Clinical Study

Oncology 2020;98:168–173 Received: November 1, 2019

Accepted: November 7, 2019
DOI: 10.1159/000504964 Published online: January 9, 2020

Does the Timing of Surgery after

Neoadjuvant Therapy in Breast Cancer

Downloaded from http://karger.com/ocl/article-pdf/98/3/168/3297560/000504964.pdf by Siriraj Medical Library, Mahidol University user on 14 August 2023
Patients Affect the Outcome?
Kausar Suleman a Osama Almalik a Emaan Haque b Ali Mushtaq b
Ahmed Badran a, c Adher Alsayed a Dahish Ajarim a Taher Al-Tweigeri a
Noha Jastaniyah a Tusneem Elhassan a Wafa Alkhayal a
aOncology Center, King Faisal Specialist Hospital and Research Center, Riyadh, Kingdom of Saudi Arabia;
bAlfaisal
University, Riyadh, Kingdom of Saudi Arabia; cDepartment of Clinical Oncology, Faculty of Medicine,
Ain Shams University, Cairo, Egypt

Keywords surgery; however, the trends of OS and DFS were poor when
Timing of surgery · Neoadjuvant chemotherapy · Breast surgery was delayed for ≥8 weeks. Median OS and median
cancer · Survival implications of time to surgical treatment DFS have not yet been reached. Of the 17% of patients that
had surgery after ≥8 weeks, 12.9% had pathological com-
plete response (pCR), while among those that received sur-
Abstract gery 4–7 weeks and <4 weeks after neoadjuvant chemother-
Background: There is a paucity of literature examining the apy, 26% and 21% had pCR, respectively (p = 0.02). ER+/HER-
impact of timing of surgery after neoadjuvant chemothera- 2+ patients had a statistically significant decrease in pCR if
py. Objective: This study aimed to analyze the impact of the surgery was performed after ≥8 weeks. Conclusion: Our pa-
time taken to initiate surgical treatment following comple- tients showed improved pCR if surgery was performed with-
tion of neoadjuvant chemotherapy on patients’ outcomes in 8 weeks, especially for ER+/HER-2+ patients. All patients
by evaluating their pathological response, overall survival had better OS and DFS trends if surgery was performed be-
(OS), and disease-free survival (DFS). Methods: This is a ret- tween 4 and 7 weeks after neoadjuvant chemotherapy.
rospective review of 611 patients diagnosed with stage II © 2020 S. Karger AG, Basel
and III breast cancer that received neoadjuvant chemother-
apy and surgery between January 2004 and December 2014.
The data was collected from a prospectively gathered regis- Introduction
try. The patients were stratified into three cohorts according
to the time of surgery after neoadjuvant chemotherapy: <4 Surgical resection is regarded as one of the main treat-
weeks, 4–7 weeks, or ≥8 weeks. Outcomes were assessed us- ment options in treating breast cancer [1]. The optimal
ing Kaplan-Meier curves, and the variables were compared time of surgical intervention in breast cancer varies de-
using log-rank statistics. Results: The 5-year OS rate was pending on whether surgery is the initial treatment after
89.6% and the 5-year DFS rate was 74%. OS and DFS were not diagnosis or it follows neoadjuvant chemotherapy.
significantly different when stratified according to timing of

© 2020 S. Karger AG, Basel Prof. Kausar Suleman

Oncology Center, King Faisal Specialist Hospital and Research Center
Zahrawi St, Al Maather
E-Mail karger@karger.com
Riyadh 12713 (Kingdom of Saudi Arabia)
www.karger.com/ocl E-Mail ksuleman @ kfshrc.edu.sa
 A review of the literature on the timing of treatment Subjects and Methods
after diagnosis found that delays have an impact on dis-
This is a retrospective review of data collected from a prospec-
ease-free survival (DFS), disease-specific survival, and tively gathered registry at King Faisal Specialist Hospital on breast
overall survival (OS) [2]. The review concluded that the cancer patients treated between January 2004 to December 2014.
optimal time from diagnosis to surgery should be <90 Patients included were female breast cancer patients diagnosed
days, while the optimal time for initiating chemotherapy with stage II and III, locally advanced breast cancer over the age of
after the time of diagnosis ought to be <120 days [2]. 18 years who had been initially treated with neoadjuvant chemo-
therapy. The neoadjuvant chemotherapy treatment regimen was
Neoadjuvant chemotherapy has become an option for FEC (5-FU, epirubicin, and cyclophosphamide) and Taxotere ±
locally advanced breast cancer as a means of downstaging Herceptin (depending on the HER-2 status of the disease). Patients
the tumor to allow for breast conservation, or in assessing were excluded from the study if they had undergone adjuvant che-
the efficacy of systemic therapy by the tumor response motherapy, had stage IV disease, or had undergone surgical exci-

