Please login first

A Mass Spectrometry-based Lipidomics Study for Early Diagnosis of clear cell Renal Cell Carcinoma
* 1 , 2 , 1 , 1 , 3 , 4 , 5 , 1
1  Centro de Investigaciones en Bionanociencias (CIBION), Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Godoy Cruz 2390, C1425FQD, CABA, Argentina
2  Instituto de Investigación en Biomedicina de Buenos Aires (IBioBA) CONICET, Instituto Partner de la Sociedad Max Planck, Godoy Cruz 2390, C1425FQD, CABA, Argentina
3  Instituto de Investigación en Biomedicina de Buenos Aires (IBioBA) CONICET, Instituto Partner de la Sociedad Max Planck, Godoy Cruz 2390, C1425FQD, CABA, Argentina.
4  Departamento de Diagnóstico y Tratamiento, Hospital Italiano de Buenos Aires, Tte. Gral. Juan Domingo Perón 4190, C1199ABB, CABA. Argentina
5  Instituto de Oncología Ángel H. Roffo, Facultad de Medicina, Universidad de Buenos Aires, Av. San Martín 5481, C1417DTB, CABA, Argentina.


Kidney cancer is fundamentally a metabolic disease.1 Renal cell carcinoma (RCC) is among the 10 most common cancers worldwide.2, 3 More than 30% of patients, often incidentally diagnosed by imaging procedures, exhibit locally advanced or metastatic RCC at the time of diagnosis.4, 5 The disease is inherently resistant to chemotherapy6 and radiotherapy.7 Clear cell RCC (ccRCC) is the most common (75%) lethal subtype, and is considered a glycolytic and lipogenic tumor.8, 9 The present work consists on a lipid profiling study of serum samples from a cohort that included patients with different ccRCC stages (stage I, II, III and, IV; n=112) and healthy individuals (n=52). A discovery-based lipidomics approach using reverse phase ultraperformance liquid chromatography coupled to quadrupole-time-of-flight mass spectrometry was implemented to investigate the potential role of lipids in sample classification. Multivariate statistical analysis was conducted on a 386-feature matrix by means of machine learning algorithms using support vector machines (SVM) coupled with the least absolute shrinkage and selection operator (Lasso) variable selection method. This analysis provided a panel of 18 features that allowed discriminating healthy individuals from ccRCC patients with 96% accuracy, 93% specificity, and 100% sensitivity in a training set under cross-validation, and 79% accuracy, 100% specificity, and 79% sensitivity in an independent test set with an AUC of 0.89. A second multivariate model trained to discriminate early stages (I and II) from late stages (III and IV) ccRCC, yielded a panel of 26 features that allowed sample classification with 84% accuracy in the training set under cross-validation, and 82% accuracy in the classification of stage I ccRCC patients from an independent test set. Preliminary putative identification of discriminant lipids was based on exact mass, isotopic pattern and database search. Significant changes in lipid levels were evaluated after correcting for multiple testing between sample classes. Phosphatidylethanolamine levels were significantly decreased (p<0.001) in serum samples from ccRCC patients relative to controls. Significantly (p<0.02) decreased levels of fatty acids were detected in serum samples from ccRCC patients compared to healthy individuals, and along disease progression from early to late ccRCC stages. Current work involves the identification of the discriminant lipid panels by tandem MS experiments and chemical standards. Serum samples were provided by the Public Oncologic Serum Bank from Instituto de Oncología “Ángel H. Roffo” and Hospital Italiano de Buenos Aires.

(1) Linehan, W. M.; Srinivasan, R.; Schmidt, L. S., The genetic basis of kidney cancer: a metabolic disease. Nat. Rev. Urol. 2010, 7, 277-85.

(2) Linehan, W. M.; Bratslavsky, G.; Pinto, P. A.; Schmidt, L. S.; Neckers, L.; Bottaro, D. P.; Srinivasan, R., Molecular Diagnosis and Therapy of Kidney Cancer. Annu. Rev. Med. 2010, 61, 329-343.

(3) IARC Globocan 2012: Estimated Cancer Incidence, Mortality and Prevalence Worldwide in 2012.

(4) Hu, B.; Lara, P. N., Jr.; Evans, C. P., Defining an Individualized Treatment Strategy for Metastatic Renal Cancer. Urol. Clin. North Am. 2012, 39, 233-249.

(5) Graves, A.; Hessamodini, H.; Wong, G.; Lim, W. H., Metastatic renal cell carcinoma: update on epidemiology, genetics, and therapeutic modalities. Immunotargets Ther. 2013, 2, 73-90.

(6) Diamond, E.; Molina, A. M.; Carbonaro, M.; Akhtar, N. H.; Giannakakou, P.; Tagawa, S. T.; Nanus, D. M., Cytotoxic chemotherapy in the treatment of advanced renal cell carcinoma in the era of targeted therapy. Crit. Rev. Oncol. Hematol. 2015, 96, 518-526.

(7) De Meerleer, G.; Khoo, V.; Escudier, B.; Joniau, S.; Bossi, A.; Ost, P.; Briganti, A.; Fonteyne, V.; Van Vulpen, M.; Lumen, N.; Spahn, M.; Mareel, M., Radiotherapy for renal-cell carcinoma. Lancet Oncol. 2014, 15, e170-e177.

(8) Hsieh, J. J.; Purdue, M. P.; Signoretti, S.; Swanton, C.; Albiges, L.; Schmidinger, M.; Heng, D. Y.; Larkin, J.; Ficarra, V., Renal cell carcinoma. Nature Reviews Disease Primers 2017, 3, 17009.

(9) Hakimi, A. A.; Pham, C. G.; Hsieh, J. J., A clear picture of renal cell carcinoma. Nat. Genet. 2013, 45, 849-850.

Keywords: lipidomic; clear cell Renal Cell Carcinoma; biomarkers; ultraperformance liquid chromatography; mass spectrometry; machine learning; suppor vector machines; lasso