Through search of the Breast Cancer Information Management System of West China Hospital, Sichuan University, a total of 65 breast cancer patients who had undergone radical mastectomy during the period from 2007 to 2010 were included in the present study. The inclusion criteria included: i) pathological diagnosis of breast cancer; ii) HER2-positive breast cancer [3+ by immunohistochemical analysis (IHC) or with gene amplification by fluorescence in situ hybridization (FISH)]; iii) availability of standard treatment containing trastuzumab at least for 1 year; and iv) >3 years of complete follow-up record upon radical mastectomy. Exclusion criteria were: i) diagnosis of metastatic breast cancer before treatment; and ii) administration of any preoperative treatment. We collected the clinical information of patients and performed telephone follow-up periodically through the Breast Cancer Information Management System (once every 6 months during the initial 1–5 years and once yearly after 5 years). The study met the recommendations of the ‘Declaration of Helsinki’ and was approved by the Ethics Committee of West China Hospital.

The 65 FFPE breast tumor tissues and 14 FFPE normal breast tissues from 65 breast cancer patients were retrospectively collected from the Pathology Department of West China Hospital. Disease-free survival (DFS) was defined as the time from radical mastectomy to tumor recurrence (28). Relapse rate was defined as the proportion of the patients with recurrent disease during the follow-up period.

The streptavidin-biotin complex (SABC) method was used to detect the expression levels of CD55, CD59 and CD46 in the FFPE specimens. The specimens were heated using a microwave at low power for 5 min for antigen retrieval. Three percent H₂O₂ deionized water solution was used to block endogenous peroxidase. Anti-CD55 (clone SM3030P; Acris Co., Gemany; 1:200), anti-CD59 (clone p282; Abcam Co., USA; 1:100) or anti-CD46 (clone H-294; Santa Cruz Biotechnology Co., USA; 1:100) antibodies were added to the samples and incubated at 4°C overnight. After PBS washes, the biotinylated secondary antibody (ZB-2010 or ZB-2020; Zhongshan Golden Bridge Co., China) was added. The samples were incubated for 30 min, and subjected to PBS washes. Afterwards, the samples were incubated with streptavidin-biotin-peroxidase labeled tertiary antibody (ZB-2404; Zhongshan) for 30 min. Color was developed using diaminobenzidine (DAB) chromogenic color development reagent. PBS instead of the primary antibody was used in the negative control group. Staining intensity of the tumor cells was classified as: −, +, + + and + + + (Fig. 1). At least 1,000 tumor cells were assessed and high CD55, CD59 or CD46 expression was defined as +++ expression of CD55, CD59 or CD46 in more than 1% of breast cancer cells. Otherwise, the breast cancer specimens were identified as having low CD55, CD59 or CD46 expression (29)

The ER⁻/HER2⁺ breast cancer cell line SK-Br-3 and the ER⁺/HER2⁺ BT-474 cell line (The Cell Bank, Chinese Academy of Sciences) (30,31), were cultured in RPMI-1640 containing 10% fetal bovine serum (FBS; Gibco, USA) and MEM (Gibco) containing 10% FBS, respectively.

SK-Br-3 or BT-474 cells (1×10⁴ cells/well) were seeded in 96-well plates and incubated overnight to allow cells to adhere. The cells were pretreated with the following drugs: Indo (20, 40, 80, 160 and 320 μM), Tam (1, 2, 4, 8 and 16 μM), LMS (0.16, 0.32, 0.64, 1.28 and 2.56 mM), sodium butyrate (0.3, 0.6, 1.2, 2.4 and 4.8 mM) (all from Sigma USA) or DDP (0.25, 0.5, 1, 2 and 4 μM; Shandong Qilu Pharmaceutical Factory, China) for 72 h. 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma) was then added and incubated for 4 h. Absorbance at 490 nm was measured using a Multiskan Spectrum (Thermo Scientific Co., USA). The maximum non-toxic doses of the drugs in the SK-Br-3 or BT-474 cells were determined.

The SK-Br-3 and BT-474 cell lines were used to study trastu-zumab-induced CDC. Normal human serum (NHS) obtained from healthy volunteers (Blood Bank of West China Hospital, Sichuan University, China) was provided voluntarily for this study. Cells were incubated with heat-inactivated NHS (56°C, 60 min) as the negative control. Phosphatidylinositol-specific phospholipase C (PI-PLC; Invitrogen Co., USA) specifically hydrolyzes and inactivates anchor proteins CD55 and CD59 (32). Cells were incubated with PI-PLC as the positive control. Breast cancer cells were pretreated with maximum non-toxic doses of Indo (40 μM), Tam (2 μM), LMS (0.16 mM), sodium butyrate (0.5 μM) and DDP (0.32 mM) for 72 h, and subsequently washed with PBS. Cells were detached by 0.125% trypsin and washed with PBS. A total of 1×10⁵ cells were then cultured with FBS-free medium containing 50% human serum (v/v) and 50 μg/ml trastuzumab for 1 h (32). Afterwards, trypan blue staining was performed to count the dead and viable cell number in a hemocytometer under a microscope. Experiments were repeated at least three times. The cell death rate was expressed as the percentage of dead cells relative to the total number of cells (viable cells and dead cells included).