Downloaded from http://karger.com/ocl/article-pdf/98/3/168/3297560/000504964.pdf by Siriraj Medical Library, Mahidol University user on 14 August 2023
[3]. It was further seen that patients with triple-negative sion of the primary tumor at an outside institution. A total of 611
patients were included in the analysis.
breast cancer (TNBC) who received neoadjuvant chemo- Breast cancer patient’s medical records were reviewed, and data
therapy had higher rates of pathological complete re- were collected according to a predesigned data sheet. The clinical
sponse (pCR), and had excellent survival rates when com- data included patient age, clinical stage, neoadjuvant chemothera-
pared to those of non-TNBC patients [4]. py regimen, and the surgical procedure performed on the patient.
Furthermore, a pooled analysis of 7 randomized trials The histopathological data included histological type, grade, lymph
node status, hormonal status, pathological nodal stage, and patient
of patients receiving neoadjuvant chemotherapy that survival data.
looked at long-term outcomes of 6,377 patients found The patients were stratified into three cohorts. The first cohort
that pCR, defined as no residual invasive or in situ disease (A) comprised patients who received surgery following neoadjuvant
in both the breast and the axilla, was seen to have a posi- chemotherapy within ≤4 weeks, the second cohort (B) received it
tive prognostic impact on survival in patients with lumi- within 4–7 weeks, and the third cohort (C) after ≥8 weeks. The sig-
nificance of the timing of surgery was evaluated on the basis of path-
nal B/Her2–, Her2+, or triple-negative disease [5]. This ological response, OS, and DFS. Furthermore, we differentiated
highlights the importance of the biology of the disease pathological response and outcome depending on the receptor sta-
and pCR for the outcome of breast cancer. tus: ER+/HER-2–, ER–/HER-2–, ER+/HER-2+, and ER–/HER-2+.
Several studies were done assessing the impact of initiat- OS was defined as the time from administration of treatment to
ing adjuvant chemotherapy after surgery, and they showed the death of the patient (or loss to follow-up). DFS was defined as
the time from diagnosis to the development of metastatic disease
that delays affected survival outcomes [6, 7]. A retrospective assessed via radiological intervention. pCR was defined as no re-
review that looked at 2,594 patients found that relapse-free sidual invasive or in situ disease in both the breast and the axilla.
survival and OS of patients were worse if chemotherapy was OS and DFS were evaluated using Kaplan-Meier curves, and the
started >12 weeks after definitive surgery [8]. A meta-anal- variables were compared using log-rank statistics. The time from
ysis and systematic review found that a 31- to 60-day delay the last dose of neoadjuvant therapy to the time of surgery was re-
corded in weeks. Variables were selected based on statistical and
of starting adjuvant chemotherapy after surgery led to clinical significance, and this included the hormone receptor status
worse OS among TNBC patients, but had no effect in the and pCR. All p values were two-tailed and considered statistically
case of hormone receptor-positive breast cancer [9]. significant at <0.05. Statistical analysis was performed using SPSS
While there are many studies analyzing the impact of version 20.
a delay in starting adjuvant chemotherapy, there is a pau-
city of literature examining the impact of timing of sur-
gery after neoadjuvant chemotherapy. A recent retro- Results
spective study that analyzed data on 1,101 patients found
that patients who underwent surgery after 8 weeks had A total of 611 patients were identified; of these, 94 pa-
worse outcomes [10]. Another retrospective review of tients (15.4%) had had surgery within the first 4 weeks,
319 patients found that if surgery was performed 21 days 424 patients (69.4%) within 4–7 weeks, and 93 patients
after neoadjuvant chemotherapy, then the patients had (15.2%) after ≥8 weeks. Within the time period of this
worse OS and relapse-free survival [11]. study, 93 patients died from any cause.
In this context, the aim of this study was to retrospec- The patient and tumor characteristics are summarized
tively analyze the impact of the timing of surgery after in Table 1. The median age at diagnosis was 44 years
neoadjuvant chemotherapy on outcomes such as DFS (range, 23–83). Most patients (99.7%) had invasive ductal
and OS, and to assess whether a certain subset of patients carcinoma. Of the 611 patients in the study, 261 (42.7%)
would benefit from earlier surgery. were ER+/HER-2–, 108 patients (17.7%) were ER–/HER-

