Genomic profiling: The foundation for personalized cancer care
As targeted cancer therapy options are increasing, it is becoming more important to choose the most appropriate treatment for patients with cancer. At a Roche-sponsored symposium held during the 2nd ESMO Asia Congress in Singapore, four experts discussed the potential of comprehensive genomic profiling (CGP) for personalizing cancer care and helping healthcare professionals to make the best treatment decisions for individual patients.
Options for the molecular management of cancers such as advanced non-small cell lung cancer (NSCLC) are increasing, and with this, there has been growing society support for molecular profiling. The 2016 NCCN NSCLC guidelines state, “the NCCN Panel strongly endorses broader molecular profiling (also known as precision medicine) to identify rare driver mutations to ensure that patients receive the most appropriate treatment.”
Comprehensive genomic profiling (CGP) is a cutting-edge sequencing technology that has several advantages over other standard methods (Figure 1). In the words of Dr James Suh, Laboratory Director at Foundation Medicine in Morrisville, North Carolina, US, “CGP is not just next generation sequencing [NGS], it is a more thoughtful approach.” There are a variety of gene alterations that may be present in cancers, however, standard methods may not be able to detect them all. “Most NGS tests can only identify base substitutions [point mutations], and some indels [insertions/deletions]. We have sequenced over 100,000 clinical cases, and almost half of the alterations we have found are copy number changes or rearrangements. You’re not going to find those unless you’re using CGP,” said Suh.
In lung cancer, there are more than 10 known gene alterations that can be targeted with approved therapies. [Lancet Oncol 2016;387:1354-1356] In many cases, these are the particular alterations that molecular testing will be aimed at finding. However, Professor Ignatius Ou Sai-Hong, Clinical Professor of Hematology-Oncology at the University of California Irvine School of Medicine in Irvine, California, US, pointed out the limitation of this approach, “if you just use one single mutation test, you may miss out on finding the others that have the potential to benefit the patient.” In the clinic, CGP has the potential to identify driver mutations for which targeted therapy recommended by the NCCN guidelines is available, as well as those for which there are FDA-approved or pending approved treatments, or treatments available on clinical trials. Dr Wong Seng Weng, Medical Director & Consultant Medical Oncologist at the Paragon Medical & Mt Elizabeth Novena Specialist Centre in Singapore said, “as oncologists, we want to pick the therapies that are destined to work, and drop those that are destined to fail.”
CGP can provide benefit for oncologists and patients
Around 5 percent of lung cancers harbour ALK rearrangements, and these are usually detected using the cytogenetic technique fluorescence in-situ hybridization (FISH). However, complex rearrangements may not always be detected by this method. CGP has been demonstrated to accurately identify these alterations, leading to optimal choice of targeted therapy, in this case crizotinib. [J Thorac Oncol 2012;7:e14-16; J Thorac Oncol 2013;8:e85-86] CGP has also been able to detect drug sensitive EGFR exon 19 deletions and EGFR fusions not identified by prior molecular testing, leading to effective treatment with EGFR inhibitors. [Clin Cancer Res 2016;22:3281-3285; Cancer Discovery 2016;6:601-611] Alterations in MET exon 4 are found across NSCLC histologies, and these may respond to treatment with crizotinib. While conventional testing with FISH can identify MET amplifications, it cannot detect splice site alterations, which can be detected using CGP. [J Thorac Oncol 2016;11:1493-1502]
For many patients with cancer, in order to perform the required molecular tests, multiple biopsies are required, and this can be painful and distressing. In one study of 31 patients who had entirely negative molecular test results prior to CGP, 71 percent had undergone multiple biopsies. With one biopsy, CGP was able to identify alterations treatable with targeted agents included in the NCCN guidelines in 29 percent of patients, and with other targeted agents available on or off clinical trial in a further 40 percent of patients. [Clin Cancer Res 2015;21:3631-3639]
Some cancers respond well to immunotherapy using checkpoint inhibitors, but predicting which patients will benefit is a challenge. Currently, measurement of PD-L1 overexpression by immunohistochemistry is the gold standard, however, this is complicated by variations in testing methods and antibodies used. Another potential use of CGP is to measure a patient’s tumour mutation burden (TMB) (Figure 2), and this may help to predict response to immunotherapy. Suh commented that, “ultimately, TMB is a better biomarker [than PD-L1] in immuno-oncology.” Patients with high TMB are found across a range of tumour types, particularly melanoma and lung cancers. In NSCLC, TMB has been shown to predict benefit of treatment with a PD-1 inhibitor. [Science 2015;348:124-128]
CGP can also help to predict resistance to targeted therapies. Using tumour DNA extracted from the patient’s blood plasma, pre-existing resistance mutations can be identified, which can predict treatment failure early and assist the oncologist with treatment decisions. [J Thorac Oncol 2017;12:145-151] “Using CGP and patient cases, we can create a database for rational use of targeted agents in the future,” said Ou.
Challenges of CGP in the clinical setting
Despite the advantages of CGP, there are still challenges faced by oncologists in the implementation and interpretation of this technology. Dr Amit Verma, Consultant of Molecular Oncology & Cancer Genetics at Max Hospital in New Delhi, India, said, “there are no guidelines, so how do we decide when to use CGP, who to test, and which site to test?” For example, should patients undergo genomic profiling early in the disease course, or after metastasis or treatment failure? Should the CGP sample be taken from the primary or recurrent site? These are the types of questions oncologists are asking. Interpretation of the results may also lead to clinical dilemmas; “When there are multiple actionable mutations, which one do we choose?” said Verma. “Although CGP may provide more therapeutic options, the level of evidence may be low,” he added. Verma proposed an “algorithm for actionability” to help oncologists to interpret the CGP report and use the information to optimize treatment decisions (Figure 3). Ethical issues may also arise, for example when a genomic alteration is detected in a familial cancer-causing gene, or a patient does not have access to the recommended clinical trial.
“Although we may find an actionable mutation, we can’t always get the drug,” said Wong. In cases where the appropriate therapy is not approved for the indication, there is a risk that the patient may deteriorate before access to the drug is granted. In some cases, the therapy may not be available at all, particularly in emerging economies.
CGP has the potential to serve as an important tool to assist healthcare professionals in their decision-making processes, and improve personalized cancer care by allowing treatment to be tailored to the specific patient. CGP has several advantages over other methods, being more sensitive, specific, and tissue-sparing. Although there are still challenges in the clinical setting, CGP offers oncologists and patients much needed options for improving cancer treatment outcomes. “There is a lot of work to be done, but I am optimistic that eventually CGP will give us oncologists what we want,” Wong said.