- Email: [email protected]
Survival after adding capecitabine and trastuzumab to neoadjuvant anthracycline-taxane-based chemotherapy for primary breast cancer (GBG 40—GeparQuattro)
original articles
Annals of Oncology
Annals of Oncology 25: 81–89, 2014 doi:10.1093/annonc/mdt410 Published online 21 November 2013
Survival after adding capecitabine and trastuzumab to neoadjuvant anthracycline-taxane-based chemotherapy for primary breast cancer (GBG 40—GeparQuattro) G. von Minckwitz1,2*, M. Rezai3, P. A. Fasching4, J. Huober5, H. Tesch6, I. Bauerfeind7, J. Hilfrich8, H. Eidtmann9, B. Gerber10, C. Hanusch11, J. U. Blohmer12, S. D. Costa13, C. Jackisch14, S. Paepke15, A. Schneeweiss16, S. Kümmel17, C. Denkert18, K. Mehta1, S. Loibl1,14 & M. Untch19 1
German Breast Group, Neu-Isenburg; 2Department of Gynaecology and Obstetrics, University Hospital, Frankfurt; 3Breast Center, Luisenkrankenhaus, Düsseldorf; Department of Gynaecology and Obstetrics, University Hospital, Erlangen; 5Department of Gynaecology and Obstetrics, University Hospital, Düsseldorf; 6Department of Onkology, Bethanien-Hospital, Frankfurt; 7Department of Gynaecology and Obstetrics, Klinikum Landshut, Landshut; 8Department of Gynaecology and Obstetrics, Eilenriede Klinik, Hannover; 9Department of Gynaecology and Obstetrics, University Hospital, Kiel; 10Department of Gynaecology and Obstetrics, University Hospital, Rostock; 11Department of Gynaecology and Obstetrics, Klinikum zum Roten Kreuz, München; 12Department of Gynaecology and Obstetrics, St. Gertrauden Krankenhaus, Berlin; 13Department of Gynaecology and Obstetrics, University Hospital, Magdeburg; 14Department of Gynaecology and Obstetrics, Klinikum Offenbach, Offenbach; 15 Department of Gynaecology and Obstetrics, University Hospital Rechts der Isar, München; 16National Center for Tumor Diseases, University Hospital, Heidelberg; 17 Department of Gynaecology and Obstetrics, Kliniken Essen-Mitte, Essen; 18Institute for Pathology, Charité Hospital, Berlin; 19Department of Gynaecology and Obstetrics, Helios-Klinikum Berlin-Buch, Berlin, Germany 4
Received 4 June 2013; revised 20 August 2013; accepted 22 August 2013
Background: The GeparQuattro study showed that adding capecitabine or prolonging the duration of anthracyclinetaxane-based neoadjuvant chemotherapy from 24 to 36 weeks did not increase pathological complete response ( pCR) rates. Trastuzumab-treated patients with HER2-positive disease showed a higher pCR rate than patients with HER2negative disease treated with chemotherapy alone. We here present disease-free (DFS) and overall survival (OS) analyses. Patients and methods: Patients (n = 1495) with cT ≥ 3 tumors, or negative hormone-receptor status, or positive hormone-receptor and clinically node-positive disease received four times epirubicin/cyclophosphamide and were thereafter randomly assigned to four times docetaxel (Taxotere), or four times docetaxel/capecitabine over 24 weeks, or four times docetaxel followed by capecitabine over 36 weeks. Patients with HER2-positive tumors received 1 year of trastuzumab, starting with the first chemotherapy cycle. Follow-up was available for a median of 5.4 years. Results: Outcome was not improved for patients receiving capecitabine (HR 0.92; P = 0.463 for DFS and HR 93; P = 0.618 for OS) as well as for patients receiving 36 weeks of chemotherapy (HR 0.97; P = 0.818 for DFS and HR 0.97; P = 0.825 for OS). Trastuzumab-treated patients with HER2-positive disease showed similar DFS (P = 0.305) but a significantly better adjusted OS (P = 0.040) when compared with patients with HER2-negative disease treated with chemotherapy alone. Recorded long-term cardiac toxicity was low. Conclusions: Long-term results, similar to the results of pCR, do not support the use of capecitabine in the neoadjuvant setting in addition to an anthracycline-taxane-based chemotherapy. However, the results support previous data showing a benefit of trastuzumab as predicted by higher pCR rates. Clinical trial number: NCT 00288002, www.clinicaltrials.gov. Key words: capecitabine, trastuzumab, neoadjuvant chemotherapy, disease-free survival, overall survival
introduction Pathological complete response ( pCR) after neoadjuvant chemotherapy has been considered as a precisely assessable predictor for disease-free (DFS) and overall survival (OS) in *Correspondence to: Prof Dr Gunter von Minckwitz, German Breast Group & Department of Gynaecology and Obstetrics, University Hospital, Frankfurt, c/o GBG Forschungs GmbH, Martin-Behaim-Straße 12, 63263 Neu-Isenburg, Germany. Tel: +49-69-6102-798740; Fax: +49-69-6102-7987440; E-mail: [email protected]
patients with primary breast cancer [1–3]. However, the large GeparTrio trial showed no improvement in pCR rates, but showed a survival advantage for the experimental treatments [4]. This demonstrates the importance to confirm the finding on pCR with corresponding survival analyses. The GeparQuattro study (clinical trial.gov NCT 00288002) is a phase III neoadjuvant study exploring the addition of capecitabine and trastuzumab to EC-T chemotherapy. Capecitabine was given in one experimental arm simultaneously to docetaxel
© The Author 2013. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].
original articles (Taxotere) and in another experimental arm in sequence to docetaxel, so that the trial could examine the effect of capecitabine as well as a 24 weeks treatment with a 36 weeks treatment. We have previously reported that EC-T with and without capecitabine as well as shorter and longer treatment resulted in similar pCR rates of in average 18.7% (ypT0 ypN0) or 25.8% (ypT0/is ypN0) [5]. The addition of trastuzumab was investigated as a second primary end point by an indirect comparison of patients with HER2-positive disease when treated in addition to trastuzumab with a reference group of patients with HER2-negative tumors treated with chemotherapy alone in the same study. This indirect comparison was considered acceptable as previous reports did not show different pCR rates for HER2-positive and HER2-negative disease when treated with anthracyclines and taxanes [6]. The results showed a pCR rate of 29.0% (ypT0 ypN0) or 41.4% (ypT0/is ypN0) for the HER2-positive/trastuzumab-treated patients and a pCR rate of 13.2% (ypT0 ypN0) or 17.8% (ypT0/is ypN0) for the HER2-negative/not-trastuzumabtreated patients [7]. We here report the corresponding survival analyses.
patients and methods
Annals of Oncology positive disease. Six monthly visits or telephone contacts were scheduled for at least 5 years to collect information on disease status. Patients with residual tumors at surgery were allowed to participate in a post-neoadjuvant randomized study on zoledronate or observation (NeoAdjuvant-Trial-Add-oN [NaTaN]) (NCT00512993) [9].
assessment of end points DFS (invasive) was calculated as the time from study registration to any invasive locoregional, contralateral or distant recurrence of breast cancer or any second primary invasive non-breast cancer or death of a patient due to any cause and OS to death of any cause [10]. During neoadjuvant chemotherapy local progression of the breast tumor was not considered as a relapse. Postprogression survival was calculated as time from first relapse to death. Response to neoadjuvant treatment was assessed clinically after four cycles of chemotherapy and before surgery as described previously [8]. Pathological response was assessed locally by examining all surgically excised breast and axillary lymphatic tissue according to the tumour–node– metastasis classification. Pathology reports were centrally reviewed at German Breast Group headquarters. Information on cardiac toxicity was collected in parallel. However, no standardized cardiac monitoring during follow-up was foreseen in the protocol.
objectives
statistics
The primary efficacy end point of the study was pathological complete response ( pCR) and has been reported together with the related toxicity earlier [8] [5]. Similar comparisons of treatments were planned as secondary objective of the study for DFS and OS as end points. Long-term cardiac assessment continued during follow-up.
