DataSupplement

Supplemental Data Table 1Clinical characteristics of patients and healthy donors

Patients with TCC
Age
(median) / % of population / Male
(n) / Female
(n) / Tumor
grade,stage* / Male
(n) / Female
(n) / Age, median
40-60 (57) / 24.6 / 11 / 4 / low grade,pTa / 10 / 3 / 70
61-70 (69) / 21.3 / 10 / 3 / high grade,
pTa-pTis / 15 / 4 / 72
71-80 (74) / 34.4 / 18 / 3 / high grade,pT1 / 11 / 1 / 70
81-90 (84) / 16.4 / 4 / 6 / high grade,
pT2-pT3 / 9 / 8 / 72
<90 (95) / 3.3 / 1 / 1
Healthy donors
Age
(median) / % of
population / Male
(n) / Female
(n)
40-60 (53) / 16.2 / 4 / 2
61-70 (65) / 51.4 / 11 / 8
71-80 (76) / 16.2 / 6 / /
81-90 (82) / 16.2 / 2 / 4

*low grade tumors = G1;high grade tumors = G2,G3

Materials and Methods

Isolation of total RNA and reverse transcription of RNA – standard protocol

Prior to RNA isolation, the pH of the buffered urine solution was adjusted to pH 7 with 1 mol/L HEPES buffer (pH 7.0) and 107 copies of RNALUC were added. Total RNA was isolated from different urine fractions using the RNeasy Midi Kit (Qiagen, Hilden, Germany) including an on-column digestion of genomic DNA with DNase I. RNA isolation was performed according to manufacturer’s instructions except that the lysis buffer described above was used instead of buffer RLT. RNA was eluted twice with 160 µL of nuclease-free H2O and then lyophilized. RNA pellets were resolved in 20 µL of nuclease-free H2O and 10 µL of RNA extract was used for cDNA synthesis and non-RT reaction. Additionally, 107 copies of RNALUC were directly reverse-transcribed to verify applied copies. Reverse transcription was performed in a total volume of 20 µL. RNA extract (10 µL), 300 ng random hexamer primer (Invitrogen, Paisley, UK) and each 0.5 mmol/L deoxynucleotide triphosphates (MBI Fermentas) were denaturated for 10 min at 75°C followed by incubation on ice for 1 min and annealing at 25°C for 10 min. Reaction buffer (1x), 200 units Superscript III Reverse Transcriptase, 5 mmol/L dithiothreitol, and 40 units RNaseOut (Invitrogen) were added and RNA was reverse transcribed at 50°C for 60 min. RT was inactivated at 70°C for 15 min. In the case of non-RT reactions, nuclease-free H2O was used instead of RNaseOut and Reverse Transcriptase.

Quantitative PCR (qPCR) and data analysis

Primers and TaqMan® probes were designed using PrimerExpress® software v. 2.0 (Applied Biosystems, Darmstadt, Germany) or Primer3 software (Steve Rozen, Whitehead Institute for Biomedical Research, Cambridge, UK) and were purchased from Metabion (Martinsried, Germany). The sequences and amplicon characteristics are listed in the supplemental Table 2. Primer sequences were checked for homology using the Blast software (ww.ncbi.nlm.nih.gov/BLAST). Amplicons were designed with primers or TaqMan® probes positioned at exon/exon boundaries (except for the ubiquitin C and the CC3 amplicon with intron-spanning primers and the BCL2L1 amplicon with primers and probe within one exon) to assure that genomic DNA was not amplified within the qPCR. All reactions were performed with the qPCRTM Core Kit (Eurogentec, Seraing, Belgium) in a total reaction volume of 10 µL (384-well plate). Reactions contained the following components: 2 µL of DNA standard, diluted cDNA or non-RT, 1x reaction buffer, 3 mmol/L MgCl2, 200 µmol/L each deoxynucleotide triphosphate, 0.025 units/µL Hot Goldstar enzyme, 200 nmol/L of forward and reverse primer, 125 nmol/L TaqMan® probe. A non-template control (nuclease-free H2O) was included for each amplicon to exclude contamination in every qPCR run. Each qPCR was performed in triplicate. Quantitative PCR was carried out in a 7900HT thermal cycler (Applied Biosystems) using a two-step protocol with an initial denaturation step at 95°C for 10 min, followed by 50 cycles at 95°C for 15 sec and at 60°C for 60 sec. PCR products from each target amplicon were purified using the MinElute PCR Purification Kit (Qiagen). Six serial 10-fold dilutions (101–106 copy numbers/reaction) were prepared in 10 mmol/L Tris/HCL (pH 8.0), 10 ng/mL polyinosinic acid potassium salt to generate standard curves. Data analysis was performed with SDS 2.1 software (Applied Biosystems) and the threshold cycle (Ct) valuesof amplified targets were transformed into absolute RNA copy numbers using the standard curves.

