The annual European DIA conference took place in Vienna this year. After the opening speech (which involved a flash mob of attendees doing a waltz across the main conference hall!), the conference kicked off with the first key note session of the week. Dave de Bronkart, aka ‘ePatient Dave’ was given 24 weeks to live upon a diagnosis of cancer. Dave described how he became one of the first ‘ePatients’, a term used for patients who use the internet to treat, and sometimes cure, their own conditions.
Things have changed since the times where medical professionals implored their patients not to use the internet to self-diagnose and treat. Dave presented a strong case for the need to engage patients and to open the internet as a tool to be used for patient empowerment. This shift will impact the industry greatly. When we set up clinical trials, we need to consider that patients are now a lot closer to their data than they used to be. When designing data capture systems and devices, we now need to ensure these will be user-friendly and will encourage patient engagement more than ever before. The FDA has even suggested that the degree of subject involvement in a study may be used as a quality indicator.
The FDA defines an adaptive design as
…an adaptive design is defined as a clinical trial design that allows for prospectively planned modifications to one or more aspects of the design based on accumulating data from subjects in the trial.
The most common adaptations of the trial design which impact on programming are:
- Change in the treatment groups
- Change in the randomisation ratio
- Focusing recruitment on patients most likely to benefit
PHASTAR’s extensive experience with CDISC standards, when combined with clear dataset specifications, can be leveraged to facilitate the changes which may be required due to Adaptive Trial Design. This could include additional data being collected and reported, or further treatment groups or regimens being added to the trial design. With standard scalable dataset specifications, it follows that forethought applied to output design and programming will aid the accommodation of changes to the data and/or trial design.
A key objective of Phase I dose escalation trials is to establish a recommended dose and/or schedule for an experimental drug or combination for further testing in Phase II. They balance avoiding unnecessary exposure of patients to subtherapeutic doses with maintaining safety and rapid enrolment.
Traditional “3+3” designs and their variations are simple to implement, based upon pre-specified rules to guide how to react to accumulating data but lack firm statistical foundation. These types of designs only “memorise” the toxicities in the previous step of dose-escalation and do not take into consideration cumulative toxicities. As a result, a large proportion of patients may receive low, non-effective doses and the estimated maximum tolerated dose (MTD), the highest dose at which a pre-specified proportion of patients experience a dose-limiting toxicity (DLT), may be inaccurate. To overcome the uncertainty about a recommended dose for future studies and slow dose escalation, model-based Bayesian adaptive designs have been developed:
- The target toxicity at the recommended dose for future studies would be explicitly defined by seeking a suitable quantile of the dose-response curve, such as the probability of toxicity at the target dose being 20%;
- The prior guess of the target dose on the pre-assumed dose-response curve before the trial starts would form the starting dose;
- The decision criteria, such as treating future patients at the current estimated MTD or minimising the variance of the dose-response model parameters, would then be used to identify the next dose after the model is updated to take consideration of all available data from the trial.
- A set of safety constraints can be put in place to control the risk of overdosing the patients, for example pre-specifying a maximum dose increase.
A day in the life of a statistician at PHASTAR
As statisticians, we love a bit of variation! A day in the life of a statistician at PHASTAR is no exception - depending on the stage of assigned projects and the nature of the studies we’re involved with, we could be providing input to the design of a study, calculating sample sizes, writing statistical analysis plans (SAPs), writing ADaM specifications, programming datasets, TLFs or efficacy analyses, or, providing expert statistical advice to our clients.
At a more senior level, statisticians at PHASTAR may also have line management responsibilities and contribute to management initiatives. We may be involved with the project management of our studies, dealing with timelines, resource projections and interacting with clients.
The following provides a glimpse into my life as a statistician working at PHASTAR, in the form of a diary describing a typical day.