Population Genetics Technologies Ltd., creator of innovative methods for genetic analyses and biomarker discovery, has successfully completed a pilot study to prove that its GenomePooling™ method is sensitive and accurate enough to detect even small percentages of gene variants within cancer tissue. This validates the method’s suitability for use in IntReALL 2010, the world’s largest genetic analysis of acute lymphoblastic leukaemia (ALL) relapse cases. ALL is the most common childhood malignant disease.
Cancer tissues are problematic because they can contain multiple tissue types and multiple cancer variants that may respond differently to treatment. Being able to identify associations between genotype and treatment responsiveness matters because, if a tumour contains highly drug resistant variants, or the patient has a genetic predisposition to relapse, alternative treatments may be appropriate. In order to demonstrate their ability to detect variants in mixed tumour tissue, Population Genetics mixed two different DNA samples and showed using their novel molecular counting technology that they could identify variants with high sensitivity.
The pilot compared data derived by Population Genetics from ALL patients and cell-line derived samples with data generated in a parallel study. The parallel study was undertaken by Dr Julie Irving, Reader in Experimental Haematology at Newcastle University, who used the conventional mutational screening method, called DHPLC (denaturing high-performance liquid chromatography).
Explained Julie Irving: “Our comparative study showed that Population Genetics’ methodology is accurate, more sensitive and thus superior to the standard DHPLC method. This gives us confidence that it is the right workflow to use for interrogating data from the large population IntReALL study which is just getting underway”.
Population Genetics sequenced 28 discrete genomic regions across a total of 50 human genomic DNA samples, 40 derived from patients suffering ALL and 10 derived from cell lines containing a mix of mutated and wild type phenotypes of two known genes. The Population Genetics method identified all the known mutations and an additional five that were not detected by DHPLC; these novel variants were then confirmed by Sanger sequencing and allele specific polymerase chain reaction.
ALL affects 4 per 100,000 children per year in Europe but intensive combination chemotherapy with stem cell transplantation has improved survival from under 20% in the 1970s to over 80% today. However treatment is complex and prolonged and it is unclear why some children do not respond to, or relapse after, existing treatments.
IntReALL 2010 comprises 23 research teams, led by Vaskar Saha, Professor of Paediatric Oncology at The University of Manchester, gathering DNA across Europe, Japan, Israel and Australia from children who have relapsed after treatment for ALL. Funded by IntReALL’s EU FP7 grant, Population Genetics will study the resulting biobank of leukaemia samples using proprietary GenomePooling workflows to identify and validate associations between genetic risk factors and treatment efficacy. It is hoped that the IntReALL study will aid in the optimisation of therapy based on genetic stratification and allow randomised controlled trials of potential new drugs.
Alan Schafer, CEO of Population Genetics, commented: “Accurately and sensitively detecting genetic variants in large numbers of samples is critical for developing genetic biomarkers for cancer. For this collaboration we created a version of GenomePooling suited to the study needs, and have established cancer as another application in which our population-scale genetic analysis methods have demonstrated their utility. Our approach enables the generation of statistically rich information from thousands of genomes with resolution down to the level of the individual. We look forward to working with the IntReALL consortium to potentially improve treatment outcomes for children suffering ALL”.
MONET: New AI tool enhances medical imaging with deep learning and text analysis