered for a meaningful comparison of treatment effects in the inflammatory BC group (N = 101), we found no indication in this type of BC that IDD chemotherapy may have a beneficial effect similar to that seen in patients with noninflammatory BC. Consistent with these findings the treatment effect of IDD retained statistical significance after adjustment for clinical tumor size, nodal status, and hormone receptor status only for noninflammatory BC and not when inflammatory BC patients were included in the analysis. The pCR rates seen in our study are somewhat lower than those reported in earlier studies, such as the NSABP B-27 study.3 However, this may be due to differences in patient populations, as reflected by the difference in the rate of patients with clinically positive axillary lymph nodes between our study and the NSABP B-27 study (54% v 30%, respectively). Furthermore, a direct comparison between both trials is limited by the fact that different methods were used for assessing primary tumor size in each study.
Most patients completed the planned number of IDD preoperative chemotherapy cycles (86% in the IDD chemotherapy group and 95% in the control group). However, IDD chemotherapy was clearly more toxic and less tolerable compared to the conventionally scheduled chemotherapy. Nevertheless, grade 3 or 4 nonhematologic toxicities, and thrombocytopenia, were manageable with standard supportive measures, and no treatment-related mortality or no long-term treatment related toxicity has been reported yet. Grade 3 or 4 neutropenia was common in both groups, however, grade 3 infections were rare and the incidence was comparable between treatment groups. At a median follow-up of 55 months, the rate of congestive heart failure was < 1% in both the IDD chemotherapy group and the control group, which is consistent with the reported incidence associated with epirubicin-based adjuvant chemotherapy.14 In summary, we do believe that the increased toxicities and risks associated with IDD chemotherapy are reasonable in light of the expected benefits in high-risk primary BC patients.
Despite a similar cumulative epirubicin and paclitaxel dose in both treatment groups shortening of intertreatment intervals with sequential application of dose-escalated epirubicin and paclitaxel allowed a 139% and 109% increase in dose intensity (mg/m2/week) of epirubicin and paclitaxel, respectively. This increase is considerably larger than that described in the Cancer and Leukemia Group B (CALGB) study 9741, which compared concurrent doxorubicin and cyclophosphamide followed by paclitaxel with sequential doxorubicin, paclitaxel, and cyclophosphamide at standard dose levels given either every 3 weeks or at 2-week intervals in lymph node-positive BC.9 In that study, dose-dense bi-weekly therapy resulted in a 50% increase of dose intensity and lead particularly in patients with the hormone receptor–negative tumors to a statistically significant but modest improvement of DFS and OS compared with therapy administered once every 3 weeks.9,15,16 In a similar concept, the Italian Gruppo Oncologico Nord Ovest-Mammella Intergruppo (GONO-MIG) trial compared six cycles of fluorouracil, epirubicin, and cyclophosphamide administered every 3 weeks versus the same regimen administered every 2 weeks with filgrastim support in patients with lymph node-positive or high-risk lymph node-negative BC.17 Although both studies had a comparable 50% increase in dose intensity the Italian study only showed a nonsignificant trend toward improved DFS and OS for women who received the dose-dense regimen, which may be due to differences in sample size (1,214 v 2,005 patients, respectively). Similarly, the role of increased dose intensity, obtained by increasing the single dose per cycle has been controversial. The CALGB study 9344 found no benefit from increased doses of doxorubicin (60, 75, or 90 mg/m2 per cycle).18 In contrast, the French Adjuvant Study Group trial 05 demonstrated a beneficial effect among lymph node-positive patients given 100 mg/m2 epirubicin instead of 50 mg/m2.6 This difference in outcome may be explained by the fact that the CALGB 9344 trial tested a moderate 50% increase in dose intensity as opposed to 100% in the French study. Preliminary results of a German adjuvant trial in 1,284 high-risk primary BC patients with four or more positive axillary lymph nodes confirm a beneficial effect of an adjuvant IDD regimen with biweekly epirubicin, paclitaxel, and cyclophosphamide on DFS and OS.19 Nevertheless, it is important to point out that in the adjuvant treatment of breast cancer, multiple studies have now shown that high-dose therapy with dose escalation to the limits of tolerability has not improved outcomes20 and that improvement in BC survival will more likely be achieved by inclusion of more effective targeted therapies with appropriate use of predictive markers. However, dose intensification, be it through dose escalation6 or dose density,9 or both8,15 may remain helpful to optimize treatment effects when using anthracyclins and paclitaxel as agents for preoperative or adjuvant chemotherapy for high-risk primary BC.
