Gemcitabine, Anthracycline, and Taxane Combinations for Advanced Breast Cancer

The treatment of advanced breast cancer remains a formidable challenge for the oncology community, as few therapeutic modalities have been developed that prolong overall survival. Nevertheless, the introduction of anthracyclines in the 1980s and taxanes in the 1990s produced significant advances in the treatment of the disease. Polychemotherapy protocols that include an anthracycline (doxorubicin or epirubicin(Drug information on epirubicin) [Ellence]) and a taxane (paclitaxel or docetaxel(Drug information on docetaxel) [Taxotere]) have contributed to prolonged cumulative survival in cohort analyses at The University of Texas M. D. Anderson Cancer Center over the past 2 decades.[ 1] Paclitaxel/Anthracycline Combinations The combination of paclitaxel and doxorubicin(Drug information on doxorubicin) has produced an overall response rate of 94%, with a 41% complete response rate (95% confidence interval [CI] = 24%-59%) and a 53% partial response rate in 32 patients.[ 2] These favorable results, however, were complicated by congestive heart failure, which occurred in more than 20% of treated patients. The combination of paclitaxel(Drug information on paclitaxel) and doxorubicin also proved to be superior to a regimen of fluorouracil(Drug information on fluorouracil) (5- FU), doxorubicin (Adriamycin), and cyclophosphamide(Drug information on cyclophosphamide) (Cytoxan, Neosar), or FAC, as first-line therapy for metastatic breast cancer by producing a significantly better overall response rate (68% vs 55%, P = .032), significantly longer time to disease progression (8.3 vs 6.2 months, P = .034), and significantly longer overall survival (23.3% vs 18.3%, P = .013). The incidence of cardiotoxicity was comparable, as a similar percentage of patients in each treatment group experienced a decrease in left ventricular ejection fraction. Grade 3/4 neutropenia occurred significantly more often in the paclitaxel/doxorubicin arm than in the FAC arm (89% vs 65%, P < .001). The incidence of febrile neutropenia was comparable in both treatment arms.[3] These data were challenged by the results of other studies showing that a combination of paclitaxel and doxorubicin was not significantly better than sequential monotherapy with both drugs,[4] or polychemotherapy with either doxorubicin and cyclophosphamide[ 5] or epirubicin and cyclophosphamide.[ 6] Thus, although paclitaxel/anthracycline combinations have shown considerable efficacy in terms of response rates, the number and percentage of complete remissions remain low, and efficacy translated into improved overall survival in only one study. Furthermore, the combination of paclitaxel and anthracyclines may be complicated by the occurrence of congestive heart failure due to the reduced clearance of anthracyclines and their metabolites following exposure to paclitaxel.[7] Finally, analysis of other studies strongly suggests that survival rates may be confounded by the switch from a nontaxane to a taxanecontaining regimen.[8] Thus, it remains to be determined whether it is possible to increase rates of complete remission and overall survival by first-line treatment without compromising safety. Gemcitabine/Paclitaxel Combinations To pursue this goal, a variety of newer antitumor drugs have been evaluated in combination with paclitaxel as potential replacements for anthracyclines. Due to its low toxicity and favorable results in 10 separate phase II studies, gemcitabine(Drug information on gemcitabine) (Gemzar) has emerged as one of the major candidates in this endeavor. Among 280 patients recruited into these 10 studies, overall response rates ranging from 12.5% to 42% were achieved with gemcitabine as either first-[9- 11] or second-line therapy.[12-16] Of note, gemcitabine has been shown to produce responses even when used as third-line treatment in patients previously exposed to anthracyclines and taxanes.[13,17,18] The quest for improved efficacy with low toxicity led to phase II trials combining gemcitabine with paclitaxel as either firstline or salvage therapy in a total of 159 patients with advanced or metastatic breast cancer.[19-21] Response rates as high as 52% were achieved, with time to disease progression ranging from 7.0 to 14.5 months. Gemcitabine has also been shown not to interfere with the pharmacokinetics or pharmacodynamics of either paclitaxel or epirubicin, suggesting a favorable toxicity profile when combined with these agents.