WuXi AppTec Communications strives to find the latest cutting-edge technologies that can potentially offer new treatments for unmet medical diseases.

Glioblastoma, a cancer of the brain, ranks high on the intractable disease list. A brand new biotech company, Divide & Conquer has a new approach to fight this deadly condition. Leveraging Co-founder and Chief Scientific Officer Dr. Frank Winkler’s groundbreaking research published in Nature in November 2015 and again in September 2019, Divide & Conquer aims to open a new front in the war on cancer by disrupting the cell-to-cell communication mechanisms of solid tumor cells. The new research shows that the ability of cancer cells to form “social networks” makes them almost invincible, which is why current drugs fail to cure many patients.

The company’s initial focus is on glioblastoma, a lethal form of brain cancer, where cancer cells have been shown to communicate with each other via structures called tumor microtubes.

These structures allow the trafficking of materials and communication between cells, which appears to be essential for tumor growth and survival as well as resistance to chemotherapy and radiotherapy.

Co-Founder David Grainger, PhD commented, “There is mounting evidence, accumulated over decades, and now taken to the next level in Dr. Winkler’s papers in Nature, that solid tumors can leverage this network effect to evade all attempts to kill them. We now have compelling evidence that it can be disrupted, with the potential to render the most lethal tumor types curable.”

In this new article on novel glioblastoma research, WuXi AppTec Communications speaks with Dr. Winkler to discuss what makes glioblastoma so difficult to treat and why the UK-based company’s new approach may find an effective therapy for this therapy-resistant disease. 

Frank Winkler is a professor of experimental neuro-oncology and managing senior physician in the Department of Neurology at the University of Heidelberg, as well as the German Cancer Research Center (DKFZ) in Heidelberg, Germany.

Dr. Winkler earned his medical degree at Freiburg University. He completed a research fellowship at the Steele Lab at Harvard University in Boston, Massachusetts. Later, he received his assistant professor degree from Ludwig-Maximilian University, and became full professor (W3) in Heidelberg in 2012.

A member of the German Association of Neurologists, German Association for Clinical Neurophysiology, and the European Organization for Research and Treatment of Cancer, Dr. Winkler also serves as a reviewer for numerous journals, including Nature, Nature Medicine, Cancer Research,the Journal of the National Cancer Institute, Lancet Neurology, and Journal of Neurology.

His work has been published in Nature, Nature Medicine, Science, Neuro-Oncology, Clinical Cancer Research, and Neurology, among others.

Dr. Winkler is involved in several Phase 2 and 3 studies, and he is the recipient of several research grants. His lab in the German Cancer Research Center (DKFZ) has pioneered novel methods for intravital imaging of brain tumor progression, which was instrumental to discover the membrane tube-connected functional networks of malignant brain tumors (glioblastomas), and their crucial role for tumor progression and therapy resistance.

WuXi AppTec:What are the challenges involved in diagnosing and treating brain cancers? How important is early detection?

Frank Winkler: The greatest challenge in treating these cancers, particularly astrocytomas and glioblastomas, is their highly diffuse growth in our most delicate organ: the brain itself. These are basically whole-brain diseases from the time of diagnosis on. MRI examinations often show that, but even brain regions that appear disease-free are already colonized by tumor cells on the microscopic level.

To make things worse, cancer cells interconnect to normal brain cells, which even includes the formation of functional synaptic contacts with them. It is easy to see that effective killing of cancer cells in this situation, without damaging neurons and other brain cells, is a formidable task. It also explains why early detection is not actually important with these particular types of cancer: unfortunately, by the time the tumor is visible on MRI, diffuse brain colonization has already happened.

WuXi AppTec: Why has it been so difficult to find druggable targets for glioblastoma?

