Journal of the National Cancer Institute Advance Access originally published online on April 8, 2008
JNCI Journal of the National Cancer Institute 2008 100(8):542-551; doi:10.1093/jnci/djn085
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© The Author 2008. Published by Oxford University Press.
ARTICLES |
Neoadjuvant Vinorelbine–Capecitabine Versus Docetaxel–Doxorubicin–Cyclophosphamide in Early Nonresponsive Breast Cancer: Phase III Randomized GeparTrio Trial
On behalf of the German Breast Group
Affiliations of authors: German Breast Group, Frankfurt, Germany (GvM, SL, KM); Universitäts-Frauenklinik, Essen, Germany (SK); Frauenklinik, Horst Schmidt Kliniken, Wiesbaden, Germany (PV); Frauenklinik vom Roten Kreuz, Munich, Germany (CH); Universitäts-Frauenklinik, Kiel, Germany (HE); Frauenklinik Henriettenstiftung, Hannover, Germany (J. Hilfrich); Universitäts-Frauenklinik, Rostock, Germany (BG); Universitäts-Frauenklinik, Tübingen, Germany (J. Huober); Universitäts-Frauenklinik, Magdeburg, Germany (SDC); Städtisches Klinikum Offenbach, Germany (CJ); Universitäts-Frauenklinik, Frankfurt, Germany (GvM, SL, MK)
Correspondence to: Gunter von Minckwitz, MD, Universitäts-Frauenklinik Frankfurt & German Breast Group, c/o GBG Forschungs GmbH, Schleussner Strasse 42, 63263 Neu-Isenburg, Germany (e-mail: minckwitz{at}germanbreastgroup.de).
 |
ABSTRACT |
|---|
 Top Abstract Context and CaveatsPatients and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Background: Among breast cancer patients, nonresponse to initial neoadjuvant chemotherapy is associated with unfavorable outcome. We compared the response of nonresponding patients who continued the same treatment with that of patients who switched to a well-tolerated non–cross-resistant regimen.
Methods: Previously untreated breast cancer patients received two 3-week cycles of docetaxel at 75 mg/m2, doxorubicin at 50 mg/m2, and cyclophosphamide at 500 mg/m2 per day (TAC). Patients whose tumors did not decrease in size by at least 50% were randomly assigned to four additional cycles of TAC or to four cycles of vinorelbine at 25 mg/m2 and capecitabine at 2000 mg/m2 (NX). The outcome was sonographic response, defined as a reduction in the product of the two largest perpendicular diameters by at least 50%. A difference of 10% or less in the sonographic response qualified as noninferiority of the NX treatment. Pathological complete response was defined as no invasive or in situ residual tumor masses in the breast and lymph nodes. Toxic effects were assessed. All statistical tests were two-sided.
Results: Of 2090 patients enrolled in the GeparTrio study, 622 (29.8%) who did not respond to two initial cycles of TAC were randomly assigned to an additional four cycles of TAC (n = 321) or to four cycles of NX (n = 301). Sonographic response rate was 50.5% for the TAC arm and 51.2% for the NX arm. The difference of 0.7% (95% confidence interval = –7.1% to 8.5%) demonstrated noninferiority of NX (P = .008). Similar numbers of patients in both arms received breast-conserving surgery (184 [57.3%] in the TAC arm vs 180 [59.8%] in the NX arm) and had a pathological complete response (5.3% vs 6.0%). Fewer patients in the NX arm than in the TAC arm had hematologic toxic effects, mucositis, infections, and nail changes, but more had hand–foot syndrome and sensory neuropathy.
Conclusion: Pathological complete responses to both regimens were marginal. Among patients who did not respond to the initial neoadjuvant TAC treatment, similar efficacy but better tolerability was observed by switching to NX than continuing with TAC.
Prior knowledge Breast cancer that is not responsive to initial neoadjuvant chemotherapy is associated with unfavorable outcome. Study design Phase III randomized trial among previously untreated breast cancer patients who did not respond to two cycles of docetaxel, doxorubicin, and cyclophosphamide (TAC) that evaluated four additional TAC cycles or four cycles of vinorelbine and capecitabine (NX) for the noninferiority of NX. The outcome was sonographic response, defined as a reduction in tumor area by 50%. A pathological complete response was no invasive or in situ tumor mass in the breast and lymph nodes. Contribution The noninferiority of NX, compared with TAC, was demonstrated. However, pathological complete responses to both regimens were marginal. Similar efficacy but better tolerability was observed by switching to NX than continuing with TAC. Implications New neoadjuvant treatment options are urgently required for patients who do not respond to current neoadjuvant therapies. Limitations This noninferiority trial was not designed to determine whether TAC and NX had equal activity, and so the results may not apply to the general population.
