Last month, Illumina announced the FDA approval of the TruSight Oncology 500 (TSO 500) test. This was the completion of a long voyage for the test, and one that frankly I had my reservations about pursuing when I was there. I wasn’t convinced that a distributed oncology panel test kit was a great strategy, and at one time suggested this be done in our CLIA lab, since we could report both IVD and clinical research content on separate reports for the same sample under the single site model. But pharma partners strongly preferred the distributed kit due to the relative ease of marketing the test outside the US, so the journey began.
Quick disclaimer before going further. The information presented in this blog is based on public information and my personal experiences in working with the FDA for over 15 years on oncology panels and companion diagnostic (CDx) tests. And while I did work at Illumina as their Head of Regulatory, that was seven years ago - a lifetime in this business. So any thoughts in this blog about the approval are based on my experience, and frankly pure speculation.
Alright, back to the blog.
My concerns in the past (and even today) were due to a strange enforcement loophole that existed prior to the FDA’s publication of the final rule on LDTs earlier this year. The FDA has always required that CDx tests, which are tied to a drug submission, needed to be FDA approved contemporaneously with the drug. That CDx content would need to be reported on an IVD report by the lab.
But the power of DNA sequencing has always been that one could gather additional information from that same sample that might be useful to the treating physician. And under CLIA that content could be reported under a separate lab report through what the agency commonly referred to at the time as “professional services”.
Contrast that model with a distributed test, where the IVD manufacturer (Illumina, Thermo Fisher, etc) would have to validate any novel variants reported with at least one clinical sample if they wanted to include it in the IVD report. I’ve run into this personally on a couple of client projects when novel content is being considered for the test report.
For a test like the TSO 500, that’s a lot of novel variants, which means you’d be masking useful content if you didn’t have clinical sample data. And in some cases, these variants are extremely rare, making the samples very valuable. That caveat puts kit manufacturers at a disadvantage versus a lab-based IVD test like Foundation One, since the lab developed test (LDT) only needed to get FDA approval on the CDx content. Therefore, the resulting project for the IVD kit manufacturer would be longer timelines and (likely) added cost to develop. With me so far?
Okay, so did my concerns pan out? Well, if we look at the public disclosures from Illumina, they got breakthrough designation from the FDA in Jan 2019 for the TSO 500. The designation was likely due to the CDx content moving through in parallel with their partner’s therapeutic at the FDA (CDx tests very often get breakthrough). Later that year Illumina announced that the first module went into the agency (Q2 2019). And five years later, they got their approval. That’s a pretty long timeline for a CDx with breakthrough designation.
In fairness to Illumina, part of that timeline was likely due to the COVID-19 pandemic, which all but crippled the FDA for a couple of years. It’s also possible that the other contributor was from Illumina having a lot of clinically relevant content, and thus wanting claims on more than just the small CDx portion. Remember the whole “Thou shalt validate all novel content with clinical samples” commandment from the FDA discussed above? We’ll have to wait to see the SSED for the TSO 500 when it comes out to know for sure (future blog!), but I’m guessing that getting at least some of those other variants approved contributed to the timeline.
Okay, so other than me getting to gloat (slightly) about the perils of IVD panel kit strategies, what exactly is my point? Well remember that earlier this year the FDA final rule on LDTs was released. What this Illumina approval means for labs (assuming the courts don’t toss the final rule out) is that LDTs will need to go through a similar process in order to get a panel test through the agency that has a lot of clinically relevant, non-CDx content. Because the CLIA loophole for separate reporting of clinically significant variants has (more than likely) been closed by the final rule.
This is significant, and something that will need to be carefully considered by both regulators and labs moving forward. Professional services reporting has been an important part of oncology panel testing by CLIA labs, even those with IVD tests. As a result, regulatory planning for future tests developed under the final rule will need to account for the change in strategy.
There are some options. First, there is a regulatory pathway through the agency that has been used before, first by Memorial Sloan Kettering, and later by companies like PGDx and ACT Genomics. Using that strategy, the clinically relevant content gets cleared for clinical trial use, with the goal of later converting it to CDx content in support of a pharma application. However, you’re talking about two different tests at that point, and that whole validating novel content with clinical samples thing still exists.
The second option is the NY State pathway that the FDA has outlined in the final rule. Oncology panel tests could get a pass from the full market authorization pathway for clinical research and trial content. But I’m guessing the agency will still want to see a PMA for the CDx variants, which might again mean two different tests.
The good news for manufacturers like Illumina is that the playing field leveled out a bit. And it’s likely that more consistent reporting of CDx content will come out of the LDT final rule. However, there’s a big question about how the agency (and labs) will handle all that other non-CDx content moving forward. Hopefully a solution can be found where patients don’t get caught in the middle.
So put another reef in the mainsail and stow everything below decks just in case. There are some rough looking seas ahead.