Tumor-Only Gene Sequencing May Expose Half of Cancer Patients to Unsuitable Treatments

Victor E. Velculescu, MD, PhD

Victor E. Velculescu, MD, PhD

Almost half of patients with cancer who receive genetic testing may be given information that leads to unsuitable targeted treatments, according to researchers from Johns Hopkins Kimmel Cancer Center. The study report, scheduled to appear April 16, 2015, inScience Translational Medicine, may represent a significant shift in practices of next-generation sequencing (NGS) and may have consequences for precision medicine in cancer treatment.¹

In the study, researchers compared genomes of tumor and normal tissue from 815 patients who had a variety of cancers, including breast, brain, renal, gastric, lung, pancreatic, blood cancers, and melanoma. They matched the tumor sample and the normal sample from each individual to identify tumor-specific mutations.

“We found that tumor-only analyses provide inaccurate information. For tumor-only analyses, we determined that even after filtering of well-known germline alterations, approximately two-thirds of the remaining mutations were part of the patient’s normal, inherited genetic variation and were not tumor-specific changes or were false-positives. When focusing on those genes that are likely to be actionable, we found that one-third of the changes in tumor-only analyses were false positives. Because each patient on average had multiple such alterations, these false-positive changes affected roughly one in every two patients analyzed. In other words, one in two patients receiving tumor-only gene panel sequencing is at risk to receive a treatment that may be inappropriate,” said lead author Victor E. Velculescu, MD, PhD, professor of oncology and pathology and co-director of the cancer biology program at the Johns Hopkins University School of Medicine in an interview withTargeted Oncology.

According to Velculescu, inaccurate genetic information from tumor-only tests can have substantial consequences like serious side effects from inappropriate therapies, lack of useful targeted therapies, and increased costs of patient care from misguided medicines.

On a positive note, the researchers found that analyses of the germline DNA from patients identified alterations in cancer-predisposing genes in a small but significant proportion of patients. This occurred despite the fact that these patients had no known family history of cancer. “These observations suggested that such tests may provide one of the simplest ways to identify such individuals and their affected family members,” Velculescu said.

The findings could represent a turning point in how NGS is performed, but there are challenges to implementing this change clinically, according to Velculescu. “One is just the logistics of getting additional normal tissue or normal DNA from a patient,” he said, noting that this may be the least of the hurdles. The second challenge involves costs. “Unfortunately, at this time insurance does not fully cover the cost of such additional sequencing. This may change in the future given the value of matched tumor–normal sequencing as we have described,” he said. The third challenge is related to privacy, but Velculescu offered solutions to resolving this problem. “One [approach] is using the normal DNA as a filter against the tumor sequencing data, so you don’t actually analyze the normal DNA to find germline mutations in a sample, but you use it as a filter against the tumor,” he said. In this way, the test provider is blinded from the germline data. “The other approach is to have really large blanket consents that allow a patient to provide this type of material and get back these types of analyses. Obviously, we can all benefit from having this kind of information from our germline DNA,” said Velculescu during a press teleconference.

As the use of NGS increases, Velculescu hopes clinicians and laboratories will reconsider the design and implementation of genomic analyses to include testing outside the tumor. He believes this study will begin a conversation about the best way to perform these analyses in terms of the mutant alterations that are present in a tumor and what can be done to ensure that crucial genetic information is found during testing.

“We are able to see these types of tests being used earlier and earlier in cancer, as targeted therapies [are] applied, and of course there is a whole other topic, which is using these mutations as a possible tool for early diagnoses or actually classifying tumors. I think we are moving more and more toward using this information in earlier-stage disease, and we would all love to do that because that is actually a better place, where we can intervene and have better outcomes,” he said.

Velculescu stressed that NGS analyses are increasingly becoming essential for physicians to make important treatment decisions, so exploring tissue outside of the tumor microenvironment may be imperative to improving patient outcomes.

“You have to do the subtraction to obtain the right somatic or tumor-specific mutations and to subtract away all the other changes that we all carry around in our germlines, which is what makes us all individuals and makes us all different from one another,” he said.