|
Neoadjuvant administration of systemic treatment to patients with early breast cancer improves surgical options and provides early information on response to treatment (1). Response to two to four initial cycles of chemotherapy can predict overall response at surgery and is a prognostic factor for long-term outcome (2). Patients with a clinical response after two to four cycles of neoadjuvant chemotherapy have a higher probability of obtaining a pathological complete response at surgery than patients without an early response (1,3,4). Multivariable analysis has demonstrated that prediction of a pathological complete response on the basis of an early response is independent of steroid receptor status, grade, age, or histological tumor type (5). In the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-27 study, 5-year disease-free survival after a median follow-up of 78 months among 2309 patients who were treated with four cycles of doxorubicin plus cyclophosphamide (with or without four cycles of docetaxel and tamoxifen) was 75.3%, 64.9%, and 50.9%, respectively, if interim response assessment showed a clinical complete response, a partial response, or no response (2). This finding on the prognostic importance of clinical response is supported, for example, by a trial of neoadjuvant infusional 5-fluorouracil, doxorubicin, and cyclophosphamide treatment for breast cancer, in which early response was a statistically significant independent prognostic factor for local recurrence–free, distant disease–free, and overall survival (6).
We generated the hypothesis that, among patients without an early response to chemotherapy with anthracyclines and taxanes (ie, doxorubicin, cyclophosphamide, and docetaxel [TAC]), the benefit from vinorelbine and capecitabine (NX), a new well-tolerated, non–cross-resistant salvage chemotherapy regimen, is similar to that from continuing the TAC regimen. We selected NX as the salvage therapy to compare with TAC because of its efficacy, especially in patients with metastatic breast cancer who have been treated with anthracyclines and taxanes and its favorable toxicity profile (7–9).
The primary aim of this study was to compare the sonographic response rate after four additional cycles of TAC with that after four cycles of NX among breast cancer patients whose tumors did not respond to two cycles of TAC as the initial preoperative treatment. Sonographic response was chosen as the best available surrogate for pathological complete response because the choice of pathological complete response as the objective would have required a large number of patients due to the anticipated low response rate in this patient population according to our pilot study (10). Secondary aims of this analysis included assessment of toxicity and compliance, the rate of breast-conserving surgery (any breast surgery not requiring reconstruction or resulting in total mastectomy), and the rate of a pathological complete response in both randomized arms. Clinical and biological markers, which were determined before the start of treatment, were evaluated for their association with a pathological complete response.
|
Patients and Methods |
|---|
|
TopAbstractContext and CaveatsPatients and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Selection of Patients
Patients with previously untreated unilateral or bilateral primary breast carcinoma were enrolled in the study if they provided written informed consent. Breast cancer diagnosis had to be confirmed histologically from a core biopsy specimen. Patients needed to have at least one of the following risk factors because patients with none of these risk factors have a very low chance of achieving a pathological complete response to neoadjuvant chemotherapy according to our previous data (10): age younger than 36 years, clinical tumor size greater than 5 cm, estrogen receptor– and progesterone receptor–negative tumor, clinical involvement of axillary lymph nodes, or undifferentiated tumor grade. The breast tumor lesion had to be measurable in two dimensions by sonography, with one diameter being at least 2 cm. For bilateral disease, the investigator had to prospectively choose one side for evaluation (ie, the breast with the tumor that was most easily measured) on the registration form. Patients with locally advanced disease that included clinical involvement of skin and/or muscle, clinical evidence of inflammatory breast cancer (T4a–T4d), or N3 stage that included supraclavicular lymph nodes were eligible for this study but were randomly assigned to treatment in a separate stratum. For patients with inflammatory disease, the area of inflammation was used for evaluation and was measured clinically. For multifocal or multicentric disease, the lesion with the largest diameter was chosen for follow-up. Other inclusion criteria were age 18 years or older, Karnofsky performance status of at least 80%, normal left ventricular ejection fraction, and sufficient hematopoietic (neutrophil count of 
2.0 x 109 cells per liter, platelet count of 100 x 109 platelets per liter, and hemoglobin of 10 g/dL), liver (total bilirubin of 1x upper normal limit; aspartate aminotransferase and alanine aminotransferase, each of 
2.5x upper normal limit; and alkaline phosphatase of 5x upper normal limit), and renal (creatinine of 175 µmol/L) function. Patients were excluded if they had evidence of distant metastases, previous chemotherapy or radiation therapy, previous serious illnesses, concurrent treatment with sex hormones or experimental drugs, or a known hypersensitivity reaction to the study compounds or if they were male. The trial started in August 2002 in 88 centers in Germany, with a total enrollment period of 33 months.
Assessments of Endpoints
At every treatment cycle and before surgery, the target lesion and regional lymph nodes were examined by palpation and hematologic and biochemical parameters were assessed. Mammography and breast sonography were repeated at the end of the second cycle and before surgery.
