Harold Varmus, the high-profile director of the US National Institutes of Health (NIH) from 1993 to 1999, returned to the biomedical agency last July as director of the National Cancer Institute (NCI). In 1989, Varmus shared the Nobel Prize in Physiology or Medicine for his studies on the genetic basis of cancer. More recently, he headed up the Memorial Sloan-Kettering Cancer Center in New York. A year after taking the reins at the NCI, Varmus spoke to Nature about his latest role and about the disease that has defined his career.

You had already headed the entire NIH. Why did you take the job as NCI chief?

When I was the NIH director, I often expressed envy of institute directors: they had the money and ran the scientific programmes. I was right — this job is more interesting.

What have been particular satisfactions or successes during your first year at the NCI?

Refurbishing the leadership team at the NCI, pursuing some important initiatives in cancer genomics and global health, identifying 'provocative questions' — important but neglected questions about many aspects of cancer — and working with some extraordinary colleagues. It is also a pleasure to be in place when important projects come to fruition: the National Lung Screening Trial, the pilot phase of the Cancer Genome Atlas, therapeutic trials for metastatic melanoma, and many smaller-scale studies of cancer biology. And I am happy to return to the best urban bike commute in the US: 12 miles through Rock Creek Park.

The NIH is facing its toughest budget in decades. How has that affected your first year at the NCI?

Some effects of this year's budget are self-evident. An actual decline in real dollars — unprecedented in my time in government — has slowed the progress that is so important to the public, caused a lot of distress in our scientific community, and required extra attention from many NCI staff, including me. I have tried to make fair decisions without losing sight of the fact that we do have a US$5-billion budget and should be able to do many things, including some new things, with those funds to take advantage of unparalleled scientific opportunities.

The public has high hopes that investment in the NCI will lead to cures for cancer. Can you deliver?

Hope is essential for public support, but it has to be framed around the reality of cancer, not founded on simplistic concepts of 'the war on cancer' or 'a cure for cancer'. Cancer is a collection of many diseases with common principles, and each disease will have to be understood and more effectively controlled on its own terms.

What should basic scientists know about your intentions and priorities for the institute?

All basic scientists who look to the NCI for funding should know that I will tolerate no retreat on the study of model systems and the pursuit of fundamental biological principles. We have just begun to catalogue genetic and epigenetic changes in cancer cells and to probe the significance of those changes. To make optimal use of that information, we need a more profound understanding of cell and developmental biology.

At a time of historically low grant-application success rates, what are you doing to protect investigator-initiated research?

We are attempting to fund about 1,100 new grants this year by making small reductions in most components of our budget; because the number of applications remains high, the success rates will be relatively low. Under these circumstances, we do not have a traditional, sharp payline. The NCI scientific programme leaders meet regularly to ensure that we are not ignoring highly original proposals and that we are not creating an unbalanced grant portfolio.

What would you say to those who fear that the NIH is moving away from investigator-initiated research and towards large-scale projects?

I don't see much evidence for the claim that the NIH is 'moving away' from investigator-initiated research. It is the largest component of the NCI's portfolio and the one that I am working hardest to protect. But there are programmes that need to be undertaken as community efforts because they are expensive and difficult, yet promise advances: the renovated clinical-trials system, genomic studies and certain translational activities. This does not mean taking away responsibilities from our investigators; it means working closely with the relevant investigators to ensure that we develop large-scale projects in a sensible way.

What is your sense of the public's understanding of cancer science and medicine?

I think it is becoming more sophisticated, especially as new genetic findings and the first targeted therapies are more widely recognized. We are also helped by Siddhartha Mukherjee's prize-winning book, The Emperor of All Maladies: A Biography of Cancer. The book explains cancer research sensibly and forcefully, and it portrays cancer in ways that show both how far we have come in treating some cancers and how much more can be done.

What have been the biggest challenges of your first year as NCI director?

The most obvious is the restricted budget. It can also be difficult to attract the best people to tough jobs at the NCI. These may look less appealing than they should because of our fiscal situation, non-competitive salaries for clinical personnel and strict ethics rules.

What was your 'provocative questions' exercise, and what has come of it?

Some colleagues and I have been conducting workshops and building an interactive website to identify important but neglected questions about many aspects of cancer. The intent has been to ensure that we are thinking about all avenues of investigation, not just doing the obvious things or complaining about the budget. Recently, the NCI's Board of Scientific Advisors approved a request for applications to address provocative questions on a selected list; more detailed information about this initiative will soon be posted on the website and published in an essay.

For 2011, all extramural grantees based at the NIH experienced a 1% cut in funding levels for existing grants, except for NCI grantees, who received a 3% cut. Why did NCI-funded researchers face deeper cuts?

The situation is a bit more complex that your question suggests. I announced in a letter to all NCI grantees on 9 May that this year we will pay 97% of the FY2010 [2010 fiscal year] levels to all grantees with non-competing awards — Type 5 renewals — whether the awards are modular or not. Other NIH institutes and centres agreed to pay 99% of last year's levels for their non-modular Type 5 awards but varied amounts, some as low as 96%, for their modular awards. As my letter explained, the NCI faced an unusually large 'commitment base' and, even with these reductions and others mentioned below, would be unable to fund as many new grants as we did in FY2009 and FY2010. Because new grants, especially to new investigators, are the lifeblood of the institute, we decided to make special efforts to issue a number of new grants that is as high as reasonably possible, probably around 1,100. The causes of our large commitment base are complex, but include a recent increase in the average size of new grants, a requirement to continue funding some grants initiated with Recovery Act funds, and at least one construction project that could not be halted.

You have also been making some deep cuts to the intramural programme and the cancer-centre programme. Why? And what has been the reaction to these cuts?

The reductions we have made this year have been significant, but I wouldn't call them deep. Moreover, they affect virtually all aspects of the NCI portfolio, not just the centres or the intramural programmes. Of course, some cuts could become deep in the future if the fiscal situation does not improve. I have made many efforts to explain those reductions through talks at society meetings, town hall events and the letter to grantees. With rare exceptions, members of our constituencies understand the predicament we are facing together and are directing their attention to the nation's economy, its choice of priorities and its representatives in Congress.

Translational and clinical research have received considerable attention recently, both in Congress and from the public. What balance of basic, translational and clinical research are you seeking at the NCI to conquer cancer as quickly as possible?

There are no simple ratios that can serve as answers. The NCI has responsibilities in all three domains and will try to make productive investments in research in all three by balancing needs for results against the opportunities for obtaining those results.

Cancer research is supported by many institutes and centres at the NIH. Has the level of cooperation and collaboration among the institutes and centres improved, remained the same or declined since you were NIH director?

Without making superficial comparisons with a different era, I would simply say that I've been impressed with two things: the collegial relations among the institute and centre directors with whom I've worked and the multiplicity of tasks that the directors have undertaken in concert. These joint tasks include governance of the NIH Clinical Center, development of plans for the NIH Common Fund, campus financial management, recruitment of new institute and centre directors, and responses to recommendations from the Scientific Management Review Board.