Introduction

The utilization of neoadjuvant chemotherapy (NACT) in breast cancer has been well established, especially after randomized controlled trials showing equal outcomes in survival compared to adjuvant chemotherapy [1, 2]. Not only did NACT allow downsizing the tumor in a group of patients who were otherwise not suitable for breast conservation [3], it also assisted in testing the tumor biology in vivo which allowed the evaluation of systemic therapy effects and tumor response. Furthermore, it has also allowed downstaging the clinically positive axilla in exceptional responders, thus performing less morbid surgeries [4, 5].

In Jordan, breast cancer is the most common cancer and the leading cause of cancer-related mortality in women, comprising 19.7% of all diagnosed cancer cases [6]. Moreover, about a third of patients present with locally advanced and or metastatic disease at a median age of 50, which is about a decade earlier than in the west. This makes the use of NACT more paramount in our population [7, 8]. The B-27 protocol (anthracycline taxanes) NACT has been used in our Jordanian patients with established efficacy and safety [9].

Limited data is available regarding the optimal timing to perform surgery after completing NACT, and major studies that addressed this issue showed contrasting results [10,11,12,13]. However, several published studies investigated the suitable time to initiate adjuvant chemotherapy, which lead several bodies of oncology to propose not delaying systemic adjuvant chemotherapy more than 6 weeks post-surgery [14,15,16,17,18,19].

In this retrospective comparative study, we aim to find the optimal time between completing NACT and having the surgery in relation to disease-free (DFS) and overall survival (OS).

Methods

Patient population

A retrospective chart review of 468 patients treated with NACT followed by surgery at King Hussein Cancer Center (KHCC) between 2006 and 2014. All patients had pathologically proven breast cancer, who received and completed the same regimen of NACT (B27 protocol [1]). Anti-Her2 agent (Trastuzumab) was added if the patients were eligible. Patients with de novo metastasis were excluded. Time to surgery was defined as the duration between the last cycle of NACT and the date of tumor removal. All patients were discussed in our specialized breast cancer multidisciplinary meeting, where recommendations were reached based on mammograms, histology and clinical presentation.

The study was conducted at KHCC Amman, Jordan, approved by the Institutional Review Board (IRB) with approval Number (17 KHCC 49).

Operative technique

All patients who underwent curative surgery were included in our analysis irrespective of the surgical technique involved.

Data collection and follow up

Patient demographics and medical histories were extracted pre-chemotherapy. We followed up with our patients with six monthly mammograms for 2 years and then every year thereafter and an annual computed tomography chest/abdomen/pelvis for 5 years.

Patients were divided into three groups according to the time to surgery: group 1: <4 weeks; group 2: 4–8 weeks; and group 3: >8 weeks. DFS and OS were calculated from the date of diagnosis to the date of recurrence or death, respectively.

Statistical analysis

Patients’ characteristics such as histology, stage and hormone status were presented in terms of counts and percentages. Duration between NACT and surgery was divide into three sequenced ranges based on ROC optimal thresholds and AUC. Accordingly, duration between NACT and surgery (<4, 4–8 and ≥8 weeks) was compared using Chi-square test, Fishers exact or Wilcoxon rank as appropriate and based on each assumption. DFS and OS were presented by Kaplan Meier curves and the comparisons were performed using Log Rank test. Significant factors were included in Cox regression and Adjusted Hazards ratios, their 95% CI and Adjusted p values were reported. All analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC).

Results

The analysis included 468 patients who underwent NACT followed by surgery between 2006 and 2014. Seventy-five patients (16%) were dead at the time of the analysis. Complete pathological response was achieved in 20% of patients, the local recurrence rate was 6.6% (n = 31) and the distant metastasis rate was 22% (n = 102). Median follow up for the whole cohort was 46 months (interquartile range 7–131). There were 142 patients in the <4 weeks group, 284 patients in the 4–8 weeks group, and 42 patients in the >8 weeks group. Their clinical characteristics are shown in Table 1. The three groups were not statistically different regarding age, cancer histology, grade, hormonal receptor status, the use of adjuvant hormonal or radiotherapy or the pathologic complete response (pCR). And the <4 weeks groups had more percentage of patients with lymphovascular invasion (LVI) (p < 0.03) and long duration of follow up (median) 4.6 years with (p = 0.022) and the >8 weeks group had fewer patients who initially presented with stage III breast cancer (p < 0.0001).

