Salivary metabolomic biomarkers for esophageal and gastric cancers by liquid chromatography–mass spectrometry

Kosuke Nakane; Koichi Yagi; Sho Yajima; Sachiyo Nomura; Masahiro Sugimoto; Yasuyuki Seto

doi:10.1111/cas.16256

. 2024 Jul 14;115(9):3089–3098. doi: 10.1111/cas.16256

Salivary metabolomic biomarkers for esophageal and gastric cancers by liquid chromatography–mass spectrometry

Kosuke Nakane ¹, Koichi Yagi ¹, Sho Yajima ¹, Sachiyo Nomura ¹, Masahiro Sugimoto ^2,³, Yasuyuki Seto ^1,^✉

PMCID: PMC11463073 PMID: 39004809

Abstract

Early detection of esophageal and gastric cancers is essential for patients' prognosis; however, optimal noninvasive screening tests are currently not available. Saliva is a biofluid that is readily available, allowing for frequent screening tests. Thus, we explored salivary diagnostic biomarkers for esophageal and gastric cancers using metabolomic analyses. Saliva samples were collected from patients with esophageal (n = 50) and gastric cancer (n = 63), and patients without cancer as controls (n = 20). Salivary metabolites were analyzed by liquid chromatography–mass spectrometry to identify salivary biomarkers. We also examined the metabolic profiles of gastric cancer tissues and compared them with the salivary biomarkers. The sensitivity of the diagnostic models based on salivary biomarkers was assessed by comparing it with that of serum tumor markers. Additionally, using postoperative saliva samples collected from patients with gastric cancer, we analyzed the changes in the biomarkers' concentrations before and after surgery. Cytosine was detected as a salivary biomarker for gastric cancer, and cytosine, 2‐oxoglutarate, and arginine were detected as salivary biomarkers for esophageal cancer. Cytidine, a cytosine nucleotide, showed decreased concentrations in gastric cancer tissues. The sensitivity of the diagnostic models for esophageal and gastric cancers was 66.0% and 47.6%, respectively, while that of serum tumor markers was 40%. Salivary cytosine concentration increased significantly postoperatively relative to the preoperative value. In summary, we identified salivary biomarkers for esophageal and gastric cancers, which showed diagnostic sensitivity at least comparable to that of serum tumor markers. Salivary metabolomic tests could be promising screening tests for these types of cancer.

Keywords: esophageal cancer, gastric cancer, LC‐MS, metabolome, saliva

We explored salivary diagnostic biomarkers for esophageal and gastric cancers using liquid chromatography–mass spectrometry. Cytosine was detected as a salivary biomarker for gastric cancer, and cytosine, 2‐oxoglutarate, and arginine were detected as salivary biomarkers for esophageal cancer. The diagnostic ability of the models constructed with the detected biomarkers was at least equivalent to that of serum tumor markers.

graphic file with name CAS-115-3089-g003.jpg

Abbreviations

AUC: area under the ROC curve
C group: control group
CA 19‐9: cancer antigen 19‐9
CEA: carcinoembryonic antigen
CI: confidence interval
EC group: esophageal cancer group
GC group: gastric cancer group
LC‐MS: liquid chromatography–mass spectrometry
ROC: receiver operating characteristic
SCC: squamous cell carcinoma antigen
TCA: tricarboxylic acid

1. INTRODUCTION

Esophageal and gastric cancers are common upper gastrointestinal tract cancers worldwide, with high morbidity and mortality rates. ¹ The mortality rate in these cancers increases with advancement of the tumor stage ² , ³ ; thus, early diagnosis is essential for improving patients' prognosis. Accordingly, establishing screening methods allowing for early detection is an urgent challenge. However, at present, optimal screening tests for early diagnosis of esophageal and gastric cancers are not available. Commonly used diagnostic methods include radiography and esophagogastroduodenoscopy, but their effectiveness as screening tests is controversial because of their invasiveness and cost‐effectiveness. ⁴ , ⁵ , ⁶ , ⁷ Serum tumor markers for esophageal and gastric cancers are not sufficiently effective because of their low sensitivity. ⁸ , ⁹ Currently, no tests using body fluid obtained with minimally invasive methods enable highly accurate screening.

Salivary tests have recently attracted attention as screening tests. The saliva has various functions and is described as blood filtrate reflecting the physiological conditions of the body. ¹⁰ Hence, salivary tests can be used to monitor patients' clinical condition and predict the development of systemic disorders. Additionally, compared with blood tests, salivary tests have the advantages of high cost‐effectiveness, safety, availability, and noninvasiveness, which can facilitate their use as novel screening methods.

Salivary metabolomics has emerged as an analytical method for diagnosing systemic metabolic disorders, including cancer, with biomarkers for oral, breast, pancreatic, lung, and colorectal cancers reported to date. ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ Metabolomics has higher quantitativeness and reproducibility, and pretreatment for the analysis is more accessible than for conventional analytical methods, such as transcriptomics or proteomics. Among several analytical techniques in metabolomics, LC‐MS can simultaneously identify and quantify a wide range of metabolites. Several types of pretreatments and LC enable comprehensive analysis of various metabolic pathways, such as energy metabolism, mainly composed of water‐soluble metabolites, as well as lipid metabolism. ¹⁷ Nonetheless, few salivary biomarkers for esophageal and gastric cancers have been reported to date, particularly ones detected by LC‐MS. ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²²

Therefore, the present study aimed to identify salivary biomarkers for esophageal and gastric cancers using LC‐MS and assess their diagnostic ability.

2. MATERIALS AND METHODS

2.1. Study design and population

This was a prospective case–control study that included patients with pathologically confirmed diagnosis of esophageal or gastric cancer between October 2021 and April 2023 at the University of Tokyo Hospital. Esophagogastric junction adenocarcinoma and Barrett esophageal adenocarcinoma were included in gastric cancer. Patients with other concurrent cancers were excluded. Additionally, patients diagnosed with hernia (inguinal, femoral, incisional, and lumbar) who did not have cancer were recruited as controls.

2.2. Sample collection

We collected three types of samples: pretreatment saliva, postoperative saliva, and surgically resected tissues. Pretreatment saliva samples were collected from all patients with esophageal and gastric cancers and all controls before treatment. Postoperative saliva samples were collected from patients who underwent curative surgery for gastric cancer approximately 1 month after surgery. We also collected cancer and normal tissues from surgically resected specimens of patients with gastric cancer who underwent curative surgery without preoperative endoscopic resection, chemotherapy, or radiotherapy within 1 month from saliva collection.

For saliva sample collection, patients were prohibited from eating after 9:00 pm on the previous day and smoking or brushing their teeth within 1 h of collection. They were required to rinse their mouths before collection in the morning and, 5 min later, drip 0.5 mL of unstimulated saliva into polypropylene tubes through polypropylene straws. ¹⁶ For tissue sample collection, cancer and normal tissues, approximately 5 mm thick, were rapidly collected from surgically resected specimens. Cancer tissues were collected from near the center of the cancer to the extent not hindering the pathological diagnosis. Saliva and tissue samples were iced after collection and immediately stored at −80°C.

