Integrating Salivary Biomarkers CST4 and With Health-Related Factors for Gastric Cancer Detection and Risk Assessment.

Introduction
Stomach cancer is the fifth most common malignancy worldwide and represents a major health challenge due to its high mortality rate.
1
The prevalence of gastric cancer (GC) varies significantly across different countries and cultural contexts, with developing countries accounting for half of all newly diagnosed cases. In certain regions of the world, particularly in Central Asian countries and Iran, as well as parts of Latin America, GC is the predominant cause of cancer-related death.2,3 Despite substantial progress in medicine, the early detection of GC continues to pose a significant challenge, frequently leading to poor prognoses and low survival rates.4,5 In contrast, countries like Japan that implement systematic screening and early detection programs exhibit improved survival rates.6-9
The etiology of GC is multifactorial, involving a complex interplay between environmental and genetic influences.
10
Crucial variables linked to GC encompass infection with Helicobacter pylori,11,12 tobacco
13
and alcohol consumption, high salt intake,
14
obesity,
15
physical inactivity,
16
and consumption of coffee.17,18 To tackle the burden of this malignancy, researchers have been actively exploring both primary and secondary prevention strategies. Primary prevention strategies emphasize lifestyle modifications,
19
and secondary prevention focuses on early diagnosis and timely treatment, particularly for individuals with a familial history of GC who are at elevated risk.
20

Endoscopic screening is recognized as a cost-effective strategy in regions with a high prevalence of GC.
21
It is important to note that endoscopy is an invasive procedure,
22
associated with potential risks such as bleeding,23,24 and mucosal perforation.
25
However, these imaging modalities face limitations, including restricted accessibility, challenges training healthcare professionals to interpret complex images, variability in imaging acquisition parameters, and concerns regarding their diagnostic accuracy.
26
Primary prevention strategies emphasize lifestyle modifications,
19
and secondary prevention focuses on early diagnosis and timely treatment, particularly for individuals with a familial history of GC who are at elevated risk.
20
Recent advances in systemic therapy, particularly immunotherapy,
27
nivolumab/chemotherapy in unresectable GC,28,29 and targeted agents,
30
have transformed survival outcomes in advanced disease. However, their efficacy remains limited to patients diagnosed at non-metastatic stages. Thus, reliable early detection tools are fundamental to enabling curative-intent treatment and maximizing therapeutic benefit from these novel regimens.
Given that stomach cancer frequently presents with severe symptoms, effective non-invasive screening for early detection is crucial in reducing mortality associated with this malignancy. Among the promising avenues of cancer research, the role of novel biomarkers has emerged as a focal point. Saliva, a non-invasive and readily accessible biofluid, has attracted considerable interest as a potential diagnostic tool for various malignancies.
31
Recent technological advancements have effectively overcome earlier limitations, positioning salivary biomarkers as promising options for early diagnosis.32,33 Cystatin S (CST4), also known as Cystatin SA-III, a non-glycosylated protein found in saliva and tears, exhibits characteristics similar to blood biomarkers, with increased expression levels noted in GC cells.34,35 CST4 has been shown to enhance the invasiveness of GC and promote its progression.
36
Notably, serum CST4 has been well-defined as a marker for detecting certain malignancies, particularly gastrointestinal tumors, where it is significantly elevated.34,37 The differential gene expression analysis tool showed that CST4 had the highest expression among the cystatin protein group in saliva.38,39 Salivary CST4 is a promising biomarker in various oral health conditions, particularly in disease such as primary Sjögren’s syndrome.40,41
CST4, found in the cytoplasm, has potential as a biomarker due to its low molecular weight and its secretion in saliva.42,43 Moreover, studies have confirmed that CST4 overexpression in tissues demonstrates high specificity and sensitivity, suggesting it could become a novel blood biomarker for GC.37,44,45 However, the clinical value of CST4, in combination with miRNA tumor markers, in diagnosing gastrointestinal malignancies has not been well-established in the literature. In conclusion, while CST4 remains a promising tumor marker, further clinical data are needed. The diagnostic potential of MicroRNA-223 (miR-223) in GC has gained considerable attention due to its strong association with tumor progression.46-48 Since miR-223 can be detected in both plasma and tissue samples, it emerges as a promising non-invasive biomarker for diagnosis GC.
48
Higher miR-223 levels are linked to worse outcomes in GC, showing a positive correlation with advanced TNM stages and lymph node metastasis. This suggests that miR-223 could be a valuable biomarker for risk-based patient stratification.49,50
Helicobacter pylori infection, a major risk factor for GC, can increase miR-223 expression through its CagA protein, connecting chronic inflammation to cancer progression.
51
When analyzed alongside demographic and clinical data, salivary CST4 and miR-223 levels’ potential diagnostic significance offers a promising opportunity to transform early detection strategies in GC. The main goal of this article is to explore the diagnostic accuracy of salivary CST4 and miR-223 levels, combined with demographic and clinical factors, for GC detection.

