Lactate in exhaled breath condensate and its correlation to cancer: challenges, promises and a call for data

Owing to its connection to cancer metabolism, lactate is a compound that has been a focus of interest in field of cancer biochemistry for more than a century. Exhaled breath volatile organic compounds (VOCs) and condensate analyses can identify and monitor volatile and non-VOCs, respectively, present in exhaled breath to gain information about the health state of an individual. This work aims to take into account the possible use of breath lactate measurements in tumor diagnosis and treatment control, to discuss technical barriers to measurement, and to evaluate directions for the future improvement of this technique. The use of exhaled breath condensate (EBC) lactic acid levels in disorders other than cancer is also discussed in brief. Whilst the use of EBC for the detection of lactate in exhaled breath is a promising tool that could be used to monitor and screen for cancer, the reliability and sensitivity of detection are uncertain, and hence its value in clinical practice is still limited. Currently, lactate present in plasma and EBC can only be used as a biomarker for advanced cancer, and therefore it presently has limited differential diagnostic importance and is rather of prognostic value.


Introduction
Based on the seminal work of Otto Warburg and his colleagues in the 1920s, it became evident that the majority of tumor cells produce energy predominantly in their cytosol through a high rate of glucose uptake and glycolysis, followed by enhanced L-lactic acid formation, even if oxygen is abundant [1][2][3]. The phenomenon, appropriately called Wartburg effect, is still under investigation since its precise mechanism and therapeutic implications remain uncertain [4][5][6].
There are two stereoisomers of lactic acid in the human body, D-and L-lactic acid. However, under normal physiological conditions, the plasma concentration of D-lactic acid (∼0.01 mM) is much lower than found for L-lactic acid (∼0.3-1.3 mM) [7,8]. Hence, when referring to lactic acid in the human body in this review it will generally be in reference to the L-lactic form, unless otherwise specified.
Under pathological conditions, the concentrations of lactic acid may be modulated. For example, lactic acid concentration in the blood of tumor bearing patients, particularly in the cases of hematologic malignancies, may increase [9,10]. Nevertheless, it is actually not clear whether the elevation of plasma lactic acid concentrations in malignancies is a general feature of the disorder or its role is only restricted to its clinical prognostic value as a state marker [9,10].
An ongoing aim in clinical medicine is to replace invasive procedures with non-invasive ones of similar clinical value. Towards this goal, an attractive alternative to invasive blood examinations is the analysis of chemical compounds contained in exhaled breath and exhaled breath condensates (EBCs) [11]. The exhaled air is suitable for the determination of volatile organic compounds (VOCs), for which more than 1500 volatiles have been identified to date (both endogenous and exogenous) [12], while EBC enables the examination of non-volatile components such as lactate, peroxynitrite and cytokines [13]. Both types of sampling are suitable for the detection of the appearance of new compounds and for following changes in distribution and concentration of otherwise normally present components resulting from changes in metabolic processes in the human body, and in this way they are of clinical value. However, to achieve reproducible results and comparability among different studies, various factors have to be considered (i) prior to the sampling among others the place and time of the breath collection, ambient air pollutants, consumed food, drinks, smoking behavior of the volunteers, (ii) parameters can differ during sampling and (iii) the variables resulting from different detection techniques [14].
In an effort to deduce the clinical value of plasma lactate, it would be convenient to be able to detect this acid through the analysis of EBCs. However, EBC investigations of breath lactate are sparse. For this reason, the aim of this article is to take into account the possible use of breath lactate measurements in tumor diagnosis and treatment control, to discuss technical barriers to measurement, and to evaluate directions for the future improvement of this technique. The use of EBC lactic acid levels in disorders other than cancer is also reviewed in short.

EBC collection methods
EBC contains mainly water condensed from the saturated exhalate, variable sized particles and droplets as aerosols of the airway lining fluid, and watersoluble volatiles absorbed in the condensing breath [15]. Standardization is an important issue in EBC collection since due to its heterogeneity, the conditions and procedures during sampling and analysis have huge influence onto the detected level of biomarkers. Despite the fact that the airway lining fluid is an unknown portion of the EBC sample highly diluted by water no single method (e.g. the use of a selected protein or conductivity [16]) is generally accepted for determination of the dilution factor.
Besides, the consumed food, drinks, smoking and also the circadian rhythm of subjects can cause pH change of EBC due to the variation of the hydrogen peroxide concentration [17]. Moreover, different methodological aspects can vary the constitution of EBC among others the use of nose clip [18] or face mask [19] during sampling, the collection through the mouth or endotracheal tubes [18], the application of inspiratory filters [20] and different exhalation maneuvers [21]. To give an example for the collected EBC volume, 10 min tidal breathing resulted in about 1-3 ml of an EBC sample arising from the cooling and condensation of the exhaled aerosol [18]. However, the sampling time differs strongly among studies [22]. Beside environmental pollutants the ambient temperature, pressure and relative humidity influence EBC volume, and pH, respectively [23].
During condensation of exhaled breath, several parameters are of utmost importance, however, the efficiency of an instrument is determined by the volume of exhaled air passed through the cold area over time and the temperature gradient between exhaled air and sampling system [14]. A number of commercially available and home-made devices are used for EBC collection in different studies [14,22] that include chambers or tubing made from various materials (Teflon, polypropylene, glass) having different length and diameter [18] and applying different cooling mechanisms (dry ice, ice-water, Peltier element, fluid nitrogen) for condensation mainly in a temperature range between −5 • C and −15 • C sometimes down to −80 • C [14,18,22].
Moreover, storage conditions such as temperature and time between collection and analysis might cause changes in concentrations of biomarkers [24].

Methods to detect lactate in breath
Lactic acid is a strong carboxylic acid with a pKa value of 3.86. Thus, it dissociates under physiological conditions and appears in its deprotonated, anionic form in biological samples. This limits its volatility. As its Henry constant (air-water partition coefficient), K H , is 9.6 × 10 −9 atm m 3 mol −1 (at 25 • C) it stays mainly in the fluid phase [25], which hinders the bloodair exchange in the lungs' alveoli. Therefore, the collection of EBC is the applicable method for lactate determination [26]. Since EBC consists of more than 99.9% water, the concentration of biomarkers such as lactate is very low, being in the µM-range [27].
The first lactate determination method, the colorimetric approach, has been applied from the first half of the last century and provided the range of lactate detection from 0.01 to 0.13 mM in animal blood and different tissues [28]. Use of liquid chromatography with ultraviolet detection, which required the derivatization of L-lactate with alpha-bromoacetophenone, provided a more sensitive technique with a detection limit down to 36 pmol. The method was optimized for detecting human plasma lactate values which were ranged to 1.5 ± 0.04 mM after overnight fasting [29]. Jackson and coworkers achieved even high analytical sensitivities for lactate in EBC in the range of 0.1 and 100 µM with hydrophilic interaction liquid chromatography coupled to mass spectrometry [30]. Given that sensors have the potential of providing cheap and highly portable devices for chemical analytics, it is worth to mention that a lactate biosensor as part of a flow injection system was used to measure blood lactate concentrations in EBC down to 1 µM, but only after preconcentration of the lactate in an ion-exchange column [31]. This biosensor was fabricated employing the highly purified enzyme lactate oxidase (EC 1.1.3.2), which converts lactate to pyruvate and H 2 O 2 . The H 2 O 2 concentration produced in the reaction was determined through amperometric detection [32]. With a new development within Healthy persons (at rest) Healthy persons (after exercise) [27] this technology, the measurement of lactate concentration as low as 5 µM became available with a linear range up to three order of magnitude [32].
Recently, a miniaturized EBC sampler was developed by Greguš and coworkers enabling a fast collection of EBC within 10 s from just one single exhalation [33,34]. In their study involving patients with various respiratory diseases, 12 inorganic anions, cations, and organic acids, among them lactate, from the ca. 70 µl sample were separated using capillary electrophoresis and detected with capacitively coupled contactless conductivity technique [35,36].
The concentration of lactate in EBC was measured to be much lower normally in the µM-range (see table 1) than that in the blood [37]. The correlation between lactate concentration in arterial blood and lactate release in exhaled air was reported by Marek and coworkers in a study using young and healthy volunteers performing exercise on an ergometer [27]. Given that during exercise lactate production is increased, compared to resting state lactate values, a wider concentration range of lactate concentrations could be compared from both collected EBC and blood samples. Lactate concentrations were measured using an enzymatic biosensor involving immobilized lactate oxidase. A significant correlation between the lactate release in EBC and in arterial blood was observed, with the lactate release in EBC increased by more than a factor of four after the exercise while lactate concentrations in arterial blood increased nearly tenfold [27].
Higher lactate concentrations in EBC were observed in patients with acute lung injury resulting from the reduction of the local pH caused by hypoxia [38]. Similarly, because of the imbalance of oxygen supplement and demand, significantly higher lactate levels were measured in EBC collected from patients suffering from either chronic obstructive pulmonary disease or asthma [39].
Whereas the acute pulmonary exacerbations in pediatric cystic fibrosis patients led to an elevated lactic acid content in EBC, compared to stable preor post-exacerbation cases, the adult patients had a lower abundance of lactic acid in the case of acute exacerbation, and changes between the stable and exacerbated groups were found to be not statistically significant [40].

Breath analysis and cancer
It has been known for a long time that the analysis of VOCs as exhaled biomarkers offers a useful tool for fast and non-invasive disease detection. The underlying idea is based on the knowledge that VOCs are the products of metabolic processes and thus can provide a mirror to the actual metabolic status of the body. In cancer cells, the biochemical shifts and altered enzymatic processes cause a quantitative change in the amount and distribution of endogenous metabolites, perhaps the best known is lactate, but the quality and quantity of VOCs also change parallel with the occurrence and progression of the disease.
Large number of studies have investigated VOCs in the exhaled breath of patients suffering with different kinds of carcinomas [41,42]. An obvious obstacle to the more wide use of VOCs in clinical practice is that there is still a very low inter-study reproducibility of results regarding the identified volatile markers combined with a poor understanding of their origin and clinical significance. The reasons for these are that in case of untargeted breath analysis a huge number of VOCs are detected due to various influencing confounding factors, including consumed food and drinks, inhaled chemicals, medications, life style and different sampling procedures [12,43]. Nonetheless, some attempts were done for standardization of sampling techniques used for breath analyses in the last years [43], because a lack of conformance between studies results in issues of intercomparison, as noted above for EBC collection, too, if the methods in the field of sample collection and analytical detection significantly varying [26].
Issitt and coworkers examined the outcome of several studies dealing with exhaled breath analysis involving lung cancer patients. Compared to healthy controls, they mainly found differences in the quantities of exhaled hydrocarbons, aldehydes and furanes. These differences were found to be sufficient to separate lung cancer patients from the healthy controls [44]. It is hypothesized that these changes are a result of polyunsaturated fatty acids undergoing a higher level of oxidative stress. For cancer patients, this comes from the elevated level of reactive oxygen species resulting in the formation of the abovementioned group of VOCs [45]. This concept was partly confirmed in the study of Schallschmidt and coworkers, who detected an increased concentration of aromatic compounds in the breath of smokers and elevated levels of oxygenated VOCs, such as aldehydes, 2-butanone and 1-butanolin, in the exhaled breath of lung cancer patients [46].
In addition to VOCs studies, it has been reported that EBCs reveal elevated levels of inflammatory and oxidative stress markers as well as gene mutations or DNA abnormalities in patients with lung cancer [47]. Peralbo-Molina and coworkers investigated untargeted metabolomics studies using gaschromatography time-of-flight mass spectrometry to measure EBC samples collected from lung cancer patients. After liquid-liquid extraction of the analytes using hexane saturated monoacylglycerols were among the detected compounds found to be useful in discriminating between the healthy and lung cancer groups [48].

Lactate, lactacidosis and clinical malignancies
It has been known for almost a century that lactate metabolism is disturbed in tumors [49]. Despite this, while the plasma lactate concentrations in relation to disorders other than cancers are commonly observed and well documented, the attempts to correlate plasma lactate and tumors in clinical practice have been fraught with difficulties. Three types of lactacidosis are defined: type A relates to tissue hypoxia and hypoperfusion (trauma, sepsis, different types of shock, seizure), type B is independent of hypoxia and hypoperfusion (drugs and toxins, diabetes mellitus, thiamine deficiency, malignancies, hereditary liver disorders, renal failure) and type D results in elevated levels of D-lactate (short bowel syndrome, diabetic ketoacidosis) [50-53]. Regardless of its origin, the higher the plasma lactate level the higher the risk of death [53,54].
As far as known, the increase in lactate production is accompanied by the acidification of the tumor environment, that suppresses normal immunological functions and results in accelerated local invasion [61][62][63]. These non-favorable experimental observations upon lactate function get supported by clinical data, too. Breast cancer patients with grade III of classification display higher lactate levels and poorer clinical prognosis than their counterparts with grade II [64]. Based on these findings, it is believed that elevated plasma lactate levels are part of the clinical spectrum of cancer in patients with poor prognosis.

A proposal for how elevated plasma lactate level may emerge in malignancies
Although elevated lactate production is a hallmark of cancer cells, with both D-and L-isomers being overproduced [65], the plasma lactate concentrations, mainly the detected L-variant, remain within the normal range for a long time. As mentioned above, the appearance of lactacidosis is closely related to hematological tumors and to the presence of liver metastases. Thus, questions to raise are: what causes the appearance of lactacidosis and what are its pathway mechanisms? Since the literature is abundant with studies investigating both the biochemical background of lactate production in tumors, particularly focusing upon Wartburg effect [66] and the effects of lactate upon tumor functioning and propagation [67], here only the questions closely in line with the topic are dealt with in the form of a moderate speculation ( figure 1, panels A and B).
In solid tumors, a metabolic heterogeneity for the cells is known, meaning a metabolic compartmentalization, or in other words symbiosis, between different types of malignant cells ( figure 1, panel A). These cells are glycolytic tumor cells and oxidative tumor cells that preferentially express monocarboxylate transporter 4 (MCT4) and MCT1, respectively. The former facilitates lactate export, while the latter enhances lactate uptake, thus providing a metabolic connection [62,71]. Such metabolic compartmentalization is not unique, because it is for example present between neurons and glial cells, too [72,73]. As known, the lactate produced by anaerobic glucose breakdown in muscles is transported to and converted in the liver thereby being recycled to glucose in the Cori cycle [74], thus the lactate overproduced in and secreted into the bloodstream from tumors can similarly be a source of glucose production ( figure 1,  panel A). An essential role in lactate metabolism is assigned to the liver accounting for about 70% of lactate clearance [8] therefore, following from the above, when the plasma lactate level exceeds the capacity of the liver to take up and handle lactate originating from tumors, then hyperlactacidemia followed by lactacidosis develops ( figure 1, panel B). Particularly, when liver damage is evident as seen in metastatic In part, this lactate is taken up by oxidative tumor cells and in part, it is secreted into the blood that, via circulation, enters the liver where it is metabolized to glucose. Glucose is released from the liver and reused up by peripheral tissues and tumor. Panel B. Relationship between plasma and EBC lactate concentrations. The slope of the rise of breath condensate lactate is low within a hitherto unknown range, but the extent of the domain possibly well exceeds the normal upper limit of plasma lactate concentration. However, when the plasma lactate level exceeds the buffering capacity of the lung tissue to absorb and metabolize lactate that is realized in the route from bloodstream to alveolar air, it starts to increase. There is a consent among clinical laboratories that the lower limit of the normal range for the serum lactate level is 0.5 mM, while the consensus among investigators is missing on that lactate levels greater than 2 mM represent hyperlactemia [68][69][70]. Though lactic acidosis is sometimes defined as serum lactate concentration above 4 mM, this value can in fact be disputed as there are obvious difficulties for giving a general definition for the lactate level representing lactacidosis as it may exist together with other acid-base abnormalities also affecting the development of the values of pH and bicarbonate below 7.35 and 20 mM, respectively [69]. The arrow indicates the expectation to develop technologies that lower the detection threshold, thus making breath condensate lactate measurements more useful technique for clinicians. Dashed line = glucose, solid line = lactate. tumors [75]. This also occurs in the case of acute tumor-lysis syndrome that is characterized by the massive destruction of malignant cells and the release in the extracellular space of cellular content, which may also lead to lactacidosis [76].
Although lactate may also be secreted into and taken up from the blood by leukemia cells (figure 1, panel A), this is different from that of other tumors owing to their MCT profile, with an MCT1 excess over MCT4 [71,77], which prefer the import rather than the export of lactate. Hence, with a high probability, a MCT related mechanism cannot be responsible for the leukocyte origin of plasma lactate. On the contrary, apoptosis of leukemia cells is able to contribute to lactacidosis in these hematologic disorders, particularly when the liver is dysfunctional [78]. The mechanism of lactacidosis developing in pheochromocytoma is, however, exceptional as it differs from the general mechanism assumed above, because in this case the effect of catecholamines upon intermediary metabolism is the pathological factor [79][80][81][82].

Conclusions and future directions
It was Linus Pauling and colleagues, who first raised half a century ago the use of breath for a more general quantitative analysis in medicine [83]. Breath sampling is a completely non-invasive and easily repeatable technique representing a useful approach to detecting and screening for disorders. Thus, its application is spreading in clinical practice. Both volatile and nonvolatile breath constituents can be profiled in exhaled breath and used in clinical work, the latter in the form of EBC detecting such compounds like lactate [11]. Although, there are several analytical methods worked out and several studies have been conducted on non-volatile compounds in EBC [84], further research and developments are needed to improve the clinical applicability of EBC approach.
In EBC, the proteomic profile of breath from cancer patients differs from that of healthy controls [13,39]. Though lactate is also determined in some clinical cases, its determination and association with cancers are not the focus of present day research [39]. However, improving of our understanding of the relationship between EBC lactate and carcinogenesis should be important in order to help evaluate not only the molecular breath patterns, but also to understand their relevance to therapy. For this reason, the technologies need to be refined, and in particular, to decrease the analytical detection threshold, which is of fundamental importance in terms of the clinical use of EBC lactate determination ( figure 1, panel B). These endeavors should include the development of more precise collection techniques and more sensitive detection methods. The crucial methodological issues in this regard are: collection devices, condensation devices, concentration of samples, avoidance of saliva contamination, de-aeration and pH monitoring [85,86]. From a clinical view point, in the course of study design the standardization of patients' groups with special attention to various confounding factors, such as comorbidity, age, food consumption, smoking and medicines has to be strongly encouraged [18,87].
Lactacidosis in cancerous patients means a poor clinical prognosis, and high lactate levels in tumors foreshadow the likelihood of metastases, tumor relapse and limits patients' survival [61,62,88,89]. Owing to the aforementioned limits in the detection techniques as well as a lack of pilot clinical studies, at the present status of knowledge the measurement of EBC lactate can only be used as a state marker of advanced cancer.
The advantage of EBC technique in determination of lactate level changes would be eminent in follow-up studies as being non-invasive it means less burden to the patients. And perhaps, the emergence of miniaturized fittings connected with smart phone may make possible a remote medical control of patients, that would be of particularly advantageous in those geographical areas where distances are great and hygienic conditions are problematic. To fulfill this function, however, further development of techniques is necessary that still awaits, but this is the future in perspective.

Data availability statement
All data that support the findings of this study are included within the article (and any supplementary files).