Glen Campbell, who was a member of The Wrecking Crew, once remarked that he had worked hard for ten years to become an overnight success in 1967, with thanks to John Hartford. Without stretching too much, this same trajectory, extended by more than ten years, has been illustrated recently in biomedical research by the case of daraxonrasib as a treatment for pancreatic cancer. Pancreatic cancer is the fourth leading cause of cancer deaths in the United States, with an estimated 67,530 cases in 2026 and 52,740 deaths. The 5-year relative survival rate for patients with metastatic pancreatic cancer is only 3%. This new drug can be fairly said to be taking the world of clinical oncology by storm:
Detailed results from the daraxonrasib clinical trial conducted by the drug’s maker, the biotech company Revolutions Medicines, were presented here at the plenary session of the annual meeting of the American Society of Clinical Oncology. The study was published simultaneously in the New England Journal of Medicine.
The deeper look at the study data confirmed what was previously announced in April by the company in a press release: Patients with advanced pancreatic cancer who received daraxonrasib as a second-line treatment achieved a median overall survival of 13.2 months, compared to 6.7 months for patients offered standard chemotherapy. Statistically, daraxonrasib reduced the risk of death by 60% compared with chemotherapy.
“Daraxonrasib will have a tremendous impact,” said Brian Wolpin, director of the Hale Family Center for Pancreatic Cancer Research at Dana-Farber Cancer Institute. He was a study investigator and is presenting the study results at ASCO on Sunday afternoon.
“In the near term, this will become a new standard treatment for patients. Longer term, this is showing the field that we can move beyond chemotherapy in this disease, and that the investment in all the science and biology over decades is finally starting to pay off,” he added.
The remarkable thing about daraxonrasib is its target, the oncogene RAS. The mutant KRAS is present in about 90% of pancreatic cancers as well as in cancers of the colon and lung and other organs, but it has been an elusive target. To understand the importance of this research, it is necessary to understand what RAS proteins do in the cell. What follows is a bit technical, but much easier to understand than “puts and calls on Wall Street,” perhaps because RAS proteins are part of my material reality (yes, I am showing both my predilections and ignorance here). RAS proteins were first described as a superfamily of small G-proteins that act as molecular switches that regulate cell proliferation, dysregulation of which is an early step in cancer progression. A short lesson in cell biology:
G-proteins are active when bound to GTP (an analog of ATP that is not used as the “energy currency” of the cell). G-proteins have an intrinsic GTPase activity that hydrolyzes the GTP to GDP. The GDP form of the protein is inactive, and the transition from GTP- to GDP-state is energetically downhill and thus irreversible in the cell. The switch cannot be simply flipped back on. Accessory proteins called GAPs (GTPase-activating protein) speed up the reaction by stimulating the intrinsic GTPase that turns the switch off. GEFs (guanine nucleotide exchange factor) replace GDP with GTP and thereby reactivate the G-protein so the cycle can begin again. The mutant KRAS was originally described as the Kirsten Rat sarcoma virus oncogene. This mutant contributes to cancer progression because cells proliferate when they are not supposed to. Subsequent mutations generally lead to a primary tumor and metastasis. Other RAS proteins are involved in cell adhesion (Rap), cell motility (Rho), and trafficking within the cell (Rab, Arf). The G-protein switch is one reason why cells “know” what they are supposed to do based on signals from other cells, with the direction dictated by transition from active (GTP-bound: on) to inactive (GDP-bound: off).
The oncogenic KRAS had long been considered an “undruggable” target. Research of Kevan Shokat’s group at the University of California-San Francisco showed this not true and their research eventually led to daraxonrasib. But the path was long and winding and presents an object lesson in scientific progress as a sequence of discoveries and digressions that can be understood only in hindsight.
