Where science meets the patient
Testing a potential new medicine in patients for the first time is a key moment in an intensive drug development process that typically takes more than a decade and costs over £1 billion.1
Does our laboratory testing translate into benefit for patients and are we pursuing a drug target that is relevant for human beings? Do we have the right drug, in the right dose, for the right patient? Does our compound produce side effects which could jeopardise its benefit in people with unmet medical need?
These are the questions that we ask in our Phase I and II clinical studies. The answers are essential for deciding whether a promising compound is good enough to progress to the large Phase III trials that inform final regulatory approval decisions and ultimately whether patients can begin accessing a new treatment.
Clinical trials are becoming ever more complex and expensive to perform, with growing risk of delay in getting breakthrough medicines to patients who need them. However, AstraZeneca is at the forefront of a revolution in the way that early phase clinical trials are carried out. We are making them smarter, faster and more cost-effective, so that patients can benefit from advances in medical science as quickly and safely as possible.
We are addressing the challenges of clinical trials through innovation in study design, biomarker-guided patient selection and state-of-the art data gathering and analysis. We are collaborating with clinicians, scientists and research centres that are internationally recognised for their expertise in conducting early clinical trials of new medicines. We are engaging with patients who participate in our trials as ‘co-scientists’ and we are gaining new knowledge from their direct input about their day-to-day experiences.
Together, we are working to translate AstraZeneca’s ground-breaking science into tomorrow’s life-changing medicines.
In the pharmaceutical industry, it typically takes about three years and costs approximately £390 million to progress a potential new medicine through Phase I and II clinical trials,1 and only one in five compounds is judged safe and effective enough to move on to Phase III studies.1 Growing complexity of studies, difficulties in patient recruitment, high volumes of paperwork for investigators and increasingly demanding data analysis are just some of the challenges that are prolonging timelines in pharmaceutical research and delaying the day a new medicine reaches patients.
To address these challenges, at AstraZeneca we are radically changing the way we perform our early clinical trials.
“We use adaptive clinical trial designs in all our patient studies. Adaptive designs allow us to get to the right dose quicker and get more patients on to that dose at an earlier stage. It’s good for patients because they are more likely to get an effective dose of drug and it’s good for us because we don’t waste time and money continuing to test doses that aren’t relevant,” explains Dr Tony Johnson, Vice President, Early Clinical Development within the IMED Biotech Unit at AstraZeneca.
“Adaptive designs allow us to change the design of a study in response to accumulating data. If an initial dose of a drug shows no activity in the first few patients or causes unwanted effects, we quickly change to another dose in the next patients tested. In traditional fixed trial design a dose is tested on a pre-agreed number of patients, even if it rapidly becomes clear that there is no benefit.”
Importantly, to gain the most value from our adaptive study designs, we have introduced many innovative study measurements, for example from novel imaging of the heart, lung, liver and tumours, as well as from blood, urine, sputum and tissue biopsies. Early dosing data can then be used in pharmacometric modelling to simulate what might happen with different doses, in different regimens and patient populations, and to plan further studies.
Adaptive designs allow us to get to the right dose quicker and get more patients on to that dose at an earlier stage.
“AstraZeneca is an industry leader in pharmacometric methodology which enables us in certain situations to avoid having to do multiple, separate studies. Pharmacometrics allows us to predict effects in particularly vulnerable populations, such as the elderly, children and those with kidney failure or liver disease. This means we put fewer patients at risk and we can get drugs to the market more quickly for the patients who need them,” says Dr Johnson.
Selecting the right patients for the right treatment has become more prevalent with the widespread introduction of biomarker testing in AstraZeneca Phase I and II clinical trials. Biomarkers are genetic, biochemical or molecular characteristics that can be used to identify patients with certain diseases and in some cases can indicate their likely response to targeted treatment.
“Integrating biomarkers into our clinical trials enables us to predict patients who are potentially more likely to respond to targeted treatment and also those who are potentially more likely to have toxicity problems,” explains Dr Shethah Morgan, Clinical Project Director at AstraZeneca. “Real time analytics in clinical trials enabling earlier science-led decisions and data-driven interpretation is a very valuable tool not just in our cancer studies but across all our therapy areas, including cardiovascular and metabolic disorders and respiratory diseases.”
AstraZeneca is innovating in the use of real-time data capture to enable more efficient and effective decision making. We are collaborating with leaders in information technology and artificial intelligence (AI) to fundamentally change the way clinical trials are performed.
iDecide is our cutting edge clinical informatics R&D framework that facilitates rapid interpretation of results, accelerates early identification of safety and efficacy signals, and improves understanding of the patient experience.
iDecide currently has four key components:
- REACT (REal Time Analytics for Clinical Trials) – which collates raw efficacy and tolerability data from patients taking part in Phase I and II trials and presents results as easy-to-interpret charts and tables for computer screens and tablets
- PROACT (Patient Reported Opinions about Clinical Tolerability) – which enables patients to send video/audio/text messages about how they are feeling and functioning while taking part in a trial including ‘real-life’ insights about how they are affected by adverse events
- WATCHER – which uses AI driven, continuous monitoring of incoming safety data from clinical trials to alert scientists to safety signals that need consideration and/or action
- FORESEE – which is in development as a comprehensive research tracking tool to keep AstraZeneca scientists continually up to date with the latest worldwide research related to the trials they are working on, including competitor data, and to predict progress timelines
REACT is currently being used in over 120 studies of investigational medicines in early and late stage development, and PROACT is increasingly being incorporated into Phase I studies.
The patient is the primary driver of everything we do.
“The patient is the primary driver of everything we do in iDecide,” explains iDecide Clinical Leader, Dr Dónal Landers. “REACT is designed to keep patients safe and support clinical trial adaptation for their benefit, while PROACT gives the patient a voice. It enables them to become an effective ‘co-scientist’ – giving us deeper, more meaningful insights into the effects of our drugs than has been possible with previous systems.”
WATCHER is already being used in some early phase studies to monitor specific safety signals, for example from liver function tests. It can recognise abnormal results in individual patients and, using pre-programmed algorithms, can search for other data that might explain the abnormality. It then alerts trial scientists to its findings so they can take appropriate action.
“WATCHER doesn’t just send us clinical notifications. It reasons over what it sees in the clinical data, using algorithms defined by clinicians. This can save us significant time in helping to decide what action to take and also provides continual clinical monitoring of the patient’s data, even when we are not using REACT,” says Dr Landers.
iDecide is the focus of a five year collaboration between AstraZeneca, Cancer Research UK’s Manchester Cancer Research Institute, The Christie, and the University of Manchester. We are uniting the world-class scientific and clinical expertise of these institutions with the unique contributions of participating patients to deliver personalised healthcare in the Experimental Cancer Medicine Centre at The Christie and in other AstraZeneca collaborating institutions for relevant clinical trials globally.
“The Manchester collaboration is a tremendous opportunity for industry and academia to work hand in hand with patients. It brings together the AstraZeneca team that created iDecide, clinicians at Europe’s largest cancer hospital who can road test it, university scientists who can build on it, and Cancer Research UK that can help roll it out across the other 18 cancer centres across the UK,” says Professor Andrew Hughes, Head of Experimental Cancer Medicine at the University of Manchester.
PROACT, the patient insights component of iDecide, is involving participants in some of our Phase I clinical trials in a unique way – enabling them to talk directly about life in a clinical trial, instead of through their doctors and nurses.
In pilot studies, patients are downloading the PROACT application onto smart phones and tablets and making real time video or audio recordings of how they are feeling. They can say what they want, when they want. Their words are anonymised under strict confidentiality guidelines and fed back directly to AstraZeneca scientists around key areas of interest, such as beneficial effects, medicine formulation, adverse events and emotions. Their insights about how a medicine affects them go far beyond what is typically collected through clinical consultations and questionnaires.
We are delighted with patient feedback about PROACT.
“Patients love it and the information we get back is much richer and more meaningful than what we receive with standard methods. When we published our results in Advances in Therapy, it was the most downloaded paper that month!” explains Dr Dónal Landers.
PROACT has already provided valuable patient insights that may impact on drug development. For example, a commonly reported rash associated with one investigational medicine which could have slowed development was rapidly shown to be tolerable and self-limiting – thanks to patient reports via PROACT. In another video, a patient reported that a medicine tasted unpleasant even though it was only in their mouth for a few seconds.
“We want to know information about taste because it is easily fixable when we are refining the formulation of a new medicine. It may be only one patient who has a problem in a small clinical trial, but it could be many more once the drug is on the market. If adding a flavouring means more patients can stay on treatment in the future, that’s really important to us,” says Dr Landers.
PROACT feedback can be quickly related to reports collected through REACT to achieve deeper understanding of what patients are experiencing. As Dr Landers concludes:
“If we can get this type of ‘soft signal’ earlier and earlier in our trials that wouldn’t be recorded by clinicians, it can have a real and lasting impact on the way we develop our medicines.”
REACT has transformed the daily collation and analysis of the thousands of raw data points collected in even the smallest Phase I clinical trial – the results of every exam, scan and lab test, the effects of every drug and dose taken, and their impact on patient health and well-being.
