PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 7973233

Efficient confidence bounds for ROC curves.

H Schäfer
View original source ↗Export metadata ↓

Abstract

The specificity and sensitivity of a quantitative diagnostic marker depends on the selected cut-off point. The ROC curve is generated by plotting sensitivity against specificity as the cut-off point runs through the whole range of possible marker values. Confidence bounds for ROC curves may be constructed by combining two separate confidence intervals for sensitivity and specificity. In the present paper, a more efficient method, which is based on a statistical test introduced by Greenhouse and Mantel, is presented. The resulting 95 per cent confidence intervals are up to 40 per cent smaller. A formula is given to calculate the sample size necessary to obtain a confidence interval of a stipulated length at a specified point of the ROC curve. As an example, the performance of a tumour marker in the diagnosis of bone marrow metastases is investigated.

Explore related subjects

Bone MarrowBone Marrow DiseasesConfidence IntervalsHumansL-Lactate DehydrogenaseMonte Carlo MethodNeoplasm MetastasisROC CurveSample SizeSensitivity and SpecificityJournal ArticleResearch Support, Non-U.S. Gov't

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

H Schäfer. 1994-08-15. Efficient confidence bounds for ROC curves.. https://doi.org/10.1002/sim.4780131506

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2014

Single cell mutational analysis of PIK3CA in circulating tumor cells and metastases in breast cancer reveals heterogeneity, discordance, and mutation persistence in cultured disseminated tumor cells from bone marrow.

BACKGROUND: Therapeutic decisions in cancer are generally guided by molecular biomarkers or, for some newer therapeutics, primary tumor genotype. However, because biomarkers or genotypes may change as new metastases emerge, circulating tumor cells (CTCs) from blood are being investigated for a role in guiding real-time drug selection during disease progression, expecting that CTCs will comprehensively represent the full spectrum of genomic changes in metastases. However, information is limited regarding mutational heterogeneity among CTCs and metastases in breast cancer as discerned by single cell analysis. The presence of disseminated tumor cells (DTCs) in bone marrow also carry prognostic significance in breast cancer, but with variability between CTC and DTC detection. Here we analyze a series of single tumor cells, CTCs, and DTCs for PIK3CA mutations and report CTC and corresponding metastatic genotypes. METHODS: We used the MagSweeper, an immunomagnetic separation device, to capture live single tumor cells from breast cancer patients' primary and metastatic tissues, blood, and bone marrow. Single cells were screened for mutations in exons 9 and 20 of the PIK3CA gene. Captured DTCs grown in cell culture were also sequenced for PIK3CA mutations. RESULTS: Among 242 individual tumor cells isolated from 17 patients and tested for mutations, 48 mutated tumor cells were identified in three patients. Single cell analyses revealed mutational heterogeneity among CTCs and tumor cells in tissues. In a patient followed serially, there was mutational discordance between CTCs, DTCs, and metastases, and among CTCs isolated at different time points. DTCs from this patient propagated in vitro contained a PIK3CA mutation, which was maintained despite morphological changes during 21 days of cell culture. CONCLUSIONS: Single cell analysis of CTCs can demonstrate genotypic heterogeneity, changes over time, and discordance from DTCs and distant metastases. We present a cautionary case showing that CTCs from any single blood draw do not always reflect metastatic genotype, and that CTC and DTC analyses may provide independent clinical information. Isolated DTCs remain viable and can be propagated in culture while maintaining their original mutational status, potentially serving as a future resource for investigating new drug therapies.

Bone Marrow↗
PubMed2007

Trisomy 8 as sole cytogenetic abnormality in a case of chronic lymphocytic leukemia.

A 49-year-old man, who had been diagnosed with chronic lymphocytic leukemia (CLL) in 2002, had a normal karyotype in his bone marrow. Trisomy 8 was demonstrated in his peripheral blood in 2005. Fluorescence in situ hybridization using an LSI CEP 8 probe performed on the archival bone marrow specimen showed three hybridization signals in 40% of 200 interphase cells scored. This confirmed that the trisomy 8 abnormality was present in both the blood and bone marrow samples. Trisomy 8 as the sole chromosomal abnormality in CLL is a very rare finding. The prognostic significance of trisomy 8 in CLL remains to be seen.

Bone Marrow↗