Noninvasive Breathomics for Developing Carbonyl Biomarkers in Esophageal Cancer

doi:10.21203/rs.3.rs-8170337/v1

Background

Despite being the gold standard for esophageal cancer (EC) diagnosis, the invasiveness and cost limit the use of upper gastrointestinal endoscopy for primary screening. We evaluated the potential of aldehyde profiling in exhaled breath condensate (EBC), a noninvasively obtained biofluid, as an EC diagnostic biomarker.

Methods

EBC, plasma, or saliva samples were collected from 152 patients with EC and 63 healthy controls (HC); aldehyde concentrations in EBC and plasma were measured using liquid chromatography–tandem mass spectrometry. Saliva microbiota was analyzed by 16S rRNA sequencing, focusing on oral acetaldehyde-producing bacteria.

Results

Among eight different aldehydes measured using EBC samples, acetaldehyde and formaldehyde concentrations were significantly higher in patients with EC than in HC, with elevations observed in both early and advanced-stage disease. Plasma acetaldehyde and formaldehyde levels were also increased in patients with EC compared with that in HC. EBC acetaldehyde and formaldehyde were identified as diagnostic biomarkers (areas under the curve: 0.77/0.87, sensitivities: 78%/89%, specificities: 70%/79%). It remains unclear whether the salivary microbiota influence the acetaldehyde concentration in EBC.

Conclusions

Acetaldehyde and formaldehyde in EBC are potential diagnostic biomarkers for EC. EBC is a noninvasive sample useful for screening with potential utility for early detection of EC.

Worldwide, esophageal cancer (EC) is ranked 11th among the most common cancers and is the 7th leading cause of cancer-related deaths [1]. EC is frequently diagnosed at an advanced stage, which has a 5-year survival rate of 15–20% [2,3,4]. EC is definitively diagnosed from tissue biopsy specimens that are obtained during upper gastrointestinal endoscopy. Early stage cancer can be detected through cancer screening, which may be associated with an improved long-term prognosis [5]. However, upper gastrointestinal endoscopy is invasive and costly, which makes it unsuitable for primary EC screening. Novel, noninvasive diagnostic methods for EC are urgently required.

Biological samples that comprise biomarkers include blood, urine, saliva, dental plaque, and exhaled breath. The clinical biomarkers for EC diagnosis include serum antigens, such as carcinoembryonic antigen and squamous cell carcinoma antigen, albeit with limited sensitivity and specificity [6,7]. Recent studies have reported noninvasive diagnostic methods using exhaled breath [6,8,9,10]. In particular, exhaled breath condensate (EBC) is a noninvasive biological specimen. The European Respiratory Society and the American Thoracic Society defined EBC as a liquid or frozen specimen obtained by cooling exhaled air for measuring both volatile and non-volatile compounds [11]. EBC consists mostly of water (99.9%), and water-soluble compounds are substantially diluted [12]. EBC contains DNA, RNA, proteins, metabolites, and volatile compounds [13], and is a source of biomarkers for various diseases [14,15]. Our laboratory has conducted metabolite analysis using EBC and demonstrated that supersulfur metabolites are useful diagnostic biomarkers for EC [16].

Aldehydes, such as acetaldehyde and formaldehyde, are both cytotoxic and carcinogenic. Heavy alcohol drinkers with inactive aldehyde dehydrogenase 2 (ALDH2) are at high risk for esophageal squamous cell carcinoma (SCC) [17-20], and numerous studies have reported an association between acetaldehyde and EC. Exhaled breath aldehydes in the gas phase are elevated in patients with esophageal adenocarcinoma (AC) [9,21]. Compared to healthy individuals, patients with EC exhibit altered oral microflora [22,23]. Among oral bacteria, the genera Neisseria, Streptococcus, and Fusobacterium produce acetaldehyde and may contribute to acetaldehyde concentrations in EBC. This study aimed to ascertain the utility of aldehydes in EBC as noninvasive biomarkers by measuring and comparing their levels in patients with EC and healthy controls (HC). We used saliva samples collected during the same period to analyze the oral commensal microbiota and to examine the relationship between oral bacteria and acetaldehyde concentrations in EBC

