This study included 119 patients diagnosed with GC through histopathologic evaluation on gastroscopic biopsy or surgical tissue specimens; 101 cases were male and 18 cases were female, ranging in age from 29 to 81 years, with a median age of 59.03 years. All patients underwent gastrectomy between 2014 and 2018 at the 940th Hospital of the Joint Security Force, Gansu, China. Patients who had preoperative treatment, such as radiotherapy, chemotherapy, or other medical interventions and those diagnosed with autoimmune diseases were excluded from the study. Formalin-fixed, paraffin-embedded (FFPE) GC tissue samples were obtained from these patients after surgery.

Demographic and clinicopathologic characteristics, including age, sex, TNM stage, histologic grade, and lymph node metastases were assessed. The TNM classification was based on the 8th edition of AJCC GC staging. The patients were followed up for 5 years by inpatient, outpatient and telephone contact. In this study, we used OS to evaluate prognosis. OS was defined as the period from the date of surgery to the date of death or last visit. The median follow-up time was 44.18 months (range 3–114 months). Informed consent was obtained from all patients.

FFPE specimens were stained with H&E and examined by experienced pathologists. At least three representative regions were selected and marked on H&E slices each tumor. The corresponding part on the wax block was selected as the material selection site according to the mark on the H&E slice. TMAs containing the 119 FFPE specimens were constructed by using a Manual Tissue Arrayer (TM1, boyikang, Bejing, China). All three marked regions on each tissue with a diameter of 2.0 mm were removed using a TMA instrument and placed into the acceptor wax block from top to bottom in a right-to-left order.

All TMA slides with a thickness of 4 μm were deparaffinized using xylene and rehydrated using graded ethanol. Then, the slides were immersed and boiled in ethylene diamine tetraacetic acid (pH 9.0) for 30 min in a pressure cooker for antigen retrieval. Endogenous peroxidase was inhibited by 3% H2O2 for 15 min. For YKL-39 staining, recombinant monoclonal rabbit anti-human YKL-39 (1:150; ThermoFisher, USA) was applied. For TAM evaluation, monoclonal mouse anti-human CD68 (ready-to-use; maixin, China) was applied. For micro vessel density (MVD) evaluation, monoclonal mouse anti-human CD34 (ready-to-use, maixin, China) was used. Specimens were incubated with all antibodies overnight at 4 °C. After washing with PBS, the slides were incubated with secondary antibody at 37 °C for 10 min. The following steps were performed: color development, counterstaining, differentiation, dehydration, and transparency. Finally, the slides fixed with neutral resin.

IHC slices were examined at a low magnification (× 100) and the most representative five high-magnification (× 400) fields were selected for staining assessment. A semi-quantitative IHC scoring criterion was used to determine the YKL-39 protein expression levels in tumor. The percent positivity of staining cells ranged from 0 to 4: 0, none; 1, 1–25%; 2, 26–50%; 3, 51–75%; 4, 76–100%. The intensity of staining was graded from 0 to 3 (0, no staining; 1, weak; 2, moderate; and 3, strong). We obtained the final IHC score by multiplying the proportion score by the intensity score. A median score of 6 (> 6 or ≤ 6) was selected as the cutoff to distinguish patients with YKL-39-positive or YKL-39-negative expression. CD68 expression was determined by the number of positive cells expressed, not the intensity of staining. The proportion of cells staining positively for CD68 to tumor interstitial cells was calculated as the CD68-positive rate which was used as a score, ranging from 0 to 10. By counting the staining results and calculating the median, it was determined that 6 was the cutoff point to distinguish between high and low expression of CD68, with a score ≥ 6 for high expression and < 6 for low expression [20]. The number of CD34-positive labeled vessels was determined with each vessel counted as one point. Five × 200 fields of view were observed and the average value was calculated. A median of 17.7 was calculated by counting the staining results to distinguish between dense and sparse angiogenesis [26].

Formalin fixed, paraffin embedded sections were deparaffinized by xylene and rehydrated. The slices were placed in a pressure cooker with sodium citrate and boiled, then cooled gradually to room temperature to complete the antigen repair. Blocking solution consisted of 0.3% Triton + 10 mg/mL bovine serum albumin + PBS and used for 1 h. The primary antibodies used include YKL-39 (1:100, Bioss, China) and CD68 (1:50, BOSTER, China) and were incubated with the tissue overnight at 4 °C. The secondary antibodies used were goat anti-mouse-Alexa Fluor 594 (1:200) and goat anti-rabbit-Alexa Fluor 488 (1:200) and were used according to instructions. The secondary antibodies were dropped onto the tissues and incubated for 2 h at room temperature and protected from light. After cleaning the slides with PBS dilution, anti-fluorescence quenching sealing liquid (including DAPI) was added and the slides were stored at 4 °C and protected from light.

Confocal laser scanning microscopy was performed with a Leica TCS SP8 laser scanning spectral confocal microscope (OLYMPUS, FV10-ASW2.1 Viewer, Japan), equipped with a 63 £ 1.32 objective. Excitation was with laser emitting at 405 nm, 488 nm, and 559 nm. All two-or three-color images were acquired using a sequential scan mode.

SPSS 25.0 was used for statistical analysis. The relationship between YKL-39, CD68, CD34, and clinicopathological characteristics of patients was tested by a χ² test. Correlation analysis used Spearman's rank correlation coefficient. Survival analyses were conducted using the Kaplan-Meier method and differences in survival were examined using the log-rank test. Univariate and multivariate analyses were conducted using the Cox proportional hazards regression models. Statistical significance was defined as a value of p < 0.05.