|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2001 American Society for Clinical Oncology Dose-Dense Doxorubicin, Docetaxel, and Granulocyte Colony-Stimulating Factor Support With or Without Tamoxifen as Preoperative Therapy in Patients With Operable Carcinoma of the Breast: A Randomized, Controlled, Open Phase IIb StudyByFrom the Department of Gynecology and Obstetrics, Goethe University, Frankfurt am Main; Red Cross Gynecological Hospital, Munich; Department of Gynecology, Charité Hospital, Humboldt University, Berlin; Department of Gynecology, University of Kiel, Kiel; Henrietten-Stift Gynecological Hospital, Hannover; am Berg Gynecological Hospital, Stuttgart; Department of Gynecology, University of Münster, Münster; Department of Radiotherapy, University of Magdeburg, Magdeburg; Bayreuth Gynecological Hospital, Bayreuth; and Department of Medical Biometry and Statistics, University Hospital Freiburg, Germany. Address reprint requests to Gunter von Minckwitz, MD, PhD, Zentrum der Frauenheilkunde und Geburtshilfe, J.W. Goethe-Universität Frankfurt am Main, Theodor-Stern Kai 7, D-60590 Frankfurt am Main, Germany; email: minckwitz{at}em.uni-frankfurt.de
PURPOSE: To investigate the effect of adding tamoxifen to a preoperative dose-dense doxorubicin and docetaxel regimen on the pathologic response of primary operable breast cancer.
PATIENTS AND METHODS: Patients (tumor size RESULTS: Within 14 months, 250 patients were included in the study at 56 centers. Of 992 planned cycles, 97.9% were administered. Pathologically complete remission (pCR) with no detectable viable tumor cells was achieved in 9.7%. There was a nonsignificant difference of -1.2% in favor of ADoc, with a 95% confidence interval of -8.6% to 6.2%. A further 2.4% had only noninvasive tumor residues, and 13.8% had focal invasive residues. Complete and partial responses detected by palpation were observed in 28.9% and 52.4%, respectively. The response rates (complete and partial) by best appropriate imaging methods were 77.5% and 67.5% for ADocT and ADoc, respectively. Breast conservation was possible in 68.8% of the patients. A tendency toward more frequent toxic events was observed with ADocT treatment. Significant predictors of pCR to chemotherapy were negative lymph node and negative estrogen receptor status. CONCLUSION: A dose-dense regimen of ADoc with G-CSF offers high compliance, moderate toxicity, and rapid efficacy as a form of preoperative chemotherapy in operable breast cancer. Concurrent treatment with tamoxifen for 8 weeks could not improve the pathologic response rate.
PREOPERATIVE chemotherapy has been shown to be as effective as postoperative chemotherapy in primary operable breast cancer.1,2 Patients with complete remission of the primary tumor have a better prognosis than patients with a partial remission, stable disease, or progressive disease.3 This has led to the hypothesis that local tumor response corresponds to the response of distant "micrometastases" and can therefore be used as a surrogate parameter for clinical outcome. Preoperative randomized trials have the advantage of including a relatively homogeneous population of untreated tumor cells, while the therapeutic effect can be evaluated precisely and at an early stage during pathologic examination at the time of surgery. Long-term follow-up is not essential for assessing efficacy, as in adjuvant studies. Preoperative chemotherapy can be used as an in vivo chemosensitivity test, and the efficacy of the treatment is evident at an early stage to both the physician and patient. In addition, studies can be designed to allow evaluation of molecular predictive markers, as tumor tissue can be obtained both before and after treatment. This setting is ideal for rapid investigation of new compounds in early-stage disease.4 We previously reported a phase IIa trial comparing conventional and dose-dense combination regimens of doxorubicin and docetaxel in primary breast tumors of 3 cm or more in diameter.5 The aims of that study were, first, to combine the two most active groups of cytotoxic agents, which have shown no cross-resistance and different toxicity profiles6 and, second, to reduce the time between diagnosis and surgery to a minimum of 8 weeks. The clinical response rate observed with the dose-dense 2-week schedule was 93%, but the pathologically complete response (pCR) rate in this subset of 42 patients was low, at 5%. However, the toxicity was acceptable, and none of the patients had to discontinue the treatment. The German Preoperative Adriamycin-Docetaxel (GEPARDO) group therefore initiated a prospective, randomized, and controlled multicenter trial to evaluate the safety and efficacy of the dose-dense schedule further and to improve response rates more by administering tamoxifen at the same time. The rationale for this study was based on (1) in vitro data showing a synergistic effect of a combination of docetaxel and tamoxifen,7 (2) evidence that tamoxifen may be able to modify multidrug resistance,8 and (3) the positive results of the Early