Jefferies
Key Takeaway
The Copaxone inequitable conduct trial kicked off yesterday. The anticipated
schedule extends through July 26, with 16 witnesses planned (6 live; 10
deposition). We believe Days 1 & 2 are key, with direct and cross-examination of
Dr. Irit Pinchasi, the project manager on Copaxone development. Following Day
1, we remain confident that the generics will have a very tough time getting an
inequitable conduct judgment vs. Teva.
Key Witness Establishes Narrative. On direct examination of Dr. Irit Pinchasi, the former
project manager of the “Copolymer-1” (COP-1) development program, Teva’s counsel
presented a coherent narrative of the development history of low-molecular weight (MW)
COP-1. Dr. Pinchasi’s testimony traced the path from the earlier Weizmann Institute’s higher
MW COP-1 (>20K Daltons mimicking endogenous myelin basic protein) which caused
worrisome sporadic “vasomotor reactions”, to the use of RBL degranulation batch screening,
to the unexpected discoveries of a positive MW-toxicity correlation and meaningful disease
activity at lower MW. Dr. Pinchasi readily conceded the variability in RBL degranulation
testing (high relative standard deviations of both precision and reproducibility), but
maintained her confidence in a clear MW-toxicity correlation (which eventually served as the
basis for the patented invention). Regarding the prior art (’550 patent from 1974 disclosing
COP-1 at higher MW), Dr. Pinchasi suggests that given the near term patent expiration at the
time (in late-1986), Teva actually planned to develop COP-1 regardless given the complex,
hard-to-reproduce nature, and expected orphan drug exclusivity. That is to suggest the low
MW formulation was safety driven rather than patent driven.
Cross Examination Less Hit, More Miss. Mylan (MYL, $24.33, NC) and Sandoz
(NOVN:VX, CHF51.50, Buy) separately cross-examined, with Mylan’s cross particularly
disjointed, disorganized, and in our view less ineffective. Between the generics, the crossX aimed to show (as far as we could deduce) that : 1) obtaining patent protection was the
driving motive for low MW, 2) the RBL cut-off defining “toxicity” was arbitrary and cherry
picked to show a MW-toxicity correlation, and that RBL was poorly correlated with in-vivo
AE’s, and 3) that TEVA’s patent app characterized RBL as “reproducible” yet Dr. Pinchasi
being quoted in writing that RBL method was not sufficiently reproducible to determine
safety. We believe Dr. Pinchasi was very strong on cross-X, consistently maintaining she had
a passive role in data selection for the patent application itself, defending her non-IP-driven
finding of MW-toxicity correlation, and generally not yielding on vague or leading questions.
Importantly, on the issue of her allegedly undermining the RBL reliability, the proceedings
closed with Dr. Pinchasi emphatically clarifying that her comments were directed to the RBL
as an insufficient “quality control” test (not that it doesn’t sufficiently correlate MW with
toxicity). We think the generics are way off base on the last issue.
A long way to go, but Inequitable conduct hurdle is very high. Sandoz will continue
cross-examination of Dr. Pinchasi tomorrow, and then there are 15 further witnesses
(including several hours of taped deposition video). An issue that will certainly rise is to
what extent data was cherry picked for patent prosecution, and to what extent it skewed
the USPTO review. For example, apparently only 4 COP-1 batches were discussed in the
patent application yet there must have been >50 batches with reams of data at the time.
Regardless, note that the Therasense en banque decision raised the inequitable hurdle very
high, such that: 1) omitted references are material only if “but for” its exclusion the patent
wouldn’t have issued, 2) there must be clear and convincing evidence of intent to deceive
by the applicant (highest burden of proof), and 3) there is no more “sliding scale”, i.e., both
materiality and intent to deceive must be clear and convincing (high materiality doesn’t
compensate for weak intent, nor vice versa) |