Methods

Ethical statement
The study was approved by the Ethical Committee of the Tehran University of Medical Sciences (Ethical Code: IR.TUMS.DENTISTRY.REC.1399.229). Following the elucidation of the study aims, all participants voluntarily furnished signed informed consent before their involvement. Individuals were made aware that they could opt out of the study without any obligation. All methods were performed in accordance with the Helsinki Declaration of 1975, as revised in 2024.
52

Samples
This retrospective case-control research involved 45 healthy persons, and 45 patients with early-stage gastric adenocarcinoma were selected as a convenience series from Imam Khomeini Hospital in Tehran. Gastroenterologists have diagnosed GC based on histological and endoscopic evaluations. This study was conducted over a period of six months, from January 2023 to June 2023. The time interval between the histopathological diagnosis and saliva sampling was within four weeks for all patients, and no clinical interventions such as surgery, chemotherapy, or radiotherapy were administered before saliva collection. Patients with a history of such treatments before saliva sampling were excluded from the study to prevent confounding effects on biomarker levels. To ensure the integrity of the study, specific exclusion criteria were established for participants. These criteria included:
Individuals exhibiting active oral or periodontal diseases.

Patients with additional malignancies or a documented history of any tumors and systemic diseases.

Patients who have undergone surgery, chemotherapy, or radiotherapy prior to saliva collection.

Patients possess a record of blood transfusions during the prior 3 years.

Women who were pregnant.

The control consisted of healthy individuals referred to Imam Khomeini Hospital (Tehran, Iran) for medical checkups, confirming the absence of current oral infections, inflammatory conditions, cancers, or systemic diseases. In addition, in the control group, individuals with blood in their stool, unexplained weight loss, hyperactive intestinal movements, or long-term fatigue were also excluded. Efforts were made to match the patients with GC and the control group for sex and age to minimize potential confounding variables (Figure 1). The reporting of this study adheres to STARD guidelines (Standard Protocol for Reporting Diagnostic Accuracy Studies) (Supplemental Table 1-STARD-2015).
53

All participants underwent a comprehensive clinical evaluation, including a review of medical history, which led to the exclusion of patients with a documented history of cancer (either self-reported or clinically diagnosed). Screening tests, including complete blood count (CBC) and tumor marker assessments (carcinoembryonic antigen (CEA),
54
alpha-fetoprotein (AFP),
55
cancer antigen 15-3 (CA15-3),
56
cancer antigen 125 (CA125),57,58 prostate-specific antigen (PSA),
59
carbohydrate antigen 19-9 (CA19-9)),60,61 were conducted to identify occult malignancies for case and control groups. Patients with a history of chemotherapy, radiotherapy, or blood transfusions within the past 3 years were excluded to reduce the risk of undiagnosed hematologic malignancies. The healthy control group reporting alarm symptoms, such as unexplained weight loss, hematemesis, melena, dysphagia, lymphadenopathy, abdominal masses, or persistent vomiting, were excluded.
During the study, participants were requested to complete a standardized and comprehensive checklist to provide their information and habits potentially related to GC progression. The data included demographic data, hereditary cancer predisposition, smoking habits, consumption of salty and fast foods, hot drinks, Helicobacter pylori infection, and the DMFT (Decayed, Missing, and Filled Permanent Teeth) index.62,63 The form-filling process was supervised by a trained author who remained neutral throughout. Clarifications were provided when necessary, ensuring that no guidance or bias influenced the participants’ responses. The information from the questionnaires was categorized for analysis, ensuring both the validity and completeness of the data.
Demographic characteristics (sex, age, BMI, occupational status, ethnicity) and clinical history (gastric ulcers) were assessed using the World Health Organization (WHO) STEPwise Approach to Surveillance (STEPS) Questionnaire,
64
with ethnicity categories standardized to Iran’s national census. Family history of malignancy was evaluated via an adapted Cancer Family History Questionnaire.
65
Smoking status (current/past) was determined using the WHO Global Adult Tobacco Survey (GATS), Section B.
66
At the same time, dietary factors (salty food, fast food, and hot drink) were quantified using the food frequency questionnaire (FFQ) of PERSIAN Cohort.67,68 The frequency score was defined as 0 for those who did not consume these items weekly and 1 for those who consumed them at least once per week.
69
The DMFT index was determined using the participants’ panoramic X-rays
62
and oral examinations by three oral medicine specialists based on the WHO criteria for DMFT.
70
Participants’ occupations were categorized into three groups: employed, homemaker, and retired.71,72 The participants were categorized into two groups according to their smoking status: tobacco users and those who had never used tobacco.73,74 All participants were encouraged to report any gastroesophageal reflux disease (GERD) history.

