Tumor Only Sequencing Could Be Misleading

Sequencing the tumors of cancer patients is an increasingly common practice among healthcare providers. The testing is done in an attempt to personalize each patient’s therapy as much as possible – but it may actually be negatively impacting treatment.

The reason, according to a study by scientists at the Johns Hopkins Kimmel Cancer Center (Baltimore, MD, USA), is that many hospitals and companies that sequence patients’ tumor genomes fail to sequence normal tissue as well, meaning that they can’t easily distinguish cancer-related mutations from those that have no deleterious effects (1). Without the ability to fully understand what’s going on inside a tumor and what genetic alterations are causing it, attempting to use tumor sequencing information to direct treatment may be inappropriate or even harmful. The study analyzed sequencing data from 815 patients with a wide variety of cancers, filtering out well-known germline mutations. Looking only at tumor sequences, researchers discovered 382 mutations. After comparing those sequences to the patients’ normal germline genomes though, they found that an average of 249 of the mutations – over 65 percent – were part of the patients’ normal background genetic variation.

But not all mutations can be therapeutically targeted. When the researchers limited their search to “actionable genes,” for which targeted treatments have already been developed, they found that many of those changes were also part of patients’ germline genomes, resulting in “false positive” sequencing results for nearly half of patients. This, warns principal investigator Victor Velculescu, could lead to inappropriate therapy for patients in whom these changes are normal. In a press release from Johns Hopkins Medicine (2), he explains that personalized therapies, which are aimed at the unique genetic changes that drive individual tumors, depend on accurate assessment of those tumors’ genomes. As his study shows, though, not all genetic changes in tumor tissue are directly related to the cancer. In order to determine which types of treatment will work for which patients, it’s important to consider every factor – and that includes comparing their tumor sequences to the genetics of their normal tissue.

Of course, implementing this knowledge in a clinical setting isn’t as easy as just acknowledging its necessity. Sequencing normal tissue alongside tumor samples doubles the cost of testing – currently several thousand dollars to sequence tumor tissue alone – and the amount of work required to collect, prepare and analyze samples. But the benefits of sequencing normal tissue extend beyond a single patient; understanding more about which mutations are directly related to cancer can increase our overall understanding of the disease and allow us to discover new genes that increase the risk or severity of disease. Ultimately, normal sequencing could help us to better identify cancer patients at greater risk of refractive disease, or screen the general population for people at risk of developing cancer in the future.