I was alerted, by both Salmon and Wolf, to a fascinating thread over on Vertex’s Yahoo! stock chatboard. It was there that I was allowed to “stand, even if only for a moment, on the shoulders of giants“, and see the teleprevir (Vertex) v. boceprevir (Schering) next-generation Hep C trials from a much-clearer vantage point.
As Wolf first correctly pointed out, right here, in our comment box, an “initial, preliminary” pass over the Vertex v. Schering study abstracts might lead one to conclude that Schering’s candidate is yielding better response rates. If one takes a moment to listen to independent experts in these matters (or, stands for a moment, on those Titans’ shoulders), it becomes clear that Vertex’s drug candidate may actually be beating Schering’s on response rates. We already know the Vertex product is far less likely to cause serious adverse events (Schering’s is three times more likely to lead to severe adverse events, in Schering’s own words). And that is the crux of the flim-flam on efficacy — it seems Schering is not counting some of its drop-outs (or adverse events) in its calculations of response rates. At least 16, for certain. How so?
Let’s read from a summary of what can be seen, from above the forest canopy, about the Schering-Plough press release of Monday, through the eyes of experts on the Vertex boards:
. . . .The Schering press releases consistently refer to a total of 595 patients in the Sprint 1 study of boceprevir. So, let’s take that as a given.
The EASL 2008 abstract refers to 59 patients in the low riba arm, and removing those leaves 536.
The EASL abstract and yesterday’s press release specify 104 in the control, which leaves 432.
The EASL abstract and yesterday’s press release specify 206 in the combined lead-in arms, leaving 226.
As far as the no-lead in arms, the EASL abstract specifies 226, while the press release specifies 210.
The EASL abstract figure of 226 agrees with all of the other numbers, and yesterday’s press release specifies 210. The EASL abstract figure of 226 agrees with all of the other numbers [See slide 12 of that PDF link — Ed.].
So, the preponderance of evidence supports the 226 figure being right. Until there is some other explanation, I will assume 226 is right.
But then we are missing 16 patients, and since this an intention to treat analysis they should be there If they had achieved SVR, Schering would certainly have included them as it would have greatly improved how boceprevir looks. But, they didn’t.
When Vertex reported their Prove 1 and 2 results they specified the number of patients who had dropped out for adverse events, as well as though who had dropped for other reasons and who had been lost to follow. Vertex counted all of these dropouts as failures, even though most of those lost to follow-yup had had an SVR at last contact. Schering, however, has only reported discontinuations due to adverse events; they have not included those lost to follow-up. Are those 16 patients lost to follow-up? If so, they should have been in the calculations. And if so they would lower the SVR rates by about 4 to 5 points in the no lead-in arms. . . .
It is curious that Schering chose to press release, rather than publish, in a peer-reviewed journal these interim boceprevir results. It is doubly curious, given that Vertex made announcements of important progress on its teleprevir efforts, in Vertex’s second quarter 2008 press release on the Friday immediately prior to Schering’s Monday presser. That knee-jerk reaction, by Schering’s External Communications Group, smelled of desperation, to me, at least — given that it is undisputed that Vertex holds a commanding lead in the race to obtain FDA approval. From these two pressers, then, we may also glean that teleprevir is showing a lower incidence-rate related to severe adverse events, at least in Phase II trials. So, why would Schering be desperate?
Well, the most recent Schering Form 10-Q, filed with the SEC, shows (at page 31, fourth full paragraph from the bottom; when added together with the figures in the second full paragraph on the top of page 32) that Schering generates more than $580 million in revenue (in just the first six-months of 2008) from treating Hepatitis C — but that revenue stream is essentially flat, to slightly decreasing. If we generously assume Schering will sell about $1 billion worldwide of its current generation Hep C treatment drugs in 2008, we begin to see why the race for the next generation Hep C candidate is so critical to Schering.
The “best case” for Schering is that it keeps — but essentially cannibalizes — this existing $1 billion in annual revenue, by being first to market with boceprevir — and is able to convert its Hep C patients to the new regimen. That will let Schering grow share, at least a little.
Now, if — as most predict — Vertex’s teleprevir beats Schering’s product to market, with at least similarly comparable efficacy — this $1 billion franchise will very quickly vanish from Schering’s revenue line.
That is why the press release — that is the hurry — that is the desperation.
Should Vertex “win out“, here, Schering will be missing another $1 billion per year in future revenue-streams. Not just opportunity cost, but erosion of existing revenue, as doctors switch Hep C patients worldwide to teleprevir, the next generation Hep C standard of care.
So this likelihood, plus the continuing decline in the Vytorin/Zetia franchise truly spell deeper woes ahead for Schering. On that last subject, Schering cautioned investors — in its most recent Form 10-Q — not to expect as robust sales of, and equity income from Vytorin/Zetia in the second half of 2008, as were seen in the first half. Scheirng’s first half 2008 was off 26 percent — and Schering itself is cautioning that will get worse — at the profit line (from the bottom of page 40):
. . . .Based on the expectation of lower U.S. sales from the Merck/Schering-Plough cholesterol joint venture, Schering-Plough expects that equity income will be lower in the second half of 2008 than it was in the first half of 2008. . . .