Supplemental Data Table 2Basic data of qPCR amplicons

Target
Symbol / Sequence accession ID / Sequence (5´ − 3´ direction)
  1. forward primer
  2. reverse primer
  3. TaqMan probe*
/ Amplicon
size (bp) / qPCR efficiency
E=10[-1/slope] / R2 of standard curve
1. TGGAGCTGCAGAGGATGATT
BAX / NM_004324 / 2. AGCTGCCACTCGGAAAAAGA / 71 / 1.99 / 0.9978
3. FAM-CCGCCGTGGACACAGACTCCC-BHQ-1
TGCGTGGAAAGCGTAGACA
BCL2L1 / NM_138578 / GCTCTAGGTGGTCATTCAGGTAAGT / 88 / 1.98 / 0.9987
FAM-TGGTGAGTCGGATCGCAGCTTG-BHQ-1
GTCTGCAGAGCTGGCAAAAG
CC3 / NM_006410 / TCAACCTTGGCTTCTACTTCTCC / 123 / 1.96 / 0.9989
YY-TGGAGGGTGCAAACATTTCAACTTGC-BHQ-1
CCAAGAAAAGACAGAAGATCAATATGAA
ETS2 / NM_005239 / GTGCCAAAACCTAATGTATTGCTG / 90 / 1.97 / 0.9992
FAM-CACCTCACCTCCGTTCCTCATTGGA-BHQ-1
GACAGTCAGCCGCATCTTCTT
GAPDH / AF261085 / TCCGTTGACTCCGACCTTC / 86 / 1.96 / 0.9987
YY-CCAGCCGAGCCACATCGCTG-BHQ-1
CTGGAGAAACTGCTGCCTCAT
RPLP0 / NM_001002 / CACCTTATTGGCCAGCAACA / 99 / 1.95 / 0.9971
FAM-CCGGGGGAATGTGGGCTTTG-BHQ-1
CGGACTTTGGGTGCGACTT
Ki-67 / NM_002417 / CAACTCTTCCACTGGGACGAT / 77 / 1.93 / 0.9973
FAM-ACGAGCGGTGGTTCGACAAGTG-TAMRA
AAAGACGCAAGTCCCATGAAG
OP18 / NM_203401 / AGCTTCCATTTTGTGGGTCAG / 146 / 1.95 / 0.9987
FAM-GCAGCTGGCTGAGAAACGAGAGCA-BHQ-1
GAACATCACGTACGCGGAATAC
RNALUC / TTTCACTGCATACGACGATTCTG / 104 / 1.96 / 0.9985
FAM-TCGAAATGTCCGTTCGGTTGGCA-BHQ-1
GATTTGGGTCGCGGTTCTT
UBC / NM_021009 / GATGGTGTCACTGGGCTCAA / 119 / 1.95 / 0.9976
FAM-TGGATCGCTGTGATCGTCACTTGA-BHQ-1
CCTGCCCTCGATGTATAACGAT
uPA / BC013575 / CCGGATAGAGATAGTCGGTAGAATTC / 92 / 1.98 / 0.9975
FAM-CCCAGTTTGGCACAAGCTGTGAGATCA-BHQ-1
CGTCATGATTGAGCAAGAATGC
UPK1A / NM_007000 / CGGAAGGCTGACGTGAAGT / 72 / 1.97 / 0.9979
FAM-TGGCACATCTGGTCCCATGGA-BHQ-1

