Diagnostic and Prognostic Values of HIF1A-AS2 and LINC00511 in Gastric Cancer with Infection.

Introduction
Gastric cancer (GC) ranks as the fifth most common malignancy globally, with approximately 660,000 new cases reported in 2020.1 Despite advances in treatment strategies—including surgery, radiotherapy, chemotherapy, immunotherapy, and complementary traditional Chinese medicine—GC remains the third leading cause of cancer-related mortality, accounting for nearly 800,000 deaths (7.7% of all cancer fatalities) in the same year.2–6 This poor prognosis is largely due to the disease’s insidious onset and late diagnosis.
A major risk factor for GC is infection with Helicobacter pylori (H. pylori), a Gram-negative bacterium classified as a Group 1 carcinogen by the WHO.7 Chronic H. pylori infection can trigger gastritis and gastric atrophy, ultimately progressing to GC, and is implicated in 65–80% of non-cardia gastric cancer cases.1,8 Although serum biomarkers such as CEA, CA19-9, and CA72-4 are routinely used to aid diagnosis and prognosis, they lack specificity for H. pylori-associated GC9–11 Thus, there is a pressing need to identify more reliable diagnostic and prognostic tools tailored to this GC subtype.
Long non-coding RNAs (lncRNAs), defined as transcripts longer than 200 nucleotides with limited protein-coding potential, have emerged as key regulators of tumor invasion, metastasis, and patient outcomes in GC.12–14 Notably, LINC00511 promotes GC progression via the miR-29c-3p/TRIP13 and AKT/mTOR pathways,15,16 while HIF1A-AS2 enhances proliferation and metastasis through the miR-429/PD-L1 axis.17 Although several lncRNAs—such as FOXD2-AS1, LINC00152, and H19—have shown promise as biomarkers, particularly in H. pylori-positive GC, their clinical utility remains incompletely explored.18,19 A systematic understanding of lncRNA involvement in H. pylori-driven GC is still lacking.
To address this gap, we integrated transcriptomic data from H. pylori-infection-related datasets and GC gene expression profiles from the GEO repository (GSE224056).20 Through this approach, we identified HIF1A-AS2 and LINC00511 as candidate lncRNAs and further validated their diagnostic and prognostic significance in H. pylori-associated GC. This study aims to establish a foundation for lncRNA-based stratification tools to guide individualized treatment strategies.

Materials and Methods

Data Resource and Processing
We acquired RNA-Seq expression data (from five GC patients) from GSE224056 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE224056) in the GEO database. Differentially expressed genes in GC were identified using GEO2R (https://www.ncbi.nlm.nih.gov/geo/geo2r/) with FDR ≤0.05, and fold change ≥2. To identify potential H. pylori infection-related differentially expressed lncRNAs, the intersection between differentially expressed genes from GSE224056 and H. pylori infection-related differentially expressed lncRNAs reported in a previous study was analyzed in Venny 2.1.0 (http://www.liuxiaoyuyuan.cn/).20

Patients and Tissue Samples
A total of 80 pairs of tumor and paracancerous tissues from patients with GC were retrospectively collected from Dongying People’s Hospital from January 2022 to December 2024. Two pathologists verified the clinicopathological features and GC diagnoses of the patients. All subjects completed informed consent forms, and their medical records, including sex, age, tumor size, pathological type, H. pylori infection, tumor node metastasis (TNM) stage, lymph node metastasis, and serum concentrations of GC biomarkers, were recorded. Patients with GC who had received radiation therapy or chemotherapy were excluded. After radical resection, GC tissues and nearby normal tissues (at least 2 cm from the primary cancer site) were removed and promptly frozen at −80°C until total RNA extraction. Patients were followed up for 60 months after surgery from June 7th, 2019, unless death occurred. This study was approved by the Ethics Committee of Dongying People’s Hospital (Approval No: 2024–082).

