Aldehyde breath test as a disease marker in patients with esophageal and hypopharyngeal squamous cell carcinoma

Background Patients with inactive aldehyde dehydrogenase 2 (ALDH2) are at high risk for esophageal squamous cell carcinoma (ESCC) and hypopharyngeal squamous cell carcinoma (HPSCC). The aldehyde breath test (ABT) may demonstrate ALDH2 gene polymorphisms. We evaluated the usefulness of the ABT in patients with ESCC and HPSCC.


Results
We found that the proportion of individuals in the SCC group with inactive ALDH2 (A/E ratio 23.3) was signi cantly higher than that in the HC group (p = 0.035), as was the A/E ratio (p < 0.001). Also, the proportion of individuals with inactive ALDH2 in the multiple subgroup was signi cantly higher than that in single subgroup (p = 0.015), as was the A/E ratio (p = 0.008).

Conclusions
In conclusion, ABT could be a useful tool to identify ESCC and HPSCC and detect multiple or double carcinomas.

Background
Hypopharyngeal squamous cell carcinoma (HPSCC) and esophageal squamous cell carcinoma (ESCC) are two of the deadliest cancers worldwide [1][2][3]. HPSCC and ESCC are often advanced when detected, the prognosis is relatively poor prognosis. [4,5]. Acetaldehyde is the rst metabolite of ethanol and a de nite carcinogen for organs affected by the abovementioned cancers [2,3]. Field cancerization is a biological process in which large areas of cells at a tissue surface or within an organ are affected by one or more carcinogenic alterations. The results of one study indicated that eld cancerization might be due to the accumulation of acetaldehyde rather than alcohol itself [6].
Aoyama et al. [22] recently reported the development of a new breath study, the acetaldehyde breath test (ABT), which can measure very low levels of acetaldehyde and alcohol quantitatively after ingestion of a very small amount of alcohol (100 mL of 0.5% ethanol) and can accurately and rapidly identify ALDH2*2 allele carriers. According to Aoyama et al., the ratio of breath acetaldehyde level to breath ethanol level (A/E ratio) could identify ALDH2*2 allele carriers very accurately; the A/E ratio in carriers of the ALDH2*2 allele was signi cantly higher than that in participants who did not have the ALDH2*2. In addition, the accuracy, sensitivity, and speci city of the A/E ratio for identifying carriers of the ALDH2*2 allele were 96.4%, 100%, and 92.5%, respectively [22].
However, there is no report on the clinical signi cance of ABT for patients with ESCC or HPSCC. In recent years, as a result of advances in endoscopic technology, many cases of super cial ESCC and super cial HPSCC have been discovered and treated with methods of endoscopic resection, such as endoscopic submucosal dissection (ESD) [23,24]. These cases may re ect the genesis of full-blown ESCC or HPSCC. In this study, we analyzed the clinical signi cance and usefulness of ABT as a disease marker for super cial ESCC and super cial HPSCC in patients treated with ESD.

Patients and methods
Two groups of subjects were compared: patients with squamous cell carcinoma (SCC group) and healthy controls (HC group). The patients in the SCC group had undergone ESD treatment for super cial ESCC or super cial HPSCC, or both, at the Kagoshima University Hospital, Kagoshima, Japan, between December 2016 and August 2018. The criteria for inclusion in the SCC group were (1) pathological diagnosis of super cial ESCC or super cial HPSCC in the resected specimens, (2) the patient's written informed consent, and (3) the patient's ability to provide breath samples. Exclusion criteria were (1) history of surgical resection of any part of the upper gastrointestinal tract, (2) history of alcohol allergy, and (3) being younger than 20 years of age.
The patients included in the HC group were examined in a medical checkup that included esophagogastroduodenoscopy and ABT at Izumi General Medical Center, Kagoshima, Japan. The criteria of inclusion in the HC group were (1) age of 20 years or older, (2) the patient's written informed consent, and (3) the patient's ability to provide breath samples. Exclusion criteria were (1) history of surgical resection of any part of the upper gastrointestinal tract, (2) history of alcohol allergy, (3) age below 20 years, (4) detection of cancer of the upper gastrointestinal tract by esophagogastroduodenoscopy, and (5) current pregnancy.
A total of 100 participants were included in the SCC group and 275 in the HC group (Fig. 1). Table 1 lists the characteristics of the participants in the SCC and HC groups. Of the patients in the SCC group, 64 had super cial ESCC, 14 had super cial HPSCC, and 22 had both conditions. Six patients were eventually excluded because of history of head and neck or esophageal cancer; 3 had undergone oral resection of pharyngeal cancer, 1 had undergone radiotherapy for pharyngeal cancer, 1 had undergone radiotherapy for vocal cord cancer, and 1 had undergone chemoradiotherapy for tongue cancer. The remaining 94 patients were subdivided into the "single" subgroup, which comprised 63 patients in whom a single lesion was initially treated with ESD, and the "multiple" subgroup, which comprised 31 patients in whom multiple lesions were initially treated with ESD.   The "single" subgroup comprised patients in whom a single lesion was initially treated with ESD; the "multiple" subgroup comprised patients in whom multiple lesions were initially treated with endoscopic submucosal dissection (ESD).
HRA, health risk appraisal.  The "single" subgroup comprised patients in whom a single lesion was initially treated with ESD; the "multiple" subgroup comprised patients in whom multiple lesions were initially treated with endoscopic submucosal dissection (ESD).
HRA, health risk appraisal. The "single" subgroup comprised patients in whom a single lesion was initially treated with ESD; the "multiple" subgroup comprised patients in whom multiple lesions were initially treated with endoscopic submucosal dissection (ESD). A/E, acetaldehyde-to-ethanol.
History regarding alcohol consumption and smoking was carefully documented in health risk appraisal (HRA) score [25], and a table of daily alcohol consumption was compiled for each age (Additional le 1).
In addition, all patients were surveyed in detail about alcohol ushing and frequency of toothbrushing. Alcohol ushing, in which the face turns red after drinking a small amount of alcohol, has been reported to be associated with esophageal cancer [25], and the frequency of toothbrushing was recently reported to be correlated with risk for head and neck cancer [26]. We scored frequency of toothbrushing as follows: 0 points for brushing almost every morning, noon, and evening; 1 point for brushing often; and 2 points for rarely or never brushing.
Squamous dysplasia, a preneoplastic lesion, is identi ed easily on Lugol chromoendoscopy as a Lugolvoiding lesion (LVL) [27]. For all patients in the SCC group, the presence of LVLs on Lugol chromoendoscopy of esophagus was documented in accordance with the report of Katada et al. as grades A (no LVLs), B (1-9 LVLs), and C ( 10 LVLs; Additional le 2) [27].

Acetaldehyde breath test
Participants were asked to drink 100 mL of 0.5% ethanol in one draught after at least 12 h of fasting and abstinence. The 0.5% ethanol was made with vodka, which contains little acetaldehyde [28]. Breath samples were collected with dedicated gas bags immediately before and 1 min after participants drank the alcohol. We used the dedicated gas bags to collect the end-tidal gas [22]. The gas bags were made of vinyl alcohol polymer and uniquely shaped to remove the gas derived from the physiological dead space. Approximately 100 mL of end-tidal gas can be collected with one breath into the bag. In this study, the breath was collected into these bags at standard temperature with air conditioning.

Measurement of breath acetaldehyde and ethanol levels
Breath acetaldehyde and ethanol levels were measured by highly sensitive gas chromatography [29] performed using the AERoChrome (Nissha FIS, Inc., Osaka, Japan), which can measure acetaldehyde and ethanol content. This device can also calculate the A/E ratio for each individual 1 min after ethanol ingestion. With the cutoff value of the A/E ratio set to 23.3, a previous study revealed that the AERoChrome was able to determine the presence of ALDH2*1 (active ALDH2) and ALDH2*2 (inactive ALDH2) with an accuracy of 96.4% [22].

Statistical analyses
The absolute and relative frequencies for qualitative variables were calculated for the SCC and HC groups.

SCC vs HC groups
We compared the SCC and HC groups with regard to characteristics and risk factors for carcinogenesis, such as alcohol consumption, smoking, measured A/E ratio, and proportion of subjects with inactive ALDH2 (A/E ratio ≥ 23.3). Patients in the SCC group were signi cantly older and had a higher proportion of men than did the HC group. In addition, HRA scores of the SCC group were higher than those of the HC group. There was no signi cant difference with regard to smoking and alcohol ushing. In the SCC group, the frequency of toothbrushing was lower (Table 1). Univariate analysis also revealed that the proportion of individuals with inactive ALDH2 (A/E ratio ≥ 23.3) was signi cantly higher in the SCC group (43%) than in the HC group (31.3%; p = 0.035). In addition, the A/E ratio in the SCC group (24.0 ± 16.5) was signi cantly higher than that in the HC group (18.8 ± 16.2; p < 0.001; Fig. 2). The A/E ratio was extracted as a factor contributing to carcinogenesis in multivariate analysis, together with age, alcohol consumption, and HRA score ( Table 2).

Single versus multiple
We compared differences in the abovementioned characteristics and carcinogenic risk factors between the single and multiple subgroups. There were no signi cant differences with regard to gender, age, alcohol consumption, smoking, alcohol ushing, HRA score, or toothbrushing frequency. The proportion of patients in the multiple subgroup with grade C LVLs was signi cantly higher than that in the single subgroup. There was no difference between the two groups with regard to other organ cancers ( Table 3). The proportion of individuals with inactive ALDH2 (A/E ratio ≥ 23.3) was signi cantly higher in the multiple subgroup (61.3%) than in the single subgroup (34.9%; p = 0.015). In addition, the A/E ratio in the multiple subgroup (31.3 ± 20.0) was signi cantly higher than that in single subgroup as well (21.2 ± 13.9; p = 0.008; Fig. 3). In multivariate analysis, the A/E ratio and LVL grade were extracted as factors contributing to carcinogenesis in the multiple subgroup (Table 4).

Discussion
In this study, we hypothesized that the ABT, which accurately identi es ALDH2 inactivity noninvasively as a disease marker for super cial ESCC and super cial HPSCC treated with ESD, would exhibit clinical signi cance and usefulness. To support this hypothesis, we demonstrated that ABT was useful for identifying patients with super cial ESCC or super cial HSCC and that the A/E ratio was useful for identifying patients with multiple lesions.
Alcohol metabolic capacity has been found to vary greatly with regard to race, and 35% of the Japanese population [30]. Individuals with inactive ALDH2 who drink heavily are at high risk for ESCC and HPSCC [17][18][19][20][21]. Genetic testing is the most reliable way to identify ALDH2*2 allele carriers; however, it is not suitable for mass screening because it is a time-consuming and cumbersome procedure. The alcohol ushing questionnaire and the ethanol patch test have been considered as alternative diagnostic tools. However, their accuracy is unsatisfactory [20,31]. The ABT could accurately identify ALDH2*2 allele carriers within 8 min; thus, people would be able to know the result shortly after the test [22].
In recent years, early esophageal cancer was reported to have a good prognosis; several studies have shown that the overall rate of 5-year survival in patients with intramucosal ESCC who underwent endoscopic resection was 89-95% [23,32,33]. Furthermore, the cause-speci c rate of 5-year survival among patients with super cial pharyngeal cancer who underwent endoscopic resection was 97% [24].
ESD as a minimally invasive treatment for super cial pharyngeal cancer was reported to be effective and safe [34]. As mentioned previously, early detection of ESCC and HPSCC is very important for long-term survival of patients with esophageal cancer.
According to a recent report, multiple LVLs in the esophagus increase the risk of multiple SCC [34].
However, Lugol chromoendoscopy is invasive and painful or uncomfortable. A major advantage of ABT is its noninvasiveness. We demonstrated that the A/E ratio and LVL grade were extracted as factors contributing to carcinogenesis in the multiple subgroup; the ABT can be used easily in healthy patients at high risk of ESCC and HPSCC. Furthermore, conducting the ABT at a young age in healthy individuals may help prevent ESCC and HPSCC by encouraging these individuals to modify drinking habits.
On the other hand, in patients with ESCC, the incidence of multiple lesions simultaneously is high. In addition, 10-50% of patients with HPSCC also have ESCC [35][36][37][38]. It is extremely important to accurately diagnose HPSCC and ESCC that have occurred at the same time. Our results indicate that ABT, which re ects alcohol metabolic ability, may be useful for identifying such patients.
This study had some limitations. First, because genetic testing is time consuming and cumbersome, it was not conducted in this study. However, Aoyama et al. [22] reported that the A/E ratio could identify ALDH2*2 allele carriers very accurately (in their study, the rate of accuracy was 96.4%). Second, patients who had undergone gastrointestinal surgery, which could have affected the breath ethanol or acetaldehyde levels, were not tested in this study. Third, the subjects of this study were patients who received only endoscopic treatment; we did not include patients who underwent surgery or chemotherapy for esophageal cancer and pharyngeal cancer. In the future, we also need to test the ABT in patients who underwent surgery and chemotherapy.

Conclusion
The results of this study suggest that the ABT is a useful, noninvasive tool for identifying individuals at high risk for pharyngeal and esophageal cancer, as well as for the detection of multiple carcinomas. In the future, the use of ABT may help prevent pharyngeal and esophageal cancer by encouraging individuals to modify their drinking habits.

Supplementary Figure Legends
Additional le 1 Daily alcohol consumption, calculated as the sum of scores A to D for each age.
Additional le 2 Grading and appearance of Lugol-voiding lesions (LVLs). The number of LVLs per endoscopic view was counted, and the grading was divided into three categories. Figure 1 Outline of this study. Flowchart of enrollment for analysis. The "single" subgroup comprised patients in whom a single lesion was initially treated with ESD; the "multiple" subgroup comprised patients in whom multiple lesions were initially treated with endoscopic submucosal dissection (ESD).