199P - Circulating tumor DNA as a dynamic marker for disease burden in patients with metastatic colorectal cancer

Date 17 December 2016
Event ESMO Asia 2016 Congress
Session Poster lunch
Topics Colon and Rectal Cancer
Rectal Cancer
Translational Research
Presenter An Hsu
Citation Annals of Oncology (2016) 27 (suppl_9): ix53-ix67. 10.1093/annonc/mdw581
Authors A. Hsu1, H. Hung-Chih2, H. Chen1, S. Chen1
  • 1Clinical Sequencing, ACT Genomics Co. Ltd., 114 - Taipei/TW
  • 2Division Of Hematology-oncology, Chang Gung Memorial Hospital-Linkou, 333 - Taoyuan/TW



Metastatic colorectal cancer (mCRC) is one of the most deadly cancer condition worldwide. Currently, treatment monitoring using CT and/or plasma CEA does not offer sufficient sensitivity and cannot be used for certain treatments. This study investigate whether circulating tumor DNA (ctDNA) can be used to monitor treatment efficacy.


Plasma samples were obtained from 16 patients with mCRC before and after treatment. Cell-free DNA were extracted for next-generation sequencing analysis. Matched tumor samples were obtained from 12 patients for concordance analysis. The sequencing panel covers more than 1500 mutation hotspots on 13 cancer-related genes. All ctDNA samples were sequenced with at least 8000X average coverage. The sequenced raw data was aligned to reference human genome to identify variants. Detection sensitivity was set at 0.5%.


ctDNA were detected in 12 of 16 (75%) pre-treatment plasma samples, with 8 samples harboring more than one ctDNA. Among the ctDNA detected, TP53 (56%) is the most common mutations, followed by KRAS (38%), NRAS (13%), PIK3CA (13%), BRAF (6%) and CTNNB1 (6%). Nine of 12 (75%) samples showed complete concordance between ctDNA and mutation detected in tissue samples. In the remaining three samples, additional mutations were detected in plasma samples but not in the matching tissue samples. All six post-treatment samples showed clear decrease in ctDNA allele frequency compared to pre-treatment sample. For samples with more than one ctDNA present in the plasma, all detected ctDNAs showed similar extent of decrease after treatment. Plasma ctDNA levels for two patients were followed for at least four months post-treatment. In both cases, changes in ctDNA levels showed a more sensitive and significant alteration compared to image data and CEA levels.


Our results indicated that plasma ctDNA can be detected in majority of patients with mCRC. Mutation spectrum detected in plasma is highly concordant with data obtained from tissue samples. The level of ctDNA significantly decrease upon treatment and sensitively reflecting tumor burden. Overall, our results illustrate the clinical feasibility of using ctDNA as a dynamic marker for disease burden in patients with mCRC.

Clinical trial indentification

Legal entity responsible for the study

Shu-Jen Chen


ACT Genomics


All authors have declared no conflicts of interest.