Dataset Information

Early Cancer Detection from Multianalyte Blood Test Results.

ABSTRACT: The early detection of cancers has the potential to save many lives. A recent attempt has been demonstrated successful. However, we note several critical limitations. Given the central importance and broad impact of early cancer detection, we aspire to address those limitations. We explore different supervised learning approaches for multiple cancer type detection and observe significant improvements; for instance, one of our approaches (i.e., CancerA1DE) can double the existing sensitivity from 38% to 77% for the earliest cancer detection (i.e., Stage I) at the 99% specificity level. For Stage II, it can even reach up to about 90% across multiple cancer types. In addition, CancerA1DE can also double the existing sensitivity from 30% to 70% for detecting breast cancers at the 99% specificity level. Data and model analysis are conducted to reveal the underlying reasons. A website is built at http://cancer.cs.cityu.edu.hk/.

SUBMITTER: Wong KC

PROVIDER: S-EPMC6548890 | biostudies-literature | 2019 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Early Cancer Detection from Multianalyte Blood Test Results.

Wong Ka-Chun KC Chen Junyi J Zhang Jiao J Lin Jiecong J Yan Shankai S Zhang Shxiong S Li Xiangtao X Liang Cheng C Peng Chengbin C Lin Qiuzhen Q Kwong Sam S Yu Jun J

iScience 20190504

The early detection of cancers has the potential to save many lives. A recent attempt has been demonstrated successful. However, we note several critical limitations. Given the central importance and broad impact of early cancer detection, we aspire to address those limitations. We explore different supervised learning approaches for multiple cancer type detection and observe significant improvements; for instance, one of our approaches (i.e., CancerA1DE) can double the existing sensitivity from ...[more]

PMID: 31103852

Similar Datasets

Project description:BackgroundEndoPredict (EP) is a clinically validated multianalyte gene expression test to predict distant metastasis in ER-positive, HER2-negative breast cancer treated with endocrine therapy alone. The test is based on the combined analysis of 12 genes in formalin-fixed, paraffin-embedded (FFPE) tissue by reverse transcription-quantitative real-time PCR (RT-qPCR). Recently, it was shown that EP is feasible for reliable decentralized assessment of gene expression. The aim of this study was the analytical validation of the performance characteristics of the assay and its verification in a molecular-pathological routine laboratory.MethodsGene expression values to calculate the EP score were assayed by one-step RT-qPCR using RNA from FFPE tumor tissue. Limit of blank, limit of detection, linear range, and PCR efficiency were assessed for each of the 12 PCR assays using serial samples dilutions. Different breast cancer samples were used to evaluate RNA input range, precision and inter-laboratory variability.ResultsPCR assays were linear up to Cq values between 35.1 and 37.2. Amplification efficiencies ranged from 75% to 101%. The RNA input range without considerable change of the EP score was between 0.16 and 18.5 ng/μl. Analysis of precision (variation of day, day time, instrument, operator, reagent lots) resulted in a total noise (standard deviation) of 0.16 EP score units on a scale from 0 to 15. The major part of the total noise (SD 0.14) was caused by the replicate-to-replicate noise of the PCR assays (repeatability) and was not associated with different operating conditions (reproducibility). Performance characteristics established in the manufacturer's laboratory were verified in a routine molecular pathology laboratory. Comparison of 10 tumor samples analyzed in two different laboratories showed a Pearson coefficient of 0.995 and a mean deviation of 0.15 score units.ConclusionsThe EP test showed reproducible performance characteristics with good precision and negligible laboratory-to-laboratory variation. This study provides further evidence that the EP test is suitable for decentralized testing in specialized molecular pathological laboratories instead of a reference laboratory. This is a unique feature and a technical advance in comparison with existing RNA-based prognostic multigene expression tests.

Project description:BackgroundMulti-cancer early detection (MCED) testing could increase detection of cancer at early stages, when survival outcomes are better and treatment costs are lower, but is expected to increase screening costs. This study modeled an MCED test for 19 solid cancers in a US population and estimated the potential value-based price (the maximum price to meet a given willingness to pay) of the MCED test plus current single cancer screening (usual care) compared to usual care alone from a third-party payer perspective over a lifetime horizon.MethodsA hybrid cohort-level state-transition and decision-tree model was developed to estimate the clinical and economic outcomes of annual MCED testing between age 50 and 79 years. The impact on time and stage of diagnosis was computed using an interception modeling approach, with the consequences of cancer modeled based on stage at diagnosis. The model parameters were mainly sourced from the literature, including a published case-control study to inform MCED test performance. All costs were inflated to 2021 US dollars.ResultsMulti-cancer early detection testing shifted cancer diagnoses to earlier stages, with a 53% reduction in stage IV cancer diagnoses, resulting in longer overall survival compared with usual care. Addition of MCED decreased per cancer treatment costs by $5421 and resulted in a gain of 0.13 and 0.38 quality-adjusted life-years across all individuals in the screening program and those diagnosed with cancer, respectively. At a willingness-to-pay threshold of $100,000 per quality-adjusted life-year gained, the potential value-based price of an MCED test was estimated at $1196. The projected survival of individuals diagnosed with cancer and the number of cancers detected at an earlier stage by MCED had the greatest impact on outcomes.ConclusionsAn MCED test with high specificity would potentially improve long-term health outcomes and reduce cancer treatment costs, resulting in a value-based price of $1196 at a $100,000/quality-adjusted life-year willingness-to-pay threshold.

Dataset Information

Early Cancer Detection from Multianalyte Blood Test Results.

Publications

Early Cancer Detection from Multianalyte Blood Test Results.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets