Increased CXCL8 Expression Is Negatively Correlated with the Overall Survival of Patients with ER-Negative Breast Cancer

Materials and Methods

Patients and tissue samples. A total of 62 patients with primary breast cancer (all females aged between 30 and 84 years, median age of 50 years) who were hospitalized in the Third Affiliated Hospital of Soochow University from January 2003 to December 2003 were recruited in the present study. All the patients underwent modified radical operations and were followed-up to the end of September 2015. All the tumor samples were excised and quickly frozen in liquid nitrogen after resection and preserved at −80°C until use for nucleotide analyses. All the tissue samples were histopathologically examined. They were all defined as invasive ductal breast cancer. In the present study, no patient was given any treatments before surgery. After surgery, all the patients received common anthracycline-based chemotherapy, but no patient was treated with humanized antibody to HER2 (trastuzumab) although trastuzumab was approved in 2002 in China. All the ERα-positive patients had routinely received a standard endocrinotherapy. However, most patients stopped continuing endocrinotherapy by themselves after 2-3 years and only a few patients with ERα-positive breast cancer continued endocrinotherapy for about 5 years. As several patients were unable to continue or complete the standardized endocrinotherapy, they were excluded from the data analyses.

The median follow-up time was 149.63 months (range=16.83-161.87 months) and terminated at September 30. 2015. The patients lost to follow-up and suspended cases were censored. The characteristics of the patients are listed in the Table I. The protocols of the present study were approved by the Institutional Ethics Committee of the Third Affiliated Hospital of Soochow University [(2001)KENo.1(R01)] and all the patients gave their written informed consent to use of their specimens and data.

Extraction of total RNA, reverse transcription and polymerase chain reaction (RT-PCR). Before RNA extraction, frozen sections of the specimens were examined histologically to ensure that they were representative of the tumor tissue. Total RNA of breast cancer tissue was extracted using total RNA isolation classic kit according to the instructions provided in the kit (SNBC, Shanghai, China) and treated with RNase-free DNase I (TaKaRa, Kyoto, Japan) for 30 min at 37°C to remove residual DNA. RNA quality and its integrity were verified using a 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA).

A standard CXCL8-containing plasmid was diluted in a 10-fold series within the range of 10⁸-10¹ with Tris-EDTA buffer. Each concentration of standard plasmid DNA sample was amplified by conventional PCR. The PCR-amplified product was separated in 1% agarose electrophoresis and stained with ethidium bromide. The size of polymerase chain reaction-amplified CXCL8 fragment was 87 bp (Figure 1). This indicated that the maximal resolution of the conventional PCR assay was 10³ copies per reaction. When the diluted CXCL8 standard solutions were quantitatively detected by real-time quantitative (q)PCR, the maximal resolution of this method was 10 copies per reaction and there was a linear relationship between the copy number of CXCL8 standard and the Ct value.

Total RNA (2 μg) was used to synthesize cDNA by reverse transcription. The first-strand cDNAs were synthesized by using Revert Aid first-stand cDNA synthesis kit (MBI Ferments Inc, Ontario, Canada) according to the instructions provided in the kit. The sequences for the primers for CXCL8 and Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and Taqman probe were designed with primer design software as follows: CXCL8: forward primer: 5’-CTCTTGGCAGCCTTCCTGATT-3’, reverse primer: 5’-TATGCACTGA CATCTAAGTTCTTTAGCA-3’; CXCL8 probe: 5’-FAM-CTTGGCAAAACTGCACC TTCACACAGA-TAMRA-3’; GAPDH: forward primer: 5’-GGAAGGTGAAGGTCGGAGTC-3’, reverse primer: 5’-CGTTCTCAGCCTT GACGGT-3’; GAPDH probe: 5’-FAM-TTTGGTCGTATTGGGCGCCTG-TAMRA-3’). These primers and probes were synthesized by Shanghai Shenyou Biotechnology, Ltd. (Shanghai, China). The PCR reaction for each target fragment was performed in total volume of 25 μl containing 2 μl of cDNA template, 2.5 μl of 10×buffer, 1.5 μl of MgCl₂ (25 mM), 0.5 μl of 4×dNTPs (10 mmol/l), 0.1 μl of 100 μM primers and probes and 0.5 μl of DNA polymerase. The thermal cycling conditions included the following steps: 180 s at 50°C, 300 s at 95°C, followed by 40 cycles of 2-step PCR consisting of 20 s at 95°C, 60 s at 60°C (collecting fluorescence signal). All PCRs were performed on a LightCycler PCR system (Roche Company, Indianapolis, IN, USA). The mRNA levels of CXCR8 and GAPDH were quantitated based on their standard curves.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