2.3. Pretreatment and analytical processes

Pretreatment procedures for saliva and tissue samples have been described previously. ²³ , ²⁴ Briefly, each saliva sample was mixed with methanol containing internal standards suitable for cation or anion analysis. After centrifugation, vortexing, and mixing with Milli‐Q water (Milli‐Q® Advantage A10;Merck & Co., Inc.), the solutions were injected into the LC‐MS device. Tissue samples were mixed with methanol, homogenized, and pretreated in almost the same way as saliva samples, and injected into the LC‐MS device. Metabolomic analyses were undertaken using an Agilent 6230B TOF LC/MS system (Agilent Technologies). ²³ , ²⁵ Raw data were processed by the MassHunter Workstation Software Quantitative Analysis (version B.08.00; Agilent Technologies). The concentrations of metabolites were calculated from the ratio of the peak areas of the metabolites to those of the corresponding internal standards.

2.4. Metabolic profiles in saliva and tissues

2.4.1. Salivary biomarkers for gastric and esophageal cancers, and metabolic profiles in gastric cancer tissues

We analyzed the pretreatment saliva samples of patients in the GC, EC, and C groups. Metabolites with significantly different concentrations between the cancer and control groups were identified as the respective cancer biomarkers.

We also compared the metabolic profiles in gastric cancer and normal tissues, and identified metabolites with concentrations that significantly differed between cancer and normal tissues. The metabolic profiles in cancer tissues were compared with the salivary biomarkers.

2.4.2. Salivary biomarkers for differentiated‐ and undifferentiated‐type gastric cancer, and relationship between salivary biomarkers and proliferating ability of gastric cancer

As different biomarkers might be detected in differentiated‐ and undifferentiated‐type gastric cancer, we compared the pretreatment saliva samples of patients with each type with those of patients in the C group. Differentiated‐ and undifferentiated‐type gastric cancer were defined according to the 2021 Japanese Gastric Cancer Treatment Guidelines (6th edition). ²⁶ The esophageal and gastric cancer stage was defined according to the Japanese Classification for Esophageal Cancer (12th edition) ²⁷ and the Japanese Classification for Gastric Cancer (15th edition). ²⁸

We also examined the relationship between salivary biomarkers for differentiated‐ and undifferentiated‐type gastric cancer and the Ki‐67 index ²⁹ because the difference in salivary biomarkers for each type might result from the proliferating capacity of gastric cancer. We used pretreatment saliva samples and formalin‐fixed gastric cancer tissues of patients who underwent curative surgery without preoperative endoscopic resection, chemotherapy, or radiotherapy. We excluded patients with a histopathological diagnosis of unclassifiable common type and special type gastric cancer. Considering the influence of the size of cancer tissues on the salivary biomarker concentrations, we undertook additional analyses using the product of the Ki‐67 index and the area of gastric cancer tissues.

All specimens were fixed in 10% formalin and embedded in paraffin wax. Paraffin blocks were sliced into 4 μm thick sections, which were dyed using H&E and Ki‐67 staining. The sections were deparaffinized with a histopathological cleaning agent (Histo‐Clear II HS‐202; National Diagnostics) three times for 5 min and with 100% ethanol (#057‐00451; FUJIFILM Wako Pure Chemical Corporation) three times for 3 min, kept in an antigen retrieval medicine (Immunosaver [#333, Nisshin EM Co., Ltd] 2 mL/Milli‐Q water 398 mL) for 20 min at 98°C, and immersed in 3% H₂O₂ (#081‐04215, FUJIFILM Wako Pure Chemical Corporation)/methanol (#137‐01823, FUJIFILM Wako Pure Chemical Corporation) solution for 10 min at room temperature to block endogenous peroxidase. Subsequently, sections were incubated with a normal blocking serum (VECTASTAIN Elite ABC Rabbit IgG Kit, PK‐6101; Vector Laboratories, Inc.) for 30 min, and then overnight at 4°C with a primary Ab (Ki‐67 Recombinant Rabbit Monoclonal Antibody [SP6], #MA5‐14520, dilution 1:300; Thermo Fisher Scientific). Next, sections were mounted with a biotinylated secondary Ab (VECTASTAIN Elite ABC Rabbit IgG Kit) for 60 min, and a peroxidase‐labeled avidin–biotin complex (VECTASTAIN Elite ABC Rabbit IgG Kit) for 30 min. The antigen–Ab complexes were visualized using a DAB Peroxidase Substrate Kit (ImmPACT DAB, SK‐4105; Vector Laboratories, Inc.). Finally, the sections were immersed in Mayer's hematoxylin (Muto Pure Chemicals Co, Ltd) for counterstaining the cell nuclei, in 100% ethanol three times for 5 min, and in Histo‐Clear II three times for 10 min. The sections were mounted using mounting media (Mount‐Quick; DAI) and observed under a microscope (BX 51; Olympus). Nuclei staining for Ki‐67 with any intensity was regarded as positive. We observed hotspots where stained cells were most frequently seen at 400× magnification, counted more than 1000 tumor cells, and calculated the ratio of stained cells to total tumor cells as the Ki‐67 index.

2.4.3. Diagnostic models for esophageal and gastric cancers and their diagnostic performance

Diagnostic models for esophageal and gastric cancers were developed with the detected biomarkers using multiple logistic regression analysis. The discrimination ability of the diagnostic models was evaluated by ROC curve analysis and the AUC. We also calculated the sensitivity and specificity of the models at an optimal cut‐off, determined by the maximum of Youden's index (sensitivity + specificity − 1), and compared the sensitivity of the models with that of serum tumor markers (esophageal cancer: SCC, CYFRA, p53; gastric cancer: CEA, CA19‐9). The cut‐off values for SCC, CYFRA, p53, CEA, and CA19‐9 were 2.5 ng/mL, 3.4 ng/mL, 1.30 U/mL, 4.9 ng/mL, and 36 U/mL, respectively.

To evaluate the generalization performance of the diagnostic models, we undertook k‐fold cross‐validation, with 100 trials performed for each random number for k (5, 10, and 20).

2.4.4. Changes in salivary biomarker concentrations before and after surgery

To examine the changes in salivary biomarker concentrations before and after surgery, we compared the pretreatment and postoperative saliva metabolic profiles in patients who underwent curative gastrectomy for gastric cancer.

2.4.5. Changes in salivary biomarker concentrations before and after chemotherapy

To evaluate the effect of chemotherapy on salivary biomarkers, we additionally collected pretreatment and postchemotherapy saliva from newly recruited patients with clinical stage IV gastric cancer and analyzed the changes in salivary biomarker concentrations through chemotherapy. Postchemotherapy samples were collected immediately before the second cycle of chemotherapy. Salivary biomarker concentrations were calculated by averaging the values obtained by three consecutive measurements for each sample because of the small number of samples. We also assessed the effectiveness of chemotherapy for gastric cancer immediately before the fourth cycle of chemotherapy using computed tomography and esophagogastroduodenoscopy, according to the RECIST 1.1. ³⁰

2.5. Statistical analysis

Fisher's exact test was used to compare categorical variables, and Student's t‐test and Mann–Whitney's U‐test were used to compare continuous variables between two groups. Wilcoxon's matched‐pairs signed‐rank test was used for a paired sample test. All statistical analyses were undertaken using JMP Pro (version 17.0.0; SAS Institute), MetaboAnlayst (version 5.0; https://www.metaboanalyst.ca/), ³¹ and Weka (version 3.6.14; The University of Waikato; https://www.cs.waikato.ac.nz/ml/weka/), with two‐sided p values less than 0.05 indicating statistical significance.

