Introduction

Breast cancer is the most common type of cancer in women [1]. Proper adjuvant treatment is necessary in indicated patients. Adjuvant treatment of breast cancer includes chemotherapy, hormonal treatment, targeted therapy and radiotherapy based on clinical and pathological characteristics of the tumor [1–5]. Although the adjuvant treatment of breast cancer has been well described, optimal timing of adjuvant treatment is not well known. It has been reported that starting time of the adjuvant treatment after surgery may affect either wound healing or prognosis of breast cancer. Some reports have shown that earlier chemotherapy has effects on drug resistance, micrometastasis and angiogenesis [6]. In the 1980s, the International Breast Cancer Study Group showed a benefit in starting CT earlier by CMF regimen in premenopausal breast cancer patients [7]. Heterogeneity of studies in the literature as a result of differences in prognostic factors of patient populations (age, menopausal status, grade, pathologic subtype, hormone receptor status, HER2, p53 etc.) may affect the role of adjuvant starting time for prognosis of breast cancer [2, 8–13]. Shannon et al. reported that adjuvant chemotherapy timing had no effect on survival in early breast cancer patients with breast sparing surgery [14]. Some benefits have been reported if adjuvant chemotherapy started before the 12th week after surgery [15]. This probable benefit may result from several factors. Micrometastasis has been shown to be doubled after surgery in animal models [16]. This effect may depend on serum transmissible growth factors [16, 17]. On the other hand, angiogenesis has been shown to increase metastasis on animal models after primary tumor had resected by decreasing angiostatin, which is a plasminogen fragment [16, 17]. Theoretically, early adjuvant treatment may inhibit this early rebound effect after surgery on micrometastatic cancer cells. In this study, we investigated optimal timing of adjuvant treatment in patients with early breast cancer.

Materials and methods

Data were collected retrospectively from 642 breast cancer patients who received chemotherapy and/or hormonal treatment at Ankara Oncology Research and Training Hospital between January 1995 and August 2002. All cases were pathologically documented invasive breast cancer in the same center. Patients receiving neo-adjuvant chemotherapy or neo-adjuvant hormonal treatment, adjuvant chemotherapy with unknown starting time, patients with primary metastatic disease and with follow-up period shorter than 6 months were excluded from the study. Four hundred and two patients were eligible and evaluated. Data on prognostic factors and treatment were collected prospectively including age, tumor size, histological type, tumor grade, hormone receptor status, number of involved lymph nodes, adjuvant irradiation to chest wall and lymph nodes, adjuvant tamoxifen and type of chemotherapy.

Chemotherapy or hormonal treatment starting time after surgery is described as delaying period. All eligible patients were classified by delaying period according to following two methods: first, patients were divided into five groups according to delaying period as shorter than 14 days (Group A, n = 130, 32.3%), between days 15 and days 29 (Group B, n = 148, 36.8%), between days 30 and days 44 (Group C, n = 66, 16.4%), between days 45 and days 59 (Group D, n = 31, 7.7%), more than 59 days (Group E, n = 27, 6.7%). Second, patients were divided into two groups as starting adjuvant treatment equal to or shorter than 44 days and longer than 44 days (n = 344, 85.6% and vs. n = 58, 14.4%, respectively). Disease-free survival (DFS) and overall survival (OS) were calculated.

Statistical analysis

Differences between the five groups were assessed by means of the chi-square test. DFS and OS were calculated using Kaplan–Meier and log-rank tests. Cox regression analysis was used for the prognostic and other treatment variables.

Results

Four hundred and two patients who received adjuvant chemotherapy and/or hormonal treatment were evaluated. Median follow-up was 50 months (range, 6–105 months). Two hundred and four (50.7%) patients received only adjuvant chemotherapy, 45 (11.2%) patients received only adjuvant tamoxifen, and 153 (38.0%) patients received tamoxifen after adjuvant chemotherapy. Two hundred and twenty-three (55.5%) patients received CMF i.v., and 129 (32.1%) patients received anthracycline-based chemotherapy. Axillary lymph node involvement was negative in 141(35.1%) patients, 1–3 positive in 110 (27.3%) patients, 4–9 positive in 83 (20.2%) patients and more than 10 nodes in 60 patients (14.9%). Patients’ characteristics and adjuvant treatment schedules are shown in Table 1. Two hundred and eighty (69.6%) of the 402 patients were without relapse during follow-up. Hundred (24.9%) patients were alive with relapse, and 22 (5.5%) patients died due to disease progression. Metastatic sites were skin (23 patients), bone (45 patients), lung (44 patients), liver (22 patients), brain (6 patients) and supraclavicular or contralateral axillary (17 patients). Six patients who were without metastasis after surgery and adjuvant treatment developed secondary contralateral breast cancer. These six patients are still without relapse after treatment for secondary breast cancer.

