The clinical impact of next-generation sequencing technology can be most keenly felt in the field of cancer. Cancer genomics continues to open up new avenues for the diagnosis and treatment of this heterogeneous disease, however there are still significant hurdles to overcome before the full potential of these new tools can be realised. In an Opinion article in Genome Medicine, Elaine Mardis from the Washington University School of Medicine, USA, discusses future applications of these tools to cancer care, as well as current barriers to clinical translation. Mardis is also a Guest Editor for the Genome Biology special issue on the genomics of cancer progression and heterogeneity. We asked Mardis what technologies she thinks hold the most promise for cancer treatment, and how advances in cancer genomics can be more readily translated to the clinic.
What do you think have been the most important developments in next-generation sequencing in the last five years that have contributed to a better understanding of cancer genomics?
Certainly improving kits, protocols and micro fluidic devices that help to embrace the challenging but incredibly informative aspects of clinical samples, such as limited numbers of cells, formalin fixation, and so on. Also the computational analysis breadth developed over that time frame has been remarkable and enabling.
What are the main hurdles to translating advances in next-generation sequencing technology to the clinic?
The challenges are numerous. Here are a few:
Firstly academic cancer centers have to commit to generating data to support the clinical benefit of these tests to patients, namely to address another challenge. Health payers have to understand why these tests and therapies help their clients so they will pay for the tests.
Coincident with these two challenges being addressed, we have got to produce data to support the argument for clinical benefit and we have to report it and share it with each other in an efficient way that largely obviates the conventional publication of results.
Lastly, pharmaceutical companies and the US Food and Drug Administration need to change the way they conduct clinical trials and approve drugs so more cancer patients benefit more rapidly, and hopefully at lower cost.
Whole-genome sequencing has been presented as a means to identify all of the genetic aberrations that contribute to cancer. Do you think this is the best way forward or are more targeted approaches needed?
The combination of whole genome sequencing and RNA sequencing is the best way forward at present. We have sacrificed a lot of discovery by doing large discovery projects by exome sequencing for the sake of doing it cheaply.
What will be needed to improve early detection of cancer using genomic and sequencing technologies?
Ideally the exciting work now happening in circulating tumor cell and nucleic acid sequencing will be applied to early detection and ultimately to prevention.
What will be needed to accelerate the application of cancer genomic diagnostics in primary clinical settings?
This will happen when these facilities have access to outsourced services that provide the sequencing, analysis and interpretation.
What new technologies across the field of cancer genomics do think hold the most promise to improve cancer treatment?
I think the most promising future therapies in cancer are happening now in the individualized immunotherapy realm. Basically, this involves a study of the most immunogenic mutant peptides in each patient’s tumor. We identify the mutations by genomic approaches, investigating only those mutations that are expressed in the RNA, and algorithmically predicting which ones are most likely to interact in a ‘non-self’ way with the patient’s HLA class I proteins. By using genomic data to identify candidate immunoantigens (also called neoepitopes), we can design personalized vaccines for patients. This is ongoing at our institution in a FDA-approved trial of melanoma patients, where several already have received their own conditioned dendritic cells as a vaccine. There are other groups also using this information to design vaccines of different types, at other institutions. I think this area holds great promise and uses our genomics capabilities in an entirely new and exciting way.
What role does open access and open data have to play in facilitating the progression of cancer genomics applied in clinical trials/case studies to standard clinical practice?
There’s a hugely important role played by rapid access to data around genes, mutations or other aberrations, and each patient’s response to therapy. The faster and more facile the sharing of these and other informative functional data, the faster we build the case for clinical benefit. This includes clinical trials data, acquired resistance mechanisms data and so-called ‘n of one’ results from individual patients.
When it comes to data sharing, what challenges do cancer researchers face in terms of the ethical implications of this?
Frankly, there are few if any ethical challenges. If we share somatic alterations only, they are unique to the tumor and not identifiable by definition. It is rare to find a cancer patient who wouldn’t want their data to contribute toward improving the disease outcome for other patients. These people are my heroes.