Shiny Toys and Trendy Technology

In an age of shiny toys and trendy technology, we know what criteria makes a mobile app a medical device (or software as a medical device). We have regulator guidance on how to apply these criteria to Artificial Intelligence/Machine Learning-based software. We’ve even updated the clinical trials ‘bible’ (ICH E6-R2 also known as ICH GCP) to accommodate the changes in technology and research techniques. So why, 19 years after the European clinical trials directive (EU-CTD = Directive 2001/20/EC) was published, do we still not have clarity or consensus on something as basic as what criteria makes a study a non-interventional study, especially since there is increasing regulator and HTA demand for real world research?

“Oh, but we do”, I hear you cry… “It’s defined in the regulations and they’ve even given us a decision tree to help us work it out”

Hmmm…me thinks it’s not that easy. The pivotal and ‘undecided’ criteria that can push a non-interventional study to a clinical trial is the same in both the currently active Clinical Trials Directive (EU-CTD = 2001/20/EC) and the (soon to be active) Clinical Trials Regulation (EU-CTR = EU/536/2014), namely ‘application of diagnostic or monitoring procedures’.

• EU-CTD decision tree asks an inverse question: “Will no diagnostic or monitoring procedures be applied to the patients included in the study, other than those which are applied in the course of current practice?”
• EU-CTR decision tree asks: “Will diagnostic or monitoring procedures be applied to the patients included in the study, other than those which are applied in the course of current practice related to the condition?”

So, classification of non-interventional studies (In Europe) is dependent on interpretation (by Researchers, RECs and Regulators) of “additional diagnostic or monitoring procedures.”

Some RECs and Regulators interpret this to include ANY additional procedures e.g., meaning even simple research procedures such as study-specific biosamples are not allowed in non-interventional studies.

Surely, in the context of a clinical trial, “diagnostic or monitoring procedures” are those procedures that are used to inform the clinical management of the patient e.g., ECG, pulse oximeter, lab tests, blood pressure monitoring etc. Therefore, in the context of a non-interventional study, the ‘non-intervention’ relates to (or should relate to) ‘not intervening in the clinical management of the patient’. Meaning? Study-specific biosampling for research purposes only are NOT “additional diagnostic or monitoring procedures”.

OK. We’ve identified the problem, so what is the solution? My (humble) suggestions is that we learn from the evolution of the clinical trial bible (ICH GCP) and the medical device (software as a Medical Device) definitions and clarify that the afore mentioned “additional diagnostic or monitoring procedures”, in the context of non-interventional studies, relate to procedures that inform the clinical management of that patient. Simple!…and…back to the real (and less simple) world 😉

>>>Learn More…