Trial Information

Early Detection of Lung Cancer - Exhaled Breath Nano-Analysis

Scientific background:

Lung cancer is the most lethal cancer, responsible for 28% of cancer deaths and killing ~1.3
million people worldwide every year. Diagnosis and treatment of lung cancer in its early
stages could increase the 5-year-survival rate 3-4 fold with a potential for cure4, 7.
Therefore, the main goal of this study is early detection of lung cancer, and specifically
focusing on the volatile biomarkers of lung cancer that will assist in easy, inexpensive
diagnosis based on our previous findings.

Currently available diagnostic tests of lung cancer are not suitable for screening, are
extremely costly and involve invasive procedures (e.g. bronchoscopy), that are not free of
complications. The goal of cancer screening is to detect tumors at an early stage in order
to give treatment a better chance of success. Recently, the biggest lung cancer screening
trial (NLST) has shown a mortality benefit of 21% per 5 years study favor low dose CT
screening protocol compare to chest X-rays1. Therefore, there is an urgent requirement for a
tool to allow a better definition of the high-risk cohort. Such a tool might be a panel of
biomarkers.

Focusing on the volatile biomarkers of lung cancer, our group has recently defined a
volatile VOCs signature that can distinguish the breath of lung cancer patients from the
breath of healthy individuals and from cancerous cells 2, 9. These significant findings have
led us to the understanding that volatile biomarkers would fit for early detection of lung
cancer and for discrimination between subtypes of lung cancer. Such discrimination will have
significant implications on clinical decisions and on patients' benefits.

Exhaled Breath Analysis as a Diagnostic Tool and Preliminary results:

Analysis of volatile organic compounds (VOCs) is a new attractive non-invasive field in
medical diagnostics. The principle behind this approach is based on the fact that cancers
cells are distinguished from normal cell in their metabolism rate, cell apoptosis pathways
and protein expression patterns and thus emit and or consume various VOCs. These VOCs can be
detected either directly from the headspace of the cancer cells or via the exhaled breath.

Together with Prof Haick group at the Technion Inst (Israel), our data shows that there is a
relation between the VOCs patterns of NSCLC and control cell lines and equivalent states in
exhaled breath. We demonstrated that there is a clear discrimination between the lung
cancer and the healthy clusters. We also analyzed the headspace of NSCLC and SCLC cell lines
and we could discriminate significantly between SCLC versus NSCLC based on their VOCs
patterns. This analysis allowed us to identify the specific VOCs consumed or omitted by
cancerous cells. This finding has clinical applications since SCLC is distinguished from
NSCLC by its sensitivity to chemotherapy and radiotherapy and other characteristics.
Therefore, a non-invasive and highly sensitive test would be extremely valuable for the
classification and early screening of lung cancer and for targeted therapy.

Breath Collection and the Artificial NOSE

In a typical collection, after normal exhalation, the subject will breath through a
mouthpiece a filtered air to remove all VOCs of any ambient contaminants. Individuals will
exhale in a constant flow rate. The exhaled air will be contained through the mouthpiece by
Mayler bags and/or will be passed through a container. The collected air breath samples will
be analysed for VOC by gas-chromatography mass spectrometry (GC-MS), highly-sensitive
nano-sensors (Nanomaterial-Based Devices, Technion - Israel Institute of Technology, Haifa,
Israel) or as online mass spectrometry (Ionimed, Austria). Further, the signals will be
analyzed by a pattern recognition algorithms, such as principal component analysis (PCA),
supported vector machines (SVM), or neuronal network analysis can then be applied on the
entire set of signals to acquire information on the identity, properties and chemical
composition of the vapor exposed to the sensors array 5, 10.

Research Objectives:

Our goal is to isolate and define a volatile signature, which allows discrimination between
lung cancer from a normal state. That will potential serve as a unique biomarker for lung
cancer.

Our Objectives are:

- To test the feasibility for detecting early stage lung cancer via analysis of exhaled
breath irrespective of the sub-histology.

- To test the feasibility of the breath biomarkers for monitoring the response to lung
cancer treatment (surgery, chemotherapy or other cancer treatments)

- To compare sensitivity and selectivity of the breath analysis to conventional cancer
markers and/or diagnostics (CT, PET, circulating tumor cells, blood markers etc.)

- Correlation of histology and/or other clinical measures to the breath signature.

Study Population

On the clinical setup, we will sample three populations:

Group A - patients with lung cancer (NSCLC and SCLC), any stage; this group will be
divided later as for:

1. SCLC (before and after therapy).

2. NSCLC:

1. Surgically treated (Before and after therapy 3 years follow up).

2. Advanced disease (Before and after therapy, 3 years follow up). Group B - high
risk patients who are undergoing investigation related to Lung cancer or pulmonary
nodule.

Group C - age and co-morbidity matched controls without proof of cancer/pre-cancer.

All collections will follow local IRB guidelines. Information will be collected from all
subjects, including epidemiologic data, histologic characteristics, tumor's metabolic
activity (SUV avidity through PET scan). The clinical information will include health
status, lung cancer subtype, pathology sub-classification and differentiation, advanced
analysis and staining if available, imaging results (including CT, PET scan and its SUV
avidity), location of the cancer, total volume of the tumor, stage of disease, genetic
classification of the tumor and epidemiological data, e.g. age, gender, smoking and family
history, family history, respiratory disease, exposure to asbestos etc.

If cancer, the selection for therapy are as per the standards of care and the routine
established care provided by the staff at the local institute. This protocol is not intended
to interfere with or dictate this process.

Examination procedure

The total duration of the study for each subject takes 10-20 minutes while subjects will
stay on followup for up to 3 years.

The study will continue for 5 years.

Newly diagnosed patients with non small cell lung cancer:

Breath tests:

1. Two tests immediately prior to any therapy (surgery/other; One week apart to test
reproducibility).

2. If was operated for cure, then at 3,6,12,18,24,36 months later.

3. If was radiated, then at mid & end of radiation and at 3,6,12,18,24,36 months.

4. If chemotherapy, then every 3 months, between cycles.

Follow up phase:

Every three months to coincide with patients regular follow up their treating physicians. As
per the standards of care, at this point every patient will be monitored with a CT scan of
chest and abdomen at regular intervals. This CT will be used for determination of disease
recurrence or to document remission.

Correlative studies

1. Sputum Cytology (Induced Sputum will be collected for cytologic examination).

2. Blood samples will be taken for Circulating tumor cells analysis and other systemic
markers.

Collection of the Breath Samples

In a typical collection, after normal exhalation, the subject will breath through a
mouthpiece a filtered air to remove all VOCs of any ambient contaminants. Individuals will
exhale in a constant flow rate. The exhaled air will be contained through the mouthpiece by
Mayler bags and/or will be passed through a container. The collected air breath samples will
be analysed for VOC by gas-chromatography mass spectrometry (GC-MS), highly-sensitive
nano-sensors (Nanomaterial-Based Devices, Technion - Israel Institute of Technology, Haifa,
Israel) or as online mass spectrometry (Ionimed, Austria). Further, the signals will be
analyzed by a pattern recognition algorithms, such as principal component analysis (PCA),
supported vector machines (SVM), or neuronal network analysis can then be applied on the
entire set of signals to acquire information on the identity, properties and chemical
composition of the vapor exposed to the sensors array 5, 10.