1. Technical advances in plasma genomic biomarkers
for mutation detection and monitoring in cancer patients
Sandra Fitzgerald1, Cherie Blenkiron1, Paula Shields1, Annette Lasham1 and Cristin Print1,2
1 Faculty Medical and Health Sciences, University of Auckland,2 Maurice Wilkins Centre, University of Auckland
Introduction
Methods
A sea-change is imminent for cancer medicine, due to the use of non-invasive genomic biomarkers in blood to inform screening, diagnosis and the selection of treatment. This technology may be used routinely in oncology within five years. Although numerous studies, including work in our laboratory, have shown that genomic analysis of blood can detect the presence and even the type of cancer, researchers have only scratched the surface of what this technology can do. There are still many technical challenges that need to be addressed before these biomarkers can be used routinely in the clinic. In our laboratory, we are generating new methods to improve the sensitivity and accuracy of non-invasive tests mutation detection in cancer patients.
Firstly, we are using a custom amplicon next generation sequencing panel – DNA QiaTarget, to screen for specific mutational hotspots or genes (Fig. 1 and 2). This strategy is useful when common mutations are absent, such as in Neuroendocrine Cancer, or to identify mutations in cancer types that have several genes mutated. Secondly, we are using Droplet Digital PCR (ddPCR) which allows the specific detection of a mutation of interest through a competitive probe assay. This method is particularly useful in cancers such as Melanoma, where discrete mutations such as BRAF V600E, are present in up to 40% of Melanomas. However, this requires the presence of each mutation to be screened for individually, and each assay requires optimisation (Figs 3 and 4).
Custom Amplicon NGS for identification of tumour mutations
Fig 1: Next Generation Custom Amplicon QiaTarget Sequencing successfully identified MEN1 mutation in the tumour of A001 Neuroendocrine patient (top panel). Sequencing of cell free DNA (cfDNA) extracted from plasma of this patient (bottom panel) also shows evidence of this mutation.
Fig 2: Neuroendocrine patient A0019 previously found to be aneuploidy for MEN1, is confirmed in the bottom panel by amplicon sequencing. Sequencing of cfDNA extracted from pre-surgical and post-surgical plasma from this patient (1st and 2rd panel) shows heterozygous SNP at this position
ddPCR assay optimisation for detection of BRAF V600E mutation – temperature gradient
A001 Tumour tissue showing ‘A’ mutation
B008 post-surgical plasma - some evidence of ‘A’ mutation
A0019 pre-surgical plasma
BRAF V600E mutation assay
51.9oC
50oC
48.7oC
48oC
54.3oC
NTC
56.3oC
Droplets detecting BRAF V600E mutation
BRAF wild type assay
56.3oC
54.3oC
51.9oC
50oC
48.7oC
48oC
NTC
Droplets detecting BRAF wild type
Fig 3: A temperature gradient has been used to determine maximum separation between positive and negative droplets for detection of BRAF V600E mutation
Sensitivity of ddPCR assays for detection of BRAF V600E melanoma mutation
(a) Fam probe detects BRAF V600E mutation in A375 DNA
(b) Hex probe detects Wt BRAF in THP1 DNA
20ng
10ng
5ng
1ng
0.5ng
0.1ng
0.01ng
NTC
Positive droplets representing decreasing BRAF V600E mutation
A375 only
10ng
15ng
19ng
20ng
NTC
Droplets detecting BRAF wild type
BRAF V600E background
(c) 2D Droplet ng A375 BRAF V600E DNA in 20ng THP1
A0019 post-surgical plasma
BRAF V600E mutation detected
Negative
BRAF wild type detected
A0019 tissue – showing aneuploidy
Fig 4: Decreasing amounts of BRAF V600E positive cell line DNA has been used to determine levels of sensitivity - 1:2000 sensitivity as seen in the individual plots for the mutant (a), wildtype (b). The 2 dimension plot (c) shows clear separation between the two probe sets.
We will continue the development of non-invasive technologies in our laboratory to further investigate the limits of detection for diagnosis of mutations in cancer, and the roles these technologies may play in the monitoring for relapse in cancer patients.
Future Directions
Acknowledgements: Sandra would like to thank Genesis Oncology Trust and William Staunton Memorial Scholarship Fund for supporting this research and Translational Medicine Trust