Clinical Next-Generation Sequencing to Guide Cancer Treatment Decisions

By Andrew J. Bredemeyer, PhD
Published Online: Wednesday, June 26, 2013
Siteman Cancer CenterAndrew J. Bredemeyer, PhD
Andrew J. Bredemeyer, PhD
 
Genomics and Pathology Services, Washington University School of Medicine in St. Louis
GPS at WUSTL
Stategic Partnership Logo
Treatment decisions in oncology are increasingly informed by the results of molecular genetic testing. In the past year, clinical laboratories have begun determining the tumor mutational status of multiple genes simultaneously using next-generation sequencing (NGS) platforms.

One such laboratory, Genomics and Pathology Services (GPS) at Washington University School of Medicine in St. Louis, in partnership with Siteman Cancer Center, has been performing clinical sequencing of more than two dozen oncogenes, tumor suppressors, and other cancer-related genes since March 2012. GPS is a Clinical Laboratory Improvement Amendments (CLIA)–certified clinical genomics laboratory that performs molecular genetic and cytogenetic testing using a variety of technologies.

“Next-generation sequencing is well-suited to oncology testing because of its sensitivity,” said John Pfeifer, MD, PhD, vice chair for Clinical Affairs for the Department of Pathology and Immunology at Washington University School of Medicine, and one of the founders of GPS. “Important somatic mutations present in only a fraction of the cells assayed—due, for instance, to tumor heterogeneity or to the presence of nonmalignant cells in the specimen—can be reliably detected well below the levels detectable by traditional sequencing methods.”

Critically, the technology is also successful at generating high-quality sequence data from formalin-fixed archival specimens, which are routinely available from standard pathology work-ups. GPS pathologists examine all specimens before sequencing and select regions of high tumor content for DNA extraction.

GPS’s current cancer NGS offering includes the full coding regions of 40 genes with demonstrated clinical relevance in a range of solid and hematologic cancers (Figure 1). “Interpreting the clinical relevance of a list of DNA variants can be daunting,” said Karen Seibert, PhD, director of GPS. “The results of our test’s comprehensive sequencing coverage are distilled into a concise clinical report by our clinical genomicists, highlighting clinically actionable mutations found in the patient’s tumor.”

Figure 1. Genomics and Pathology Services at Washington University offers a 40-gene next-generation sequencing test designed to provide predictive, prognostic, and diagnostic information for solid and hematologic cancers.
 40-gene next-generation sequencing test
Results of GPS’s cancer sequencing test frequently identify mutations, such as activating EGFR mutations in lung cancer, which predict response to targeted therapies. Many of the genes tested hold prognostic value, particularly in hematologic cancers and myelodysplastic syndromes. In some disease settings, test results can yield a more definitive diagnosis or stratify cancer subtypes.

The most common use of the test to date is in the setting of recurrent or metastatic disease, including lung, colorectal, pancreatic, and brain cancers, as well as sarcomas. However, Seibert noted that GPS regularly receives requests at the time of primary diagnosis, particularly for cancer types that lack an effective standard of care.

“Reimbursement success with third-party payers will drive the use of these sorts of tests in the short term,” said Pfeifer. “We have found that most insurers are receptive if there is ample medical evidence, in the cancer type being tested, of the predictive, prognostic, or diagnostic value of the focused set of actionable genes we assay.”

Most of the genes in GPS’s cancer sequencing test have precedents for reimbursement as single-gene tests, according to Pfeifer. “NGS analysis of many genes simultaneously is a cost-effective approach. Insurers may find that paying for NGS-based testing of gene sets offers substantial savings over sequential single-gene testing.”

Importantly, multigene tests on an NGS platform also make it easy to identify mutations known to be actionable in cancer types other than the one being tested. Such a finding can offer a treatment option when other options are exhausted. Testing at GPS has revealed multiple such cases, including an EGFR mutation in a sinonasal squamous cell carcinoma and a KIT mutation in a thymic carcinoma. In the latter case, treatment with a tyrosine kinase inhibitor yielded a radiographic response, and after 10 months the patient’s disease remains stable.

Figure 2. GPS’s cancer gene sequencing test yields actionable information in approximately 40% of cases.
 actionable results in approximately 40% of cases
In GPS’s experience, testing yields an actionable mutation in about 40% of cases (Figure 2). However, “Cases in which no mutation is found can also inform treatment decisions,” said Pfeifer. “Oncologists may rule out certain treatments with this information, potentially saving patients from toxicity and from time spent trying ineffective therapies.”

GPS performs tests at the request of ordering physicians nationally and internationally. Upon receiving an order, GPS staff contact the patient’s insurer for test authorization. GPS also coordinates acquisition of surgical materials for testing from the institution of origin. Ordering clinicians need only submit a requisition form available at GPS’s website, gps.wustl.edu/info.

Physicians who would like to order a test or who seek more information can contact GPS at 866-450- 7697 or at gps@wustl.edu.


Online CME Activities
Free CME from PER
Community Practice Connections: 13th Annual International Congress on the Future of Breast Cancer®
Highlights From the 13th Annual International Congress on the Future of Breast Cancer®
Pancreatic Cancer: Novel and Emerging Approaches to Early Detection and Treatment
Oncology Best Practice: Challenges in Treating Chronic Lymphocytic Leukemia – The Older Patient and Comorbid Conditions
More Reading
$auto_registration$