The first discovery in this chain was that a viral oncogene is a normal cellular protein from the host that was hijacked by a cancer-causing virus and then mutated during insertion into the viral genome and/or viral replication. Upon infection of the host by the virus carrying the mutation, cancer can be the result. Peyton Rous identified Rous Sarcoma Virus as a cause of cancer in chickens in 1916. He was awarded a Nobel Prize fifty years later in 1966. Harold Varmus and J. Michael Bishop identified c-src (cellular src) as the normal cognate gene of the cancer-causing v-src (viral src) in 1976 and were awarded a Nobel Prize in 1989. G-proteins were discovered as trimeric signal transduction proteins in 1980 by Alfred Gilman and Martin Rodbell, who were awarded a Nobel Prize in 1994. Small monomeric G-proteins of the RAS superfamily were identified soon after. RAS was identified as critical for the regulation of cell proliferation as the eukaryotic cell cycle was described beginning in the 1980s. This research involved thousands of researchers, nearly all of whom were supported by the National Institutes of Health and similar organizations in other countries, plus another agency as we will see. Without this foundation, there would be no daraxonrasib as a rational, if unforeseen, outcome of this research.
Going back to Kevan Shokat and his research group, they published a short comment in Nature Chemical Biology (October 2024; technical but very readable) on how daraxonrasib was developed as a description of how most scientific discoveries are made:
RAS family proteins are small GTPases that cycle between an inactive GDP-bound conformation and an active GTP-bound conformation to relay signals to downstream pathways. Despite their central role in driving cancer, which was identified in the early 1980s, RAS proteins were widely considered to be undruggable until only recently. This is in stark contrast with perhaps the most ‘druggable’ class of cancer drivers, protein kinases, which also bind to nucleotide triphosphates.
…
The early 2000s saw an outpouring of cancer genomic data, and among these, the most frequent RAS mutation in lung cancer stood out to us — KRAS(G12C). Cysteine is the most reactive amino acid, and this mutation gave us a chemical handle right next to the nucleotide pocket and regions of RAS that are most critical for cancer signaling. In 2013, we, in the Shokat lab, published a proof-of-concept for KRAS inhibition by directly targeting the mutant cysteine of oncogenic KRAS(G12C), starting with a disulfide tethering fragment-based screen (Fig. 1). Over the 10 years since our report, the field has seen an explosion of KRAS-directed preclinical and clinical drug development, including accelerated US Food and Drug Administration (FDA) approval of two KRAS(G12C) inhibitors in lung cancer and more than 30 RAS inhibitors now in clinical trials (Supplementary Table 1; gene in italics, protein in plain text).
There is a lot hidden here except to the very few biochemists among us. The key is that the mutant KRAS has the amino acid cysteine (C) at position 12 in the protein chain of amino acids instead of a glycine (G). Cysteine is highly reactive due to its sulfhydryl group (-SH). Therefore it can be attacked by any number of reactive compounds, some of which inhibit KRAS function. The normal RAS gene in the cell has glycine (-H) which is reactive with nothing under biological conditions. One advance led to another. The KRAS(G12C) mutant protein is primarily in the GDP-bound state and therefore in the “off” conformation. Further research showed how to make a multi-component inhibitor complex of KRAS-GTP in the “on” conformation. Daraxonrisib was the eventual result. But this outcome was not a foregone conclusion:
While RAS as the most frequently mutated oncogene always generates interest from funding agencies, the feasibility of a particular approach is often the main driver of the funding decision. Even acknowledging the advantage of the mutant cysteine, it is without question that our approach had slim odds of success, and perhaps it is unsurprising that our initial application for funding did not advance. Thankfully, we were able to proceed with the help of funding focused on individuals and not the specific project. In particular, the flexibility of Howard Hughes Medical Institute funding was essential for us to undertake this risky adventure in the hopes of making a groundbreaking discovery.
While supporting researchers directly fosters a culture of innovation and enables high-risk/high-reward science, institutional and public funding also play a crucial part. Having state-of-the-art equipment available (such as NMR, FACS, synchrotron beamline access, and so on), along with databases (such as COSMIC, PubChem and SGC), was also crucial to the success of our work. [1] Perhaps more important than funding, facilities and databases was the constant encouragement of colleagues with decades long passion for RAS, including Frank McCormick, Mariano Barbacid, Channing Der, Kevin Shannon, Julian Downward, Fred Wittinghofer and Roger Goody.