Quantitative Reverse Transcriptase Polymerase Chain Reaction (qRT-PCR)
Total RNA from the tumor and adjacent tissues was extracted with the TRIzol™ Plus RNA Purification Kit (12183555, ThermoFisher, Waltham, MA, USA), and RNA quality was identified by the A260/280 ratio using an ultraviolet spectrophotometer. The SuperScript™ VILO™ complementary DNA (cDNA) Synthesis Kit (11754050, ThermoFisher) was used to generate cDNA). Afterwards, the cDNA and primers were used for qRT-PCR with the help of Fast SYBR™ Green Master Mix (4385612, Thermo Fisher) in a QuantStudio™ 7 Pro real-time PCR instrument (A43165, Thermo Fisher), with the following conditions: 95°C for 20s, and 40 cycles at 95°C for 3s and 60°C for 30s. Finally, all obtained data were processed according to the 2−ΔΔCT method with GAPDH as the internal control. The primer sequences used are listed in Table 1.21

Detection of CEA, CA 19-9, and CA 72-4
Assay kits for CEA (V517440, Snibe, Shenzhen, China), CA 19–9 (V517453, Snibe), and CA 72–4 (V253864, Mindray, Shenzhen, China) were purchased. The concentrations of the GC biomarkers CEA, CA 19–9, and CA 72–4 in the serum of patients were detected using the electrochemiluminescence method on a Centro XS3 LB 960 Microplate Luminometer (Stuttgart, Germany).

Statistical Analyses
Data analysis was performed using GraphPad Prism 8.4 (GraphPad, Inc., La Jolla, California, USA) and SPSS Statistics 27.0 software (IBM Corp., Armonk, NY, USA). Pearson’s χ2 test or two-sided Fisher’s exact test was used to evaluate associations between discrete variables. An independent t-test was used to examine expression variations between the two groups. Pearson’s correlation coefficient was used to analyze the correlation between the two continuous variables. Utilizing a receiver operating characteristic (ROC) curve, the diagnostic value (specificity and sensitivity) was assessed using a receiver operating characteristic curve. The time between a patient’s date of curative surgery and their death or their final contact date if they were still alive was referred to as overall survival (OS). The Kaplan-Meier method was used to study the probability of survival. Additionally, in multivariate studies, the prognostic value of different covariates was determined using the Cox proportional hazards model. Statistical significance was set at P < 0.05.

Results

Identification of Differentially Expressed lncRNAs in H. pylori Infection-Related GC Cancer
A total of 24,253 differentially expressed genes were identified in the GSE224056 dataset (Figure 1A). These genes were intersected with 73 differentially expressed lncRNAs in H. pylori-infected GC cell lines, resulting in 24 candidate lncRNAs (Figure 1B). Therein, we found that HIF1A-AS2 and LINC00511 who were in top 50 H. pylori infection-associated differentially expressed lncRNAs have been experimentally verified to be highly expressed in GC by literature screening. Subsequently, we extracted the expression profiles of HIF1A-AS2 and LINC00511 from the GSE224056 dataset. As shown in Figure 1C, HIF1A-AS2 was highly expressed in tumor tissues compared to that in paracancerous tissues (para-tissue). In addition, upregulation of LINC00511 was observed in tumor tissues relative to paracancerous tissues (Figure 1D). In short, HIF1A-AS2 and LINC00511 were abnormally highly expressed in GC tumor tissues.

Validation and Clinical Values of HIF1A-AS2 and LINC00511
To verify the high expression levels of HIF1A-AS2 and LINC00511, we examined their levels in 80 pairs of tumor and adjacent non-tumor tissues from patients with GC by qRT-PCR. As shown in Figure 2A and B, we observed markedly high expressions of HIF1A-AS2 and LINC00511 in tumor tissues (P < 0.001). Further, we divided patients into two groups (Low and High) with median values as cut-off values to analyze the correlation between clinicopathological factors and HIF1A-AS2 or LINC00511 expression. We found that high levels of HIF1A-AS2 or LINC00511 were related to tumor size, H. pylori infection, TNM stage, and lymph node metastasis (Table 2). The bar graphs affirmed that HIF1A-AS2 and LINC00511 are highly expressed in patients with tumor size >5 cm, H. pylori infection, stage III and IV, and lymph node metastasis (Figure 2C–J, P < 0.01). These findings indicate that high expression levels of HIF1A-AS2 and LINC00511 are involved in the clinicopathological factors of GC patients.

Correlations Between HIF1A-AS2 and LINC00511 with Concentrations of Serum CEA, CA 19-9, and CA 72-4
Moreover, the Pearson correlation coefficient showed that HIF1A-AS2 expression positive correlated with the levels of the GC biomarkers CEA (Figure 3A, P < 0.001, r = 0.705), CA 19–9 (Figure 3B, P < 0.001, r = 0.637), and CA 72–4 (Figure 3C, P < 0.001, r = 0.710). In addition, positive correlations between LINC00511 expression and CEA (Figure 3D, P < 0.001, r = 0.600), CA 19–9 (Figure 3E, P < 0.001, r = 0.705), and CA 72–4 (Figure 3F, P < 0.001, r = 0.632) levels were found in GC patients. These results demonstrated that HIF1A-AS2 and LINC00511 expression is positively correlated with the levels of GC serum biomarkers. Also, HIF1A-AS2 expression was positive correlated with LINC00511 expression (Figure 3G, P < 0.001, r = 0.590).