Breast Cancer Trialists’ Collaborative Group on the combination of adjuvant chemoendocrine treatment in steroid receptor–positive patients.9 All patients, irrespective of their receptor status, had to be randomized to evaluate the predictive value of the steroid receptors as well as other factors on the effect of tamoxifen. The primary aim of the present study was to determine whether the addition of tamoxifen to preoperative dose-dense doxorubicin and docetaxel combination chemotherapy is capable of improving the rate of pCRs. A pCR was defined as no microscopic evidence of viable tumor cells in the resected breast specimen.10 Secondary aims were the comparison of partial and complete clinical responses detected by palpation or by the best appropriate imaging method, rates of breast-conserving surgery, and the frequency of grade 3 and 4 toxicity in both arms of the study. The predictive values of menopausal status, tumor size, nodal status, grading, and estrogen and progesterone receptor content for achieving pCR were also evaluated.
Patient Population A patient had to meet the following inclusion criteria to be enrolled onto the trial: unilateral primary carcinoma of the breast, confirmed histologically by core-cut needle or incisional biopsy (fine-needle aspiration was not sufficient); tumor measurable in two dimensions, either by mammography, breast ultrasound, or breast magnetic resonance imaging (MRI); primary tumor  3 cm in largest diameter by palpation or imaging method (in patients with multifocal or multicentric breast cancer, the largest lesion was measured); no evidence of distant metastases (as confirmed by chest radiography, liver ultrasound, and bone scintigraphy); age between 18 and 70 years; life expectancy of at least 10 years, disregarding the diagnosis of cancer; Karnofsky index of 70%; adequate hematologic, renal, and hepatic function (WBC count > 4,000, platelet count > 100 000, and bilirubin, serum creatinine, and transaminase levels within the normal range); history or electrocardiographic evidence of normal cardiac function without or with medication, confirmed by echocardiography or multiple gated acquisition (MUGA) scan; negative pregnancy test and appropriate nonhormonal contraception in fertile women; and written informed consent and presumed compliance on the part of the patient. Patients were excluded from participation if there was evidence of locally advanced cancer (stage T4) or bilateral, metastatic, or inflammatory breast cancer; if they had previously been treated for breast cancer, including surgery, radiotherapy, and cytotoxic or endocrine treatments (surgical diagnostic procedures were allowed); if there had been a previous malignancy other than breast cancer or noninvasive breast lesions with a disease-free interval of less than 10 years; if they had previously undergone cytotoxic treatment for any condition; if there was preexisting neurotoxicity greater than grade 2 (World Health Organization criteria), active infection, or other significant illness that might influence the tolerability of the treatment; if they were currently receiving treatment with sex hormones (treatment had to be discontinued before the start of systemic therapy); or if there was psychiatric illness or drug addiction that would prevent them from providing informed consent. Participating centers had to confirm that the trial was conducted according to the protocol recommendations and had to apply for approval from an ethics committee. Source data verification was carried out at bimonthly visits by an external clinical research organization (SKM, Wiesbaden, Germany).
Study Treatment Antibiotic prophylaxis (eg, trimethoprim and sulfamethoxazole) could be given to patients with an absolute neutrophil count less than 0.5 x 109/L. Additional antiemetic treatment could be prescribed if needed.
Dose Modification
Surgery and Radiotherapy
Trial Assessments All patients had to undergo core-cut needle or incisional biopsy. The diagnosis of invasive breast carcinoma was confirmed and the nuclear grade was assessed by histopathologic evaluation. Estrogen and progesterone receptor content was determined using immunohistochemistry. Further examinations included a gynecologic examination, ultrasound scan of the liver and small pelvis, chest radiography (posteroanterior and lateral), bone scintigraphy, electrocardiography and echocardiography or MUGA radionuclide ventriculography scanning, and the following laboratory tests: hemoglobin, WBC count including differential blood count, platelets, bilirubin, serum creatinine, alkaline phosphatase, AST, ALT, and pregnancy test (if appropriate). Before each cycle and before surgery, the breast tumor and axillary lymph nodes were measured by palpation, toxicities were documented, and a weekly differential blood count and biweekly testing of bilirubin, serum creatinine, alkaline phosphatase, AST, and ALT were performed. Electrocardiography and echocardiography or MUGA scanning, as well as the best appropriate imaging method, were repeated before surgery.
