Norman E. Sharpless
“The notion that cancer’s not one disease, but thousands of diseases is really starting to sink in, and the implications of that fact are being felt throughout [NCI], and it means we have to change how we do everything. I hope that the early days of the Sharpless administration will be remembered as a time when we really bought into that reality and did some things differently,” NCI Director Norman “Ned” Sharpless said in a conversation with The Cancer Letter.
A video of the conversation is posted above.
“My plan is I think the first people that need to hear about my new vision for the NCI are the intramural program and NCI staff,” Sharpless said. “So, I plan to have a town hall sometime in March, date to be determined, followed up by some sort of more national event after that where I can talk about this.
“Both at open forums, but I really want to make sure that the NCI people hear it from me directly first and have an opportunity to ask questions about it.”
Sharpless spoke with Paul Goldberg, editor and publisher of The Cancer Letter.
Thank you very much for sitting down to talk with me. How do you like the job so far?
So far, so good. I’m enjoying the new job greatly. The colleagues are wonderful, the sense of mission at the National Cancer Institute is very strong, and so, working with people who are so talented and so passionate is really a privilege.
My guess is that you’re playing a long game rather than a short game, so 10 years from now, when we all look back at the Sharpless years, you may still be there—or here, rather—what do you want people to say?
I think you’re right about that. The NCI lends itself to gradual change and the nature and size of the culture and the institution is such that the long game is really the way to go. I think it is likely what we’ll say 10 years from now is that this is a time when things were really changing in cancer research.
This notion that cancer’s not one disease, but thousands of diseases is really starting to sink in, and the implications of that fact are being felt throughout the institution, and it means we have to change how we do everything. So, I hope that the early days of the Sharpless administration will be remembered as a time when we really bought into that reality and did some things differently.
And then what happens?
We’ll have to change how we do clinical trials. We’ll have to change how we even hope to gain further knowledge in cancer research. So, for example, I’ve spoken a lot about my belief in the power of Big Data and data aggregation.
The reason I’m fervently passionate about that idea is that it really is one of the only ways that we can make some of the progress I hope to make by—if it’s a thousand diseases, some of those diseases are quite rare, and so the traditional paradigm of a 500-patient clinical trial doesn’t work so well in the new reality of fragmented cancer diagnoses and instead what we’ll need is large aggregated data sets to see and discover the behavior of those sorts of tumor types. Those will inform basic biology, cancer disparities research, survivor research. You know, whatever your interest is in cancer, it’s addressed by aggregated, annotated data sets.
What role should NCI play in bioinformatics? What should you do? What should the industry do?
Well, I think that’s right that there are a number of players in this field that have an important role. I think the National Cancer Institute’s particular role—it can fund research and play a role in standard setting so that if a bunch of different industry players develop different sort of data silos that don’t work together, that doesn’t really advance the research mission to the extent we would like.
I think the NCI can say, “If you’re going to get data elements from an electronic health record, these are the ones you should get in this format. If you’re going to use a sort of strategy to uniquely identify patients, here’s one that works. If you’re going to store data in certain formats, here’s a good way to do it so that they can be aggregated and talk to each other.”
I think that we have a role to play in standard-setting and incentivizing the top research. If you have these huge data sets, how do you mine them? You can’t use your Excel spreadsheet and click on “sort” or whatever for terabytes of data, so making sense of those large data sets is going to require a significant research effort.
I think that that’s probably a better fit for us than trying to create the uber, all things, ginormous, Skynet-type data bank. That, I think, would not be a good idea, because that would preclude innovation. That would prevent the different small companies from doing what they’re doing and the academic entities from trying to develop new structures.
So, I think facilitating research and letting this grow organically and trying to figure out and identify the winners and then force them to work together and aggregate is really a useful role for the NCI.
Is that more of an honest broker role, or is it more of a standard-setter, in a kind of more planning sense?
We certainly can play both of those roles. We have an important honest broker role for example in the [Surveillance, Epidemiology, and End Results Program] dataset, where we hold a lot of de-identified data from multiple different SEER contract sites and make that available to the research community. We similarly have a very important standard-setting role that I alluded to, but I think it’s a lot more than just those two functions.