3. RESULTS

3.1. Patient characteristics

A total of 50 and 63 patients with esophageal and gastric cancer, respectively, and 20 patients without cancer were recruited. The patients' clinicopathologic characteristics are summarized in Table 1. All patients with esophageal cancer had squamous cell carcinoma, and one patient with gastric squamous cell carcinoma was included in the EC group from a histopathological viewpoint. Five patients with Barrett's esophageal adenocarcinoma were included in the GC group: four with differentiated and one with undifferentiated adenocarcinoma. The clinical stage at the time of saliva collection was used. No significant differences were observed in the patient characteristics among the groups.

TABLE 1.

Clinicopathologic characteristics of the total study population.

	Control	Esophageal cancer	Gastric cancer
	n = 20	n = 50	n = 63
Age (years); median (25% percentile, 75% percentile)	72.0 (66.0, 81.5)	72.5 (66.8, 75.5)	69.0 (60.0, 75.0)
Sex
Male	18 (90.0)	41 (82.0)	44 (69.8)
Female	2 (10.0)	9 (18.0)	19 (30.2)
Pathophysiology
Differentiated type	–	–	29 (46.0)
Undifferentiated type	–	–	26 (41.3)
Unclassifiable common type	–	–	7 (11.1)
Special type	–	–	1 (1.6)
Clinical stage
0/I	–	19 (38.0)	32 (50.8)
II	–	7 (14.0)	11 (17.5)
III	–	18 (36.0)	7 (11.1)
IV	–	6 (12.0)	13 (20.6)
Helicobacter pylori Ab
Positive	–	–	25 (39.7)
Negative	–	–	31 (49.2)
Unknown	–	–	7 (11.1)

Open in a new tab

Note: Data are shown as n (%), unless otherwise indicated.

3.2. Salivary biomarkers for gastric cancer and metabolic profiles in gastric cancer tissues

The differences in the metabolite concentrations between the GC and C groups are shown in Figure 1. Cytosine had a significantly lower concentration in the GC group than in the C group (p = 0.0389). Box plots and the ROC curve of cytosine are also shown in Figure 1 (AUC, 0.658; 95% CI, 0.510–0.798).

Volcano plots showing the difference in each salivary metabolite concentration (Conc.) between the gastric cancer (GC) and control (C) groups. The x‐axis is the log₂‐transformed value of fold changes of metabolites, and metabolites with x > 0 means that the mean concentration of the metabolites is higher in the GC group. The y‐axis is the −log₁₀‐transformed value of the p value calculated by the Mann–Whitney' U‐test, and y > 1.3 means p < 0.05. Box plots and the receiver operating characteristic (ROC) curve of cytosine are also shown. AUC, area under the ROC curve.

The results of the metabolomic analyses in gastric cancer and normal tissues collected from seven patients are shown in Figure 2 and Table S1. Compared with those in normal tissues, in gastric cancer tissues, the concentrations of cytosine, uracil, polyamines including N ¹,N ¹²‐diacetylspermine and N ¹‐acetylspermine, phenylalanine, and choline were significantly higher, and that of cytidine, a nucleotide of cytosine, was significantly lower.

Volcano plots showing the difference in each metabolite concentration between cancer and normal tissues in gastric cancer (n = 7). The x‐axis is the log₂‐transformed value of fold changes of metabolites, and metabolites with x > 0 means that the mean concentration of the metabolites is higher in the cancer tissues. The y‐axis is the −log₁₀‐transformed value of the p value calculated by the Mann–Whitney U‐test, and y > 1.3 means p < 0.05.

3.3. Salivary biomarkers for esophageal cancer

The difference in salivary metabolite concentrations between the EC and C groups is shown in Figure 3. The concentration of 2‐oxoglutarate was significantly higher (p = 0.0258), and those of arginine and cytosine were significantly lower (p = 0.0200 and 0.0496, respectively) in the EC group than in the C group. Box plots and ROC curves of the metabolites are also shown in Figure 3.

Volcano plots showing the difference in each salivary metabolite concentration (Conc.) between the esophageal cancer (EC) and control (C) groups. The x‐axis is the log₂‐transformed value of fold changes of metabolites, and metabolites with x > 0 means that the mean concentration of the metabolites is higher in the EC group. The y‐axis is the −log₁₀‐transformed value of the p value calculated by the Mann–Whitney U‐test, and y > 1.3 means p < 0.05. Box plots and receiver operating characteristic (ROC) curves of the three metabolites with significantly different concentrations between the groups are also shown. AUC, area under the ROC curve.

3.4. Salivary biomarkers for differentiated‐ and undifferentiated‐type gastric cancer, and relationship between salivary biomarkers and Ki‐67 index

The results of the subgroup analyses for differences in metabolite concentrations between differentiated‐ (n = 29) and undifferentiated‐type gastric cancer (n = 26) and the C group for each metabolite are shown in Figures S1 and S2. No salivary metabolites showed significantly different concentrations between differentiated‐type gastric cancer and the C group, with a tendency of low concentration of cytosine. In contrast, cytosine showed a significantly lower concentration in undifferentiated‐type gastric cancer than in the C group.

The analysis of the relationship between salivary cytosine concentrations and the proliferative ability of gastric cancer tissues (Ki‐67 index) included samples from 38 patients. Their clinicopathologic characteristics are shown in Table S2. We found no correlation between salivary cytosine concentrations and the Ki‐67 index of gastric cancer tissues (R = 0.018; Figure 4). This finding was also confirmed in the subgroup analyses of differentiated‐ and undifferentiated‐type gastric cancer, as well as in the additional analyses using the product of the Ki‐67 index and the area of gastric cancer tissues (Figures S3 and S4).

Relationship between salivary cytosine concentrations and the Ki‐67 index of gastric cancer tissues.

3.5. Diagnostic models for esophageal and gastric cancers

For gastric cancer, the sensitivity and specificity of the diagnostic model with cytosine were 47.6% and 80.0% at an optimal cut‐off, respectively (Figure 1; AUC, 0.658; 95% CI, 0.510–0.798). The sensitivity of CEA and CA19‐9 in the GC group was 38.7% and 22.6%, respectively.

For esophageal cancer, the odds ratios of the detected candidate biomarkers and the ROC curve of the diagnostic model using the biomarkers are shown in Figure 5 (AUC, 0.788; 95% CI, 0.677–0.899). The sensitivity and specificity of the model were 66.0% and 90.0% at an optimal cut‐off, respectively. The sensitivity of SCC, CYFRA, and p53 in the EC group was 26.0%, 18.0%, and 40.0%, respectively, which showed the higher sensitivity of the developed diagnostic model. The k‐fold cross‐validation (k = 5, 10, and 20) was performed for the diagnostic model for esophageal cancer. The average AUCs of the 100 trials were 0.691, 0.695, and 0.691 for k = 5, 10, and 20, respectively.

Odds ratios of the detected biomarkers and the receiver operating characteristic (ROC) curve of the diagnostic model for esophageal cancer. Area under the ROC curve, 0.788 (95% confidence interval, 0.677–0.899).

3.6. Changes in salivary cytosine concentration before and after surgery

The changes in salivary cytosine concentration in eight patients with gastric cancer before and after surgery are shown in Figure 6. The concentration of cytosine, which was significantly lower in the GC group than in the C group before surgery, increased significantly after surgery.

Changes in the salivary cytosine concentration (Conc.) in patients with gastric cancer before and after surgery (n = 8). **p < 0.01 (Wilcoxon's matched‐pairs signed‐rank test).