So, had Kevan Shokat and his team not been generously supported by the Howard Hughes Medical Institute to follow their collective intuition, progress may have stopped at inhibitors attached to that reactive cysteine in KRAS. Instead, daraxonrasib is now available as the first drug with great promise against pancreatic cancer. The drug is not a cure, but as a second-line treatment it provides a path forward for many patients and clinical oncologists and cancer researchers. For example, the significant life extension provided by daraxonrasib might give a patient the time necessary for a personalized cancer vaccine approach to work as a cure (we have discussed these here previously). And this:
The term ‘undruggable’ has morphed from a deterrent to a motivation, often simply used to describe a target that has recently been drugged for the first time. But even though the term no longer means the same thing, the sentiment remains. If we are to drug the undruggable targets, it will be essential to continue looking at scientific dogma with a skeptical eye. New approaches, new technology, or just a new set of eyes can upend historical understanding of even well-studied proteins such as RAS. If we as a community are to take on these challenges, greater security and flexibility of funding will be essential. This means funding creative individuals and teams and giving them the leeway to pursue well-reasoned, but risky projects. The innovative hub-and-spoke model pioneered by the US National Cancer Institute RAS Initiative has provided invaluable support for the past decade of resurgence in RAS drug discovery.
…
It would be impossible to count the number of times fortune swung in our favor along the way, from the frequency of the G12C mutation in lung cancer, to the biophysical properties of this mutant and the fragments in the library we screened (elements of the top hit from our screen even persisted into both FDA-approved G12C inhibitors). Perhaps the most gratifying moment of all came in 2020, when Jon saw his first patient with KRAS(G12C)-mutant lung cancer and could share with Kevan and Ulf that the patient had had a dramatic response to a KRAS inhibitor. As a team, we have had the good fortune to experience first-hand the full arc of KRAS drug development, from identifying initial hits in a fragment screen to seeing patients benefitting in the clinic. Despite remarkable progress in direct RAS inhibitors, we have yet to maximize their potential. As we begin to see efficacy against non-cysteine RAS mutants and in RAS-driven cancers other than lung, and as we hone in on effective combination therapies, it is clear that the best is yet to come.
The overnight success of daraxonrasib was a long time in coming.
There is also another critical lesson implicit in the long chain of discoveries that led to daraxonrasib. Virtually every scientific advance in biology and biomedical science is incremental, despite “incremental” being a kiss of death in a reviewer’s report. This is primarily because every protein has at least 2.5 billion years of evolutionary history behind it. Evolution is the work of a tinkerer, not an engineer (e.g., consider the GAPs and GEFs covered above as ad hoc additions to the basic system that make it tunable). Had conventional wisdom been followed, the incremental advances discussed here would have stopped long ago. Kevan Shokat and his research team keyed in on the chemically reactive mutant in their attack on KRAS. They and their colleagues, generously acknowledged and well known to those of us who have followed them from a distance, are the men and women entrusted with our science. In a short introduction to a Chemical Reviews special issue “Drugging the Undruggable,” Shokat and Ziyang Zhang note that, “To this day, the word “undruggable” has yet to appear in any major dictionary. We hope it never does.” As do we all, including former Senator Ben Sasse of Nebraska and tens of thousands of other pancreatic cancer patients and their families.
Finally, getting back to the current funding climate, one might reasonably ask: Is the RAS Initiative at the Frederick (Maryland) National Laboratory for Cancer Research one of the “scientific priorities” of the current administration, which are set to become the deciding factors in who gets what support for “high-priority, gold-standard (sic)” science? Once again, we can only hope so.
But we shouldn’t have to. It goes without saying that no one can ever tell what the answer is ex ante or from where the most important advances will come. That’s why the experiments are done. Whether we can afford to do the experiments is a simple political question. The opportunity costs of not doing the experiments will remain unknowable but they are very large. Only 25% of our heralded F-35s are mission ready at any given time. If only 25% of our experiments work, we will still receive a great benefit at a much lower aggregate cost. And the experiments that “did not work” will have told us something important by showing that a mechanism inconsistent with expectations is at work. The idle 75% of F-35s sitting in their hangars with their various hoods up are just dead weight, or spare parts repositories.
Note
[1] This state-of-the-art equipment is funded in large part by the indirect costs (overhead) that come with NIH support and have exercised politicians for the past fifteen months. Without this instrumentation and other resources, the research cannot be done. While private agencies limit overhead to 10-15%, no scientist without underlying NIH support receives these generally smaller basic and clinical research awards.
By
