Computer simulations offer great promise in drug development
Global experts from industry and academia explored the emerging science of quantitative systems pharmacology – mathematically designed models and simulations that show how medicines work on various systems of the body – during a recent Roche symposium held at Roche Innovation Center New York.
From predicting the best aircraft design to forecasting the weather, interactive computer models have long been used in numerous applications.
Now, the quantitative approach is emerging as a valuable tool for answering key questions in drug discovery and development. Its potential was explored during a two-day Roche symposium, "Increasing Drug Development Success Through Quantitative Systems Pharmacology," held in early November at the Roche Innovation Center New York. The unique forum brought together diverse academic and industry leaders from both the experimental and computational sides and generated significant exchanges among the 100 select attendees. "Systems pharmacology has tremendous potential to dramatically improve success rates in drug development," said Richard Peck, Roche Global Head of Clinical Pharmacology, who shared the following insights about quantitative systems pharmacology (QSP) and the symposium.
What is Quantitative Systems Pharmacology?
The body is a dynamic, constantly changing system, not a single pathway. We need to understand the whole system in order to discover and develop the most effective medicines. At its most basic, QSP takes empirical, clinical and other data from numerous sources and applies mathematics and biostatistics techniques to map human biological networks and simulate how drugs actually work in the body. For the first time, we are able to "see" how our medicines impact the biological processes in diseases, in the immune system and in the interactions between them.
Why is this valuable?
We have made unprecedented progress in understanding disease biology and the immune system. For example, we now know the same diagnosis in two people may stem from completely different genetic causes. We also know that two individuals with the same disease may have different responses to the same treatment. But we still need to address critical unknowns in order to improve medical outcomes. Why are there such varied responses to identical treatment? What triggers an antibiotic to stop working or a tumor to stop responding to treatment? What in the immune-disease-drug interactions leads to drug resistance or treatment failure over time? Quantitative systems pharmacology not only can illustrate the biological differences in disease progression and immune response in people, but, more importantly, can help us understand which specific differences in this complex network really matter. This insight can guide many drug development decisions.
Let’s look at combinations in cancer immunotherapy (CIT), which uses the body’s immune system to fight tumor cells. CIT is believed to transform cancer treatment and we are seeing very promising treatment results. But we are still faced with drug resistance and ultimate treatment failure. As we learn more about immune response, it is becoming increasingly apparent that drug combinations will likely play an important role in targeted cancer therapy. Benjamin Ribba of Roche, who shared his work on modeling the cancer immunotherapy cycle, highlighted current challenges in developing safe and effective CIT combination therapies. He pointed out that exploring a number of drugs in a number of cancer indications could lead to hundreds, if not thousands, of potential combinations, far too many to actually test. The computer-based platform he is developing for CIT offers great potential for helping select which combinations to test so we hopefuly have the right combination therapy for the right patients.
What were other areas of discussion at the symposium?
A recurring theme was the immune system and the role it plays in drug response and resistance. We also had several experts in the areas of cancer and infectious diseases describe their use of modeling to show that changing the timing, duration or dose of standard treatment can dramatically decrease drug resistance or improve response rates. There were many examples of models predicting behavior that we previously didn’t know. It is very exciting, and only the beginning.
What do you mean by ‘only the beginning’?
Quantitative systems pharmacology is just emerging as a discipline in the pharmaceutical industry. We are at the very early stages of seeing where biological thinking can inform computational modelling and computational modelling can inform biological thinking. Each needs the other to be most effective. We purposely designed the forum so speakers with biological expertise alternated with computational experts because we know it will take a network of scientists to understand the networks in our bodies. We wanted to ignite those connections, and we were delighted to see extensive cross-pollination not only across disciplines, but also across therapeutic areas, with virology researchers getting new ideas from work being done in oncology and vice verse.
Cancer immunotherapy simulation shows systems pharmacology promise
Cancer immunotherapy (CIT), which harnesses the body’s immune system to fight tumor cells, is ushering in a new era of cancer treatment. It is believed that cancer grows and spreads when one or more steps in what is known as the "cancer immunity cycle" do not work properly. The aim of CIT is to find drugs or combinations of drugs that correct failures and reestablish the loop so the immune system can do its job against the cancer cells. Systems pharmacology holds promise for identifying the cause of disruptions in the cycle and how best to address them.
Benjamin Ribba of Roche Clinical Pharmacology is developing a cancer immunity cycle model to simulate and integrate the steps of the immune process. In initial work with the groundbreaking model, he asked it to identify what would lead to the greatest expansion of T-cells to fight cancer cells within a specific kind of tumor. Contrary to what he expected, the model indicated that to increase T-cell infiltration, it is more important to improve steps earlier rather than later in the cancer immunity cycle. While the model will change depending on the cancer indication, it is designed to highlight priorities for expanding T-cells in the tumor microenvironment, potentially driving choices about drug development.
"Everything is connected," he added. "When we develop network models, when we learn more about the system as a whole, we can better understand and navigate through its complexity to help support clinical decisions."