3.7. Changes in salivary cytosine concentration before and after chemotherapy

A 64‐year‐old man and a 51‐year‐old woman with clinical stage IV undifferentiated‐type gastric cancer were newly recruited. Both patients were treated with S‐1, oxaliplatin (130 mg/m²), and nivolumab (360 mg). The dosage of S‐1 was determined according to the body surface area. The tumor response to chemotherapy according to RECIST 1.1 was partial response in both patients.

The salivary cytosine concentrations in these patients before and after chemotherapy are shown in Figure 7. While the salivary cytosine concentration remained unchanged in one of the patients, it decreased after chemotherapy in the other patient, although the cancer size was reduced.

Changes in the salivary cytosine concentration (Conc.) in patients with clinical stage IV gastric cancer before and after chemotherapy (n = 2).

4. DISCUSSION

In the present study, by analyzing saliva samples of patients with esophageal and gastric cancers using LC‐MS, we identified cytosine as a candidate salivary biomarker for gastric cancer, and cytosine, 2‐oxoglutarate, and arginine as candidate salivary biomarkers for esophageal cancer. The diagnostic ability of the models constructed with the detected biomarkers was at least equivalent to that of serum tumor markers.

Cytosine is a base made up of nucleic acid. Salivary cytosine was found in a lower concentration in esophageal and gastric cancers compared with that in controls. As nucleotide biosynthesis is often promoted in cancer cells, ³² cytosine might be used to synthesize nucleotides, resulting in reduced salivary concentration. Cancer cells depend on de novo biosynthesis of nucleotides to sustain nucleotides for RNA production and DNA replication. De novo biosynthesis of nucleotides is a complex metabolic pathway involving many metabolic pathways, such as the pentose phosphate pathway, one‐carbon metabolism, and TCA cycle. ³³ These characteristics might explain why only cytosine concentration among bases of nucleotides changed significantly in the saliva. Nonetheless, further studies are required, particularly from a biochemical viewpoint. A previous metabolomic study suggested that salivary cytosine levels decreased in patients with lung cancer, although not significantly. ¹⁵ The present study also detected cytosine as a candidate salivary biomarker for both esophageal and gastric cancers despite the pathohistological difference. Thus, cytosine is expected to be a salivary biomarker for other cancers as well.

Cytidine, a nucleotide of cytosine with a ribose ring, was found in a significantly lower concentration in gastric cancer than in normal tissues, while cytosine and uracil were found in significantly higher concentrations. These results indicate that cytosine is chemoattracted to cancer tissues for cell proliferation and diluted in the saliva, and that synthesizing cytidine may be a rate‐limiting stage in cancer. This speculation might also explain the lack of correlation between salivary cytosine concentration and the cancer cell proliferation rate. Chemoattractive factors for cytosine may exist in many types of cancer.

The concentrations of N ¹,N ¹²‐diacetylspermine and N ¹‐acetylspermine were significantly higher in gastric cancer than in normal tissues. N ¹,N ¹²‐diacetylspermine and N ¹‐acetylspermine are acetylated polyamines. Ornithine is converted to putrescine by ornithine decarboxylase in the urea cycle, and then spermidine, spermine, and other polyamines are synthesized from putrescine. Polyamines are acetylated by spermidine/spermine‐N ¹‐acetyltransferase and excreted from cells. ³⁴ Activation of the MYC oncogene was shown to enhance ornithine decarboxylase. Consequently, polyamines have attracted attention as diagnostic biomarkers for cancer because of their relationship to the oncogene MYC. Indeed, N ¹,N ¹²‐diacetylspermine concentrations have been reported to increase in the saliva, serum, and urine of patients with various cancers. ¹⁶ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ N ¹‐acetylspermine concentration was also increased in the saliva of patients with breast, colon, and pancreatic cancer. ¹³ , ¹⁴ , ¹⁶ Choline, a metabolite in the one‐carbon metabolism, and phenylalanine concentrations have also been reported to increase in the saliva of patients with various cancers. ¹¹ , ¹² , ¹³ However, these metabolites were not identified as salivary biomarkers, which might suggest that the salivary metabolome is different from the metabolome in gastric cancer tissues.

The subgroup analyses for differentiated‐ and undifferentiated‐type gastric cancer suggested that cytosine was a candidate biomarker for undifferentiated‐type gastric cancer, although its concentration in differentiated‐type gastric cancer was also relatively lower than in normal tissues. In the correlation analysis, no correlation was observed between salivary cytosine concentrations and the Ki‐67 index regardless of the cancer differentiation status or cancer tissue volume. Thus, the difference in salivary cytosine concentrations between differentiated‐ and undifferentiated‐type gastric cancer might be due to mechanisms other than tumor cell proliferation, such as the rate‐limiting stage in cytidine synthesis mentioned above. Another possible reason is that the Ki‐67 index, calculated by tumor cells of local hotspots in cancer tissues, might not correctly represent the proliferating capacity of the entire tumor. Finally, these results might also be due to the small sample size in this subgroup; hence, further studies are needed to elucidate this relationship.

To date, 2‐oxoglutarate, which is an intermediate metabolite in the TCA cycle, has not been reported as a biomarker for any cancer. Oxidative phosphorylation in the mitochondria, including the TCA cycle, is promoted in cancer cells, which might have led to the increase in 2‐oxoglutarate concentration in the present study. ³³

Arginine is an essential mediator of immune defense for tumor cells, and arginase activation is thought to be a mechanism of tumor‐induced immunosuppression in tumor cells. ⁴¹ , ⁴² Additionally, arginine is converted to ornithine in the urea cycle and used to synthesize polyamines. Reportedly, plasma arginine concentrations decreased in breast, colon, and pancreatic cancer, and salivary arginase was activated in breast cancer, ⁴² , ⁴³ which does not contradict our result that salivary arginine concentrations decreased in patients with esophageal cancer.

The sensitivity of the diagnostic models for esophageal and gastric cancers in this study was 66.0% and 47.6%, respectively, and the sensitivity of serum tumor markers was almost the same as that previously reported. ⁸ , ⁹ Furthermore, in the current study, the AUC for gastric cancer was 0.658 and that for esophageal cancer was 0.788. These AUC values are not lower than those of serum tumor markers reported in prior studies for CEA, CA19‐9, SCC, CYFRA, and p53 (0.689–0.767, 0.566–0.697, 0.665–0.709, 0.626–0.708, and 0.74, respectively). ⁴⁴ , ⁴⁵ , ⁴⁶ , ⁴⁷ , ⁴⁸ , ⁴⁹ , ⁵⁰ Therefore, the diagnostic models in this study showed diagnostic ability comparable to that of serum tumor markers, with the model for esophageal cancer deemed a candidate to be adopted as a screening test. The k‐fold cross‐validation also showed a higher generalizability of the diagnostic model for esophageal cancer.

In the evaluation of the changes in salivary cytosine concentrations before and after surgery, cytosine concentrations increased significantly after surgery, with a tendency to reach the average value of controls. In contrast, no obvious relationship was observed between the tumor response to chemotherapy and the changes in salivary cytosine concentration before and after chemotherapy. One of the main chemotherapeutic agents for gastrointestinal cancer is 5‐fluorouracil, an antimetabolite that reduces the synthesis of cytosine, making it difficult to see the skewed distribution of cytosine in the body after chemotherapy.

