Site-specific relapse pattern of the triple negative tumors in Chinese breast cancer patients

A total of 1662 patients were selected retrospectively from a large database of patients who underwent surgery between January 1, 2000 and March 31, 2004 at the Cancer Hospital, Fudan University, Shanghai, China.

Before surgery, the following evaluations were mandatory for each patient as complete physical examination, chest radioscopy, bilateral mammography, ECG, ultrasonography of breasts, axillary fossa, cervical parts, abdomen, and pelvis, complete blood count, and routine biochemical tests to make an exact staging. Thereafter, each patient received lumpectomy or mastectomy followed by adjuvant therapy according to the guidelines or recommendations used at the time of surgery.

Eligibility criteria for this analysis, similar to other relevant reports [17, 18], included female gender, an initial diagnosis of unilateral primary breast cancer without distant metastases, at least 2 months of follow-up information for disease recurrence, and complete data on the following: age, tumor size, number of involved axillary lymph nodes, status of estrogen receptor (ER), progesterone receptor (PR) as well as ERBB2. All eligible patients in the database were included. All data were entered into a computerized database and verified to minimize errors in data entry.

Follow-up information regarding tumor recurrence and survival status was accomplished through patients' clinic visits with records kept in the computerized database of the outpatient department, personal contact with the patient as well as the assistance of Shanghai Center for Disease Control and Prevention (CDC). Thereinto, personal contact with the patient referred to routine correspondence or telephone visits, which were carried out at the Cancer Hospital, Fudan University every 3 months during the first two years, every 6 months during the next two years and once a year thereafter. As this was a retrospective study without any medical intervention, the ethical approval was not required.

Immunohistochemical staining was carried out as a standard operating procedure in the pathology department of Cancer Hospital, Fudan University, Shanghai, China. All primary monoclonal antibodies for estrogen receptor (ER), progesterone receptor (PR) and ERBB2 were from Dako. The staining results were assessed by at least two pathologists, using a semiquantitative scoring system, where integrated the proportion score and the intensity score. The proportion score, indicating the percentage of tumor cells stained, was interpreted such that a score of 0 required no staining seen, 1 required ≤25% of cells positive, 2 required 25-50% of cells stained, 3 required 50-75% of positive cells and 4 required >75% of staining cells. As to the intensity score, a negative result was defined as a score of 0, weakly positive as 1, moderately positive as 2, and strongly positive as 3. The final score was calculated as the product of the proportion score and the intensity score. Thereby, staining results ranged from score 0 to 12. The scoring system for ER and PR was defined as negative for score 0 and positive for scores of 1~12 with the nucleic staining of carcinoma cells, whereas ERBB2 was defined as negative for scores of 0~8 (namely, 0, 1+ and 2+ in the DAKO scoring system) and positive for strong membranous staining with scores of 9~12 (namely DAKO score 3+).

Clinicopathological parameters were compared between different subgroups using one-way analysis of variance (ANOVA) test for continuous variables, chi-square test for unordered categorical variables and nonparametric Kruskal-Wallis rank test for ordinal categorical variables.

Site-specific relapse-free survival (RFS) was defined as the time from surgery to the earliest occurrence of an event, including locoregional, visceral and bone relapse. Locoregional failure was defined as a first relapse on the chest wall or in the ipsilateral breast, the ipsilateral axilla, the ipsilateral supraclavicular or infraclavicular fossa, or the ipsilateral internal mammary region [19], which was required to be identified by biopsy or fine needle aspiration. Visceral metastasis was established if there was any radiological evidence of metastases to viscera (including lung, liver and brain). Bone relapse was defined as metastases to bone without visceral metastasis. Any suspicious lesions on bone scan should be confirmed by further X-ray or CT/MRI examination. Those without any evidence of event were censored at the last date they were known to be alive.

Survival analyses were estimated by the Kaplan-Meier method and were compared using the log-rank test. Multivariate Cox proportional hazards regression analyses were applied to modeling the relationship between subgroup and RFS, adjusted for age (≤50 vs. >50), tumor size (≤2 cm vs. >2 cm), number of axillary lymph nodes (ALNs) involved (0, 1-3, ≥4), histological grade (I/II vs. III) and systemic treatment (yes vs. no). Hazard ratios (HRs) were presented with their 95% confidence intervals (CIs). For graphical display of RFS, annual hazard rates were estimated using a Kernel method of smoothing. All statistical tests were two sided and P < 0.05 was considered to be statistically significant. All statistical analyses were performed with Stata statistical software package (release 10.0; Stata Corporation, College Station, Texas, USA).

The mean age at diagnosis was 52.7 years old (range 24-90). Median follow-up was 3.15 years, ranging from 2 months to 7.8 years. According to different combinations of HR and ERBB2 status, 1662 patients were classified into the three subgroups as follows: triple negative (19.31%), ERBB2+ (26.72%) and HR+/ERBB2- (53.97%, HR+ referred to ER+ or PR+). When compared with triple negative and ERBB2+ tumors, less relapse events were observed in HR+/ERBB2- patients (15.89% and 14.64% vs. 8.81%, P < 0.001; Table 1).

We found a statistically significant difference in locoregional RFS between the three subgroups (log-rank P = 0.007, Fig. 1; adjusted P < 0.05, data not shown). Compared with the triple negative patients, locoregional RFS was significantly better for HR+/ERBB2- counterparts (P = 0.027; HR = 0.521, 95% CI 0.293-0.927; Table 2), but nearly comparable to ERBB2+ tumors (P = 0.221; HR = 0.686, 95% CI 0.375-1.254; Table 2). Besides, the survival curves of locoregional failure for triple negative and ERBB2+ patients were virtually superimposable. Somewhat differently in the hazard curves, the former showed an early major surge reaching the maximum at approximately 2 years after surgery, while the corresponding peak for the latter was at 1.5 years (Fig. 2).

A statistically significant difference was yielded in visceral RFS among the three subgroups (log-rank P = 0.025, Fig. 3; adjusted P < 0.05, data not shown). The multivariate Cox regression analysis demonstrated that triple negative patients, similar to ERBB2+ counterparts (P = 0.889; HR = 0.952, 95% CI 0.479-1.893; Table 2), tended to have more visceral relapses than HR+/ERBB2- women (P = 0.005; HR = 0.339, 95% CI 0.158-0.724; Table 2) across all time periods. Furthermore, the early relapse peak appeared to occur later in triple negative phenotype than that in ERBB2+ subgroup, as opposed to a relatively flat hazard rate until 4 years for HR+/ERBB2- category (Fig. 4).

There was no difference in RFS among the three groups in the univariate survival analysis (log-rank P = 0.180; Fig 5). Nevertheless, when compared to triple negative patients, a significantly lower risk of developing bone relapse was discerned in the multivariate Cox regression analysis for ERBB2+ women (P = 0.048; HR = 0.384, 95% CI 0.148-0.991), with the borderline significance for HR+/ERBB2- breast cancers (P = 0.058; HR = 0.479, 95% CI 0.224-1.025; Table 2). This finding was confirmed by the further analysis of hazard function (Fig. 6).