Impact of Time From Completion of Neoadjuvant Chemotherapy To Surgery On Survival Outcomes in Breast Cancer Patients
Impact of Time From Completion of Neoadjuvant Chemotherapy To Surgery On Survival Outcomes in Breast Cancer Patients
Impact of Time From Completion of Neoadjuvant Chemotherapy To Surgery On Survival Outcomes in Breast Cancer Patients
DOI 10.1245/s10434-015-5020-3
1
Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, Houston, TX; 2Department of
Health Services Research, The University of Texas MD Anderson Cancer Center, Houston, TX; 3Department of Breast
Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX; 4Department of Breast Surgical
Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX
Whereas preclinical models have suggested a biological hyperlipidemia, and diabetes (0, 1, 2–3). Multivariable Cox
basis for better outcomes with shorter intervals between proportional hazards models evaluating TTS as a continu-
definitive surgery and the initiation of adjuvant ous variable in days and weeks were implemented; for
chemotherapy, no such biological framework supports a clinical relevance and ease of interpretation, the data for
shorter interval between completion of NSC and sur- discrete intervals of B4, 4–6, and 6–24 weeks are pre-
gery.8–10 We hypothesized that the time to surgery (TTS) sented. Because a majority of patients (1031 of 1101,
after completion of NSC would not impact survival based 93.6 %) underwent surgery between 0 and 8 weeks after
on the early therapeutic effects of NSC on micrometastatic completion of NSC, a sensitivity analysis comparing out-
disease that may develop into frank metastases and ulti- comes among patients undergoing surgery between 0–
mately drive breast cancer-specific mortality. 8 weeks and 8–24 weeks was performed. Statistical anal-
yses were performed using SAS EG6.1 (SAS Institute Inc.,
METHODS Cary, NC) and S-Plus 8.2 (TIBCO Software Inc.). The
study was approved by the MD Anderson Institutional
A retrospective review of a prospectively maintained Review Board.
research database at The University of Texas MD Ander-
son Cancer Center identified 1449 patients diagnosed with RESULTS
stage I-III breast cancer treated with anthracycline- or
taxane-based NSC between June 1995 and April 2007. Patient and clinical characteristics are summarized in
Patients whose NSC was received outside of our institution Table 1. A total of 335 of 1,101 patients (30.4 %) had
(n = 318) and whose time interval from NSC to surgery surgery within 4 weeks of their last dose of NSC, 524
was unknown (n = 24) were excluded. To exclude outliers (47.6 %) in 4–6 weeks, and 242 (22.0 %) after [6 weeks.
and clinical scenarios not consistent with usual clinical Median age was 50 (range 24–83) years, median TTS was
practice, six patients with a TTS interval C24 weeks were 33 (range 8–159) days, and median follow-up time was 94
excluded. The final analysis included 1101 patients. (range 3–178) months. Patients whose TTS was B4 weeks
Patients were divided into three groups based on TTS: were more likely to have more advanced stage (p = 0.003)
(a) B4 weeks; (b) [4–6 weeks; and (c) [6–24 weeks. and higher tumor grade (p = 0.04).
Patient and clinical characteristics, including age, race, At the time of this analysis, 300 patients (27.2 %) had
clinical stage, nuclear grade, histology, presence of lym- died, 310 (28.1 %) had experienced a recurrence, and 78
phovascular invasion (LVI), tumor subtype (TNBC, HER2- (7.1 %) had experienced a locoregional recurrence.