Timing of Surgery in Breast Cancer Oncology 2020;98:168–173 169

Patients DOI: 10.1159/000504964
Table 1. Patient, clinical, and treatment characteristics at baseline stratified by time to surgery after neoadjuvant chemotherapy

All patients <4 weeks 4–7 weeks ≥8 weeks p value

(N = 611) (n = 94) (n = 434) (n = 93)
n % n % n % n %

Age
<40 years 219 35.8 30 31.9 156 36.8 33 35.5 0.219
40–50 years 238 39.0 36 38.3 158 37.3 44 47.3
>50 years 154 25.2 28 29.8 110 25.9 16 17.2
Clinical tumor stage
T1 6 1.0 1 1.1 3 0.7 2 2.2 0.590

Downloaded from http://karger.com/ocl/article-pdf/98/3/168/3297560/000504964.pdf by Siriraj Medical Library, Mahidol University user on 14 August 2023
T2 155 25.4 21 22.3 111 26.2 23 24.7
T3 220 36.0 37 39.4 156 36.8 27 29.0
T4 227 37.2 35 37.2 151 35.6 41 44.1
Tx 3 0.5 0 0.0 3 0.7 0 0.0
Clinical stage
IIA 30 4.9 6 6.4 20 4.7 4 4.3 0.822
IIB 141 23.1 19 20.2 101 23.8 21 22.6
IIIA 191 31.3 30 31.9 137 32.3 24 25.8
IIIB 245 40.1 38 40.4 163 38.4 44 47.3
Unstageable 4 0.7 1 1.1 3 0.7 0 0.0
Histopathological grade
G1 7 1.1 3 3.2 4 0.9 0 0.0 0.321
G2 293 48.0 49 52.1 199 46.9 45 48.4
G3 286 46.8 40 42.6 201 47.4 45 48.4
Gx 25 4.1 2 2.1 20 4.7 3 3.2
Hormone receptor status
ER+/PR+ 294 48.1 37 39.4 215 50.7 42 45.2 0.304
ER+/PR– 78 12.8 11 11.7 51 12.0 16 17.2
ER–/PR+ 7 1.1 2 2.1 4 0.9 1 1.1
ER–/PR– 232 38.0 44 46.8 154 36.3 34 36.6
HER2– 392 64.2 55 58.5 280 66 57 61.3 0.319
HER2+ 219 35.8 39 41.5 144 34.0 36 38.7
ER+/HER2– 261 42.7 33 35.1 189 44.6 39 41.9 0.355
ER–/HER2+ 108 17.7 24 25.5 67 15.8 17 18.3
ER+/HER2+ 111 18.2 15 16.0 77 18.2 19 20.4
ER–/HER2– 131 21.4 22 23.4 91 21.5 18 19.4
Type of treatment received
Anthracyclines 392 64.2 55 58.5 280 66.0 57 61.3 0.319
Anthracyclines + trastuzumab 219 35.8 39 41.5 144 34.0 36 38.7

2+, 111 patients (18.2%) were ER+/HER-2+, and 131 months, with a range of 40–92 months. The 5-year OS
(21.4%) were ER–/HER-2–. rate was estimated to be 88.8%, 85.9%, and 70% among
The median time to surgery from diagnosis was 27 the patients who had surgery within the first 4 weeks,
weeks, and the median time to surgery following comple- within 4–7 weeks, and after ≥8 weeks, respectively. How-
tion of neoadjuvant chemotherapy was 5 weeks. ever, the difference between rates was not statistically sig-
Among all patients, the OS rate at 5 years was 89.6% nificant. Figure 1b and d indicates the Kaplan-Meier es-
(Fig. 1a) and the DFS rate at 5 years was 74% (Fig. 1c). timates of OS and DFS for the three different groups.
Median OS and median DFS have not yet been reached; Only 12.9% of the patients who received surgery after
OS and DFS were not significantly different when strati- ≥8 weeks had pCR; this is in comparison to a pCR of 26%
fied according to the time of surgery; however, the trends among the patients who received surgery within 4–7
of DFS and OS were poor when surgery was delayed for weeks after completion of neoadjuvant chemotherapy
≥8 weeks. The median duration of follow-up was 62 (p = 0.02) (Table 2).

