In a retrospective study design, 150 primary tumors were collected consecutively from patients diagnosed with RCC who underwent surgery between 2001 and 2008 at the Hiroshima University Hospital (Hiroshima, Japan). Only patients without preoperative radiotherapy or chemotherapy were enrolled in the study. Of 150 patients, 129 were at stages I, II and III, while 21 were at stage IV. The 129 patients with stage I, II and III RCC underwent curative resection. Of the 129 patients with stage I, II and III RCC, 20 received adjuvant therapy (interferon-α) (15). Of the 21 patients with stage IV RCC, 17 received postoperative therapy using interferon-α (15). Postoperative follow-up was scheduled every 1, 2 or 3 months during the first 2 years after surgery, and every 6 months thereafter unless more frequent follow-up was deemed necessary. Chest-abdominal computed tomography scan and serum chemistries were performed during each 6 month visit at least. Patients were monitored by their physician until they succumbed to the disease or the date of the last documented contact. The median follow-up period was 64 months (range, 2–120). Operative mortality was defined as death within 30 days of the patient leaving the hospital, and these patients were omitted from the analysis.

For qRT-PCR analysis, archival FFPE tissues were used. Histological classification was based on the World Health Organization system. RCC cases were classified into clear cell RCC (ccRCC) and non-ccRCC. Tumor staging was performed according to the TNM grouping system. Since written informed consent was not obtained, for reasons of strict privacy protection, identifying information for the samples was removed prior to analysis. This procedure was in accordance with the Ethical Guidelines for Human Genome/Gene Research of the Japanese Government.

FFPE samples were sectioned (10 μm), deparaffinized and stained with hematoxylin and eosin (H&E) to ensure that the sectioned block contained tumor cells. The tumor areas in the adjacent sections were marked under a light microscope without hematoxylin staining. Tumor areas were macrodissected with sterile disposable scalpels and subjected to RNA isolation using the Recover All™ Total Nucleic Acid Isolation kit (Ambion, Austin, TX, USA), according to the manufacturer’s instructions. Expression levels of miR-486 and RNU6B were measured using TaqMan® assays for miRNA (Applied Biosystems, Austin, TX, USA) in such a manner that the identity and clinical outcomes of samples were blinded. Complementary DNA (cDNA) was synthesized using miRNA-specific primers and the TaqMan^® MicroRNA Reverse Transcription kit (Applied Biosystems) according to the manufacturer’s instructions. Briefly, 40 ng of RNA was reverse transcribed in a 20 μl reaction with gene-specific RT probes. qRT-PCR was performed using the ABI 7900 Version 2.3 Sequence Detection System (Applied Biosystems). RNU6B was used as an endogenous normalization control for miR-486. The assays were performed in triplicate. Quantification of miR-486 relative expression was calculated using the RQ manager 1.2 (Applied Biosystems).

FFPE samples were sectioned, deparaffinized and stained with H&E to ensure that the sectioned block contained tumor cells. Adjacent sections were then stained immunohistochemically with the Dako EnVision+ Mouse Peroxidase Detection System (DakoCytomation, Carpinteria, CA, USA). Antigen retrieval was performed by microwave heating in citrate buffer (pH 6.0) for 30 min. After peroxidase activity was blocked with 3% H₂O₂ methanol for 10 min, sections were incubated with normal goat serum (DakoCytomation) for 20 min to block non-specific antibody binding sites. Sections were incubated with primary antibodies against olfactomedin 4 (1:50 dilution, using an anti-olfactomedin 4 antibody raised in our laboratory) (16) for 1 h at room temperature, followed by incubations with EnVision+ anti-mouse peroxidase for 1 h. Staining was completed with 10-min incubation with the substrate-chromogen solution. Sections were counterstained using 0.1% hematoxylin.

Associations between clinicopathological characteristics and miR-486 expression were analyzed using the Fisher’s exact test. To evaluate the associations between clinical covariates and cancer-specific mortality univariate and multivariate Cox regression analysis was used and conducted using SPSS software (SPSS Inc., Chicago, IL, USA). Hazard ratio (HR) and 95% confidence interval (CI) were estimated from the Cox proportional hazard models. For the analyses, age was treated as a categorical variable (65 plus >65 vs. <65 years old). For the final multivariate Cox regression models, the variables that were moderately associated (P<0.10) with cancer-specific mortality were included. Differences in miR-486 expression levels between the two groups were determined by the Mann-Whitney U test using Graphpad Prism 5.0 (Graphpad Software, Inc., San Diego, CA, USA). Kaplan-Meier survival curves were constructed for high- and low-miR-486 patients using the Graphpad Prism 5.0 software. Differences between survival curves were tested for statistical significance using the log-rank test. P<0.05 was considered to indicate a statistically significant difference.

