Mark E. Robson, MD
Although the potential for next-generation sequencing of breast cancer tumors to improve treatment strategies is widely recognized, questions swirl about the optimal use of such increasingly available technologies in clinical practice for today’s patients.
During the 31st Annual Miami Breast Cancer Conference
, two leading researchers debated key aspects of the potential benefits and limitations of employing knowledge gained through current genomic profiling technologies to manage patient care.
Mark E. Robson, MD, attending physician in Clinical Genetics & Breast Cancer Medical Services at Memorial Sloan Kettering Cancer Center (MSK) in New York City, argued that next-generation sequencing is a powerful tool that thus far has proved invaluable in oncology research but has not been established as a useful addition to daily clinical practice.
Kimberly L. Blackwell, MD, a professor of Medicine and director of the Breast Cancer Program at Duke Cancer Institute in Durham, North Carolina, maintained that clinicians could use the technology to steer patients into clinical trials and to identify actionable mutations. “I see the promise of next-gen sequencing every day in my own clinic,” she said.
The Case Against Routine Use
Robson, who specializes in treating young women with hereditary breast cancer and in researching targeted therapies for patients with BRCA 1/2
mutations, began his presentation with an endorsement of the potential for next-generation sequencing to fuel therapeutic advancements.
“I think that we’re in the midst of an incredibly exciting time when it comes to genetics and the application of genetic technologies to cancer treatment,” said Robson. “We’re basically testing this enormous hypothesis that we can move beyond the use of our conventional histopathology in staging and bring genomics in, and by defining the pattern of genomic aberrations within a tumor, improve treatment outcomes.
“But I would emphasize that right now this is very clearly a hypothesis,” he added. “This is not in any way proven that on a routine basis this is necessarily going to be useful.”
At MSK, mutation profiling is conducted using its IMPACT process in which sequences from an individual tumor are analyzed in the context of 341 preselected genes. Somatic mutations including base substitutions, small indels, copy number alterations, and possibly select rearrangements are identified. Robson said there are three potential clinical applications of information gained through tumor sequencing: (1) a variant is identified that is clearly linked to response from an approved drug; (2) a variant is identified that would make the patient eligible for a clinical trial; and (3) a variant is identified that is potentially predictive of response to an approved drug off-study.
In the first instance, Robson said, clinicians do not need “agnostic profiling” to determine whether a patient is a candidate for treatment with an FDA-approved drug since the target already is known and a companion diagnostic test typically is available.
When it comes to clinical trials, Robson said that clinicians can find a list of studies targeting a given mutation using sequencing results but that there may be barriers to participation for the patient, particularly having to travel hundreds of miles to enroll. The patient’s financial resources, performance status, and stage of disease also might rule out trials. Robson noted that it may be appropriate to conduct upfront tumor profiling in large academic centers with broad portfolios of targeted therapy trials, but that clinicians in daily practice and their patients might find trials that are better matches geographically, financially, and physically by using databases and other sources.
As far as treating a patient with a targeted agent identified through sequencing outside the boundaries of a clinical trial, Robson discouraged this practice. “Most potentially risky is the idea that you have a patient, either early stage or late stage, you have a genomic profile conducted, and then you use that information to ad hoc pick a targeted agent that you think is going to be useful for the individual and do the treatment,” said Robson. “How do you know whether or not the particular alteration that you found is going to predict response to the particular agent you would like to use?”