neu positive or hormone-receptor positive), presence or Table 2 summarizes the 5-year OS, RFS, and LRFS
absence of three comorbidities (diabetes, hypertension and according to TTS. For the whole cohort, 5-year OS, RFS,
hyperlipidemia), presence of pathologic complete response and LRFS were 83, 76, and 94 %, respectively. The 5-year
(pCR), and type of surgery (breast-conserving surgery or OS estimate was 79, 87, and 81 % in patients who under-
mastectomy) were tabulated and compared between groups went surgery in 0–4, 4–6, and 6–24 weeks after NSC,
using v2 test. respectively (p = 0.03). The groups did not differ in 5-year
Locoregional recurrence-free survival (LRFS), recur- RFS or LRFS. Figure 1 demonstrates Kaplan–Meier esti-
rence-free survival (RFS), and overall survival (OS) were mates of OS, RFS, and LRFS according to TTS. Compared
measured from the date of diagnosis to the date of the first with an interval of 4–6 weeks, patients with triple-negative
documented locoregional recurrence, recurrence (either (p = 0.05) and HER2-neu positive (p = 0.04) breast can-
locoregional or distant, whichever came first), or death, cer who underwent surgery at B4 or [6 weeks had worse
respectively. Patients not experiencing the relevant end OS, but this difference was not reflected in RFS or LRFS
point were censored at last follow-up. Five-year LRFS, and the number of patients in each group was small. When
RFS, and OS were calculated using the Kaplan–Meier patients were grouped into those who did and did not
product limit method; groups were compared with the log- achieve a pCR, OS, RFS, and LRFS did not differ among
rank statistic. Cox proportional hazards regression models the three TTS intervals (Table 3).
were fit to determine the association between TTS interval In multivariable analysis, using as a reference patients
and survival outcomes after adjustment for potential con- with TTS B4 weeks, patients who underwent surgery at 4–
founders. Variables in the model were selected based on 6 weeks or at [6 weeks had equivalent OS, LRFS, and
statistical and clinical significance and included age (as a RFS (Table 4). Multivariable analysis including TTS as a
continuous variable), race, clinical stage (III vs. I–II), LVI, continuous variable did not demonstrate a statistically
breast cancer subtype, pCR, type of breast surgery, and significant difference in OS, LRFS, or RFS (data not
number of comorbidities including hypertension, shown).
Breast Cancer Neoadjuvant to Surgery Interval 1517
TABLE 1 Patient and clinical characteristics by interval from neoadjuvant therapy to surgery
All patients Time to surgery 0–4 weeks Time to surgery 4–6 weeks Time to surgery [6 weeks
(N = 1101) (n = 335) (n = 524) (n = 242)
N % N % N % N % p
Age (years)
B50 541 49.1 177 52.8 259 49.4 105 43.4
[50 560 50.9 158 47.2 265 50.6 137 56.6 0.08
Race
White 744 67.6 228 68.1 357 68.1 159 65.7
African-American 149 13.5 48 14.3 60 11.5 41 16.9
Hispanic 146 13.3 45 13.4 73 13.9 28 11.6
Other 62 5.6 14 4.2 34 6.5 14 5.8 0.35
Clinical stage
I/II 669 60.8 180 53.7 339 64.7 150 62.2
III 431 39.2 155 46.3 185 35.3 91 37.8 \0.01
Nuclear grade
I/II 405 37.6 104 31.9 204 39.8 97 40.6
III 673 62.4 222 68.1 309 60.2 142 59.4 0.04
Histology
Ductal 907 82.4 276 82.4 434 82.8 197 81.4
Other 194 17.6 59 17.6 90 17.2 45 18.6 0.89
Lymphovascular invasion
Negative 759 70.2 219 67.0 368 71.5 172 72.0
Positive 322 29.8 108 33.0 147 28.5 67 28.0 0.31
Subtype
HR-positive 629 57.6 190 57.4 298 57.3 141 58.5
Her2-neu-positive 188 17.2 53 16.0 91 17.5 44 18.3
TNBC 275 25.2 88 26.6 131 25.2 56 23.2 0.89
Comorbidities*
0 774 70.3 244 72.8 364 69.5 166 68.6
1 258 23.4 73 21.8 123 23.5 62 25.6
2–3 69 6.3 18 5.4 37 7.1 14 5.8 0.67
Pathologic complete response
No 906 82.3 278 83.0 417 79.6 211 87.2
Yes 195 17.7 57 17.0 107 20.4 31 12.8 0.03
Surgery
BCS 391 35.5 122 36.4 192 36.6 77 31.8
Mastectomy 710 64.5 213 63.6 332 63.4 165 68.2 0.40
Adjuvant radiation
No 220 20 51 15.2 103 19.7 66 27.3
Yes 881 80 284 84.8 421 80.3 176 72.7 \0.01
HR hormone receptor, TNBC triple-negative breast cancer, BCS breast-conserving surgery, AN anthracycline-based, TX taxane-based
*Hypertension, hyperlipidemia, and diabetes
All 1101 300 0.83 (0.81, 0.85) 310 0.76 (0.73, 0.79) 78 0.94 (0.92, 0.95)
Weeks from neoadjuvant chemotherapy to surgery
0–4 weeks 335 93 0.79 (0.74, 0.83) 94 0.74 (0.69, 0.78) 25 0.93 (0.9, 0.96)
4–6 weeks 524 129 0.87 (0.84, 0.9) 140 0.78 (0.74, 0.82) 32 0.96 (0.93, 0.97)
[6 weeks 242 78 0.81 (0.75, 0.85) 0.03 76 0.74 (0.68, 0.8) 0.28 21 0.92 (0.87, 0.95) 0.31
Weeks from neoadjuvant chemotherapy to surgery
0–8 weeks 1031 271 0.84 (0.82, 0.86) 284 0.76 (0.74, 0.79) 71 0.94 (0.93, 0.96)
[8 weeks 70 29 0.72 (0.59, 0.81) 0.001 26 0.71 (0.58, 0.8) 0.04 7 0.89 (0.78, 0.94) 0.23
Age (years)
B50 541 130 0.85 (0.81, 0.87) 161 0.75 (0.71, 0.78) 47 0.93 (0.9, 0.95)
[50 560 170 0.82 (0.79, 0.85) 0.03 149 0.77 (0.73, 0.81) 0.27 31 0.95 (0.93, 0.97) 0.05
Race
White 744 202 0.84 (0.81, 0.86) 200 0.77 (0.74, 0.8) 50 0.94 (0.92, 0.96)
African-American 149 59 0.72 (0.64, 0.79) 60 0.67 (0.59, 0.74) 19 0.9 (0.83, 0.94)
Hispanic 146 29 0.91 (0.85, 0.95) 35 0.79 (0.71, 0.85) 7 0.96 (0.91, 0.98)
Other 62 10 0.86 (0.74, 0.93) \0.01 15 0.76 (0.63, 0.85) \0.01 2 0.96 (0.86, 0.99) 0.01
Clinical stage
I 46 8 0.93 (0.81, 0.98) 8 0.93 (0.81, 0.98) 8 0.93 (0.81, 0.98)
II 623 135 0.89 (0.86, 0.91) 135 0.89 (0.86, 0.91) 135 0.89 (0.86, 0.91)
III 431 156 0.75 (0.7, 0.79) \0.0001 156 0.75 (0.7, 0.79) \0.0001 156 0.75 (0.7, 0.79) \0.0001
Nuclear grade
I/II 405 89 0.92 (0.88, 0.94) 93 0.83 (0.79, 0.86) 13 0.98 (0.96, 0.99)
III 673 205 0.78 (0.75, 0.81) \0.01 212 0.72 (0.68, 0.75) \0.01 63 0.91 (0.89, 0.93) \0.0001
Histology
Ductal 907 260 0.82 (0.79, 0.84) 262 0.75 (0.72, 0.78) 68 0.94 (0.92, 0.95)
Other 194 40 0.9 (0.84, 0.93) 0.02 48 0.81 (0.75, 0.86) 0.13 10 0.96 (0.92, 0.98) 0.20
Lymphovascular invasion
Negative 759 177 0.87 (0.85, 0.89) 181 0.8 (0.77, 0.83) 41 0.96 (0.94, 0.97)
Positive 322 121 0.73 (0.68, 0.78) \0.0001 125 0.66 (0.6, 0.71) \0.0001 36 0.9 (0.85, 0.93) \0.01
Subtype
HR-positive 629 147 0.88 (0.86, 0.91) 157 0.8 (0.76, 0.83) 28 0.96 (0.95, 0.98)
