Abstract 1264: Small volatile organic compounds extracted from advanced pancreatic cancer patient blood samples: Development towards a diagnostic method

Proceedings: AACR 103rd Annual Meeting 2012‐‐ Mar 31‐Apr 4, 2012; Chicago, IL Introduction Analysis of biological samples for volatile organic compound (VOC) profiles or fingerprints can potentially be utilized as a non-invasive, early stage diagnostic tool for disease related to metabolic processes. A hypothesis was presented in 1989 that dogs could detect malignant tumors by scent and several studies of canine scent detection for cancer followed. Our team has taken an analogous approach using solid phase microextraction in combination with capillary gas chromatography time of flight mass spectrometry (SPME GC-TOFMS) in place of canine scent detection to discover indicators for pancreatic cancer, a deadly cancer difficult to detect in its early development stage. VOCs are not commonly considered in “biomarker” analysis of clinical samples which usually require extensive pre-processing. The VOC fingerprint from a biological sample presumes many such substances are products, by-products, and waste resulting from human metabolism (and/or symbiotic microbiom) of nutrients and intermediates from endogenous processes, absorbed environmental contaminants, and endo/exogenous bacterial metabolism. Methods IRB approval was obtained to enroll patients with advanced pancreatic cancer and healthy volunteers. Blood was collected in pre-conditioned vacutainers from 11 patients and 5 healthy volunteers, all non-smokers, no chemotherapy in preceding 5 days, and analyzed within 2 weeks. SPME GC-TOFMS is a means for eliminating the biological matrix and pre-concentrating specific ranges of complex volatile fraction polarities at low concentrations and a very sensitive analytical instrumental platform for the characterization of VOC chromatographic profiles extracted from the 3ml blood samples. Multivariate statistical methods were used to analysis retention time exact mass observations and integrated peak areas. Results Even with the numerous variables contributing to an already diverse underlying metabolism of each participant, our chemometric approach did discriminate between “healthy” volunteers and the cancer patients with a 5% prediction error using a “leave one out” statistical analysis for 37 samplesb from the 16 subjects. Key retention time and exact mass observations from a fiber specific SPME were associated to preliminary molecule identifications via NIST data base searches. Our procedures for system tune, calibration, and monitoring achieved a response stability over a 1 month period of ∼10% rsd and accurate mass measurement of at least 5 mDa (30 - 300 Da) using benzene-d6, BFB, and naphthalene-d8. Conclusions VOC fingerprints from a SPME GC-TOFMS analysis of blood can be successful in discovering differentiating indicators related to APC. Future work will involve other SPME fibers and more samples. (Supported by the IBIS & Scottsdale Health Care Foundations) Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 1264. doi:1538-7445.AM2012-1264