A complete response by palpation and by sonography was defined as no clinical or sonographic sign or symptom of disease in the breast at the time of examination. A partial response was defined as a reduction in the product of the two largest perpendicular diameters of the primary tumor size by at least 50%. No change was defined as a reduction in tumor size of less than 50% or an increase of less than 25%. Progressive disease was defined as an increase in tumor size of more than 25% or the detection of a new breast lesion. Involvement of the axillary lymph nodes was not taken into account.
All excised tissues from the breast and regional lymph nodes were assessed for pathological response according to the following modified regression grading system described by Sinn et al. (11): regression grade 5 or a pathological complete response = no microscopic evidence of residual viable tumor cells (invasive or noninvasive) in all resected specimens of the breast and lymphatic nodes; regression grade 4 = no residual tumor in breast tissue but involved lymph nodes; regression grade 3 = only residual noninvasive (in situ) tumor in breast tissue irrespective of lymph node involvement; regression grade 2 = focal invasive tumor measuring 5 mm or less; and regression grade 0 or 1 = no substantial effect of chemotherapy detectable. If new lesions were detected, the response was graded as 0. The histological response was evaluated locally in the single institutions, but the pathology reports were centrally reviewed in the German Breast Group headquarters, Neu-Isenburg, Germany.
For the analysis of compliance and safety, the number of occurrences of all documented dose reductions, dose delays, and adverse events was reported. The severity of toxic effects was evaluated by use of the National Cancer Institute Common Terminology Criteria, version 3.0.
Treatment
An overview of the GeparTrio Trial design is given in Figure 1. All patients were scheduled to receive two 3-week cycles of a TAC regimen (doxorubicin at 50 mg/m2, cyclophosphamide at 500 mg/m2, and docetaxel at 75 mg/m2—all administered on day 1). Supportive treatment consisted of 20 mg of dexamethasone administered intravenously immediately before docetaxel and 4 mg of dexamethasone administered orally twice a day on days 2–4. Granulocyte colony-stimulating factor (filgrastim at 5 µg/kg or lenograstim at 150 µg/m2) and/or 500 mg of ciprofloxacin orally twice a day, each given on days 5–10, were used initially as both treatment and secondary prophylaxis; these drugs were used later as primary prophylaxis, by replacing granulocyte colony-stimulating factor with 6 mg of pegfilgrastim (Amgen Corp, Thousand Oaks, CA) once on day 2, after it was registered in October 2003 for febrile neutropenia or infection during all TAC cycles, as described previously (12). Early tumor response was determined, if possible, by sonography (alternatively by clinical examination, mammography, or magnetic resonance imaging) during the third week of the second cycle. When a (sonographic) response was identified, the patient was identified as a responder and TAC treatment for that patient was continued for four or six more cycles (Figure 1); results for these patients are reported in the accompanying article (13). If a sonographic response was not identified (ie, if tumor size decreased by <50%), the patient was identified as a nonresponder and was randomly assigned to either four more 3-week cycles of TAC or four 3-week cycles of an NX regimen (vinorelbine at 25 mg/m2 on days 1 and 8 plus capecitabine at 1000 mg/m2, administered orally twice a day on days 1–14). These nonresponding patients are the subject of this analysis. Patients whose tumor increased in size of 25% or more were removed from the study and treated at the discretion of the investigator.
|
Within 21 days after completion of chemotherapy and after overall assessment of response, patients had to undergo surgery and postoperative treatment according to standard recommendations (14). Trastuzumab was not recommended as a treatment to patients after surgery because it was not approved as a treatment for breast cancer until most of the patients had completed therapy.
Patients were categorized as having breast-conserving surgery (in contrast to mastectomy) as the final surgical procedure if they had a tumorectomy, segmentectomy, or quadrantectomy. Breast-conserving surgery was not achieved if the patients received a total or partial mastectomy with or without reconstruction procedure.
Statistical Methods
All 622 patients who did not respond to the initial two cycles of TAC were included in the efficacy and toxicity analyses (Figure 2). In addition, patients who agreed to participate and were centrally confirmed for eligibility but who were either randomly assigned as responder or not randomly assigned are listed in Figure 2.
|
The treatment groups were compared according to the study endpoints by use of two-sided P values from Pearson
2 test. Patients with missing response data were counted as having no response. For the primary endpoint of sonographic response rates (SRRs; complete or partial response), the shifted null hypothesis [ie, H0 = SRR(NX) – SRR(TAC) = –10%] was tested at the noninferiority margin at which the SRR for NX was 10% lower than the SSR for TAC, with a 2 test of H0 [ie, using a normal approximation to test whether SRR(NX) – SRR(TAC) = –10%].