Participants

Patients with a diagnosis of EC confirmed at the Tohoku University Hospital between July 2021 and June 2024 and who provided informed consent were included in this study (Fig. 1). Specimens from these patients were collected prior to any initial treatment, including chemotherapy, radiotherapy, and surgery, or dental interventions. The exclusion criteria were as follows: (1) history of cancer, (2) presence of synchronous multiple primary cancers at different sites, and (3) incomplete clinical data. Healthy individuals without a history of cancer were recruited as volunteers for sample collection and served as HC. Written informed consent was obtained from all participants. The clinical information for patients with EC was obtained from electronic medical records whereas the information of HC was collected through questionnaires. This study was approved by the Ethics Committee of the Tohoku University Graduate School of Medicine (approval number: 2021-1-167, 2024-1-770). All procedures followed were in accordance with the Helsinki Declaration of 1964 and later versions.

Collection and Analysis of EBC

EBC was collected by applying a nose clip while the subject was seated and breathing normally at rest for 5 min. If the EBC collected after 5 min was insufficient, an additional 2-min collection was performed. The collected sample was rapidly cooled to −20 ℃ using a recovery device (Supplementary Fig. 1) and subsequently stored at −80 ℃ [16,24]. No strict restrictions were imposed on fasting, smoking, or alcohol consumption prior to the sample collection. Quantification of aldehydes in EBC used the dinitrophenylhydrazine (DNPH) method [25]. Frozen EBC samples were thawed prior to analysis; 50 µL EBCs were reacted with 2.3 mM DNPH in 200 mM hydrogen chloride at 37 ℃ for 20 min for derivatization. The reaction mixtures were injected into the liquid chromatography–electrospray ionization-tandem mass spectrometry (LC-ESI-MS/MS, LCMS-8060NX; Shimadzu, Kyoto, Japan). Each carbonyl compound was separated by YMC-Triart C18 column (50 x 2.0 mm inner diameter) under the following elution conditions: mobile phase A (0.1% formic acid) with a gradient of mobile phase B (0.1% formic acid in acetonitrile) from 30 to 100% (30-100%B: 0-13 min, 100-100%B: 13-15 min) at a flow rate of 0.3ml/min at 40 ℃. MS spectra were obtained at each temperature of the ESI probe at 300 ℃, desolvation line at 250 ℃ and heat block at 400 ℃; nebulizer, heating, and drying nitrogen gas flows were set at 3, 10, and 10 l/min, respectively. Each metabolite was identified and quantified by means of multiple reaction monitoring mode. The multiple reaction monitoring parameters of target metabolites were indicated at Supplementary Table 1. The concentration of each carbonyl compound was calculated from a standard curve obtained from the 16 aldehydes-DNPH mixed standard solution (FUJIFILM Wako Chemicals, Hiratsuka, Japan).

Collection and Analysis of plasma

After blood collection, plasma was promptly separated by centrifugation at 3,500 rpm for 10 min at 4 ℃, and the plasma was stored at −80 ℃ until analysis. Frozen plasma samples were thawed prior to analysis; 20 µL plasmas were mixed with 2.7 mM DNPH in 50% methanol and 200 mM hydrogen chloride at 37 ℃ for 20 min for derivatization. The reaction mixtures were then centrifuged at 14,000 rpm for 10 min at 4 ℃. Next, 20 µL supernatants were diluted with 20 µL ultrapure water. Measurements were performed by injecting 10 µL of the sample into the LC-ESI-MS/MS. The LC-ESI-MS/MS settings were identical to those used for EBC analysis. As with EBC, the concentration of each carbonyl compound was calculated from a standard curve obtained from the 16 aldehydes-DNPH mixed standard solution.

Oral Microbiota Analysis

Saliva samples were collected and subjected to 16S rRNA gene sequencing analysis. Saliva samples from patients with EC were collected in the Division of Perioperative Oral Health Management prior to treatment initiation and stored at −80 ℃ until use. The sequencing of the V1–V2 region was performed using the genetic analysis service provided by Bioengineering Lab (Bioengineering Lab. Co., Ltd. Kanagawa, Japan).

Statistical Analysis

Statistical analyses were performed using GraphPad Prism version 10 (GraphPad Software, Boston, MA, USA). Between-group comparisons were performed using the Mann–Whitney U test whereas multigroup comparisons were conducted using one-way ANOVA. Multivariate analysis was performed using logistic regression. The receiver operating characteristic (ROC) curves and areas under the curve (AUC) were calculated using EZR (Easy R) version 1.68 [26].

Background and Clinicopathological Characteristics of the Participants

This study included 152 patients with EC and 63 HC. Figure 1a comprises a flowchart of the enrolled patients and HC. Ultimately, the EBC analysis included 106 EC and 47 HC; the plasma analysis included 50 EC and 13 HC; and the saliva analysis included 33 EC and 56 HC.

The clinicopathological characteristics of the patients are summarized in Table 1. Tumor histology revealed SCC, AC, and other EC subtypes in 91, 23, and 5 cases, respectively. In clinical staging according to the UICC TNM 8th edition [27], we classified 48 cases as cStage I/II and 71 cases as cStage III/IV. Statistically significant differences were observed between the EC and HC groups regarding age, sex, smoking history, Brinkman Index, and alcohol consumption.

Validation of Aldehyde Concentrations in EBC

Using a 16 aldehydes-DNPH mixed standard solution, 8 of the 16 aldehydes were detected. Among the 8 aldehydes that were quantified by LC-ESI-MS/MS, acetaldehyde and formaldehyde exhibited statistically significant differences between the EC and HC groups (Fig. 1b). The acetaldehyde concentration in EBC was higher in patients with EC (median: 1.88 µM) than in HC (median: 1.31 µM; P < 0.0001). The formaldehyde concentration in EBC was elevated in patients with EC (median: 8.54 µM) compared with that in HC (median: 4.53 µM, P < 0.0001). The results of all aldehyde analyses are shown in Supplementary Fig. 2.

Subgroup Analysis for EBC in Patients with EC Compared with HC

Subgroup analysis based on histological type (Fig. 2a) showed that the acetaldehyde and formaldehyde concentrations in EBC were higher in patients with SCC and AC as compared with HC. There was no statistically significant difference between the SCC and AC groups. Subgroup analysis based on clinical stage (Fig. 2b) showed that acetaldehyde and formaldehyde concentrations in EBC were increased in patients with both early stage and advanced-stage EC compared with that in the HC group. No statistically significant difference was observed between early stage and advanced-stage groups.

Plasma Aldehyde Levels and Their Correlation with EBC Concentrations

The acetaldehyde and formaldehyde concentrations in plasma were measured (Fig. 3). Acetaldehyde and formaldehyde levels in plasma were elevated in patients with EC compared with that in HC. To investigate the origin of aldehydes in EBC, we analyzed the correlation between acetaldehyde and formaldehyde concentrations in both EBC and plasma among participants with available data for both matrices. No significant correlation was observed between plasma and EBC aldehyde concentrations, which may partly be due to the small number of HC (n = 13) (Supplementary Fig. 3). In addition, aldehyde production by oral microbiota may also contribute to the EBC aldehyde levels.

Association Between Oral Microbiota Composition and Aldehyde Levels in EBC

Based on the analysis of the oral microbiota, 19 acetaldehyde-producing bacterial species [28] were identified (Fig. 4). The proportion of acetaldehyde-producing bacteria within the overall oral microbiota was compared between the HC and EC groups, but no significant differences were observed (Fig. 4a). However, an analysis at the species level revealed that Fusobacterium nucleatum was significantly more prevalent in the EC group, whereas Streptococcus mitis was significantly more prevalent in the HC group (Fig. 4b). Based on the data available at present, no apparent association was observed between the salivary microbiota and acetaldehyde concentration in EBC.

EBC’s Usefulness as a Diagnostic Biomarker of Acetaldehyde and Formaldehyde in EBC

Multivariate analysis was performed to assess the influence of patient background variables (Table 2). Both acetaldehyde and formaldehyde levels showed significant differences, which suggested that these two metabolites could potentially help distinguish between EC and HC.

To verify the diagnostic accuracy of EC based on acetaldehyde and formaldehyde concentrations in EBC, ROC curve analyses were performed (Fig. 1c). The sensitivity, specificity, and AUC for acetaldehyde were 78%, 70%, and 0.77, respectively, with a cutoff value of 1.47 µM. For formaldehyde, the sensitivity, specificity, and AUC were 89%, 79%, and 0.87, respectively, with a cutoff value of 5.91 µM.

