In total, 300 patients were included in the present study, which included 100 patients with primary osteosarcoma who were treated in Xuzhou Children's Hospital (Xuzhou, China) between March 2012 and May 2015. Osteosarcoma was diagnosed via clinical and microanatomical examination, and by histological assessment of specimens from osteosarcoma tissue. Participants did not have any history of other tumors and did not undergo any other pre-treatment when the initial diagnosis of osteosarcoma was confirmed. To select patients with benign bone tumors, 100 patients with benign bone tumors, including osteoenchondroma, ecchondrosis, bone cysts and fibrous osteoma, were included. These bone-related benign tumors were diagnosed via histological examination, similar to osteosarcoma. An additional 100 healthy age- and gender-matched subjects constituted the control group. These individuals were not diagnosed with any orthopedic issue or tumor. The experiments and all follow-up evaluations gained approval from the Ethics Committee of Xuzhou Children's Hospital (no. 20130116) and all participants provided written informed consent.

For each patient, his/her demographic information, tumor site, metastasis position, and pathological pattern were recorded upon diagnosis. All patients with osteosarcoma were treated with standard therapeutic pro Ethics Committee of cesses involving new accessorial chemotherapy and wide-margin surgical excision coupled with adjuvant chemotherapy or radiotherapy. Chemotherapy-induced reactions were categorized into two types on the basis of histological assessment of the osteosarcoma specimens: Poor, when tumor cell necrosis rate was <90%; good, when tumor cell necrosis rate was >90%. Furthermore, all patients with osteosarcoma were monitored postoperatively. The continuous disease-free survival period was recorded from the date of surgery until the time of tumor recurrence or the patient succumbed to mortality. The final follow-up record was on May 2017 and the median value of overall survival was 40.1 months in a total range of 10–75 months.

Venous blood samples from patients with osteosarcoma were obtained upon diagnoses. The blood samples were stored in serum separator tubes for coagulation, followed by centrifugation at a speed of 7500 × g for 20 min at 4°C to separate serum into other tubes at −80°C prior to ELISA for circulating MIC-1. The level of circulating MIC-1 was assessed via a sensitive in-house ELISA (RAB0204; Sigma-Aldrich; EMD Millipore, Billerica, MA, USA). Remixed MIC-1 (50 µl) in different concentrations and serum were added in a 96-well plate, with wells coated with 5 µg/ml anti-MIC-1 antibody. Thereafter, 50 µl of 0.2 µg/ml rabbit anti-MIC-1 biotinylated antibody was mixed in and incubated for 1 h at 37°C. Streptavidin-HRP was added following cleaning the plates with washing buffer (cat. no. RAB0204; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany), followed by incubation for 30 min at 37°C. Eventually, the optical density of every well was measured on a microplate reader at 450 nm. Based on the standard curve plotted from stepwise dilution of the recombinant MIC-1, circulating MIC-1 levels from each individual were calculated. All samples were assayed in duplicate.

Resected osteosarcoma tissue samples and non-cancer bone tissues were obtained and incubated at −80°C. RNA was isolated from the frozen tissues using TRIzol (Invitrogen; Thermo Fisher Scientific, USA), in accordance with the manufacturer's protocol. Total RNA (1 µg of every sample) was used for cDNA synthesis, using a reverse transcription kit (Takara Bio, Inc., Otsu, Japan). Subsequently, RT-qPCR analysis was performed for synthesized cDNA under the following cycling conditions: 2 min pre-treatment at 95°C, followed by 60°C for 15 sec, 72°C and 95°C each for 30 sec, in 35 cycles, and extension at 72°C for 5 min. The samples were then maintained at 4°C. RT-qPCR reactions were performed using an ABI Prism 7000 sequence detection system (Applied Biosystems; Thermo Fisher Scientific, Inc., Waltham, MA, USA) with the GoTaq qPCR Master Mix (Promega Corporation, Madison, WI, USA). The following primers were used: MIC-1, forward 5′-CGGAATTCATGGCGCGCAACG-3′ and reverse 5′-CCCTCGAGTATGCAGTGGCAGT-3′; β-actin, forward 5′-AGGCACCAGGGCGTGAT-3′ and reverse 5′-GCCCACATAGGAATCCTTCTGAC-3′. Relative mRNA levels of all genes were normalized against the levels of β-actin using the ΔΔCq method (14). The experiments were repeated three times.