Because the sample size was originally calculated for a one-sided test, corresponding one-sided P values are also quoted for the primary endpoint. A secondary sensitivity analysis of efficacy was performed in patients who were correctly randomly assigned as nonresponders, received the planned number of treatment cycles, and had an available sonographic efficacy assessment. Parameters included in univariate and multivariable binary logistic regression analyses of pathological complete response were chosen and dichotomized (age: 50 vs <50 years; tumor size: <4 vs 4 cm; histology: lobular invasive vs nonlobular; grading: 1 and 2 vs 3; lymph node status: positive vs negative; estrogen receptor [ER] and/or progesterone receptor [PgR] status: positive vs negative; HER2 status: positive vs negative; and allocated treatment) as previously described in the analysis of the GeparDuo study (15). All P values reported are from two-sided tests unless otherwise specified.
It was estimated that among patients not responding to the first two cycles of TAC, the rate of achieving a sonographic response after four additional cycles of TAC (p0) would be approximately 53% (10). Noninferiority of the treatment with NX was defined as a response rate up to 10% lower than the response rate for TAC (p1 > 43%). If we assumed equal efficacy of TAC and NX, a sample size of 618 patients would have been needed to confirm noninferiority of NX relative to TAC with an 
value of .05 and a β value of .20 by use of a one-sided test. Only one adaptive interim analysis for total sample size adjustment and without any stopping rule according to Fisher (16) was planned and performed after 300 evaluable nonresponding patients; results of this analysis did not lead to a change in sample size.
Central randomization by fax with a 1:1 ratio of patients assigned to the TAC arm vs the NX arm was stratified according to participating site and presence of locally advanced disease. The protocol was reviewed by all responsible local ethics committees. The conduct of the trial was supervised by an independent data monitoring committee. The trial registration number at clinicaltrials.gov is NCT 00544765.
|
Results |
|---|
|
TopAbstractContext and CaveatsPatients and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Baseline
The GeparTrio trial (NCT 00544765) started in August 2002 in 88 centers in Germany, with a total of 2090 patients who were confirmed to be eligible during the following 33 months (Figure 2). Of these 2090 patients, 2072 patients received at least one cycle of TAC. Of these 2072 patients, 1390 (67.1%) were randomly assigned as responders—who are the subjects of another analysis in an accompanying article (13)—and 622 (30.0%) were randomly assigned as nonresponders—who are the subjects of this analysis. For 60 (2.9%) patients, treatment was terminated early. Primary tumor progression with an increase in tumor size of more than 25% occurred in 14 (0.7%) patients. Three patients died during the first cycle of TAC due to neutropenic enterocolitis, which led to a more intense primary prophylaxis with pegfilgrastim and ciprofloxacin (12). Another patient died during the second cycle of TAC from myocardial infarction.
Baseline characteristics of the 622 patients randomly assigned as nonresponders are shown in Table 1. The median age for all patients was 51.8 years (range = 24–80 years); 267 (43%) of the patients were premenopausal. The median tumor size was 4.0 cm (range = 1.0–33.0 cm) if measured by palpation and 2.9 cm (range = 0.9–30.0 cm) if measured by sonography. For 11 patients, the tumor was clinically measured as less than 2 cm but sonographic measurement anticipated a much larger tumor size. In addition, 102 patients (16.7%) had stage T4 disease. Undifferentiated tumors were found more frequently in patients randomly assigned to continue with TAC than to switch to NX.
|
Compliance
A total of 1860 of 1926 planned TAC cycles were given to the 321 patients randomly assigned to TAC, and 621 TAC cycles and 1091 of 1204 planned NX cycles were given to the 301 patients randomly assigned to NX (Figure 2). The exact planned number of cycles was given to 290 (90.3%) and to 257 (85.4%) of patients in the TAC and NX arms, respectively (P = .1). Day 1 of a cycle was delayed for more than 3 days for 46 (2.5%) of the 1860 cycles in the TAC arm and for 99 (5.9%) of the 1712 cycles in the NX arm (difference = 3.3%; 95% confidence interval [CI] = 2.0% to 4.6%; P < .001) (Table 2). A dose reduction of any of the drugs occurred in 83 (4.5%) of 1860 cycles in the TAC arm and 162 (9.5%) of 1712 cycles in the NX arm (difference = 5%; 95% CI = 3.3% to 6.7%; P < .001). The higher frequency of dose reductions observed in the NX arm was attributable mainly to more reductions in the dose of capecitabine.
