Background

Plasma ctDNA (circulating tumor DNA) has emerged as a novel biomarker for detecting genomic alterations and for longitudinal monitoring of colorectal cancer (CRC) patients. However, nearly 30% of patients show no mutations detected, potentially missing opportunities for companion therapy. Single circulating tumor cell (sCTC) genomics can provide greater sensitivity in detecting actionable targets. We report comprehensive genomic profiling (CGP) of live sCTCs and paired ctDNA from an advanced CRC patient population.

Methods

Retrospectively, live sCTCs and CTC clusters were isolated from six patients with stage IV CRC using OncoRADAR technology. Whole genomes of sCTCs were amplified and target-enriched using hybridization capture with OncoIndx, a comprehensive 1080-gene panel, to generate sequencing libraries. These libraries were sequenced on the Illumina NextSeq 2000 platform in paired-end mode with a sequencing depth of 500×. Raw sequence alignment and variant calling were performed using iCare software. Paired ctDNA samples were processed similarly but sequenced at a higher depth of 5500×.

Results

A total of 22 sCTCs were isolated, including four CTC clusters. The combined mutational landscape revealed 142 clinically relevant mutations, including 65 missense (45.77%), 25 nonsense (17.61%), 16 frameshift (11.27%), 7 indels (4.93%), 10 splice variants (7.04%), and 19 structural variants (13.38%). NRAS was the most frequently mutated gene, occurring in 52% of samples, followed by SMO (47.6%), TAP1 (42.85%), and TP53 (42.5%). In paired ctDNA samples, TP53 (66%), KRAS (50%), and TAP1 (33.33%) were the most frequently mutated genes. At the individual gene level, a 40% concordance was observed between sCTC and ctDNA. The genomic profile of sCTCs was particularly enriched with mutations in proliferative and stemness-maintenance signaling pathways, including NRAS:p.A146T and SMO:p.V392G, suggesting potential therapy evasion mechanisms. CTCs also showed a higher accumulation of immunotherapy resistance signatures, including loss-of-function mutations in STK11 and STAT5B, which were not detected in paired ctDNA samples.

Conclusions

The genomic profile of sCTCs exhibited enriched mutations in proliferative and stemness-maintenance signaling pathways. Therapy resistance signatures were more prevalent in sCTCs compared to ctDNA and may provide important clinical insights, particularly for patients who cannot provide tissue biopsy samples or show negative ctDNA results.