The GeparQuattro study was an open-label, randomized, phase III trial with two primary objectives. Allocation to capecitabine and chemotherapy duration was carried out by central randomization with a 1:1:1 ratio for the three chemotherapy treatments and a stratification according to the participating site, clinical response after four cycles of EC, HER2-status, ER/PgRstatus and disease stage. Allocation to trastuzumab was based on local HER2 status assessment of the breast tumor. Assumptions for sample size calculation have published previously [8]. All analyses were carried out on an intent-to-treat approach including all patients having received one cycle chemotherapy and/or trastuzumab in the analysis for trastuzumab by HER2 status; and all patients randomly assigned after four cycles of EC in the capecitabine and duration analyses (CONSORT statement). Survival analysis was planned after median follow up of five years. Time to event outcome parameters were estimated using the Kaplan-Meier product-limit method and treatment groups were compared using the log-rank test. Univariate Cox proportional hazards models were used to calculate hazard ratios (HRs); 95% confidence intervals (CI) are provided. Multivariate Cox model was used to adjust for relevant baseline characteristics. Univariate Cox model was carried out in predefined subgroups similar to those used for the analysis of the pCR primary end point [9] and Cox model including interaction between subgroup factor and therapy was used to test for heterogeneity across subgroups. All statistical tests were two-sided by default; the significance level was set to 0.05 unadjusted for multiple comparisons.
selection of patients Female patients with previously untreated, unilateral or bilateral primary breast carcinoma were enrolled in the study if they provided a written informed consent. Tumor lesions needed to have a palpable size of ≥2 cm or a sonographical size of ≥1 cm in diameter. Locally advanced tumors with cT4 or cT3; or estrogen- (ER) and progesterone- (PgR) receptor-negative tumors; or ER or PgR positive tumors which are cN+ (for cT2) or pNSLN+ (for cT1) were eligible. Further eligibility criteria have been described previously [8].
treatment All patients were scheduled to four cycles of EC (epirubicin 90 mg/m2, cyclophosphamide 600 mg/m2, all on day 1, every 3 weeks). Thereafter, patients were randomly assigned to either docetaxel (Taxotere) 100 mg/m² day 1 q day 21 for four cycles (EC-T), or docetaxel 75 mg/m² day 1 q day 21 for four cycles plus capecitabine 900 mg/m² twice daily days 1–14 q day 21 for four cycles (EC-TX), or docetaxel 75 mg/m² day 1 q day 21 for four cycles followed by capecitabine 900 mg/m² twice daily days 1–14 q day 21 for four cycles (EC-T-X). Patients with HER2-positive tumors received trastuzumab 6 mg/kg i.v. every 3 weeks concomitantly to all cytotoxic treatment, starting with a loading dose of 8 mg/kg i.v. on day 1 of the first EC cycle. Recommended postoperative treatment consisted of radiotherapy to the ipsilateral remaining breast for all patients treated by breast-conserving surgery and radiotherapy to the chest wall with or without regional nodes in case of mastectomy for patients with an initial tumor size of >5 cm or clinically suspect axillary nodes before treatment or histologically involved nodes after neoadjuvant treatment. Endocrine treatment was given for at least 5 years in patients with ER- and/or PgR-positive tumors, and trastuzumab treatment was completed for a total duration of 12 months in patients with HER2-
| von Minckwitz et al.
results Between August 2005 and December 2006, 1495 patients were registered in the study and started neoadjuvant treatment (CONSORT statement). After completion of chemotherapy and surgery, 1408 patients were followed up for a median of 5.4 years until November 9th, 2012. Forty patients died without relapse of disease. Baseline characteristics of patients were well balanced between treatment groups as previously shown [1, 2].
Volume 25 | No. 1 | January 2014
original articles
Annals of Oncology
A
1.0
++
Disease-free survival (probability)
0.8
0.6
0.4
EC-T (N = 116 events) EC-TX (N = 102 events) EC-T-X (N = 113 events)
0.2
Capecitabine: Hazard ratio, 0.92 (95% CI 0.73 to 1.15) P = 0.463 log rank test Duration: Hazard ratio, 0.97 (95% CI 0.78 to 1.22) P = 0.818 log rank test
No. at risk EC-T EC-TX EC-T-X
471 479 471 0
B
429 435 428
383 383 395
1
2
346 345 349
300 285 301
4 3 Time since randomization (years)
262 253 258
40 63 42
5
6
7
1.0
Overall survival (probability)
0.8
0.6
0.4
EC-T (N = 68 events) EC-TX (N = 60 events) EC-T-X (N = 68 events)
0.2 Capecitabine: Hazard ratio, 0.93 (95% CI0.69 to 1.25) P = 0.618 log rank test Duration: Hazard ratio, 0.97 (95% CI 0.72 to 1.30) P = 0.825 log rank test No. at risk EC-T EC-TX EC-T-X
471 479 471 0
443 451 439
406 403 406
1
2
370 370 371
322 308 319
4 3 Time since randomization (years)
287 279 278
46 71 49
5
6
7
Figure 1. Disease-free (DFS) (A) and overall survival (OS) (B) after neoadjuvant anthracycline-taxane based chemotherapy with or without capecitabine and after 24 or 36 weeks of treatment duration.
Volume 25 | No. 1 | January 2014
doi:10.1093/annonc/mdt410 |
original articles
Annals of Oncology
DFS and OS were similar for patients treated with the three different chemotherapy regimens (Figure 1). There was no statistically significant difference in survival for patients receiving capecitabine or not together with or after anthracycline-taxane based chemotherapy (P = 0.463 for DFS and P = 0.618 for OS). Similarly there was no difference in outcome for patients receiving 24 weeks or 36 weeks of chemotherapy (P = 0.818 for DFS and P = 0.825 for OS). A multivariate analysis (Table 1) including age, clinical tumor size, clinical nodal status, histological tumor type, tumor grade, hormone-receptor status, and HER2 status/trastuzumab treatment to adjust for variations in baseline characteristics showed comparable results. Hazard ratios were 0.95 for DFS and 0.98 for OS regarding the capecitabine comparison, and 0.95 for DFS and 0.96 for OS regarding the duration comparison, with all 95% confidence intervals broadly crossing 1.0. Subgroup analysis (Figure 2) could not identify a subgroup for which an effect of capecitabine on DFS might be postulated. Patients with hormone-receptor-positive tumors
showed the lowest hazard ratio of 0.8 with an upper CI of 1.09 marginally crossing one in favor of adding capecitabine (test for interaction P = 0.832). Similarly no subgroup according to conventional risk factors was found favoring 24 or 36 weeks of chemotherapy duration. Comparing DFS and OS in patients having HER2-positive tumors and being treated with chemotherapy plus trastuzumab with patients having HER2-negative tumors and being treated with chemotherapy alone, no statistically significant difference was found (P = 0.305 for DFS and P = 0.074 for OS) (Figure 3). The baseline-factor adjusted survival analysis revealed a HR of 0.83 (P = 0.145) for DFS and a HR of 0.70 (P = 0.040) for OS in favor of the HER2-positive/trastuzumab-treated group (Table 1). As for the HER2-positive/trastuzumab-treated group a better outcome was observed only for OS but not for DFS, we compared in a post-hoc analysis post-progression survival between the two groups for those patients with documented relapse. In fact, post-progression survival was significantly better
Table 1. Baseline characteristics of patients and multivariate analysis on disease-free (DFS) and overall survival (OS) to adjust treatment effects by baseline characteristics of patients Parameter
Capecitabine treatment No Yes Non-randomized Treatment duration 24 weeks 36 weeks Non-randomized HER2 status/trastuzumab treatment Negative/no Positive/yes Age (years) <40 ≥40 Clinical tumor size (mm) <40 ≥40 Not known Tumor type Ductal invasive/other Lobular invasive Tumor grade I or II III Not known Clinical nodal status Negative Positive ER/PgR status Negative Positive
Total n
Relapses n(%)
DFS HR (95% CI) 0.95 (0.73–1.25)
471 950 74
116 (24.6) 215 (22.6)
479 942 74
113 (23.6) 218 (23.1)
1049 446
256 (24.4) 100 (22.4)
222 1273
56 (25.2) 300 (23.6)
674 798 23
124 (18.4) 225 (28.2)
1327 168
320 (24.1) 36 (21.4)
872 572 51
184 (21.1) 169 (29.4)
673 822
130 (19.3) 226 (27.5)
531 964
169 (31.8) 187 (19.4)
P value
Deaths n (%)
0.713
OS HR (95%CI)
P value
0.98 (0.69–1.40)
0.927
0.96 (0.67–1.37)
0.821
0.70 (0.50–0.98)
0.040
1.12 (0.73–1.72)
0.591
1.75 (1.29–2.39)
<0.001
1.24 (0.76–2.01)
0.391
0.14 (1.05–2.00)
0.025
1.61 (1.19–2.19)
0.002
0.53 (0.38–0.74)
<0.001
68 (14.4) 128 (13.5) 0.95 (0.73–1.25)
0.732 68 (14.2) 128 (13.5)
0.83 (0.65–1.07)
0.145 160 (15.3) 54 (12.1)
0.88 (0.65–1.20)
0.419 28 (12.6) 186 (14.6)
1.56 (1.24–1.96)
<0.001 68 (10.1) 139 (17.4)
1.14 (0.79–1.66)
0.481 192 (14.5) 22 (13.1)
1.31 (1.02–1.67)
0.032 104 (11.9) 107 (18.7)
1.50 (1.19–1.88)
0.001 71 (10.5) 143 (17.4)
0.61 (0.47–0.78)
<0.001 109 (20.5) 105 (10.9)
HR, hazard ratio; ER, estrogen receptor; PgR, progesterone receptor.
| von Minckwitz et al.