|
Efficacy
Sonographic assessment of response before surgery was possible in 508 (81.7%) of the 622 patients. Overall sonographic response rate was 50.5% (95% CI = 45.0% to 56.0%) with TAC and 51.2% (95% CI = 45.6% to 56.8%) with NX. Thus, the difference between NX and TAC was 0.7% (95% CI = –7.1% to 8.5%), and the noninferiority of NX (ie, response rates were no more than 10% worse for NX than for TAC) was statistically significant (for noninferiority, two-sided P = .008 and one-sided P = .004) and thus demonstrated (Table 3). A complete response was achieved in 29 (9.0%) of the 301 patients in the NX arm and in 20 (6.6%) of the 321 patients in the TAC arm. Twenty-five (4.0%) of the 622 patients in both arms (14 in the TAC arm and 11 in the NX arm) developed progressive disease during this second phase of treatment.
|
Response assessment by physical examination was recorded for 549 (88.3%) of the 622 patients (Table 3). Overall response rates as assessed by this method were 66.4% for TAC and 61.1% for NX (difference = 5.3%; 95% CI = –2.2% to 12.8%). No evidence of disease by palpation (ie, complete response) was reported for 61 (19.0%) of the 321 patients in the TAC arm and for 48 patients (15.9%) of the 301 patients in NX arm. An increase in size of more than 25% or a new lesion was reported for 14 (2.3%) of the 622 patients.
Investigators were asked also to assess response by the method considered by themselves to be most appropriate (Tables 1 and 3). A complete response was reported for 47 (14.6%) of the 321 patients in the TAC arm and 37 (12.3%) of the 301 patients in the NX arm. A partial response was reported for 176 (54.8%) in the TAC arm and 151 (50.2%) in the NX arm. Overall response rates were 69.5% in the TAC arm and 62.5% in the NX arm. Tumor progression was reported in 17 (2.7%) of all 622 patients.
A pathological complete response (ie, no invasive or in situ residual tumor in the breast and no involvement of the excised lymph nodes [regression grade 5]) was found in 17 (5.3%) of the 321 patients treated with TAC and 18 (6.0%) of the 301 patients treated with NX (Table 3). In situ residual tumor masses in the breast and/or tumor involvement of axillary lymph nodes were found in 18 and five additional patients in the TAC and NX arms, respectively.
After the systemic treatment, 612 (98.4%) of the 622 patients underwent surgery. For seven patients, the type of surgery was unknown. Breast-conserving surgery was possible in 184 (57.3%) of the 321 patients receiving TAC and 180 (59.8%) of the 301 patients treated with NX. Segmentectomy was the most common type of resection; it was used in 94 (29.3%) of the 321 patients in the TAC arm and 86 (28.6%) of the patients in the NX arm (Table 3). In 71 (22.2%) patients in the TAC arm and 50 (16.6%) patients in the NX arm, multiple surgeries were necessary to obtain tumor-free margins. Patients with tumor reduction detected by physical examination were more likely to have breast-conserving surgery than those in whom it was detected by sonography, but the difference was not statistically significant. Patients who received breast-conserving surgery were statistically significantly more likely to have a histological assessment with a regression grade of 4 or 5 than patients who underwent mastectomy (Table 4).
|
The association of conventional clinical and histological parameters with pathological complete response was analyzed by univariate and multivariable regression analyses. Hormonal receptor status, HER2 status, tumor grade, and size were statistically significantly associated with a pathological complete response in a univariate analysis (Table 5). The highest pathological complete response rate of 9.5% (95% CI = 5.1% to 13.9%) was achieved in the group of 168 patients with undifferentiated tumors. In the multivariable model, only a negative hormonal receptor status (odds ratio [OR] = 3.22; 95% CI = 1.01 to 10.29; P = .04) and a tumor size of less than 4 cm (OR = 3.49; 95% CI = 1.12 to 10.92; P = .03) were statistically significantly associated with pathological complete response (regression grade 5). In sensitivity analyses in which patients with missing baseline parameters were re-randomly distributed 1:1 to the two treatment arms or were analyzed according to their observed distribution and also when pathological complete response was replaced by regression grades of 3–5 in the definition of response, similar results were obtained when using a strict definition of regression grade 5 only (data not shown).
|
Safety
Statistically significantly fewer toxic effects were observed in the NX arm than in the TAC arm (Table 6). Grade 3–4 anemia, thrombocytopenia, and febrile neutropenia were statistically significantly more pronounced in patients who were assigned to continue on TAC than in those who were assigned to NX (P = .008, .005, and .049, respectively). Switching treatment to NX led also to a statistically significant decrease in various nonhematologic adverse events, including nausea or vomiting, mucositis of various organs, and nail changes. However, adverse events of capecitabine (eg, hand–foot syndrome) and vinorelbine (eg, peripheral neuropathy) were more frequent with NX. Differences between arms for other chronic toxic effects (eg, alopecia) could not be observed because all patients had been pretreated with TAC before randomization and almost all patients had alopecia at the time of randomization.
|
|
Discussion |
|---|
|
TopAbstractContext and CaveatsPatients and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Among patients without early response to treatment with two TAC cycles, similar efficacy was observed by switching to four NX cycles and by continuing with four more TAC cycles. Four NX cycles were better tolerated than four more TAC cycles. However, low pathological complete responses to both cytotoxic regimens (a total of six TAC cycles or two TAC cycles followed by four NX cycles) were observed in this setting.