Diagnose Values of HIF1A-AS2 and LINC00511
The ROC curve was generated to assess the prognostic significance of HIF1A-AS2 and LINC00511 in GC patients. For HIF1A-AS2, the area under the curve (AUC) was 0.759. The specificity and sensitivity for GC were 75% and 70%, respectively (Figure 4A and Table 3, P < 0.001). For LINC00511, the AUC was 0.816, with a specificity of 72.5% and sensitivity of 80% (Figure 4B and Table 3, P < 0.001). We then predicted the combined diagnostic value of HIF1A-AS2 and LINC00511. As shown in Figure 4C and Table 3, the AUC was 0.839 (P < 0.001), and the specificity and sensitivity for GC were 71.25% and 81.25%, respectively. Altogether, the high expression of HIF1A-AS2 and LINC00511 showed good diagnostic value for GC, and the combined diagnosis of HIF1A-AS2 and LINC00511 had a superior effect.

Prognostic Values of HIF1A-AS2 and LINC00511
Furthermore, according to the median expression of HIF1A-AS2/LINC00511, patients were divided into two groups, high expression and low expression, to evaluate the prognostic value of HIF1A-AS2 and LINC00511 alone or in combination. The survival curves in Figure 5A and B showed that increased expressions of HIF1A-AS2 and LINC00511 was associated with poor 5-year OS (P < 0.05). Similarly, patients with high HIF1A-AS2/low LINC00511 (Figure 5C) showed significantly reduced survival (P < 0.05). The high LINC00511/low HIF1A-AS2 subgroup (Figure 5D) exhibited a moderately elevated risk of mortality (P < 0.05). In contrast, the low HIF1A-AS2/low LINC00511 group (Figure 5E) and high HIF1A-AS2/high LINC00511 expression group (Figure 5F) were not associated with the worst survival. Univariate and multivariate Cox regression analyses were performed to explore the prognostic values of HIF1A-AS2 and LINC00511. As shown in Table 4, the expression of HIF1A-AS2 (P=0.014; HR=1.855; 95% CI=1.135–3.031) and LINC00511 (P=0.018; HR=1.808; 95% CI=1.107–2.955) were significantly associated with OS of GC patients, and other clinicopathological factors including tumor size, TNM stage, and lymph node metastasis were also presented relevant with OS.