Volume 25 | No. 1 | January 2014
original articles
Annals of Oncology
A
Subgroup
N patients 1421
Hazard ratio Test for interaction (95% CI) 0.92 (0.73–1.15) 0.9077 0.94 (0.53–1.67) 0.92 (0.72–1.18) 0.8606 0.85 (0.58–1.23) 1.01 (0.76–1.35) 0.7139 0.93 (0.74–1.19) 0.80 (0.39–1.63) 0.8948 0.89 (0.65–1.23) 0.97 (0.70–1.35) 0.9054 1.04 (0.71–1.53) 0.87 (0.66–1.15) 0.8266 1.07 (0.76–1.51) 0.80 (0.59–1.09) 0.7667 0.95 (0.73–1.25) 0.84 (0.55–1.28)
Overall Age (years) <40 211 ≥40 1209 Clinical tumor size <40 642 ≥40 751 Histological type Ductal or other 1260 Lobular 158 Grade 1 or 2 822 3 542 Clinical nodal status Negative 640 Positive 777 ER/PgR status Negative 497 Positive 919 Trastuzumab treatment No 994 Yes 426
0.4
0.6
0.8 1.0
1.5
2.0
With capecitabine better Without capecitabine better CI confidence interval
B
Subgroup
N patients
Hazard ratio (95% CI)
Test for interaction
Overall Age (years) <40 ≥40 Clinical tumor size <40
1421
0.97 (0.78–1.22)
211 1209
0.69 (0.40–1.21) 1.04 (0.81–1.33)
642 751
0.76 (0.52–1.10) 1.15 (0.86–1.53)
0.8974 0.5322
Histological type Ductal or other 1260 Lobular 158 Grade 1 or 2 822 3 542 Clinical nodal status Negative 640 Positive 777 ER/PgR status Negative 497 Positive 919 Trastuzumab treatment No 994 Yes 426
0.2893 0.95 (0.74–1.20) 1.20 (0.60–2.38) 0.1816 0.85 (0.62–1.17) 1.16 (0.83–1.62) 0.4221 0.93 (0.64–1.35) 0.98 (0.73–1.31) 0.3102 0.96 (0.68–1.34) 0.96 (0.71–1.31) 0.6834 0.95 (0.73–1.24) 1.04 (0.67–1.60) 0.4
0.6
0.8 1.0
36 weeks duration better
1.5
2.0
24 weeks duration better
CI confidence interval
C
Subgroup
N patients
Hazard ratio (95% CI)
Overall Age (years) <40 ≥40 Clinical tumor size <40 ≥40 Histological type Ductal or other Lobular Grade 1 or 2 3 Clinical nodal status Negative Positive ER/PgR status Negative Positive
1495
0.89 (0.70–1.12)
218 1273
0.66 (0.36–1.24) 0.93 (0.72–1.20)
674 787
0.96 (0.66–1.41) 0.88 (0.65–1.18)
1327 164
0.89 (0.70–1.13) 0.45 (0.11–1.87)
864 572
1.16 (0.85–1.59) 0.64 (0.45–0.90)
665 822
0.87 (0.59–1.30) 0.85 (0.64–1.13)
522 964
0.66 (0.47–0.92) 1.05 (0.76–1.44)
0.1049 0.1094 0.1488 0.1044 0.1042 0.0969
0.4 CI confidence interval
Test for interaction
0.6
0.8 1.0
HER2-positive/ trastuzumab better
1.5
2.0
HER2-negative/ no trastuzumab better
Figure 2. Treatment effects of capecitabine (A), chemotherapy duration (B) and trastuzumab in HER2-positive disease versus no trastuzumab in HER2-negative disease (C) on DFS by subgroups.
Volume 25 | No. 1 | January 2014
doi:10.1093/annonc/mdt410 |
original articles A
Annals of Oncology
1.0
Disease-free survival (probability)
0.8
0.6
0.4
HER2+ with trastuzumab (N = 100 events) HER2– without trastuzumab (N = 256 events)
0.2 Hazard ratio, 0.89 (95% CI 0.70 to 1.12) P = 0.305 log rank test No. at risk + trastuzumab 446 – trastuzumab 1049 0
B
412 923
374 824
1
2
333 737
276 634
4 3 Time since randomization (years)
238 557 5
47 105 6
7
1.0
Overall survival (probability)
0.8
0.6
0.4
HER2+ with trastuzumab (N = 54 events) HER2– without trastuzumab (N = 160 events) 0.2 Hazard ratio, 0.76 (95% CI 0.56 to 1.03) P = 0.074 log rank test
No. at risk + trastuzumab 446 – trastuzumab 1049 0
423 956
389 864
1
2
360 784
302 676
264 605
4 3 5 Time since randomization (years)
55 120 6
7
Figure 3. DFS (A) and OS (B) after neoadjuvant chemotherapy in combination with trastuzumab in patients with HER2-positive disease and without trastuzumab in patients with HER2-negative disease.
| von Minckwitz et al.
Volume 25 | No. 1 | January 2014
original articles
Annals of Oncology
A
B
1.0
0.8 Disease-free survival (probability)
Disease-free survival (probability)
0.8
1.0
0.6
cCR after 4xEC (N = 12 events) cPR after 4xEC (N = 212 events) cNC after 4xEC (N = 107 events) cPD after 4xEC (N = 14 events)
0.4
Hazard ratio, 1, 0.41 (95% CI 1.19 to 1.68) P < 0.0001 log rank test
0.2 No. at risk cCR cPR cNC cPD
98 918 405 36 0
92 838 361 20 1
83 755 322 16
75 677 287 15
67 563 255 10
63 490 219 9
15 88 42 1
2
3
4
5
6
0.6
0.4
0.2 No. at risk cCR cPR cNC cPD 7
cCR at surgery (N = 54 events) cPR at surgery (N = 172 events) cNC at surgery (N = 69 events) cPD at surgery (N = 42 events) Hazard ratio, 1.85 (95% CI 1.62 to 2.12) P < 0.0001 log rank test 78 70 66 16 98 94 86 719 603 536 101 918 858 784 49 405 380 344 313 275 241 1 36 22 16 16 13 11
0
1
Time since randomization (years)
C
2
3
5
6
7
1.0
ypT0/is (N = 64 events) ypT1 (N = 113 events) ypT2 (N = 83 events) ypT3 (N = 35 events) ypT4 (N = 27 events)
0.8 Disease-free survival (probability)
4
Time since randomization (years)
0.6
0.4
0.2 No. at risk ypT0/is ypT1 ypT2 ypT3 ypT4
Hazard ratio, 1.58 (95% CI 1.43 to 1.75) P < 0.0001 log rank test 424 528 313 87 43 0
400 493 274 71 30 1
364 448 247 64 18
325 405 218 55 15
286 343 180 44 12
259 295 154 36 11
2 3 4 5 Time since randomization (years)
53 59 24 6 0 6
7
Figure 4. DFS analysis according to the clinical response after four cycles of epirubicin/cyclophosphamide (EC) (A), clinical response at surgery (B) and pathological ypT stage (C).
(P = 0.039) for relapsed patients with HER2-positive tumors compared with relapsed patients with HER2-negative tumors (data not shown). Subgroup analysis (Figure 2) shows that patients with grade 3 tumors (HR 0.64; test for interaction P = 0.1044) or hormone-receptor-negative tumors (HR 0.66; test for interaction P = 0.0969) had a trend for better DFS when the tumor was HER2-positive and treated with trastuzumab compared with patients with HER2-negative tumors. During the current observation period cardiac events were reported for 9 (2.0%) patients receiving trastuzumab in concurrence to EC-T and in 26 (2.5%) patients receiving EC-T alone (congestive heart failure (0 versus 3), cardiac ischemia (3 versus 3),
Volume 25 | No. 1 | January 2014
rhythm disorders (4 versus 14) and other cardiac events (2 versus 6)). As the comparison of clinical response after four cycles EC and clinical response as well as histopathological response at surgery was a predefined end point, we also analyzed the prognostic information of these three methods of response evaluation (Figure 4). Clinically complete response, partial remission or no change, and tumor progression after four treatment cycles resulted in different DFS (P ≤ 0.0001) (Figure 4A). By assessing clinical response at surgery, patients with complete response or with tumor progression showed a long-term prognosis comparable to the earlier assessment. In addition, DFS of patients
doi:10.1093/annonc/mdt410 |
original articles with a partial remission appeared more favorable compared with patients with no change in clinical tumor size (Figure 4B). In fact, DFS according to the histological tumor stage at surgery shows a similar separation with patients having a pCR (ypT0/is ypN0) showing the best prognosis and patients with gross residual disease (ypT3 or ypT4) showing 5 year event rates of 55% and higher. However, the frequency of patients with complete remission increased over the assessment time points (n = 98 (6.7%) with cCR after 4 cycles EC, n = 347 (24.5%) with cCR at surgery, n = 424 (30.4%) with pCR) (Figure 4C).
discussion Survival analysis confirmed the findings on pCR rates of the various treatments of the GeparQuattro study. No difference in DFS and OS was found for the neoadjuvant chemotherapy arms with or without capecitabine as well as for 24 weeks and 36 weeks of treatment duration. The only subgroup with a trend towards better outcome by the addition of capecitabine was the one with hormone-receptor-positive disease. The results are in concordance with other capecitabine trials conducted in the adjuvant setting. The USON 01062 trial added capecitabine to AC-T given at identical doses and reported no increase in DFS after the statistical plan was changed from an event-driven to a time-driven analysis due to a low event rate [3]. However, a significant advantage for OS was observed which has to be interpreted with caution, given the lowered event rate and power of study. The Finish Breast Group compared docetaxel followed by 5-fluorouracil/epirubicin/cyclophosphamide (FEC) with the same regimen but simultaneous use of capecitabine with docetaxel and with 5-fluorouracil/epirubicin [11]. A HR of 0.66 in favor of the capecitabine arm had been observed in a 3-year interim analysis [12]. However, the final analysis showed a nonsignificant HR of 0.79 between arms [7]. A recent metaanalysis that combined both studies showed significantly improved survival for the addition of capecitabine at a HR of 0.83 (0.71;0.98) [8]. The addition of trastuzumab to neoadjuvant chemotherapy and the postsurgical phase lead to a similar DFS of patients with HER2-positive disease when compared with a reference group of patients with HER2-negative disease receiving chemotherapy alone in the same study. As the natural course of HER2-positive disease is worse than HER2-negative disease, this could be interpreted as an improvement in outcomes. Comparison of OS revealed a significantly higher adjusted survival probability for patients with HER2-positive disease. However, it is likely that this effect at least in part derived from palliative treatment as post-progression survival was significantly more favorable in the HER2-positive subgroup. Surrogacy of pCR for survival cannot be adequately addressed in the GeparQuattro study due to the very different tumor biology and patient management of HER2positive and negative diseases. However, the results correspond to findings of the Neoadjuvant Herceptin study [5], where 118 trastuzumab untreated HER2-positive patients showed a worse event-free survival compared with 99 HER2-negative patients treated with the same chemotherapy in the same trial. As patients with HER2-positive tumors were randomly treated with trastuzumab, the effect of trastuzumab on pCR and survival could be precisely described and HER2-positive/trastuzumab-
| von Minckwitz et al.
