Diagnosis of Common Oral Diseases by Signature Volatile Profiles

Physicians in ancient Greece understood that smelling the breath of their patients can assist in diagnosis. However, while this rather primitive diagnostic method hasn't changed much over the centuries, there has been an increasing interest in recent years in improving methods of noninvasive early diagnosis for numerous metabolic and infectious diseases.

Currently, many diseases are missed and exacerbated because of delayed diagnosis. In dental medicine, patients tend to avoid visiting their dentist even for regular checkups, and thus they only appear at the clinic at a later, advanced stage of the disease, suffering of pain and discomfort that could have easily been prevented, and presenting late stage disease, thus jeopardizing the treatment outcome.

For these reason, there is an urgent need for an inexpensive and minimally invasive technology that would serve as a diagnostic aid, allowing 1) efficient early detection and 2) treatment customization.

Diagnostic modalities based on the detection of volatile organic compounds (henceforth VOCs), i.e. organic compounds with relatively high vapor pressure at room temperature, may be the answer for the aforementioned need.

"Electronic Noses" for the Detection of Volatile Organic Compounds in the Exhaled Air Disease-specific volatile organic compounds are produced mainly through changes in specific biochemical pathways in the body. Following their production by either human or bacterial cells, VOCs may be found in body fluids, including infected cells and/or their microenvironment, blood, breath, and saliva, among others. Thus, VOCs released from their origin can be directly detected from blood and the headspace of cells.

Numerous worldwide research groups are investigating the possibility of non-invasive VOC detection in the exhaled air, enabling the diagnosis of various systemic diseases6, including carcinomas (lung, breast and colorectal), tuberculosis, kidney failure and asthma. The VOCs in question are alkanes (C4-C20), methyl-alkenes and benzene derivatives, found in healthy individuals in concentrations of 1-20ppb (part per billion), while in disease, the concentration of disease-specific VOCs is increased. VOCs are suggested to be produced by the distressed tissue due to oxidative stress, by the liver as part of a metabolic reaction, or by the immune system.

An "Electronic Nose" is a device capable to recognize the volatile composition of an air sample. The device consists of an array of nanoreceptors capable to transform physical or chemical information into an electric signal. Every receptor in the array reacts differently to the materials in the analyzed sample, and the combination of these responses from several receptors produces a unique volatile profile, distinguishable by quality (namely, which molecules are found in the sample?) and quantity (namely, what is the exact concentration of each molecule?).

An electronic nose system for air sample analysis should usually consist of three key components:

1. Breath / headspace collection device

2. Processing device

3. Classification algorithm. The first and most widespread such devices are based on air sample analysis in a Gas-Chromatography / Mass-Spectrometry (GC-MS) system. Advanced systems include methyl-oxide sensors (MOS) or nanoreceptors.

A nanoreceptor-based electronic nose system (NaNose) is being developed, at the Laboratory of the co-PI (Prof. Haick, Faculty of Chemical Engineering, Technion, Haifa, Israel), capable of detecting traces of VOCs in concentrations of single parts-per-billion (1µg/l). Paired with a computerized system able to analyze large amounts of data from hundreds of molecules, detected simultaneously by an array of 20 or more receptors, this system surpasses the other "electronic nose" devices in terms of sensitivity and efficiency. Contrary to costly and slow GC-MS systems, an ideal nanomaterial-based sensor for breath testing should be sensitive at very low concentrations of volatile organic compounds, even in the presence of environmental or physiological confounding factors. It should also respond rapidly and proportionately to small changes in concentration and provide a consistent output that is specific to a given compound. When not in contact with the compounds in question, the sensor should quickly return to its baseline state, or be simple and inexpensive enough to be disposable.

Air sampling with such receptors should be comparably simple, and its results may be interpreted automatically, which makes it suitable for cost effective screening of a large populations. Only positively tested patients will require conventional and unpleasant diagnostics to confirm the early diagnosis, before a treatment is suggested.

Nevertheless, all existing VOC based diagnostic modalities tend to forget that the exhaled air is a combination of numerous origins - lungs, the upper GI tract, and the oral cavity with small nasal component. While the composition and volatile profiles of air from the lungs were thoroughly researched in the recent decade, as means to find an early diagnostic approach to cancer10,15, the oral component remained largely untouched.

