EXCEPTIONAL RESPONDERS TO CANCER TREATMENT

Most clinicians will have at least one anecdotal incident of a patient who showed an exceptional response to treatment. A new initiative at the National Cancer Institute (NCI) is aimed at finding out why some patients respond so dramatically.

The first-ever Exceptional Responders Initiative has been launched to examine tumors from more than 100 cancer patients who had good responses to treatments that benefited, overall, less than 10% of treated patients. Coleader Barbara A. Conley, MD, associate director of the NCI Cancer Diagnosis Program, and colleagues want to determine whether there are molecular biomarkers in the tumor that can predict response to the drug tested or to drugs with similar mechanisms of action.

"The NCI has been interested in how current molecular analysis capability can be used to improve how we treat patients," Dr Conley told Medscape Medical News.

"NCI Director Harold E. Varmus, MD, met with experts from across the country and heard from Dr Solit and others who had found molecular reasons for a particular response," she explained.

In 2012, a group headed by David B. Solit, MD, used tumor genome sequencing to explain why the tumor of a bladder cancer patient had an exceptionally good response to everolimus treatment.

In addition, researchers trawling the NCI Cancer Therapy Evaluation Program (CTEP) phase 2 clinical trial database turned up about 100 cases of exceptional responses to otherwise unimpressive regimens.

"Since NCI has been supporting early-phase drug studies for cancer for decades, it seemed logical to review those response databases to see how many patients had complete responses or long-lasting responses to drugs that did not go on to approval in that indication," Dr. Conley explained. "Those of us who are clinicians and who have participated in therapeutics development studies all had anecdotal instances of exceptional responses, both in patients who had received investigational treatments and in those who had been treated with standard regimens."

Clinical researchers from oncology specialty societies have expressed support for this approach. Charles D. Blanke, MD, chair of the SWOG cooperative group, told Medscape Medical New that "SWOG finds the concept of looking for factors underlying exceptional efficacy and toxicity fascinating and believes this trial to be very promising." Dr. Blanke is professor of medicine at Oregon Health Science University's Knight Cancer Institute in Portland.

Collection of Tissue and Clinical Data

Dr. Conley said that tissue and clinical data from exceptional responders will be obtained from NCI-supported trials, and potentially from other clinical trials. The remaining samples and data will come from standard therapy settings, such as community practices, where there are reliable outcome data, and from pharmaceutical industry trials or other sources. She said that letters are being sent to cancer centers and clinicians in the United States to ask them to assist in this effort.

The researchers are looking for patients who had either a complete response or a durable partial response (more than 6 months) to a treatment in which less than 10% of patients achieve such responses.

Patients must have given consent for future research on their tumor tissue. If the patient has died, the tumor can be used for genomic analysis unless previous consent denied this use. Available tumor tissue must have been collected prior to the administration of the treatment drug.

Clinicians can propose a potential exceptional responder by emailing a description of the case (without patient identifiers) to the initiative, Dr Conley explained.

Information should be provided on the diagnosis; the treatment that resulted in the exceptional response; the response type and duration; the patient's vital status; a signed informed consent form in which the patient agrees to future research on the specimen; the availability of tissue; the nature of tissue sample, if known (e.g., FFPE block, FFPE core biopsy, fresh frozen, age of tissue); and contact information for the submitting party (email address, phone number, institution address, and an NCT identifier number if the patient was part of a clinical trial).

DNA and RNA from tissue samples will be isolated at the Biospecimen Core Resource at Nationwide Children's Hospital in Columbus, Ohio. The isolates will then be shipped to a DNA sequencing and analysis center at Baylor University in Houston.

An expert panel has been recruited to assess whether proposed cases are exceptional responders. According to Dr. Conley, a second panel, which will assess the correlation of molecular and response data, will consist of experts from the NCI or The Cancer Genome Atlas (TCGA), professionals who performed the tissue preparation and assay, statisticians, and the proposers of the case, if they want to participate.

"Thirteen cases have been proposed, some of which are in review and some of which require additional information. We have had interest from some academic sites that will be proposing cases as well. This is a promising beginning," Dr. Conley reported.

Results of the initiative are likely to have implications for clinical trial design and clinical practice.

"Currently, trials have embraced designs where patients are assigned to treatment based on the molecular features of their tumor that may predict response, if there is such a marker. This study may allow hypotheses about response markers in a broader array of systemic therapies, not just those targeted at a specific abnormality or pathway, but those where we do not know exactly what molecular features predict response," Dr. Conley said.

"The feasibility of this approach is supported by reports in the literature of relevant mutations in tumor specimens from patients who experienced an exceptional response to a drug in a clinical trial, even though that drug failed to meet the trial's end point for clinical benefit," Louis Staudt, PhD, MD, director of the NCI Center for Cancer Genomics and coleader of the initiative, said in a statement.

The ability to identify molecular markers that predict clinical response is likely to advance the current trend toward more individualized care based on the recognition that many superficially similar tumor types comprise subsets that respond differently to different anticancer treatments, Dr. Conley noted.

It is expected that the Exceptional Responders Initiative will be completed in 3 to 4 years. According to statement from NCI, the researchers hope that other investigators will be able to use the data generated to test hypotheses on specimens from clinical trials or to compare their own datasets with the full genomic annotation of 100 cases of exceptional responders, which will be placed in the public domain.