Annals of Oncology
treated patients showed a better event-free survival than HER2negative/not-trastuzumab-treated patients. Analyses of the GeparQuattro study have several limitations: the three-arm design makes interpretation of data highly complex. Especially the comparison of trastuzumab-treated patients with HER2-positive disease with the reference group of patients with HER2-negative disease should be interpreted with caution due to the non-randomized, indirect type of analysis. Similar to the USON [8] and FINXX [13] study, the GeparQuattro study design used different doses of docetaxel in the control and experimental arms, so that the assessment of the capecitabine effect is biased by a potentially lower efficacy of docetaxel in the experimental arms [13]. Not randomizing 41 patients without response to EC might somewhat select a more favorable group of patients for the chemotherapy comparison. Finally, left ventricular ejection fraction measurement stopped before surgery, so that cardiac toxicity might be underestimated. In conclusion, the results of the GeparQuattro study do not support the use of capecitabine in the neoadjuvant setting in addition to an anthracycline-taxane-based chemotherapy. However, the results support a sustained patient’s benefit of trastuzumab when given together with neoadjuvant chemotherapy.
acknowledgements Conference information: Presented in part at the 35th San Antonio Breast Cancer Symposium, December 5–8, 2012, San Antonio, TX. The following centers and additional investigators contributed significantly to the conduct of the trial: Aalen (K. Gnauert); Amberg (A. Scharl); Aschersleben (G. Hasslbauer); Bergisch Gladbach (B. Liedtke); Berlin (J.-U. Blohmer, B. Brückner, P. Klare, A. Kleine-Tebbe, J. Potenberg, G. Sergius, A. Till, M. Untch); Bielefeld (M. Just); Böblingen (E. Weiss); Bochum (U. Bückner); Bonn (A. Hocke); Braunschweig (R. Lorenz); Bremen (G. Doering, S. Feidicker); Chemnitz (P. Krabisch); Coesfeld (M. Glados); Cuxhaven (U. Deichert); Deggendorf (D. Augustin); Düsseldorf (M. Rezai, G. von Minckwitz); Ebersberg (C. Höß); Eschweiler (P. Staib); Frankfurt am Main (F. Khandan, V. Möbus, C. Solbach, H. Tesch); Freiburg (E. Stickeler); Fürstenwalde (G. Heinrich); Fürth (J. Wilke); Gera (D.-M. Zahm); Gifhorn (T.-H. Dewitz); Göppingen (C. Heiss); Greifswald (A. Belau); Grevenbroich (E. Harms); Halle (T. Lantzsch, C. Thomssen); Hamburg (U. Mattner, V. Müller, A. Nugent); Hamm (F. Holms); Hanau (T. Müller); Hannover (J.-U. Deuker; K. M. Lübbe, S. Noeding); Herne (D. Strumberg); Hildesheim (C. Uleer); Jena (I. Runnebaum); Karlsruhe (O. Tomé, H.-U. Ulmer); Kassel (B. Conrad); Kiel (H. Eidtmann); Langen (E. Krapfl); Leipzig (D. Langanke); Lich (U. Kullmer); Limburg (P. Scheler); Lübeck (D. Fischer); Magdeburg (S. D. Costa, C. Kahl); Mainz (M. Schmidt, W. Wiest); Mannheim (M. W. Sütterlin); Meiningen (H. Graf ); München (N. Harbeck, B. Rack); Neunkirchen (G. P. Breitbach); Pforzheim (D. Kramer); Potsdam (F. Dreßler); Quedlinburg (S.-T. Graßhoff ); Regensburg (O. Ortmann); Reutlingen (I. Thalmann); Rodewisch (A. S. Schlosser); Rosenheim (T. Beck); Rostock (T. Reimer); Rottweil (G. Bartzke); Saarbrücken (M. Deryal); Stadthagen (S. Lemster); Stendal (A. Stefek); Stralsund (F. Ruhland); Stuttgart (U. Karck); Stuttgart (H. Stehle); Torgau (E. Simon); Traunstein
Volume 25 | No. 1 | January 2014
original articles
Annals of Oncology
(U. Kronawitter); Trier (M. R. Clemens); Tübingen (T. Fehm); Ulm (W. Janni); Unna (K. Latos); Wertheim (S. Pätzold); Westerstede (J. Mayer); Wiesbaden (F. Lorenz-Salehi); Wiesbaden (V. Heyl, G. Hoffmann) and Witten (J. Hackmann).
funding
4. 5.
6.
The trial received funding support from Roche and SanofiAventis. The funders had no access to the study database and were not involved in the analysis and interpretation of the results. No grant number applicable.
7.
disclosure
8.
GvM has received speaker honoraria research grants from Roche and Sanofi-Aventis. PAF has received speaker honoraria from Novartis and Genomic Health and research grants from Novartis. JH has received speaker honoraria from Roche, Sanofi-Aventis and GSK and research grants from GSK. AS has received speaker honoraria and research grants from Roche. All the other authors have declared no conflicts of interest.
references 1. von Minckwitz G, Untch M, Blohmer JU et al.. Definition and impact of pathological complete response on prognosis after neoadjuvant chemotherapy in various intrinsic breast cancer subtypes. J Clin Oncol 2012; 30: 1796–1804. 2. Cortazar P, Zhang L, Untch M et al.. Meta-analysis results from the Collaborative Trials in Neoadjuvant Breast Cancer (CTNeoBC). San Antonio Breast Cancer Conference San Antonio Texas, USA 2012; S1–S11. 3. O’Shaughnessy J, Paul D, Stokoe C et al.. First efficacy results of a randomized, open-label, phase II study of adjuvant doxorubicin plus cyclophosphamide,
Volume 25 | No. 1 | January 2014
9.
10.
11.
12.
13.
followed by docetaxel with or without capecitabine, in high-risk early breast cancer. Cancer Res 2010; 70(24 Suppl): Abstract # S4–2. von Minckwitz G, Blohmer JU, Costa SD et al.. Response-Guided neoadjuvant chemotherapy for breast cancer. J Clin Oncol; doi: 10.1200/JCO.2012.45.0940. von Minckwitz G, Rezai M, Loibl S et al.. Capecitabine in addition to anthracycline/ taxane-based neoadjuvant treatment in patients with primary breast cancer: The Phase III GeparQuattro Study. J Clin Oncol 2010; 28: 2015–2023. Huober J, von Minckwitz G, Denkert C et al.. Effect of neoadjuvant anthracyclinetaxane-based chemotherapy in different biological breast cancer phenotypes: overall results from the GeparTrio study. Breast Cancer Res Treat 2010; 124: 133–140. Untch M, Rezai M, Loibl S et al.. Neoadjuvant treatment with trastuzumab in HER2-positive breast cancer: results from GeparQuatro Study. J Clin Oncol 2010; 28: 2024–2031. Jiang Y, Yin W, Zhou L et al.. First efficacy results of capecitabine with anthracycline- and taxane-based adjuvant therapy in high-risk early breast cancer: a meta-analysis. PLoS One 2012; 7(3): e32474. von Minckwitz G, Zahm MD, Eidtmann H et al.. Zoledronic acid (ZOL) as add on therapy in patients with tumor residuals after neoadjuvant chemotherapy for primary breast cancer—first interim safety analysis of the NATAN study (GBG 36). Breast Cancer Symposium, Proceeding Book 2009; Abstract # 216, ASCO Breast 2009. Hudis CA, Barlow WE, Costantino JP et al.. Proposal for standardized definitions for efficacy end points in adjuvant breast cancer trials: the STEEP system. J Clin Oncol 2007; 25: 2127–2132. Joensuu H, Kellokumpu-Lehtinen PL, Huovinen R et al.. Adjuvant capecitabine, docetaxel, cyclophosphamide, and epirubicin for early breast cancer: final analysis of the randomized FinXX trial. J Clin Oncol 2012; 30: 11–18. Joensuu H, Kellokumpu-Lehtinen PL, Huovinen R et al.. Adjuvant capecitabine in combination with docetaxel and cyclophosphamide plus epirubicin for breast cancer: an open-label, randomised controlled trial. Lancet Oncol 2009; 10: 1145–1151. von Minckwitz G, Untch M, Nüesch E et al.. Impact of treatment characteristics on response of different breast cancer phenotypes: pooled analysis of the German neo-adjuvant chemotherapy trials. Breast Cancer Res Treat 2011; 125: 145–156.
