Levels of circulating tumor DNA correlate with tumor volume in gastro-intestinal stromal tumors: an exploratory long-term follow-up study

Abstract

Patients with gastro-intestinal stromal tumors (GISTs) undergoing tyrosine kinase inhibitor therapy are monitored with regular computed tomography (CT) scans, exposing patients to cumulative radiation. This exploratory study aimed to evaluate circulating tumor DNA (ctDNA) testing to monitor treatment response and compare changes in ctDNA levels with RECIST 1.1 and total tumor volume measurements. Between 2014 and 2021, six patients with KIT proto-oncogene, receptor tyrosine kinase (KIT) exon-11-mutated GIST from whom long-term plasma samples were collected prospectively were included in the study. ctDNA levels of relevant plasma samples were determined using the KIT exon 11 digital droplet PCR drop-off assay. Tumor volume measurements were performed using a semi-automated approach. In total, 94 of 130 clinically relevant ctDNA samples were analyzed. Upon successful treatment response, ctDNA became undetectable in all patients. At progressive disease, ctDNA was detectable in five out of six patients. Higher levels of ctDNA correlated with larger tumor volumes. Undetectable ctDNA at the time of progressive disease on imaging was consistent with lower tumor volumes compared to those with detectable ctDNA. In summary, ctDNA levels seem to correlate with total tumor volume at the time of progressive disease. Our exploratory study shows promise for including ctDNA testing in treatment follow-up.

Keywords: circulating tumor DNA; follow‐up; gastro‐intestinal stromal tumor; treatment; tumor volume.

Fig. 1

Selection of prospectively collected plasma samples. Time from diagnosis to last follow‐up (= light blue bar). Plasma samples that were not analyzed by KIT exon 11 digital droplet polymerase‐chain‐reaction (ddPCR) drop‐off assay (= red crosses). Clinically relevant plasma samples that were analyzed by ddPCR without treatment (= black triangles). Plasma samples that were analyzed by ddPCR under imatinib treatment (= green triangles), sunitinib treatment (= yellow triangles) or regorafenib treatment (= purple triangles), respectively. † = dead due to disease.

Fig. 2

ctDNA versus tumor volume and RECIST. Figure shows the circulating tumor DNA (ctDNA) in copies·mL⁻¹ (red line), analyzed by KIT exon 11 digital droplet polymerase chain reaction (ddPCR) drop‐off assay, in association with total 3D tumor volume in mL (blue line). All patients had metastasized disease at first time of detection of ctDNA. All blood samples in which ctDNA was measured were taken within 3 weeks of imaging. The x‐axis shows time in weeks from first taken blood sample. The left y‐axis shows the scale of copies·mL⁻¹ and the right y‐axis of total 3D tumor volume (mL). The colors in the background correspond to the different treatment lines. Response evaluations according to Response Evaluation Criteria In Solid Tumors (RECIST 1.1): B, baseline; CR, complete response; PD, progressive disease; PR, partial response; SD, stable disease.

Fig. 3

Tumor volume versus RECIST and ctDNA. Figure shows the relation of circulating tumor DNA (ctDNA) in copies·mL⁻¹ with total 3D tumor volume in mL of all samples of all patients that were taken at time of PR (green), SD (brown) or PD (red). SD = stable disease (n = 4), PR = partial response (n = 14), PD = progressive disease (n = 15), according to Response Evaluation Criteria In Solid Tumors (RECIST 1.1). Only blood samples taken within 3 weeks of imaging were included. Samples at time of complete response (CR) according to RECIST were excluded (n = 6).