[7] Gemcitabine/Anthracycline Combinations Two phase II studies have been reported that combined gemcitabine with doxorubicin[22] and epirubicin[ 23] in women with advanced or metastatic breast disease. Overall response rates were 75% with gemcitabine/ doxorubicin and 60% with gemcitabine/epirubicin. Both regimens were associated with low treatment-related toxicity. These observations and the favorable outcomes reported with gemcitabine/paclitaxel regimens led to the idea of combining gemcitabine with both an anthracycline and paclitaxel. Three-Drug Regimens Containing Gemcitabine Two different three-drug regimens have been evaluated in patients with metastatic breast cancer. One regimen consists of gemcitabine at 2,500 mg/m², doxorubicin at 30 mg/m², and paclitaxel (Taxol) at 135 mg/m² (GAT).[24] The other regimen incorporates gemcitabine at 1,000 mg/m², epirubicin at 90 mg/m², and paclitaxel at 175 mg/m² (GET).[25,26] The GAT regimen administered at 14-day intervals produced an overall response rate of 82.9% (95% CI = 67.9%-92.8%), including a 43.9% complete response rate and a 39% partial response rate, as well as an impressive response rate in patients with visceral metastases (lung metastases, 72%; liver metastases, 50%). The median duration of response was 14.1 months, median time to disease progression 13.9 months; and median overall survival 26.2 months.[27] Significant toxicity (grades 3/4) included neutropenia, which occurred in 10% of all cycles in 44% of patients, and thrombocytopenia, which occurred in 1% of cycles in 7% of patients. In the GET regimen, gemcitabine is administered on days 1 and 4, epirubicin on day 1, and paclitaxel on day 1 every 21 days.[25,26] The GET regimen led to a 31% complete remission rate and a 61% partial remission rate, resulting in an overall response rate of 92% (95% CI = 77.5%-98.2%) in 36 patients. Six courses of GET were followed by high-dose chemotherapy in 25 patients who responded to primary treatment resulting in conversion to a better response in 39% of patients. Ultimately, complete remissions occurred in 58% of 36 patients, and the overall response rate increased to 96%.[26] CECOG Study BM1 Based on the favorable results achieved with the GET regimen, the Central European Cooperative Oncology Group (CECOG) initiated a controlled, prospective, randomized, multicenter phase III trial of GET vs a three-drug regimen containing 5-FU, epirubicin, and cyclophosphamide (FEC) in October 1999. The GET regimen was administered as described above; the FEC regimen consisted of 5-FU at 500 mg/m², epirubicin at 90 mg/m², and cyclophosphamide at 500 mg/m², all given on day 1 of a 21-day cycle. FEC was chosen as the comparator arm owing to its response rate of 45% to 54%, response duration of 11 to 14 months, time to disease progression of 9 to 11 months, and overall survival of 15 to 20 months.[28-30] These results have led to the acceptance of FEC as standard treatment for advanced breast cancer in many parts of the world, including countries in eastern and southeastern Europe. The first interim toxicity analysis of the trial was presented in 2001.[31] The final data are scheduled for presentation at the annual meeting of the American Society of Clinical Oncology in 2003.[32] A total of 260 patients were recruited between October 1999 and February 2002 by the clinical investigators listed in the Appendix. Inclusion criteria consisted of the following: histologically confirmed metastatic breast cancer, prior adjuvant chemotherapy restricted to non-anthracycline- containing regimens, the present treatment constituted first-line chemotherapy for metastatic disease (previous hormonal treatment was permitted), presence of measurable disease, patient age between 18 and 75 years, an Eastern Cooperative Oncology Group performance status of 0 to 1, life expectancy exceeding 12 weeks, and adequate hematologic, renal, and cardiac function. The primary objective of the study is time to disease progression; secondary objectives include response rate, survival, quality of life, and treatment toxicity. An interim toxicity analysis was performed in February 2001 after the recruitment of 123 patients, 57 of whom were randomized to GET and 66 to FEC. At that time, 84 patients whose dose intensity exceeded 95% in both treatment arms of the study were evaluable. The favorable results achieved in this interim toxicity analysis (Table 1) encouraged continued patient recruitment until the original goal of 260 patients had been achieved. By the time 117 more patients were recruited, 52 had completed treatment. Thirty-nine patients interrupted treatment for various reasons, including study drug toxicity (n = 8), disease-related death (n = 3), death due to a non–disease-related cause (n = 1), the physician's decision (n = 10), the patient's decision (n = 8), loss to follow-up (n = 4), or other reasons (n = 5). As of this writing, demographic data on 260 patients ranging in age from 29 to 74 years are available.