Cancer is a category of diseases, each of them broadly characterized by the type of cells from which the cancer originates. Key features that all types of cancer share are that cancer cells divide out of control of normal cellular mechanisms and do not differentiate into normal, functional cells. Commonly, cancer is treated with powerful chemotherapy drugs and radiation therapies that target dividing cells. This approach may destroy cancer cells, but it also can have significant health impacts on patients when healthy, noncancerous dividing cells are also destroyed. Additionally, such treatment regimens may leave behind residual cancer cells that are more aggressive than the cancer cells eliminated by the therapy.
Newer cancer therapies that target specific cancer pathways have fewer serious side effects. However, to deliver the correct therapy, clinicians must be able to determine which pathway is involved in a particular cancer case. Current next-generation sequencing (NGS) technologies are ideal for screening hundreds or thousands of variants, making these methods appealing for assessing the substantial genetic diversity of cancers.
Using xGen® Lockdown® Probes, Xie et al. describe the development of a complete genomic cancer panel for NGS that enriches for 115 genes that are either diagnostic for cancer or known to impact the effectiveness of specific cancer therapies. The scientists also developed a bioinformatics pipeline for identification of indels, base substitutions, and copy number variation—mutations that can affect gene expression and contribute to cancer development.
Xie et al. tested the performance of their assay using 14 patient samples that included formalin-fixed, paraffin-embedded (FFPE) tumor tissue, blood, and pleural fluid. Blood samples were sequenced at 50–100X and tumor samples at 200–500X mean coverage.
The researchers also describe some of the challenges of working with cancer samples, including low DNA content from small biopsy tissues or from mixed tissue specimens containing both healthy and tumor cells. In addition, many tumor biopsies are preserved as FFPE samples, which frequently contain fragmented nucleic acids that have been degraded by such factors as crosslinking and low pH. The authors present their optimization and library preparation methods for these challenging FFPE samples that allowed them to generate usable libraries from as little as 25 ng of genomic DNA.
The authors' optimized cancer panel and bioinformatics analysis protocols provided good sequencing uniformity and significantly reduced PCR duplicates. They reported 77–88% on-target performance for the 115 genes represented in the target capture panel.
Xie et al. note that the continued discovery of new cancer genes and genes with therapeutic implications will make it necessary to continually update their panel. For this, they emphasize the flexibility of xGen Lockdown Probes and the ease with which they can add new, custom DNA probes as these discoveries are made.