Cells were pretreated with 2 μM Tam for 6, 12 and 24 h. Total cellular RNA was extracted using the BIOzol total RNA extraction kit (BioFlux Co., Japan) according to the manufacturer’s instructions. First-strand cDNAs were synthesized from total RNA using M-MLV transcriptase and oligo-dT primer. Quantitative RT-PCR (qRT-PCR) was performed using duplicate SsoFast™ Eva Green^® Supermix kit (Bio-Rad, USA) in Chromo4 real-time PCR detector (Bio-Rad) following the manufacturer’s instructions. The primer sequences used are presented in Table I (33). The relative quantitation of gene expression was performed using the comparative Ct (ΔΔCt) method with GAPDH as a housekeeping gene. Experiments were repeated at least twice.

After pretreatment with 2 μM (maximum non-toxic doses) Tam for 72 h, SK-Br-3 and BT-474 cells were lysed at 4°C in RIPA lysis buffer (Biyuntian Co., China). Total protein concentration was determined using the Bradford protein assay (Bio-Rad). Equal amounts of cell lysates were separated on 12% SDS-PAGE and transferred to PVDF membranes. PVDF membranes were blocked with 5% nonfat milk in PBS-T (0.5% of Tween-20 in PBS). The membranes were incubated at 4°C overnight with the primary antibodies against CD55 (BRIC216), CD59 (YTH53.1) or β-actin (all from Santa Cruz Biotechnology, USA). After PBS-T washing, horseradish peroxidase-conjugated secondary antibodies (Santa Cruz Biotechnogy) were added. The membranes were then incubated at room temperature for 1 h. Individual protein bands were visualized using chemiluminescence reagent (Millipore, USA). The film was exposed in a dark room and scanned with HP LaserJet 3055 scanner. Volume densitometry of the immunoreactive bands was quantitated with One Quantity software (Bio-Rad, USA).

SK-Br-3 and BT-474 cells were pretreated with the maximum non-toxic doses of Indo, Tam, LMS, DDP or sodium butyrate for 72 h and washed with PBS. Cells were incubated on ice for 30 min with the FITC-conjugated anti-CD55 antibody, FITC-conjugated anti-CD59 antibody (Beijing Jingmei Gene Valley Technology Co., China) or FITC-conjugated anti-CD46 antibody (Biolegend Co., USA). IgG1 isotype control/FITC (Beijing Jingmei Gene Valley Technology Co., China) was used as the negative control antibody. Then, the cells were washed, and resuspended in formaldehyde fixing solution (1% BSA-PBS + 1% polyphosphate formaldehyde), and subjected to flow cytometry (FCM).

Statistical significance of the correlation between the clinicopathologic characteristics or prognosis and the patients with mCRP high expression and low expression was determined with a χ² test comparison. Kaplan-Meier survival analysis was used for estimation of DFS, and log-rank test was used for inter-group comparison. Cox proportional hazards model analysis was used for estimation of univariate hazard ratios (HRs) for DFS and multivariate analysis. Menopausal status, tumor size (cm), tumor grade, stage, lymph node metastasis status, hormone receptor status and mCRP expression were included in the model. The CDC of trastuzumab was expressed as mean ± SEM. Independent samples t-test was used for comparison between two groups. P<0.05 was considered to indicate a statistically significant difference. SPSS software version 16.0 (SPSS Inc., USA) was used for the statistical analysis.

The clinicopathological characteristics of the 65 breast cancer patients who underwent postoperative adjuvant treatment containing trastuzumab are summarized in Table II. Patient age ranged from 21 to 67 years (with a median age of 49 years). Adjuvant chemotherapy plus trastuzumab was administered to 43 patients and adjuvant endocrine therapy plus adjuvant chemotherapy and trastuzumab to 22 patients. Trastuzumab was given once every three weeks with a starting dose of 8 mg/kg and a maintenance dose of 6 mg/kg, following chemotherapy. Trastuzumab therapy was maintained for at least 1 year. The median follow-up duration was 47 months (range from 24 to 75 months). Relapse occurred in 13 patients during the follow-up period. Two patients relapsed within 1 year, 10 patients within 1–3 years, and 1 patient within 3–5 years. Among them, 5 patients had lymph node metastasis, 4 patients had lung metastasis, 2 patients had bone metastasis, and 2 patients had gastrointestinal metastasis.