Frank Winkler: From my point of view, there are four key reasons:

First, we have to face a disease with a very high molecular and cellular heterogeneity, both on an intra- and intertumoral level. That means, when you look for druggable targets by studying specific molecular alterations in glioblastoma cells, it is very unlikely that you end up with one single drug that will benefit all glioblastoma patients. The ongoing umbrella trials like the N2M2 trial in Heidelberg address this very problem by searching for druggable molecular alterations that are present in a few percent of glioblastoma patients, and only these patients will receive the drug.

Second, the blood-brain barrier is preventing sufficient concentrations of many drugs and, even more so, therapeutic antibodies at the tumor cell site.

Third, a subset of tumor cells is displaying high levels of resistance to all available therapeutics, and we are just at the beginning of starting to understand their features. We believe that the tumor cell networks we are targeting at Divide & Conquer are a key factor in this respect.

Last but not least, we might be most successful in the end when we focus on basic cellular mechanisms of progression and resistance, and the microenvironmental interactions of glioblastoma. Again, that’s exactly what we do at Divide & Conquer.

WuXi AppTec: Has genomic analysis of glioblastoma improved drug discovery? If so how?

Frank Winkler: There are different examples of genomic analysis of incurable glioma types that have led to drug discovery. The most notable is the IDH1/2 mutation in astrocytic gliomas, where inhibitors have been developed and are being clinically tested. Additionally, there is the K27M mutation, which makes up a new class of malignant incurable brain tumors in which histone deacetylase (HDAC) inhibitors and other strategies are also being tested.

However, for the large proportion of glioblastomas, no such driver mutations have been identified so far. Thus, drugs developed for other tumors entities have been tried with glioblastomas too without any success thus far in controlled clinical trials, which includes immunotherapies.

That brings us to the point that we need to do now, from my point of view: understand the crucial cellular and molecular factors of tumor progression and resistance, without looking too much into what is already known from other cancers, and develop completely new treatment strategies.

WuXi AppTec: How did you choose to focus on glioblastoma at Divide & Conquer?

Frank Winkler: The “why” is very clear: because of exactly the reasons stated above. To be more specific about the “how,” we discovered that glioblastoma cells extend ultra-long, tubular cell membrane protrusions that are used to invade the brain in a scanning mode to colonize it and to interconnect to a highly functional, communicating multicellular network that protects tumor cells from dying after radiotherapy and chemotherapy. This is apparently achieved by an improved homeostasis in these networks that supports cellular resilience. To make things worse, the network is able to detect damage to itself, and to ultimately repair itself in a highly coordinated fashion, which is apparently also responsible for rapid local recurrences after surgical resection.

By studying all this, the scales fell from our eyes. If you want to make a real difference in this disease, you need to disconnect the tumor cells. Isolate them from each other, and thereby transform an incurable disease into a potentially curable one. There is actually an example from man: there is a closely related brain tumor type, called oligodendroglioma, in which cells lack the ability to form resistant networks because of a particular genetic alteration (1p/19q codeletion). Happily, this glioma type is not incurable: most patients experience relapse-free survival over decades when treated with radiochemotherapy.

WuXi AppTec: How did you develop your drug candidate?

Frank Winkler: We have developed an in vitro assay for tumor cell connectivity that nicely reflects the anatomical and functional networks we see in patients and in vivo in mice. We are using this system to screen for drugs that inhibit the membrane tube extensions that connect tumor cells.

By doing this, we verified some “typical suspects” interfering with actin and microtubule dynamics that, however, are not druggable. Most importantly, we discovered that the PKC pathway is a master regulator of tumor cell network connectivity, and we have identified lead compounds that target this pathway, and–importantly–can be given to mice with acceptable toxicity too. At Divide & Conquer we are testing all lead compounds in refined mouse models that allow us to study both tumor cell disconnection on a microscopic level, and anti-tumor effects at the same time, using long-term in vivo microscopy combined with MRI and survival studies. That ensures that we really hit the target we want.

WuXi AppTec: What is the mechanism of action?