We also have a research role of identifying the key scientists working in data science and saying, “If we make these sort of datasets available to you, how are you going to use them? What are you going to do with them?” So, I think the research efforts are really important. And, similarly, I think we have a, for want of a better term, a sort of role in data quality assurance in the greater world so that we want to provide large sets of data that we know to be of very high quality, they’re properly annotated and linked that can then be used by the research community for particular endeavors.
The Genomic Data Commons, for example, is not merely a repository of data that the NCI runs, but it’s also a repository of quality, validated data that can be used for these research efforts.
One of the most interesting stories I’m covering now, I think, is the spread of the academic model into the community setting. How does that affect NCI? What should you be doing differently? Is there anything you need to be doing differently?
Yes, I agree with you. This is a really important development and a really big story nationally. I think it makes total sense once you realize that cancer is thousands of diseases. No single institution is going to have enough of these certain rare subtypes to really do meaningful clinical research so multi-institutional studies is the next thing that happens, but that only takes you so far and pretty soon, you realize you’re going to have to do clinical trials and translation of best practices in the community setting.
That’s where things are going, I agree. The ability to do enrollment into clinical trials at community oncology practices is a big new development, and the ability to do sort of cutting-edge research and translate novel therapy in the community setting is a big development. It’s good for patients. They don’t want to drive three hours to the cancer center.
It’s good for research, because clinical trial enrollments in adult oncology is about 3 to 5 percent of patients and I think the only way we’re going to make that better is by enrolling patients in the community oncology setting. So, it’s good for patients, it’s good for research, and I think community oncologists like it. They don’t want to have to refer patients for tertiary care. They prefer to keep patients home and under their care.
So, I think it’s a win-win-win, but how to make it happen? As you know, the National Cancer Institute has a big investment in the so-called National Community Oncology Research Program, NCORP. That is an entity that is a very successful program that we have pledged new investment in, so, we plan to grow the NCORP. It’s where a lot of the research for the NCI-MATCH, meaning the patients on the [Molecular Analysis for Therapy Choice] trial were enrolled at NCORP sites, and similarly, it’s a great fit for certain kinds of research like the TMIST trial, which is this mammography study that will be done over the next seven years.
Not every trial works in the NCORP setting, but some work really, really well. I think that a commitment to continued research in the community setting and figuring out how to grow that and figuring out how to do a more diverse set of trials in the community setting are future challenges.
As I look at this—and I can only look at it as a reporter, but you’re a former cancer center director and you’re now the NCI director—what does all of this, all of the things you’ve just described do to the definition of a cancer center? Does it need to be redefined? Maybe this is a premature question to ask.
No, no, I think it’s a good question. I don’t think the move to do more care closer to home is in any way a threat to cancer centers. I think that most cancer centers have fairly large catchment areas, in fact, and are really challenged by the NCI to figure out how to do research throughout the patients in their entire catchment area. So, they’ve all, many of them have started thinking about how to use satellite facilities or how to collaborate with more local partners.
I think this, as I said, makes sense, because cancer centers are very much judged on the metric of incident number of cases, enrollment to clinical trials and sort of novel research efforts and, to do that in the modern era of thousands of different diseases requires outreach.
I don’t think any sensible cancer center feels like the old days where everybody drove three hours to see them is the model going forward.
What about clinical trials? What kind of clinical trials should NCI do? What kind of clinical trials, more importantly, shouldn’t NCI do? What should the role for the industry be? Where’s the line between the industry and NCI? Where should it be?
Yes. Industry has a huge role to play in this, and certainly significant resources to develop to address this question. Industry doesn’t do certain kinds of trials where there are important research questions and I think that’s really where NCI plays its most critical role.
So, for example, the willingness of pharmaceutical companies to test their drugs in pediatric populations has been probably, I would say the appetite has been lower to do that than in adult populations. So, leadership in pediatric oncology has been an important thing for the NCI to do.
