**5. Biomarkers in diagnostic of disease**

Using the techniques of extraction and identification of VOC allows the discovery of biomark‐ ers that are measurable biomolecules that represent the normal physiological state or changes of an individual related with a pathophysiological process or treatment. Generally biomarkers may be (a) proteins that represent the cellular and enzymatic changes or (b) small organic molecules called metabolites. These compounds may help in early detection and monitoring the progress of a disease, and its response to the use of a drug or therapeutic intervention [104].

Currently biomarker discovered using "omics" technologies, such as proteomics and metabo‐ lomics wherein sample biochemical profiles, are determined (mainly biofluids) from patients and healthy individuals. Compounds that after an appropriate statistical analysis appear over or under-expressed in samples from patients when compared to healthy controls may be considered as potential biomarkers of disease [105].

In the area of Leishmaniasis there is only one work reported in literature about VOC extraction. HS-SPME was used by de Oliveira et al. [41] for extraction of VOC from hair of dog healthy and infected by *Leishmania infantum*, followed by GC-MS analysis which detected about 274

After extraction VOC from samples it is necessary to identify each component to obtain quantitative and qualitative information about samples. Since the invention and development of gas chromatography technique the analysis of volatile compounds in biological becomes

GC is an excellent analytical technique for separating compounds which can be volatilized at the temperature applied to the injector device. Therefore, when GC is coupled to a detector it

Nowadays gas chromatography coupled to flame ionization detector (GC-FID) and GC-MS are the main techniques used for VOC analysis and the GC-MS technique is more powerful

In recent years, the development of multi-dimensional CG has significantly improved separation of VOC from complex samples, making easier than before to obtain bio-information related to compounds present in samples [101]. Bean et al. [102] applied comprehensive twodimensional chromatography coupled to time-of-flight mass spectrometry (GCxGC-TOF-MS) technique to identify the profile of VOC from the metabolism of *Pseudomonas aeruginosa* bacteria. The method enabled better separation and identification of a wider number of compounds compared to the GC-MS technique, enabling the discovery of 28 new VOC

Another innovative application was proposed by Stadler et al. [103] that used TD × GC-TOF-MS for the identification of VOC released from a pig carcass, which allowed to define of the

Using the techniques of extraction and identification of VOC allows the discovery of biomark‐ ers that are measurable biomolecules that represent the normal physiological state or changes of an individual related with a pathophysiological process or treatment. Generally biomarkers may be (a) proteins that represent the cellular and enzymatic changes or (b) small organic molecules called metabolites. These compounds may help in early detection and monitoring the progress of a disease, and its response to the use of a drug or therapeutic intervention [104].

Currently biomarker discovered using "omics" technologies, such as proteomics and metabo‐ lomics wherein sample biochemical profiles, are determined (mainly biofluids) from patients and healthy individuals. Compounds that after an appropriate statistical analysis appear over

profile of compounds released during the decomposition of different tissues

becomes a powerful technique for analysis of VOC in biological samples [100].

due to the great identification ability of mass spectrometry (MS).

compounds, mostly ketones, aldehydes and hydrocarbons

326 Leishmaniasis - Trends in Epidemiology, Diagnosis and Treatment

**4.3. Methods of analysis**

characteristic of *P. aeruginosa*.

**5. Biomarkers in diagnostic of disease**

much easier.

It is highlighted that finding metabolites present at significantly different levels in samples from patients does not necessarily make them useful biomarkers. The route for the validation and use of biomarkers requires extensive studies. This validation must address two issues: a) the metabolite whose concentration levels differ significantly between samples of a test population is indicative of a specific pathophysiology? b) there is an abnormal level of this specific metabolite clearly stating a specific pathophysiology when all other symptoms are considered? In addition, biomarkers need to provide adequate levels of sensitivity and specificity (> 80%) for correct detection and classification of disease [106].

The strategy most often used to develop biomarkers involves a discovery phase in a restrict number of samples, followed by validation of biomarker with potential using a larger number of samples from patients before this compound can be adopted as a clinical tool.

The use of chromatographic techniques combined with the power of mass spectrometry allows the identification of dozens of compounds and this when coupled with the large number of samples studied generates an immense amount of data that must be processed, analyzed and interpreted.

The purpose of the data processing is to extract quantitative information from data related to detected metabolites obtained from chromatographic and mass spectrometry. This extracted information can be arranged in an matrix which later be will used in the multivariate analy‐ sis[107].

The main multivariate techniques used for biomarker discovery are Principal Component Analysis (PCA), Partial Least-Squares – Discriminant Analysis (PLS-DA) and orthogonal PLS-DA (OPLS-DA) [108]. PCA is commonly implemented as a starting point for the data manip‐ ulation to obtain an overview about cluster tendencies and to detect outliers.

PLS-DA is used to deal with specific studies class where a linear relationship between metabolites in the matrix X and category information in the matrix Y is established. This technique can enhance the separation of classes and improve the interpretability of statistical models [109]. In OPLS-DA model partition variance matrix X in predictive and non-predictive classes (orthogonal classes) not only improves the discrimination metabolic profiles but also the transparency of models [110].

Many biomarkers of disease from different samples have been discovered by the application of different techniques for extraction and analysis. Table 1 summarizes some of these bio‐ markers.

Table 1 shows already discovered biomarkers for various diseases such as Melitus diabetes, myocardial ischemia, galactosemia, breast cancer, ovarian carcinoma, renal cell carcinoma, colorectal carcinoma, etc [4].


**Table 1.** Biomarkers identified for different diseases.

Most of the identified biomarkers are non-volatile, such as proteins, lipids and other high molecular weight compounds. However, several volatile biomarkers have been discovered in recent years.

In this context, the discovery of VOC that may serve as biomarkers of infectious diseases such as leishmaniasis is presented as a promising possibility. The first work in this direction has been reported in the literature by Oliveira et al. [41] which proposed a noninvasive diagnostic method for detection of visceral leishmaniasis in dogs. HS-SPME/GC-MS techniques were developed and applied to identify the VOC emitted from hair of 8 healthy dogs and 16 dogs infected by *Leishmania infantum* and 274 compounds were detected (Figure 4).

**Figure 4.** Chromatograms obtained by HS-SPME/GC–MS analysis of canine hair samples, for all VOC [41].

So, multivariate Principal Component Analysis (PCA) and Soft Independent Modeling of Class Analogy (SIMCA) were used to discriminate healthy and infected groups of dogs. Figure 5 shows the PCA scores graph.

The majority of identified VOC belonged to the class of ketones, aldehydes and hydrocarbons, and compounds such as benzaldehyde, 2-hexanone and 2,4-nonadienal showed greater discriminatory power between the two groups, therefore potential this later compounds are candidates for biomarkers of canine visceral leishmaniasis.

Continuing this work, Magalhães-Junior et al [42] investigated that the profile of VOC emitted from infected dogs change according to symptomatology or not of their conditions. For this 36 dogs living in endemic areas for leishmaniasis were studied, which were divided into three groups according to (1) absence of the disease, (2) presence of clinical infection or (3) subclinical *Leishmania infantum*. In this study, infection was confirmed for parasitological test by culture and microscopy and PCR.

Most of the identified biomarkers are non-volatile, such as proteins, lipids and other high molecular weight compounds. However, several volatile biomarkers have been discovered in

Breath Cysticfibrosis GC-MS Carbonyl sulphide, alkanes Phillips et al. [121]

TD-GC-TOF-MS

Urine Lung cancer GC-MS 2-heptanone, o-toluidine,

**Sample Disease Technique Biomarkers References** Urine Colorectal carcinoma UPLC-TOF-MS Nucleosides, carnitines Wang et al [111]

Dried blood Galactosemia LC-MS/MS Hexose monophosphates Jensen et al. [117]

Onchocerciasis LC-MS Fatty acids, protein, sterol

UPLC and

Blood plasma Myocardial ischemia HPLC-MS Sugars, ribonucleotides,

Blood serum Renal cell carcinoma LC-MS Phospholipids, phenylalanine

Leucine, isoleucine, valine, phenylalanine, tyrosine

GC-TOF-MS Lipids and polar metabolites Oresic et al [114]

HPLC-MS Hormones, nucleosides Woo et al. [115]

Propane,carbon disulfide, 2 propenal, ethylbenzene and isopropyl alcohol

Hexanal, nonanal, decanal, undecanal, pentadecanal, dodecanal, etc

amino acids Sabatine et al [116]

lipid, etc Denery et al. [119]

nitromethane, etc Matsumura et al. [122]

Rudnicka et al. [120]

Basanta et al. [3]

and cholesterol metabolites Liu et al. [118]

Wang et al [112] Langenberg and Savage [113]

Blood serum Diabetes melitus LC-MS

328 Leishmaniasis - Trends in Epidemiology, Diagnosis and Treatment

Breast cancer, ovarian carcinoma, cervical carcinoma

Breath Lung cancer GC-TOF-MS

Chronic ostructive pulmonary disease

**Table 1.** Biomarkers identified for different diseases.

Blood serum Type I diabetes

Urine

Blood plasma and serum

Breath

In this context, the discovery of VOC that may serve as biomarkers of infectious diseases such as leishmaniasis is presented as a promising possibility. The first work in this direction has been reported in the literature by Oliveira et al. [41] which proposed a noninvasive diagnostic method for detection of visceral leishmaniasis in dogs. HS-SPME/GC-MS techniques were developed and applied to identify the VOC emitted from hair of 8 healthy dogs and 16 dogs

infected by *Leishmania infantum* and 274 compounds were detected (Figure 4).

recent years.

The authors used HS-SPME/GC-MS techniques for identification of VOC from dogs and after multivariate analysis using PCA and PLS-DA they found that the profile of these compounds emitted differ between the three groups. There still differences among infected dogs showing clinical disease from those with subclinical infection. They were also identified 10 new compounds with potential use as biomarkers of infection. These studies show that the use of VOC as biomarkers of infection by *Leishmania sp*. as a promising approach. However, many studies must still be performed until confirmation of biomarkers capable of being used in clinical practice, including potential application for humans.

**Figure 5.** PCA scores graph based on chromatographic peak areas of VOC of the hair of leishmaniasis infected (A) and non-infected dogs (B) [41].
