NEFL mRNA could not be detected in any of the 11 normal breast tissues. Of the breast cancer samples, 91.7% (165/180) expressed NEFL mRNA as measured by real-time PCR analyses, and expression ranged from 5.54×10⁻⁸ to 2.79×10⁻⁴. NEFL mRNA was expressed in all of the 14 lymph node metastasis samples, and expression ranged from 5.52×10⁻⁸ to 9.46×10⁻⁶. The distribution of NEFL mRNA expression in breast tissues did not accord with a normal distribution. Based on the ROC analysis, the mRNA value (2.30×10⁻⁶) capable of distinguishing patients with relapse or distant metastasis from the patients with DFS in five years was used to group all of the samples into two groups: “NEFL-low” group (less than 2.30×10⁻⁶) and “NEFL-high” group (more than 2.30×10⁻⁶).

NEFL mRNA was not expressed in all of the normal breast tissues (11/11), and lower than it in their paired primary breast cancers (P<0.001). Moreover, NEFL mRNA was expressed in 97% primary cancer tissues and 100% lymph node metastasis samples, and the difference of NEFL mRNA levels between the malignant and normal breast tissues was statistically significant (P<0.001).

For 14 of the patients with primary breast cancers, the paired lymph node metastases were available. NEFL mRNA was down-regulated more than 1.5-fold (from 1.97 to 78.36) in 71.4% (10/14) of the lymph node samples than in their paired primary cancer tissues (P = 0.011). And the NEFL mRNA expression levels were lower in the primary cancer specimens with positive lymph nodes than in cancers with negative lymph nodes. NEFL mRNA was highly expressed in 56.8% (42/74) of the lymph node-negative patients, but was highly expressed only in 39.6% (42/106) of the node-positive cases. This difference in the level of NEFL mRNA of breast cancer specimens between lymph node-positive and lymph node-negative cases was statistically significant (P = 0.023, Table 1).

No significant differences in the NEFL mRNA level were found for any of the different clinicopathological factors, including menopausal status, tumor size, clinical stage, nuclear grade, ER status, PR status, and HER2 status (P>0.05, Table 1).

In the 174 cases with follow-up data for more than three years, the 3-year DFS rate was 77.2% (71/92) in patients with low-expressed NEFL and 87.8% (72/82) in the patients with high-expressed NEFL. The 5-year DSF rates were 54.3% (38/70) and 80.4% (45/56) in patients with low-expressed NEFL and patients with high-expressed NEFL, respectively. Kaplan and Meier survival analysis suggests that the DFS time of patients with low-expressed NEFL was shorter than the DFS of patients with high-expressed NEFL (P = 0.004, Figure 1A). The sensitivity and specificity of NEFL mRNA expression levle to predict the clinical outcome of breast cancer patients were 74.4% and 53.3%, respectively (Table 2). Next, tumor size, clinical stage, histological grade, lymph node status, ER status, PR status, HER2 status, and NEFL level were analyzed in a Cox's multivariate analysis. As a result, tumor size greater than 5 cm [OR = 2.26 (95% CI 0.92–4.99), P = 0.079], high nuclear grade [OR = 2.70 (95% CI 1.28–5.71), P = 0.009], negative PR [OR = 2.32 (95% CI 1.05–5.13), P = 0.038], and low-expressed NEFL [OR = 2.69 (95% CI 1.24–5.88), P = 0.013] were independent factors in predicting the relapse or distant metastasis of breast cancer patients (Table 3).

When the survival status of the patients with different NEFL expression levels and different stages of progression was analyzed, NEFL mRNA expression level was found to be a prognostic factor to predict DFS of early-stage breast cancer patients, including patients with clinical stage I/II disease (P = 0.0004, Figure 1B), patients with negative lymph nodes (P = 0.008, Figure 1C), and patients with histological grade I/II tumors (P = 0.006, Figure 1D). However, NEFL mRNA had a low predictive power to determine the DFS of late-stage breast cancer patients (P>0.05, Figure 1). Both the sensitivity and specificity to predict relapse or distant metastasis were higher in clinical stage I/II patients (85.7% and 54.1%, respectively), in node-negative patients (88.9% and 63.1%, respectively), or in histological grade I/II patients (76.9% and 57.0%, respectively) than them in predicting relapse or distant metastasis in overall breast cancer patients (74.4% and 53.3%, respectively; Table 2). In patients with clinical stage I/II, negative lymph nodes, and histological grade I/II, the hazard of relapse or distant metastasis of patients with low-expressed NEFL was 5.13-, 12.20-, and 2.78-fold higher, respectively, than in patients with high-expressed NEFL (Table 3).

