Skip to main content
. 2023 Nov 29;24(23):16931. doi: 10.3390/ijms242316931

Table 2.

List of non-blood circulating protein markers for GC diagnosis reported over the past 10 years.

Biomarker(s) (a) Proteomic Technology (b) Patients Major Findings Ref.
Tumor Characteristics (c) Group (nr) (d)
Ascitic fluid
PGC and POSTN • LC-MS/MS • ELISA • Clinical stage IV GC (85) In total, 299 differentially expressed proteins were quantified, 81 and 218 of which were up- and downregulated, respectively, in the ascitic fluids of GC. PGC and POSTN proteins distinguished malignant ascites from benign ones and were verified by ELISA, being thus potential candidate biomarkers of disease state. [88]
Gastric juice
SNCG
(also from serum)
• ELISA Not reported GC (87), CPL (38), and C (44) SNCG levels were higher in GC vs. CPL or C in both gastric juice and serum. The expression of SNCG in GJ and serum was significantly associated with tumor node metastasis stage, lymph node metastasis, tumor size, and drinking. The detection of SNCG in gastric juice and serum was reported as an ideal method of clinical diagnostic value for the early diagnosis of GC, with high specificity (90.91%) and sensitivity (83.91%) (positive predictive value = 94.81%; negative predictive value = 74.07%; 95% CI: 0.869–0.971; p< 0.0001). [59]
Ela3A, PepA, GastL, Gastricsin, and CystD • iTRAQ labeling • LC-MS/MS • WB • Histology: 30% diffuse, 33% intestinal, 37% unknown/mix • invasion depth: 54% I–II, 46% III–IV GC (70) and benign gastritis (17) An increase in Ela3A together with a decrease in PepA, Gast, Gastricsin, and CystD occurred in gastric fluids of GC patients with high confidence. A three-biomarker panel of CystD + PepA + Ela3A was sufficient for initial GC diagnosis (sensitivity = 95.7%; specificity = 76.5%). [89]
S100A9, GIF, and AAT • 2DE • MS • WB • Clinical stages: discovery cohort→33.3% I (early GC), 33.3% III 33.3% IV (late GC); validation cohort→51% early GC, 49% advanced GC Discovery cohort: GC (9) and gastritis (3); validation cohort: GC (43) and gastritis (17) Out of the 15 differential proteins identified, levels of S100A9, GIF, and AAT correlated with GC status. S100A9 and AAT (AUC = 0.81; p = 0.0013) were promising biomarker pairs for early GC diagnosis. [90]
Saliva
CSTB, TPI1, and DMBT1 • 1DE • 2D-DIGE • TMT • LC-MS/MS • ELISA • Clinical stages: 75% I–II, 25% III–IV Discovery cohort: GC (20) and C (20); validation cohort: GC (20) and C (20) A total of 48 differential proteins were found (p < 0.05) between GC and C, including 7 up- and 41 downregulated proteins. Three proteins were successfully validated (CSTB, TPI1, and DMBT1). These proteins differentiated GC (p < 0.05) and, combined, showed a sensitivity of 85% and a specificity of 80% in GC detection with an accuracy of 0.93. [91]
Urine
SORT1 and VTN • TMT • LC-MS/MS • Grade 2 or grade 3 adenocarcinoma Discovery cohort: GC (5) and C (5); validation cohort: GC (19) and C (12) A total of 246 proteins were differentially expressed in GC cases. Some proteins more abundant in GC vs. C are already known to play crucial roles in GC progression (ephrin A1, pepsinogen A3, sortilin 1, SORT1, and VTN). Others had not previously been linked to GC (shisa family member 5, mucin-like 1, and leukocyte cell-derived chemotaxin 2). The overexpression of SORT1 and VTN in GC urines was confirmed in an independent set of urine samples. [92]
ANXA11, CDC42, NAPA, and SLC25A4 • LC-MS/MS Not detailed Discovery cohort: GC (14) and patients with gastric lesions (109; SG, CAG, IM, or LGIN); validation cohort: GC (18) and patients with gastric lesions (114; SG, CAG, IM, or LGIN) Urinary levels of ANXA11, CDC42, NAPA, and SLC25A4 were positively associated with gastric lesion progression; they may potentially predict the progression of gastric lesions and risk of GC occurrence. [93]
ADAM12 and TFF1 • LC-MS/MS Not detailed discovery cohort: 18 patients; training cohort: 176 patients; validation cohort: 88 patients. A urinary biomarker panel combining TFF1, ADAM12, and H. pylori significantly discriminated early GC vs. C in both training and validation cohorts. [94]
ADAM12 and MMP-9/NGAL complex • Protein array analysis • substrate gel electrophoresis • ELISA • Histology: 23% diffuse, 77% intestinal • clinical stages: 75% I–II, 25% III–IV • invasion depth: 69% T1–T2, 31% T3–T4 GC (35) and C (35) Urinary levels of the MMP-9/NGAL complex and ADAM12 were higher in GC vs. C (p < 0.001). Both the MMP-9/NGAL complex and ADAM12 were significant, independent diagnostic biomarkers for GC by multivariate analysis and distinguished between GC and C samples (AUC = 0.825, p < 0.001) in an ROC analysis. [95]
EL • WB • Histology: 43% intestinal, 57% diffuse • clinical stages: 33% I–II, 67% III–IV • invasion depth: 33% T1–T2, 67% T3–T4 • degree of differentiation: 48% high or moderate, 52% poor or undifferentiated GC (90) and C (57) The EL content decreased by a ~9.9-fold average in GC vs. C (p < 0.0001), achieving a 0.967 AUC value for the ROC curve, demonstrating its high accuracy as a promising diagnostic marker for GC. [96]

Abbreviations: 1-DE, one-dimensional electrophoresis; 2DE, two-dimensional electrophoresis; AAT, α-1-antitrypsin; ADAM12, disintegrin and metalloproteinase domain-containing protein 12; ANXA11, annexin A11; CAG, chronic atrophic gastritis; CDC42, cell division control protein 42 homolog; CPL, gastric precancerous lesions; CSTB, cystatin B; CystD, cystatin D; Da, Dalton; DMBT1, deleted in malignant brain tumors 1 protein; EL, endothelial lipase; Ela3A, elastase 3A; IM, intestinal metaplasia; MMP-9, matrix metalloproteinases-9; GastL, gastric lipase; GIF, gastric intrinsic factor; LGIN, low-grade intraepithelial neoplasia; MALDI-TOF, matrix-assisted laser desorption ionization time-of-flight; MMP9, matrix metallopeptidase 9; MS, mass spectrometry; NAPA, NSF attachment protein α; NGAL, neutrophil gelatinase-associated lipolalin; PepA, pepsin A; PGC, progastriscin; POSTN, periostin; ROC, receiver operating characteristic; S100A9, S100 calcium-binding protein A9; SLC25A4, solute carrier family 25 member 4; SG, superficial gastritis; SNCG, synuclein-gamma; SORT1, sortilin 1; TFF1, trefoil factor 1; TMT, tandem mass tags; TPI1, triosephosphate isomerase; VTN, vitronectin. (a) Protein name abbreviation is followed by the UniProtKB ID; (b) proteomics techniques used to analyze proteins are reported; (c) among the clinical characteristics detailed by authors, only those about tumor clinical stages (from I to V), histological types (intestinal, diffuse, mixed), gastric anatomic subsites (proximal, distal), pathological stages with the TNM system (pTNM stages, from T1 to T4), and presence of metastasis (M0, M1) are, if present, reported; (d) nr, number of individuals per group.