Similarly, if you think about it, there’s a lot of incentive for a pharma company to test their new drug, their new molecular entity, but there’s less incentive for them to test new models of surgery or radiation oncology. So, those kinds of trials, multi-modality trials that involve surgery, radiation, and chemotherapy are a good fit for the NCI.
There are a lot of challenges to the modern clinical trial operation. Nothing has felt this reality of the many different kinds of cancer fact more than clinical trials, because the old paradigm was relatively simple, from an administrative point of view. You randomize 500 patients one way or the other. That’s a pretty simple structure.
But now what happens is some of the lung cancer [patients] come in, they get sequenced for 400 genes. If they have this one event, they can go in this trial. If they have another event, they can go on that trial. So, there’s this huge burden of patients that are screened and not even treated, but there’s still costs and expenses with those trials. The per-patient costs for accrual are much higher.
So, the clinical trials enterprise has really had to adapt and modernize, and it has not finished with that process. It is a bit painful and has caused the process to go more slowly than we would like in places, but I think there’s pretty uniform agreement now among the really good clinical trials in the United States that things have to change and that what the vision of the future looks like is, I think, coming into focus.
For years now, really for many decades, the glib thing to say has been, “Here’s how you free up some NCI money. Kill the intramural program. Get rid of Frederick. What have they ever done?” I’m not sure that argument holds water anymore, with HPV and with CAR-T. What size and role do you see for the intramural research program and what size and role do you see for the Frederick lab?
Yes. Several things to talk about there. So, as a long-time veteran of the extramural program, I have heard this statement. I would be disingenuous if I pretended that no one had ever said anything bad about the intramural program. Having said that, even before I got here, I think I had an appreciation for the quality of the science that goes on intramurally and now, as NCI director, I really have really become very impressed.
One thing to know is that the science effort, the clinician scientists and basic scientists within the intramural program, that number has shrunk significantly over the last decade. It’s now 270 scientists. We’ve looked on an analysis using hard metrics, publications, patents, citations by clinical trials of that cohort of scientists, and if you compare those NCI intramural clinician scientists and basic scientists to any group of extramural researchers, they compare very favorably.
I mean, they’re an awesome group of scientists. It doesn’t—you know, 275 people—they don’t work on everything, but they work very, very well on certain topics. You alluded to HPV. I believe the world’s leading HPV research has gone on here in the intramural program, led by Doug Lowy, my predecessor and deputy and wonderful sherpa in this endeavor.
I think the CAR-T effort, Steve Rosenberg [chief of the Surgery Branch, senior investigator, and head of the Tumor Immunology Section at NCI’s Center for Cancer Research] work, has been transformative, and is now starting to see realization and FDA-approved therapies. Marston Linehan [chief of the Urologic Oncology Branch and senior investigator at NCI’s Center for Cancer Research] has done this marvelous research on familial genitourinary cancer that’s been fantastic.
There’s also the intramural health services research arm, the Division of Cancer Epidemiology and Genetics, that is a national treasure. They do kinds of work that is just not doable anywhere else in the world, looking at the role of sort of types of occupational exposures and carcinogens and cancer risk, aggregating enormous data sets for GWAS [genome-wide association] studies, doing very elegant research on both survivorship and cancer therapy in the real world. It’s an effort that is terrific and I think of great value and easy to defend.
Frederick is a different thing. Frederick National Lab is not really part of the intramural program, although it supports intramural research. Frederick has existed since the days of the National Cancer Act, has gone through some iterations and I think it was Dr. Harold Varmus, who decided that if we were going to have a national lab in Frederick, Md., that it had to be awesome, that it had to be really, really great and he invested both the finances and the administrative discipline to try and transform Frederick. He set it on a very good trajectory that I believe continues today.
Frederick now does some very exciting cutting-edge research like the RAS Initiative. It has these very important support capabilities that are quite marvelous where you can spin up assays and new therapeutic production facilities there in a very short time, at least for federal academia standards. I think Frederick is a valuable entity.