Her2-neu-positive 188 52 0.84 (0.78, 0.88) 59 0.73 (0.66, 0.79) 19 0.93 (0.88, 0.96)
TNBC 275 100 0.71 (0.65, 0.76) \0.0001 92 0.69 (0.63, 0.74) \0.01 30 0.89 (0.84, 0.92) \0.0001
Comorbidities*
0 774 203 0.83 (0.8, 0.86) 223 0.75 (0.72, 0.78) 58 0.93 (0.91, 0.95)
1 258 70 0.85 (0.8, 0.89) 63 0.8 (0.75, 0.85) 18 0.95 (0.91, 0.97)
2–3 69 27 0.76 (0.64, 0.85) 0.02 24 0.7 (0.57, 0.79) 0.11 2 0.96 (0.86, 0.99) 0.42
Pathologic complete response
No 906 281 0.81 (0.78, 0.83) 288 0.73 (0.7, 0.76) 67 0.94 (0.92, 0.95)
Yes 195 19 0.95 (0.91, 0.98) \0.0001 22 0.92 (0.87, 0.95) \0.0001 11 0.96 (0.92, 0.98) 0.28
Surgery
BCS 391 76 0.87 (0.83, 0.9) 82 0.83 (0.79, 0.87) 22 0.95 (0.93, 0.97)
Mastectomy 710 224 0.81 (0.78, 0.84) \0.0001 228 0.72 (0.69, 0.75) \0.0001 56 0.93 (0.91, 0.95) 0.10
HR hormone receptor, TNBC triple-negative breast cancer, BCS breast-conserving surgery
* Hypertension, hyperlipidemia and diabetes
Breast Cancer Neoadjuvant to Surgery Interval 1519
0.8
0.7 analysis suggested worse outcomes in patients who
0.6 underwent surgery at [8 weeks. Our results indicate that
0.5 intervals to surgery of up to 8 weeks will not compromise
0.4 survival outcomes patients treated with NSC.
0.3 NSABP B-18 and B-27 were seminal multi-institution
0.2 Time to Sugery 0-4 Weeks
Time to Sugery 4-6 Weeks
trials of preoperative chemotherapy whose results led to the
Time to Sugery >6 Weeks
0.1 adoption of NSC in clinical practice. Neither trial evaluated
0 the interval from completion of NSC to surgery.11–13
0 2 4 6 8 10 12 14
Time After Diagnosis (years)
Individual studies of NSC have reported intervals from 3 to
Time to Sugery
0-4 Weeks 335
Number of patients at risk
307 266 222 144 64 24 8
5 weeks, while some do not report the interval used.14–19 In
4-6 Weeks 524 491 452 400 273 149 67 5
>6 Weeks 242 249 193 164 109 62 25 1 the absence of data, clinicians have been left to extrapolate
from the adjuvant therapy interval data, although its
B Recurrence-Free Survival applicability to the neoadjuvant period is unclear.
1 In our study, the majority of patients (1,031 of 1,101,
P = 0.28 93.6 %) underwent surgery 0–8 weeks after completion of
Probability of Recurrence-Free Survival
0.9
0.8 chemotherapy. A sensitivity analysis suggested worse OS
0.7 and RFS with no difference in LRFS in patients undergoing
0.6
surgery in 8–24 weeks (n = 70); only OS remained sta-
0.5
tistically significantly different in multivariate analysis.
0.4
These data suggest that beyond 8 weeks, survival outcomes
0.3
may be compromised; to what extent this is due to breast
Time to Sugery 0-4 Weeks cancer-specific mortality versus comorbid illness is
0.2 Time to Sugery 4-6 Weeks
Time to Sugery >6 Weeks unknown. We evaluated the distribution of hypertension,
0.1
hyperlipidemia, and diabetes (categorized as 0, 1, or 2–3
0
0 2 4 6 8 10 12 14 comorbidities) and found no significant differences in the
Time After Diagnosis (years) number of comorbidities in patients who underwent sur-
Time to Sugery Number of patients at risk
0-4 Weeks 335 283 240 204 128 55 20 5 gery in 0–8 weeks versus 8–24 weeks (data not shown).
