Diagnosis of Urological Cancer by 1H NMR Based Metabonomics Urinalysis: A Pilot Study
Leslie Clifford Noronha Araújo *
Urologic Service, Clinical Hospital, Universidade Federal de Pernambuco, UFPE, Brazil.
Flávia Cristina Morone Pinto
Postgraduate Program in Surgery, Department of Surgery, Center for Health Sciences, Universidade Federal de Pernambuco, UFPE, Brazil.
Tássia Brena Barroso Carneiro Costa
Department of Fundamental Chemistry, Universidade Federal de Pernambuco, UFPE, Brazil.
Ronaldo Dionísio Silva
Department of Fundamental Chemistry, Universidade Federal de Pernambuco, UFPE, Brazil.
Salvador Vilar Correia Lima
Postgraduate Program in Surgery, Department of Surgery, Center for Health Sciences, Universidade Federal de Pernambuco, UFPE, Brazil.
Ricardo Oliveira Silva
Department of Fundamental Chemistry, Universidade Federal de Pernambuco, UFPE, Brazil.
*Author to whom correspondence should be addressed.
Abstract
Aims: The most prevalent urological malignancies are prostate cancer (PC), bladder cancer (BC) and renal cancer (RC). The diagnosis of each of these diseases is conducted, in most cases, invasively and each procedure may lead to complications. The method of metabonomic spectrometry by nuclear magnetic resonance of hydrogen (1H NMR) provides pathways of diagnostic information that can identify pathologies without invasive procedures. The possibility of using this method for the diagnosis of those cancers by a single sample of urine has not been described yet.
Study Design: Prospective, observational.
Place and Duration of the Study: Department of Urology and Department Fundamental Chemistry of Universidade Federal de Pernambuco (UFPE), between July of 2015 to February of 2016.
Methodology: A sample of 3 ml of urine was collected from 25 volunteers distributed into 4 groups: A control group (07 volunteers), a PC (08 volunteers), a BC (05 volunteers), and an RC (05 volunteers). All samples underwent 1H MRI to generate spectra. A multivariate statistics analysis for the development of metabonomic models and comparison analysis groups was performed.
Results: These models showed a slight separation between the control group and each of the three groups of patients with oncological diseases. For the elaboration of the definitive models it was necessary to incorporate the volunteers of the BC and RC into one group (BC/RC). The metabonomic method when compared to control group, shown sensitivity of 90.9%, specificity of 100%, 100% PPV and NPV of 85.7% for CB/CR and sensitivity, specificity, PPV and NPV of 100% for the PC.
Conclusion: This pilot study demonstrates that the method is feasible with easy execution, showing simplicity besides being not invasive and allowing the diagnosis of oncological diseases with a single urine collection.
Keywords: Metabonomics, nuclear magnetic resonance, prostate cancer, bladder cancer, kidney cancer, diagnosis.