It has new leadership with Ethan Dmitrovsky, who I think is a very good person to play a role in leading Frederick, and I believe that Frederick is great, but we have to continually ask what can it do, and what should it do, and how can we use its special capabilities most optimally? And so, we spend a lot of time thinking about new things for Frederick to do.
Should both of these entities, maybe they should or shouldn’t be grouped. Sorry if I was unfair. Should they be roughly where they are? Should they grow? Should they shrink? Or is this a premature question to ask?
No, I think the intramural program has shrunk, and it’s probably gotten to a size that makes sense. The clinical enterprise here, intramurally, should not get any smaller. In fact, I think probably should be a little larger—not 50% larger—10% larger. There are some areas where we need to do some recruitment. The research enterprise, I think, is a good size. It’s certainly big enough to feel like it’s a dynamic environment where they’re wonderful collaborators. It feels like any sort of academic research program in the country, but yet not so big that it’s diverting resources away from the rest of the extramural world.
I think Frederick is a different animal. The size of Frederick depends on what Frederick is doing. If we figure out for Frederick to do new things we might increase capacity up there and if we pare back some of the activities we might see decrease in Frederick’s size. Right now, the portfolio of Frederick I think is pretty strong.
The RAS Initiative is very good, some of the other efforts related to… There’s a GP facility that does important production of therapeutic modalities for people at the NCI, there’s a nanotechnology catheterization lab, there are a bunch of capabilities that support the intramural program. It’s doing a bunch of useful things that I think are strong.
In fact, one of the things I’m most excited about Frederick is we’re building this sort of cryo-EM support facility, because we have appreciated that cryo-electron microscopy is so important nationally, but a lot of places, even though they’re buying and investing in the technology, really still aren’t able to solve structures to the level of detail that they would like and so the NCI is a facility where people can now send us their matrices, their lattices and we will do some of the imaging for them and then help them with the data interpretation, which I think is a well-used resource and is going to grow a lot.
At the most recent National Cancer Advisory Board meeting (The Cancer Letter, Feb. 16) you spoke quite a bit about the RPG pool and the size of it and how it should be structured. It was a long and interesting discussion. If you were to summarize what the issues are and what the outcome is, if there is an outcome, what is it?
Yes. The Research Project Grant pool, or the RPG pool, is an acronym I didn’t understand until recently—it’s one of many—is the pool of money whereby the NCI supports investigator-initiated science.
So, for the most part, it’s not completely exclusive, but for the most part, R01s and P01s and some of the U01s, the grants that an investigator, working on their own, dreams up in the outside world and says, “I’d like a million or $2 million to work in this area,” and applies to the NCI, that’s where it comes from.
I would argue it’s been a good investment, that some of the most important advances in cancer research have come from investigator-initiated science funded through R01s, through individuals. But it’s an expensive pot of money, so of the sort of $5.6 billion NCI budget, roughly $2 billion goes to the RPG pool and that’s for $500 million-ish for new awards, for competing and new awards, and then the rest, $1.5 billion for renewals, because the grants are typically four to five years, non-competing awards.
So, whenever we fund a grant, it has four out-years as well, generally. Decisions we make in the RPG pool, even if they’re quite small in size, like we’d like to increase everybody’s grant from—typically on the non-competing awards we would cut them a little bit, so, instead of funding them at 100% we’d fund them at 98% or 94%.
So, a couple of percentage points in the RPG pool turns into $15 million to $100 million annually. So, even minor decisions to the RPG pool have big implications for the NCI budget in the present year and for future years.
Against that backdrop, we have looked at the trends of the rate at which people are sending us new grants and that has gone up sharply over the last few years. We have a lot of people with great ideas in cancer research, they’re sending those proposals to the NCI, and we want to fund them. But if you get more awards and you don’t put in more money, then the paylines go down. That is very frustrating for the extramural community. I lived that. I know all about low paylines.
