A Complete Pipeline for Untargeted Urinary Volatolomic Profiling with Sorptive Extraction and Dual Polar and Nonpolar Column Methodologies Coupled with Gas Chromatography Time-of-Flight Mass Spectrometry

Abstract

Volatolomics offers an opportunity for noninvasive detection and monitoring of human disease. While gas chromatography–mass spectrometry (GC–MS) remains the technique of choice for analyzing volatile organic compounds (VOCs), barriers to wider adoption in clinical practice still exist, including: sample preparation and introduction techniques, VOC extraction, throughput, volatolome coverage, biological interpretation, and quality control (QC). Therefore, we developed a complete pipeline for untargeted urinary volatolomic profiling. We optimized a novel extraction technique using HiSorb sorptive extraction, which exhibited high analytical performance and throughput. We achieved a broader VOC coverage by using HiSorb coupled with a set of complementary chromatographic methods and time-of-flight mass spectrometry. Furthermore, we developed a data preprocessing strategy by evaluating internal standard normalization, batch correction, and we adopted strict QC measures including removal of nonlinearly responding, irreproducible, or contaminated metabolic features, ensuring the acquisition of high-quality data. The applicability of this pipeline was evaluated in a clinical cohort consisting of pancreatic ductal adenocarcinoma (PDAC) patients (n = 28) and controls (n = 33), identifying four urinary candidate biomarkers (2-pentanone, hexanal, 3-hexanone, and p-cymene), which can successfully discriminate the cancer and noncancer subjects. This study presents an optimized, high-throughput, and quality-controlled pipeline for untargeted urinary volatolomic profiling. Use of the pipeline to discriminate PDAC from control subjects provides proof of principal of its clinical utility and potential for application in future biomarker discovery studies.

Rapid Diagnosis of Urinary Tract Cancers on a LEGO-Inspired Detection Platform via Chemiresistive Profiling of Volatile Metabolites

“…Those values were well below the clinically significant VOC levels in the headspace of urine (25− 200 ppm). 27,28 To demonstrate its usefulness in long-term and recyclable volatile monitoring, the sensor array was tested against 10 ppm of hexanal at room temperature, which generally displayed prompt (response time <30 s, Figure 1h) and repeatable (over 20 cycles, Figure S11) response−recovery kinetics. Such results were comparable to a variety of engineered, high-quality MXene-based gas sensors for clinical monitoring.…”

“…Chemiresistive changes of the MXene–TMDC heterojunction declined linearly as a function of the concentration of VOCs, and the limits of detection for the above three VOCs were 3.2, 4.7, and 4.9 ppm, respectively (Figures g and S10). Those values were well below the clinically significant VOC levels in the headspace of urine (25–200 ppm). , To demonstrate its usefulness in long-term and recyclable volatile monitoring, the sensor array was tested against 10 ppm of hexanal at room temperature, which generally displayed prompt (response time <30 s, Figure h) and repeatable (over 20 cycles, Figure S11) response–recovery kinetics. Such results were comparable to a variety of engineered, high-quality MXene-based gas sensors for clinical monitoring. , It is worth noting that the doping of TMDCs and glycerol as a protecting agent prevents MXene from oxidation and effectively extends the shelf life of those nanosensors to over 60 days (Figure S12).…”

Global Urinary Volatolomics with (GC×)GC-TOF-MS

“…Siloxanes (artifacts generated either from chromatographic columns or extraction sorbents), features with signal-to-noise ratio (S/N) < 3, and annotated features whose reverse matched factor (RMF) < 800 were not further analyzed. A pooled quality control (QC) approach, which has been described elsewhere, was applied where preprocessing steps were evaluated with principal component analysis (PCA) . Briefly, features with either (i) CV > 30% in the pooled QC samples or (ii) blank average levels < 30% in nanopure water compared to their corresponding levels in the pooled QC or (iii) 1-tailed Spearman’s rho > 0.7 in the dilution series and q value < 0.05 after Benjamini–Hochberg correction were removed from further analysis.…”

“…Clinical applicability was assessed by analyzing the urine of 28 pancreatic ductal adenocarcinoma (PDAC) patients and 33 cancer-free control patients with benign pancreatic pathologies (REC reference 17/WA/016 and 14/LO/1136). Cohort details are described elsewhere . Briefly, the inclusion criteria were as follows: (i) adult PDAC patients and (ii) adult control patients with normal a upper gastrointestinal tract on computed tomography (CT).…”

“…4 We had previously developed a pipeline for urinary volatolomics focusing on the nonpolar fraction of the volatolome and by employing single-dimension gas chromatography online with time-of-flight mass spectrometry (GC-TOF-MS). 6 Herein, we report an updated methodology for urinary VOC analysis, where we (i) achieve a global coverage of the volatolome by using multiple sorbent materials with varied polarities, (ii) increase throughput by consolidating our previous two single-dimension GC methods to a unified one, (iii) minimize urine sample consumption 5-fold (from 2 to 0. AAXX-5149), and DiffLok TD caps were provided from Markes International (Llantrisant, UK).…”