Saliva sample collection
To mitigate circadian rhythm influences on salivary secretions, saliva was obtained at 9:00 and 11:00 a.m. A 90-minute abstention from drinking, smoking, eating, and mouth hygiene was mandated prior to the sample to remove gustatory or mechanical stimulation. Following a comprehensive dental and periodontal examination, unstimulated whole saliva was obtained the spitting. Participants were instructed to accumulate saliva in their oral cavity and spit into a sterilized container every minute up to 15 minutes. Centrifuge the saliva samples, designated with coded labels for confidentiality, to isolate the liquid contaminants. Further examination was conducted utilizing the supernatant. The index tests were performed blinded to the reference standard results and clinical information. Samples were processed using coded identifiers, and laboratory personnel had no access to participants’ diagnostic status during the analysis. The reference standard results were only unblinded after completion of all laboratory measurements for final comparative analysis. Following centrifugation, samples were stored at −80°C to preserve their integrity.

Measurement of salivary CST4
Samples were preserved at −80°C until subjected to enzyme-linked immunosorbent assay (ELISA) analysis. Salivary CST4 levels were quantified utilizing a 96-well human CST4 ELISA kit (Cat. No. ZB-12150 C-H9648 (PRO-1097), ZellBio GmbH, Germany) in accordance with the manufacturer’s instructions. In this process, the CST4 protein was introduced into wells pre-coated with anti-human CST4 monoclonal antibodies. Subsequently, biotin-labeled anti-CST4 antibodies were added, which then formed an immune complex by binding with streptavidin-HRP. With a sensitivity of 2.5 ng/ml, the CST4 ELISA kit measured 50–1600 ng/ml. A Hyperion ELISA microplate reader measured sample absorbance. The spectrometer software computed CST4 concentrations using standard curves. The average of triplicate sample measurements was given.

Extraction and Purification of RNA
To achieve sample homogenization, 500 μL was transferred to sterile microtubes, followed by the addition of 800 μL of TRIzol reagent (Selecta, Spain). The mixture was thoroughly vortexed to achieve homogeneity and allowed to stand for 5 minutes. The solution was subjected to 200 μL of chloroform (EMSURE) to initiate phase separation. Following a 5-minute incubation, the tubes underwent centrifugation at 12,000 revolutions per minute for a duration of 10 minutes in a refrigerated centrifuge. This resulted in three distinct layers: the upper aqueous layer containing RNA, the middle layer including DNA, and the lower organic layer with proteins and other organic compounds.
To precipitate and wash RNA, the top liquid layer was moved to a fresh sterile microtube, and 800 μL of pre-cooled −20°C isopropanol was added. To ensure thorough mixing, the tubes were gently shaken and centrifuged at 12,000 rpm for 15 minutes. After a white RNA pellet had developed at the tube bottom, the supernatant was eliminated. A centrifuge was run for 3 minutes at 14,000 rpm after 500 μL of 70% ethanol was added to remove any leftover contaminants. This ethanol washing step was repeated once more. After removing ethanol, the tubes were air-dried for 2–3 minutes. The RNA pellet was then left to dry at room temperature on a dry block. The RNA was dissolved in 20 μL of RNase-free water (GENEX), protecting it from enzymatic degradation (DEPC-treated), and the solution was stored in the freezer for long-term preservation.