It’s a tough message. If the Congress is giving us more money and we have the Moonshot funding but paylines are going down, that seems somewhat discordant. So, we have wanted very much to preserve pay lines but to also fully fund grants at 100 percent for a variety of reasons. These small little cuts that we’ve been imposing every year add up. We take a big cut in the beginning of the grant and then small little cuts every year and it is very frustrating to the extramural community as well. So, we’re trying to preserve pay lines and fully fund non-competing renewals.
Two laudable goals that I think everybody in the extramural community would like; I proposed to our advisory board that we do that. That is something internally we’ve talked about among our senior program leaders and there is enthusiasm to do that within the NCI. As I said, the belief that the RPG pool is a good investment is pervasive throughout the institution, but we also wanted to sound out that with our advisory boards. And they were basically enthusiastic. They thought that these cuts really are detrimental to science, that to the extent possible, we should fund grants fully when able, that we should try and preserve paylines.
So, what that translates to in reality, though, is about $125 million more for the RPG pool this year. That’s kind of a one-time thing. It won’t go up that much every year, but because of a bunch of things—out-year commitments made in prior years—this year is a particularly striking year. I don’t think we would put $125 million in the RPG pool every year, but that’s what it means for this year. So, we would like to do that.
We’ll know if we’ll be able to do that when we learn what our budget is, 2018 enacted budget is from Congress, which we do not know yet. We are hopeful that we’ll be able to put significant additional resources to the RPG pool this year.
That’s fascinating. So, you’re saying that the RPG pool should overall stay where it is.
Or even, it actually grows a bit. I think it does, the con is that it does decrease our flexibility in 2018. And, of course, if we fund more grants this year, then those grants have out-year costs as well, so, it limits our flexibility this year and in future years. So, it’s not something that we take lightly. There are a lot of things that we could do with those funds if we didn’t use them for the RPG pool.
As I mentioned at the board meeting, I’ve asked my senior program leaders for new ideas and what they would cost, and I’ve met with each one of the divisions and centers independently and gone through their budget and their future budget, and I’ve been proposed $300 million of new great ideas. There’s no shortage of good stuff to do, so, if we didn’t use the RPG pool money for that, we could do some of these other ideas.
But I think the RPG pool is sufficiently important and the preservation of that fund is sufficiently important to the extramural community and cancer research engine in general, that that’s sort of got to be priority number one this year.
And that may stay that way for a while?
Well, like next year we envision more flexibility. Next year we would have to put additional money into the RPG pool again, but it would be much less. It would not be this size. It would be more like $50 million.
If we had the same kind of budget in 2019 that we might get in 2018, then we might be able to do other things. Of course, now we’re talking about the hypotheticals of congressional funding more than 12 months away, and that’s bad. That’s hard to do, right?
Let’s not do that. So, you’ve mentioned that you’ve been on a listening tour that began in the summer and it ends in March. Which day in March, so we can all mark our calendars? How will you know?
I don’t know the exact day. I’ll explain how I envision this. It really started before I was sworn in. I was able to go around and meet with a lot of people, but because of laws about things that you can tell a non-federal employee, I really couldn’t begin it in earnest until I was sworn in October 17. So, I felt like six months was about the right length of time.
I think I’m feeling now that I am comfortable in meetings. I know what most of the acronyms stand for. I know who everybody is, so I feel like now I’m having not just information receipt, but I’m actually able to make decisions and I think the nature of the institution, if I extended it much longer than that it would probably start to frustrate people. I need to at some point to become a little more definitive about my goals for the NCI.
My plan is, I think, the first people that need to hear about my new vision for the NCI are the intramural program and NCI staff. So, I plan to have a town hall sometime in March, date to be determined, followed up by some sort of more national event after that where I can talk about this. Both at open forums, but I really want to make sure that the NCI people hear it from me directly first and have an opportunity to ask questions about it.
Can you summarize what you’ve heard during your listening tour?
Wow. That’s challenging. It borders from the incredibly weedy but important, like how SEER works, or how the RPG pool works, or how Frederick works. Some of these things have, you know, Frederick has a lot to do with the occult practices of government accounting, government contracting, so it ranges from stuff like that that’s very, very important but I would be hard pressed to, I think I would bore the room to tears if I were to try and explain some of those details.
