Research Paper Volume 12, Issue 1 pp 978—995

Metabolomic profiling of dried blood spots reveals gender-specific discriminant models for the diagnosis of small cell lung cancer

Li Yu 1, , Kefeng Li 2, , Xiangmin Li 1, , Chao Guan 1, , Tingting Sun 1, , Xiaoye Zhang 1, ,

  • 1 Department of Clinical Oncology, Shengjing Hospital of China Medical University, Shenyang, Liaoning 110004, China
  • 2 School of Medicine, University of California, San Diego, CA 92103, USA

received: September 27, 2019 ; accepted: December 24, 2019 ; published: January 12, 2020 ;

https://doi.org/10.18632/aging.102670
How to Cite

Copyright © 2020 Yu et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY 3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Abstract

The accurate diagnosis of small cell lung cancer (SCLC) at initial presentation is essential to ensure appropriate treatment. No validated blood biomarkers that could distinguish SCLC from non-small cell lung cancer (NSCLC) has yet been developed. Dried blood spot (DBS) microsampling has gained increasing interest in biomarkers discovery. In this study, we first performed metabolomic profiling of DBS samples from 37 SCLC, 40 NSCLC, and 37 controls. Two gender-specific multianalyte discriminant models were established for males and females, respectively to distinguish SCLC from NSCLC and controls. The receiver operator characteristic (ROC) curve analysis showed the diagnostic accuracy of 95% (95% CI: 83%-100%) in males SCLC using five metabolites in DBS and 94% (95% CI: 74%-100%) for females using another set of five metabolites. The robustness of the models was confirmed by the random permutation tests (P < 0.01 for both). The performance of the discriminant models was further evaluated using a validation cohort with 78 subjects. The developed discriminant models yielded an accuracy of 91% and 81% for males and females, respectively, in the validation cohort. Our results highlighted the potential clinical utility of the metabolomic profiling of DBS as a convenient and effective approach for the diagnosis of SCLC.

Abbreviations

2-AG: 2-arachidonylglycerol; 2-OH: 2-hydroxy; 9-HETE: 9-hydroxyeicosatetraenoic acid; 13-HODE: 13-Hydroxyoctadecadienoic acid; Cer: Ceramide; CI: confidence interval; COPD: chronic obstructive pulmonary disease; CVD: cardiovascular diseases; CVs: cross-validations; DBS: dried blood spot; ESI: electrospray ionization; HBECs: human bronchial epithelial cells; IHC: immunohistochemistry; IMP: inosine monophosphate; IR: ionizing radiation; LASSO: least absolute shrinkage and selection operator; LC-MS/MS: liquid chromatography coupled with tandem mass spectrometry; LOOCV: leave-one-out cross-validation; MTHFR: methylenetetrahydrofolate reductase; NPV: negative predictive value; NSCLC: non-small cell lung cancer; NSE: neuron-specific enolase; PE: phosphatidylethanolamine; PEMT: phosphatidylethanolamine N-methyltransferase; PI: Phosphatidylinositol; PLS-DA: partial least square discriminant analysis; PPV: positive predictive value; ROC: receiver operator characteristic; SCLC: small cell lung cancer; SD: standard deviation; sFas: soluble Fas; SM: Sphingomyelin; sMRM: scheduled multiple reaction monitoring; VIP: variance in projection.