170 Oncology 2020;98:168–173 Suleman et al.

DOI: 10.1159/000504964
 Color version available online
Cumulative overall survival probability Cumulative disease-free survival probability

1.0 1.0

0.8 0.8
Cumulative survival

Cumulative survival
OS = 89.6%
0.6 0.6
DFS = 74%

0.4 0.4

Downloaded from http://karger.com/ocl/article-pdf/98/3/168/3297560/000504964.pdf by Siriraj Medical Library, Mahidol University user on 14 August 2023
0.2 0.2

0 0

0 50 100 150 200 0 50 100 150 200

a Time, months c Time, months

Cumulative overall survival probability Cumulative disease-free survival probability

1.0 1.0 <4 weeks
<4 weeks
4–7 weeks 4–7 weeks
≥8 weeks ≥8 weeks

0.8 0.8
Cumulative survival

Cumulative survival

0.6 0.6

0.4 0.4

0.2 0.2

p value = 0.5 p value = 0.3

0 0

0 50 100 150 200 0 50 100 150 200

b Time, months d Time, months

Fig. 1. Kaplan-Meier curves comparing cumulative overall survival in all three cohorts (a) and in the subgroups
(b), as well as disease-free survival in all three cohorts (c) and in the subgroups (d).

In terms of receptor status, a Pearson χ2 test performed Discussion

for ER+/HER-2+ patients was found to be statistically sig-
nificant for this subgroup of patients. They had a de- Timing of adjuvant chemotherapy has been shown
creased pCR if surgery was performed ≥8 weeks after neo- to affect outcome in breast cancer patients, and a clear
adjuvant chemotherapy (Table 3). However, the ER+/ window of time has been defined. However, the win-
HER-2–, ER–/HER-2+, and ER–/HER-2– patients had dow of time for surgery after neoadjuvant chemothera-
no statistically significant difference in pCR. py still remains to be defined. In our review, we found

Timing of Surgery in Breast Cancer Oncology 2020;98:168–173 171

Patients DOI: 10.1159/000504964
Table 2. pCR among patients who received surgery <4 weeks, 4–7 Table 3. pCR in relation to ER+/HER-2+ biological subtype pa-
weeks, and ≥8 weeks following neoadjuvant chemotherapy tients who received surgery <4 weeks, 4–7 weeks, and ≥8 weeks
following neoadjuvant chemotherapy
Patients <4 weeks 4–7 weeks ≥8 weeks p value
Patients <8 weeks ≥8 weeks p value
No pCR 74 (78.7%) 313 (73.8%) 81 (87.1%) 0.020
pCR 20 (21.3%) 111 (26.2%) 12 (12.9%) No pCR 61 (66.3%) 17 (89.5%) 0.044
pCR 31 (33.7%) 2 (10.5%)
Total 94 (100%) 424 (100%) 93 (100%)
Total 92 (100%) 19 (100%)
pCR, pathological complete response.
pCR, pathological complete response.