The agarose electrophoresis image of conventional polymerase chain reaction-amplified C-X-C motif chemokine ligand 8 (CXCL8) fragment.

CXCL8 standard curve was established by plotting the known copy numbers of the positive CXCL8 template against the corresponding Ct values. For each sample, the PCR assay was repeated three times and the mean value was taken as the expression of CXCL8 in that sample. The RT-qPCR reaction system for GAPDH was the same as that for CXCL8. The relative expression levels of CXCL8 gene were calculated by normalizing CXCL8 levels with the corresponding levels of GAPDH (CXCL8/GAPDH) and calculated based on CXCL8 standard curve to obtain the relative expression of CXCL8 gene.

Immunohistochemistry. Formalin-fixed and paraffin-embedded tissues were cut into 3-μm-thick consecutive sections, dewaxed in xylene and rehydrated in graded ethanol solutions. Polyclonal rabbit antibody against human HER2 and P53 (DAKO Company, Fuzhou, Fujian, China) and corresponding polyclonal mouse antibodies against human ERα and PR (MXB Biotechnologies, Fuzhou, PR China) were used. There were two steps in EnVision IHC staining and color development, 3,3’-diaminobenzidine (DAB) was used as the color reagent, and phosphate-buffered saline was used as substitute for the primary antibodies and taken as the negative control. Tumors were considered positive when there were at least 1% positively stained tumor nuclei in the sample on testing in the presence of expected reactivity of internal (normal epithelial elements) and external controls (16, 17).

Immunostaining was scored by two pathologists. The intensity (I) of staining was graded on a scale of 0-3+ with 0 representing no detectable staining and 3+ representing the strongest staining. The four strongest staining regions were randomly selected under a 400× field. In each of the four regions, the rate of positive cell staining (R) under a 400× field was calculated and defined as follows: 0. no staining; 1, ≤10% tumor cells with staining; 2, 11-50% tumor cells with staining; 3, 51-75% tumor cells with staining; and 4, >75% tumor cells with staining. Samples with scores <3 were considered negative, while those with scores ≥3 were considered positive. Histochemistry score=I × R (18).

Tumors were considered as HER2-positive when cells exhibited strong membranous staining (3+). Tumors exhibiting 0 or 1+ staining for HER2 protein expression were considered to be negative. In cases of equivocal membrane staining (score 2+) for HER2, fluorescence in situ hybridization was used to evaluate HER2/neu gene amplification (16, 17).

Statistical analyses. Statistical analyses were performed with the R software, version 3.0.3 (http://journal.r-project.org/). Breast carcinoma-specific OS was defined as the time from the date of diagnosis to death if the patient died from breast carcinoma. Wilcoxon rank-sum test was used to examine the distribution of quantitative CXCL8 levels according to clinicopathological parameters. Survival curves were estimated by the Kaplan–Meier method, and the curves were compared with the log-rank test in univariate analyses. To examine the association of CXCL8 and clinicopathological parameters with OS, the Cox proportional hazards regression model was applied for calculating the hazard ratios (HR) and for corresponding 95% confidence interval (95% CI). Forward stepwise regression was used to calculate the multivariate HRs. A p-value of less than 0.05 was considered as statistically significant. The receiver operating characteristic (ROC) curve was defined as a plot of test sensitivity as the Y-axis versus its 1-specificity as the X-axis, which was conducted to determine the cut-off value of CXCL8 expression level for predicting 10-year OS of breast cancer patients. Then, the area under the curve (AUC) of the ROC was used to assess the predictive validity.