As a dedicated IRT provider for more than a decade, Veracity Logic has deployed many systems for subject randomization, subject management, and management of study drug for all phases of clinical trials, including numerous Phase 1,1b, and 2 exploratory trials.
But a lingering conundrum in the industry with regard to these smaller studies is this: At what point does it stop making sense, in terms of per patient cost and savings in manual effort, to use an IRT system at all?
We help clients sort out the pros and cons of this question before irreversible commitments are made one way or another. The key factors for Sponsors to consider when deciding whether or not to use an IRT, while obvious in one sense, can get a bit muddy in others. Grids like the one below can help sort things out.
|Study is randomized||+1||-1|
|Study uses > 5 sites||+1||-1|
|Study has > 3 visits||+1||-1|
|Study needs Returns tracked at content level||+1||-1|
|Study has > 20 subjects||+1||-1|
And so on.
Based on the above grid, a perfect YES (a score of 6) to the question of whether it makes sense to use an IRT for a small study would be the following: A study that must be randomized, involves more than 5 sites, has more than 3 study visits in the protocol, needs to track drug returns, has complex dosing options, and has 40-50 subjects planned. The wisdom of using an IRT would be true even if the cost per subject, due to the relatively small number of subjects, might make people flinch in a way they would not have to do for a 300-subject study.
On the other hand, a study that is open label, involves less than 5 sites, has only 2 treatment visits, involves an IV infusion, and has only 18 subjects planned, approaches zero reasons to use an IRT. Veracity Logic has advised clients to proceed with manual processes in these types of situations.
But what of studies that mix it up a bit. For example, what about a study that is open label, non–randomized, will be conducted at only two sites, but requires 10 weekly treatments per subject, multiple escalating dose levels in capsule form (returns to be recorded), and a total of 36 subjects in three cohorts? The score here is 2 No’s and 4 Yes’s. The per-subject cost may very well be outweighed by the gains in efficiencies (and confidence) that using an IRT provides.
Of course, for randomized studies a randomization process/package of some sort must always be applied. To maintain a high confidence in the quality of study data (when independence and blinding requirements are met), external management is always a smart call, whether an IRT is the software of choice or not.
In addition to the grid shown above, intangibles of service, quality, experience, and reputation offered by IRT vendors is often that extra factor that tips the scale.
Share this Post