A chart review was retrospectively performed on 371 patients who underwent surgery for EBC in the Department of Breast Surgery of the First Hospital of China Medical University (CMU; Shenyang, China) between August 2011 and March 2016.

Inclusion criteria used were as follows: i) Patients with an established diagnosis of EBC, defined as histological diagnosis of operable stage I to stage III invasive ductal carcinoma with achievement of an R₀ resection; ii) patients aged between 18 and 75 years; and iii) patients who underwent standard surgical treatment and adjuvant chemotherapy. All patients included in the study received at least one of the following: Chemotherapy, radiotherapy, endocrine therapy and/or anti-human epidermal growth factor receptor 2 (HER2) therapy, according to the NCCN guidelines (6,7). This included those with N0 cancer, who received chemotherapy due to high-risk factors such as an estrogen receptor (ER) expression level of <50% and/or a Ki67 level ≥30%. PXC used by the study population was manufactured by Xi'an Charoen Pokphand Pharmaceutical, Co., Ltd., (batch no. 1712140), which is the sole manufacturer and distributor of PXC. Patients were divided into the PXC group (patients who received PXC treatment) and the non-PXC group (patients who did not receive PXC) per chart review.

The present study adjusted for confounders using the principle of restriction and matching. Subjects in the two cohorts (PXC vs. non-PXC) were matched in terms of age, tumor stage, axillary metastasis, hormonal receptor status, HER2 status and molecular subtypes to eliminate potential confounding factors. In addition, only those who had received PXC treatment within 1 year of diagnosis and had received standard dosing of PXC (4–8 capsules, three times a day for >1 month, for a maximum of 6 months; the dosage of each capsule is 230 milligrams) were included. Patients must also have completed any indicated standard treatment in addition to surgery (i.e. chemotherapy, radiotherapy, endocrine therapy accordingly) before PXC treatment, and those who did not comply with standard treatments were excluded from the study. Disease-free survival (DFS) was defined as the time from the date of surgery to either tumor recurrence or death due to any cause, and overall survival (OS) was defined as the interval between the date of surgery and the date of patient death or the last follow-up visit.

The study protocol was reviewed by the CMU Ethics Committee Institutional Review Board (approval no. 2020 NO.203). Approval was granted prior to the commencement of the study and the requirement for written consent was waived.

All subjects underwent chemotherapy. A total of 105 patients were treated with a docetaxel + anthracyclines + cyclophosphamide (TEC or EC-T) regimen, 12 patients were treated with an anthracyclines + cyclophosphamide (EC) regimen, 88 patients were treated with a docetaxel + cyclophosphamide (TC) regimen, 20 patients were treated with a docetaxel + platinum (DP) regimen, 121 patients were treated with an anthracyclines + cyclophosphamide + docetaxel + herceptin (EC-TH) regimen, 14 patients were treated with a 5-fluorouracil + anthracyclines + cyclophosphamide (CEF) regimen and 11 patients were treated with a CEF-T regimen. For all patients, treatments were initiated within 6 weeks of surgery according to NCCN. In addition, for patients with ER+ expression ≥10%, regardless of PR expression, endocrine therapy should be completed for 5 years. Data for AEs was obtained through chart review of routine documentation. All patients with breast cancer were routinely followed up in the Outpatient Clinic immediately before starting chemotherapy and all patients underwent routine laboratory tests. Patients were also followed up at an interval of 14 days after starting chemotherapy and then at the end of the first treatment cycle, 21 days after starting chemotherapy, during which they were admitted for observation. AEs were documented for every cycle of treatment by phone or during follow-up visits using the Common Terminology Criteria for Adverse Events version 4.0 (8). Treatment was either suspended or terminated for patients with AEs that were of grade 2 or higher during chemotherapy, and the period for which the agent was administered at the prescribed dose (not including the period when the dose was reduced) was recorded. Incidence of AEs, and the completion rate of the first cycle of chemotherapy at the prescribed dose were tabulated.

The primary objective of the present study was to assess the efficacy and survival benefit of PXC-based adjuvant therapy, and if this was significant, to identify prognostic factors for its use. Participant demographics and characteristic such as age, tumor stage, axillary metastasis status, hormonal receptor status, HER2 status and molecular subtypes were reviewed and tabulated. Descriptive statistics were reported as proportions and medians. Frequencies of tumor characteristics and response rates were compared using the χ² test or Fisher's exact test. DFS and OS were demonstrated using the Kaplan-Meier survival curve. The log-rank test was used to compare two or more survival curves, and Cox's proportional hazards regression models were used to analyze the independent predictors for recurrence. Survival duration was calculated starting from the time of surgery. P≤0.05 was considered to indicate a statistically significant difference. All statistical analyses were performed with SPSS 12.0 for Windows (SPSS Inc.).