Table 1 Clinical and pathological characteristics and adjuvant treatment schedules of 402 patients with breast cancer
Full size table

Median time to starting time of adjuvant treatment after surgery is day 21 (range, day 4 to day 258). Most of the patients have received first cycle of chemotherapy between second and fourth weeks. Patients’ distribution according to starting time of adjuvant treatment is shown in Fig. 1.

Fig. 1
figure 1

Patient distribution according to starting time of adjuvant treatment after surgery

Full size image

DFS for all patients was 89 and 84% at third and fifth years, respectively. OS for all patients was 97 and 95% at third and fifth years, respectively. The most frequent metastatic sites in decreasing order were bone, lung, skin and liver. Negative receptor status, tumor size and axillary involvement were found to be significantly negatively correlated with prognosis (P < 0.016, P < 0.05, and P < 0.001, respectively) for all patients.

Patient subgroups were similar according to age, menopausal status, receptor status, tumor size, axillary status, surgery type and adjuvant treatments (P > 0.05) according to two methods. OS and DFS were not shown significantly different between 5 groups (P > 0.05) as described in “Materials and method” section. One hundred and twenty-nine patients (32.1%) received anthracycline-based chemotherapy, two hundred and twenty-three (55.5%) patients received CMF regimen, and five patients were given other chemotherapy regimen (P = 0.78).

The patients who received adjuvant treatment within 44 days after surgery had significantly superior OS compared to the patients who received adjuvant treatment 44 days after surgery (92% vs. 83.3% for 5 years, respectively, P = 0.03, Fig. 2) but not significant for DFS. The patients receiving CMF i.v. or anthracycline-based chemotherapy (excluding only adjuvant hormonal treatment patients) with shorter delaying period (shorter or equal to 44 days) had significantly superior OS but not DFS for 5 years (91.4 vs. 83.4% P < 0.05 and 83.4 vs. 82.2% P = 0.03, respectively).

Fig. 2
figure 2

Overall survival according to adjuvant chemotherapy starting time before and after day 44

Full size image

Discussion

Some of the animal studies have shown that starting adjuvant treatment earlier has some benefit on micrometastasis, drug resistance and angiogenesis. Some clinical trials have shown a similar positive effect on survival [18–20], but others have not [14, 21]. In this study, two different methods were used for examining adjuvant treatment starting time. One of the methods used 5 subgroups. However, these 5 subgroups were not significantly different for OS or DFS. On the other hand, to find an optimal starting time of adjuvant treatment, patients were divided into two groups according to 44 days cut-off, which has shown a statistically significant difference in OS between the 2 groups (Fig. 2).

Pronzota et al. have reported that the patients whose adjuvant treatment started before 35 days after surgery had significantly better OS at 4 years with CMF regimen (88 vs. 69%; [19]). International Breast Cancer Study Group (IBCSG) has reported starting adjuvant chemotherapy before 20 days after surgery with CMF regimen had better outcome in OS (60 vs. 34% at 10 years) in node positive, premenopausal, ER negative 226 patients [18]. Moreover, IBCSG has evaluated adjuvant CMF with or without preoperative CMF regimen in node positive premenopausal women. In ER negative 101 patients, they showed better DFS in patients who received perioperative CT than ER + patients (48 vs. 38%) at median 11 years follow-up [7]. On the other hand, Sertoli et al. have shown a benefit with perioperative CEF regimen in ER negative patients, but this effect was not apparent in ER positive patients [22].

In another study, starting adjuvant treatment early at 21 days did not improve either OS or DFS in all patients or in the ER negative group [14]. Endocrine manipulation may act a more positive role than effect of earlier initiation of chemotherapy in ER positive patients. Starting chemotherapy earlier may improve outcomes especially in ER negative patients. However, our report did not show any difference between ER positive and ER negative or node positive and node negative patients. Furthermore, early chemotherapy may help initiating earlier hormonal treatment after chemotherapy, which is especially important in hormone responsive breast cancer patients.

In conclusion, our study showed that starting adjuvant systemic treatment within 44 days improves OS in patients with early breast cancer. A larger number of studies, especially those with subgroup analyses, are needed. Adjuvant treatment of breast cancer is very important for both DFS and OS. Some factors such as wound healing and performance status may be reason to delay adjuvant treatment. Our study puts forward an optimal starting time for adjuvant therapy and describes the possibility of poor outcomes if adjuvant therapy is delayed.