The GeparTrio study was designed to evaluate early response to two initial cycles of TAC as an in vivo chemosensitivity test in previously untreated breast cancer patients. Interim response evaluation is a unique opportunity deriving from the use of preoperative chemotherapy. Two distinct strategies were followed: a further increase in the number of TAC cycles was assessed in patients with an early response (13), and a switch to a non–cross-resistant chemotherapy (ie, NX) was assessed in patients who did not have an early response to two cycles of TAC (this study).
The three-drug TAC regimen was chosen as the induction therapy because it is the only anthracycline–taxane combination therapy that was better than a three-drug anthracycline–based regimen in the adjuvant setting (17). The specific design of the trial did not allow use of a sequential anthracycline–taxane regimen. The GeparTrio's pilot study (10), which included 285 patients, reported pathological complete response rates of 17.9% in the whole population and 19.8% in patients with operable tumors. This success rate appears to be in the upper range of reported results when a conservative definition of pathological complete response is used (1).
Patients who do not respond to the initial cycles of chemotherapy have an unfavorable prognosis. In an early retrospective study from the Royal Marsden Hospital, lack of response after preoperative chemotherapy was an independent negative predictor for pathological complete response at surgery and also translated into a poorer long-term outcome (18). In the NSABP B-27 study, disease relapsed in 49.1% of the patients who did not have a clinical response to four cycles of doxorubicin and cyclophosphamide during the 78-month observation period, a rate that was statistically significantly higher than that in the patients who had a partial response (relapse rate = 35.1%) or in patients who had a complete response (relapse rate = 24.7%). Disease relapse was also independent of whether cytotoxic treatment with docetaxel was continued (2). In a French phase II trial (19) that included 278 patients, the 5-year overall survival rate of 67% for those who did not have a clinical response to three cycles of a combination of vinblastin, thiotepa, methotrexate, 5-fluorouracil, and doxorubicin was poorer than the survival rate of those who responded (ie, 82%). However, the overall survival rate of nonresponding patients who showed a response after completion of chemotherapy was as good as that of patients who had initially responded.
The first prospective randomized study that used early response as a decision aid was the randomized phase II Aberdeen study (20). In that study, patients (n = 159) were treated preoperatively with a doxorubicin-containing regimen for four cycles. After 12 weeks of treatment, the 68% of patients who had a clinical response were randomly assigned to either four more cycles of the same regimen or four cycles of docetaxel. All patients who did not have a clinical response were treated with docetaxel. The change to docetaxel increased the clinical response rate after eight cycles from 68% to 94% and the pathological complete response rate from 16% to 34% (including responses of patients with small foci of residual invasive disease). However, the pathological complete response rate in nonresponding patients was only 2%, and these nonresponders had statistically significantly worse progression-free survival at 2 years than responders.
The design of the GeparTrio study differs from that of the Aberdeen study in two important respects: first, induction chemotherapy already included docetaxel to provide maximum efficacy during the first two cycles and, second, randomization to a non–cross-resistant regimen took place among patients who had not responded to treatment at an early assessment. We therefore explored whether our design was practical in a pilot study (10); this study found that 208 (74%) of the 280 patients were responders and 72 (26%) were nonresponders and estimated that in nonresponding patients, the probability of a late sonographic response was 53% and that of a histological response was 5.2%. Because of statistical considerations regarding the low pathological complete response rate, the sonographic response was chosen as the primary endpoint for the phase III trial.
The experimental combination regimen of vinorelbine and capecitabine was chosen mainly because of its results in the setting of metastatic breast cancer. Both drugs have been tested as single agents in several phase II studies of patients after the failure of anthracycline–taxane–based first-line metastatic treatments, resulting in response rates of 26% (7–9). This two-drug combination appeared to be appropriate because there was no overlap in their resistance and safety profiles (1,4).
This randomized phase III study clearly confirmed the preliminary results of the pilot phase. Efficacy of NX was not inferior to that of TAC. However, the observed pathological complete response rate was very low with both regimens. Consequently, it has to be assumed that the subset of patients who did not have an early response to chemotherapy probably suffered from chemotherapy-resistant disease. In addition, because the chemotherapy resistance in these patients did not appear to be complete, as indicated by the clinical and sonographic response rates, and because the patients might have better surgical options if tumor size was reduced, the low pathological complete response rates did not justify stopping chemotherapy early. Moreover, long-term follow-up is necessary to obtain definitive data on the efficacy of both regimens.