A total of 371 patients (age range, 27–65 years; median age, 50) met the inclusion criteria and were included in this study. Patients were followed up from 22 to 137 months, with the median follow-up time being 101 months. During the follow-up period, 122 patients (32.9%) experienced recurrence and 73 patients (19.7%) died. There were 154 patients (41.5%) in the PXC treatment group and 217 (58.5%) in the non-PXC group. Of those in the PXC group, 62 (16.7% of total) patients received treatment for 1 month, and 92 (24.8% of total) patients received treatment for ≥3 months, with an overall range of treatment duration of 1–6 months.

Subjects in the two cohorts (PXC vs. non-PXC) were similar in terms of age (P=0.172), tumor stage (P=0.075), axillary metastasis (P=0.259), hormonal receptor status (P=0.206), HER2 status (P=0.520) and molecular subtype (P=0.509), as demonstrated in Table I. Most patients had tumor ER⁺ expression ≥10% (79% in PXC group, 73% in non-PXC group) and >70% of patients in each group had PR⁺ tumors. Of the patients in the PXC group with ER⁺ tumors, 100% patients received endocrine therapy when ER⁺ expression ≥10%, and of the patients in the non-PXC group with ER⁺ tumors, 96% received endocrine therapy. Most HER2⁺ patients (98.0% of patients in the PXC group and 96.1% of patients in the non-PXC group) received anti-HER2 therapy. All cases of TNBC received 6–8 cycles of chemotherapy with TEC, EC-T or DP regimens (data not shown).

At the conclusion of the study period, the overall DFS rate was 72.1% in the PXC group compared with 63.6% in the non-PXC group. However, Kaplan-Meier analysis showed no significant difference between the DFS rates of the PXC and non-PXC groups (P=0.098; Fig. 1A). In addition, OS was not significantly different between the PXC and non-PXC groups (P=0.557; Fig. 1B).

Subgroup analyses were performed to assess the effects of the administration duration of PXC, with subjects divided into groups who have received 1–3 months of PXC therapy and those with ≥3 months of PXC therapy, both of which were compared with the non-PXC group (Fig. 2). The results showed that the DFS rates were significantly higher in the PXC group with at least 3 months of treatment compared with those in the non-PXC group (76.1 vs. 63.6%, respectively; P=0.043 Fig. 2A). However, such a difference was not observed for OS (P=0.258 Fig. 2B). There was no significant difference in DFS (P=0.703) or OS (P=0.707) between those who used PXC for 1–3 months and those who did not use PXC (Fig. 2C and D).

Subgroup analysis was conducted between those who took PXC for ≥3 months and those who did not. In patients with hormone ER-negative tumors, the DFS rate in the PXC group (90.9%) was significantly higher than that in the non-PXC group (66.1%; P=0.011, Fig. 3B) regardless of node status; but there was no significant difference in DFS with PXC use in those with ER⁺ tumors (P=0.469, Fig. 3A). As per HER2 status, there was no statistically significant difference in both DFS and OS between the PXC group and the non-PXC group (Fig. 4). In summary, DFS was significantly improved in patients with hormone receptor-negative tumors and in those with ≥3 months of treatment with PXC.

Univariate and multivariate analyses for the risk of recurrence are summarized in Table II. Among the six variables in both univariate and multivariate analyses, three variables were identified to have prognostic significance: Tumor stage (P<0.001 and P=0.001, respectively), axillary metastasis (P<0.001 and P<0.001, respectively), and HER2 status (P=0.009 and P=0.018, respectively), all of which were associated with a higher risk of recurrence. By contrast, PXC use was associated with significantly lower risk of recurrence upon univariate analysis (P=0.046).

The results of univariate and multivariate analysis for DFS are summarized in Table III. Among the six variables in the univariate analysis, tumor stage (P=0.002) and axillary metastasis (P<0.001) were identified to have prognostic significance. Multivariate analysis using Cox's proportional hazard model identified axillary metastasis (P<0.001) to be an independent prognostic factor for DFS.

The following incidences of AEs (only those rated grade 2 or higher were included) were observed in the non-PXC group and the PXC group with ≥3 months of treatment: 6.0 and 3.3% for anemia (P=0.409); 41.5 and 23.9% for neutropenia (P=0.003), 4.1 and 0.0% for nausea (P=0.062); 6.0 and 8.7% for fatigue (P=0.388), 5.1 and 5.4% for oral mucositis (P>0.999); 3.7 and 1.1% for diarrhea (P=0.289); and 6.0 and 3.3% for anorexia (P=0.409), as detailed in Table IV. The incidences of neutropenia (P=0.003) of grade 2 or higher were significantly lower in the PXC group than those in the control group. Compared with that in the non-PXC group, the incidence of nausea (≥ grade 2) in the PXC group was markedly reduced (4.1 vs. 0%), although it did not reach statistical significance (P=0.062; Fisher's exact test). Notably, all patients had no obvious side effects caused by PXC. These results suggest that PXC treatment may play a role in preventing neutropenia and alleviating nausea in patients with EBC receiving chemotherapy.