In a recent series of articles, I explored what Real-World Evidence (RWE) is, and why it is needed. In the final part of the series, I gave a worked example of what is possible and hinted that (rather obviously) different styles of study (and different questions) fit at different points in the drug development cycle. This article explores this in more detail, hopefully showing that RWE is not just something to think about after your drug has launched but can support and enhance the process throughout the life cycle.
Firstly, there is a slight shift in thinking required. RWE, when used to its full potential, is not a discreet project, with specific questions and answers, but rather a ‘living’ dataset that should be added to over time, and allows you to answer your questions as they arise. On a conceptual level, for a single drug, this starts during phase II (when you are starting to get an inkling this drug has legs), however considering most pharmaceutical companies will have multiple assets in the same disease area, this should become a main stay of their wider R&D budget, keeping the data set updated through time.
At this point, these questions tend to be largely epidemiological in nature (patient characterisation, pathway analysis, disease and treatment progression etc.). Depending on the disease area, most of this can be answered with a reasonably simple, Electronic Health Record (EHR) retrospective dataset. However, given the on-going nature of these questions, it might be worth considering investing a long-term, prospective data set. More complicated disease areas may also require the integration of multiple Secondary or Specialist care datasets as well, allow cause and effect to be considered across different healthcare providers. All this information can then be synthesised together to help improve patient recruitment for your Phase III trials, reducing operational time, and hopefully, ultimately, saving money.
In practical terms, by this point, the dataset should give a reasonably accurate overview of the disease area as a whole, what the patients look like and how they interact with the healthcare system, regardless of the patient’s current treatment.
Once into Phase III, you can start adding your treatment specifically into the dataset. Most simply, this can just be a case of adding your clinical trial (RCT) data into the dataset. More broadly, you could look at extracting the patient’s EHR as part of their patient history, to further enrich the dataset, and added outcome parameters after the study (follow-ups).
Another key question you can start to consider at this point is your value proposition. On a financial level, what is the impact of your treatment has on the healthcare eco-system as a whole (Does your treatment reduce A&E attendances or prevent/reduce the need for rescue medication etc.)? All of these effects can be quantified for both your on-going clinical trial, but also tallied up in a financial sense for your value proposition.
Note: Linking data sources
This might be a good point to make a quick observation about linking datasets. In the UK, every patient has a unique NHS number (or CHI number in Scotland). By hashing and salting (sort of like a one-way encryption), from this number you can create a unique ID for every patient in the UK which can be used as a linking ID regardless of the source of a data set (because anywhere in the NHS, the patient will be referred to by their NHS number). This is detached from their NHS number (once created), unique to your dataset, and doesn’t allow you, or anyone else, get back to that patient’s NHS number (i.e. makes them anonymous). What this means in reality is that a patient might be added to the dataset as part of a retrospective analysis during Phase II, have their details updated with the clinical trial data during phase III, then consent to being part of a prospective safety study after launch for Phase IV (more information on this below).
Of course, the more data that is added to the dataset, the more refined your ongoing epidemiology questions can become. This intern helps to refine your study design and patient recruitment strategies for later studies creating a ‘snowball’ effect of added value and savings.
Once your drug has launched, it is customary to conduct a Post-Authorisation Safety Study, to track the long-term safety of the treatment in a larger population than is realistically possible during Phase III. This is a huge opportunity to also collect efficacy data both for pricing negotiations and marketing purposes.
Imagine a ‘light touch’ study, where patients are seen in a baseline clinic (to take key measurements and be consented), given your treatment, and left to experience the healthcare eco-system as they would normally. They are then invited back, 12 months later, for their outcome visit, again taking some key measurements for comparison. During the trial period, you then take feeds from the patients primary, secondary (and if applicable tertiary) care records, to capture all interactions with the Healthcare ecosystem.
Another tangential opportunity sits within Innovative Pricing and Patient Access Schemes (bet you were wondering when I would put both topics in one article). One of the biggest challenges faced by new treatments (particularly launching into a crowded marketplace such as Asthma or Diabetes), is to gain awareness, and/or market share. Both of the above options are a good way of helping to push your treatment out into the consciousness of prescribers, but it also offers an opportunity to collect more detail. One example might be to run an outcome based scheme (‘we will rebate the value of the treatment if it doesn’t work’). This helps to promote the treatment to prescribers while continuing to paint a positive picture about the pharma company caring about outcomes. However, what it also does is provide an excellent environment within which to collect outcomes data.
Secondary to the variables of interest for the pricing scheme (the measure of outcome for the contract), the infrastructure required to run these types of contracts, can be utilised to collect additional information that might be of interest. It is easy to forget that the questions your brand managers and sales teams have are (give or take) the exact same questions asked by the Medicines Management Teams within the NHS. How many patients are on this treatment? Is it effective? Are there any more patients who would benefit from this treatment? Are there any patients on this treatment that probably shouldn’t be? And so on and so forth. All of these can potentially be answered through the same infrastructure system, on an on-going and rolling basis (for both parties).
Of course, the Nth iteration of this line of thinking are Beyond the Pill type initiatives. This, in general terms, is when a pharmaceutical company funds additional services that equate to an added value delta on top of their treatment. This works particularly well for treatments at the end of their life cycle and are coming under threat from generics and biosimilars.
Why should I spend more on this branded treatment, when I can pay 30% less and get an equivalent efficacy of the drug?
Because for that slightly higher price, the pharma company will also help improve patient diagnosis, support patient management, and pay for additional training for your staff.
Of course, this also generates data. Every time you see a patient, with permission you can record more information about them, feed it into their medical records. Firstly this helps the physician, within reason, more information is always better. However, it can also be recaptured as part of a prospective or retrospective study, for your use. It also helps raise the bar on the standard of care generally, which in turn improves patient outcomes, and increases (in measurable terms) the efficacy of the treatments.
And so the cycle continues. All of the data, knowledge and insight gathered from the early phase II epidemiology datasets and studies, can now be leveraged for the next great leap forward in the disease area. Helping to bring about a new class of treatments that can truly revolutionise medicine.
Hopefully, this article has gone some way to help convince you that it might be worth looking at RWE a little earlier in the drug life cycle. If you have any questions or would like to look at some of these integrated options for RWE or Innovative Pricing schemes, send me an email at firstname.lastname@example.org or visit our website for more information