Frank Winkler: Next to the PKC pathway, we are looking into other drug classes that target master regulators of cellular connectivity. Among them are ones that are used during (neuro) development to connect cells with each other to form functional syncytia, but are later downregulated and should no longer be critically relevant for the adult organism.

In general, we know that tumors recapitulate many molecular and cellular processes of normal development. Thus, at Divide & Conquer, we believe that the establishment of a functional multicellular network can be targeted in a tumor cell-specific way, at least in adults, without unacceptable toxicity to normal tissues. Whether this works out in the end, we will see. It remains a completely new therapeutic concept, and thus therapeutic ratios and windows need to be explored.

WuXi AppTec: What regulatory challenges do you face in clinical development?

Frank Winkler: Since no working therapies other than surgery, radiation and chemotherapy with alkylating agents exist for the treatment of glioblastoma, there is a great need for novel approaches. In other words: the standard of care is not great. For the large (about 70%) subgroup of patients without MGMT promotor hypermethylation, chemotherapy is basically ineffective, too. That really fosters clinical development, which should ideally be in the primary setting, with the new drug combined with radiotherapy, with or without chemotherapy, depending on the MGMT promotor methylation status. The trial concepts are clear and have worked out very nicely so far in large Phase 3 trials. Only the effective drug is missing.

WuXi AppTec: Have you worked with patients in developing your drug development strategy? If so how?

Frank Winkler: I work with patients every day and know that the vast majority of them share our view that we need to make true progress in this disease, and that we need to take our thinking in really new directions. As a matter of fact, many of them actively ask for all kind of unproven and often even harmful additional therapies, which from my point of view mostly reflects the lack of therapies with satisfying effectiveness today. The novel concept of disconnecting tumor cell networks is very plausible to many, and I do not foresee any general acceptance problems.

WuXi AppTec: What lessons have you learned during the drug development process?

Frank Winkler:  As a scientist and clinician, you need to pair with strong partners to get things going. We tried to work on drug screenings ourselves for years, without great success. With Divide & Conquer, we have founded a company with an environment where all the expertise and experience are there to make fast progress.

WuXi Apptec: How soon will we have an effective treatment for glioblastoma? Will effective treatments require combinations of drugs?

Frank Winkler: If everything goes as planned, we can easily have a drug candidate that can be put forward to a clinical trial in one year from now. Combinations are possible, but not necessarily. More drugs generally mean more toxicity, and often complex interactions. Particularly for early preclinical and clinical development of a completely novel antitumor strategy that can be tricky. However drug screenings continue, and if we discover a truly synergistic mechanism of action of drug candidates, this might be a way to go.

WuXi Apptec: Can you comment on the specific progress or lack of progress in treating this disease in the last ten or 20 years?

Frank Winkler: The only true progress in the last 20 years was the addition of the alkylating chemotherapy temozolomide to the standard of care – even though it is relevantly active only in a smaller subgroup of patients. Moreover, a physical therapy with alternating electric fields (TTF) has been introduced as well, but is only available in some countries, and costs and compliance are issues. As I mentioned before, the lack of progress is due to disease-inherent and brain-inherent factors.

WuXi AppTec: What would be the one thing that has the most potential to lead a paradigm shift from treatment to cure for cancer patients?

Frank Winkler: Target general principles of tumor cell resistance, which are most likely of a basic cellular nature, and/or target tumor growth modulation by the nonmalignant organ microenvironment which is less heterogeneous and can less effectively escape drug actions compared to cancer cells. One example for this strategy is the disconnection of glioblastoma cells we are currently working on at Divide & Conquer. Outside glioblastoma, the success story of immunotherapies is another.

The current “mainstream” drug development strategy is different: target singular molecular alterations of cancer cells. This will only prove effective over the long-term when a) those are present in every cancer cell (which frequently is not the case), b) the tumor has no opportunity for escaping the drug action (which is also rare), and c) a targetable molecular alteration exists at all. Here, the hurdles appear higher.