Correlation between EBC Acetaldehyde Concentration and Alcohol Flushing Response

The presence or absence of flushing, as determined by a questionnaire survey, was used as an indirect indicator of ALDH2 polymorphism to infer its relationship with acetaldehyde concentration. The relationship between alcohol flushing response and EBC acetaldehyde concentrations was examined in patients with EC. The median EBC acetaldehyde concentration was 2.01 µM in patients with alcohol flushing and 1.76 µM in those without flushing. Although the difference was not statistically significant (P = 0.078), there was a trend toward higher acetaldehyde concentrations in patients with flushing (Supplementary Fig. 4).

This study investigated potential early diagnostic biomarkers for EC. The results suggest that acetaldehyde and formaldehyde concentrations in exhaled breath may serve as useful noninvasive biomarkers for the early detection of EC. To the best of our knowledge, only three studies have reported on aldehyde-based biomarkers for EC using gas analysis of exhaled breath [8,9,29]. One study reported a diagnostic model for esophageal and gastric cancer that combined four volatile organic compounds—hexanoic acid, phenol, methylphenol, and ethylphenol—with an ROC curve AUC of 0.91 [8]. Another study proposed a diagnostic model for esophageal adenocarcinoma using volatile organic compounds, such as carboxylic acids, phenol, and aldehydes—including butanal, pentanal, hexanal, heptanal, octanal, nonanal, and decanal—which achieved an AUC of 0.90 [9]. The third study demonstrated elevated aldehydes, including acetaldehyde, in exhaled breath in patients with EC, but did not report ROC curve [29]. In contrast, the distinguishing feature of the present study is the use of EBC, which is a sample that is noninvasive and easy to collect and store. This is the first report of aldehyde measurement in EBC for EC diagnosis. Volatile organic compounds in exhaled breath in the gaseous phase are commonly measured using gas chromatography mass spectrometry. However, this approach has limitations in quantitative accuracy due to factors such as incomplete sample recovery during breath collection, adsorption within the column, and solvent elution. Analysis of EBC offers advantages, as it enables accurate collection of steady-state exhaled breath, reflects the status of the lower airways, allows absolute quantification by LC-ESI-MS/MS. Notably, in this study, formaldehyde in EBC alone achieved an AUC of 0.85, which is comparable to previous diagnostic models that relied on the combined analysis of multiple metabolites. In addition, both SCC and AC were studied as histological subtypes of EC, and the concentrations of acetaldehyde and formaldehyde in EBC were significantly elevated in each subtype compared to HC group.

Furthermore, the plasma acetaldehyde and formaldehyde concentrations were higher in patients with EC than in HC, which is consistent with the results of EBC measurements. This was attributed to a correlation between blood aldehyde concentrations and EBC to some extent, but no clear correlation was found between blood aldehyde concentrations and EBC in individual patient groups where both were measured (Supplementary Fig. 3). This may be attributable to the limited sample size, variability in specimen collection times and potential influence of medications on detoxification metabolic pathways. Furthermore, with respect to acetaldehyde in EBC, acetaldehyde production by oral bacteria may have been involved. However, in our analysis of oral microbiota, the relative abundance of acetaldehyde-producing bacteria did not differ significantly between HC and patients with EC. Nevertheless, variations in particular bacterial species, notably an elevated presence of Fusobacterium nucleatum, may indicate that differences in acetaldehyde-producing capacity among various taxa may influence overall acetaldehyde concentrations. Therefore, the origin of aldehydes in EBC warrants further investigation. Moreover, ALDH2 is a well-known enzyme that is involved in acetaldehyde metabolism. Although ALDH2 gene polymorphisms were not examined in this study, they may play a role in the elevated acetaldehyde concentrations observed in the EBC of patients with EC.

Formaldehyde is produced in vivo by various reactions, including demethylation of histones and nucleic acids [30,31]. Formaldehyde is produced during the synthesis of glycine from serine, and serine has been reported to be necessary for cancer cell growth [32]. Cancer cells overexpress serine hydroxymethyltransferase —an enzyme that is involved in the de novo serine synthesis pathway [33,34]. During the conversion of serine to glycine, formaldehyde is produced, and formaldehyde levels in cancer cells possibly increase owing to the upregulation of the one-carbon metabolism cycle [30,35]. This increase in intracellular formaldehyde possibly leads to elevated concentrations of formaldehyde in both blood and EBC. Alcohol dehydrogenase 5 (ADH5) is a well-known enzyme that is involved in formaldehyde metabolism [36,37], and, similar to acetaldehyde, genetic polymorphisms of ADH5 may influence formaldehyde levels. We plan to investigate these gene polymorphisms using next-generation sequencing analysis in future studies.

The main limitations of this study are as follows. First, the age, sex, smoking history, and drinking history of the participants were not matched between the EC and HC groups. In particular, age and sex represent significant confounding factors. Second, this study did not include patients with other cancer types; therefore, the specificity of the biomarkers for EC remains uncertain. Third, the effects of diet, alcohol consumption, and smoking could not be fully controlled, as fasting, alcohol abstinence, and smoking cessation prior to specimen collection were not strictly enforced in this study to reflect real-world clinical conditions. Especially, the influence of recent alcohol intake on acetaldehyde concentration could not be ruled out. Fourth, this study did not evaluate the presence of genetic polymorphisms in ALDH2 and ADH5—the enzymes responsible for metabolizing acetaldehyde and formaldehyde, respectively. Instead, the presence or absence of alcohol flushing response, as determined by a questionnaire survey, was employed as an indirect indicator of ALDH2 status with respect to acetaldehyde concentration. Fifth, blood aldehyde concentrations were measured using plasma samples. Although acetaldehyde is more abundant in erythrocytes than in plasma [38], its concentration in erythrocytes was not measured in this study. Sixth, the mechanism underlying the elevation of aldehydes in patients with cancer remains unclear. Although quantitative analysis demonstrated its potential usefulness as a diagnostic marker, the elucidation of the underlying mechanism requires further investigation.

In conclusion, acetaldehyde and formaldehyde concentrations in the EBC of patients with EC were significantly higher than those in HC, which indicates their potential utility as diagnostic biomarkers for EC. Furthermore, the noninvasive nature and ease of sample preservation make EBC a promising candidate for primary screening prior to upper gastrointestinal endoscopy. In future studies, we plan to examine ALDH2 and ADH5 gene polymorphisms using next-generation sequencing analysis to validate the origin of acetaldehyde and formaldehyde in EBC.

Acknowledgements

We thank all of the patients who participated in this trial. We appreciate Bioengineering Lab for analyzing saliva samples, and Elsevier Language Services (https://webshop.elsevier.com/language-editing/) for the English language editing. This work was supported by Yuka Takamatsu (Tohoku University Clinical Research Coordinator), who assisted with specimen collection, storage, and management.

Authors' contributions

KF: investigation, data curation, formal analysis, visualization, writing-original draft; YO: resources, writing-review and editing, project administration, funding acquisition, supervision; SO: methodology, validation; HO: resources; TT: methodology; MM: writing-review and editing; JY: methodology; TI: methodology; SA: resources, investigation; YT: resources; CS: resources, HI: resources; NU: resources; TAr: resources; YG: resources; TAb: funding acquisition; HM: funding acquisition; MI: resources; TK: supervision; TAk: conceptualization, writing-review and editing, funding acquisition, supervision

Ethics approval and consent to participate

The study was approved by the Ethics Committee of the Tohoku University Graduate School of Medicine (approval number: 2021-1-167, 2024-1-770). Informed consent was obtained from all participants. All procedures followed were in accordance with the Helsinki Declaration of 1964 and later versions.

Consent for publication

The manuscript contains no individual person’s data in any form.

Data availability

Data supporting the findings of this study are available from the corresponding author upon request.

Competing interests

The authors declare no conflict of interest.

Funding information

This research was supported by the grants from the Japan Society for the Promotion of Science KAKENHI Grant Numbers JP21H05263, JP23K20040, JP24H00063, and 22K19397 to T. Akaike, JP24H00605 to H. Motohashi, JP23K08206 to Y. Ozawa, 23K14333 to S. Ogata, 23K06094 to T. Takata, 23K06145 to M. Morita, 23K06386 to T. Ida; by the Japan Science and Technology Agency (JST) CREST Grant Number JPMJCR2024 to T. Akaike; and by the Japan Agency for Medical Research and Development (AMED) Grant Number JP21zf0127001 to T. Akaike and 24zf0127001h0004 to T. Abe.

Bray, F., Laversanne, M., Sung, H., Ferlay, J., Siegel, R.L., Soerjomataram, I. et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA. Cancer J. Clin. 74, 229–263 (2024).
Short, M.W., Burgers, K.G., Fry, V.T. Esophageal cancer. Am. Fam. Physician 95, 22–28 (2017).
Pennathur, A., Gibson, M.K., Jobe, B.A., Luketich, J.D. Oesophageal carcinoma. Lancet. 381, 400–412 (2013).
Enzinger, P.C., Mayer, R.J. Esophageal cancer. N. Engl. J. Med. 349, 2241–2252 (2003).
Pennathur, A., Farkas, A., Krasinskas, A.M., Ferson, P.F., Gooding, W.E., Gibson, M.K. et al. Esophagectomy for T1 esophageal cancer: outcomes in 100 patients and implications for endoscopic therapy. Ann. Thorac. Surg. 87, 1048–1054; discussion 1054-5 (2009).
Yazbeck, R., Jaenisch, S.E., Watson, D.I. From blood to breath: New horizons for esophageal cancer biomarkers. World J. Gastroenterol. 22, 10077–10083 (2016).
Zhang, J., Zhu, Z., Liu, Y., Jin, X., Xu, Z., Yu, Q. et al. Diagnostic value of multiple tumor markers for patients with esophageal carcinoma. PLoS One 10, e0116951 (2015).
Kumar, S., Huang, J., Abbassi-Ghadi, N., Španěl, P., Smith, D., Hanna, G.B. Selected ion flow tube mass spectrometry analysis of exhaled breath for volatile organic compound profiling of esophago-gastric cancer. Anal. Chem. 85, 6121–6128 (2013).
Kumar, S., Huang, J., Abbassi-Ghadi, N., Mackenzie, H.A., Veselkov, K.A., Hoare, J.M. et al. Mass spectrometric analysis of exhaled breath for the identification of volatile organic compound biomarkers in esophageal and gastric adenocarcinoma. Ann. Surg. 262, 981–990 (2015).
Zou, X., Zhou, W., Lu, Y., Shen, C., Hu, Z., Wang, H. et al. Exhaled gases online measurements for esophageal cancer patients and healthy people by proton transfer reaction mass spectrometry. J Gastroenterol Hepatol. 31, 1837–1843 (2016).
Horváth, I., Barnes, P.J., Loukides, S., Sterk, P.J., Högman, M., Olin, A.C. et al. A European Respiratory Society technical standard: exhaled biomarkers in lung disease. Eur. Respir. J. 49, 1600965 (2017).
Konstantinidi, E.M., Lappas, A.S., Tzortzi, A.S., Behrakis, P.K. Exhaled breath condensate: technical and diagnostic aspects. ScientificWorldJournal 2015, 435160 (2015).
Campanella, A., De Summa, S., Tommasi, S. Exhaled breath condensate biomarkers for lung cancer. J. Breath Res. 13, 044002 (2019).
López-Sánchez, LM., Jurado-Gámez, B., Feu-Collado, N., Valverde, A., Cañas, A., Fernández-Rueda, JL. et al. Exhaled breath condensate biomarkers for the early diagnosis of lung cancer using proteomics. Am J Physiol Lung Cell Mol Physiol. 313, L664-L676 (2017).
Soyer, T., Birben, E., Türer, Ö.B., Kahveci, M., Tuğcu, G.D., Soyer, Ö.U. et al. MicroRNA levels in exhaled breath condensate of patients with esophageal atresia. Dis. Esophagus 34, doaa082 (2021).
Asamitsu, S., Ozawa, Y., Okamoto, H., Ogata, S., Matsunaga, T., Yoshitake, J. et al. Supersulfide metabolome of exhaled breath condensate applied as diagnostic biomarkers for esophageal cancer. Cancer Sci. 116, 1023–1033 (2025).
Lee, C.H., Lee, J.M., Wu, D.C., Goan, Y.G., Chou, S.H., Wu, I.C. et al. Carcinogenetic impact of ADH1B and ALDH2 genes on squamous cell carcinoma risk of the esophagus with regard to the consumption of alcohol, tobacco and betel quid. Int. J. Cancer 122, 1347–1356 (2008).
Yang, C.X., Matsuo, K., Ito, H., Hirose, K., Wakai, K., Saito, T. et al. Esophageal cancer risk by ALDH2 and ADH2 polymorphisms and alcohol consumption: exploration of gene-environment and gene-gene interactions. Asian Pac. J. Cancer Prev. 6, 256–262 (2005).
Yokoyama, A., Muramatsu, T., Ohmori, T., Higuchi, S., Hayashida, M., Ishii, H. Esophageal cancer and aldehyde dehydrogenase-2 genotypes in Japanese males. Cancer Epidemiol. Biomarkers Prev. 5, 99–102 (1996).
Yokoyama, A., Omori, T., Yokoyama, T., Sato, Y., Mizukami, T., Matsushita, S. et al. Risk of squamous cell carcinoma of the upper aerodigestive tract in cancer-free alcoholic Japanese men: an endoscopic follow-up study. Cancer Epidemiol. Biomarkers Prev. 15, 2209–2215 (2006).
Markar, S.R., Wiggins, T., Antonowicz, S., Chin, S.T., Romano, A., Nikolic, K. et al. Assessment of a noninvasive exhaled breath test for the diagnosis of oesophagogastric cancer. JAMA Oncol. 4, 970–976 (2018).
Kawasaki, M., Ikeda, Y., Ikeda, E., Takahashi, M., Tanaka, D., Nakajima, Y. et al. Oral infectious bacteria in dental plaque and saliva as risk factors in patients with esophageal cancer. Cancer 127, 512–519 (2021).
Li, Y., Wei, B., Xue, X., Li, H., Li, J. Microbiome changes in esophageal cancer: implications for pathogenesis and prognosis. Cancer Biol. Med. 21, 163–174 (2023).
Matsunaga, T., Sano, H., Takita, K., Morita, M., Yamanaka, S., Ichikawa, T. et al. Supersulphides provide airway protection in viral and chronic lung diseases. Nat Commun. 14, 4476 (2023).
EPA. Determination of carbonyl compounds. Environ. Prot. Agency 0100, 1–34 (1996).
Kanda, Y. Investigation of the freely available easy-to-use software 'EZR' for medical statistics. Bone Marrow Transplant. 48, 452–458 (2013).
Brierley, J.D., Gospodarowicz, M.K., Wittekind, C., editors. TNM Classification of Malignant Tumours. 8th ed. Oxford: Wiley-Blackwell; 2017.
Moritani, K., Takeshita, T., Shibata, Y., Ninomiya, T., Kiyohara, Y., Yamashita, Y. Acetaldehyde production by major oral microbes. Oral Dis. 21, 748–754 (2015).
Zhu, B., Wei, Y., Zheng, X., Tang, C., Xie, X., Lv, Y. Enhanced breath aldehyde analysis by dual-membrane-assisted charge tagging, enrichment, and onsite elution NanoESI-MS. Anal Chem. 97, 8021–8029 (2025)
Burgos-Barragan, G., Wit, N., Meiser, J., Dingler, F.A., Pietzke, M., Mulderrig, L. et al. Mammals divert endogenous genotoxic formaldehyde into one-carbon metabolism. Nature 548, 549–554 (2017).
Walport, L.J., Hopkinson, R.J., Schofield, C.J. Mechanisms of human histone and nucleic acid demethylases. Curr. Opin. Chem. Biol. 16, 525–534 (2012).
Maddocks, O.D., Berkers, C.R., Mason, S.M., Zheng, L., Blyth, K., Gottlieb, E. et al. Serine starvation induces stress and p53-dependent metabolic remodelling in cancer cells. Nature 493, 542–546 (2013).
Snell, K. Enzymes of serine metabolism in normal and neoplastic rat tissues. Biochim. Biophys. Acta 843, 276–281 (1985).
Snell, K., Weber, G. Enzymic imbalance in serine metabolism in rat hepatomas. Biochem. J. 233, 617–620 (1986).
Newman, A.C., Maddocks, O.D.K. One-carbon metabolism in cancer. Br. J. Cancer 116, 1499–1504 (2017).
Nakamura, J., Holley, D.W., Kawamoto, T., Bultman, S.J. The failure of two major formaldehyde catabolism enzymes (ADH5 and ALDH2) leads to partial synthetic lethality in C57BL/6 mice. Genes Environ. 42, 21 (2020).
Kasamatsu, S., Nishimura, A., Alam, M.M., Morita, M., Shimoda, K., Matsunaga, T. et al. Supersulfide catalysis for nitric oxide and aldehyde metabolism. Sci. Adv. 9, eadg8631 (2023).
Baraona, E., Di Padova, C., Tabasco, J., Lieber, C.S. Red blood cells: a new major modality for acetaldehyde transport from liver to other tissues. Life Sci. 40, 253–258 (1987).

Table 1. Background characteristics of healthy controls and patients with esophageal cancer

	EC n=119	HC n=57	p-value
Age, years (median, range)	67 (40–82)	48 (24–63)	<0.0001
Sex (male/female)	100/19	24/33	<0.0001
BMI, kg/m² (median, range)	22.4 (14.3–31.6)	21.5 (16.6–30.3)	0.27
Smoker (±)	98/21	24/33	<0.0001
Never/former/current smoker	19/79/21	33/17/7	<0.0001
Brinkman Index (median, range)	500 (0–2100)	0 (0–1000)	<0.0001
Alcohol consumption (±)	102/17	37/20	0.0027
Flushing (±)	64/37	-	-
PS (0/1)	109/10	-	-
Histology (SCC/AC/others)	91/23/5	-	-
Stage (Ⅰ/Ⅱ/Ⅲ/Ⅳ)	27/21/47/24	-	-
cT (1/2/3/4)	31/23/62/3	-	-
cN (0/1/2/3)	41/49/28/1	-	-
cM (0/1)	100/19	-	-
Location (Ce/Ut/Mt/Lt/Jz/G)	1/18/43/35/21/1	-	-

Abbreviations: EC, esophageal cancer; HC, healthy controls; SCC, squamous cell carcinoma; AC, adenocarcinoma.

Table 2. Multivariate analysis of acetaldehyde and formaldehyde concentrations in EBC

Acetaldehyde
Characteristics	Participants		Univariate analysis	Multivariate analysis
	EC n=106	HC n=47	p-value	OR	95% CI	p-value
Age, years (median, >61 and ≤61)	67	43	<0.0001	50.9	10.4–249.2	<0.0001
Sex (male/female)	88/18	18/29	<0.0001	2.91	0.92–9.25	0.070
Smoker (±)	86/20	19/28	<0.0001	2.14	0.67–6.87	0.20
Drinking habit (±)	92/14	31/16	0.004	2.21	0.64–7.60	0.21
Acetaldehyde in EBC (µM; >1.74 and ≤1.74)	1.88	1.31	<0.0001	4.12	1.45–11.67	0.0078
Formaldehyde
Characteristics	Subjects		Univariate analysis	Multivariate analysis
	EC n=106	HC n=47	p-value	OR	95% CI	p-value
Age, years (median, >61 and ≤61)	67	43	<0.0001	52.9	10.4–268.2	<0.0001
Sex (male/female)	88/18	18/29	<0.0001	2.91	0.88–9.60	0.080
Smoker (±)	86/20	19/28	<0.0001	1.42	0.42–4.79	0.57
Drinking habit (±)	92/14	31/16	0.004	2.23	0.63–7.93	0.22
Formaldehyde in EBC (µM; >7.50 and ≤7.50)	8.54	4.53	<0.0001	6.60	2.16–20.2	0.0009

Abbreviations: EC, esophageal cancer; HC, healthy controls.

No competing interests reported.

Supplementaryinformation.docx

Noninvasive Breathomics for Developing Carbonyl Biomarkers in Esophageal Cancer

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Background

Methods

Results

Discussion

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1