Monitor tumor burden with circulating tumor DNA

Abstract

There is a need to identify better biomarkers to monitor diseases and/or assess therapeutic responses. For those with cancer, one can identify DNA fragments that contain somatic mutations originating in the tumor DNA in plasma or serum. There have been several early studies suggesting that advances in sequencing technologies will allow identification of somatic genomic alterations that can be used to monitor tumor dynamics. Dawson et al. investigated circulating cell-free DNA carrying tumor specific alterations in patients with breast cancer. The authors compared CT imaging from 30 women with metastatic breast cancer receiving treatment, using two assays for circulating tumor DNA, CA 15-3, and CTCs. Taken the two methods together circulating tumor DNA was detected in 29 or 30 women (97%) and 115 of 141 plasma samples (82%). Circulating tumor DNA levels showed a greater dynamic range and greater correlation with changes in tumor burden than did CA 15-3 or CTC. The relatively small study showed that circulating tumor DNA has a superior sensitivity to other circulating biomarkers and a dynamic range that correlates with tumor burden.

There is a need to identify better biomarkers to monitor diseases and/or assess therapeutic responses. The development and validation of these mechanism-based biomarkers could serve as novel surrogate endpoints in drug development and clinical decision-making.¹ The components of biomarker development include discovery, qualification, verification, research assay optimization, clinical validation, and commercialization.² Unfortunately, very few solid tumors have a validated biomarker. Prostate specific antigen (PSA) is fraught with problems both for diagnosis as well as therapeutic monitoring.³ Likewise, CEA for colon cancer and CA125 for ovarian cancer have limitations.⁴ In breast cancer, cancer antigen 15-3 (CA 15-3) is a serum biomarker that is clinically useful in a subset of patients, but has limited sensitivity.⁴ Circulating tumor cells (CTCs) has also been put forward as promising biomarker to assess tumor burden.⁵ However, the sensitivity concerns for the assay for detecting circulating tumor cells lies between 1 and 5 cells per 7.5 mL of blood.

Circulating fragmented DNA is well documented in the serum/plasma of individuals.⁶ For those with cancer, one can also identify DNA fragments that contain somatic mutations originating in the tumor DNA. There have been several early studies suggesting that advances in sequencing technologies have allowed identification of somatic genomic alterations that can be used to monitor tumor dynamics.⁷ Dawson et al. investigated circulating cell-free DNA carrying tumor specific alterations in patients with breast cancer.⁸

The authors compared CT imaging from 30 women with metastatic breast cancer receiving treatment, using two assays for circulating tumor DNA, CA 15-3, and CTCs. DNA was extracted from archival-tumor tissue samples to identify somatic genomic mutations by two approaches, targeted deep sequencing to screen for point mutations in PIK3CA and TP53 and whole-genome paired-end sequencing to identify mutations and structural variants in tumor-tissue specimens vs. matched normal-tissue specimens. They identified genomic alterations in tissue from 30 of 52 women. Twenty-two patients had mutations only, 5 had structural variants, and 3 patients had both mutations and structural variants.⁸

In the 30 women with somatic mutations or structural variants, 141 serial plasma samples were quantified for circulating tumor DNA by either digital PCR or tagged-amplicon deep sequencing. With a sensitivity of detecting one mutation molecule in a background of 1000 wild-type molecules, digital PCR was used to detect circulating tumor DNA, which was identified in 18 of 19 women and in 82% of the plasma samples analyzed. As a high-throughput alternative to digital PCR, 44 samples (from 11 patients) were analyzed using tagged-amplicon deep sequencing. The sensitivity of tagged-amplicon deep sequencing allowed for the detection of a mutant allele fraction of >0.14%. Circulating tumor DNA was detected in all 11 patients and in 80% of the plasma samples analyzed. In a subset of the plasma samples both methods were used (digital PCR and taggedamplicon deep sequencing). Taken together circulating tumor DNA was detected in 29 or 30 women (97%) and 115 of 141 plasma samples (82%). The median quantity of circulating tumor DNA was 150 amplifiable copies/mL of plasma.⁸

CA 15-3 was detected in 78% of patients and CTC in 87%. Circulating tumor DNA levels showed a greater dynamic range and greater correlation with changes in tumor burden than did CA 15-3 or CTC. Circulating tumor DNA was detected and showed a serial change in 95% of patients in treatment responses. Similar findings were seen for both circulating tumor cells and CA 15-3; however, if the number of CTC was less than 5 per 7.5 mL or the CA 15-3 was less than 50 U per mL, that correlation was disrupted.⁸

Dawson et al. made a significant step forward in the development of circulating tumor DNA as a potential biomarker.⁸ The relatively small study showed that circulating tumor DNA has a superior sensitivity to other circulating biomarkers and a dynamic range that correlates with tumor burden. However, one limitation of the study was that somatic mutations or structural variants were detected in only about 60% of patients. While monitoring circulating tumor DNA requires the identification of somatic mutations in individual patients, there is no single locus that was commonly mutated in this breast cancer study population. Thus, substantial sequencing will be required for probe design or the identification of possible common gene signatures. Hopefully, in the near future we will have a better handle on the somatic mutations for solid tumor to allow us to validate this biomarker further.

Disclosure of Potential Conflicts of Interest

No potential conflict of interest was disclosed.