Table 1 Patient characteristics
Full size table

The 5-year survival rates were 89%, 77%, and 85.6% for the groups <4 weeks, 4–8 weeks and >8 weeks, respectively. The Kaplan Meier curves for OS and DFS for all groups are shown in Figs. 1 and 2.

Fig. 1
figure 1

The Kaplan Meier Plot for all patient groups’ overall survival. Duration is between the neoadjuvant therapy and surgeryThe Kaplan Meier Plot for all patient groups’ overall survival. Duration is between the neoadjuvant therapy and surgery

Full size image
Fig. 2
figure 2

The Kaplan Meier Plot for all patient groups’ event free survival. Duration is between the neoadjuvant therapy and surgery

Full size image

In the univariable analysis, Table 2, TNM clinical and pathological stage, LVI, nuclear grade, pathological response and Hormonal receptor status affected DFS. Whereas time to surgery, TNM clinical and pathological stage, LVI, nuclear grade and Hormonal receptor status were statistically significant factors for OS. Breast reconstruction did not have an influence on DFS nor OS.

Table 2 Univariate analysis, disease free survival and overall survival
Full size table

In the multivariate Cox regression, Table 3, only the presenting clinical stage (III vs. I/II) and LVI were shown to be associated with worse OS, whereas in DFS, grade, LVI and the clinical stage were shown to be significant detrimental factors. Time to surgery did not have an influence on neither DFS nor OS.

Table 3 Multivariate Cox regression analysis showing overall survival and disease-free survival of all patients
Full size table

In comparing the groups’ OS and DFS (Tables 4 and 5, respectively), patients who underwent surgery beyond 8 weeks post NACT had worse OS than the 4–8 weeks group.

Table 4 A comparison between the overall survival between patient groups
Full size table
Table 5 A comparison between disease free survival among patient groups
Full size table

Discussion and conclusion

This retrospective comparative study aims to study the effect of time of breast cancer surgery after NACT on patients’ survival. Our data showed that when comparing patients who had surgery beyond 8 weeks to patients who had the surgery between 4 and 8 weeks, the former group showed worse OS. Also, when the >8 weeks group was compared to the combined cohort of <8 weeks post treatment, there was a statistical trend towards worse OS.

Scheduling patients who completed NACT for surgery is dictated by many factors; toxicity from chemotherapy, which was shown to be the major cause of delay to surgery [3], patients’ medical status and anxiety can play a role in delaying surgery. In addition, operating theatre slot availability and surgical waiting lists might also be taken into consideration.

In the literature, there is no clear-cut evidence on the optimal time to perform the surgery post NACT.

A published study by Sanford et al. was the first to address this issue [10]. It was a retrospective review of 1101 patients diagnosed with stages I, II and III. Time between NACT and surgery was categorized as <4 weeks (30.4% of patients), 4–6 weeks (47.6%), or >6 weeks (22%). The 5-year OS estimate was 79%, 87% and 81% respectively (p = 0.03). They did not find a difference between the three groups in terms of DFS. Patients who underwent surgery at ≤4 or >6 weeks had worse OS but not DFS.

Another study from Modena Italy showed a better OS and DFS outcome if surgery was performed before the end of 3 weeks post treatment. However the groups were mismatched in number [11]. Authors from the Kaiser Permanente Los Angeles Medical Center conducted their own review on 58 patients and found that delaying surgery beyond 60 days affects recurrence but not survival [12].

As far as we know, this is the second analysis of its kind in our region, the Middle East, outside of Europe and the United States. One of our aims was to know if findings could be replicated in our specific cohort of patients.

The retrospective nature of this study may have an adverse impact on the outcome especially with a small sample size in the >8 weeks group. It is also a single center experience and overgeneralizing results should be taken with caution. Theoretically to define the optimal time post NACT to surgery, prospective randomized trials are needed which are perhaps very difficult to conduct as they may harbor ethical issues.

With the limited data available on the subject, delaying surgery up to 8 weeks does not impact outcomes and this might give surgeons reassurance regarding scheduling their patients for surgery.