The expression of miR-486 and the corresponding non-neoplastic kidney tissue samples were assessed in 86 RCC cases using qRT-PCR. The expression of miR-486 in RCC samples was ∼2.7-fold higher compared to that in corresponding non-neoplastic kidney samples (P<0.0001, Mann-Whitney U-test).

Expression of miR-486 was examined in 64 additional RCC tissue samples using qRT-PCR. We investigated the association between the clinicopathological characteristics and miR-486 expression levels in 150 RCC cases (Table I). These 150 RCC cases were divided into high- and low-miR-486 cases. When low-miR-486 expression was classified according to the lowest quartile, the number of high- and low-miR-486 cases was 112 and 38, respectively. Expression of miR-486 was not associated with clinicopathological characteristics such as gender, T, N and M classifications, tumor stage or venous invasion.

The association between miR-486 expression levels and cancer-specific mortality was evaluated. The Kaplan-Meier analysis for the 150 RCC cases (stages I–IV) was performed. As shown in Fig. 1A, cancer-specific mortality did not statistically vary in high- and low-miR-486 cases.

Patients with RCC at stage I and II have a good rate of survival, whereas it is difficult to predict patient survival with stage III and IV RCC. These patients would benefit greatly from prognostic biomarkers. Therefore, we analyzed the prognostic value of miR-486 expression in patients with stage III and IV RCC (n=46). Using the Kaplan-Meier analysis, we found that RCC cases with high-miR-486 expression had significantly worse cancer-specific mortality compared to those with low-miR-486 expression (P=0.0331, log-rank test, Fig. 1B). Since treatment with interferon-α for patients with stage III and IV RCC may affect patient survival, individuals who did not receive interferon-α treatment (n=14) were assessed. Although cancer-specific mortality among patients with high- and low-miR-486 expression exhibited no statistically significant difference (P=0.0574, log-Rank test, Fig. 1C), cancer-specific mortality of patients with high-miR-486 expression tended to be worse compared to that of patients with low-miR-486 expression.

To evaluate the potential for miR-486 expression as a prognostic predictor in patients with stage III and IV RCC, univariate and multivariate Cox proportional hazards analyses were used to evaluate the association of miR-486 expression with cancer-specific mortality (Table II). Findings of the univariate analysis showed that N (HR, 2.73; 95% CI, 1.10–6.24; P=0.0315) and M classifications (HR, 3.16; 95% CI, 1.46–7.08; P=0.0035), tumor stage (HR, 4.03; 95% CI, 1.82–9.54; P=0.0005) and miR-486 expression (HR, 3.38; 95% CI, 1.18–14.26; P=0.0202) were significantly associated with cancer-specific mortality. According to the multivariate model, which included N and M classifications, tumor stage and miR-486 expression, expression of miR-486 was an independent prognostic classifier of cancer-specific mortality (HR, 4.33; 95% CI, 1.45–18.71; P=0.0064).

Survival analysis was performed using an alternatively defined cut-off point. When low-miR-486 expression was defined on the basis of lower than median levels of expression of miR-486, the number of high- as well as low-miR-486 miR-486 cases was 75. Kaplan-Meier analysis revealed that cancer-specific mortality did not exhibit statistically significant differences in high- and low-miR-486 cases (data not shown). When low-miR-486 expression was classified according to the lowest tertile, the number of high- and low-miR-486 cases was 100 and 50, respectively. Kaplan-Meier analysis revealed that cancer-specific mortality did not exhibit statistically significant differences in high- and low-miR-486 cases (data not shown).

Biomarkers that predict therapeutic outcomes may provide tools to allow physicians to better stratify patients to more effective treatments. Therefore, we analyzed the association between miR-486 expression and therapeutic outcomes in stage III and IV RCC patients treated with interferon-α (n=32). However, expression of miR-486 was not significantly associated with therapeutic outcome (P=0.2931, log-rank test, Fig. 1D).

OLFM4, which encodes the OLFM4 protein, has been reported as one of the direct targets of miR-486 (14). Although alteration of OLFM4 expression has been reported in several types of human cancer (16–18), expression of OLFM4 in RCC has not been investigated. Therefore, the expression of OLFM4 was analyzed using immunohistochemistry in 86 RCC cases. In non-neoplastic kidney, OLFM4 staining was observed in uriniferous tubules but not in glomeruli (Fig. 2A). By contrast, OLFM4 staining was not expressed in RCC cells (Fig. 2B). No OLFM4 staining was detectable in the 86 RCC cases.