doi:10.1093/annonc/mdt410 |
Annals of Oncology
Annals of Oncology 25: 81–89, 2014 doi:10.1093/annonc/mdt410 Published online 21 November 2013
Survival after adding capecitabine and trastuzumab to neoadjuvant anthracycline-taxane-based chemotherapy for primary breast cancer (GBG 40—GeparQuattro) G. von Minckwitz1,2*, M. Rezai3, P. A. Fasching4, J. Huober5, H. Tesch6, I. Bauerfeind7, J. Hilfrich8, H. Eidtmann9, B. Gerber10, C. Hanusch11, J. U. Blohmer12, S. D. Costa13, C. Jackisch14, S. Paepke15, A. Schneeweiss16, S. Kümmel17, C. Denkert18, K. Mehta1, S. Loibl1,14 & M. Untch19 1
German Breast Group, Neu-Isenburg; 2Department of Gynaecology and Obstetrics, University Hospital, Frankfurt; 3Breast Center, Luisenkrankenhaus, Düsseldorf; Department of Gynaecology and Obstetrics, University Hospital, Erlangen; 5Department of Gynaecology and Obstetrics, University Hospital, Düsseldorf; 6Department of Onkology, Bethanien-Hospital, Frankfurt; 7Department of Gynaecology and Obstetrics, Klinikum Landshut, Landshut; 8Department of Gynaecology and Obstetrics, Eilenriede Klinik, Hannover; 9Department of Gynaecology and Obstetrics, University Hospital, Kiel; 10Department of Gynaecology and Obstetrics, University Hospital, Rostock; 11Department of Gynaecology and Obstetrics, Klinikum zum Roten Kreuz, München; 12Department of Gynaecology and Obstetrics, St. Gertrauden Krankenhaus, Berlin; 13Department of Gynaecology and Obstetrics, University Hospital, Magdeburg; 14Department of Gynaecology and Obstetrics, Klinikum Offenbach, Offenbach; 15 Department of Gynaecology and Obstetrics, University Hospital Rechts der Isar, München; 16National Center for Tumor Diseases, University Hospital, Heidelberg; 17 Department of Gynaecology and Obstetrics, Kliniken Essen-Mitte, Essen; 18Institute for Pathology, Charité Hospital, Berlin; 19Department of Gynaecology and Obstetrics, Helios-Klinikum Berlin-Buch, Berlin, Germany 4
Received 4 June 2013; revised 20 August 2013; accepted 22 August 2013
Background: The GeparQuattro study showed that adding capecitabine or prolonging the duration of anthracyclinetaxane-based neoadjuvant chemotherapy from 24 to 36 weeks did not increase pathological complete response ( pCR) rates. Trastuzumab-treated patients with HER2-positive disease showed a higher pCR rate than patients with HER2negative disease treated with chemotherapy alone. We here present disease-free (DFS) and overall survival (OS) analyses. Patients and methods: Patients (n = 1495) with cT ≥ 3 tumors, or negative hormone-receptor status, or positive hormone-receptor and clinically node-positive disease received four times epirubicin/cyclophosphamide and were thereafter randomly assigned to four times docetaxel (Taxotere), or four times docetaxel/capecitabine over 24 weeks, or four times docetaxel followed by capecitabine over 36 weeks. Patients with HER2-positive tumors received 1 year of trastuzumab, starting with the first chemotherapy cycle. Follow-up was available for a median of 5.4 years. Results: Outcome was not improved for patients receiving capecitabine (HR 0.92; P = 0.463 for DFS and HR 93; P = 0.618 for OS) as well as for patients receiving 36 weeks of chemotherapy (HR 0.97; P = 0.818 for DFS and HR 0.97; P = 0.825 for OS). Trastuzumab-treated patients with HER2-positive disease showed similar DFS (P = 0.305) but a significantly better adjusted OS (P = 0.040) when compared with patients with HER2-negative disease treated with chemotherapy alone. Recorded long-term cardiac toxicity was low. Conclusions: Long-term results, similar to the results of pCR, do not support the use of capecitabine in the neoadjuvant setting in addition to an anthracycline-taxane-based chemotherapy. However, the results support previous data showing a benefit of trastuzumab as predicted by higher pCR rates. Clinical trial number: NCT 00288002, www.clinicaltrials.gov. Key words: capecitabine, trastuzumab, neoadjuvant chemotherapy, disease-free survival, overall survival
introduction Pathological complete response ( pCR) after neoadjuvant chemotherapy has been considered as a precisely assessable predictor for disease-free (DFS) and overall survival (OS) in *Correspondence to: Prof Dr Gunter von Minckwitz, German Breast Group & Department of Gynaecology and Obstetrics, University Hospital, Frankfurt, c/o GBG Forschungs GmbH, Martin-Behaim-Straße 12, 63263 Neu-Isenburg, Germany. Tel: +49-69-6102-798740; Fax: +49-69-6102-7987440; E-mail: [email protected]
patients with primary breast cancer [1–3]. However, the large GeparTrio trial showed no improvement in pCR rates, but showed a survival advantage for the experimental treatments [4]. This demonstrates the importance to confirm the finding on pCR with corresponding survival analyses. The GeparQuattro study (clinical trial.gov NCT 00288002) is a phase III neoadjuvant study exploring the addition of capecitabine and trastuzumab to EC-T chemotherapy. Capecitabine was given in one experimental arm simultaneously to docetaxel
© The Author 2013. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].
original articles (Taxotere) and in another experimental arm in sequence to docetaxel, so that the trial could examine the effect of capecitabine as well as a 24 weeks treatment with a 36 weeks treatment. We have previously reported that EC-T with and without capecitabine as well as shorter and longer treatment resulted in similar pCR rates of in average 18.7% (ypT0 ypN0) or 25.8% (ypT0/is ypN0) [5]. The addition of trastuzumab was investigated as a second primary end point by an indirect comparison of patients with HER2-positive disease when treated in addition to trastuzumab with a reference group of patients with HER2-negative tumors treated with chemotherapy alone in the same study. This indirect comparison was considered acceptable as previous reports did not show different pCR rates for HER2-positive and HER2-negative disease when treated with anthracyclines and taxanes [6]. The results showed a pCR rate of 29.0% (ypT0 ypN0) or 41.4% (ypT0/is ypN0) for the HER2-positive/trastuzumab-treated patients and a pCR rate of 13.2% (ypT0 ypN0) or 17.8% (ypT0/is ypN0) for the HER2-negative/not-trastuzumabtreated patients [7]. We here report the corresponding survival analyses.
patients and methods
Annals of Oncology positive disease. Six monthly visits or telephone contacts were scheduled for at least 5 years to collect information on disease status. Patients with residual tumors at surgery were allowed to participate in a post-neoadjuvant randomized study on zoledronate or observation (NeoAdjuvant-Trial-Add-oN [NaTaN]) (NCT00512993) [9].
assessment of end points DFS (invasive) was calculated as the time from study registration to any invasive locoregional, contralateral or distant recurrence of breast cancer or any second primary invasive non-breast cancer or death of a patient due to any cause and OS to death of any cause [10]. During neoadjuvant chemotherapy local progression of the breast tumor was not considered as a relapse. Postprogression survival was calculated as time from first relapse to death. Response to neoadjuvant treatment was assessed clinically after four cycles of chemotherapy and before surgery as described previously [8]. Pathological response was assessed locally by examining all surgically excised breast and axillary lymphatic tissue according to the tumour–node– metastasis classification. Pathology reports were centrally reviewed at German Breast Group headquarters. Information on cardiac toxicity was collected in parallel. However, no standardized cardiac monitoring during follow-up was foreseen in the protocol.
objectives
statistics
The primary efficacy end point of the study was pathological complete response ( pCR) and has been reported together with the related toxicity earlier [8] [5]. Similar comparisons of treatments were planned as secondary objective of the study for DFS and OS as end points. Long-term cardiac assessment continued during follow-up.
The GeparQuattro study was an open-label, randomized, phase III trial with two primary objectives. Allocation to capecitabine and chemotherapy duration was carried out by central randomization with a 1:1:1 ratio for the three chemotherapy treatments and a stratification according to the participating site, clinical response after four cycles of EC, HER2-status, ER/PgRstatus and disease stage. Allocation to trastuzumab was based on local HER2 status assessment of the breast tumor. Assumptions for sample size calculation have published previously [8]. All analyses were carried out on an intent-to-treat approach including all patients having received one cycle chemotherapy and/or trastuzumab in the analysis for trastuzumab by HER2 status; and all patients randomly assigned after four cycles of EC in the capecitabine and duration analyses (CONSORT statement). Survival analysis was planned after median follow up of five years. Time to event outcome parameters were estimated using the Kaplan-Meier product-limit method and treatment groups were compared using the log-rank test. Univariate Cox proportional hazards models were used to calculate hazard ratios (HRs); 95% confidence intervals (CI) are provided. Multivariate Cox model was used to adjust for relevant baseline characteristics. Univariate Cox model was carried out in predefined subgroups similar to those used for the analysis of the pCR primary end point [9] and Cox model including interaction between subgroup factor and therapy was used to test for heterogeneity across subgroups. All statistical tests were two-sided by default; the significance level was set to 0.05 unadjusted for multiple comparisons.
