'Tis the season when we all feel a bit more hopeful, and when our faith in miracles swells. Which probably explains the resurgence in the dream that many drug developers are clinging to: "if I can just get a LDT out there, this whole companion diagnostics thing will go away." In fairness to those that espouse this belief, FDA has allowed this in the distant past (pre-CoDx days), with Monogram/Selzentry being a favorite example. But we'd like to offer the opposite and more recent example of ChemGenex and OmaPro , where ODAC decided a test was required and ChemGenex ended up with a third line claim instead of a first line claim because they failed to handle the CoDx portion of the trial seriously.
Times have changed, and as faithful readers of the Myraqa blogs already know, there have been a number of high profile actions by the agency of late that are designed to make the life of LDT manufacturers less comfortable. And the FDA (both CDER and CDRH) is still very clear that no drug requiring a diagnostic for its safe and effective use shall slip into the market without the test going through the submission process, even if the drug has breakthrough status.
This leaves those working on drug development stuck with a difficult reality. On their right, they have executives and investors proclaiming that all new compounds shall have a biomarker. And to their left, there's the reality that CoDx development is costly and time consuming. The latter is further exacerbated by the fact that the approved test which they paid for might never be used, since LDTs for the same marker will (at least to this point) remain on the market. Hence the "Hail Mary" regulatory strategy of hoping the drug will have such astounding performance that FDA will approve it without a CoDx, even if it's for the good of the patients.
The difficult part of this discussion is that it is possible this Hail Mary strategy might work. FDA even addressed this situation in the draft CoDx guidance "FDA may decide to approve a therapeutic product even if its IVD companion diagnostic device is not yet approved or cleared when the therapeutic product is intended to treat a serious or life-threatening condition for which no satisfactory alternative treatment exists and the benefits from the use of the therapeutic product with an unapproved or uncleared IVD companion diagnostic device are so pronounced as to outweigh the risks from the lack of an approved or cleared IVD companion diagnostic device."
But as any good football coach knows, while you need the Hail Mary in your playbook, you shouldn't expect to win many games with it. It is the play you use when everything else has failed and you're running out of time. In a CoDx context, it is the equivalent of betting your drug development program on a strategy of brinksmanship with the FDA.
This might be a good time for everyone to collectively take a step back and re-evaluate their regulatory strategy around CoDx. First, if you've got a selected clinical trial design (i.e. only marker positive patients are included), then you've incorporated your regulatory strategy into the trial design, because you only know the performance in the selected population. So pick a CoDx partner and move forward.
For all those other cases, you're going to need to do some old fashioned risk analysis. You'll need to evaluate whether treatment decisions could be made on the test (CoDx!), how insurers feel about reimbursing a test with vague clinical utility (CoDx?), and of course the ever-ambiguous LDT regulatory landscape to decide whether or not to pursue a companion diagnostic strategy.
So in summary, there's really no substitute for a good regulatory plan. This typically involves making sure you have done everything to ensure a FDA approvable test is available. Should your CoDx partner come up short at the end, FDA can probably work with you to get the drug on the market with the worst case being the CoDx lags slightly behind. Just remember that if your current plan rests solely on the hope that your drug will be just too good for FDA to delay, well you might want to refine the strategy a bit more.