Performance Comparison of Droplet Digital PCR and Next-Generation Sequencing for Circulating Tumor DNA Detection in Non-Metastatic Rectal Cancer

Abstract

Background and objectives: Circulating tumor DNA (ctDNA) can potentially identify rectal cancer patients benefiting from neoadjuvant and adjuvant therapy. This study compared droplet digital PCR (ddPCR) and next-generation sequencing (NGS) for ctDNA detection in localized rectal cancer before and after surgery.

Methods: Pre-therapy plasma and rectal tumor samples were collected from a development group (n = 41) and a validation group (n = 26). Mutations in tumor samples were identified using NGS, and ctDNA detection was performed with both ddPCR and NGS. Recurrence was assessed 1 year after surgery in the development group.

Results: In the development group, ddPCR detected ctDNA in 24/41 (58.5%) and NGS panel in 15/41 (36.6%; p = 0.00075) of the baseline plasma. In the validation group, 21/26 (80.8%) patients had detectable ctDNA in the pre-therapy plasma. A positive ctDNA result was associated with higher clinical tumor stage and with lymph node positivity as detected by MRI. Postoperative ddPCR did not detect ctDNA before most recurrences.

Conclusions: We demonstrated a practical oligomarker ctDNA test for localized rectal cancer suitable for clinical workflow, and that ddPCR detects ctNA from pre-therapy plasma at a satisfactory level in advanced rectal cancers. Detecting ctDNA with ddPCR may help to assess the local severity, but the clinical utility of this approach should be evaluated in clinical trials.

Keywords: NGS; circulating tumor DNA; ctDNA; ddPCR; rectal cancer.

FIGURE 1

(A) A flowchart of the development cohort, patient enrollment, and study design. NGS, next‐generation sequencing; ddPCR, droplet digital PCR; short radiotherapy, short‐course radiotherapy 5 × 5 Gy; chemoradiotherapy; 50.4/1.8 Gy radiotherapy with capecitabine radiosensitizer. (B) Illustration of the discrepancy between cTNM‐status and (y)pTNM‐status with respect to baseline ctDNA status and neoadjuvant therapy. Sixteen patients had consistent staging, while 13 patients exhibited upstaging and 12 downstaging. cTNM‐status, TNM‐stage in the pre‐therapy rectal MRI. (y)pTNM‐stage, TNM‐stage in the final pathological examination. cTNM‐status, clinical TNM‐status; (y)pTNM‐status, TNM‐status in the final pathological examination.

FIGURE 2

The proportion of baseline ctDNA‐positive and ‐negative patients according to the initial clinical TNM‐stage (A, B) and cT‐status (C) when both methods were utilized. cT‐status, T‐stage in the initial rectal MRI; ctDNA, circulating tumor DNA. The p‐values are calculated with two‐tailed Fisher's exact test.

FIGURE 3

A Kaplan–Meier estimate of cancer‐related death (A) and recurrence‐free survival (B) based on baseline ctDNA status by using ddPCR alone. (C) Longitudinal ctDNA and clinical follow‐up with a swim track plot of patients with recurrence during the follow‐up. *The illustration also includes two patients diagnosed with metastatic disease before or at the time of surgery (#3 and #30). Information on the length of oncological therapies was unavailable for two patients (#3 and #11). ctDNA, circulating tumor DNA; ddPCR, droplet digital PCR. The p‐values are determined by the Log‐rank test.

FIGURE 4

The validation cohort study design and results. (A) An illustration of the SYNCOPE trial. Baseline ctDNA status stratified by initial clinical TNM‐stages (B), cT‐status (C), and cN‐status (D). CAPOX, capecitabine‐oxaliplatin combination chemotherapy; cN‐status, N‐stage in the initial rectal MRI; cTNM‐status, clinical TNM‐status; cT‐status, T‐stage in the initial rectal MRI; ddPCR, droplet digital PCR; SYNCOPE, SYstemic Neoadjuvant and adjuvant COntrol by PrEcision medicine in rectal cancer; XELOX, Oxaliplatin and capecitabine.