Mesothelioma MESO
Of 60,498 genes, 1,378 are higher in the tumor than in normal lung and in whole blood, and are not made by blood immune cells. 629 are protein-coding, and 618 of those have supporting plasma Cell-free RNA (cfRNA)RNA fragments that circulate in blood plasma outside cells. Most come from blood cells; a small share comes from other tissues, including tumors. evidence; the rest are mostly non-coding genes the plasma sources cannot check.
Samples compared
- Tumor
- 87
- TCGA primary tumors
- Normal tissue
- 288
- 288 GTEx lung
- Whole blood
- 337
- GTEx samples from 328 donors
- Survival
- 85
- patients, 73 events (Survival endpointsCurated TCGA outcomes (Liu et al. 2018): OS is overall survival, PFI the progression-free interval and DSS disease-specific survival. LB2 uses whichever has the most events in each cancer type.)
Results in brief
- 1,378
- candidate genes: 618 with plasma cfRNA evidence, 12 protein-coding without, 748 non-coding
- 45
- genes in the tissue panel, Held-out AUCHow well an elastic-net model separates tumor from normal tissue, measured with nested cross-validation on samples the model did not train on. It describes tissue, not the accuracy of a blood test. 0.996
- 40
- genes associated with overall survival (FDR < 0.05, 150 tested)
Candidates
Filter by plasma evidence or discrimination, find a gene, and select any point or row for its full record. The table holds all 1,378 candidates.
How the candidates were selected
60,498 genes tested; 2,251 higher in tumor than in normal lung; 2,216 also higher than in whole blood; 1,378 remain after removing genes expressed in blood immune cells; 618 of these have plasma cfRNA evidence, 12 protein-coding candidates have none, and 748 are non-coding genes the plasma sources cannot list.
Each comparison uses a two-sided Wilcoxon rank-sum test. A gene passes when its q-value is below 0.05, its log2 fold change is at least 1, and the lower bound of the 95% confidence interval for its AUC is at least 0.70.
Genes made by any of 18 sorted blood immune cell types above 1 nTPM are then removed, because blood cells supply most plasma RNA. Genes missing from that reference, mostly non-coding, are kept.
Plasma evidence only reorders the list. A candidate not yet seen in plasma stays in, because every dataset misses genes. Read the full methods.
Tissue classifier panel
An elastic-net logistic regression was trained on the 300 top-ranked candidates to tell tumor from normal tissue. Its Held-out AUCHow well an elastic-net model separates tumor from normal tissue, measured with nested cross-validation on samples the model did not train on. It describes tissue, not the accuracy of a blood test. is 0.996, from nested five-fold cross-validation.
Every normal sample here comes from GTEx, so the TCGA-versus-GTEx offset cannot be separated from the tumor/normal contrast and no study correction was possible: the classifier may partly learn study differences.
This number describes tumor and normal tissue, where separation is expected to be near perfect. It is not the accuracy of a blood test: that has to be measured in plasma from patients and controls.
Association with survival
- KIF4A2.12 (1.58–2.83)1.4 × 10−5
- EXO12.11 (1.61–2.77)5.0 × 10−6
- CENPA2.10 (1.59–2.78)1.1 × 10−5
- CENPF1.98 (1.46–2.68)3.1 × 10−4
- OXTR1.93 (1.52–2.45)5.0 × 10−6
- GINS11.89 (1.42–2.52)3.1 × 10−4
- PBK1.79 (1.35–2.39)0.0013
- UCHL11.70 (1.26–2.29)0.0045
- CD2761.63 (1.30–2.02)3.1 × 10−4
- APCDD1L1.63 (1.24–2.13)0.0039
- RAC31.62 (1.25–2.11)0.0034
- COL11A11.59 (1.25–2.04)0.0028
- TWIST11.56 (1.22–2.01)0.0045
- GPRIN11.55 (1.25–1.93)0.0014
- COL1A11.48 (1.19–1.85)0.0045
- CALB20.70 (0.57–0.86)0.0055
- CFB0.69 (0.56–0.85)0.0045
- EFNA50.67 (0.54–0.84)0.0039
- C4B0.62 (0.47–0.80)0.0034
- C4A0.61 (0.47–0.79)0.0028
Show the other 20 genes
- ZNF6951.63 (1.20–2.20)0.0094
- C1QL41.51 (1.18–1.93)0.0077
- FGD11.49 (1.15–1.93)0.012
- ZNF4691.48 (1.18–1.86)0.0058
- FOXL21.47 (1.14–1.89)0.013
- GNG41.47 (1.15–1.87)0.0094
- CHRNA51.46 (1.15–1.84)0.0094
- CRABP21.44 (1.15–1.82)0.0094
- DIAPH31.44 (1.14–1.82)0.012
- CAMK2N21.41 (1.13–1.77)0.012
- ASPM1.41 (1.12–1.78)0.017
- SOX111.39 (1.09–1.77)0.026
- SLC12A81.37 (1.09–1.72)0.025
- HSF2BP1.33 (1.09–1.63)0.020
- WT10.76 (0.62–0.92)0.023
- SLC13A30.75 (0.60–0.94)0.049
- PRR150.71 (0.57–0.88)0.0094
- MUC160.70 (0.56–0.88)0.010
- SLC7A40.70 (0.55–0.89)0.015
- TMEM151A0.69 (0.55–0.87)0.0094
For the 150 top-ranked candidates, a Cox model relates tumor expression, as a continuous value, to overall survival in 85 patients (73 events). No high/low cutpoint is searched for, since optimized cutpoints inflate false positives.
A Hazard ratio per SDFrom a Cox model with the gene’s tumor expression as a continuous variable: the change in hazard for each one-standard-deviation increase. Above 1, higher expression goes with a shorter time to the event; below 1, with a longer one. above 1 means higher expression goes with a shorter overall survival. These are associations in tissue, not evidence that a blood level predicts outcome.
Caveats for MESO
- GTEx has no pleura or peritoneum, so normal lung stands in for the tissue of origin.
- Every normal sample comes from GTEx, a different study from the tumors, so study differences cannot be corrected in the tissue comparison.
- With only 87 tumors, confidence intervals are wide; treat close ranks as ties.
- The blood comparison sets TCGA tumors against GTEx blood. Study and biology cannot be separated there, which is why genes made by blood immune cells are removed as well.