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
Employment or Leadership Position: None Consultant or Advisory Role: Michael Untch, Bristol-Myers Squibb (C), Pfizer (C), Amgen (C); Christoph Thomssen, Amgen (C), Bristol-Myers Squibb (C), Pfizer (C); Nadia Harbeck, Bristol-Myers Squibb (C), Pfizer (C); Andreas Schneeweiss, Amgen (C); Rolf Kreienberg, AstraZeneca (C), Novartis (C); Hans-Joachim Lück, Amgen (C), Bristol-Myers Squibb (C), Pfizer (C); Gottfried E. Konecny, Genentech (C), sanofi-aventis (C) Stock Ownership: None Honoraria: Michael Untch, Bristol-Myers Squibb, Pfizer, Amgen; Volker Möbus, Amgen, Bristol-Myers Squibb, GlaxoSmithKline, Novartis, AstraZeneca, Pfizer, Roche; Christoph Thomssen, Amgen, Bristol-Myers Squibb, Pfizer; Ingo Bauerfeind, Amgen, Bristol-Myers Squibb, Pfizer; Nadia Harbeck, Amgen, Bristol-Myers Squibb, Pfizer; Andreas Schneeweiss, Amgen; Hans-Joachim Lück, Amgen, Bristol-Myers Squibb, Pfizer; Gottfried E. Konecny, Amgen, Bristol-Myers Squibb, sanofi-aventis, Genentech, Roche Research Funding: Michael Untch, Bristol-Myers Squibb, Pfizer, Amgen; Volker Möbus, Amgen, Bristol-Myers Squibb, Roche, Johnson & Johnson; Christoph Thomssen, Amgen, Bristol-Myers Squibb, Pfizer; Gottfried E. Konecny, Amgen, Bristol-Myers Squibb, GlaxoSmithKline Expert Testimony: None Other Remuneration: None
AUTHOR CONTRIBUTIONS
Conception and design: Michael Untch, Walther Kuhn, Gottfried E. Konecny
Provision of study materials or patients: Michael Untch, Volker Möbus, Walther Kuhn, Bernd Rudolph Muck, Christoph Thomssen, Ingo Bauerfeind, Nadia Harbeck, Christoph Werner, Annette Lebeau, Andreas Schneeweiss, Stephen Kahlert, Franz von Koch, Karl Ulrich Petry, Diethelm Wallwiener, Rolf Kreienberg, Ute-Susann Albert, Hans-Joachim Lück, Fritz Jänicke
Collection and assembly of data: Michael Untch, Volker Möbus, Walther Kuhn, Christoph Thomssen, Ingo Bauerfeind, Nadia Harbeck, Annette Lebeau, Andreas Schneeweiss, Stephen Kahlert, Franz von Koch, Karl Ulrich Petry, Diethelm Wallwiener, Rolf Kreienberg, Ute-Susann Albert, Hans-Joachim Lück, Axel Hinke, Fritz Jänicke, Gottfried E. Konecny
Data analysis and interpretation: Michael Untch, Axel Hinke, Gottfried E. Konecny
Manuscript writing: Michael Untch, Gottfried E. Konecny
Final approval of manuscript: Michael Untch, Volker Möbus, Walther Kuhn, Christoph Thomssen, Ingo Bauerfeind, Nadia Harbeck, Christoph Werner, Annette Lebeau, Andreas Schneeweiss, Stephen Kahlert, Franz von Koch, Karl Ulrich Petry, Diethelm Wallwiener, Rolf Kreienberg, Ute-Susann Albert, Hans-Joachim Lück, Axel Hinke, Fritz Jänicke, Gottfried E. Konecny
Footnotes
-
Supported by Amgen, Bristol- Myers Squibb, and Pfizer, Germany.
-
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
- Received September 24, 2008.
- Accepted January 23, 2009.
