ABSTRACT
BACKGROUND: It was recently demonstrated that the size of cell-free DNA (cfDNA) fragments that originates from tumor cells are shorter than cfDNA fragments that originates from non-malignant cells. We investigated whether cfDNA fragment size and cfDNA levels might have prognostic value in patients with advanced pancreatic cancer. METHODS: Blood samples were obtained from patients with advanced pancreatic cancer, before (n = 61) initiation of chemotherapy and after the first cycle of chemotherapy (n = 39). Samples were separated with density centrifugation and plasma DNA was isolated. Mode cfDNA fragment size and cfDNA levels were then determined using a 2100 Bioanalyzer. A cohort of partially age-matched healthy volunteers (n = 28) constituted the control group. RESULTS: Both a pre-treatment cfDNA fragment size of ≤ 167 bp (mode) and high pre-treatment cfDNA levels were associated with shorter progression-free survival (PFS) (p = 0.002 and p < 0.001, respectively) and overall survival (OS) (p = 0.001 and p = 0.001, respectively). Furthermore, multivariable Cox regression analyses demonstrated that pre-treatment cfDNA levels could independently predict prognosis for both PFS (HR = 3.049, p = 0.005) and OS (HR = 2.236, p = 0.028). CONCLUSION: This study demonstrates that cfDNA fragment size and cfDNA levels can be used to predict disease outcome in patients with advanced pancreatic cancer. The described approach, using a rapid, economic and simple test to reveal prognostic information, has potential for future treatment stratification and monitoring.
Subject(s)
Cell-Free Nucleic Acids/blood , Cell-Free Nucleic Acids/chemistry , Pancreatic Neoplasms/blood , Pancreatic Neoplasms/pathology , Adult , Aged , Aged, 80 and over , Case-Control Studies , Female , Humans , Kaplan-Meier Estimate , Male , Middle Aged , Multivariate Analysis , Neoplasm Staging , Prognosis , Proportional Hazards Models , Survival AnalysisABSTRACT
Most current methods of circulating tumour cell (CTC) enrichment target the epithelial protein EpCAM, which is commonly expressed in adenocarcinoma cells. However, such methods will not recover the fraction of CTCs that have a non-epithelial phenotype due to epithelial-mesenchymal transition. For phenotype-independent CTC enrichment, we developed a new enhanced negative depletion strategy-termed MINDEC-that is based on multi-marker (CD45, CD16, CD19, CD163, and CD235a/GYPA) depletion of blood cells rather than targeted enrichment of CTCs. Here we validated the performance of MINDEC using epithelial and mesenchymal cancer cell lines, demonstrating a mean recovery of 82 ± 10%, high depletion (437 ± 350 residual white blood cells (WBCs)/mL peripheral blood), linearity between spiked and recovered cells (correlation coefficient: r = 0.995), and a low detection limit (≥1 cell recovered in all four replicates spiked with 3 cells). For clinical validation of this method, we enumerated CTCs in peripheral blood samples from patients with metastatic pancreatic cancer, detecting CTCs in 15 of 21 blood samples (71%) from 9 patients. The promising performance of the MINDEC enrichment strategy in our study encourages validation in larger clinical trials.
Subject(s)
Antigens, Neoplasm/blood , Cell Adhesion Molecules/blood , Cytological Techniques , Epithelial Cell Adhesion Molecule/metabolism , Neoplastic Cells, Circulating/pathology , Antigens, CD/blood , Antigens, CD19/blood , Antigens, Differentiation, Myelomonocytic/blood , Biomarkers, Tumor/blood , Breast Neoplasms/blood , Cell Line, Tumor , Cell Separation/methods , Epithelial-Mesenchymal Transition , Female , GPI-Linked Proteins/blood , Glycophorins/blood , Humans , Leukocyte Common Antigens/blood , Limit of Detection , Lung Neoplasms/blood , MCF-7 Cells , Mesothelioma/blood , Pancreatic Neoplasms/blood , Phenotype , Receptors, Cell Surface/blood , Receptors, IgG/blood , Reproducibility of ResultsABSTRACT
We used KRAS mutations to investigate the clinical relevance of circulating tumor DNA (ctDNA) measurements in patients with advanced pancreatic cancer. Fifty-three blood samples were collected from 14 prospectively recruited patients prior to chemotherapy (gemcitabine or FOLFIRINOX) and subsequently every month during treatment. Samples were processed by density centrifugation and plasma DNA isolation. A Peptide-nucleic acid-clamp PCR was then used to detect KRAS mutations (present in >90% of pancreatic cancers) as a surrogate marker for ctDNA. Plasma samples from 29 healthy individuals were analyzed as a reference group. Results were compared to conventional monitoring measures and survival data. Median follow-up time was 3.7 months (range 0.6-12.9 months). Ten (71%) patients had a positive KRAS status in the plasma samples obtained prior to chemotherapy, indicating the presence of ctDNA. Among the patients who were ctDNA-positive before chemotherapy, nine (90%) experienced disease progression during follow-up, compared to one (25%) of four ctDNA-negative patients (P = 0.01). The pre-therapy ctDNA level was a statistically significant predictor of both progression-free and overall survival (P = 0.014 and 0.010, respectively). Of the 14 patients, ten had ≥2 follow-up samples; in several of these patients, the ctDNA level changed substantially during the course of chemotherapy. Changes in ctDNA levels corresponded both with radiological follow-up data and CA19-9 levels for several patients. This pilot study supports the hypothesis that ctDNA may be used as a marker for monitoring treatment efficacy and disease progression in pancreatic cancer patients. Recruitment of more patients is ongoing to corroborate these findings.