The type of surgery, classification of the histologic response using the criteria proposed by Sinn et al10 (grade 0, no effect; grade 1, resorption and tumor sclerosis; grade 2, minimal focally invasive residues of
Statistical Evaluation
For sample size calculation, a pCR rate of approximately 10% was assumed for the group treated by chemotherapy alone,9,11 and a rate of approximately 25% was assumed for the group receiving chemoendocrine treatment. The null hypothesis that pCR rates might be identical under chemotherapy and chemoendocrine treatment was tested at a significance level of alpha = 5%. The power to detect a difference of 10% versus 25% had to be no less than 1 - beta = 80%. Thus, using a two-sided
Compliance and treatment outcomes were analyzed according to treatment as randomized, using an intent-to-treat approach. Statistical treatment comparisons were performed with respect to pCR rates, partial or complete remission rates by best appropriate imaging methods, and breast conservation rates. This was done by calculating a two-sided 95% confidence interval (CI) for the treatment difference using normal approximation, which is equivalent to the All patients who received at least one dose of chemotherapy were assessable for toxicity. Toxicities are reported descriptively as summaries of the type and grade according to the treatment actually given, and no statistical tests were performed.
Baseline Results Two hundred fifty patients were recruited into the trial between April 1998 and June 1999 at 56 participating centers all over Germany. The projected sample size of 200 was exceeded, in order to ensure that the subsequent study would be able to start directly after closure of the trial reported here. One hundred twenty-two patients were randomized to receive chemoendocrine therapy (ADocT), and 128, to chemotherapy alone (ADoc). One patient randomized to ADoc was withdrawn from the study immediately after randomization, as a revised pathology report showed that she did not have breast cancer but chronic mastitis. Another patient in the same arm declined preoperative chemotherapy (no postrandomization data are available). Due to a mistake, one patient was randomized twice, first to ADoc and then to ADocT. The second randomization result was ignored, and the patient received ADoc. The progress of patients throughout the trial is indicated in Table 1.
In addition to the one patient without breast cancer, minor violations of entry criteria (incomplete screening, screening completed after randomization, laboratory values out of range) were found for 42 patients. A neutrophil count was missing for 43 patients. The informed consent form was signed after randomization by seven patients who had consented orally before the start of the study. Randomization to treatment with or without tamoxifen was performed shortly after the application of the first cycle of ADoc in 10 patients. In the 248 patients for whom postrandomization data were available, the median age at diagnosis was 48 years (range, 27 to 67 years), with 46.4% of the patients being peri- or postmenopausal. Median body surface area before the onset of chemotherapy was 1.74 m2 (range, 1.38 to 2.32 m2). The Karnofsky index was 70% or higher in all patients and 90% to 100% in 96.3% of the patients. The median palpable maximum tumor diameter was 4.0 cm (range, 0 to 19 cm), with 23.4% of the tumors being larger than 5 cm. The median surface area was 15 cm2 (range, 0 to 34.2 cm2) based on palpation. A total of 126 patients (50.8%) had no clinically enlarged axillary lymph nodes. In 189 patients (76.2%), ultrasound scanning was indicated as the best appropriate imaging method for assessing the breast tumor at the time of randomization. In 44 patients (17.7%), mammography was considered best, and in 15 patients (6.0%), MRI or computed tomography was preferred. The median surface area based on ultrasound was 8.6 cm2 (0.7 to 34.2 cm2, n = 236), on mammography, 10.5 cm2 (1.4 to 100 cm2, n = 176), and on MRI or computed tomography, 11.3 cm2 (1.1 to 300 cm2, n = 46). The diagnosis was ascertained using core-cut needle biopsy in 93.5% of the patients and by incisional biopsy in 6.5%; 76.4% of the patients were diagnosed as having a ductal invasive lesion, 13.8% as having a lobular invasive lesion, and 9.8% as having other subtypes. The tumors were not classified in two patients. Histologic grading was determined in 218 tumors: 4.6% were found to be of grade 1, 49.1% of grade 2, and 46.3% of grade 3. Estrogen and progesterone receptor status was assessed in 196 and 195 patients, and the findings were positive in 59.2% and 54.9%, respectively. All characteristics were found to be well balanced in the two treatment arms, using logistic regression analysis (Table 2).
