FTC Suing to Block Illumina’s Acquisition of GRAIL

Back when Illumina first announced they were going to acquire GRAIL, the general feeling in the M&A field was that this wouldn’t trigger any antitrust issues. I was asked if I could think of any difficulties. I didn’t claim back then that I was an antitrust expert (and still don’t), but I did wonder if the FTC would be concerned about all of the other liquid biopsy players being at a disadvantage since they’re all using Illumina’s sequencers. More idle curiosity on my part rather than an actual prediction. It turns out the FTC is a bit concerned. Concerned enough that they announced today that they’re suing to block the deal (WSJ, GenomeWeb, FTC complaint PDF). Unlike the Illumina-PacBio merger, I don’t really have a strong opinion about this one, but there are several interesting points.

1. While the FTC’s concerns aren’t completely surprising, this announcement came out of the blue. As far as I can tell they didn’t telegraph their intentions or even mention that they were going to look into it. This is in contrast to the proposed PacBio acquisition where they went through two rounds of reporting and analysis (with public commenting periods) before declaring they would sue to prevent the merger. This time they brought the hammer down right from the start - NO MERGER FOR YOU!
2. This time around the FTC is worried about “vertical integration” (while the PacBio acquisition concerns were about “horizontal integration”). Apparently it’s quite rare as there’s only been one other case in the last 40 years (the AT&T - Time Warner merger).
3. Again, no strong opinion on whether this should be blocked or not, but It seems pretty reasonable for Illumina’s liquid biopsy customers to be nervous. Trying to compete against someone who controls your access to the platforms and reagents you need to operate means you aren’t starting from a position of power. Apparently Illumina has committed to dropping their prices for competitors by 40% by 2025. That’s nice, but their internal costs are lower than that, so I’d still be nervous.
4. What would have happened if Illumina had never spun GRAIL out? Would the FTC still be coming after them if this had been maintained as an internal project, forcing them to divest it into an independent company? 
5. Why wasn’t this an issue when they acquired Verinata and dominated the NIPT market?
6. What other options do GRAIL’s competitors have? It needs to be high throughput and low cost, so that rules out PacBio. BGI/MGI would fit the bill from a technical aspect, but as of right now the MGI platforms aren’t broadly available (litigation pending, check back later). Then there’s Oxford Nanopore. With their PromethION system they certainly seem to have the cost and throughput necessary to compete. But despite the strong strides they’ve made in improving the read quality, early stage cancer detection via liquid biopsy is one of the more demanding applications. You’re looking for a rare signal and any sequencing error makes that harder. Has their error rate dropped low enough yet? The other potential issue is this application can’t really take advantage of ONT’s super long reads, so there would have to be some sort of prep to turn the short ctDNAs/RNAs into long concatemers. But I haven’t seen anyone try using ONT for liquid biopsies yet.