4-6 Weeks 524 452 410 354 244 130 57 2
>6 Weeks 242 205 176 142 93 55 23 1 However, it is possible that a difference in comorbidities
existed that was not captured with this simple analysis.
C Locoregional Recurrence-Free Survival Previous studies have demonstrated a subtype-specific
benefit from earlier initiation of adjuvant therapy in TNBC
Probability of Loco-Regional Recurrence-Free Survival
Patients with pCR 195 19 0.95 (0.91, 0.98) 22 0.92 (0.87, 0.95) 11 0.96 (0.92, 0.98)
0–4 weeks 57 6 0.91 (0.8, 0.96) 6 0.91 (0.8, 0.96) 3 0.95 (0.84, 0.98)
4–6 weeks 107 9 0.96 (0.9, 0.99) 10 0.92 (0.85, 0.96) 6 0.96 (0.9, 0.99)
[6 weeks 31 4 1 0.50 6 0.9 (0.71, 0.97) 0.25 2 0.97 (0.78, 1) 0.96
Patients without pCR 906 281 0.81 (0.78, 0.83) 288 0.73 (0.7, 0.76) 67 0.94 (0.92, 0.95)
0–4 weeks 278 87 0.77 (0.71, 0.81) 88 0.7 (0.65, 0.76) 22 0.93 (0.89, 0.95)
4–6 weeks 417 120 0.85 (0.81, 0.88) 130 0.75 (0.7, 0.79) 26 0.95 (0.93, 0.97)
[6 weeks 211 74 0.78 (0.72, 0.83) 0.17 70 0.72 (0.66, 0.78) 0.73 19 0.91 (0.86, 0.94) 0.34
Patients with HR-positive tumors 629 147 0.88 (0.86, 0.91) 157 0.8 (0.76, 0.83) 28 0.96(0.95, 0.98)
0–4 weeks 190 41 0.87 (0.81, 0.91) 43 0.8 (0.74, 0.85) 8 0.97 (0.93, 0.99)
4–6 weeks 298 71 0.89 (0.85, 0.92) 75 0.79 (0.74, 0.84) 13 0.97 (0.94, 0.98)
[6 weeks 141 35 0.89 (0.82, 0.93) 0.90 39 0.8 (0.72, 0.86) 0.68 7 0.95 (0.89, 0.97) 0.94
Patients with HER2-neu-positive 188 52 0.84 (0.78, 0.88) 59 0.73 (0.66, 0.79) 19 0.93 (0.88, 0.96)
tumors
0–4 weeks 53 16 0.81 (0.68, 0.89) 17 0.71 (0.56, 0.81) 5 0.92 (0.8, 0.97)
4–6 weeks 91 18 0.87 (0.78, 0.92) 25 0.78 (0.68, 0.86) 7 0.96 (0.89, 0.99)
[6 weeks 44 18 0.81 (0.66, 0.9) 0.04 17 0.67 (0.51, 0.79) 0.40 7 0.88 (0.73, 0.95) 0.26
Patients with TNBC 275 100 0.71 (0.65, 0.76) 92 0.69 (0.63, 0.74) 30 0.89 (0.84, 0.92)
0–4 weeks 88 36 0.61 (0.5, 0.71) 34 0.61 (0.5, 0.71) 12 0.85 (0.74, 0.91)
4–6 weeks 131 40 0.82 (0.74, 0.87) 38 0.75 (0.67, 0.82) 11 0.92 (0.86, 0.96)
[6 weeks 56 24 0.62 (0.48, 0.74) 0.05 20 0.66 (0.52, 0.77) 0.21 7 0.86 (0.73, 0.93) 0.37
pCR pathologic complete response, HR hormone receptor, TNBC triple-negative breast cancer
[4, 6] weeks versus [0, 4] weeks 0.89 (0.68, 1.16) 0.38 0.97 (0.74, 1.26) 0.80 0.80 (0.47, 1.36) 0.42
[6 weeks versus [0, 4] weeks 1.14 (0.84, 1.56) 0.40 1.16 (0.85, 1.58) 0.35 1.26 (0.70, 2.27) 0.45
[8 weeks versus [0, 8] weeks 1.62 (1.07, 2.36) 0.02 1.42 (0.92, 2.10) 0.09 1.75 (0.75, 3.50) 0.16
Variables included in multivariable model: age, race, clinical stage, LVI, breast cancer subtype, pCR, type of breast surgery, and number of
comorbidities including hypertension, hyperlipidemia and diabetes
Transitions between modalities of care in breast cancer mutually acceptable to the patient, surgical oncologist, and
can be challenging. Patients may be anxious to expedite reconstructive surgeon.22,23 Our data showing that delays
their treatment and avoid any perceived gaps in therapy. In in surgery of up to 8 weeks do not impact outcomes should
addition, surgeons must balance the morbidity associated allow patients and surgeons to rest assured in scheduling
with surgery immediately following immunosuppressive surgery during this window. Because metastatic disease is
chemotherapy and its impact on patient conditioning, a usually the driver of breast cancer-specific mortality,