Levels of CD55, CD59 and CD46 expression in the FFPE breast tumor tissues from 65 patients who underwent postoperative adjuvant therapy containing trastuzumab were detected by immunohistochemistry. CD55, CD59 and CD46 were expressed on the membrane of the breast cancer cells (Fig. 1). Among the patients, 30/65 (46.2%) had high expression levels of CD55, 29 (44.6%) had a high expression level of CD59, and 29 (44.6%) had a high expression level of CD46. In comparison, only 2/14 (14.3%) normal breast tissue specimens had high expression levels of CD55, CD59 or CD46; the remaining 12 (85.7%) specimens had low expression levels.

As shown in Table III, patients with CD55 or CD59 overexpression had significantly higher relapse rates than the patients with low expression of CD55 (33.3 vs. 8.6%; P=0.013) or CD59 (31.0 vs. 11.1%; P=0.046). The relapse rate in the CD46 overexpression group was slightly higher than that in the CD46 low expression group (24.1 vs. 16.7%). However, the difference was not statistically significant (P=0.454). Other clinicopathological characteristics, including menopausal status, histological classification, tumor stage, lymph node metastasis, and hormone receptor status did not differ significantly between patients with high and low expression of mCRPs (P>0.05).

Mean DFS of 65 patients included in our study was 64 months (95% CI, 59–69 months). We found that the mean DFS of patients with CD55 or CD59 high expression was significantly shorter when compared to the patients with low expression of CD55 (56 vs. 70 months; log-rank test, P=0.008) or CD59 (56 vs. 69 months; log-rank test, P=0.033). In contrast, the mean DFS was not significantly different between patients with high and low expression of CD46 (66 vs. 60 months; log-rank test, P=0.516; Fig. 2). Risk factors of DFS including menopausal status, tumor size (cm), tumor grade and stage, lymph node metastasis, hormone receptor status and mCRP expression were analyzed using Cox proportional hazards model. Univariate analysis revealed that overexpression of CD55 (HR=4.822; 95% CI, 1.323–17.57; P=0.02) or CD59 (HR=3.343; 95% CI, 1.027–10.878; P=0.05), but not CD46 (HR=1.43; 95% CI, 0.48–4.257; P=0.52), was associated with a high risk of relapse. Multivariate analysis showed that CD55 overexpression was an independent risk factor of relapse (HR=4.757; 95% CI, 0.985–22.974; P=0.05). However, CD59 overexpression was not significantly associated with risk of relapse (HR=2.378; 95% CI, 0.543–10.414; P=0.25) (Table IV).

Our retrospective analysis of clinical samples revealed that CD55 overexpression is an independent risk factor of relapse in breast cancer patients treated with trastuzumab. It is possible that a high CD55 expression level suppresses trastuzumab-mediated CDC, and thus leads to poor prognosis. Therefore, we aimed to ascertain whether downregulation of mCRP expression could sensitize breast cancer cells to trastuzumab-mediated CDC. We confirmed the maximum non-toxic doses of Indo, Tam, LMS, DDP and sodium butyrate (40, 2, 0.16, 0.5 and 0.32 mM, respectively) in SK-Br3 and BT-474 cells by MTT. Breast cancer cells were pretreated with these drugs at their maximum non-toxic doses. Flow cytometric results showed that only Tam inhibited expression levels of CD55 in the SK-BR-3 and BT-474 cells (Fig. 3A). Real-time PCR and western blotting results confirmed the inhibitory effect of Tam on both the mRNA and protein expression levels of CD55 (Fig. 4). In contrast, CD59 expression levels in both breast cancer cell lines did not change significantly after Tam pretreatment.

To investigate the effect of mCRP inhibition of trastuzumab-mediated CDC, we pretreated breast cancer cells with Tam, DDP, Indo, LMS or sodium butyrate for 72 h, and then incubated the cells with 50% normal human serum (NHS) and trastuzumab for 1 h. Trastuzumab with NHS induced an insignificant cytotoxic response (4.8±0.5 vs. 4.6±0.3%, respectively in the SK-Br3 and BT-474 cells), compared with the negative control group (4.5±0.01 vs. 4.1±0.04%, respectively), while Tam pretreatment significantly enhanced the cell death rate induced by trastuzumab and NHS in the SK-Br3 (14.33±1.7%, P<0.01) and BT-474 cells (9.0±0.6%, P<0.01) (Fig. 3B and C). In contrast, DDP, Indo, LMS, or sodium butyrate pretreatment did not significantly affect trastuzumab-mediated CDC (P>0.05; Fig. 3B and C).