Table 3.
Studies on volatile organic compounds for the identification of prostate cancer biomarkers.
| Sample Groups | Analytical Approach | Biomarker’s Candidates | Prediction Model | Validation Characteristics | Reference |
|---|---|---|---|---|---|
| Urine | |||||
| PCa: 59 HC: 43 |
HS-SPME/GC–MS | 2,6-dimethyl-7-octen-2-ol, pentanal, 3-octanone, 2-octanone | Repeated 10-Fold CV, Repeated Double CV |
NA | [87] |
| PCa: 58 HC: 60 |
HS-SPME/GC–MS | hexanal, 2,5-dimethylbenzaldehyde, 4-methylhexan-3-one, dihydroedulan IA, methylglyoxal, 3-phenylpropionaldehyde | PLS-DA, ROC | Sensitivity: 89% Specificity: 83% AUC: 0.904 |
[39] |
| PCa: 20 BC: 20 RC: 20 HC: 20 |
HS-SPME/GC–MS | methylglyoxal, hexanal, 3-phenylpropionaldehyde, 4-methylhexan-3-one, 2,5-dimethylbenzaldehyde, dihydroedulan IA, ethylbenzene, heptan-2-one, heptan-3-one, 4-(2-methylpropoxy)butan-2-one, methyl benzoate, 3-methyl-benzaldehyde | PLS-DA | Sensitivity: 76% Specificity: 97% AUC: 0.90 |
[80] |
| PCa: 32 HC: 32 |
GC–MS | VOMs involved in amino acids, purine, glucose, urea, Krebs cycle biochemical pathways | PCA, PLS-DA | NA | [88] |
| PCa: 108 HC: 75 |
SBSE/TD-GC–MS | 11 VOMs | ROC | Sensitivity: 87% Specificity: 77% AUC: 0.86 |
[89] |
| PCa: 29 BPH: 21 |
HS-SPME/GC–MS | furan, p-xylene | - | NA | [90] |
| BC: 15 PCa: 55 HC: 36 |
GC–TOF-MS and GC–IMS | 35 VOMs | ROC, Repeated 10-Fold CV | GC–IMS method Sensitivity: 87% Specificity: 92% AUC: 0.95 GC–TOF-MS method Sensitivity: 78% Specificity: 88% AUC: 0.94 |
[55] |
| PCa: 88 HC: 86 |
Urine headspace conditioning, followed by e-nose analysis | The study tested the ability of urinary volatilome profiling to distinguish patients with PCa from HC | ROC | Sensitivity: 85.2% Specificity: 79.1% AUC: 0.82 |
[92] |
| PCa: 133 HC: 139 |
Urine headspace conditioning, followed by e-nose analysis (Cyranose C320) | The study tested the ability of urinary volatilome profiling to distinguish patients with PCa from HC | PCA, ROC | Sensitivity: 82.7% Specificity: 88.5% AUC: 0.90 |
[54] |
| PCa: 132 HC: 60 |
Urine headspace conditioning, followed by e-nose analysis (Cyranose C320) | The study tested the ability of urinary volatilome profiling to distinguish patients with PCa from HC | PCA | Sensitivity: 82% Specificity: 87% AUC: NA |
[93] |
| HCC: 31 PCa: 62 BC: 29 HC: 18 |
SPME, followed by analysis with polymer tabs sensor |
The study tested the ability of urinary volatilome profiling to distinguish patients with PCa from HC | PCA, ROC | PCa detection Sensitivity: 70% Specificity: NA AUC: 0.70 |
[94] |
| Exhaled breath | |||||
| LC: 30 CC: 26 BTC: 22 PCa: 18 HC: 22 |
HS-SPME/GC–MS | 6 VOMs for LC, 6 VOMs for CC, 5 VOMs for BTC, 4 VOMs for PCa | PCA | NA | [91] |
| PCa: 32 HC: 53 |
E-nose analysis (Cyranose C320) | The study tested the ability of exhaled breath volatilome profiling to distinguish patients with PCa from HC | ANN | Sensitivity: 84% Specificity: 70 AUC: 0.79 |
[95] |
Legend: ANN, artificial neural network; AUC, area under the ROC curve; BC, bladder cancer; BPH, benign prostate hyperplasia; BTC, breast cancer; CC, colon cancer; CV, cross-validation; GC–IMS, gas chromatography–ion mobility spectrometry; GC–MS, gas chromatography–mass spectrometry; GC–TOF-MS, gas chromatography coupled to time-of-flight mass spectrometry; HC, healthy control; HCC, hepatocellular cancer; HS-SPME, headspace solid-phase microextraction; kNN, k-nearest neighbor; LC, lung cancer; NA, not analyzed; PCA, principal component analysis; PLS-DA, partial least-squares discriminant analysis; RC, renal cancer; ROC, receiver operating characteristic; SBSE, stir bar sorptive extraction; TD-GC–MS, thermal desorption gas chromatography–mass spectrometry.