But ranging to the what should the future of cancer research be? The very big picture—what are we doing right? What are we doing wrong? Where do things have to change and how can we do things faster? Really, I think one of the most important parts of that conversation, frankly, for the bigger picture stuff, has been with the patient advocates.
They’ve really argued that patients have to be the north star of this endeavor, that we really have to keep in mind what’s best for patients with cancer to reduce the burden of cancer through prevention and screening, early detection and therapy. What may be convenient for extramural researchers or good for the intramural program or fortunate for the NCI director, none of that is as important as does it or does it not improve our success against cancer in this sort of fight against cancer for our patients.
You’re talking about people who knocked on your door and came to see you, or are you talking about people located for you through the NCI advocacy outreach program?
A lot of both. I’ve answered a lot of email. I’ve had brunch at Silver Diner in Bethesda with a granddad who wanted to tell me about his grandchild who died of glioma. I’ve met with organized advocacy groups.
Just the passion around cancer among survivors and advocates and family members and loved ones is just unbelievable. Until you’ve sat with these people and asked them what frustrates them and what they’d like to see the NCI do better, that exercise has proven phenomenally informative.
How’s that different from what you heard as a clinician and as a center director? Is that a different level?
No. That’s a good question, because actually I thought I knew. I have this joke that I like to tell that my wife and I lived together before we were married and we thought we knew what marriage was like. Then when we got married, we realized, “Are you going to make that noise with a spoon for the rest of your life?”
So, I thought, doing cancer one-on-one as a doctor, I had learned a lot about the burdens of a specific patient one at a time. As a physician you try and often talk your patients into therapies that have a lot of toxicity and have both side effects and benefits and it really involves knowing that particular patient and what they want in terms of what’s the right decision. Often it’s not so much medical as patient-tailored.
So, I had had that sense of individual one-on-one with patients of leukemia for a long time, and then as a cancer center director, I think the scope of reference increased substantially. But still that was really focused around the patients within a specific hospital or a specific catchment area. The things that were really good at the University of North Carolina were areas of particular focus, for example.
But now, as NCI director, there’s, sort of, no kind of cancer in the United States that I can’t be interested in. There’s no issue around cancer, be it prevention, survivorship, Big Data usage, therapy, prevention, all of those things are important and have to be balanced, because everything has trade offs. This kind of increase in the scope of interest has been quite striking and surprising, in a good way.
And advocates are probably some of the most well-informed people about science, too, because they’re kind of generalists.
No, I think that’s true. I think some of these advocates have worked in certain areas for 10, 20 years and they have learned everything one can learn about the cancer that interests them and they are quite sophisticated in some instances.
So, the listening tour is over, almost. We will find out when. What comes next?
In the near future, what I expect to happen is we will sort of begin to focus on a few specific areas. I’ve been sort of adopting this three-bin philosophy. There are things that we have to do, like the clinical trials endeavor. I alluded to this, that it works, but it needs some structural changes to make it work as well as everyone would like. There are things that we would like to do. There are areas of new effort, new investment for the NCI and that’s particularly fun to think about as a new director.
And then there are things that we’re already doing that need continued to significant investment. Those can be kind of nice. A prior NCI director told me, “One of the good things about the job is you get to take credit for what your predecessors did,” and so, some of these great programs have been going on for years, but they’re coming to fruition in the next few years. They’ll be important to our cancer research endeavor.
So, that’s sort of the three bins and we have plans for all of those. It involves making choices. If the budget only goes up a little bit and we have a lot of new ideas, then that means that we necessarily have to stop doing some things or increase efficiency, which I also am eager to try and do. And that has, by the way, been a strong signal from this administration.
The secretary of Health and Human Services and the White House have both sent the clear signal to the NIH that they like the NIH, they value biomedical research, but they want it to be done in an efficient and competent manner and so, there will be emphasis on that as well.
Well, thank you. I’m looking forward to covering all of this.
Thank you for the pleasure of sitting down today.