The tumor tissues and their adjacent tissues were homogenized in ice and lysed in RIPA buffer with protease inhibitors (Sigma; EMD Millipore) and phenylmethylsulfonyl fluoride (Sigma; EMD Millipore). The dissolved solid was sonicated and centrifuged at a speed of 7,500 × g at 4°C for 5 min. Supernatant liquor was obtained following precipitation. Proteins were extracted their concentrations were determined using a BCA Protein kit (Thermo Fisher Scientific, Inc.). Protein (50 µg) was then separated via sodium dodecyl sulfate polyacrylamide gel electrophoresis (10% resolving gel) and subsequently electroblotted onto polyvinylidene fluoride membranes (EMD Millipore), which were washed with Tris-buffered saline and 0.05% Tween 20, followed by blocking with 5% skimmed milk. The membrane was then incubated with anti-MIC-1 (Thermo Fisher Scientific, Inc.; cat. no. PA5-71579, 1:1,000) and β-actin antibodies (Santa Cruz Biotechnology, Inc., Dallas, TX, USA, cat. no. sc-130656, 1:1,000) at 4°C overnight. The membrane was then incubated with HRP-conjugated secondary antibody (Santa Cruz Biotechnology, Inc., cat. no. sc-2004, 1:4,000) at room temperature for 1 h. Following washing, protein bands were detected using an Enhanced Chemiluminescence Western Blotting kit (Thermo Fisher Scientific, Inc.) and then quantified by Image J 1.43 software (National Institutes of Health, Bethesda, MD, USA).

One-way analysis of variance was performed to determine differences in circulating MIC-1 levels in the osteosarcoma patient group, benign tumor group, and control group. A t-test was performed to quantify the mRNA expression of MIC-1 in tumor tissues and associated peri-carcinomatous tissue. Diagnostic efficiency was evaluated via receiver operating characteristic (ROC) analysis. Survival rates were determined via Kaplan-Meier survival analysis. Statistically significant differences were measured via the log-rank test. Multivariable assessment of factors affecting prognosis was performed via Cox's regression analysis model. Statistical analysis was performed using Prism 5 software (GraphPad Software, Inc., La Jolla, CA, USA). P<0.05 was considered to indicate a statistically significant difference.

The present study investigated the association between circulating levels of MIC-1 and clinicopathological characteristics in patients with osteosarcoma. The clinical features of the samples are shown in Tables I and II. There were no considerable differences among patient features. As shown in Fig. 1, the circulating MIC-1 levels were elevated in patients with osteosarcoma compared with those in the benign bone tumor group and the healthy controls. In addition, the levels of MIC-1 in patients with benign tumors were marginally, higher than those of the healthy controls.

The present study investigated the correlation between the level of circulating MIC-1 and clinicopathological features in patients with osteosarcoma. The levels of circulating MIC-1 did not correlate with age, gender, or chemotherapeutic response. Patients with osteosarcoma with thigh bone tumors had a marginally, but not significantly, higher level of circulating MIC-1 than those with tumors at other locations (data not shown). The analysis of tumor size indicated that circulating MIC-1 levels were increased significantly in tumors >8 cm, compared with those that were <8 cm (P<0.05; Fig. 2A). Circulating MIC-1 levels differed among patients at various clinical stages; patients with stage III tumors had a significantly higher level of circulating MIC-1 than those with stage IIA and IIB tumors (P<0.05 for both). Similarly, as shown in Fig. 2B, patients with stage IIB tumors had significantly higher levels of circulating MIC-1 than those with stage IIA osteosarcoma (P<0.05). Circulating MIC-1 was higher in patients with osteosarcoma with distant metastases and in low tumor grades (P<0.05; Fig. 2C and D).

Assessment of the area under the ROC curves (ROC/AUC) determined the sensitivity and specificity of circulating MIC-1 at different concentrations. Circulating expression levels of MIC-1 were considered a biological indicator for distinguishing patients with osteosarcoma from healthy controls (Fig. 3A). This measurement may also assist in differentiating between patients with osteosarcoma and those with benign bone tumors (Fig. 3B). As shown in Fig. 3C, circulating MIC-1 may assist in differentiating patients with late-stage cancer, for example stage III, from those with early-stage cancer, for example stage IIA. The aforementioned findings indicated that circulating MIC-1 levels also assist in distinguish among different stages of osteosarcoma; therefore, MIC-1 may serve as a potential biomarker of disease progression.

All 100 patients with osteosarcoma included in the present study received a complete set of unified treatments and complete follow-up evaluation. On categorizing patients with osteosarcoma in accordance with average circulating levels of MIC-1, two types of patient group were obtained: High and low. Significant differences in overall survival rate were detected between the high circulating MIC-1 expression group and the low MIC-1 expression group (Fig. 4). Patients with high circulating MIC-1 levels tended to have a shorter survival rate, compared with those with low circulating MIC-1 levels. To determine whether MIC-1 is a potential prognostic indicator of osteosarcoma, Cox's univariate and multivariate regression analyses were performed, which suggested that a high level of MIC-1 served as a functional biomarker in the poor prognosis group (Table III).

The levels of MIC-1 in tumor tissue and adjacent tissue were detected via RT-qPCR analysis. As shown in Fig. 5A, the mRNA levels of MIC-1 were considerably higher in the osteosarcoma tumor tissue than in the tumor-adjacent tissue. Concurrently, western blot analysis revealed an enhancement in the protein levels of MIC-1 in osteosarcoma cancer tissues (Fig. 5B). These data suggested that the increased production of MIC-1 may be a potential source of circulating MIC-1.