• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

Table 12Evidence Table. Analytic validity, H/I ratio

Study, yearMeasureConclusions
Goetz, 200662Context: Validation of the H/I ratio signature using patients from the randomized NCCTG 89-30-52 trial on tamoxifen treatment. Out of the 256 eligible patients, FFPE were available for 227 patents
  • LMC was performed prior total RNA preparation
  • RT-PCR expression values for each gene were normalized using a standard curve obtained analyzing the human universal total RNA (Stratagen, La Jolla, CA)
  • No reference genes were used
  • 206 patients out of 256 successfully analyzed, for the following reasons:
 No specimen available: 29/256 patients (11%)
 Tumor content sufficient: 211/227 patients (93%)
 Failed RT-PCR: 5/211 patients (2.4%)
 Successful assays: 206/211 patients (97.6%)
Jansen, 200772Context: Validation of the H/I ratio signature by measuring expression levels normalized to a different set of control genes respect to MA et al 200661 using fresh frozen samples. From a population of 1693, 1,252 were considered eligible and were successfully evaluated
  • RT-PCR was use to measure the expression levels of HOXB13 and IL17BR
  • Two pairs of primers-probes were used to evaluated the IL17BR gene, aligning to the 3′ and the 5′ end of the genes and corresponding to the region of the transcripts respectively assayed by Ma et al 200464 and Reid et al 200569
  • Three reference genes different from those used by Ma et 200661 were used (porphobilinogen deaminase, hypoxanthine-guanine phospho-ribosyltransferase, and beta-2-microglobulin)
In this study analytic validity data were not reported
Jerevall, 200763Context: Validation of the H/I ratio signature. Expression levels were normalized to b-actin using fresh frozen samples. Patients were collected from two distinct institutions; of 373 tumor samples analyzed, RNA expression data were obtained from 357 tumors
  • RT-PCR reactions in the two institutions were performed using the same sets of primers/probes and two distinct instruments
  • In each reaction serial dilutions of a standard sample were used to construct a standard curve used to quantify gene expression prior normalization
  • The reproducibility between the two institution was assessed by Pearson's correlation
  • 357 patients out of 373 successfully analyzed, for the following reasons:
 Insufficient RNA yield: 16/373 patients (4.3%)
 Overall success rate: 357/373 patients (95.7%%)
  • Reproducibility between institutions, Pearson's correlations, 10 patients:
 HOXB13:b-actin, r = 0.96, P < 0.001
 IL17BR:b-actin, r = 0.87, P = 0.002
 HOXB13:IL17BR, r = P < 0.001
  • Median values for all the samples analyzed at the two centers:
 HOXB13:b-actin, 0.086 vs. 0.081
 IL17BR:b-actin, 1.4 vs. 1.3)
 HOXB13:IL17BR, 0.074 vs. 0.055
Ma, 200464Context: This study used the H/I ratio signature for the development and preliminary validation of the of the two-gene ratio signature. The training set was analyzed by microarray (n=60) and by RT-PCR (n=59)
  • Comparison of microarray based and RT-PCR based ratio by Pearson's correlation on 59/60 patients
  • Correlations between array and RT-PCR:
 HOXB13, r = 0.83
 IL17BR, r = 0.93
 HOXB13/IL17BR, r = 0.83
Ma, 200661Context: Development of the HOXB13:IL17BR two-gene index. Out of the 1,002 eligible The FFPE tissue microarrays were 4 years old, and obtained from specimens originally stored as fresh frozen tissuesAccording to the authors these results confirmed the significant correlations between mRNA and protein levels for ER and PR and provided validation of the FFPE gene expression assay platform.
  • IHC for ER and PR was performed90,91
  • IHC Allred92 scores of 3 to 8 were considered positive for ER or PR93
  • Concordance between RT-PCR and IHC results by k statistics
  • Since both ER and PR mRNA RT-PCR measurements were bimodal; the following midpoints were used as cutoffs:
 2.5 CT for ER
 5.9 for PR
  • The quality of these tissue microarrays FFPE specimens was comparable with those without the intervening snap-freeze step (data not shown)
  • 852 patients out of 1002 successfully analyzed, for the following reasons:
 Tumor content < 10%: 132/1002 patients (13.2%)
 Poor RNA yield: 18/870 patients (2%)
 Successful assays: 852/870 (98%)
  • Agreement between IHC vs RT-PCR:
 ER, 91% concordance, k = 0.83, P value = .0001
 PR, 85% concordance, k = 0.70, P value = .0001

NCCTG= North Central Cancer Treatment Group; FFPE = formalin fixed paraffin embedded; LMC= laser micro-dissection; RNA= ribonucleic acid; RT-PCR = reverse transcriptase polymerase chain reaction; IHC = immunohistochemistry; ER = estrogen receptor; PR = progesterone receptor; mRNA= messenger ribonucleic acid.

From: Appendix I: Evidence Tables

Cover of Impact of Gene Expression Profiling Tests on Breast Cancer Outcomes
Impact of Gene Expression Profiling Tests on Breast Cancer Outcomes.
Evidence Reports/Technology Assessments, No. 160.
Marchionni L, Wilson RF, Marinopoulos SS, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.