Discussion
Comparatively, lncRNAs are more tissue-specific and are increasingly used as leading-edge diagnostic and prognostic indicators. The earliest study on lncRNA expression in GC was published in 1997, in which heterozygosity for H19 was observed in 28 patients.22 To date, lncRNA H19 overexpression has been shown to promote the carcinogenesis and metastasis of GC and is regarded as a potential diagnostic marker in GC.23,24 In addition, a large number of lncRNAs have been reported to be involved in GC progression through a variety of pathways, such as epigenetic, chemoresistance, and metabolic reprogramming.25–27 However, only a few studies have focused on lncRNAs in H. pylori-related GC. Our study identified two key lncRNAs, HIF1A-AS2 and LINC00511, and explored their diagnostic and prognostic value in H. pylori-associated GC.
HIF1A-AS2 serves as a specific negative regulator of hypoxia-inducible factor-1 alpha (HIF-1α).19 Yang et al found that HIF1A-AS2 was highly expressed in GC tumor tissues from 50 patients and cell lines and suggested that the upregulation of HIF1A-AS2 is involved in poor 5-year survival.17 Furthermore, HIF1A-AS2 was reported to promote the proliferation and metastasis of GC cells through miR-429/PD-L1.17 A previous study indicated that overexpression of HIF1A-AS2 was correlated with various clinical indicators, including TNM classification, tumor invasiveness, lymphatic spread, and adverse prognosis.28 Chen et al reported that the AUC of the ROC curve for HIF1A-AS2 was as high as 0.673.28 Similarly, we also suggest that HIF1A-AS2 expression is related to clinicopathological characteristics and has great potential as a diagnostic biomarker of GC.
In addition, high LINC00511 expression has been observed in GC tumor tissues and cells, which leads to GC progression. For instance, Cui et al demonstrated that LINC00511 expression in tumor tissues was higher than that in adjacent tissues from 50 GC patients, and LINC00511 active signal transducer and activator of transcription 3 (STAT3) pathways expedites GC cell promotion and migration.29 Wang et al revealed that LINC00511 mediates PTEN promoter methylation to enhance the proliferation, migration, and stemness of GC cells by activating the AKT/mTOR pathway.30 Importantly, the correlation between LINC00511 expression and the clinical features of patients with GC has attracted attention. In a previous study enrolling 25 patients with GC, increased LINC00511 expression was reported to be related to larger tumor size (> 5 cm) and advanced TNM stage.31 We observed that some larger tumors (>5 cm) showed low HIF1A-AS2 and LINC00511 expression. This may be due to tumor heterogeneity or alternative pathways making these lncRNAs less critical in advanced stages. Here, we further confirmed that high LINC00511 expression has a strong correlation with clinicopathological characteristics including tumor size, TNM stage, H. pylori infection, lymph node metastasis, and serum GC biomarkers, and suggested that LINC00511 has a huge diagnostic value in GC. In addition, our study also suggested the diagnostic value of HIF1A-AS2 and LINC00511 in H. pylori-related GC. The PI3K/AKT and JAK/STAT pathways were activated in H. pylori-induced GC, which may explain the link between HIF1A-AS2 and LINC00511 with H. pylori-related GC.32 Unexpectedly, elevated lncRNA levels also occurred in some H. pylori-negative or node-negative samples. This could stem from prior undetected H. pylori exposure or other non-H. pylori drivers such as EBV or metabolic factors.
There has been little clinical research on the role of LINC00511 in GC. Only Huang et al revealed that the increase in LINC00511 expression in tumor tissues is associated with poor OS in patients with GC using the Kaplan-Meier method.33 Our study suggests a correlation between high LINC00511 expression and poor prognosis in GC patients in 5 years. The prognostic value of LINC00511 has also been identified in other cancers. A meta-analysis involving 1024 cancer patients revealed that patients with malignant tumors may have lower OS and disease-free survival (DFS) if their LINC00511 expression levels are elevated.34 Specifically, high levels of LINC00511 are associated with worse OS in multi-system cancers, such as breast cancer, renal cell cancer, glioma, hepatocellular carcinoma, pancreatic ductal adenocarcinoma, non-small-cell lung cancer, ovarian cancer, and cervical cancer osteosarcoma.34 In addition, HIF1A-AS2 has been regarded as a novel prognostic biomarker for triple-negative breast cancer.35 Guan et al reported that HIF1A-AS2 is a potential prognostic factor in human papillomavirus-related cervical cancer.36 In lung adenocarcinoma specimens, HIF1A-AS2 levels were elevated, which is indicative of worse OS and DFS.37 These findings suggest that HIF1A-AS2 and LINC00511 are associated with poor prognosis in patients with cancer, suggesting that HIF1A-AS2 and LINC00511 are prognostic markers in multiple cancers.
This study innovatively identified HIF1A-AS2 and LINC00511 as key lncRNAs significantly associated with H. pylori-infected GC. Their elevated expression correlates strongly with aggressive clinicopathological features and serves as a powerful diagnostic and prognostic biomarker, highlighting their critical role in H. pylori-driven GC progression and offering new potential for precise patient stratification. However, despite the analytical exploration of the clinical correlation of HIF1A-AS2 and LINC00511, additional limitations must be acknowledged when interpreting the findings of the current study. First, we collected tumor tissues and adjacent tissues of GC patients for expression level detection but did not include normal individuals in this study. This is insufficient to fully clarify the clinical value of HIF1A-AS2 and LINC00511. Second, inadequate validation testing without animal models and GC cell lines may necessitate future refinement of the experimental design for more precise biological observations. Finally, elucidation of the molecular mechanisms regulating the effects of HIF1A-AS2 and LINC00511 on GC tumorigenesis and progression requires additional research.

Conclusion
In conclusion, this study revealed that high expression of HIF1A-AS2 and LINC00511 is associated with clinicopathological factors in gastric cancer patients. HIF1A-AS2 and LINC00511 showed high sensitivity and specificity in the diagnostic models for gastric cancer. Elevated expression of LINC00511 and HIF1A-AS2 has been linked to low 5-year overall survival rates. These findings provide compelling evidence that HIF1A-AS2 and LINC00511 are potential biomarkers for the diagnosis and prognosis of gastric cancer.