Compliance A total of 481 (98.6%) of 488 planned cycles in the ADocT arm and 490 (97.2%) of 504 planned cycles in the ADoc arm were administered. The numbers of patients per randomized arm in whom there was a divergence from four cycles are shown in Table 1. One patient per arm received five cycles for organizational reasons. Another patient in the ADoc arm received additional epirubicin and cyclophosphamide during the third cycle of ADoc, after which chemotherapy was discontinued. The second, third, or fourth ADoc cycle was postponed for more than 2 days 25 times (7.0%) in 23 patients allocated to ADocT and 24 times (6.6%) in 22 patients randomized to ADoc alone. An impermissible dose reduction to less than 90% of the planned dose per cycle occurred in two patients, one by chance and one because of elevated liver enzymes after the third cycle. Tamoxifen was started after the onset of the first cycle of ADoc in three patients and interrupted or stopped before the end of chemotherapy in three patients, and two women randomized to ADocT received no tamoxifen. None of the patients allocated to ADoc alone received tamoxifen. Surgery was performed more than 28 days after administration of the last cycle of chemotherapy in 26 patients in each arm, respectively. In 32 cases, this was due to organizational reasons or patient’s wish. Surgery was performed earlier than 14 days after the last drug administration in 10 cases. One patient allocated to ADocT refused further chemotherapy after three cycles and surgery.
Primary End Point: pCR Eleven (9.1%) of 121 patients had a pCR after treatment with ADocT, and 13 (10.3%) of 126 had a pCR after ADoc. The difference between the two treatment arms was -1.2%, with a 95% CI of -8.6% to +6.2%. If there is still a difference, it can be concluded that it is definitively below the anticipated difference of 15%. This remained true when the effect of tamoxifen was adjusted in a logistic regression model including the above-mentioned prognostic factors.
Secondary End Points A tumor response was assessed using imaging methods in 243 patients, with the same imaging technique being used before and after systemic therapy. The method initially chosen was not repeated before surgery in 10 of these patients; therefore, an alternative method was analyzed (Table 2). Partial or complete remission was observed in 77.5% (93 of 120) and 67.5% (83 of 123) of the patients treated with or without tamoxifen, respectively. The difference in favor of tamoxifen was 10.0%, with a 95% CI of -1.2% to +21.3%. This difference was due mainly to a higher complete response rate of 12.5% in the tamoxifen group compared with 5.7% in the chemotherapy-only group. We further analyzed to what extent palpation, ultrasound, mammography, or MRI directly before surgery can predict a pCR. The positive predictive value for a pCR in case of a total clinical or sonographic disappearance of the breast lesion was only between 24.3% and 26.1%. However, if tumor was still palpable or visible at ultrasound scan, the probability of detecting histologic tumor residues was 92.8% to 96.3%. The low number of cases limits the evidence of this analysis for mammography or MRI. So far, none of the methods seems to be useful in the future for identifying patients who do not need breast surgery (Table 3).
There were 245 patients for whom both clinical examination of axillary lymph nodes after start of treatment and histologic results at surgery were documented. While 124 (50.6%) had no evidence of involved nodes before the start of chemotherapy, 192 (78.4%) were clinically node-negative after treatment. At surgery, 119 (48.6%) showed no histologic infiltration of axillary nodes. Of the patients with clinically free axilla before surgery, 105 (54.7%) showed no histologic involvement at surgery. A median of 16 lymph nodes was removed, and in patients with positive nodes, a median of four (range, one to 28) was infiltrated by tumor cells.
Rate of breast-conserving surgery.
It was possible to conserve the affected breast in 68.8% of the patients. The rate was identical in the two treatment groups (68.6% and 69.0%, with a 95% CI for the difference of -12.0% to +11.1%). The rates of breast conservation were 83.1%, 73.2%, and 53.6% in tumors with an initial palpable size of less than 4 cm (n = 77),
Toxicity Toxicity was analyzed according to the treatment actually given; 120 patients received ADocT and 128 received ADoc alone. All hematologic tests were performed in over 98% of the cycles, except for neutrophil counts, which were examined in 651 (66.9%) of 973 cycles (87.9% of the patients had at least one measurement taken). Anemia of grade 3 or 4 was observed at least once in 2.5% of the patients in the ADocT group, with a maximum of 1.8% reached in the fourth cycle, and in 2.4% of those in the ADoc group, in whom the maximum of 2.4% occurred in the third cycle. Thrombocytopenia reached a maximum of grade 3 or 4 in 1.7% and 0% of the two groups, respectively. Leukopenia reached grade 3 or 4 in 40.0% of the patients receiving ADocT, with a maximum of 23.5% in the second cycle, and in 35.4% of those receiving ADoc alone, with a maximum of 18.6% in the third cycle. Neutrophil counts reached maximum grades of 3 or 4 in 35.6% and 24.6% of the patients receiving ADocT and ADoc, respectively. In the tamoxifen group, the decrease in leukocytes and neutrophils was more pronounced in all cycles except the first. Table 5 shows the figures for World Health Organization grades 1/2 and 3/4 hematologic toxicity per cycle.
Maximum World Health Organization grades for nonhematologic toxicity are shown for both treatments in Table 6. Basically, there was no difference in toxicity between the two randomized groups. There seemed to be a higher incidence of severe infections associated with the higher rate of grade 3/4 neutropenia in the ADocT group. However, the incidences of febrile neutropenia were similar, at 8.3% in the ADocT group and 8.7% in the ADoc group. The most common severe forms of toxicity, apart from alopecia, were fatigue and loss of appetite. All other toxicities occurred in less than 5% of the cycles.
Left ventricular ejection fraction was measured before surgery in 156 patients (62.9%). The fraction fell below 50% in only two patients (a decrease from 57% to 48% and from 77% to 37%). In three patients (1.2%), clinical signs of chronic heart failure became evident (one of grade 4 and two of grade 3). Another patient died during chest wall radiotherapy due to chronic heart failure, although the cardiac ultrasound examination had been normal before surgery. Three of these four events were observed in the tamoxifen group. Thromboembolic events (two pulmonary embolisms, two thromboses of the subclavian vein, and one deep venous thrombosis) were reported in five patients, four of whom were receiving tamoxifen treatment. Overall, grade 4 events occurred in 36 patients (30.0%) receiving ADocT and in 31 patients (24.2%) receiving ADoc alone. In both groups, the most frequent grade 4 event was neutropenia. The following deviations from the original treatment regimen were due to toxicity. The second, third, or fourth cycle of ADocT was postponed for more than 2 days for medical reasons 14 times (3.9%) in 14 patients (11.7%), and ADoc was postponed 16 times (4.4%) in 15 patients (11.7%). Chemotherapy was discontinued for medical reasons in three patients (2.5%) and six patients (4.7%) receiving ADocT and ADoc, respectively. Tamoxifen was interrupted or stopped because of toxicity in two patients (1.7%). Surgery had to be postponed for medical reasons in 14 patients (11.7%) treated with ADocT and in 11 patients (8.6%) receiving ADoc.
Prediction of pCR
Breast cancer is widely regarded as a systemic disease with a local component. Preoperative administration of systemic therapy therefore represents a logical step in improving adjuvant therapy. The timing of systemic therapy seems to have no impact on survival, but preoperative administration can significantly improve the rate of breast-conserving surgery,14 and new regimens can be evaluated rapidly and more precisely. For this purpose, prospective randomized trials have to be performed using pCR as a primary objective and comparing different preoperative systemic treatments. The randomized phase II trial reported here shows that a dose-dense, 8-week combination therapy consisting of doxorubicin and docetaxel can achieve a substantial rate of pCRs. The effect is clinically evident in the 68.8% rate of breast-conserving therapies in a population with a median initial tumor size by palpation of 4 cm. However, simultaneous application of tamoxifen does not have a significant impact on the effectiveness of the treatment. This might be explained by the short duration of treatment, so that a small positive effect might not be detected by a trial of this size. Indirect comparison of the observed pCR rate of 9.7% with rates reported in other preoperative chemotherapy trials is difficult, as each study includes populations with different risk profiles. Since other trials have also found that clinical nodal status significantly correlates with the pathologic response rate of preoperative treatment,1 the rate of node-positive patients can be used to compare the risk profile in these studies. In addition, the definitions of pCR varied. Separate reporting of the grade of pathologic remission has been given only in recent trials. Bonadonna et al15 reported on their experience at the Milan Cancer Institute with 536 nonrandomized patients with a node-positive rate of 37%. Using various regimens containing cyclophosphamide, methotrexate, and fluorouracil, as well as doxorubicin, epidoxorubicin, or mitoxantrone, they observed a pCR rate with no viable tumor cells of only 3%. The percentage of patients with clinically positive nodal status in the National Surgical Adjuvant Breast and Bowel Project B18 study,1 at 26% of 693 assessable patients, was lower than that in the GEPARDO study. A pathologic remission rate of 12.8% for the combination of doxorubicin and cyclophosphamide was reported. However, one third of these patients still had noninvasive tumor residues. In another phase II trial of 372 patients treated with fluorouracil, doxorubicin, and cyclophosphamide (FAC) at the M.D. Anderson Cancer Center in Houston, TX, a rate of pathologically detected regression of 16% was reported.3 No information regarding residual in situ tumors was provided for this less favorable group of patients (80% node-positive). This demonstrates that preoperative regimens can be evaluated only by direct comparison, in the manner selected for the GEPARDO study design. At present, only three studies with a comparable design have been reported. Buzdar et al16 reported a single-center study conducted at the M.D. Anderson Cancer Center in which 176 patients (as part of a larger trial) were preoperatively treated either with FAC or with a paclitaxel regimen. Sixty-two percent of the patients had clinically enlarged lymph nodes. Disease-free survival was used as a conventional primary study aim. The response of the primary tumor and the type of surgery used were only secondary objectives in the preoperatively treated patients. The clinical response rate was equal in the two arms (complete or partial remission 79% with FAC and 80% with paclitaxel), but there was a tendency for more breast conservations to be possible in the paclitaxel group (38% v 27%). However, there was a higher rate of pCRs (including in situ residues) after treatment with FAC (18% v 6%). The second study, by a French group, has only been published in abstract form.17,18 Two hundred patients with T2 and T3 tumors were treated preoperatively with either doxorubicin (60 mg/m2) and paclitaxel (200 mg/m2) (AT; n = 133 patients) or doxorubicin and cyclophosphamide (AC; n = 67 patients) every 3 weeks in an unbalanced 2:1 randomization. Sixty-three percent of the breast tumors were 2 to 5 cm in size, and 44% of the patients had no palpable axillary lymph nodes. A clinical response, with a more than 50% reduction in the maximum diameter, was observed in 85% of the AT group and in 66% of the AC group. None of the 200 patients showed signs of clinical progression. The primary aim of the study was the pCR rate, but this was evaluated using two different classifications. Using the classification given by Chevallier et al,19 grade 1 remission, ie, with no macroscopic or microscopic tumor residues, was observed in 6% in the AC group and 15% in the AT group; grade II remission, with only a residual in situ component, was documented in a further 4% and 5%, respectively. When the histology was classified using the criteria of Sataloff et al,20 with grade A remission indicating total or subtotal effectiveness of the tumor treatment, the difference between the two arms was more pronounced (23% for AT and 9% for AC). The rate of breast conservations was low in both arms, at 56% and 45% for