All of the 180 breast cancer patients who were used in the present study underwent complete dissection of the breast and axillary lymph nodes without preoperative chemotherapy at Tianjin Medical University Cancer Institute and Hospital (TMUCIH), China, between January 2001 and November 2004. After surgery, 165 breast cancer cases were treated with chemotherapy; 102 cases with positive ER status were treated with tamoxifen as a hormone therapy; and 97 cases were treated by radiotherapy. All of the breast cancer patients were followed up until May of 2009. DFS was defined as the time interval from surgery to first local relapse/distant organ metastases (patients with relapse or distant metastasis) or to the last follow-up visit (patients with disease-free survival). Of the 180 breast cancer cases, 174 cases were followed for more than three years (31 cases with relapse or distant metastasis and 143 cases with DFS), and 126 cases were followed for more than five years (43 cases with relapse or distant metastasis and 83 cases with DFS). The median follow-up time was 65 months.

All the specimens used in the present study, 11 normal breast tissue samples, 180 primary tumors and 14 lymph node metastasis samples, were collected from the 180 breast cancer patients. Tissue samples were snap-frozen in liquid nitrogen and stored at −80°C. All samples were examined by hematoxylin-eosin (H&E) staining, and only the normal tissue samples with 50% or more epithelial cells and tumor samples that consisted of 75% or more cancer cells were selected for real-time RT-PCR. ER expression and PR expression were determined as positive when more than 1% of the nuclei were stained by immunohistochemical staining. HER2 was defined as positive when more than 10% of the membrane was stained by an immunohistochemical assay. The study protocol was approved by the Institutional Review Board and the Research Ethics Committee of TMUCIH and written consent was obtained from all participants.

RNA was extracted with TRIZOL reagent (Invitrogen, Gaithersburg, MD, USA) according to the manufacturer's instructions. Then, 5 µg of total RNA was used to perform reverse transcription (RT) for first-strand cDNA synthesis. RNA was denatured for 5 min at 65°C and snap cooled on ice in the presence of 0.5 µg Oligo(dT) and 10 mmol dNTP mix. The sample was then incubated at 4°C for 50 min with First-Strand Buffer, 0.2 µmol DTT, 40 units of RNaseOUT ribonuclease inhibitor and 200 units of SuperScript II in a total volume of 20 µL. The reactions were stopped by incubation at 70°C for 15 min. All of the reagents used for RT were from Invitrogen.

Real-time RT-PCR was performed using the Platinum Quantitative PCR SuperMix-UDG System (Invitrogen). We quantified the transcripts of the GAPDH housekeeping gene as a control as previously described [30]. Primers and TaqMan probes for NEFL were as follows: 5′-CCTGGAAATCGAAGCAT-3′, 5′-ATTTCACTCTTTGTGGTCCTC-3′, and 5′-(FAM) ATTTGTTGATCGTGTCCTGCATAGC (TAMRA)-3′. Assays were performed with the ABI 7500 TaqMan system (Applied Biosystems, Foster City, CA, USA). PCR was carried out after incubation at 50°C for 2 min and pre-denaturing at 95°C for 3 min, followed by 40 cycles at 95°C for 30 sec and 62°C for 1 min. The relative quantification was given by the C_T values, determined by triplicate reactions for all of the samples for both NEFL and GAPDH. The triplicate C_T values of NEFL were averaged, and the C_T value of GAPDH was subtracted to obtain ΔC_T. The relative expression level of NEFL mRNA was determined as 2^−ΔCT.

RNA was extracted from cancer tissues taken from 10 breast cancer patients and pooled equally as the quality control RNA. Quality control RNA and Diethylpyrocarbonate (DEPC)-treated water, served as the positive and negative control samples, respectively,were used to perform RT and real-time PCR with each of the different batches of assays. If the expression levels of NEFL or GAPDH in the negative control samples were detectable or the expression level in the positive control RNA was beyond the 95% confidence interval of the mean NEFL or GAPDH expression level of the quality control RNA, the expression levels in that batch of samples were assayed again.

The distribution of NEFL mRNA expression in breast tissues did not accord with normal distribution, therefore, the relationship between NEFL and various clinicopathological variables was analyzed by the chi-square test or the Fisher's exact test, as appropriate. The differences of NEFL mRNA levels between normal breast tissues and paired primary breast cancer samples and between primary cancer samples and paired lymph node metastases were calculated using Wilcoxon signed-rank test. The cut-off value for distinguishing patients with a poor prognosis from patients with a good prognosis was determined by calculating the receiver operating characteristic (ROC) curve and area under curve (AUC). Survival analysis was carried out according to the methods of Kaplan and Meier and log-rank test. Multivariate survival analysis was performed by a backward stepwise Cox proportional hazards regression model. All calculations were performed with the SPSS for Windows statistical software package (SPSS Inc, Chicago, IL, USA).