National Cancer Institute Director Promises New Directions to Accelerate Route to Cure

Theodore Bosworth


January 04, 2019

Of the four strategies being pursued by the National Cancer Institute (NCI) to advance their goal of cancer eradication, the effort to create large pools of readily shared data appears to be the one dearest to the heart of NCI director Norman E. Sharpless, MD. In an outline of his priorities presented after his first year in office,[1] Sharpless spoke about this specific issue with particular passion. He believes many of the clues to better outcomes are hidden in data that have already been collected and are just waiting to be found.

To drive this point home, Sharpless told a story. Two years ago, he picked up a copy of the New England Journal of Medicine to learn that p53 mutations are an important predictor of response to decitabine in older patients with acute myeloid leukemia (AML). Although there were only 116 patients in this study,[2] he described the results as "black and white" in regard to the clarity of their meaning in clinical practice.

"Now, one could say at this point that this is great news. 'Hooray, this is medical progress. We now have a better understanding of how to treat AML,' but I did not see this that way," Sharpless recounted. "I saw this as a clear failure of data aggregation. If we had been aggregating data in a deliberative way with clinical and generative data from the get-go in AML, a result like this would have fallen out immediately."

Before assuming his current post, Sharpless led the University of North Carolina's Lineberger Comprehensive Cancer Center as a hematologist/oncologist. This experience influenced his views on data aggregation, which has become one of his priorities in leading the NCI. He describes this undertaking as one of "four pillars" of his agenda, developed after speaking broadly with many stakeholders during his first few months in office. At ASH, he outlined each of them, but spoke with particular fervor about the power of aggregated data.

"I would argue that no matter what your interest in cancer research—be it pathogenesis, basic science, prevention, or survivorship, or cancer health disparities—any of these topics—is empowered by big data," he said. In solid tumors and hematologic malignancies, where personalized medicine is now expected to dominate clinical care, big data promise to unlock complex interrelationships among genomics, molecular pathways of cancer growth, and signaling systems that increase resistance to cancer drugs. Obscure or invisible without data aggregation, these connections have huge potential for clinical advances.

"The p53 mutation is common, but what about a mutation that might drive therapy in 5% of patients or just 0.5% of patients? I don't see how we are going to discover those therapeutic subsets without aggregating data. This is why we have to explain this to the greater community," he said.

Making this case is important to engage allies. Sharpless has told the p53 story to gather support on many occasions, including to members of Congress. To achieve his vision of a cloud-based repository, containing huge quantities of data researchers could readily share and query, he would not only need to overcome obstacles in logistics and funding, but also of cooperation between entities.

"The overarching goal is to create large, linked, multimodal data sets that have histology, radiology, genetics, and clinical outcomes," Sharpless explained. This means enlisting many, often competing institutions to provide data that can be organized in useful ways for research purposes without violating data security.

"While data privacy and security are important—in fact, critical—there are also costs to not aggregating data and not sharing data for research," Sharpless said. Referring back to the p53 story, he believes that many of the next advances reside in data already captured or accumulating and waiting to be found. He argues that the very real costs of slowing the discovery process inherent in failing to aggregate and share large quantities of data "are simply too great."

Of the remaining three priorities that Sharpless described in his efforts as director designed to "sharpen the focus" of the NCI, one is at least tangentially related. Data sharing is likely to become increasingly important within the context of new strategies to test emerging drugs.

"Every one of today's successful therapies for malignant disease came about because of a successful clinical trial, but we need to bring these trials up to date," Sharpless said. Owing to the rapid growth in personalized treatment choices based on the molecular profile of the patient's cancer, the classic randomizations have become ill-suited for many types of cancer, where the specific characteristics of each malignancy may be important in determining response and optimal treatment.

"The NCI [is addressing this by] increasing funding for certain kinds of trials, developing novel clinical trial designs, and incentivizing data sharing for clinical trials," said Sharpless, referring to adjustments he feels are needed to improve clinical value. For this same reason, "we are beginning more and more to open trials in the community, away from large cancer centers but closer to home where patients live, in order to establish that these therapies work in the real world for diverse patient populations."

The final two priorities are related. One is to reenergize support for basic science, and the other is to provide the incentives and funding for workforce development in cancer research. The NCI has never abandoned basic science research, because it is an important part of its portfolio, but Sharpless said that new political will to invest in cancer research has been the basis for NCI increasing funding to both clinical and basic science research.

"At the request of Congress, we have prioritized support for early-stage investigators," Sharpless said. Although not exclusively relevant to basic science, funding for research has been increased, with substantial amounts being directed toward an increase in Research Project Grants (RPGs; also known as R01s), which are often key for early research concepts to move forward.

"Our goal was to increase the number of R01s in 2018 by at least 25%, and we far surpassed that goal. We also started to fund a new award for early-stage investigators called the R-37, which is a 7-year award as opposed to the usual 5-year R01, the idea being to determine whether longer periods of funding are particularly beneficial [to supporting the cancer research workforce]," Sharpless reported.

For basic science, specifically, the funds for RPGs were increased $100 million in 2018. The RPG is a pool of money reserved to fund a broad array of investigator-initiated research activities outside of the walls of the NCI. Sharpless described RPG applications as where "some of the strangest and craziest but also some of the best ideas" in cancer research emerge.

Appropriations from the federal government to the NCI have been increasingly steadily for several years. Although this can be traced at least partially back to the Moonshot Initiative championed by the last presidential administration, Sharpless suggested that the political will to fund cancer research is also based on the startling number of recent advances that have captured the public imagination.

"We are making progress at a rate that is faster and greater than at any point in my career as a medical oncologist," Sharpless said. He hopes his strategies for the NCI will continue spur the sense of mission in the clinical and research communities, which he acknowledged is already at a high level.

Sharpless concluded his speech on a positive note, summing up the accomplishments of the NCI in 2018: "Here is one clear fact: It is a great time to be a cancer scientist and a cancer doctor in the United States." said.

To accept his post as NCI director, Dr Sharpless agreed to divest from investment interests, withdraw from industry advisory relationships, and relinquish titles to patents.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.