Abstract 1216P
Background
Employing a rapid liquid biopsy platform that can support clinicians in the diagnosis of different cancers, particularly for patients who develop cancers not targeted in current screening programs, would cause a paradigm shift in cancer diagnostics. Current liquid biopsies focus on single tumor derived biomarkers, such as circulating tumor DNA (ctDNA), which limits test sensitivity, especially for early-stage cancers that do not shed enough genetic material.
Methods
The Dxcover® Cancer Liquid Biopsy has been assessed upon its ability to predict individual cancers in organ-specific classifications: brain, breast, colorectal, kidney, lung, ovarian, pancreatic, and prostate cancer. The test uses Fourier transform infrared (FTIR) spectroscopy to build spectral profiles of serum samples, and machine learning algorithms to predict disease status. We also made a further exploratory evaluation of the ability to differentiate the signature from any one of the 8 cancers from non-cancer patient samples. We assessed the test performance when the cancer samples were grouped together to mimic patients with non-specific symptoms where the cancer site was uncertain. Additionally, we have examined non-generative data augmentation methods to improve machine learning performance.
Results
Area under the receiver operating characteristic curve (AUROC) values were calculated for 8 cancer types v symptomatic non-cancer controls: most classifiers achieved AUROC values above 0.85. The cancer v asymptomatic non-cancer classification detected 64% of stage I cancers when specificity was 99% (overall sensitivity 57%). When tuned for higher sensitivity, this model identified 99% of stage I cancers (with specificity 59%). For the colorectal cancer dataset, data augmentation using a WGAN led to an increase in AUROC from 0.91 to 0.96, demonstrating the impact data augmentation can have on deep learning performance, which could be useful when the amount of real data available for model training is limited.
Conclusions
This spectroscopic blood test can effectively detect early-stage cancer, and could facilitate the requisite earlier diagnosis when treatment can be more effective.
Clinical trial identification
Legal entity responsible for the study
Dxcover Limited.
Funding
Dxcover Limited.
Disclosure
M.J. Baker: Financial Interests, Personal, Full or part-time Employment: Dxcover Limited; Financial Interests, Personal, Leadership Role: Dxcover Limited; Financial Interests, Personal, Stocks or ownership: Dxcover Limited. J.M. Cameron, A. Sala, G. Antoniou, J.J.A. Conn, R.G. McHardy: Financial Interests, Personal, Full or part-time Employment: Dxcover Limited. D.S. Palmer: Financial Interests, Personal, Research Funding, Dxcover, GSK, Endophotonics: Dxcover Limited; Financial Interests, Personal, Stocks or ownership: Dxcover Limited; Financial Interests, Personal, Leadership Role: Dxcover Limited; Financial Interests, Personal, Full or part-time Employment: Dxcover Limited; Financial Interests, Personal, Advisory Role: Dxcover Limited.
Resources from the same session
1220P - Comprehensive diagnose of programmed death-ligand 1 from two-dimensional to three-dimensional in breast cancer with computer-aided artificial intelligence system
Presenter: Yi-Hsuan Lee
Session: Poster session 14
1221P - The functional domain of BRCA1/2 pathogenic variants (PVs) as potential biomarkers of second tumor and domain-related sensitivity to PARP-inhibitors
Presenter: Lorena Incorvaia
Session: Poster session 14
1222P - Detection of androgen-receptor splice variant 7 messenger RNA in circulating tumor cells of prostate cancer by in vitro assay
Presenter: Hoin Kang
Session: Poster session 14
1223P - Homologous recombination deficiency (HRD) testing on ovarian cancer ascites: A feasibility study
Presenter: Alberto Ranghiero
Session: Poster session 14
1224P - Detection of circulating tumor DNA (ctDNA) in untreated patients (pts) with cancer: Implications for early cancer detection (ECD)
Presenter: Yoshiaki Nakamura
Session: Poster session 14
1225P - Combining ctDNA and tissue-based-genomic profiling in advanced cancer: A real-world evidence prospective study in non-Western patients treated at Gustave Roussy cancer campus
Presenter: Tony Ibrahim
Session: Poster session 14
1226P - Multi-site validation of a deep learning solution for HER2 profiling of breast cancer from H&E-stained pathology slides
Presenter: Salim Arslan
Session: Poster session 14
1227P - Novel in vivo photonics-immunoassay system, inPROBE, for the rapid detection of HER2 in breast cancer
Presenter: Magdalena Staniszewska
Session: Poster session 14
1228P - A circulating tumor cell (CTC) based assay for diagnostic immunocytochemistry profiling of lung cancer
Presenter: Nitesh Rohatgi
Session: Poster session 14
1229P - Selective phenotypic and genotypic evaluation of circulating glial cells for improved diagnosis of glial malignancies
Presenter: Sewanti Limaye
Session: Poster session 14