AT and AC, respectively. The investigators were convinced of the superiority of the AT treatment and therefore stopped recruitment before the planned number of 240 patients was reached. This led to the considerably lower number of patients in the AC arm. A formal statistical analysis of the difference in efficacy between the two arms has not yet been presented. Both of the above studies show that differences in tumor response can be detected using prospective, randomized preoperative therapeutic trials in primary operable breast cancer. Histologic complete remission seems to be a valid surrogate marker for survival, but the criteria defining this can vary between different trials. However, the differences between the comparators are smaller than expected in the sample size calculations, so that these studies do not reach statistical significance. In addition, it has not been proven whether differences in pCR do in fact affect survival. In a recent initial report, Smith et al21 compared continuous-infusion (ci) epirubicin, cisplatin, and fluorouracil (ECisF) with conventional AC in a phase III trial involving 426 patients. After a median follow-up of 30 months, ci ECisF and AC resulted in similar rates of clinical response and mastectomy, but there was a significant survival benefit for ci ECisF. However, no pathologic response rates for either regimen were reported. For this reason, longer follow-up of all of these trials is necessary. The toxicity of the regimen investigated in the present trial was moderate, and adherence to the treatment was very high. This may have been due to the short duration of the treatment and the high level of compliance shown by patients able to observe for themselves that the tumor was becoming softer and smaller during therapy. In addition, clinicians able to observe shrinkage of the tumor using imaging methods are less likely to consider discontinuing the treatment, in comparison with adjuvant treatment, in which monitoring of the therapeutic effect is not possible. Compared with the pilot study,5 we observed a higher rate of hematologic and nonhematologic toxicity. Severe anemia or thrombocytopenia was not observed more frequently, whereas the incidence of grade 3/4 neutropenia increased from 2% to 30%. This may have been due to a reduction in the period of administration of granulocyte colony-stimulating factor (G-CSF) from 10 days to 6 days. However, cases of febrile neutropenia were still infrequent. In a study using an every-3-weeks combination of the same dosage of doxorubicin and docetaxel without the use of G-CSF, an 96% incidence of grade 3/4 neutropenia was reported in 214 patients with metastatic disease.22 The application of a median of eight cycles was associated with febrile neutropenia in 7% of the patients. Four cycles of dose-dense ADoc also resulted in a generally lower incidence of nonhematologic toxicity. The incidence of clinical chronic heart failure and the decrease in left ventricular ejection fraction were also lower in the present study in comparison with the metastatic regimen. This may have been due to the low cumulative dose of doxorubicin. A combination of doxorubicin and paclitaxel administered preoperatively using an every-3-weeks schedule without G-CSF led to an incidence of grade 3/4 neutropenia of 79%.17 The nonhematologic toxicity rate was comparable to that with the dose-dense ADoc regimen. However, a reduction in the left ventricular ejection fraction of less than 15% was observed in 64% of the patients, and a reduction of more than 15% was seen in 18%. This is in agreement with other reports on doxorubicin-paclitaxel regimens and is a very important finding in relation to treatment with curative intent. The present trial proved the feasibility, efficacy, and moderate toxicity of the dose-dense combination of doxorubicin, docetaxel, and G-CSF support. The short duration of treatment had a positive impact on the compliance of patients with preoperative systemic treatment. However, additional treatment with tamoxifen could not significantly improve the short-term end points related to the eradication of the primary breast tumor. In this clinical situation, synergistic effects among tamoxifen, docetaxel, and doxorubicin as postulated in in vitro studies7,8 do not seem to be as relevant. The antiestrogenic effect of tamoxifen might be more relevant in regimens with longer duration.
Supported in part by Aventis, Bad Soden, and Amgen, Munich, Germany.
1. Fisher B, Bryant J, Wolmark N, et al: Effect of preoperative chemotherapy on the outcome of women with operable breast cancer. J Clin Oncol 16: 2672-2685, 1998[Abstract] 2. Scholl SM, Fouquet A, Asselain B, et al: Primary versus adjuvant chemotherapy in premenopausal patients with tumors considered too large for breast conserving surgery: Preliminary results of a randomised trial. Eur J Cancer 30: 645-652, 1994
3.
Kuerer HM, Newman LA, Smith TL, et al: Clinical course of breast cancer patients with complete pathologic primary tumor and axillary lymph node response to doxorubicin-based neoadjuvant chemotherapy. J Clin Oncol 17: 460-469, 1999 4. Costa SD, von Minckwitz G, Eiermann W, et al: The role of Taxotere in the neoadjuvant chemotherapy of breast cancer. Semin Oncol 26: 24-31, 1999[Medline]
5.
von Minckwitz G, Costa SD, Eiermann W, et al: Maximized reduction of primary breast tumor size using preoperative chemotherapy with doxorubicin and docetaxel. J Clin Oncol 17: 1999-2005, 1999 6. Bourgeois H, Gruia G, Dieras V, et al: Docetaxel in combination with doxorubicin as 1st-line CT of metastatic breast cancer (MBC): A phase I dose-finding study. Proc Am Soc Clin Oncol 15: 148, 1996 (abstr 259) 7. Ferlini C, Scambia G, Distefano M, et al: Synergistic antiproliferative activity of tamoxifen and docetaxel on three oestrogen receptor-negative cancer cell lines is mediated by the induction of apoptosis. Br J Cancer 75: 884-891, 1997[Medline] 8. Pommerenke E, Mattern J, Volm M: Modulation of doxorubicin toxicity by tamoxifen in multidrug-resistant tumor cells in vitro and in vivo. J Cancer Res Clin Oncol 120: 422-426, 1994[Medline] 9. Early Breast Cancer Trialists’ Collaborative Group: Tamoxifen for early breast cancer: An overview of the randomised trials. Lancet 351: 1451-1467, 1998[Medline] 10. Sinn HP, Schmid H, Junkermann H, et al: Histologic regression of breast cancer after primary (neoadjuvant) chemotherapy. Geburtshilfe Frauenheilkd 54: 552-558, 1994[Medline] 11. Powles TJ, Hardy JR, Ashley SE, et al: Chemoprevention of breast cancer. Breast Cancer Res Treat 14: 23-31, 1989[Medline] 12. Fleiss JL: Statistical Methods for Rates and Proportions. New York, NY, Wiley, 1981 13. Ingelfinger JA: Biostatistics in clinical medicine. New York, NY, Macmillan Publishing Company, 1987
14.
Wolff AC, Davidson NE: Primary systemic therapy in operable breast cancer. J Clin Oncol 18: 1558-1569, 2000
15.
Bonadonna G, Valagussa P, Brambilla C, et al: Primary chemotherapy in operable breast cancer: Eight-year experience at the Milan Cancer Institute. J Clin Oncol 16: 93-100, 1998
16.
Buzdar AU, Singletary SE, Theriault RL, et al: Prospective evaluation of paclitaxel versus combination chemotherapy with fluorouracil, doxorubicin, and cyclophosphamide as neoadjuvant therapy in patients with operable breast cancer. J Clin Oncol 17: 3412-3417, 1999 17. Penault-Llorca F, Sastre X, Fiche M, et al: Pathological response to neoadjuvant chemotherapy (CT): Final results of a prospective randomized trial of 4AT vs. 4AC as induction therapy in patients with operable breast cancer using Sataloff classification. Breast Cancer Res Treat 57: A248, 1999 (abstr) 18. Dieras V, Fumoleau P, Romieu G, et al: A randomized, parallel study of doxorubicin/Taxol (paclitaxel) (AT) and doxorubicin/cyclophosphamide (AC) as neoadjuvant treatment of breast cancer. Presented at the 21st Annual San Antonio Breast Cancer Symposium, San Antonio, TX, December 12-15, 1998. Breast Cancer Res Treat 50: 233, 1998 (abstr 25) 19. Chevallier B, Roche H, Olivier JP, et al: Inflammatory breast cancer: Pilot study of intensive induction chemotherapy (FEC-HD) results in a high histologic response rate. Am J Clin Oncol 16: 223-228, 1993[Medline] 20. Sataloff DM, Mason BA, Prestipino AJ, et al: Pathologic response to induction chemotherapy in locally advanced carcinoma of the breast: A determinant of outcome. J Am Coll Surg 180: 297-306, 1995[Medline] 21. Smith IE, A’Hern RP, Howell A, et al: Preoperative continuous infusional ECisF (epirubicin, cisplatin and infusional 5 FU) vs. conventional AC chemotherapy for early breast cancer: A phase III multicentre randomised trial by the Topic Trial Group. Proc Am Soc Clin Oncol 19: 84a, 2000 (abstr 320) 22. Nabholtz JM, Falkson G, Campos D, et al: A phase III trial comparing doxorubicin (A) and docetaxel (T) (AT) to doxorubicin and cyclophosphamide (AC) as first line chemotherapy in MBC. Proc Am Soc Clin Oncol 18: 127a, 1999 (abstr 485) Submitted August 22, 2000; accepted April 26, 2001. This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
0.5 x 109/L for 
2 test, a sample size of at least 100 patients per group was needed,
