放射治疗后呼气中挥发性有机物的分析

Analysis of volatile organic compounds in exhaled breath after radiotherapy 放射治疗后呼气中挥发性有机物的分析 1 2 first-author 1 corresp Xue ZOU, xzou@cmpt.ac.cn Xue ZOU, xzou@cmpt.ac.cn xzou@cmpt.ac.cn https://orcid.org/0000-0001-9540-6764 1 2 1 2 1 2 1 1 1 3 1 1 1 3 corresp corresp Xue ZOU, xzou@cmpt.ac.cn Xue ZOU, xzou@cmpt.ac.cn Pei WANG, 61775618@qq.com Pei WANG, 61775618@qq.com 61775618@qq.com https://orcid.org/0000-0002-8613-6480 1 3 1 Anhui Province Key Laboratory of Medical Physics and Technology, Institute of Health and Medical Technology, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei 230031, China Anhui Province Key Laboratory of Medical Physics and Technology, Institute of Health and Medical Technology, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei 230031, China University of Science and Technology of China, Hefei 230026, China University of Science and Technology of China, Hefei 230026, China Department of Laboratory Medicine, Hefei Cancer Hospital, Chinese Academy of Sciences, Hefei 230031, China Department of Laboratory Medicine, Hefei Cancer Hospital, Chinese Academy of Sciences, Hefei 230031, China 放射治疗具有一定风险，主要是因为其在杀死癌细胞的同时可对正常细胞造成严重损伤。呼气中的挥发性有机化合物（VOCs）可以提供人体内生理以及病理相关过程的信息，具有监测机体损伤的潜力。本研究收集了10名肺癌患者的110份呼气样本，使用固相微萃取气相色谱-质谱技术（SPME-GC/MS）进行检测。通过热图分析可发现28种VOCs（包括芳香族化合物、烷烃、醛类、酸类、酮类、醇类和酯类）在放疗后6小时内发生规律性的变化。通过对放疗前后0.5小时的呼气样本进行Mann-Whitney U 检验和正交偏最小二乘判别分析（OPLS-DA），丙酸、三乙酸甘油酯和吲哚可被确定为放疗的呼气标志物。所建立的OPLS-DA模型的预测优度参数 Q 2 为0.53，显示出良好的预测能力。综上所述，呼气VOCs可用于监测放射治疗造成的机体损伤，有助于制定个体化治疗方案。 放射治疗 呼气 挥发性有机物 气相色谱质谱 生物标志物 Radiotherapy uses high-energy X-rays or other particles to destroy cancer cells and medical practitioners have used this approach extensively for cancer treatment ( Hachadorian et al., 2020 Hachadorian et al., 2020 ). However, it is accompanied by risks because it seriously harms normal cells while killing cancer cells. The side effects can lower cancer patients' quality of life and are very unpredictable due to individual differences ( Bentzen, 2006 Bentzen, 2006 ). Therefore, it is essential to assess a patient's body damage after radiotherapy to formulate an individualized recovery treatment plan. Exhaled volatile organic compounds (VOCs) can be changed by radiotherapy and thus used for medical diagnosis ( Vaks et al., 2012 Vaks et al., 2012 ). During treatment, high-energy X-rays can induce apoptosis; meanwhile, cell membranes are damaged due to lipid peroxidation, converting unsaturated fatty acids into volatile metabolites ( Losada-Barreiro and Bravo-Díaz, 2017 Losada-Barreiro and Bravo-Díaz, 2017 ). At the same time, radiotherapy oxidizes water, resulting in reactive oxygen species (ROS) that can increase the epithelial permeability of pulmonary alveoli, enabling the respiratory system to exhale volatile metabolites ( Davidovich et al., 2013 Davidovich et al., 2013 ; Popa et al., 2020 Popa et al., 2020 ). These exhaled VOCs can be used to monitor body damage caused by radiotherapy. In our study, 110 breath samples were obtained from ten lung cancer patients who were receiving radiotherapy. The gender, age, tumor location, cancer stage, and daily dose of the patients are shown in Table 1 Table 1 . We stored the samples in our lab-made glass bottles away from light to avoid interference from sampling bags or tubes. The VOCs in the samples were detected using solid phase micro-extraction-gas chromatography/mass spectrometry (SPME-GC/MS). Ethanol was used as the internal standard. We evaluated the changing trends of exhaled VOCs and identified the radiotherapy biomarkers through multivariate statistical analysis. The daily dose was 2 Gy. No. Gender Age (years) Tumor location Cancer stage 1 Male 72 Upper right lung T3N3M0 IIIC 2 Female 58 Upper right lung T2N3M0 IIIB 3 Male 71 Left lung T1bN3M0 IIIB 4 Male 66 Lower right lung T2N2M1c IVB 5 Male 65 Right lung T2N2M1b IVA 6 Male 70 Right lung T2N2M0 IIIA 7 Male 49 Right lung T4N3M0 IIIC 8 Male 56 Lower left lung T4N2M0 IIIB 9 Male 62 Upper left lung T4N0M0 IIIA 10 Male 72 Upper right lung T4N3M1c IVB As shown in the heatmap in Fig. 1 1 , 28 kinds of VOCs regularly changed after radiotherapy. The data are arranged horizontally in five columns, representing the concentrations of exhaled VOCs 0.5 h before and 0.5, 2.0, 4.0, and 6.0 h after radiotherapy. We discovered three change trends which occurred together: (1) increasing in the 6.0 h after treatment, including hexanal, butyric acid, and propionic acid; (2) decreasing in the 6.0 h after treatment, including triacetin, indene, cyclohexanone, methylbenzene, ethylbenzene, 1,‍3-dimethylbenzene, and 1,‍4-dimethylbenzene; (3) increasing followed by decreasing, including acetic acid, valeric acid, hexanoic acid, nonanoic acid, acetone, decanal, 2-methylbutanoic acid, 2-ethylhexanal, 2-ethylhexanol, L-menthol, dimethyl adipate, dodecane, tetradecane, caprolactam, p -cresol, 6-methylheptan-2-one, 6-methyl-5-hepten-2-one, and indole. However, for propionic acid in (1), the concentration before treatment was higher than that 0.5 h after treatment, which did not entirely follow the first change trend. Meanwhile, for triacetin in (2), the concentration abnormally increased 2.0 h after treatment, which was not entirely in line with the second change trend. The 28 VOCs could be classified into aromatic compounds, alkanes, aldehydes, acids, ketones, alcohols, and esters. Aromatic compounds, alkanes, and aldehydes are biomarkers of lipid oxidation of polyunsaturated fatty acids in cell membranes ( Phillips et al., 2013 Phillips et al., 2013 ). The increase of alkanes and aldehydes may be due to lipid oxidation accelerated by radiotherapy or irradiation ( Phillips et al., 2013 Phillips et al., 2013 ). However, the reason for the decreased concentration of aromatic compounds in this study was unclear. Meanwhile, acids, ketones, alcohols, and esters are products of the reaction between glutathionesulfonic acid and glucose, or between glutathione and glucose. Nakajima (2015) Nakajima (2015) observed that glutathionesulfonic acid and alutathione increased in the liver after radiation exposure to protect the organ against oxidative stress, which was inducible even by low-dose radiation exposure. We analyzed the breath samples 0.5 h before and after treatment to identify the exhaled biomarkers of radiotherapy. Based on the Mann-Whitney U test, seven kinds of VOCs ( P < 0.05) were significantly different: acetone, propionic acid, 2-methylbutanoic acid, indene, dimethyl adipate, triacetin, and indole. We selected these VOCs for the orthogonal partial least squares-discriminant analysis (OPLS-DA) on SIMCA-P14.1. Fig. 2 Fig. 2 a presents the corresponding integral matrix; the integral values are concentrated in the T2 ellipse of the scatter plot at a 95% confidence interval. The before-and-after breath samples exhibited a promising separation trend. R 2 X (cum), which we used to evaluate the quality of the OPLS-DA model, was 0.912. On the other hand, R 2 Y (cum), which assessed the proportion of variance of the response variable, was 0.631. Meanwhile, the statistical value Q 2 (cum) calculated from sevenfold cross-validation described the model's predictive ability. In this model, Q 2 =0.53 > 0.50, indicating that the original model had a high predictive capability. The supervised pattern-recognition method tended to generate an overfitting phenomenon while enlarging the differences between the groups. Therefore, it was necessary to apply the permutation test to prove the model's validity. We obtained validation plots from 200 permutation tests. In Fig. 2 Fig. 2 b, R 2 and Q 2 regression lines intersect the left Y -axis. The R 2 Y (cum) and Q 2 Y (cum) values of the original OPLS-DA model were more significant than all R 2 Y and Q 2 Y values of the permutated models, thus validating the original model's suitability and validity. The respective intercepts of R 2 Y and Q 2 Y were 0.0472 and -‍0.1900 (both of which were less than 0.4), signifying a non-overfitted model. The goal of determining the variable importance in projection (VIP) scores was to select the VOCs that most significantly contributed to the OPLS-DA model, revealing three determinant VOCs with VIP scores greater than 1.0: propionic acid, triacetin, and indole. We identified these VOCs as the biomarkers of radiotherapy. Fig. S1 presents the quantitative analysis results for the three biomarkers. Meanwhile, Fig. 3 Fig. 3 illustrates the median concentration values and interquartile ranges (IQRs) of the three characteristic VOCs. The concentrations of propionic acid and triacetin in the exhaled breath of subjects decreased after radiotherapy (Figs. 3a 3a and 3b 3b ), while that of indole increased ( Fig. 3 Fig. 3 c). Exhaled propionic acid is a potential lung cancer biomarker produced from lung cancer cells ( Wang et al., 2014 Wang et al., 2014 ). It can also originate from the fermentation of undigested carbohydrates by intestinal anaerobes ( Lin, 2004 Lin, 2004 ). From the heatmap in Fig. 1 Fig. 1 , it is evident that its concentration immediately decreased after radiotherapy before gradually increasing. Presumably, the initial decrease was due to lung cancer cell death caused by the treatment, while the subsequent increase was due to the rising number of intestinal anaerobes; intestinal anaerobes multiplied because of the intestinal inflammation caused by radiotherapy ( McKelvey et al., 2018 McKelvey et al., 2018 ). Nonspecific lipases and esterases can hydrolyze triacetin to release glycerol and acetic acid. In this study, triacetin concentration decreased after treatment. Glycerol can undergo conversion to glyceraldehyde-3-phosphate for glycolysis, while acetic acid can produce acetyl-coenzyme A (acetyl-CoA) ( Reisenauer et al., 2011 Reisenauer et al., 2011 ; Long et al., 2015 Long et al., 2015 ). Radiotherapy can increase the activity of the hypoxia-inducible factor 1 (HIF-1) transcription factor ( Zhong et al., 2013 Zhong et al., 2013 ), which then stimulates glycolysis and consumes glycerol ( Semenza, 2011 Semenza, 2011 ). Additionally, radiotherapy can cause the overexpression of cytoplasmic acetyl-CoA synthetase in tumors ( Chung et al., 2017 Chung et al., 2017 ), which consumes acetic acid. Thus, it can be inferred that glycerol and acetic acid consumption causes a decrease in triacetin concentration. Indole is related to tryptophan metabolism in intestinal flora ( de Vietro et al., 2020 de Vietro et al., 2020 ). In this study, its concentration immediately increased after radiotherapy before gradually decreasing. Moreover, it can bind to blood albumin ( Gambhir et al., 1975 Gambhir et al., 1975 ). However, the molecular conformation of albumin can change under ionizing radiation ( Kirilova et al., 2011 Kirilova et al., 2011 ). Therefore, we assumed that the initially increased concentration was caused by a decrease in the ability of indole to bind to albumin. In another study, indole produced by intestinal flora decreased when exposed to radiation ( Broin et al., 2015 Broin et al., 2015 ). This may be why we found that exhaled indole concentration decreased 2.0 h after radiotherapy. Notably, the time intervals between the most significant concentration differences for the 28 VOCs before and after radiotherapy were different ( Fig. 1 Fig. 1 ). For instance, the time interval for dodecane was 2.0 h, while that for methylbenzene was 6.0 h. This situation raises the question: what is the appropriate time point to sample exhaled gases? We suggest recruiting more volunteers and sampling breath more frequently, which we plan to undertake in subsequent studies. In summary, we found that 28 kinds of VOCs changed regularly after radiotherapy, including aromatic compounds, alkanes, aldehydes, acids, ketones, alcohols, and esters. Using the Mann-Whitney U test and OPLS-DA analysis, we identified three radiotherapy biomarkers. Furthermore, we established a discriminant model with a Q 2 value of 0.53, indicating good predictive ability. Based on heatmap analysis, we pinpointed a problem regarding the appropriate time point to sample exhaled gases. The time intervals between the most significant concentration differences before and after radiotherapy for the 28 VOCs were different. This context requires further study involving more breath samples and higher sampling frequency. Materials and methods 1 Detailed methods are provided in the electronic supplementary materials of this paper. References Bentzen SM , 2006 . Preventing or reducing late side effects of radiation therapy: radiobiology meets molecular pathology . Nat Rev Cancer , 6 ( 9 ): 702 - 713 . https://doi.org/10.1038/nrc1950 https://doi.org/10.1038/nrc1950 Preventing or reducing late side effects of radiation therapy: radiobiology meets molecular pathology Broin PÓ , Vaitheesvaran B , Saha S , et al. , 2015 . Intestinal microbiota-derived metabolomic blood plasma markers for prior radiation injury . Int J Radiat Oncol Biol Phys , 91 ( 2 ): 360 - 367 . https://doi.org/10.1016/j.ijrobp.2014.10.023 https://doi.org/10.1016/j.ijrobp.2014.10.023 Intestinal microbiota-derived metabolomic blood plasma markers for prior radiation injury Chung YH , Tsai CK , Wang CC , et al. , 2017 . Early response monitoring following radiation therapy by using [ 18 F]FDG and [ 11 C]acetate PET in prostate cancer xenograft model with metabolomics corroboration . Molecules , 22 ( 11 ): 1946 . https://doi.org/10.3390/molecules22111946 https://doi.org/10.3390/molecules22111946 Early response monitoring following radiation therapy by using [18F]FDG and [11C]acetate PET in prostate cancer xenograft model with metabolomics corroboration Davidovich N , DiPaolo BC , Lawrence GG , et al. , 2013 . Cyclic stretch-induced oxidative stress increases pulmonary alveolar epithelial permeability . Am J Respir Cell Mol Biol , 49 ( 1 ): 156 - 164 . https://doi.org/10.1165/rcmb.2012-0252OC https://doi.org/10.1165/rcmb.2012-0252OC Cyclic stretch-induced oxidative stress increases pulmonary alveolar epithelial permeability de Vietro N , Aresta A , Rotelli MT , et al. , 2020 . Relationship between cancer tissue derived and exhaled volatile organic compound from colorectal cancer patients . Preliminary results. J Pharm Biomed Anal , 180 : 113055 . https://doi.org/10.1016/j.jpba.2019.113055 https://doi.org/10.1016/j.jpba.2019.113055 Relationship between cancer tissue derived and exhaled volatile organic compound from colorectal cancer patients Gambhir KK , McMenamy RH , Watson F , 1975 . Positions in human serum albumin which involve the indole binding site . Sequence of 107-residue fragment. J Biol Chem , 250 ( 17 ): 6711 - 6719 . https://doi.org/10.1016/S0021-9258(19)40991-5 https://doi.org/10.1016/S0021-9258(19)40991-5 Positions in human serum albumin which involve the indole binding site Hachadorian RL , Bruza P , Jermyn M , et al. , 2020 . Imaging radiation dose in breast radiotherapy by X-ray CT calibration of Cherenkov light . Nat Commun , 11 : 2298 . https://doi.org/10.1038/s41467-020-16031-z https://doi.org/10.1038/s41467-020-16031-z Imaging radiation dose in breast radiotherapy by X-ray CT calibration of Cherenkov light Kirilova EM , Kalnina I , Zvagule T , et al. , 2011 . Fluorescent study of human blood plasma albumin alterations induced by ionizing radiation . J Fluoresc , 21 ( 3 ): 923 - 927 . https://doi.org/10.1007/s10895-010-0608-2 https://doi.org/10.1007/s10895-010-0608-2 Fluorescent study of human blood plasma albumin alterations induced by ionizing radiation Lin J , 2004 . Too much short chain fatty acids cause neonatal necrotizing enterocolitis . Med Hypotheses , 62 ( 2 ): 291 - 293 . https://doi.org/10.1016/S0306-9877(03)00333-5 https://doi.org/10.1016/S0306-9877(03)00333-5 Too much short chain fatty acids cause neonatal necrotizing enterocolitis Long PM , Tighe SW , Driscoll HE , et al. , 2015 . Acetate supplementation as a means of inducing glioblastoma stem-like cell growth arrest . J Cell Physiol , 230 ( 8 ): 1929 - 1943 . https://doi.org/10.1002/jcp.24927 https://doi.org/10.1002/jcp.24927 Acetate supplementation as a means of inducing glioblastoma stem-like cell growth arrest Losada-Barreiro S , Bravo-Díaz C , 2017 . Free radicals and polyphenols: the redox chemistry of neurodegenerative diseases . Eur J Med Chem , 133 : 379 - 402 . https://doi.org/10.1016/j.ejmech.2017.03.061 https://doi.org/10.1016/j.ejmech.2017.03.061 Free radicals and polyphenols: the redox chemistry of neurodegenerative diseases McKelvey KJ , Hudson AL , Back M , et al. , 2018 . Radiation, inflammation and the immune response in cancer . Mamm Genome , 29 ( 11 ): 843 - 865 . https://doi.org/10.1007/s00335-018-9777-0 https://doi.org/10.1007/s00335-018-9777-0 Radiation, inflammation and the immune response in cancer Nakajima T , 2015 . Roles of sulfur metabolism and rhodanese in detoxification and anti-oxidative stress functions in the liver: responses to radiation exposure . Med Sci Monit , 21 : 1721 - 1725 . https://doi.org/10.12659/MSM.893234 https://doi.org/10.12659/MSM.893234 Roles of sulfur metabolism and rhodanese in detoxification and anti-oxidative stress functions in the liver: responses to radiation exposure Phillips M , Byrnes R , Cataneo RN , et al. , 2013 . Detection of volatile biomarkers of therapeutic radiation in breath . J Breath Res , 7 ( 3 ): 036002 . https://doi.org/10.1088/1752-7155/7/3/036002 https://doi.org/10.1088/1752-7155/7/3/036002 Detection of volatile biomarkers of therapeutic radiation in breath Popa C , Maia Bratu A , Petrus M , 2020 . Special issue “applications of photoacoustic spectroscopy” . Molecules , 25 ( 21 ): 5116 . https://doi.org/10.3390/molecules25215116 https://doi.org/10.3390/molecules25215116 Special issue “applications of photoacoustic spectroscopy” Reisenauer CJ , Bhatt DP , Mitteness DJ , et al. , 2011 . Acetate supplementation attenuates lipopolysaccharide-induced neuroinflammation . J Neurochem , 117 ( 2 ): 264 - 274 . https://doi.org/10.1111/j.1471-4159.2011.07198.x https://doi.org/10.1111/j.1471-4159.2011.07198.x Acetate supplementation attenuates lipopolysaccharide-induced neuroinflammation Semenza GL , 2011 . Oxygen sensing, homeostasis, and disease . N Engl J Med , 365 ( 6 ): 537 - 547 . https://doi.org/10.1056/nejmra1011165 https://doi.org/10.1056/nejmra1011165 Oxygen sensing, homeostasis, and disease Vaks VL , Domracheva EG , Sobakinskaya EA , et al. , 2012 . Using the methods and facilities of nonsteady-state spectroscopy of the subterahertz and terahertz frequency ranges for noninvasive medical diagnosis . J Opt Technol , 79 ( 2 ): 66 - 69 . https://doi.org/10.1364/jot.79.000066 https://doi.org/10.1364/jot.79.000066 Using the methods and facilities of nonsteady-state spectroscopy of the subterahertz and terahertz frequency ranges for noninvasive medical diagnosis Wang CS , Dong R , Wang XY , et al. , 2014 . Exhaled volatile organic compounds as lung cancer biomarkers during one-lung ventilation . Sci Rep , 4 : 7312 . https://doi.org/10. 1038/srep07312 https://doi.org/10.1038/srep07312 Exhaled volatile organic compounds as lung cancer biomarkers during one-lung ventilation Zhong J , Rajaram N , Brizel DM , et al. , 2013 . Radiation induces aerobic glycolysis through reactive oxygen species . Radiother Oncol , 106 ( 3 ): 390 - 396 . https://doi.org/10.1016/j.radonc. 2013.02.013 https://doi.org/10.1016/j.radonc.2013.02.013 Radiation induces aerobic glycolysis through reactive oxygen species Dianlong GE, Pei WANG, Xue ZOU, and Chengyin SHEN contributed to the conception and design of this study. Dianlong GE, Yajing CHU, Jijuan ZHOU, Wei XU, Yue LIU, and Qiangling ZHANG performed the experiments. Dianlong GE, Yan LU, Lei XIA, and Aiyue LI performed the data analysis. Dianlong GE wrote the manuscript. Chaoqun HUANG, Pei WANG, Xue ZOU, Chengyin SHEN, and Yannan CHU revised the manuscript. All authors have read and approved the final manuscript, and therefore, have full access to all the data in the study and take responsibility for the integrity and security of the data. This research was supported by the National Natural Science Foundation of China (Nos. 21876176, 21777163, 21705152, and 22076190), the Youth Innovation Promotion Association, Chinese Academy of Sciences (No. 2019432), the Presidential Foundation of Hefei Institute of Physical Sciences of Chinese Academy of Sciences (No. YZJJZX202009), the Functional Development Program of Instrument and Equipment in Chinese Academy of Sciences (No. Y9BS0C1291), the Anhui Provincial Key R&D Program (No. 202104d07020003), and the Joint Fund of the Second Affiliated Hospital of Anhui Medical University and the Center of Medical Physics and Technology of Hefei Institute of Physical Sciences of the Chinese Academy of Sciences (No. LHJJ2020006). 10.1631/jzus.B2100447 1673-1581（2022）02-0153-05 This research was supported by the National Natural Science Foundation of China (Nos. 21876176, 21777163, 21705152, and 22076190), the Youth Innovation Promotion Association, Chinese Academy of Sciences (No. 2019432), the Presidential Foundation of Hefei Institute of Physical Sciences of Chinese Academy of Sciences (No. YZJJZX202009), the Functional Development Program of Instrument and Equipment in Chinese Academy of Sciences (No. Y9BS0C1291), the Anhui Provincial Key R&D Program (No. 202104d07020003), and the Joint Fund of the Second Affiliated Hospital of Anhui Medical University and the Center of Medical Physics and Technology of Hefei Institute of Physical Sciences of the Chinese Academy of Sciences (No. LHJJ2020006). 2021-05-23 2021-08-15 2022-02-15 2022-04-19 2021-12-17 Copyright © Zhejiang University and Springer-Verlag GmbH Germany, part of Springer Nature 2022 2022 浙江大学出版社 Dianlong GE, Xue ZOU, Yajing CHU, Jijuan ZHOU, Wei XU, Yue LIU, Qiangling ZHANG, Yan LU, Lei XIA, Aiyue LI, Chaoqun HUANG, Pei WANG, Chengyin SHEN, and Yannan CHU declare that they have no conflict of interest. All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000(5). Informed consent was obtained from all patients for being included in the study. Compliance with ethics guidelines Supplementary information 1 Materials and methods; Fig. S1 浙江大学学报（英文版）（B辑：生物医学和生物技术） 2 23