Abstract 65P
Background
Changes of circulating tumor DNA (ctDNA) detection have been revealed to predict the changes in tumor volume and therapeutic effects in colorectal cancer. However, some patients suffered from “negative false” issue in ctDNA detection in many studies, which may result in a delay in effective clinical intervention. We explored if there were preference of genes or other factors in ctDNA detection in a large Chinese colorectal cancer cohort.
Methods
A total of 577 colorectal cancers were included. All the tumor tissues and plasma underwent 189-gene next generation sequencing.
Results
Tumor tissue and recent plasma from 245 pts were evaluated for the consistence in mutations. A total of 887 mutations were detected in tumors and 204 mutations of them, from 70 pts (defined as group1) were also detected in plasma of the same patient. 683 mutations from 175 patients (defined as group2) were not detected in plasma samples. In 204 plasma-detected mutations (defined as group1 muts), the mutational frequency of APC, TP53 and KRAS genes were 25%, 22% and 6% respectively, while in 683 non-plasma-detected mutations (defined as group2 muts), the frequency was 3%, 0.3% and 0.1% respectively. We compared the cfDNA levels in plasma between group1 and group2 and observed a significant lower trend in samples from group2. To further explore the mutations status in plasma, another colorectal cancer cohort consisting 332 patients with longitudinal ctDNA detections (median maximum monitoring time is 123 days) were analyzed. A total of 127 group1 muts were detected in 259 patients. Among them, 106 mutations were detected in subsequent plasma samples in the same patient while 21 mutations not. Interestingly, the mutational frequency of APC, TP53 and KRAS genes were 25%, 26%, 9% in 106 mutations vs 36%, 20%, 0.04% in 21 mutations, respectively. Notably, “neo-RAS” were observed in two patients out of 102 KRAS-mt patients. The KRAS-G13D in their tumor were not detected in subsequent plasma.
Conclusions
In recurrent mutated colorectal cancer driver genes, mutations in APC were more likely not to be detected in ctDNA compared with TP53 and KRAS. The result indicates a more targeted panel design when choosing mutations monitored via ctDNA in colorectal cancer.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
This work is supported by Guangdong Natural Science Foundation (No. 2018A0303130184， 2019A1515011678), Science and Technology Programme of Guangzhou Municipal Government (No. 201707010304).
Disclosure
All authors have declared no conflicts of interest.