Downloaded from http://karger.com/ocl/article-pdf/98/3/168/3297560/000504964.pdf by Siriraj Medical Library, Mahidol University user on 14 August 2023
that all patients had better OS and DFS trends if surgery survival rates between patients with surgery <4 weeks,
was performed between 4 and 7 weeks after neoadju- 4–6 weeks, and >6 weeks after neoadjuvant chemothera-
vant chemotherapy. It was also seen that the pCR rates py [10].
were higher if surgery was performed within 8 weeks, In contrast, another study that looked at the impact of
especially for ER+/HER-2+ patients. surgery after neoadjuvant chemotherapy among 319 pa-
Several factors determine scheduling for surgery; tients divided the patients into two groups: 61 patients in
this could be related to the time needed to recover from group A who received surgery within 21 days, and 258
the toxicity of the systemic therapy to withstand sur- patients in group B that had surgery after 21 days, with
gery, patient comorbidities, and coordination with re- the majority of the patients having surgery within 5 weeks
constructive surgeons. All these factors may cause pa- [11]. It was seen that OS (hazard ratio = 3.1, 95% CI 1.1–
tients to experience anxiety and to consider this wait 8.6, p = 0.03) and relapse-free survival (hazard ratio = 3.1,
time a delay in their treatment plan. However, these 95% CI 1.3–7.1, p = 0.008) was significantly lower among
wait times may be necessary and may not impact the the patients who received surgery after 21 days (group B)
patients’ outcome in the setting after neoadjuvant che- than among those in group A [11]. One of the limitations
motherapy. of that study, however, is the division of the cohort, with
In a SEER review of 95,544 patients that analyzed the 81% being in group B [11].
time from diagnosis to breast cancer surgery before sys- In our analysis, all patients showed a better pCR rate,
temic therapy, Bleicher et al. [12] demonstrated that a as well as improved OS and DFS trends, if surgery was
delay of over 60 days led to lower OS among patients performed before 8 weeks, especially for ER+/HER-2+ pa-
with stage I and II disease but not among those with tients, with pCR indirectly translating into improved OS
stage III disease. However, a possible reason for the dif- in some subtypes of breast cancer. Delaying surgery for >8
ference in findings in stage III patients could be that weeks showed a lower pCR, as was demonstrated not only
these patients tend to receive neoadjuvant rather than in our study but also in the MD Anderson Cancer Center
adjuvant chemotherapy as the first modality of treat- review of the impact of time of surgery after neoadjuvant
ment, hence explaining the findings seen in this article. chemotherapy [10]. A study done by Wagner et al. [13]
This suggests that the timing of starting definitive treat- that looked at the effect of time to primary surgery follow-
ment – whether surgery or systemic therapy – impacts ing breast cancer diagnosis showed that delays in surgery
survival. were not associated with tumor size progression.
The median time to surgery following completion of However, a recent study analyzing the impact of time
neoadjuvant chemotherapy was 5 weeks in our cohort. to starting adjuvant chemotherapy after surgery found
This is similar to what was found in a review of the im- that delaying the start of adjuvant chemotherapy by >120
pact of surgery time after neoadjuvant chemotherapy on days after diagnosis was associated with worse OS (hazard
survival that was conducted by Sanford et al. [10]; the ratio = 1.29, p < 0.001) [14].
majority of their cohort had had surgery between 4 and Furthermore, a clinical feasibility study was done eval-
6 weeks after neoadjuvant chemotherapy. Furthermore, uating omitting surgery for patients with pCR for neoad-
in their multivariate analysis, they found equivalent OS, juvant chemotherapy [15]. That study has led to a clinical
locoregional recurrence-free survival, and relapse-free trial that is enrolling patients with triple-negative disease

172 Oncology 2020;98:168–173 Suleman et al.

DOI: 10.1159/000504964
and her2/neu+ disease who receive neoadjuvant chemo- gery was performed before 8 weeks, especially for ER+/
therapy and attained complete radiological and biopsy- HER-2+ patients.
proven pathological response to omit surgery [15].
While the timing of surgery for breast cancer from the
time of diagnosis until definitive surgery is important, Statement of Ethics
multiple studies have shown that the time of starting (ei- The study was carried out in compliance with the principles
ther adjuvant or neoadjuvant) chemotherapy from diag- laid down in the Declaration of Helsinki. The study was initiated
nosis affects OS. This suggests that the biology of breast only after all required legal documentation was reviewed and ap-
cancer is important and that the timing of starting sys- proved by the respective institutional review board/independent
ethics committee and competent authority according to national
temic therapy is more relevant to survival than the timing
and international regulations.
of surgery, in the neoadjuvant setting.

Downloaded from http://karger.com/ocl/article-pdf/98/3/168/3297560/000504964.pdf by Siriraj Medical Library, Mahidol University user on 14 August 2023
Limitations Disclosure Statement
This study is retrospective in nature and was conducted
at a single large tertiary care center. The patients’ initial dis- The authors have no conflict of interest nor funding to declare
related to this study.
ease status could have been the reason for the response ob-
served, and not necessarily the timing of surgery. Finally,
our study did not take into consideration the new chemo- Funding Sources
therapy protocols, which include dose-dense and weekly
chemotherapy where recovery of the bone marrow is faster. The authors received no funding for this study.