selection of patients Female patients with previously untreated, unilateral or bilateral primary breast carcinoma were enrolled in the study if they provided a written informed consent. Tumor lesions needed to have a palpable size of ≥2 cm or a sonographical size of ≥1 cm in diameter. Locally advanced tumors with cT4 or cT3; or estrogen- (ER) and progesterone- (PgR) receptor-negative tumors; or ER or PgR positive tumors which are cN+ (for cT2) or pNSLN+ (for cT1) were eligible. Further eligibility criteria have been described previously [8].
treatment All patients were scheduled to four cycles of EC (epirubicin 90 mg/m2, cyclophosphamide 600 mg/m2, all on day 1, every 3 weeks). Thereafter, patients were randomly assigned to either docetaxel (Taxotere) 100 mg/m² day 1 q day 21 for four cycles (EC-T), or docetaxel 75 mg/m² day 1 q day 21 for four cycles plus capecitabine 900 mg/m² twice daily days 1–14 q day 21 for four cycles (EC-TX), or docetaxel 75 mg/m² day 1 q day 21 for four cycles followed by capecitabine 900 mg/m² twice daily days 1–14 q day 21 for four cycles (EC-T-X). Patients with HER2-positive tumors received trastuzumab 6 mg/kg i.v. every 3 weeks concomitantly to all cytotoxic treatment, starting with a loading dose of 8 mg/kg i.v. on day 1 of the first EC cycle. Recommended postoperative treatment consisted of radiotherapy to the ipsilateral remaining breast for all patients treated by breast-conserving surgery and radiotherapy to the chest wall with or without regional nodes in case of mastectomy for patients with an initial tumor size of >5 cm or clinically suspect axillary nodes before treatment or histologically involved nodes after neoadjuvant treatment. Endocrine treatment was given for at least 5 years in patients with ER- and/or PgR-positive tumors, and trastuzumab treatment was completed for a total duration of 12 months in patients with HER2-
| von Minckwitz et al.
results Between August 2005 and December 2006, 1495 patients were registered in the study and started neoadjuvant treatment (CONSORT statement). After completion of chemotherapy and surgery, 1408 patients were followed up for a median of 5.4 years until November 9th, 2012. Forty patients died without relapse of disease. Baseline characteristics of patients were well balanced between treatment groups as previously shown [1, 2].
Volume 25 | No. 1 | January 2014
original articles
Annals of Oncology
A
1.0
++
Disease-free survival (probability)
0.8
0.6
0.4
EC-T (N = 116 events) EC-TX (N = 102 events) EC-T-X (N = 113 events)
0.2
Capecitabine: Hazard ratio, 0.92 (95% CI 0.73 to 1.15) P = 0.463 log rank test Duration: Hazard ratio, 0.97 (95% CI 0.78 to 1.22) P = 0.818 log rank test
No. at risk EC-T EC-TX EC-T-X
471 479 471 0
B
429 435 428
383 383 395
1
2
346 345 349
300 285 301
4 3 Time since randomization (years)
262 253 258
40 63 42
5
6
7
1.0
Overall survival (probability)
0.8
0.6
0.4
EC-T (N = 68 events) EC-TX (N = 60 events) EC-T-X (N = 68 events)
0.2 Capecitabine: Hazard ratio, 0.93 (95% CI0.69 to 1.25) P = 0.618 log rank test Duration: Hazard ratio, 0.97 (95% CI 0.72 to 1.30) P = 0.825 log rank test No. at risk EC-T EC-TX EC-T-X
471 479 471 0
443 451 439
406 403 406
1
2
370 370 371
322 308 319
4 3 Time since randomization (years)
287 279 278
46 71 49
5
6
7
Figure 1. Disease-free (DFS) (A) and overall survival (OS) (B) after neoadjuvant anthracycline-taxane based chemotherapy with or without capecitabine and after 24 or 36 weeks of treatment duration.
Volume 25 | No. 1 | January 2014
doi:10.1093/annonc/mdt410 |
original articles
Annals of Oncology
DFS and OS were similar for patients treated with the three different chemotherapy regimens (Figure 1). There was no statistically significant difference in survival for patients receiving capecitabine or not together with or after anthracycline-taxane based chemotherapy (P = 0.463 for DFS and P = 0.618 for OS). Similarly there was no difference in outcome for patients receiving 24 weeks or 36 weeks of chemotherapy (P = 0.818 for DFS and P = 0.825 for OS). A multivariate analysis (Table 1) including age, clinical tumor size, clinical nodal status, histological tumor type, tumor grade, hormone-receptor status, and HER2 status/trastuzumab treatment to adjust for variations in baseline characteristics showed comparable results. Hazard ratios were 0.95 for DFS and 0.98 for OS regarding the capecitabine comparison, and 0.95 for DFS and 0.96 for OS regarding the duration comparison, with all 95% confidence intervals broadly crossing 1.0. Subgroup analysis (Figure 2) could not identify a subgroup for which an effect of capecitabine on DFS might be postulated. Patients with hormone-receptor-positive tumors
showed the lowest hazard ratio of 0.8 with an upper CI of 1.09 marginally crossing one in favor of adding capecitabine (test for interaction P = 0.832). Similarly no subgroup according to conventional risk factors was found favoring 24 or 36 weeks of chemotherapy duration. Comparing DFS and OS in patients having HER2-positive tumors and being treated with chemotherapy plus trastuzumab with patients having HER2-negative tumors and being treated with chemotherapy alone, no statistically significant difference was found (P = 0.305 for DFS and P = 0.074 for OS) (Figure 3). The baseline-factor adjusted survival analysis revealed a HR of 0.83 (P = 0.145) for DFS and a HR of 0.70 (P = 0.040) for OS in favor of the HER2-positive/trastuzumab-treated group (Table 1). As for the HER2-positive/trastuzumab-treated group a better outcome was observed only for OS but not for DFS, we compared in a post-hoc analysis post-progression survival between the two groups for those patients with documented relapse. In fact, post-progression survival was significantly better
Table 1. Baseline characteristics of patients and multivariate analysis on disease-free (DFS) and overall survival (OS) to adjust treatment effects by baseline characteristics of patients Parameter
Capecitabine treatment No Yes Non-randomized Treatment duration 24 weeks 36 weeks Non-randomized HER2 status/trastuzumab treatment Negative/no Positive/yes Age (years) <40 ≥40 Clinical tumor size (mm) <40 ≥40 Not known Tumor type Ductal invasive/other Lobular invasive Tumor grade I or II III Not known Clinical nodal status Negative Positive ER/PgR status Negative Positive
Total n
Relapses n(%)
DFS HR (95% CI) 0.95 (0.73–1.25)
471 950 74
116 (24.6) 215 (22.6)
479 942 74
113 (23.6) 218 (23.1)
1049 446
256 (24.4) 100 (22.4)
222 1273
56 (25.2) 300 (23.6)
674 798 23
124 (18.4) 225 (28.2)
1327 168
320 (24.1) 36 (21.4)
872 572 51
184 (21.1) 169 (29.4)
673 822
130 (19.3) 226 (27.5)
531 964
169 (31.8) 187 (19.4)
P value
Deaths n (%)
0.713
OS HR (95%CI)
P value
0.98 (0.69–1.40)
0.927
0.96 (0.67–1.37)
0.821
0.70 (0.50–0.98)
0.040
1.12 (0.73–1.72)
0.591
1.75 (1.29–2.39)
<0.001
1.24 (0.76–2.01)
0.391
0.14 (1.05–2.00)
0.025
1.61 (1.19–2.19)
0.002
0.53 (0.38–0.74)
<0.001
68 (14.4) 128 (13.5) 0.95 (0.73–1.25)
0.732 68 (14.2) 128 (13.5)
0.83 (0.65–1.07)
0.145 160 (15.3) 54 (12.1)
0.88 (0.65–1.20)
0.419 28 (12.6) 186 (14.6)
1.56 (1.24–1.96)
<0.001 68 (10.1) 139 (17.4)
1.14 (0.79–1.66)
0.481 192 (14.5) 22 (13.1)
1.31 (1.02–1.67)
0.032 104 (11.9) 107 (18.7)
1.50 (1.19–1.88)
0.001 71 (10.5) 143 (17.4)
0.61 (0.47–0.78)
<0.001 109 (20.5) 105 (10.9)
HR, hazard ratio; ER, estrogen receptor; PgR, progesterone receptor.
| von Minckwitz et al.