Real-Time Quantitative Polymerase Chain Reaction (qPCR) Assessment
The samples were extracted using a nanodrop technique, and RNA concentration was measured. For a blank measurement, 2 μL of DEPC-treated water and samples were added to the BioTek Epoch (BioTek, Gen5) microplate. The Gen program was set to measure nucleic acids using the quantification function. Absorbance was measured at 260 nm for nucleic acids and 280 nm for proteins. A reference range of 1.8–2 was used to compute the nucleic acid to protein absorption ratio, and RNU6 (U6) was used as the internal control. Stem-loop primers were used for the target to appropriately adjust the data (Supplemental File). Polymerase chain reaction (PCR) was conducted on two saliva samples, and each primer was validated against them. To confirm primer specificity, the PCR results of these two samples were subjected to electrophoresis on an agarose gel.

Synthesis of Complementary DNA (cDNA)
The production of cDNA is essential for amplifying genes through PCR, as cDNA serves as the substrate for DNA polymerase. For the synthesis of cDNA from RNA templates, we employed the FIREScript RT cDNA Synthesis Kit provided by Solis BioDyne. This kit contains reverse transcriptase, an enzyme that reads RNA sequences and synthesizes cDNA strands. Consequently, the amount of synthesized cDNA is directly related to the expression levels of the target RNA.
Initially, mix RNA with primer, combine, and dNTPs for each gene. The samples were subsequently placed in a thermal cycler and modified according to the kit’s specifications, with program instructions adhered to. Following the reaction, the samples were frozen.

Real-Time PCR
Real-time PCR was conducted by the Ampliqon SYBR assay technique, which enables the monitoring of reaction progress in real time through the fluorescent dye SYBR Green. Prepare total reaction by mixing water, GENEX Master Mix, primers, and synthesized cDNA. This solution was then placed in instrument-compatible microtubes. Capped microtubes were put in the Step One Plus device (Applied Biosystems).
Samples were heated to 95°C for 10 minutes to ensure the complete inactivation of any remaining reverse transcriptase enzyme. The amplification process consisted of 45 cycles, each including a 15-second denaturation step at 95°C and a 30-second annealing phase at 60°C to support primer binding and enzyme function. Data, including amplification curves, were generated by the instrument software as outlined in the protocol. Each sample was analyzed in duplicate, and consistent Ct values were obtained for analysis.

Helicobacter pylori Infection
Stool samples (fresh without macroscopic evidence of blood) were collected in sterile containers. Stool samples were preserved at −20°C until subjected to ELISA analysis. Helicobacter pylori antigen levels were quantified using a commercial ELISA kit (PISHTAZ TEB, Tehran, Iran) according to the manufacturer’s protocol.
75

Statistical analysis
The statistical analysis used SPSS 18.0.0 (SPSS Inc., USA) and GraphPad Prism 9.0 (GraphPad Software, USA). A p-value (p) below 0.05 was deemed statistically significant. For comparing quantitative factors between the two groups, a Student’s t-test was used. The relationship between the presence of GC and explanatory variables was assessed using multiple logistic regression (MLR) analyses. To determine the diagnostic value of salivary CST4 and miR-223 in distinguishing patients with GC, a receiver operating characteristic (ROC) curve was generated. The optimal cutoff points for the ROC curves were identified using Youden’s index. Furthermore, principal component analysis (PCA) and MLR were implemented to investigate the cumulative impacts of salivary microRNAs (miRNAs), clinical characteristics, and demographic data. The sample size of 45 participants per group was determined using PASS software v15.0 (NCSS, LLC, USA) for ROC curve analysis. Assuming an AUC of 0.85, a significance level of 0.05, and a 95% confidence interval (CI) width of 0.161, this sample size provided adequate power to detect statistically significant differences between groups.

Results

Patient demographics
Table 1 provides a summary of the demographics, clinical features, and laboratory results for cases and controls. The two groups exhibited significant differences in relation to a positive history of gastric ulcers (P < 0.001), hot drinks (P < 0.001), salivary CST4 level, ng/mL (P < 0.001), salivary miR-223 level (2−ΔΔCt) (P < 0.001), DMFT (P = 0.003), salty food consumption (P = 0.005), body mass index (BMI) (P = 0.005), and Helicobacter pylori infection (P = 0.025) (Table 1).
The analysis revealed no statistically significant distinctions between the case and control groups regarding sex, age, smoking status (current or past), fast food consumption, or family history of malignancy.

Salivary CST4 and miR-223-3p
The levels of salivary miR-223-3p (Figure 2A) and CST4 (Figure 2B) were analyzed in GC and healthy individuals. ROC curve analysis (Figure 2C) was employed to evaluate the salivary miR-223-3p and CST4, as well as their combined MLR model for differentiating cases from controls. The area under the ROC curve (AUC) values were determined as 0.84 (95% CI: 0.76–0.93) for miR-223-3p, 0.78 (95% CI: 0.69–0.88) for CST4, and 0.89 (95% CI: 0.83–0.96) for their MLR model. CST4 and miR-223-3p cutoffs were set using Youden’s index at 4.77 (2-∆∆Ct) and 0.035 (ng/mL), respectively.
PCA was conducted on salivary CST4 and miR-223 levels, alongside demographic data, to define the important factors for differentiation of GC cases from controls. The analysis combined salivary CST4 and miR-223 levels, gastric ulcer experience, DMFT status, and Helicobacter pylori infection (Figure 3).
The ROC curve was constructed based on PCA model to assess the ability to differentiate between GC cases and controls. The PCA score distinguishes GC cases with a high sensitivity of 0.80, specificity of 0.71, and an AUC of 0.80 (95% CI = 0.71–0.89). The results of distinguishing between GC cases and controls using salivary miR-223-3p, CST4, and clinodemographic data through MLR are shown in Table 2.

Discussion
GC ranks among the most fatal malignancies worldwide, frequently being diagnosed at advanced stages. Consequently, studies highlight the need for implementing early detection and control programs for GC in regions with high prevalence and mortality rates associated with the disease.
76
Development of efficient early detection and diagnostic methods is consistent with this overall objective. In this context, salivary biomarkers, which include proteins, genes, and metabolic components, are crucial.
77
The discovery of miRNAs has spurred research aimed at identifying biomarkers in oral biofluids that can be utilized for diagnosing various cancers, including miRNAs in salivary exosomes. The identification of miRNA salivary biomarkers presents significant potential for the early diagnosis of malignancies of the gastrointestinal tract, including pancreatic cancer and GC.78,79 Moreover, studies have confirmed that CST4 overexpression in tissues demonstrates high specificity and sensitivity, suggesting it could become a novel blood biomarker for GC.37,44,45 Compared to the numerous studies that have focused on blood and tissue biomarkers for diagnosing GC, limited studies have examined the role of salivary biomarkers in GC detection.78,80,81 However, when integrated with demographic and clinical data, salivary CST4 and miR-223 levels show significant promise in enhancing early detection strategies for GC.
Li et al
48
conducted a study on miR-223 as a potential biomarker for early GC detection, revealing significantly increased miR-223 in the blood of patients with GC relative to controls. Increased levels of serum miR-223 have been reported in patients diagnosed with GC, establishing a significant correlation between miR-223 expression and TNM stage.
82
Aalami et al
83
further associated serum miR-223 expression with cancer grade. Our analysis revealed an elevation of salivary miR-223, aligning with previous observations of increased expression of miR-223 in blood or tissues.
48
,82-84
Li and colleagues demonstrated a significant association between hot tea consumption and an increased risk of GC.
85
In ours, the habit of daily hot tea consumption was more than twice as common in the patients with GC compared with the control group, with 34 individuals in the patients with GC versus 16 in the control group. Although the underlying mechanism of this association is unclear, hot tea consumption may potentially damage the gastric mucosa, increasing its vulnerability to the development of cancer.86,87 It is also possible that hot tea affects the equilibrium of the gastric microbiota, resulting in inflammatory responses and oncogenesis.85,88,89
A recent meta-analysis found a strong link between high salt intake and increased GC risk.
90
Our study’s findings of higher salted food consumption in patients with GC support the need for dietary interventions to lower this risk. In our study, we found that 11 out of 45 patients with GC had a history of gastric ulcer, whereas none of the 45 individuals in the control group had gastric ulcers. This finding aligns with the observed association between gastric ulcer and increased GC risk,
91
further supporting the notion that gastric ulcer may play a significant role in GC incidence.
According to our findings, there was a statistically significant difference in the DMFT score index between the control group and patients with GC. As a measure of dental health, the DMFT score was significantly higher in patients with GC that aligns with findings from other studies.92-94 This study revealed that persons who cleaned their teeth less than once daily had a markedly increased chance of acquiring stomach cardia adenocarcinoma.
92
GC and tooth loss may be associated with microbiota that facilitates the production of carcinogenic compounds. Specifically, absent teeth may harbor an oral microbiome that converts nitrate to nitrite. Nitrite may interact with bodily amines to produce carcinogenic nitrosamines.
95
Among the models examined, the MLP model, including miR-223 levels, CST4, clinical features, and risk variables, produced the greatest diagnostic odds ratio (DOR). Conversely, concentrating just on salivary CST4 resulted in the lowest diagnostic DOR, underscoring the significance of integrating many prognostic parameters.
The integration of machine learning models that incorporate salivary biomarkers, imaging data, and electronic health records will enable dynamic and personalized GC risk prediction, moving beyond traditional static diagnostics. In addition, rapid salivary tests, such as CRISPR-based miRNA detection, will transition from laboratories to point-of-care settings, facilitating large-scale community-level screening, especially in resource-limited environments. Moreover, microbiome-targeted prevention strategies, including probiotic or antibiotic interventions to modulate oral-gastric microbial flux, are expected to emerge as important adjunct therapies for high-risk individuals, potentially reducing disease progression and improving outcomes.
Salivary biomarkers like CST4 and miR-223 show promise in diagnosing GC, but challenges and limitations remain. First, the absence of endoscopic screening in our rigorously matched healthy controls precludes definitive exclusion of asymptomatic pre-malignant lesions (e.g., atrophic gastritis, intestinal metaplasia) or early-stage GC, potentially leading to underestimation of biomarker performance. Second, despite adjustment for known risk factors, residual confounding from unmeasured or imprecisely quantified variables, including exact dietary salt intake, socioeconomic status, specific occupational exposures, and medication use, may influence observed associations. Third, salivary biomarker levels are inherently susceptible to pre-analytical variability. However, stringent standardization was applied, and external validation using harmonized methodologies across independent laboratories remains essential. Our exclusion of oral diseases enhances internal validity but may reduce Generalizability to populations with a high oral inflammation burden. Further studies in dentally diverse cohorts are needed to verify clinical utility. Fourth, the significant association between higher DMFT index and GC, while mechanistically plausible via oral microbiome dysbiosis and nitrosamine production, remains fundamentally observational; this study cannot establish causality or define specific microbial pathways. Fifth, prospective validation in larger, multicentre cohorts is critical to confirm diagnostic performance in real-world screening contexts. Finally, the role of oral microbiota in GC is being explored through metagenomic sequencing of paired saliva and gastric samples to identify pro-carcinogenic pathways. To evaluate real-world utility, prospective multicenter trials are investigating the performance of these biomarkers in diverse populations, including those with dental compromise. In addition, standardization of saliva collection and storage protocols is critical for reproducibility, with international efforts like SALIVARES developing evidence-based guidelines.

Conclusion
This study highlights the potential of salivary biomarkers, particularly miR-223 and CST4, as promising tools for GC early diagnosis. Significant differences in the salivary miR-223 and CST4 were detected between GC cases and controls, with biomarkers showing strong diagnostic performance in differentiating the two groups. The MLR model, incorporating these biomarkers along with clinical and demographic factors, demonstrated high sensitivity, specificity, and accuracy, making it a promising tool for GC detection. In addition, the study identified several lifestyle and clinical factors, such as hot drink consumption, salty food intake, and a history of gastric ulcers, that were significantly associated with GC risk, reinforcing the importance of these factors in GC pathogenesis. The combination of salivary biomarkers and clinodemographic data offers a comprehensive approach for early detection, which is critical in improving patient outcomes given the typically late-stage diagnosis of GC. Our findings support the further exploration of salivary biomarkers, particularly in combination with clinical risk factors, as a non-invasive strategy for GC diagnosis, potentially enhancing early detection programs in high-risk groups.

