Kevin Marks, PhD, has been involved in successful endeavors to create meaningful medicines over his 20-year career as a cancer drug hunter, including spells as the head of oncology drug discovery at Novartis Institutes for BioMedical Research (NIBR) and the head of biology at Agios Pharmaceuticals. Still, Marks was losing sleep over the problem of cancer drug resistance.
“All too often, the actual benefits to patients are short-lived because tumors are heterogeneous and drug resistance emerges quickly,” said Marks. His commitment to the cause led him to a position as an entrepreneur-in-residence at Google Ventures (GV). “I felt that to have a chance at doing something truly unique, I needed to start with a blank piece of paper.”
Two years ago, Marks decided that he needed to flip the 30-year-old script for cancer drug discovery—find the oncogene, inhibit the oncogene. Instead, he set about seeking ways to activate oncogene pathways.
Together with Bill Sellers, MD, director of the cancer program of the Broad Institute, and Mike Dillon, PhD, former CSO of IDEAYA Biosciences and former global head of discovery chemistry for oncology at NIBR, Marks has co-founded Delphia Therapeutics to develop targeted cancer medicines based on cancer’s surprising vulnerability to oncogene overactivation.
Behind a $67-million series A financing led by GV, Nextech Invest, Polaris Innovation Fund, and Alexandria Venture Investments, Delphia integrates tumor genetics, novel functional genomic approaches, and studies of inhibitor drug resistance to identify targets that drive oncogene overactivation.
“Cancers cannot tolerate excessive activation of oncogenes, which is surprising but powerful,” Marks stated. “It turns out that the ancient Greeks considered ‘nothing in excess’ to be one of the most important concepts about how to live. It was so significant that they chiseled it in large letters above the most important building in the most important city of early Greece, the Temple of Delphi—the birthplace of mankind and civilization. They thought it was important enough to write on the walls, and we agreed! So we named our company Delphia.”
Written in the stars
At GV, Marks had the time and bandwidth to begin looking for evidence of other vulnerabilities and characteristics that become key features of drug-resistant cancers.
This search led Marks down a rabbit hole in the scientific literature about some of the common outcomes when cancers become resistant to oncogenic inhibitors. This led him to some elegant work from Sellers, who happened to work across the street from GV’s Cambridge office. Marks called Sellers to invite him over to discuss a couple of ideas.
“[Bill] entered the room, leaned in, and asked, ‘Is this about over-activating oncogenes, not inhibiting them?” Marks recalled. “With a slight hush, I just smiled. He smiled as well and opened his laptop; it turns out his lab had been following the same trail of breadcrumbs.”
Sellers spent several years conducting fundamental cancer genetics research to predict which drugs to use in oncology. Using human genetics, Sellers discovered that, in addition to how oncogenes drive cancer biology, there are patterns that, when overactivated, are harmful to tumorigenesis—known as activation lethality.
“Some levels of activity are too high for the cancer to tolerate,” Marks explained. “It is written in the code of tumor genetics, which means that some of the same fundamental data that predicts success in oncology can be used and leveraged in our research.”
“Bill had clearly demonstrated that this is not just true in one or two outlier case studies, but that it is a widespread phenomenon observed for some of the most important and prevalent oncogenes. While we had been in similar orbits, we had never met. But from that first meeting we knew that we should team up on Delphia.”
That initial meeting at GV sparked a collaboration between Marks and Sellers, with Sellers bringing in Dillon, his longtime collaborator. It was a perfect fit. After all, the name Delphia Therapeutics evokes the image of a powerful ancient oracle.
Polar reversal
Delphia Therapeutics’ process begins with identifying the pathways that are vulnerable to overactivation. Any pathway in cells that is critical for their growth and survival, as is the case with almost any cancer pathway, has multiple ways for the cell to regulate it—it has regulators that can turn it on and off.
“That homeostatic control, the endogenous regulation, is supposed to be important, and normal cells have multiple means to do it,” said Marks. “They have multiple ways of turning the pathway up and down. The mutations that drive cancer that drive those pathways break and disable much of that regulation. The limited remaining regulation then becomes a vulnerability (or Achilles heel) in the cancer.”
Delphia’s platform employs a set of integrating techniques to identify pathways with genetic evidence of activation lethality. It then employs functional genomics and other tools to identify regulatory weaknesses. Then, whatever the best drugging strategy is for tweaking that node, they bring it into the lab and continue drug discovery.
Delphia plans to develop a genetic patient selection biomarker due to biology’s fundamental roots in human genetics, according to Marks.
“What we are doing scientifically is the inverse of these three decades of targeted therapy in precision medicine and oncology, but the benefits of that precision medicine playbook still apply to what we are doing—that is, prospectively identify the patients most likely to respond,” Marks explained.
“We will enroll them early in our clinical trials, so we will have the advantage of precision medicine oncology, which is the ability to obtain early human evidence of whether the medicine is working and use that to make decisions. So we have all the characteristics of precision medicine. We are simply switching the polarity of the medicine.”
The tools of the time
Marks said that as his team has built out the story, the response from leading cancer biologists has not been one of surprise. “They’re like, ‘That makes sense, but we had never really thought about or figured out how to turn it into medicines yet,’” said Marks.
Delphi’s approach is currently feasible due to two factors. The first is the ability to collect evidence of this phenomenon in tumor genetics, which is defined as the absence of something rather than the presence of a mutation or signal.
For example, when two different oncogenes can activate a given pathway, they are never found together because combining them results in too much activity. That is known scientifically as mutual exclusivity, and it can only be discovered when looking for the absence of something. With sufficient statistical power and data from sequenced patients, analyses can detect not only increased activity but also what is missing from the data set. Five or ten years ago, there was insufficient data to identify activation lethality candidates in human genetic data. However, in a 2023 Nature Genetics article, Sellers and his team discovered new pathway hyperactivation dependencies in subsets of APC-mutant colorectal cancers where further activation of the WNT pathway by APC knockdown or direct β-catenin overexpression led to robust antitumor effects in xenograft and patient-derived organoid models.
Tools are an important component of the second factor. Over the past 5–10 years, numerous step-function improvements have been made to the current methods for small-molecule discovery, such as interfering with a protein-protein interaction or identifying an allosteric pocket on a target. Delphia uses cutting-edge functional genomics tools to translate this insight into drug discovery.
Delphia has made rapid progress, according to Marks, due to the strength and clarity of the science. Despite the headwinds of 2023, Delphia had a successful year developing drug discovery programs that fueled their series A, which will propel each candidate through preclinical testing and into the clinic. Furthermore, Delphia’s biological thesis is based on fundamental biology, which allows it to avoid technological, commercial, and competitive risk.
“We are the first movers in a whole new field, pursuing some of the biggest oncogenes and thus the biggest patient populations in oncology,” Marks explained. “It is part of what has made it an appealing opportunity, in addition to the novel science, that as we progress, each step incrementally reduces our biology risk, which is the one area that we will hold as we move forward. We are following the playbook of the greatest hits in oncology drug development to bring this new approach to patients.”
