Prognostic and Predictive Impact of Circulating Tumor DNA in Gastric Cancer Treatment

Introduction The cell lysis phenomena tumor releases DNA that can be detected in the blood and in other biological fluids such as lymph, urine or stool. An increase in plasma levels of tumor DNA is described in patients with cancer, and recently, monitoring of circulating tumor DNA that has been proposed as a relevant potential marker to assess the prognosis and the early response to treatment of colon or breast cancers (Dawson et al, 2013; Spindler et al, 2013; Tie et al, 2015). However, there is no data on the prognostic and predictive impact of circulating tumor DNA in context of gastric cancer treatment.

The aim of this study is to evaluate the correlation between the level of circulating tumor DNA and prognosis or response to treatment of localized and advanced gastric cancer.

Patients and Methods:

Three university hospitals of "Assistance Publique Hôpitaux de Paris" (AP-HP) will participate in this prospective study: European Georges Pompidou hospital, Pitié-Salpêtrière hospital and Cochin hospital. Inclusion criteria were all patients aged over 18 years with localized or advanced adenocarcinoma of gastric or gastro-oesophageal junction, histologically proven. The patients will be enrolled over a period of 2 years after receiving and signed a specific consent information form.

This is a non-interventional study who does not change the management of patients. There will be no additional invasive procedures to those already scheduled for routine care. Blood samples will be made at the time of chemotherapy sessions from the Huber needle previously implanted in the port-a-cath for the administration of chemotherapy agents.

The circulating tumor DNA is analyzed and quantified by sequencing proton from somatic genetic alterations identified in the tumor (Inserm Unit 775 UMR_S, Professor Pierre Laurent-Puig).

The data related to the patient (age at diagnosis, sex, weight, height, WHO performance status), tumor (tumor markers CEA and CA 19-9, date of diagnosis of gastric cancer, histological type and tumor differentiation, tumor stage, and metastatic sites) and treatment (resection of the primary tumor, date of surgery, chemotherapy protocol) will be collected anonymously. Monitoring data concern the efficacy of chemotherapy (tumor response, the date of disease progression, survival), as well as the possible dates of tumor recurrence or death.

The Statistical analysis will be based on a survival model in order to predict the responder or non-responder status, including parameters normally associated with risk of recurrence and death. The association between changing in circulating tumor DNA levels in responders and non-responders will be performed with a Cox model; the DNA circulating levels will be considered as a variable dependent of time. An estimation of 100 patients is planned for each cohort (localized and advanced diseases) with a recruitment period of approximately 2 years.

Expected Results

• For the cohort of patients with a localized tumor: loss (or decrease) in circulating tumor DNA after curative treatment for patients who do not exhibit tumor recurrence; OR no loss (or increase) in circulating tumor DNA after curative treatment in patients with tumor recurrence.
• For the cohort of patients with advanced tumor: early and significant increase in the level of circulating tumor DNA in non-responders to chemotherapy; OR early and significant reduction in the level of circulating tumor DNA in patients who respond to the treatment.