Abstract 623P
Background
Cancer ranks second globally in causes of death, accounting for 21% of all fatalities. However, many types of cancer can be cured if diagnosed and treated during early stages. We propose a liquid biopsy cancer analysis method that uses deep learning and a methylation-sensitive restriction enzyme digestion followed by sequencing method to detect and classify the most common cancers worldwide at early stages.
Methods
We developed a selective methylation sensitive restriction enzyme sequencing (MRE-Seq) method combined with a prediction model based on deep neural network (DNN) learning on data from 63,266 CpG sites to identify global hypomethylation patterns. The methylation dataset was made from 96 colon cancer samples, 95 lung cancer samples, 122 gastric cancer samples, 136 breast cancer samples, and 183 control samples. To eliminate batch bias, the ANOVA test was performed during feature selection. A DNN was adopted as a classifier, and 5-fold cross validation was performed to verify the classification performance.
Results
Across four cancer types, colorectal cancer had the highest predictive performance at 0.98, followed by breast cancer at 0.97, gastric cancer at 0.96, and lung cancer at 0.93. At 95% specificity, the sensitivity for detecting early-stage cancers varied widely, with lung cancer at 50% and breast cancer at 83%. Two different metrics were used to evaluate the model's performance. The cancer classifier (performance in detecting cancer) had a sensitivity of 95.1% and a specificity of 66.7%, indicating better performance in correctly identifying cancer samples. The cancer type classifier (performance in classifying the cancer type) utilized the precision metric to evaluate the accuracy of cancer classification. Notably, breast cancer achieved the highest precision at 95.8%, followed by lung cancer at 83.3%, gastric cancer at 79.1%, and colon cancer at 69.0%.
Conclusions
The proposed classification model based on the MRE-Seq method can reliably identify cancer and normal samples and differentiate between different cancer types using only methylation information obtained from patient's blood. This approach could be used in clinical practices to help medical experts diagnose cancer earlier and at the individual level.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.
Resources from the same session
122P - Distinct transcriptomic immune profiling and clinicopathological features of cribriform morphology in colorectal adenocarcinomas
Presenter: Abdelhakim Khellaf
Session: Poster Display
Resources:
Abstract
123P - Spatial molecular profiling identifies FGF20 upregulation on cancer-associated fibroblast and FGFR2-PI3K/Akt activation in tumor cells of sporadic early-onset colon cancer
Presenter: Dave Hoon
Session: Poster Display
Resources:
Abstract
124P - Characteristics, prognosis and therapeutic effects of non-V600 BRAF mutated colorectal cancer
Presenter: Lalida Arsa
Session: Poster Display
Resources:
Abstract
125P - Final results of APOLLON-11 and SOYUZ-APOLLON study: Multicentre prospective observational post-authorization study of bevacizumab biosimilar in patients with metastatic colorectal cancer in real-world practice
Presenter: Alexey Tryakin
Session: Poster Display
Resources:
Abstract
126P - From tumor height (TH) to tumor regression grade (TRG) in locally advanced rectal cancers (LARC) during total neadjuvant therapy (TNT): A retrospective analysis
Presenter: Valeria Pusceddu
Session: Poster Display
Resources:
Abstract
127P - A meta-analysis of efficacy and safety from head-to-head first-line (1L) trials of epidermal growth factor receptor inhibitors (EGFRIs) versus bevacizumab in combination with chemotherapy (CT) doublets in patients with RAS wild-type (WT) metastatic colorectal cancer (mCRC) by sidedness
Presenter: Takayuki Yoshino
Session: Poster Display
Resources:
Abstract
128TiP - A phase II study of cadonilimab + FOLFOXIRI and bevacizumab as initial therapy for unresectable proficient mismatch repair/microsatellite stable (pMMR/MSS) metastatic colorectal cancer (mCRC)
Presenter: Rongbo Lin
Session: Poster Display
Resources:
Abstract
140P - Prevalence of claudin-18 isoform 2 (CLDN18.2) positivity in locally advanced (LA) unresectable or metastatic gastric or gastroesophageal junction (mg/GEJ) adenocarcinoma in patients (pts) in the Asia region: Phase III SPOTLIGHT and GLOW studies
Presenter: Hoo Hwoei Fen Soo
Session: Poster Display
Resources:
Abstract
141P - Early phase trials outcomes in refractory upper GI cancers: A 10-year analysis from the SCRI UK phase I unit
Presenter: Antonella Cammarota
Session: Poster Display
Resources:
Abstract
142P - The survival impact of the addition of durvalumab to cisplatin/gemcitabine in advanced biliary tract cancer: A real-world, retrospective, multicentric study
Presenter: Silvia Foti
Session: Poster Display
Resources:
Abstract