Micro niches in the oral cavity of healthy individuals' harbor biofilm that may contain over 1000 different bacterial species. In the absence of active mechanical or chemical cleansing, biofilm accumulates, changes and matures, causing soft tissue inflammation, known as gingivitis. In some individuals and under certain conditions, inflammation progresses to periodontal disease, and the biofilm composition changes to a higher proportion of anaerobic Gram-negative bacteria with increased virulence and tissue-breakage capabilities. Aside of biofilm microorganisms, additional sources of VOCs in the oral cavity may include serum, gingival exudate, inflamed gum tissue, sores and lesions, pathologies associated with salivary glands, sinuses and nasal cavity, gastrointestinal reflux, interdental trapped food debris, and environmental pollutes. Hertel et al. has recently reported unique volatile profiles in the headspace of specific oral bacteria and fungi.

We hypothesize that nanoreceptor-based volatolomics can be used as a diagnostic modality for non-invasive early diagnosis of oral diseases, based on a sample of exhaled air.

The proposed research aims to lay the foundation for a fast, user-friendly and non-invasive diagnostic modality for common oral diseases in exhaled air samples, based on the NaNose. This will serve in the foreseeable future as a base for creation of a marketable clinical or home-based diagnostic device.

Specific aims:

1. Detection and analysis of VOCs connected with common oral diseases in the headspaces of microorganism samples from the oral cavities of patients diagnosed with gingivitis, periodontal diseases or caries, and compared to a control group of healthy patients.

2. Characterization of volatile profiles for common oral diseases and defining specified diagnostic nanoreceptors.

Desired Outcomes

1. The results of the proposed research may lead to a revolution among dental professionals, as well as at home care. To date, the only way for an individual to learn of his oral condition is to schedule an appointment with a dentist, yet a significant proportion of the population avoids regular checkups and treatment. An easy tool enabling a simple and non-invasive air sampling at home or in a drugstore will significantly increase patient compliance and curability rates, and decrease healthcare expenditure.

2. VOC mapping of the air originating in the oral cavity may increase the diagnostic value of air samples originating in the lungs, thus improving the precision, sensitivity and specificity of non-invasive early diagnosis of cancer, kidney failure and other internal diseases.

Methods

Stage 1 - Detection and analysis of VOCs connected with common oral diseases Known bacterial strains (Streptococcus mutans, Streptococcus sanguinis, Porphyromonas gingivalis, Fusobacterium nucleatum, Acinetobacter actinomycetemcomitans, Tannerella forsythia, Lactobacillus acidophilus) will be cultivated separately in sealed bottles containing appropriate media, enabling collection and analysis of the unique headspace inside the bottles. After cultivation, headspace will be collected into a Tenax absorbent tube (Sigma-Aldrich, 28718-U SUPELCOTenax® TA / Carboxen® 1018) and analyzed in GC-MS (Gas Chromatography-Mass Spectrometry). A unique volatile profile of every bacterial strain will be determined via a statistical analysis. The complete trial protocol is available in Addendum 1 below.

Three (3) repetitions will be performed with each strain.

Stage 2 - Characterization of volatile profiles for common oral diseases and defining specified diagnostic nanoreceptors The tested samples will be of oral bacterial plaque and infected dentin, cultivated in sealed bottles similarly to Stage 1.

The patients will originate at the students' clinics, and the samples will be taken as part of their dental treatment, namely: removal of caries and infected dentin for future restoration, and dental calculus and plaque removal for the treatment of periodontal disease.

1. In the beginning of the treatment session, an informed consent will be obtained from the patient, agreeing to the usage of removed plaque / infected dentin in research. No deviation from the standard treatment protocol is expected.

1. The plaque sample of periodontal patients will be taken before gross scaling has commenced, with a manual tool (Gracey 5-6 curette), from the cervical area of mandibular incisors.

2. The infected dentin sample of carious teeth will be taken with a manual tool (Spoon Excavator), after the initial form of the prepared cavity was outlined.

2. The collected samples will be then cultivated in sealed bottles containing Wilkins-Chalgren medium, at 37°C for 48 hours, similarly to Stage 1 above. A bottle with sterile medium will be used as control.

3. Headspace will be then collected and analyzed similarly to points 5-8 at Stage 1 above.

4. At this point, the unique volatile profile of each patient will be linked with his clinical diagnosis (extent of periodontal disease / caries), and analyzed for statistical consistency.

5. Data gathered in stages 1-2 will be statistically analyzed as a whole.