known contribution to postoperative complications.20,21 prompt preoperative administration of effective systemic
The data for breast surgery after NSC has consistently therapy to target micrometastatic disease likely has a
demonstrated low complication rates, but additional greater impact on survival than the interval at which sur-
logistical issues may influence scheduling a date that is gery follows systemic therapy.
Breast Cancer Neoadjuvant to Surgery Interval 1521
Our study is subject to the limitations of retrospective 9. Fisher B, Gunduz N, Coyle J, Rudock C, Saffer E. Presence of a
studies; however, a prospective trial to address this ques- growth-stimulating factor in serum following primary tumor
removal in mice. Cancer Res. 1989;49(8):1996–2001.
tion would be difficult to conduct. Our ability to make 10. Segatto I, Berton S, Sonego M, et al. Surgery-induced wound
nuanced determinations about patient health was limited by response promotes stem-like and tumor-initiating features of breast
the use of a finite number of comorbidity search terms in cancer cells, via STAT3 signaling. Oncotarget. 2014;5(15):6267–79.
our database and by the possibility of incomplete docu- 11. Wolmark N, Wang J, Mamounas E, Bryant J, Fisher B. Preop-
erative chemotherapy in patients with operable breast cancer:
mentation; nonetheless, the patients in our cohort appear to nine-year results from National Surgical Adjuvant Breast and
have had a less-than-average burden of comorbidities. An Bowel Project B-18. J Natl Cancer Inst Monogr. 2001;(30):96–
additional limitation is the absence of data regarding 102.
treatment delays. 12. Bear HD, Anderson S, Brown A, et al. The effect on tumor
response of adding sequential preoperative docetaxel to preop-
In this study, we demonstrated no relationship between erative doxorubicin and cyclophosphamide: preliminary results
the NSC to surgery interval on OS, LRFS, or RFS. A from National Surgical Adjuvant Breast and Bowel Project Pro-
possible detrimental effect of TTS[8 weeks was observed; tocol B-27. J Clin Oncol. 2003;21(22):4165–74.
in clinical practice the scheduling of surgery within 13. Bear HD, Anderson S, Smith RE, et al. Sequential preoperative or
postoperative docetaxel added to preoperative doxorubicin plus
8 weeks of completion of NSC is usually feasible barring cyclophosphamide for operable breast cancer: National Surgical
an unexpected clinical event. As more patients are treated Adjuvant Breast and Bowel Project Protocol B-27. J Clin Oncol.
with NSC, multidisciplinary coordination of care at the 2006;24(13): 2019–27.
time of transition to surgery will become increasingly 14. Mauriac L, Durand M, Avril A, Dilhuydy JM. Effects of primary
chemotherapy in conservative treatment of breast cancer patients
common. These data should prove reassuring to clinicians with operable tumors larger than 3 cm. Results of a randomized
coping with ever-increasing patient volumes. trial in a single centre. Ann Oncol. 1991;2(5): 347–354.
15. Von Minckwitz G, Kümmel S, Vogel P, et al. Neoadjuvant
DISCLOSURES None vinorelbine-capecitabine versus docetaxel-doxorubicin-cyclophos-
phamide in early nonresponsive breast cancer: phase III randomized
GeparTrio trial. J Natl Cancer Inst. 2008;100(8):542–51.
