Background: Early cancer diagnosis might improve survival rates. As circulating tumor DNA (ctDNA) carries cancer-specific modifications, it has great potential as a noninvasive biomarker for detection of incipient tumors.
Patients and methods: We collected cell-free DNA (cfDNA) samples of 1002 elderly without a prior malignancy, carried out whole-genome massive parallel sequencing and scrutinized the mapped sequences for the presence of (sub)chromosomal copy number alterations (CNAs) predictive for a malignancy. When imbalances were detected, 6-monthly clinical follow-up was carried out.
Results: In 3% of participants chromosomal imbalances were detected. Follow-up analyses, including whole-body MRI screening, confirmed the presence of five hematologic malignancies: one Hodgkin lymphoma (HL), stage II; three non-HL (type chronic lymphocytic leukemia, Rai I-Binet A; type SLL, stage III; type mucosa-associated lymphoid tissue, stage I) and one myelodysplastic syndrome with excess blasts, stage II. The CNAs detected in cfDNA were tumor-specific. Furthermore, one case was identified with monoclonal B-cell lymphocytosis, a potential precursor of B-cell malignancy. In 24 additional individuals, CNAs were identified but no cancer diagnosis was made. For 9 of them, the aberrant cfDNA profile originated from peripheral blood cells. For 15 others the origin of aberrations in cfDNA remains undetermined.
Conclusion(s): Genomewide profiling of cfDNA in apparently healthy individuals enables the detection of incipient hematologic malignancies as well as clonal mosaicism with unknown clinical significance. CNA screening of cellular DNA of peripheral blood in elderly has established that clonal mosaicism for these chromosomal anomalies predicts a 5- to 10-fold enhanced risk of a subsequent cancer. We demonstrate that cfDNA screening detects CNAs, which are not only derived from peripheral blood, but even more from other tissues. Since the clinical relevance of clonal mosaics in other tissues remains unknown, long-term follow-up is warranted. Taken together, this study demonstrates that genomewide cfDNA analysis has potential as an unbiased screening approach for hematological malignancies and premalignant conditions.