As found by various other neoadjuvant trials, low levels of estrogen and progesterone receptors are reliable predictors for a pathological complete response and for having breast-conserving surgery (1,21). A similar observation has been made repeatedly for patients with undifferentiated tumors. Surprisingly, our data indicate that these two response predictors maintained their importance in the subgroup without an early response and so might be of help when a clinician must decide which nonresponding patients should start neoadjuvant chemotherapy and which should continue the same therapy.
The trial has several limitations. There is no proof that the equal activity of TAC and NX after early nonresponsive breast cancer is a general phenomenon or applies only to patients who did not respond to two initial cycles of TAC. It can therefore not be postulated that the results apply to all patients, especially those with an early response to chemotherapy. The design of the trial also does not allow the conclusion that both regimens are equally effective in achieving a pathological complete response because the primary aim of the trial was to demonstrate that NX was not inferior to TAC in achieving a sonographic response before surgery.
To fully assess the risks and benefits from further chemotherapy in this patient subset, the toxicity profiles have to be taken into account. NX appears to have a more favorable profile than TAC with regard to hematologic and nonhematologic toxic effects. Only hand–foot syndrome and sensory neuropathy were more pronounced in the NX arm than in the TAC arm.
Tumors will recur frequently in patients with tumors that do not show an early response. New treatment options are therefore urgently required. The GeparTrio study is, to our knowledge, the first prospective randomized trial to address this population; however, it was not able to identify a successful improvement in treatment efficacy. New biological agents, preferably those that have a mechanism of action involving the induction of drug sensitivity, should therefore be developed for use in this setting. An additional advantage of evaluation in the neoadjuvant setting (1) is the quick and precise efficacy assessment of these new molecules in breast cancer.
|
Funding |
|---|
|
TopAbstractContext and CaveatsPatients and MethodsResultsDiscussionFundingReferencesNotes |
|---|
This study received unrestricted financial support from Amgen, Germany; Roche, Germany; and Sanofi-Aventis, Germany.
|
NOTES |
|---|
|
TopAbstractContext and CaveatsPatients and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Dr von Minckwitz is conducting research sponsored by Sanofi-Aventis and Roche. Dr Huober is in the speaker's bureaus for Sanofi-Aventis, GlaxoSmithKline, and Roche. Dr Jackisch has received honoraria from Roche-Pharma, Germany. Dr Kaufmann has received honoraria for lectures from Pfizer, Astra-Zeneca, Novartis, Sanofi-Aventis, Roche, and Amgen.
Dr Erika Graf provided independent external statistical review of the manuscript.
The authors had full responsibility in the design of the study, the collection of the data, the analysis and interpretation of the data, the decision to submit the manuscript for publication, and the writing of the manuscript.
|
REFERENCES |
|---|
|
TopAbstractContext and CaveatsPatients and MethodsResultsDiscussionFundingReferencesNotes |
|---|
1. Kaufmann M, Hortobagyi GN, Goldhirsch A, et al. Recommendations from an international expert panel on the use of neoadjuvant (primary) systemic treatment of operable breast cancer: an update. J Clin Oncol (2006) 24(12):1940–1949.
2. Bear HD, Anderson S, Smith RE, et al. Sequential preoperative or postoperative docetaxel added to preoperative doxorubicin plus cyclophosphamide for operable breast cancer: National Surgical Adjuvant Breast and Bowel Project Protocol B-27. J Clin Oncol (2006) 24(13):2019–2027.
3. Beresford MJ, Stott D, Makris A. Assessment of clinical response after two cycles of primary chemotherapy in breast cancer [published online ahead of print July 11, 2007]. Breast Cancer Res Treat. doi:10.1007/s10549-007-9644-2.
4. von Minckwitz G, Costa SD, Raab G, et al. 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 study. J Clin Oncol (2001) 19(15):3506–3515.
5. von Minckwitz G, Sinn HP, Raab G, et al. Clinical response after two cycles compared to HER2, Ki-67, p53, and bcl-2 in independently predicting a pathological complete response after preoperative chemotherapy in patients with operable carcinoma of the breast [NCT00543829]. Breast Cancer Res. In press.
6. Moon YW, Rha SY, Jeung HC, Yang WI, Suh CO, Chung HC. Neoadjuvant chemotherapy with infusional 5-fluorouracil, adriamycin and cyclophosphamide (iFAC) in locally advanced breast cancer: an early response predicts good prognosis. Ann Oncol (2005) 16(11):1778–1785.
7. O'Shaughnessy JA, Blum JL. Capecitabine/taxane combination therapy: evolving clinical utility in breast cancer. Clin Breast Cancer (2006) 7(1):42–50.[Web of Science][Medline]
8. Seidman AD, O'Shaughnessy J, Misset JL. Single-agent capecitabine: a reference treatment for taxane-pretreated metastatic breast cancer? Oncologist (2002) 6(7, suppl):20–28.[CrossRef]
9. Zelek L, Barthier S, Riofrio M, et al. Weekly vinorelbine is an effective palliative regimen after failure with anthracyclines and taxanes in metastatic breast carcinoma. Cancer (2001) 92(9):2267–2272.[CrossRef][Web of Science][Medline]
10. von Minckwitz G, Blohmer JU, Raab G, et al. In vivo chemosensitivity-adapted preoperative chemotherapy in patients with early-stage breast cancer: the GEPARTRIO pilot study. Ann Oncol (2005) 16(1):56–63.
11. Sinn HP, Schmid H, Junkermann H, et al. Histological regression of breast cancer after primary (neoadjuvant) chemotherapy [German]. Geburtshilfe Frauenheilkd (1994) 54(10):552–558.[Medline]
12. von Minckwitz G, Kummel S, du Bois A, et al. Pegfilgrastim {+/–} ciprofloxacin for primary prophylaxis with TAC (docetaxel/doxorubicin/cyclophosphamide) chemotherapy for breast cancer. Results from the GEPARTRIO study. Ann Oncol (2007) 19(2):292–298.[CrossRef][Web of Science][Medline]
13. von Minckwitz G, Kummel S, Vogel P, et al. Intensified neoadjuvant chemotherapy in early-responding breast cancer: phase III randomized GeparTrio Study. J Natl Cancer Inst (2008) 100(8):552–562.
14. http://www.ago-online.de/index.php?site=mamma_guide&lang=de. Accessed October 17, 2007.
15. von Minckwitz G, Raab G, Caputo A, et al. Doxorubicin with cyclophosphamide followed by docetaxel every 21 days compared with doxorubicin and docetaxel every 14 days as preoperative treatment in operable breast cancer: the GEPARDUO study of the German Breast Group [NCT00543829]. J Clin Oncol (2005) 23(12):2676–2685.
16. Fisher LD. Self-designing clinical trials. Stat Med (1998) 17(14):1551–1562.[CrossRef][Web of Science][Medline]
17. Martin M, Pienkowski T, Mackey J, et al. Adjuvant docetaxel for node-positive breast cancer. N Engl J Med (2005) 352(22):2302–2313.
18. Verrill MW, Ashley SE, Walsh GA. Royal Marsden Hospital, London. Pathological complete response (pCR) in patients treated with neoadjuvant chemotherapy for operable breast cancer.
19. Khayat D, Antoine EC, Nizri d Auclerc G, Lorenzi I. Neoadjuvant chemotherapy in breast cancer: simple prognostic or useful predictive factor? In: ASCO Educational Book (2001) Alexandria, VA: American Society of Clinical Oncology. 510–515.
20. Smith IC, Heys SD, Hutcheon AW, et al. Neoadjuvant chemotherapy in breast cancer: significantly enhanced response with docetaxel. J Clin Oncol (2002) 20(6):1456–1466.
21. Loibl S, von Minckwitz G, Raab G, et al. Surgical procedures after neoadjuvant chemotherapy in operable breast cancer: results of the GEPARDUO trial. Ann Surg Oncol (2006) 13(11):1434–1442.
22. TNM Classification of Malignant Tumours (2002) 6th ed. Hoboken, NJ: John Wiley & Sons.
23. Bloom HJ, Richardson WW. Histological grading and prognosis in breast cancer; a study of 1409 cases of which 359 have been followed for 15 years. Br J Cancer (1957) 11(3):359–377.[Web of Science][Medline]
24. AGO Breast Commission, ed. Diagnostic And Treatment of Patients with Primary and Metastatic Breast Cancer. Complete Guidelines of AGO Breast Commission. Munich, Germany; (2008) http://www.ago-online.org/index.php?lang=de&site=mamma_guide_08_1_0&topic=mamma_guide.
Manuscript received October 24, 2007; revised February 11, 2008; accepted February 28, 2008.
Editorial about this Article
Related Article in JNCI
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2008 100: 521-523.
J Natl Cancer Inst 2008 100: 519.
This article has been cited by other articles:
|
|
 |
|
  V. Valero The Role of Systemic Treatment for Patients with Residual Disease after Neoadjuvant Chemotherapy ASCO Educational Book, January 1, 2009; 2009(1): 30 - 33. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Berruti, M. P. Brizzi, D. Generali, M. Ardine, L. Dogliotti, P. Bruzzi, and A. Bottini Presurgical Systemic Treatment of Nonmetastatic Breast Cancer: Facts and Open Questions Oncologist, November 1, 2008; 13(11): 1137 - 1148. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. J. Esteva and G. N. Hortobagyi Can Early Response Assessment Guide Neoadjuvant Chemotherapy in Early-Stage Breast Cancer? J Natl Cancer Inst, April 16, 2008; 100(8): 521 - 523. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||