16. Gianni L, Pienkowski T, Im YH, et al. Efficacy and safety of
REFERENCES neoadjuvant pertuzumab and trastuzumab in women with locally
advanced, inflammatory, or early HER2-positive breast cancer
1. Mauri D, Pavlidis N, Ioannidis JPA. Neoadjuvant versus adjuvant (NeoSphere): a randomised multicentre, open-label, phase 2 trial.
systemic treatment in breast cancer: a meta-analysis. J Natl Lancet Oncol. 2012;13(1):25–32.
Cancer Inst. 2005;97(3):188–94. 17. Von Minckwitz G, Raab G, Caputo A, et al. Doxorubicin with
2. Rastogi P, Anderson SJ, Bear HD, et al. Preoperative cyclophosphamide followed by docetaxel every 21 days com-
chemotherapy: updates of National Surgical Adjuvant Breast and pared with doxorubicin and docetaxel every 14 days as
Bowel Project Protocols B-18 and B-27. J Clin Oncol. preoperative treatment in operable breast cancer: the GEPAR-
2008;26(5):778–85. DUO study of the German Breast Group. J Clin Oncol.
3. Mittendorf EA, Buchholz TA, Tucker SL, et al. Impact of 2005;23(12):2676–85.
chemotherapy sequencing on local-regional failure risk in breast 18. Makris A, Powles TJ, Ashley SE, et al. A reduction in the
cancer patients undergoing breast-conserving therapy. Ann Surg. requirements for mastectomy in a randomized trial of neoadju-
2013;257(2):173–9. vant chemoendocrine therapy in primary breast cancer. Ann
4. Jara Sánchez C, Ruiz A, Martı́n M, et al. Influence of timing of Oncol. 1998;9(11):1179–84.
initiation of adjuvant chemotherapy over survival in breast can- 19. Untch M, Rezai M, Loibl S, et al. Neoadjuvant treatment with
cer: a negative outcome study by the Spanish Breast Cancer trastuzumab in HER2-positive breast cancer: results from the
Research Group (GEICAM). Breast Cancer Res Treat. GeparQuattro study. J Clin Oncol. 2010;28(12):2024–31.
2007;101(2):215–23. 20. Teoule P. Trojan J, Bechstein W, Woeste G. Impact of neoad-
5. Cold S, Düring M, Ewertz M, Knoop A, Møller S. Does timing of juvant chemotherapy on postoperative morbidity after
adjuvant chemotherapy influence the prognosis after early breast gastrectomy for gastric cancer. Dig Surg. 2015;32(4):229–37.
cancer? Results of the Danish Breast Cancer Cooperative Group 21. Berkel AE, Woutersen DP, van der Palen J, Klaase JM. Prog-
(DBCG). Br J Cancer. 2005;93(6):627–32. nostic factors for postoperative morbidity and tumour response
6. Shannon C, Ashley S, Smith IE. Does timing of adjuvant after neoadjuvant chemoradiation followed by resection for rectal
chemotherapy for early breast cancer influence survival? J Clin cancer. J Gastrointest Surg. 2014;18(9):1648–57.
Oncol. 2003;21(20):3792–7. 22. Warren Peled A, Itakura K, Foster RD, et al. Impact of chemotherapy
7. Gagliato D, Gonzalez-Angulo AM, Lei X, et al. Clinical impact on postoperative complications after mastectomy and immediate
of delaying initiation of adjuvant chemotherapy in patients with breast reconstruction. Arch Surg. 2010;145(9):880–5.
breast cancer. J Clin Oncol. 2014;32(8):735–44. 23. Song J, Zhang X, Liu Q, et al. Impact of neoadjuvant
8. Gunduz N, Fisher B, Saffer EA. Effect of surgical removal on the chemotherapy on immediate breast reconstruction: a meta-anal-
growth and kinetics of residual tumor. Cancer Res. 1979;39(10): ysis. PLoS One. 2014;9(5):e98225.
3861–5.