Volume 25 | No. 1 | January 2014
original articles
Annals of Oncology
A
Subgroup
N patients 1421
Hazard ratio Test for interaction (95% CI) 0.92 (0.73–1.15) 0.9077 0.94 (0.53–1.67) 0.92 (0.72–1.18) 0.8606 0.85 (0.58–1.23) 1.01 (0.76–1.35) 0.7139 0.93 (0.74–1.19) 0.80 (0.39–1.63) 0.8948 0.89 (0.65–1.23) 0.97 (0.70–1.35) 0.9054 1.04 (0.71–1.53) 0.87 (0.66–1.15) 0.8266 1.07 (0.76–1.51) 0.80 (0.59–1.09) 0.7667 0.95 (0.73–1.25) 0.84 (0.55–1.28)
Overall Age (years) <40 211 ≥40 1209 Clinical tumor size <40 642 ≥40 751 Histological type Ductal or other 1260 Lobular 158 Grade 1 or 2 822 3 542 Clinical nodal status Negative 640 Positive 777 ER/PgR status Negative 497 Positive 919 Trastuzumab treatment No 994 Yes 426
0.4
0.6
0.8 1.0
1.5
2.0
With capecitabine better Without capecitabine better CI confidence interval
B
Subgroup
N patients
Hazard ratio (95% CI)
Test for interaction
Overall Age (years) <40 ≥40 Clinical tumor size <40
1421
0.97 (0.78–1.22)
211 1209
0.69 (0.40–1.21) 1.04 (0.81–1.33)
642 751
0.76 (0.52–1.10) 1.15 (0.86–1.53)
0.8974 0.5322
Histological type Ductal or other 1260 Lobular 158 Grade 1 or 2 822 3 542 Clinical nodal status Negative 640 Positive 777 ER/PgR status Negative 497 Positive 919 Trastuzumab treatment No 994 Yes 426
0.2893 0.95 (0.74–1.20) 1.20 (0.60–2.38) 0.1816 0.85 (0.62–1.17) 1.16 (0.83–1.62) 0.4221 0.93 (0.64–1.35) 0.98 (0.73–1.31) 0.3102 0.96 (0.68–1.34) 0.96 (0.71–1.31) 0.6834 0.95 (0.73–1.24) 1.04 (0.67–1.60) 0.4
0.6
0.8 1.0
36 weeks duration better
1.5
2.0
24 weeks duration better
CI confidence interval
C
Subgroup
N patients
Hazard ratio (95% CI)
Overall Age (years) <40 ≥40 Clinical tumor size <40 ≥40 Histological type Ductal or other Lobular Grade 1 or 2 3 Clinical nodal status Negative Positive ER/PgR status Negative Positive
1495
0.89 (0.70–1.12)
218 1273
0.66 (0.36–1.24) 0.93 (0.72–1.20)
674 787
0.96 (0.66–1.41) 0.88 (0.65–1.18)
1327 164
0.89 (0.70–1.13) 0.45 (0.11–1.87)
864 572
1.16 (0.85–1.59) 0.64 (0.45–0.90)
665 822
0.87 (0.59–1.30) 0.85 (0.64–1.13)
522 964
0.66 (0.47–0.92) 1.05 (0.76–1.44)
0.1049 0.1094 0.1488 0.1044 0.1042 0.0969
0.4 CI confidence interval
Test for interaction
0.6
0.8 1.0
HER2-positive/ trastuzumab better
1.5
2.0
HER2-negative/ no trastuzumab better
Figure 2. Treatment effects of capecitabine (A), chemotherapy duration (B) and trastuzumab in HER2-positive disease versus no trastuzumab in HER2-negative disease (C) on DFS by subgroups.
Volume 25 | No. 1 | January 2014
doi:10.1093/annonc/mdt410 |
original articles A
Annals of Oncology
1.0
Disease-free survival (probability)
0.8
0.6
0.4
HER2+ with trastuzumab (N = 100 events) HER2– without trastuzumab (N = 256 events)
0.2 Hazard ratio, 0.89 (95% CI 0.70 to 1.12) P = 0.305 log rank test No. at risk + trastuzumab 446 – trastuzumab 1049 0
B
412 923
374 824
1
2
333 737
276 634
4 3 Time since randomization (years)
238 557 5
47 105 6
7
1.0
Overall survival (probability)
0.8
0.6
0.4
HER2+ with trastuzumab (N = 54 events) HER2– without trastuzumab (N = 160 events) 0.2 Hazard ratio, 0.76 (95% CI 0.56 to 1.03) P = 0.074 log rank test
No. at risk + trastuzumab 446 – trastuzumab 1049 0
423 956
389 864
1
2
360 784
302 676
264 605
4 3 5 Time since randomization (years)
55 120 6
7
Figure 3. DFS (A) and OS (B) after neoadjuvant chemotherapy in combination with trastuzumab in patients with HER2-positive disease and without trastuzumab in patients with HER2-negative disease.
| von Minckwitz et al.
Volume 25 | No. 1 | January 2014
original articles
Annals of Oncology
A
B
1.0
0.8 Disease-free survival (probability)
Disease-free survival (probability)
0.8
1.0
0.6
cCR after 4xEC (N = 12 events) cPR after 4xEC (N = 212 events) cNC after 4xEC (N = 107 events) cPD after 4xEC (N = 14 events)
0.4
Hazard ratio, 1, 0.41 (95% CI 1.19 to 1.68) P < 0.0001 log rank test
0.2 No. at risk cCR cPR cNC cPD
98 918 405 36 0
92 838 361 20 1
83 755 322 16
75 677 287 15
67 563 255 10
63 490 219 9
15 88 42 1
2
3
4
5
6
0.6
0.4
0.2 No. at risk cCR cPR cNC cPD 7
cCR at surgery (N = 54 events) cPR at surgery (N = 172 events) cNC at surgery (N = 69 events) cPD at surgery (N = 42 events) Hazard ratio, 1.85 (95% CI 1.62 to 2.12) P < 0.0001 log rank test 78 70 66 16 98 94 86 719 603 536 101 918 858 784 49 405 380 344 313 275 241 1 36 22 16 16 13 11
0
1
Time since randomization (years)
C
2
3
5
6
7
1.0
ypT0/is (N = 64 events) ypT1 (N = 113 events) ypT2 (N = 83 events) ypT3 (N = 35 events) ypT4 (N = 27 events)
0.8 Disease-free survival (probability)
4
Time since randomization (years)
0.6
0.4
0.2 No. at risk ypT0/is ypT1 ypT2 ypT3 ypT4
Hazard ratio, 1.58 (95% CI 1.43 to 1.75) P < 0.0001 log rank test 424 528 313 87 43 0
400 493 274 71 30 1
364 448 247 64 18
325 405 218 55 15
286 343 180 44 12
259 295 154 36 11
2 3 4 5 Time since randomization (years)
53 59 24 6 0 6
7
Figure 4. DFS analysis according to the clinical response after four cycles of epirubicin/cyclophosphamide (EC) (A), clinical response at surgery (B) and pathological ypT stage (C).
(P = 0.039) for relapsed patients with HER2-positive tumors compared with relapsed patients with HER2-negative tumors (data not shown). Subgroup analysis (Figure 2) shows that patients with grade 3 tumors (HR 0.64; test for interaction P = 0.1044) or hormone-receptor-negative tumors (HR 0.66; test for interaction P = 0.0969) had a trend for better DFS when the tumor was HER2-positive and treated with trastuzumab compared with patients with HER2-negative tumors. During the current observation period cardiac events were reported for 9 (2.0%) patients receiving trastuzumab in concurrence to EC-T and in 26 (2.5%) patients receiving EC-T alone (congestive heart failure (0 versus 3), cardiac ischemia (3 versus 3),
Volume 25 | No. 1 | January 2014
rhythm disorders (4 versus 14) and other cardiac events (2 versus 6)). As the comparison of clinical response after four cycles EC and clinical response as well as histopathological response at surgery was a predefined end point, we also analyzed the prognostic information of these three methods of response evaluation (Figure 4). Clinically complete response, partial remission or no change, and tumor progression after four treatment cycles resulted in different DFS (P ≤ 0.0001) (Figure 4A). By assessing clinical response at surgery, patients with complete response or with tumor progression showed a long-term prognosis comparable to the earlier assessment. In addition, DFS of patients
doi:10.1093/annonc/mdt410 |
original articles with a partial remission appeared more favorable compared with patients with no change in clinical tumor size (Figure 4B). In fact, DFS according to the histological tumor stage at surgery shows a similar separation with patients having a pCR (ypT0/is ypN0) showing the best prognosis and patients with gross residual disease (ypT3 or ypT4) showing 5 year event rates of 55% and higher. However, the frequency of patients with complete remission increased over the assessment time points (n = 98 (6.7%) with cCR after 4 cycles EC, n = 347 (24.5%) with cCR at surgery, n = 424 (30.4%) with pCR) (Figure 4C).
discussion Survival analysis confirmed the findings on pCR rates of the various treatments of the GeparQuattro study. No difference in DFS and OS was found for the neoadjuvant chemotherapy arms with or without capecitabine as well as for 24 weeks and 36 weeks of treatment duration. The only subgroup with a trend towards better outcome by the addition of capecitabine was the one with hormone-receptor-positive disease. The results are in concordance with other capecitabine trials conducted in the adjuvant setting. The USON 01062 trial added capecitabine to AC-T given at identical doses and reported no increase in DFS after the statistical plan was changed from an event-driven to a time-driven analysis due to a low event rate [3]. However, a significant advantage for OS was observed which has to be interpreted with caution, given the lowered event rate and power of study. The Finish Breast Group compared docetaxel followed by 5-fluorouracil/epirubicin/cyclophosphamide (FEC) with the same regimen but simultaneous use of capecitabine with docetaxel and with 5-fluorouracil/epirubicin [11]. A HR of 0.66 in favor of the capecitabine arm had been observed in a 3-year interim analysis [12]. However, the final analysis showed a nonsignificant HR of 0.79 between arms [7]. A recent metaanalysis that combined both studies showed significantly improved survival for the addition of capecitabine at a HR of 0.83 (0.71;0.98) [8]. The addition of trastuzumab to neoadjuvant chemotherapy and the postsurgical phase lead to a similar DFS of patients with HER2-positive disease when compared with a reference group of patients with HER2-negative disease receiving chemotherapy alone in the same study. As the natural course of HER2-positive disease is worse than HER2-negative disease, this could be interpreted as an improvement in outcomes. Comparison of OS revealed a significantly higher adjusted survival probability for patients with HER2-positive disease. However, it is likely that this effect at least in part derived from palliative treatment as post-progression survival was significantly more favorable in the HER2-positive subgroup. Surrogacy of pCR for survival cannot be adequately addressed in the GeparQuattro study due to the very different tumor biology and patient management of HER2positive and negative diseases. However, the results correspond to findings of the Neoadjuvant Herceptin study [5], where 118 trastuzumab untreated HER2-positive patients showed a worse event-free survival compared with 99 HER2-negative patients treated with the same chemotherapy in the same trial. As patients with HER2-positive tumors were randomly treated with trastuzumab, the effect of trastuzumab on pCR and survival could be precisely described and HER2-positive/trastuzumab-
| von Minckwitz et al.
