NCI Corner – Interview with Matthew Holderfield, PhD

NCI to Team Up with Cancer Research UK and the Cancer Research Technology Pioneer Fund on RAS Research

On April 19, 2017, Cancer Research UK (CRUK) and the Cancer Research Technology (CRT) Pioneer Fund announced a commitment of £2.5 million, or about $3.2 million, for a collaboration with the National Cancer Institute (NCI) as part of the RAS Initiative.1

The trio of human RAS oncogenes (KRAS, NRAS, and HRAS) are the most commonly mutated gene family in cancer, and about 35% of lung cancers are driven by activating mutations of KRAS.2 Unlike oncogenic kinases such as ALK and EGFR, RAS proteins have not yet been successfully targeted by therapeutics and have been called “undruggable.”

In 2013, the NCI launched the RAS Initiative and created a hub at the Frederick National Laboratory for Cancer Research (FNLCR) in Maryland to facilitate national and global collaboration among the RAS research community. IASLC Lung Cancer News spoke with Matthew Holderfield, PhD, the RAS Drug Discovery Group Lead at the FNLCR, about this new collaboration and what it means to the RAS Initiative as a whole.

Can you describe the purpose of the NCI RAS Initiative?
In short, the purpose is to develop effective treatments for patients with KRAS mutated cancers. The NCI RAS initiative has several strategies to achieve this. We conduct high throughput small molecule screening to identify new chemical leads that may develop into drugs. We do this ourselves or in collaboration with pharmaceutical companies with an interest in RAS. We also collaborate with academics and companies with ongoing drug discovery or basic research programs focused on targeting KRAS mutated cancers. We can provide reagents, technical support, or start a full collaborative research program, depending on the topic and level of interest.

To what extent is KRAS responsible for the growth and metastatic potential of advanced non-small cell lung cancer (NSCLC)?
This is a really interesting question. We know that KRAS is mutated in about 20%–30% of NSCLC. EGFR and NF1 mutations are also quite common in NSCLC, both of which promote growth by activating KRAS without mutating the KRAS gene itself. So, the actual number of tumors where KRAS plays a role is probably far greater than those with KRAS mutations. However, we also know that not all KRAS mutated cells are KRAS dependent, meaning that if KRAS is removed, some of those cancers will still divide. There are many redundant pathways that can promote cell division, and presumably some tumors will have mutations in other oncogenes independent of KRAS. We are really only starting to understand these mechanisms well enough to make predictions about KRAS dependence and there is a lot more work to be done in this area. Regardless, there are a lot of patients with tumors that activate KRAS, particularly in NSCLC.

Why is RAS considered “undruggable”? Is it a “fellow passenger” as opposed to an “oncogenic driver”?
RAS is definitely an oncogenic driver. The high number of mutations in cancer, and all the preclinical biology, validate that KRAS is an excellent and wellvalidated target. In this case, the term “undruggable” just means that people have tried to develop small molecule inhibitors without success. Kinases and G-protein coupled receptors are among the favorite targets because we understand how to design very good drugs for these proteins. Both types of enzymes have large clefts in the surface of the protein where small molecule substrates can bind. These clefts can be used as a “pocket” for a small molecule inhibitor to bind and disrupt protein function. Small proteins with few pockets or with very high affinity for substrates are not easy to drug. RAS seems to meet all these criteria for a difficult target. It is a relatively small protein with no obvious pockets for a small molecule to bind, with the exception of a single site that binds to GTP. Unfortunately, GTP binds very tightly to RAS, and there is a lot of GTP in the cell, so you would need a lot of drug to overcome GTP. It’s no question that RAS is a difficult target, but I don’t think it’s impossible.

What will the CRUK and the CRT Pioneer Fund contribute to the RAS Initiative?
The CRUK Beatson Institute has already done some NMR screening to find fragments that bind to RAS. These are small compounds that aren’t quite big enough to be a drug, but might be a starting point for developing a drug. This is an approach that we have not taken. So, it’s a completely different way to get to a drug than we are currently pursuing and it’s not something we would otherwise be working on. Additionally, they have enough chemistry support to really make some progress on the project. The NCI RAS initiative is focused primarily on RAS biology, and we depend on our collaborators for chemistry support. In this case, the CRUK Beatson Institute is well resourced for the project and they have some really smart scientists on the team.

Why has it been difficult to develop reliable assays for potential RAS agents?
I think the main reason is because there are no RAS inhibitors. It’s difficult to develop an assay to find a RAS inhibitor if there are no RAS inhibitors to validate the assay. It’s a chicken-or-egg problem that has really plagued the field. One way around this problem is to develop a lot of assays that test the same biology using orthogonal methods. If a potential RAS inhibitor scores positive in 1 assay, it could be an artifact; but a compound that scores in 5 different assays, each testing the function of the compound in slightly different ways, is much more likely to be an interesting drug candidate.

Are there realistic prospects for active intervention? Why have agents like the farnesyl transferase inhibitors failed in the past?
Definitely. I would not be working in the field if I wasn’t optimistic about the possibilities. One significant challenge, and the reason FT inhibitors failed for KRAS mutated cancers, is redundancy. There are actually some very good FT inhibitors that effectively and potently prevent RAS farnesylation. However, mammals have evolved genetic and biochemical redundancies that cause KRAS to be get geranylgeranylated when FT is inhibited. Essentially, the backup system kicks in and keeps KRAS functioning. So, even though the inhibitors work exactly as designed, KRAS dependent cells do not respond to FT inhibitors. Interestingly, another RAS isoform, HRAS, does not get geranyl-geranylated, and HRAS dependent cells are very sensitive to farnesyl transferase inhibitors. Unfortunately, HRAS mutations are far less common than KRAS mutations. Yet FT inhibitors are being evaluated for patients with HRAS mutated cancers. So I think the mechanism is still valid. We just have to find a similar vulnerability for KRAS, the protein that drives cancer progression in 20%–30% of cancers.

New resources and technologies are being brought to bear on the riddle of targeting RAS with small molecule drugs. Should the collaborators at the NCI’s FNLCR and the CRUK Beatson Institute succeed in solving the chicken-or-egg dilemma of developing effective RAS drugs, as well as crafting reliable assays for their analysis, it will open new avenues for RAS drug discovery. Considering the current need for such agents, any progress toward targeting RAS proteins has the potential to have a significant impact in outcomes for many patients with lung cancer and other cancers.

1. The NCRI partnership launches new five-year strategy to accelerate progress in cancer research. CRUK Press Release. April 25, 2017. Available at cancer-news/press-release/2017-04-25-the-ncri-partnershiplaunches- new-five-year-strategy-to-accelerate-progress-in-cancerresearch- 0.

2. NCI RAS Initiative website available at research/key-initiatives/ras.