Supplemental Material

sj-docx-1-onc-10.1177_11795549251375898 – Supplemental material for Integrating Salivary Biomarkers CST4 and miR-223 With Health-Related Factors for Gastric Cancer Detection and Risk Assessment

Supplemental material, sj-docx-1-onc-10.1177_11795549251375898 for Integrating Salivary Biomarkers CST4 and miR-223 With Health-Related Factors for Gastric Cancer Detection and Risk Assessment by Maryam Koopaie, Saba Manafi, Soheila Manifar, Shima Younespour, Sajad Kolahdooz and Mahdi Karimipour Pareshkooh in Clinical Medicine Insights: Oncology

sj-docx-2-onc-10.1177_11795549251375898 – Supplemental material for Integrating Salivary Biomarkers CST4 and miR-223 With Health-Related Factors for Gastric Cancer Detection and Risk Assessment

Supplemental material, sj-docx-2-onc-10.1177_11795549251375898 for Integrating Salivary Biomarkers CST4 and miR-223 With Health-Related Factors for Gastric Cancer Detection and Risk Assessment by Maryam Koopaie, Saba Manafi, Soheila Manifar, Shima Younespour, Sajad Kolahdooz and Mahdi Karimipour Pareshkooh in Clinical Medicine Insights: Oncology

sj-jpg-3-onc-10.1177_11795549251375898 – Supplemental material for Integrating Salivary Biomarkers CST4 and miR-223 With Health-Related Factors for Gastric Cancer Detection and Risk Assessment

Supplemental material, sj-jpg-3-onc-10.1177_11795549251375898 for Integrating Salivary Biomarkers CST4 and miR-223 With Health-Related Factors for Gastric Cancer Detection and Risk Assessment by Maryam Koopaie, Saba Manafi, Soheila Manifar, Shima Younespour, Sajad Kolahdooz and Mahdi Karimipour Pareshkooh in Clinical Medicine Insights: Oncology

sj-jpg-4-onc-10.1177_11795549251375898 – Supplemental material for Integrating Salivary Biomarkers CST4 and miR-223 With Health-Related Factors for Gastric Cancer Detection and Risk Assessment

Supplemental material, sj-jpg-4-onc-10.1177_11795549251375898 for Integrating Salivary Biomarkers CST4 and miR-223 With Health-Related Factors for Gastric Cancer Detection and Risk Assessment by Maryam Koopaie, Saba Manafi, Soheila Manifar, Shima Younespour, Sajad Kolahdooz and Mahdi Karimipour Pareshkooh in Clinical Medicine Insights: Oncology

sj-jpg-5-onc-10.1177_11795549251375898 – Supplemental material for Integrating Salivary Biomarkers CST4 and miR-223 With Health-Related Factors for Gastric Cancer Detection and Risk Assessment

Supplemental material, sj-jpg-5-onc-10.1177_11795549251375898 for Integrating Salivary Biomarkers CST4 and miR-223 With Health-Related Factors for Gastric Cancer Detection and Risk Assessment by Maryam Koopaie, Saba Manafi, Soheila Manifar, Shima Younespour, Sajad Kolahdooz and Mahdi Karimipour Pareshkooh in Clinical Medicine Insights: Oncology

sj-jpg-6-onc-10.1177_11795549251375898 – Supplemental material for Integrating Salivary Biomarkers CST4 and miR-223 With Health-Related Factors for Gastric Cancer Detection and Risk Assessment

Supplemental material, sj-jpg-6-onc-10.1177_11795549251375898 for Integrating Salivary Biomarkers CST4 and miR-223 With Health-Related Factors for Gastric Cancer Detection and Risk Assessment by Maryam Koopaie, Saba Manafi, Soheila Manifar, Shima Younespour, Sajad Kolahdooz and Mahdi Karimipour Pareshkooh in Clinical Medicine Insights: Oncology