The present study has several limitations. First, we could not conduct external validation of the diagnostic models. Further study is needed with larger saliva sample collections of patients with cancer and controls to verify the diagnostic ability of the models. Second, this was a case–control study; thus, confounding factors cannot be ruled out. Third, we could not examine the AUCs of serum tumor markers and statistically compare the diagnostic ability of the developed models with that of serum tumor markers. However, we evaluated the sensitivity of the models and tumor markers, which is deemed more important than specificity in screening tests. In addition, we set the sensitivity and specificity of the models just at the optimal cut‐off using the Youden index to reduce the arbitrariness. Fourth, we could not analyze the metabolic profiles in esophageal cancer tissues because of the shortage of esophageal cancer tissues without chemotherapy. Finally, the prevalence of esophageal and gastric cancer in the present study does not reflect their real‐world prevalence. Thus, the suitability of the developed diagnostic models for screening tests in the general population needs to be verified.

In conclusion, we identified salivary metabolomic biomarkers for esophageal and gastric cancers, which indicated diagnostic ability at least comparable to that of existing tumor markers. Therefore, salivary metabolomics could provide potential screening tests for cancer.

AUTHOR CONTRIBUTIONS

Kosuke Nakane: Formal analysis; investigation; writing – original draft. Koichi Yagi: Supervision; writing – review and editing. Sho Yajima: Writing – review and editing. Sachiyo Nomura: Resources; writing – review and editing. Masahiro Sugimoto: Conceptualization; formal analysis; methodology; resources; software; writing – original draft. Yasuyuki Seto: Conceptualization; project administration; writing – review and editing.

CONFLICT OF INTEREST STATEMENT

Masahiro Sugimoto is a board member of SalivaTech Co. Ltd., and serves as a consultant to Human Metabolome Technologies Inc. The other authors have no conflicts of interest to declare. All authors had full access to all the data in the study and take responsibility for the decision to submit the manuscript for publication.

ETHICS STATEMENTS

Approval of the research protocol by an Institutional Review Board: The study conformed to the provisions of the Declaration of Helsinki and was approved by the Ethical Review Board of the University of Tokyo (approval No. 2021205NI).

Informed consent: Written informed consent was obtained from all patients before sample collection.

Registry and the Registration No. of the study/trial: N/A.

Animal studies: N/A.

Supporting information

Figure S1.

CAS-115-3089-s002.tif^{(838.4KB, tif)}

Figure S2.

CAS-115-3089-s005.tif^{(862.5KB, tif)}

Figure S3.

CAS-115-3089-s004.tif^{(860.7KB, tif)}

Figure S4.

CAS-115-3089-s003.tif^{(892.4KB, tif)}

Tables S1–S2.

CAS-115-3089-s001.docx^{(23.3KB, docx)}

ACKNOWLEDGMENTS

We thank the staff of the University of Tokyo for collecting samples. We would also like to thank Editage for English language editing.

Nakane K, Yagi K, Yajima S, Nomura S, Sugimoto M, Seto Y. Salivary metabolomic biomarkers for esophageal and gastric cancers by liquid chromatography–mass spectrometry. Cancer Sci. 2024;115:3089‐3098. doi: 10.1111/cas.16256