* FAM = 6-carboxyfluorescein; BHQ = Black Hole Quencher™; YY = Yakima Yellow; TAMRA = 5-Carboxytetramethylrhodamine

Statistical analysis of marker ratios

After univariate investigation of the marker ratios, we performed multivariate analyses for a dichotomous group variable. Classification trees have been used to select the most important predictors for further analysis using the R-package random Forest (computing 5000 trees and with an importance-statement) (1). Linear relationships of the selected marker ratios were confirmed using multivariable fractional polynomials (2). We analyzed the selected ratios in a logistic regression model with backward elimination considering all interactions of the identified ratios.

Additionally, we investigated the selected marker ratios using a nonparametric diagonal linear discriminant analysis (DLDA). We used a k-nearest neighbor rule with k = 3 as nonparametric method for classification. We analyzed two models, a DLDA including all identified marker ratios and a DLDA including only one ratio. The results of the group classification of the DLDA for each patient were compared with the real group membership (true or false). Differences between the results of the two models were tested using an exact McNemar test.

References

1. Breiman, L. Random Forests. Machine Learning 2001;45:5-32.

2. Sauerbrei W, RoystonP.Building multivariable prognostic and diagnostic models: Transformation of the predictors by using fractional polynomials. Journal of the Royal Statistical Society A 1999;162: 71-94. Corrigendum: JRSS 2002;165: 399-400.

Supplemental Data Table 3AUCs of investigated RNA marker ratios using whole urine from 37 healthy donors and 61 tumor patients

RNA marker ratio / AUC / Asymptotic confidence interval, 95%
Upper limit / Lower limit
ETS2/uPA / 0.929 / 0.882 / 0.976
ETS2/CC3 / 0.757 / 0.657 / 0.857
ETS2/BAX / 0.755 / 0.656 / 0.853
ETS2/BCL2L1 / 0.727 / 0.622 / 0.832
ETS2/UPK1A / 0.665 / 0.556 / 0.773
Ki-67/uPA / 0.657 / 0.546 / 0.768
Ki-67/BCL2L1 / 0.643 / 0.528 / 0.758
ETS2/OP18 / 0.638 / 0.520 / 0.756
ETS2/GAPDH / 0.638 / 0.527 / 0.750
Ki-67/CC3 / 0.627 / 0.509 / 0.746
Ki-67/UPK1A / 0.625 / 0.511 / 0.738
Ki-67/OP18 / 0.614 / 0.494 / 0.733
OP18/UPK1A / 0.590 / 0.478 / 0.702
CC3/uPA / 0.588 / 0.473 / 0.704
OP18/BAX / 0.576 / 0.463 / 0.690
OP18/uPA / 0.575 / 0.459 / 0.691
OP18/CC3 / 0.574 / 0.460 / 0.689
Ki-67/GAPDH / 0.572 / 0.450 / 0.693
BAX/uPA / 0.570 / 0.455 / 0.686
OP18/BCL2L1 / 0.564 / 0.450 / 0.679
CC3/BAX / 0.559 / 0.445 / 0.674
BAX/UPK1A / 0.557 / 0.443 / 0.672
BCL2L1/UPK1A / 0.527 / 0.410 / 0.644
BCL2L1/uPA / 0.527 / 0.410 / 0.644
CC3/UPK1A / 0.524 / 0.410 / 0.638
Ki-67/ETS2 / 0.519 / 0.390 / 0.649
CC3/BCL2L1 / 0.519 / 0.401 / 0.637
BAX/BCL2L1 / 0.514 / 0.391 / 0.637
OP18/GAPDH / 0.447 / 0.333 / 0.561
BCL2L1/GAPDH / 0.373 / 0.263 / 0.482
CC3/GAPDH / 0.370 / 0.261 / 0.478
UPK1A/GAPDH / 0.367 / 0.258 / 0.477
BAX/GAPDH / 0.366 / 0.258 / 0.473
uPA/GAPDH / 0.359 / 0.250 / 0.469
BAX/Ki-67 / 0.353 / 0.236 / 0.470