ABSTRACT
To compare results from messenger RNA (mRNA)-based TargetPrint testing with those from immunohistochemistry (IHC) and in situ hybridization (ISH) conducted according to local standard procedures at hospitals worldwide. Tumor samples were prospectively obtained from 806 patients at 22 hospitals. The mRNA level of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) was assessed by TargetPrint quantitative gene expression readouts. IHC/ISH assessments were performed according to local standards at the participating hospitals. TargetPrint readout showed a high concordance with IHC/ISH of 95 % (kappa 0.81) for ER, 81 % (kappa 0.56) for PR, and 94 % (kappa 0.76) for HER2. The positive/negative agreement between TargetPrint and IHC for ER, PR, and HER2 was 96 %/87 %, 84 %/74 %, and 74 %/98 %, respectively. The concordance rate in IHC/ISH results between hospitals varied: 88-100 % for ER (kappa 0.50-1.00); 50-100 % for PR (kappa 0.20-1.00); and 90-100 % for HER2 (kappa 0.59-1.00). mRNA readout of ER, PR, and HER2 status by TargetPrint was largely comparable to local IHC/ISH analysis. However, there was substantial discordance in IHC/ISH results between different hospitals. When results are discordant, the use of TargetPrint would improve the reliability of hormone receptor and HER2 results by prompting retesting in a reference laboratory.
Subject(s)
Breast Neoplasms/metabolism , Breast Neoplasms/pathology , Receptor, ErbB-2/metabolism , Receptors, Estrogen/metabolism , Adult , Aged , Aged, 80 and over , Biomarkers, Tumor/analysis , Female , Gene Expression Regulation, Neoplastic , Humans , Immunohistochemistry/methods , Middle Aged , Receptors, Progesterone/metabolism , Reproducibility of Results , Young AdultABSTRACT
BACKGROUND: The 4Kscore combines measurement of four kallikreins in blood with clinical information as a measure of the probability of significant (Gleason ≥7) prostate cancer (PCa) before prostate biopsy. OBJECTIVE: To perform the first prospective evaluation of the 4Kscore in predicting Gleason ≥7 PCa in the USA. DESIGN, SETTING, AND PARTICIPANTS: Prospective enrollment of 1012 men scheduled for prostate biopsy, regardless of prostate-specific antigen level or clinical findings, was conducted at 26 US urology centers between October 2013 and April 2014. INTERVENTION: The 4Kscore. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: The primary outcome was Gleason ≥7 PCa on prostate biopsy. The area under the receiver operating characteristic curve, risk calibration, and decision curve analysis (DCA) were determined, along with comparisons of probability cutoffs for reducing the number of biopsies and their impact on delaying diagnosis. RESULTS AND LIMITATIONS: Gleason ≥7 PCa was found in 231 (23%) of the 1012 patients. The 4Kscore showed excellent calibration and demonstrated higher discrimination (AUC 0.82) and net benefit compared to a modified Prostate Cancer Prevention Trial Risk Calculator 2.0 model and standard of care (biopsy for all men) according to DCA. A possible reduction of 30-58% in the number biopsies was identified with delayed diagnosis in only 1.3-4.7% of Gleason ≥7 PCa cases, depending on the threshold used for biopsy. Pathological assessment was performed according to the standard of care at each site without centralized review. CONCLUSION: The 4Kscore showed excellent diagnostic performance in detecting significant PCa. It is a useful tool in selecting men who have significant disease and are most likely to benefit from a prostate biopsy from men with no cancer or indolent cancer. PATIENT SUMMARY: The 4Kscore provides each patient with an accurate and personalized measure of the risk of Gleason ≥7 cancer to aid in decision-making regarding the need for prostate biopsy.