Instead of searching through multiple listings and Excel spreadsheets for patterns and trends indicating success or failure for an investigational medicine, drug development specialists, such as Dr Shethah Morgan, switch on their laptop and watch fully-processed data build into graphs and charts across their screen. With a few clicks on their keyboard, they can drill deep into the data for each patient, compare results across the study population and check for all-important indicators of safety and efficacy.
“While it used to take me a day to prepare data for discussion at a meeting, now it takes me an hour. REACT has taken the ‘manual labour’ out of data management and analysis, and enables us to focus on interpreting the findings. It makes time for science!” says Dr Morgan.
REACT represents a revolution for safety and efficacy analysis. Some technology moves forward in baby steps but REACT is progressing in leaps and bounds.
One of the powerful features of REACT is that it enables scientists to relate dosing, lab tests, adverse events, and other data on a single timeline. They can see the sequence of events and work out potential cause and effect. Did a patient’s headache happen before or after they took a medicine? Was it the paracetamol they took for their headache that affected their liver function test or the test medicine they took the same day?
“That sort of analysis used to take us months of work, and we’d only be able to do it once or twice in a study. REACT allows us to do it all the time – continually updating the relationships between events happening in a study,” explains Dr Landers.
Bringing all the trial data together as the study progresses, REACT informs key decision-making – not least the all-important ‘go or no-go’ commitment to progress an investigational medicine into the next phase of clinical trials.
Our lung cancer medicine, osimertinib,2 was one of our first investigational compounds to use REACT to guide decision-making during clinical trials. The unprecedented speed with which it moved from the start of clinical trials to approval in just 2½ years is ample testimony to the value of this important new tool for data integration and analysis.
Imagine a day when patients taking part in clinical trials don’t need to travel long distances to specialist centres for tests and treatment. A day when they can visit mobile scanners parked near their home. A day when they wear sensors that constantly monitor their vital signs and biochemical changes and transfer data electronically to trial centres.
For some trials, that day is already here.
Taking clinical trials closer to patients – in their homes and at local hospitals and health centres – is easier and more convenient for participants and better use of research resources.
Bedside biomarker testing, ‘wearable’ technology and remote monitoring have great potential for boosting patient recruitment to clinical trials of the new generation of personalised medicines. These treatments are targeted at highly specific genetic mutations and biochemical and molecular abnormalities, bringing a greater chance of successful treatment than the previous ‘one size fits all’ approach. However, this means that the pool of patients who meet the requirements of a trial protocol may be very small. With many patients currently failing protocol requirements and low recruitment rates across the pharmaceutical industry, the need for improvement is clear.
“Going to patients instead of getting them to come to us is better for them, and it means that we can boost recruitment and use fewer centres to perform our studies. Generating data more easily, coupled with advances in data generation technology, analysis and integration, can improve decision making and dramatically reduce the cost of our trials,” explains Dr Tony Johnson.
He suggests that such advances in early phase study design and management, patient recruitment and data handling could meaningfully impact the way that subsequent, much larger Phase III outcomes studies are carried out to support faster regulatory approval of new medicines.
“If applied to Phase III trials, these innovations may eventually have the potential to cut costs by as much as 90%, an observation also made by the USA FDA Commissioner, Dr Robert Califf, at the American Heart Association scientific meeting in November, 2016. Looking further ahead, the use of ‘wearables’ by patients will allow us to collect much larger amounts of data for analysis through integrated systems, such as iDecide. As a result, we might even reach a point where we have all the insights we need for marketing approval of a new drug, without the need for traditional, large Phase III trials,” suggests Dr Johnson.
Moving forward, AstraZeneca is taking an increasingly hands-on approach to its clinical trial programmes in early clinical development, instead of fully outsourcing to clinical research organisations (CROs) to manage studies. It was an important decision, and sets us apart from most other pharmaceutical companies, but we expect to see significant benefits from much closer relationships with key investigators and clinical research sites.
“We want to work directly with patients, clinicians and scientists at every important step of our new medicines development,” concludes Dr Johnson. “We believe this will help us better understand the challenges for patients, investigators and clinicians as well as the outcomes of our studies and their implications for future research by AstraZeneca, bringing important new medicines to patients who need them.”
1. Inside Innovation: The Medicines Development Process. Association of British Pharmaceutical Industries (ABPI)
2. Tagrisso (Osimertinib): Summary of Product Characteristics. EMA, 2016. (http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/004124/WC500202022.pdf)
Be among our employees who continue to make us an innovation-driven company that stands firmly among the world’s leaders in biopharmaceuticals.
Date of Preparation: January 2017
Date of Expiry: January 2018
Atlas ID: 1033118.011