REFERENCES

1. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68(6):394‐424. [DOI] [PubMed] [Google Scholar]
2. Watanabe M, Toh Y, Ishihara R, et al. Comprehensive registry of esophageal cancer in Japan, 2014. Esophagus. 2022;19(1):1‐26. [DOI] [PMC free article] [PubMed] [Google Scholar]
3. Suzuki S, Takahashi A, Ishikawa T, et al. Surgically treated gastric cancer in Japan: 2011 annual report of the national clinical database gastric cancer registry. Gastric Cancer. 2021;24(3):545‐566. [DOI] [PubMed] [Google Scholar]
4. Lee KJ, Inoue M, Otani T, et al. Gastric cancer screening and subsequent risk of gastric cancer: a large‐scale population‐based cohort study, with a 13‐year follow‐up in Japan. Int J Cancer. 2006;118(9):2315‐2321. [DOI] [PubMed] [Google Scholar]
5. Matsumoto S, Yamasaki K, Tsuji K, Shirahama S. Results of mass endoscopic examination for gastric cancer in Kamigoto Hospital, Nagasaki Prefecture. World J Gastroenterol. 2007;13(32):4316‐4320. [DOI] [PMC free article] [PubMed] [Google Scholar]
6. Gastric cancer screening guidelines 2014 based on the effectiveness evaluation (in Japanese). http://canscreen.ncc.go.jp/guideline/iganguide2014_150421.pdf. Accessed December 18, 2023.
7. Arnal MJD, Arenas ÁF, Arbeloa ÁL. Esophageal cancer: risk factors, screening and endoscopic treatment in Western and Eastern countries. World J Gastroenterol. 2015;21(26):7933‐7943. [DOI] [PMC free article] [PubMed] [Google Scholar]
8. Yajima S, Suzuki T, Oshima Y, et al. New assay system elecsys anti‐p53 to detect serum anti‐p53 antibodies in esophageal cancer patients and colorectal cancer patients: multi‐institutional study. Ann Surg Oncol. 2021;28(7):4007‐4015. [DOI] [PubMed] [Google Scholar]
9. Shimada H, Noie T, Ohashi M, Oba K, Takahashi Y. Clinical significance of serum tumor markers for gastric cancer: a systematic review of literature by the Task Force of the Japanese Gastric Cancer Association. Gastric Cancer. 2014;17(1):26‐33. [DOI] [PubMed] [Google Scholar]
10. Hyvärinen E, Savolainen M, Mikkonen JJW, Kullaa AM. Salivary metabolomics for diagnosis and monitoring diseases: challenges and possibilities. Metabolomics. 2021;11(9):587. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Sugimoto M, Wong DT, Hirayama A, Soga T, Tomita M. Capillary electrophoresis mass spectrometry‐based saliva metabolomics identified oral, breast and pancreatic cancer‐specific profiles. Metabolomics. 2010;6(1):78‐95. [DOI] [PMC free article] [PubMed] [Google Scholar]
12. Ishikawa S, Sugimoto M, Kitabatake K, et al. Identification of salivary metabolomic biomarkers for oral cancer screening. Sci Rep. 2016;6:31520. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Murata T, Yanagisawa T, Kurihara T, et al. Salivary metabolomics with alternative decision tree‐based machine learning methods for breast cancer discrimination. Breast Cancer Res Treat. 2019;177(3):591‐601. [DOI] [PubMed] [Google Scholar]
14. Asai Y, Itoi T, Sugimoto M, et al. Elevated polyamines in saliva of pancreatic cancer. Cancers. 2018;10(2):43. [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Takamori S, Ishikawa S, Suzuki J, et al. Differential diagnosis of lung cancer and benign lung lesion using salivary metabolites: a preliminary study. Thorac Cancer. 2021;13:460‐465. doi: 10.1111/1759-7714.14282 [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Kuwabara H, Katsumata K, Iwabuchi A, et al. Salivary metabolomics with machine learning for colorectal cancer detection. Cancer Sci. 2022;113(9):3234‐3243. [DOI] [PMC free article] [PubMed] [Google Scholar]
17. Sugimoto M. Salivary metabolomics for cancer detection. Expert Rev Proteomics. 2020;17(9):639‐648. [DOI] [PubMed] [Google Scholar]
18. Wu ZZ, Wang JG, Zhang XL. Diagnostic model of saliva protein finger print analysis of patients with gastric cancer. World J Gastroenterol. 2009;15(7):865‐870. [DOI] [PMC free article] [PubMed] [Google Scholar]
19. Xiao H, Zhang Y, Kim Y, et al. Differential proteomic analysis of human saliva using tandem mass tags quantification for gastric cancer detection. Sci Rep. 2016;6:22165. doi: 10.1038/srep22165 [DOI] [PMC free article] [PubMed] [Google Scholar]
20. Bel'skaya LV, Sarf EA, Shalygin SP, Postnova TV, Kosenok VK. Identification of salivary volatile organic compounds as potential markers of stomach and colorectal cancer: a pilot study. J Oral Biosci. 2020;62(2):212‐221. [DOI] [PubMed] [Google Scholar]
21. Hirayama A, Kami K, Sugimoto M, et al. Quantitative metabolome profiling of colon and stomach cancer microenvironment by capillary electrophoresis time‐of‐flight mass spectrometry. Cancer Res. 2009;69(11):4918‐4925. [DOI] [PubMed] [Google Scholar]
22. Du J, Zhang L. Analysis of salivary microRNA expression profiles and identification of novel biomarkers in esophageal cancer. Oncol Lett. 2017;14(2):1387‐1394. [DOI] [PMC free article] [PubMed] [Google Scholar]
23. Saeki Y, Takenouchi A, Otani E, et al. Long‐term mastication changed salivary metabolomic profiles. Metabolites. 2022;12(7):660. [DOI] [PMC free article] [PubMed] [Google Scholar]
24. Shimizu H, Usui Y, Wakita R, et al. Differential tissue metabolic signatures in IgG4‐related ophthalmic disease and orbital mucosa‐associated lymphoid tissue lymphoma. Invest Ophthalmol Vis Sci. 2021;62(1):15. [DOI] [PMC free article] [PubMed] [Google Scholar]
25. Shimizu H, Usui Y, Asakage M, et al. Serum metabolomic profiling of patients with non‐infectious uveitis. J Clin Med. 2020;9(12):3955. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Japanese Gastric Cancer Association . Japanese gastric cancer treatment guidelines 2021: the 6th edition. Gastric Cancer. 2023;26:1‐25. [DOI] [PMC free article] [PubMed] [Google Scholar]
27. Mine S, Tanaka K, Kawachi H, et al. Japanese classification of esophageal cancer: the 12th edition: Part I. Esophagus. 2024;21(3):179‐215. [DOI] [PMC free article] [PubMed] [Google Scholar]
28. Japanese Gastric Cancer Association . Japanese Classification of Gastric Carcinoma: The 15th Edition. Kanehara Shuppan; 2017. [DOI] [PubMed] [Google Scholar]
29. Shinohara T, Ohshima K, Murayama H, Kikuchi M, Yamashita Y, Shirakusa T. Apoptosis and proliferation in gastric carcinoma: the association with histological type. Histopathology. 1996;29(2):123‐129. [DOI] [PubMed] [Google Scholar]
30. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer. 2009;45(2):228‐247. [DOI] [PubMed] [Google Scholar]
31. Pang Z, Chong J, Zhou G, et al. MetaboAnalyst 5.0: narrowing the gap between raw spectra and functional insights. Nucleic Acids Res. 2021;49(W1):W388‐W396. [DOI] [PMC free article] [PubMed] [Google Scholar]
32. Tong X, Zhao F, Thompson CB. The molecular determinants of de novo nucleotide biosynthesis in cancer cells. Curr Opin Genet Dev. 2009;19(1):32‐37. [DOI] [PMC free article] [PubMed] [Google Scholar]
33. DeBerardinis RJ, Chandel NS. Fundamentals of cancer metabolism. Sci Adv. 2016;2(5):e1600200. [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Gerner EW, Meyskens FL Jr. Polyamines and cancer: old molecules, new understanding. Nat Rev Cancer. 2004;4(10):781‐792. [DOI] [PubMed] [Google Scholar]
35. Hiramatsu K, Takahashi K, Yamaguchi T, et al. N ¹,N ¹²‐Diacetylspermine as a sensitive and specific novel marker for early‐ and late‐stage colorectal and breast cancers. Clin Cancer Res. 2005;11(8):2986‐2990. [DOI] [PubMed] [Google Scholar]
36. Yamaguchi K, Nakamura M, Shirahane K, et al. Urine diacetylspermine as a novel tumour maker for pancreatobiliary carcinomas. Dig Liver Dis. 2005;37(3):190‐194. [DOI] [PubMed] [Google Scholar]
37. Enjoji M, Nakamuta M, Arimura E, et al. Clinical significance of urinary N ¹,N ¹²‐diacetylspermine levels in patients with hepatocellular carcinoma. Int J Biol Markers. 2004;19(4):322‐327. [DOI] [PubMed] [Google Scholar]
38. Wikoff WR, Hanash S, DeFelice B, et al. Diacetylspermine is a novel prediagnostic serum biomarker for non‐small‐cell lung cancer and has additive performance with pro‐surfactant protein B. J Clin Oncol. 2015;33(33):3880‐3886. [DOI] [PMC free article] [PubMed] [Google Scholar]
39. Takahashi Y, Sakaguchi K, Horio H, et al. Urinary N ¹,N ¹²‐diacetylspermine is a non‐invasive marker for the diagnosis and prognosis of non‐small‐cell lung cancer. Br J Cancer. 2015;113(10):1493‐1501. [DOI] [PMC free article] [PubMed] [Google Scholar]
40. Niemi RJ, Roine AN, Häkkinen MR, et al. Urinary polyamines as biomarkers for ovarian cancer. Int J Gynecol Cancer. 2017;27(7):1360‐1366. [DOI] [PubMed] [Google Scholar]
41. Albaugh VL, Pinzon‐Guzman C, Barbul A. Arginine‐dual roles as an onconutrient and immunonutrient. J Surg Oncol. 2017;115(3):273‐280. [DOI] [PMC free article] [PubMed] [Google Scholar]
42. Vissers YLJ, Dejong CHC, Luiking YC, Fearon KCH, von Meyenfeldt MF, Deutz NEP. Plasma arginine concentrations are reduced in cancer patients: evidence for arginine deficiency? Am J Clin Nutr. 2005;81(5):1142‐1146. [DOI] [PubMed] [Google Scholar]
43. Bel'skaya LV, Sarf EA, Kosenok VK. Indicators of L‐arginine metabolism in saliva: a focus on breast cancer. J Oral Biosci. 2021;63(1):52‐57. [DOI] [PubMed] [Google Scholar]
44. He CZ, Zhang KH, Li Q, Liu XH, Hong Y, Lv NH. Combined use of AFP, CEA, CA125 and CAl9‐9 improves the sensitivity for the diagnosis of gastric cancer. BMC Gastroenterol. 2013;13:87. [DOI] [PMC free article] [PubMed] [Google Scholar]
45. Ning S, Wei W, Li J, et al. Clinical significance and diagnostic capacity of serum TK1, CEA, CA 19‐9 and CA 72‐4 levels in gastric and colorectal cancer patients. J Cancer. 2018;9(3):494‐501. [DOI] [PMC free article] [PubMed] [Google Scholar]
46. Liu W, Li Y, Zhang Y, et al. Circulatinglong non‐coding RNA FEZF1‐AS1 and AFAP1‐AS1 serve as potential diagnostic biomarkers for gastric cancer. Pathol Res Pract. 2020;216(1):152757. [DOI] [PubMed] [Google Scholar]
47. Lu Z, Zheng S, Liu C, et al. S100A7 as a potential diagnostic and prognostic biomarker of esophageal squamous cell carcinoma promotes M2 macrophage infiltration and angiogenesis. Clin Transl Med. 2021;11(7):e459. [DOI] [PMC free article] [PubMed] [Google Scholar]
48. Zheng Q, Zhang L, Tu M, et al. Development of a panel of autoantibody against NSG1 with CEA, CYFRA21‐1, and SCC‐Ag for the diagnosis of esophageal squamous cell carcinoma. Clin Chim Acta. 2021;520:126‐132. [DOI] [PubMed] [Google Scholar]
49. Wang J, Yu L, Sun Y, et al. Development and evaluation of serum CST1 detection for early diagnosis of esophageal squamous cell carcinoma. Cancer Manag Res. 2021;13:8341‐8352. [DOI] [PMC free article] [PubMed] [Google Scholar]
50. Zhang J, Xu Z, Yu L, Chen M, Li K. Assessment of the potential diagnostic value of serum p53 antibody for cancer: a meta‐analysis. PLoS One. 2014;9(6):e99255. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1.

