Virtual Library
Start Your Search
J. Van Cleemput
Author of
-
+
MINI 24 - Epidemiology, Early Detection, Biology (ID 140)
- Event: WCLC 2015
- Type: Mini Oral
- Track: Thymoma, Mesothelioma and Other Thoracic Malignancies
- Presentations: 1
- Moderators:J. Creaney, M. Carbone
- Coordinates: 9/08/2015, 16:45 - 18:15, 102+104+106
-
+
MINI24.08 - Breath Analysis by Ion Mobility Spectrometry Allows Discrimination of Pleural Mesothelioma Patients From Controls (ID 1508)
17:25 - 17:30 | Author(s): J. Van Cleemput
- Abstract
Background:
Despite the use of asbestos has been banned in many countries, 125 million people worldwide are still exposed to asbestos and at risk for developing malignant pleural mesothelioma (MPM). Since MPM is a lethal tumor, with diagnosis mainly at advanced stage due to non-specific symptoms and investigations, it is thought that only an early diagnosis will improve patient’s outcome (van Meerbeeck et al., 2011). Breathomics has emerged as a new research field, allowing to detect volatile organic compounds (VOCs) in breath which can be used as non-invasive markers for disease (Lamote et al., 2014). We investigated if asbestos induces VOCs and how breath VOCs could help discriminating MPM patients from occupational asbestos-exposed and non-exposed controls.
Methods:
Twenty-three MPM patients, ten healthy asbestos-exposed (AEx) individuals and twelve healthy non-exposed (HC) persons were included. After a fasting period of 2 hours before the breath sampling, participants breathed tidally for 3 minutes through a mouthpiece connected to a bacteria filter. Subsequently, ten ml alveolar air was sampled via a CO~2~-controlled ultrasonic sensor and analyzed using the BioScout Multicapillary Column/Ion Mobility Spectrometer (MCC/IMS, B&S Analytik, Dortmund, Germany). Per subject a background sample was taken. VOCs were visually selected and their intensity (V) was calculated via on-board VisualNow 3.7 software. After calculating the alveolar gradient, we performed a lasso regression in R to search for peaks that have the most discriminative power to distinguish MPM patients from controls. Predictions were made by leave-one-out cross-validation. The use of breath VOCs on the diagnostic performance was investigated by ROC-analysis.
Results:
Eighty-nine VOCs were selected in breath and background samples. The VOCs P25, P8 and P7 were able to discriminate HC from AEx controls with 91% accuracy (AUC~ROC~=0.95), yielding a sensitivity, specificity and positive (PPV) and negative predictive value (NPV) of respectively 90%, 92%, 90% and 92%. MPM patients were discriminated from AEx by the VOCs P5, P3, P30, P1 and P54 with 82% accuracy (AUC~ROC~=0.73), yielding a sensitivity, specificity and PPV and NPV of respectively 91%, 60%, 84% and 75%. When discriminating MPM patients from pooled HC and AEx controls, the VOCs P5, P3 and P1 were found to be important, yielding 73% accuracy (AUC~ROC~=0.71) and a sensitivity, specificity and PPV and NPV of respectively 70%, 77%, 76% and 71%.
Conclusion:
Breath analysis can discriminate MPM patients from healthy asbestos-exposed persons with 82% accuracy and from combined asbestos-exposed and non-exposed controls with 73% accuracy while healthy asbestos-exposed persons can be discriminated from non-exposed persons with 91% accuracy. The VOCs P25, P8 and P7 seem markers for asbestos-exposure while VOCs P5, P1 and P3 seem linked to MPM pathogenesis after exposure. For screening and to rule out diagnosis, a high sensitivity and NPV are mandatory and to rule in diagnosis, a high specificity and PPV are mandatory, which can enrich a population at risk for follow-up with (annual) CT scans or chest radiography. Hence, our results hold promise to use the breath test for screening of asbestos-exposed healthy seniors.