Author Contributions
Conclusion
The authors are fully responsible for all content and editorial
All patients had better OS and DFS trends if surgery decisions, were involved in all stages of manuscript development,
was performed between 4 and 7 weeks after neoadjuvant and have approved the final version of the paper.
chemotherapy. Higher pCR rates were observed if sur-

References
1 Rostas JW, Dyess DL. Current operative man- vant chemotherapy in patients with breast erable breast cancer patients. Eur J Surg On-
agement of breast cancer: an age of smaller cancer. J Clin Oncol. 2014;32(8):735–44. col. 2017;43(4):613–8.
resections and bigger cures. Int J Breast Can- 7 Chavez-MacGregor M, Clarke CA, Lichten- 12 Bleicher RJ, Ruth K, Sigurdson ER, Beck JR,
cer. 2012;2012:516417. sztajn DY, Giordano SH. Delayed initiation of Ross E, Wong YN. Time to surgery and breast
2 Bleicher RJ. Timing and delays in breast can- adjuvant chemotherapy among patients with cancer survival in the United States. JAMA
cer evaluation and treatment. Ann Surg On- breast cancer. JAMA Oncol. 2016;2(3):322–9. Oncol. 2016;2(3):330–9.
col. 2018;25(10):2829–38. 8 Lohrisch C, Paltiel C, Gelmon K, Speers C, 13 Wagner JL, Warneke CL, Mittendorf EA,
3 Thompson AM, Moulder-Thompson SL. Taylor S, Barnett J. Impact on survival of time Bedrosian I, Babiera GV, Kuerer HM. Delays
Neoadjuvant treatment of breast cancer. Ann from definitive surgery to initiation of adju- in primary surgical treatment are not associ-
Oncol. 2012;23 Suppl 10:x231–6. vant chemotherapy for early-stage breast can- ated with significant tumor size progression
4 Liedtke C, Mazouni C, Hess KR, André F, cer. J Clin Oncol. 2006;24(30):4888–94. in breast cancer patients. Ann Surg.2011;
Tordai A, Mejia JA. Response to neoadjuvant 9 Zhan QH, Fu JQ, Fu FM, Zhang J, Wang C. 254(1):119–24.
therapy and long-term survival in patients Survival and time to initiation of adjuvant 14 Kupstas AR, Hoskin TL, Day CN, Haber-
with triple-negative breast cancer. J Clin On- chemotherapy among breast cancer patients: mann EB, Boughey JC. Effect of surgery type
col. 2008;26(8):1275–81. a systematic review and meta-analysis. Onco- on time to adjuvant chemotherapy and im-
5 von Minckwitz G, Untch M, Blohmer JU, target. 2017;9(2):2739–51. pact of delay on breast cancer survival: a Na-
Costa SD, Eidtmann H, Fasching PA. Defini- 10 Sanford RA, Lei X, Barcenas CH, Mittendorf tional Cancer Database analysis. Ann Surg
tion and impact of pathologic complete re- EA, Caudle AS, Valero V. Impact of time from Oncol. 2019;26(10):3240–9.
sponse on prognosis after neoadjuvant che- completion of neoadjuvant chemotherapy to 15 Kuerer HM, Rauch GM, Krishnamurthy S,
motherapy in various intrinsic breast cancer surgery on survival outcomes in breast cancer Adrada BE, Caudle AS, DeSnyder SM. A clin-
subtypes. J Clin Oncol. 2012; 30(15): 1796– patients. Ann Surg Oncol. 2016; 23(5): 1515– ical feasibility trial for identification of excep-
804. 21. tional responders in whom breast cancer sur-
6 Gagliato Dde M, Gonzalez-Angulo AM, Lei 11 Omarini C, Guaitoli G, Noventa S, Andreotti gery can be eliminated following neoadjuvant
X, Theriault RL, Giordano SH, Valero V. A, Gambini A, Palma E. Impact of time to sur- systemic therapy. Ann Surg.2018;267(5):946–
Clinical impact of delaying initiation of adju- gery after neoadjuvant chemotherapy in op- 51.

Timing of Surgery in Breast Cancer Oncology 2020;98:168–173 173

Patients DOI: 10.1159/000504964