CAS-115-3089-s002.tif^{(838.4KB, tif)}

Figure S2.

CAS-115-3089-s005.tif^{(862.5KB, tif)}

Figure S3.

CAS-115-3089-s004.tif^{(860.7KB, tif)}

Figure S4.

CAS-115-3089-s003.tif^{(892.4KB, tif)}

Tables S1–S2.

CAS-115-3089-s001.docx^{(23.3KB, docx)}

[cas16256-bib-0001] 1. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68(6):394‐424. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0002] 2. Watanabe M, Toh Y, Ishihara R, et al. Comprehensive registry of esophageal cancer in Japan, 2014. Esophagus. 2022;19(1):1‐26. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0003] 3. Suzuki S, Takahashi A, Ishikawa T, et al. Surgically treated gastric cancer in Japan: 2011 annual report of the national clinical database gastric cancer registry. Gastric Cancer. 2021;24(3):545‐566. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0004] 4. Lee KJ, Inoue M, Otani T, et al. Gastric cancer screening and subsequent risk of gastric cancer: a large‐scale population‐based cohort study, with a 13‐year follow‐up in Japan. Int J Cancer. 2006;118(9):2315‐2321. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0005] 5. Matsumoto S, Yamasaki K, Tsuji K, Shirahama S. Results of mass endoscopic examination for gastric cancer in Kamigoto Hospital, Nagasaki Prefecture. World J Gastroenterol. 2007;13(32):4316‐4320. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0006] 6. Gastric cancer screening guidelines 2014 based on the effectiveness evaluation (in Japanese). http://canscreen.ncc.go.jp/guideline/iganguide2014_150421.pdf. Accessed December 18, 2023.