Annals of Oncology
treated patients showed a better event-free survival than HER2negative/not-trastuzumab-treated patients. Analyses of the GeparQuattro study have several limitations: the three-arm design makes interpretation of data highly complex. Especially the comparison of trastuzumab-treated patients with HER2-positive disease with the reference group of patients with HER2-negative disease should be interpreted with caution due to the non-randomized, indirect type of analysis. Similar to the USON [8] and FINXX [13] study, the GeparQuattro study design used different doses of docetaxel in the control and experimental arms, so that the assessment of the capecitabine effect is biased by a potentially lower efficacy of docetaxel in the experimental arms [13]. Not randomizing 41 patients without response to EC might somewhat select a more favorable group of patients for the chemotherapy comparison. Finally, left ventricular ejection fraction measurement stopped before surgery, so that cardiac toxicity might be underestimated. In conclusion, the results of the GeparQuattro study do not support the use of capecitabine in the neoadjuvant setting in addition to an anthracycline-taxane-based chemotherapy. However, the results support a sustained patient’s benefit of trastuzumab when given together with neoadjuvant chemotherapy.
acknowledgements Conference information: Presented in part at the 35th San Antonio Breast Cancer Symposium, December 5–8, 2012, San Antonio, TX. The following centers and additional investigators contributed significantly to the conduct of the trial: Aalen (K. Gnauert); Amberg (A. Scharl); Aschersleben (G. Hasslbauer); Bergisch Gladbach (B. Liedtke); Berlin (J.-U. Blohmer, B. Brückner, P. Klare, A. Kleine-Tebbe, J. Potenberg, G. Sergius, A. Till, M. Untch); Bielefeld (M. Just); Böblingen (E. Weiss); Bochum (U. Bückner); Bonn (A. Hocke); Braunschweig (R. Lorenz); Bremen (G. Doering, S. Feidicker); Chemnitz (P. Krabisch); Coesfeld (M. Glados); Cuxhaven (U. Deichert); Deggendorf (D. Augustin); Düsseldorf (M. Rezai, G. von Minckwitz); Ebersberg (C. Höß); Eschweiler (P. Staib); Frankfurt am Main (F. Khandan, V. Möbus, C. Solbach, H. Tesch); Freiburg (E. Stickeler); Fürstenwalde (G. Heinrich); Fürth (J. Wilke); Gera (D.-M. Zahm); Gifhorn (T.-H. Dewitz); Göppingen (C. Heiss); Greifswald (A. Belau); Grevenbroich (E. Harms); Halle (T. Lantzsch, C. Thomssen); Hamburg (U. Mattner, V. Müller, A. Nugent); Hamm (F. Holms); Hanau (T. Müller); Hannover (J.-U. Deuker; K. M. Lübbe, S. Noeding); Herne (D. Strumberg); Hildesheim (C. Uleer); Jena (I. Runnebaum); Karlsruhe (O. Tomé, H.-U. Ulmer); Kassel (B. Conrad); Kiel (H. Eidtmann); Langen (E. Krapfl); Leipzig (D. Langanke); Lich (U. Kullmer); Limburg (P. Scheler); Lübeck (D. Fischer); Magdeburg (S. D. Costa, C. Kahl); Mainz (M. Schmidt, W. Wiest); Mannheim (M. W. Sütterlin); Meiningen (H. Graf ); München (N. Harbeck, B. Rack); Neunkirchen (G. P. Breitbach); Pforzheim (D. Kramer); Potsdam (F. Dreßler); Quedlinburg (S.-T. Graßhoff ); Regensburg (O. Ortmann); Reutlingen (I. Thalmann); Rodewisch (A. S. Schlosser); Rosenheim (T. Beck); Rostock (T. Reimer); Rottweil (G. Bartzke); Saarbrücken (M. Deryal); Stadthagen (S. Lemster); Stendal (A. Stefek); Stralsund (F. Ruhland); Stuttgart (U. Karck); Stuttgart (H. Stehle); Torgau (E. Simon); Traunstein
Volume 25 | No. 1 | January 2014
original articles
Annals of Oncology
(U. Kronawitter); Trier (M. R. Clemens); Tübingen (T. Fehm); Ulm (W. Janni); Unna (K. Latos); Wertheim (S. Pätzold); Westerstede (J. Mayer); Wiesbaden (F. Lorenz-Salehi); Wiesbaden (V. Heyl, G. Hoffmann) and Witten (J. Hackmann).
funding
4. 5.
6.
The trial received funding support from Roche and SanofiAventis. The funders had no access to the study database and were not involved in the analysis and interpretation of the results. No grant number applicable.
7.
disclosure
8.
GvM has received speaker honoraria research grants from Roche and Sanofi-Aventis. PAF has received speaker honoraria from Novartis and Genomic Health and research grants from Novartis. JH has received speaker honoraria from Roche, Sanofi-Aventis and GSK and research grants from GSK. AS has received speaker honoraria and research grants from Roche. All the other authors have declared no conflicts of interest.
references 1. von Minckwitz G, Untch M, Blohmer JU et al.. Definition and impact of pathological complete response on prognosis after neoadjuvant chemotherapy in various intrinsic breast cancer subtypes. J Clin Oncol 2012; 30: 1796–1804. 2. Cortazar P, Zhang L, Untch M et al.. Meta-analysis results from the Collaborative Trials in Neoadjuvant Breast Cancer (CTNeoBC). San Antonio Breast Cancer Conference San Antonio Texas, USA 2012; S1–S11. 3. O’Shaughnessy J, Paul D, Stokoe C et al.. First efficacy results of a randomized, open-label, phase II study of adjuvant doxorubicin plus cyclophosphamide,
Volume 25 | No. 1 | January 2014
9.
10.
11.
12.
13.
followed by docetaxel with or without capecitabine, in high-risk early breast cancer. Cancer Res 2010; 70(24 Suppl): Abstract # S4–2. von Minckwitz G, Blohmer JU, Costa SD et al.. Response-Guided neoadjuvant chemotherapy for breast cancer. J Clin Oncol; doi: 10.1200/JCO.2012.45.0940. von Minckwitz G, Rezai M, Loibl S et al.. Capecitabine in addition to anthracycline/ taxane-based neoadjuvant treatment in patients with primary breast cancer: The Phase III GeparQuattro Study. J Clin Oncol 2010; 28: 2015–2023. Huober J, von Minckwitz G, Denkert C et al.. Effect of neoadjuvant anthracyclinetaxane-based chemotherapy in different biological breast cancer phenotypes: overall results from the GeparTrio study. Breast Cancer Res Treat 2010; 124: 133–140. Untch M, Rezai M, Loibl S et al.. Neoadjuvant treatment with trastuzumab in HER2-positive breast cancer: results from GeparQuatro Study. J Clin Oncol 2010; 28: 2024–2031. Jiang Y, Yin W, Zhou L et al.. First efficacy results of capecitabine with anthracycline- and taxane-based adjuvant therapy in high-risk early breast cancer: a meta-analysis. PLoS One 2012; 7(3): e32474. von Minckwitz G, Zahm MD, Eidtmann H et al.. Zoledronic acid (ZOL) as add on therapy in patients with tumor residuals after neoadjuvant chemotherapy for primary breast cancer—first interim safety analysis of the NATAN study (GBG 36). Breast Cancer Symposium, Proceeding Book 2009; Abstract # 216, ASCO Breast 2009. Hudis CA, Barlow WE, Costantino JP et al.. Proposal for standardized definitions for efficacy end points in adjuvant breast cancer trials: the STEEP system. J Clin Oncol 2007; 25: 2127–2132. Joensuu H, Kellokumpu-Lehtinen PL, Huovinen R et al.. Adjuvant capecitabine, docetaxel, cyclophosphamide, and epirubicin for early breast cancer: final analysis of the randomized FinXX trial. J Clin Oncol 2012; 30: 11–18. Joensuu H, Kellokumpu-Lehtinen PL, Huovinen R et al.. Adjuvant capecitabine in combination with docetaxel and cyclophosphamide plus epirubicin for breast cancer: an open-label, randomised controlled trial. Lancet Oncol 2009; 10: 1145–1151. von Minckwitz G, Untch M, Nüesch E et al.. Impact of treatment characteristics on response of different breast cancer phenotypes: pooled analysis of the German neo-adjuvant chemotherapy trials. Breast Cancer Res Treat 2011; 125: 145–156.
doi:10.1093/annonc/mdt410 |