[32] All but three patients allocated to GET or FEC treatment (98.8%) took at least one dose of the study drug. Median follow-up time was 11.48 months for patients in the GET arm and 10.30 months for those in the FEC arm (P = .702). Fifty-eight patients (22.66%), including 24 (19.35%) on GET and 34 (25.76%) on FEC, died during the study. Demographic analysis of all included patients showed a balanced distribution between treatment arms (GET, 124 patients; FEC, 135 [P = .573]), as well as the distribution of other variables in patients allocated to the two treatment arms, including initial pathologic diagnosis and disease stage (P = .632 and P = .230, respectively), duration of disease-free interval (P = .938), hormone-receptor status (estrogen receptors, P = .836; progesterone(Drug information on progesterone) receptors, P = .825), ECOG performance status (P = .411), distribution of metastases to visceral organs including liver and lung (P = .111), and prior treatment (hormonal therapy, P = .252; adjuvant chemotherapy, P = .619). A slight imbalance in distribution was found in menopausal status, in that fewer premenopausal patients were randomized to the GET arm (14.52%) than to the FEC arm (28.15%; P = .035). Detailed demographic data are presented in Table 2. Data on the primary and secondary end points of the study will be presented shortly.[32] Conclusions Based on currently available data, we can conclude that the toxicity profile of either the GET or FEC regimen is acceptable and that both are well tolerated. Myelotoxicity and peripheral polyneuropathy represented major side effects of GET, although both were of limited clinical relevance. Interim toxicity analyses showed that the vast majority of patients completed treatment within the framework of the CECOG trial, with toxicity constituting a rare reason for treatment interruption. While the final data of this phase III study are expected shortly, it appears that the inclusion of gemcitabine in a triple-drug chemotherapy regimen offers potentially promising efficacy as first-line treatment with limited toxicity in patients with advanced or metastatic breast cancer. Appendix Investigators of the GET vs FEC study (CECOG BM1), listed according to the number of recruited patients: Mrsic Zrinka, MD, Department of Medical Oncology, University Hospital, Zagreb, Croatia; Semir Beslija, MD, Institute of Oncology, Sarajevo, Bosnia; Jacek Jassem, MD, Department of Oncology and Radiotherapy, Medical University of Gdansk, Poland; Christoph Wiltschke, MD, Clinical Division of Oncology, Department of Medicine I, University Hospital, Vienna, Austria; Zsuzsanna Kahan, MD, Onkotherapias Klinika, Szeged, Hungary; Mislav Grgic, MD, University Hospital Rebro, Zagreb, Croatia; Valentina Tzekova, MD, University Hospital "Queen Joanna," Sofia, Bulgaria; Moshe Inbar, MD, Oncology Department, Sourasky Medical Center, Tel Aviv, Israel; Jozica Cervek, Institute of Oncology, Ljubljana, Slovenia; Constanta Timcheva, MD, National Oncologic Center, Sofia, Bulgaria; Janos Szanto, MD, Institute of Oncology, Debrecen Medical University, Debrecen, Hungary; Maria Wagnerova, MD, Fakultna Nemocnica Luisa Pasteura, Kosice, Slovakia; Stanislav Spanik, MD, Nemocnica Sv. Alzbety, Narodny Onkologicky Ustav, Bratislava, Slovakia; Nicolae Ghilezan, MD, Institutul Oncologic Cluj, Cluj-Napoca, Romania; Janusz Pawlega, MD, Klinika Onkologii CMUJ, Krakow, Poland; Damir Vrbanec, MD, Department of Pathophysiology, University Hospital, Zagreb, Croatia; Tamas Pinter, MD, Petz Aladar County Hospital, Györ, Hungary; Jerzy Zaluski, MD, Department of Chemotherapy, Great Poland Cancer Center, Poznan, Poland; Antoaneta Tomova, MD, Regional Dispensary for Oncology, Plovdiv, Bulgaria; Nil Molinas Mandel, MD, Department of Medical Oncology, Cerrahpa A Medical School, Istanbul University, Turkey; Jerzy Tujakowski, MD, Regional Oncology Center, Bydgoszcz, Poland; Ivan Koza, MD, National Cancer Institute, Bratislava, Slovakia; Milan Kuta, MD, Department of Oncology and Radiotherapy, Hospital Chomutov, Czech Republic; Nazan Günel, MD, Department of Medical Oncology, Faculty of Medicine, Gazi University, Ankara, Turkey; Osman Manavoglu, MD, Faculty of Medicine, Uludag University, Bursa, Turkey; Lubos Petruzelka, MD, Department of Oncology, Charles University, Prague, Czech Republic; Adi Shani, MD, Oncology Institute, Kaplan Medical Center, Rehovot, Israel; Yilmaz Ugur, MD, Department of Medical Oncology, Faculty of Medicine, University of Izmir, Turkey; and Erikisi Melek, MD, Faculty of Medicine, Cukurova University, Adana, Turkey. CECOG Head Office: Christoph C. Zielinski, MD, Coordinator; Thomas Brodowicz, MD, Co-Coordinator; Irmgard Resch, MD, Clinical Research Associate; Margit Landsgesell, Administrative Director; and Dagmar Just, Administrative Assistant. CRO: Dr. R. Kobelt, Innopharm. Contacts: cecog@akhwien. ac.at, www.cecog.org