[cas16256-bib-0007] 7. Arnal MJD, Arenas ÁF, Arbeloa ÁL. Esophageal cancer: risk factors, screening and endoscopic treatment in Western and Eastern countries. World J Gastroenterol. 2015;21(26):7933‐7943. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0008] 8. Yajima S, Suzuki T, Oshima Y, et al. New assay system elecsys anti‐p53 to detect serum anti‐p53 antibodies in esophageal cancer patients and colorectal cancer patients: multi‐institutional study. Ann Surg Oncol. 2021;28(7):4007‐4015. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0009] 9. Shimada H, Noie T, Ohashi M, Oba K, Takahashi Y. Clinical significance of serum tumor markers for gastric cancer: a systematic review of literature by the Task Force of the Japanese Gastric Cancer Association. Gastric Cancer. 2014;17(1):26‐33. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0010] 10. Hyvärinen E, Savolainen M, Mikkonen JJW, Kullaa AM. Salivary metabolomics for diagnosis and monitoring diseases: challenges and possibilities. Metabolomics. 2021;11(9):587. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0011] 11. Sugimoto M, Wong DT, Hirayama A, Soga T, Tomita M. Capillary electrophoresis mass spectrometry‐based saliva metabolomics identified oral, breast and pancreatic cancer‐specific profiles. Metabolomics. 2010;6(1):78‐95. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0012] 12. Ishikawa S, Sugimoto M, Kitabatake K, et al. Identification of salivary metabolomic biomarkers for oral cancer screening. Sci Rep. 2016;6:31520. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0013] 13. Murata T, Yanagisawa T, Kurihara T, et al. Salivary metabolomics with alternative decision tree‐based machine learning methods for breast cancer discrimination. Breast Cancer Res Treat. 2019;177(3):591‐601. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0014] 14. Asai Y, Itoi T, Sugimoto M, et al. Elevated polyamines in saliva of pancreatic cancer. Cancers. 2018;10(2):43. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0015] 15. Takamori S, Ishikawa S, Suzuki J, et al. Differential diagnosis of lung cancer and benign lung lesion using salivary metabolites: a preliminary study. Thorac Cancer. 2021;13:460‐465. doi: 10.1111/1759-7714.14282 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0016] 16. Kuwabara H, Katsumata K, Iwabuchi A, et al. Salivary metabolomics with machine learning for colorectal cancer detection. Cancer Sci. 2022;113(9):3234‐3243. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0017] 17. Sugimoto M. Salivary metabolomics for cancer detection. Expert Rev Proteomics. 2020;17(9):639‐648. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0018] 18. Wu ZZ, Wang JG, Zhang XL. Diagnostic model of saliva protein finger print analysis of patients with gastric cancer. World J Gastroenterol. 2009;15(7):865‐870. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0019] 19. Xiao H, Zhang Y, Kim Y, et al. Differential proteomic analysis of human saliva using tandem mass tags quantification for gastric cancer detection. Sci Rep. 2016;6:22165. doi: 10.1038/srep22165 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0020] 20. Bel'skaya LV, Sarf EA, Shalygin SP, Postnova TV, Kosenok VK. Identification of salivary volatile organic compounds as potential markers of stomach and colorectal cancer: a pilot study. J Oral Biosci. 2020;62(2):212‐221. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0021] 21. Hirayama A, Kami K, Sugimoto M, et al. Quantitative metabolome profiling of colon and stomach cancer microenvironment by capillary electrophoresis time‐of‐flight mass spectrometry. Cancer Res. 2009;69(11):4918‐4925. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0022] 22. Du J, Zhang L. Analysis of salivary microRNA expression profiles and identification of novel biomarkers in esophageal cancer. Oncol Lett. 2017;14(2):1387‐1394. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0023] 23. Saeki Y, Takenouchi A, Otani E, et al. Long‐term mastication changed salivary metabolomic profiles. Metabolites. 2022;12(7):660. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0024] 24. Shimizu H, Usui Y, Wakita R, et al. Differential tissue metabolic signatures in IgG4‐related ophthalmic disease and orbital mucosa‐associated lymphoid tissue lymphoma. Invest Ophthalmol Vis Sci. 2021;62(1):15. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0025] 25. Shimizu H, Usui Y, Asakage M, et al. Serum metabolomic profiling of patients with non‐infectious uveitis. J Clin Med. 2020;9(12):3955. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0026] 26. Japanese Gastric Cancer Association . Japanese gastric cancer treatment guidelines 2021: the 6th edition. Gastric Cancer. 2023;26:1‐25. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0027] 27. Mine S, Tanaka K, Kawachi H, et al. Japanese classification of esophageal cancer: the 12th edition: Part I. Esophagus. 2024;21(3):179‐215. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0028] 28. Japanese Gastric Cancer Association . Japanese Classification of Gastric Carcinoma: The 15th Edition. Kanehara Shuppan; 2017. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0029] 29. Shinohara T, Ohshima K, Murayama H, Kikuchi M, Yamashita Y, Shirakusa T. Apoptosis and proliferation in gastric carcinoma: the association with histological type. Histopathology. 1996;29(2):123‐129. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0030] 30. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer. 2009;45(2):228‐247. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0031] 31. Pang Z, Chong J, Zhou G, et al. MetaboAnalyst 5.0: narrowing the gap between raw spectra and functional insights. Nucleic Acids Res. 2021;49(W1):W388‐W396. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0032] 32. Tong X, Zhao F, Thompson CB. The molecular determinants of de novo nucleotide biosynthesis in cancer cells. Curr Opin Genet Dev. 2009;19(1):32‐37. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0033] 33. DeBerardinis RJ, Chandel NS. Fundamentals of cancer metabolism. Sci Adv. 2016;2(5):e1600200. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0034] 34. Gerner EW, Meyskens FL Jr. Polyamines and cancer: old molecules, new understanding. Nat Rev Cancer. 2004;4(10):781‐792. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0035] 35. Hiramatsu K, Takahashi K, Yamaguchi T, et al. N ¹,N ¹²‐Diacetylspermine as a sensitive and specific novel marker for early‐ and late‐stage colorectal and breast cancers. Clin Cancer Res. 2005;11(8):2986‐2990. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0036] 36. Yamaguchi K, Nakamura M, Shirahane K, et al. Urine diacetylspermine as a novel tumour maker for pancreatobiliary carcinomas. Dig Liver Dis. 2005;37(3):190‐194. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0037] 37. Enjoji M, Nakamuta M, Arimura E, et al. Clinical significance of urinary N ¹,N ¹²‐diacetylspermine levels in patients with hepatocellular carcinoma. Int J Biol Markers. 2004;19(4):322‐327. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0038] 38. Wikoff WR, Hanash S, DeFelice B, et al. Diacetylspermine is a novel prediagnostic serum biomarker for non‐small‐cell lung cancer and has additive performance with pro‐surfactant protein B. J Clin Oncol. 2015;33(33):3880‐3886. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0039] 39. Takahashi Y, Sakaguchi K, Horio H, et al. Urinary N ¹,N ¹²‐diacetylspermine is a non‐invasive marker for the diagnosis and prognosis of non‐small‐cell lung cancer. Br J Cancer. 2015;113(10):1493‐1501. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0040] 40. Niemi RJ, Roine AN, Häkkinen MR, et al. Urinary polyamines as biomarkers for ovarian cancer. Int J Gynecol Cancer. 2017;27(7):1360‐1366. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0041] 41. Albaugh VL, Pinzon‐Guzman C, Barbul A. Arginine‐dual roles as an onconutrient and immunonutrient. J Surg Oncol. 2017;115(3):273‐280. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0042] 42. Vissers YLJ, Dejong CHC, Luiking YC, Fearon KCH, von Meyenfeldt MF, Deutz NEP. Plasma arginine concentrations are reduced in cancer patients: evidence for arginine deficiency? Am J Clin Nutr. 2005;81(5):1142‐1146. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0043] 43. Bel'skaya LV, Sarf EA, Kosenok VK. Indicators of L‐arginine metabolism in saliva: a focus on breast cancer. J Oral Biosci. 2021;63(1):52‐57. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0044] 44. He CZ, Zhang KH, Li Q, Liu XH, Hong Y, Lv NH. Combined use of AFP, CEA, CA125 and CAl9‐9 improves the sensitivity for the diagnosis of gastric cancer. BMC Gastroenterol. 2013;13:87. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0045] 45. Ning S, Wei W, Li J, et al. Clinical significance and diagnostic capacity of serum TK1, CEA, CA 19‐9 and CA 72‐4 levels in gastric and colorectal cancer patients. J Cancer. 2018;9(3):494‐501. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0046] 46. Liu W, Li Y, Zhang Y, et al. Circulatinglong non‐coding RNA FEZF1‐AS1 and AFAP1‐AS1 serve as potential diagnostic biomarkers for gastric cancer. Pathol Res Pract. 2020;216(1):152757. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0047] 47. Lu Z, Zheng S, Liu C, et al. S100A7 as a potential diagnostic and prognostic biomarker of esophageal squamous cell carcinoma promotes M2 macrophage infiltration and angiogenesis. Clin Transl Med. 2021;11(7):e459. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0048] 48. Zheng Q, Zhang L, Tu M, et al. Development of a panel of autoantibody against NSG1 with CEA, CYFRA21‐1, and SCC‐Ag for the diagnosis of esophageal squamous cell carcinoma. Clin Chim Acta. 2021;520:126‐132. [DOI] [PubMed] [Google Scholar]

[cas16256-bib-0049] 49. Wang J, Yu L, Sun Y, et al. Development and evaluation of serum CST1 detection for early diagnosis of esophageal squamous cell carcinoma. Cancer Manag Res. 2021;13:8341‐8352. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cas16256-bib-0050] 50. Zhang J, Xu Z, Yu L, Chen M, Li K. Assessment of the potential diagnostic value of serum p53 antibody for cancer: a meta‐analysis. PLoS One. 2014;9(6):e99255. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Salivary metabolomic biomarkers for esophageal and gastric cancers by liquid chromatography–mass spectrometry

Kosuke Nakane

Koichi Yagi

Sho Yajima

Sachiyo Nomura

Masahiro Sugimoto

Yasuyuki Seto

Abstract

Abbreviations

1. INTRODUCTION

2. MATERIALS AND METHODS

2.1. Study design and population

2.2. Sample collection

2.3. Pretreatment and analytical processes

2.4. Metabolic profiles in saliva and tissues

2.4.1. Salivary biomarkers for gastric and esophageal cancers, and metabolic profiles in gastric cancer tissues

2.4.2. Salivary biomarkers for differentiated‐ and undifferentiated‐type gastric cancer, and relationship between salivary biomarkers and proliferating ability of gastric cancer

2.4.3. Diagnostic models for esophageal and gastric cancers and their diagnostic performance

2.4.4. Changes in salivary biomarker concentrations before and after surgery

2.4.5. Changes in salivary biomarker concentrations before and after chemotherapy

2.5. Statistical analysis

3. RESULTS

3.1. Patient characteristics

TABLE 1.

3.2. Salivary biomarkers for gastric cancer and metabolic profiles in gastric cancer tissues

FIGURE 1.

FIGURE 2.

3.3. Salivary biomarkers for esophageal cancer

FIGURE 3.

3.4. Salivary biomarkers for differentiated‐ and undifferentiated‐type gastric cancer, and relationship between salivary biomarkers and Ki‐67 index

FIGURE 4.

3.5. Diagnostic models for esophageal and gastric cancers

FIGURE 5.

3.6. Changes in salivary cytosine concentration before and after surgery

FIGURE 6.

3.7. Changes in salivary cytosine concentration before and after chemotherapy

FIGURE 7.

4. DISCUSSION

AUTHOR CONTRIBUTIONS

CONFLICT OF INTEREST STATEMENT

ETHICS STATEMENTS

Supporting information

ACKNOWLEDGMENTS

REFERENCES

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases