Cancer treatment monitoring using cell-free DNA fragmentomes

Abstract

Circulating cell-free DNA (cfDNA) assays for monitoring individuals with cancer typically rely on prior identification of tumor-specific mutations. Here, we develop a tumor-independent and mutation-independent approach (DELFI-tumor fraction, DELFI-TF) using low-coverage whole genome sequencing to determine the cfDNA tumor fraction and validate the method in two independent cohorts of patients with colorectal or lung cancer. DELFI-TF scores strongly correlate with circulating tumor DNA levels (ctDNA) (r = 0.90, p < 0.0001, Pearson correlation) even in cases where mutations are undetectable. DELFI-TF scores prior to therapy initiation are associated with clinical response and are independent predictors of overall survival (HR = 9.84, 95% CI = 1.72-56.10, p < 0.0001). Patients with lower DELFI-TF scores during treatment have longer overall survival (62.8 vs 29.1 months, HR = 3.12, 95% CI 1.62-6.00, p < 0.001) and the approach predicts clinical outcomes more accurately than imaging. These results demonstrate the potential of using cfDNA fragmentomes to estimate tumor burden in cfDNA for treatment response monitoring and clinical outcome prediction.

Fig. 1. DELFI-tumor fraction (DELFI-TF) as a mutation-independent approach for tumor monitoring.

a Tumors from patients with treatment-naive non-operable liver-only mCRC who enrolled in the CAIRO5 phase III trial were tested for hotspot mutations in KRAS, NRAS, and BRAF. Blood samples were collected at baseline, during treatment, and at the time of disease progression or last follow-up. Patients carrying KRAS, NRAS, or BRAF driver mutations were monitored with ddPCR and DELFI-TF assays. Patients with wild-type KRAS, NRAS, and BRAF tumors were monitored with DELFI-TF only. b Heatmap representation of genomic features depicts deviations of cfDNA fragment ratios and chromosomal arm-level z-scores across baseline and on-treatment time points of 149 patients having a liquid biopsy within 60 days of a RECIST1.1 evaluation, along with DELFI-TF values and clinical and demographic characteristics. c cfDNA genome-wide fragmentation profiles in 504 non-overlapping 5 Mb genomic regions at baseline and at time points within 60 days of imaging assessment by RECIST1.1 show marked heterogeneity at baseline and for patients who exhibited disease progression compared to patients who experienced stable disease or radiologic response after initiating first-line systemic therapy. Correlations for fragment ratios across 504 genomic regions for each sample compared to identical genomic regions in 153 non-cancers are shown in the color scale.

Fig. 2. DELFI-TF accurately predicted cfDNA tumor burden in patients with metastatic colorectal cancer.

a Model schematic. Plasma aliquots were collected from patients and used for cfDNA isolation. From each time point of patients with tissue-confirmed RAS/BRAF mutations, duplicate cfDNA samples were utilized for ddPCR and low-coverage WGS. WGS fragment-sequencing statistics were calculated per sample at a given time point. A random forest model was trained against the MAFs called by ddPCR readouts of the tumor-specific RAS/BRAF variants in all longitudinal cfDNA samples to generate the DELFI-TF values. b Patients with mCRC exhibit a wide range of DELFI-TF values at baseline (T0; n = 128) and reduced tumor fractions at the first time point after treatment commencement (T1; n = 151). Tumor fractions increase from treatment commencement (T1) to progression (n = 73). Non-cancer controls exhibit remarkably low DELFI-TF values (n = 153). No significant difference was observed between mutant type (MT) and wild-type (WT) samples for either time point (T0, p = 6.10e-01, T1, p = 5.14e-01, progression, p = 4.57e-01, two-sided Wilcoxon rank-sum). c DELFI-TF strongly correlates with detectable MAF values measured by ddPCR (n = 177, r = 0.90, p = 8.83e-65, two-sided Pearson correlation). Plasma time points with undetectable MAF (n = 132; n = 5 at T0, 40 at T1, 28 at T2, 59 at T3–T9) are indicated in red and included those with high DELFI-TF values in patients which were independently validated as having measurable tumor burden. The middle hinge in the boxplots corresponds to the median, while the lower and upper hinges correspond to the first and third quartiles. The upper whisker extends from the hinge to the largest value no further than the 1.5 * interquartile range from the hinge. Ribbons around the regression line in correlation plots represent the 95% confidence level interval for predictions.

Fig. 3. DELFI-TF as a non-invasive biomarker for disease burden, systemic treatment response, and prognostic outcome.

a Tumor fractions assessed by DELFI-TF and MAF at baseline were significantly lower in patients with a later confirmed partial response (PR) or complete response (CR) by the first two consecutive RECIST1.1 measurements at follow-up (tan, n = 8), compared to cases with consecutive progressive disease (PD) or stable disease (SD) (lavender, n = 38) (DELFI-TF, two-sided Wilcoxon rank-sum, p = 1.60e-02; MAF, two-sided Wilcoxon rank-sum, p = 2.2e-02). b Tumor fractions assessed by DELFI-TF and MAF at baseline were significantly different among patients who were eventually treated with complete resection (orange; n = 18), incomplete resection (yellow; n = 12) or no resection (blue; n = 35) after receiving first-line systemic treatment (DELFI-TF, p = 4.54e-02, Kruskal–Wallis; MAF, p = 1.09e-02, Kruskal–Wallis). c Colorectal cancer patients with metachronous metastases (gray; n = 16) exhibit lower tumor fractions assessed by DELFI-TF at baseline than patients who presented with synchronous metastases (green; n = 112) (p = 5.08e-04, two-sided Wilcoxon rank-sum). d, Kaplan–Meier curves for overall survival (OS) according to baseline DELFI-TF and MAF values. Patients with low DELFI-TF (green) or low MAF (yellow) experienced significantly longer OS than patients with high DELFI-TF (red) or high MAF (blue) (p = 6.23e-04, log-rank). Low DELFI-TF and MAF values were categorized as below the 25th percentile distribution of tumor fraction among the population of RAS/BRAF mutant individuals (n = 65). The middle hinge in the boxplots corresponds to the median, while the lower and upper hinges correspond to the first and third quartiles. The upper whisker extends from the hinge to the largest value no further than the 1.5 * interquartile range from the hinge.

Fig. 4. Dynamic changes in DELFI-TF were associated with longitudinal clinical outcomes in colorectal cancer patients.

a DELFI-TF slopes were calculated amongst all patients with a post-treatment blood draw (T1) and a blood draw within 60 days of progression (Tn Progression:60). A linear regression model was applied using the difference in days between T1 and all time points up to progression (n = 80). Left, DELFI-TF slopes are colored based on results below (blue) or above (orange) the median value across all DELFI-TF slopes from eligible patients. Right, swimmer plot encompassing RECIST1.1, cfDNA testing events, surgery, and death events for patients according to time on the study since registration. Bar segments are colored according to the RECIST1.1 readouts. b Kaplan–Meier curves for progression-free survival (PFS) according to DELFI-TF slopes below (blue) or above (orange) the median among patients with at least one blood draw within 60 days of disease progression (n = 80) (p = 7.32e-04, log-rank). c, Kaplan–Meier curves for OS according to DELFI-TF slopes below (blue) or above (orange) the median among patients with at least one blood draw within 60 days of disease progression (n = 80) (p = 3.33e-04, log-rank).