This post is contributed by Melissa Wojtylak, of ReedSmith, one of our regular guest bloggers whom we’re trying to convince to join us on a more formal basis.  Take it away Melissa:

******************

 The Southern District of Illinois delivered a giant lump
of coal when it denied the defendant’s motion for summary judgment in a Zometa
case last week.  In Rutz v. Novartis Pharmaceuticals Corp., 2012 U.S.
Dist. LEXIS 177779 (S.D. Ill. Dec. 17, 2012), plaintiff claimed that his
decedent developed bisphosphonate-related osteonecrosis of the jaw (“BRONJ”)
after taking Zometa for 3 years while undergoing treatment for breast
cancer.  In considering defendant’s
summary judgment motion, the court looked first at the evidence on specific
causation, then considered the evidence on warnings.   Because the warnings discussion is the most
troubling – and basically obviates the need for a specific causation discussion
– we’ll look at the court’s warnings analysis first. 

The record indicated that plaintiff’s decedent took
Zometa from June 2002 through August 2005. 
Id. at *6.   According to the opinion, the defendant learned of the drug’s association with BRONJ in 2002, and subsequently amended
its labeling in September 2003, February 2004 and September 2004; it also sent
Dear Doctor letters in September 2004.  Id. at *12.  Ms. Rutz’s doctor got this letter, but continued
to prescribe the drug for her until August 2005.  Id. at
*5.   That’s right: he continued to
prescribe for two years after the first label change.  Perhaps not surprisingly, the defendant
argued that there was no evidence that it should have acted more quickly to
implement a different warning.    Id. at
*12.   The court merely found that the
defendant’s actions created an issue of fact as to whether it should have
warned sooner or differently.  Id. at *14-15.  Completely absent is any discussion of the prescriber’s  testimony on the warnings, or why he decided
to continue prescribing the drug through three label changes and a Dear Doctor
letter – for nearly a year after the latter, in fact.  While we don’t know the doctor’s thinking, as
Bexis noted in a prior post,  where the alternative to therapy
is likely to be death from cancer, it’s tough to prove that a doctor would
choose not to prescribe a drug based on the possibility of a lesser injury.

Plaintiff argued that the prescriber was not a learned
intermediary, and also, that he would have heeded a warning if it had been
adequate.  Id. at * 17.  Loose language in some Illinois cases permits plaintiffs to argue that the learned intermediary doctrine should be set aside (what they really mean is that the warning can be found inadequate) if the jury determines
that the prescriber was not sufficiently warned, but the court didn’t take this
on.  Instead, it punted on deciding as a
matter of law whether this prescriber was a learned intermediary.  Id.
at 18.  This muddied the waters for the
discussion on heeding presumption, as the court admitted that if the
doctor was a learned intermediary, it was not clear whether a heeding
presumption would apply.  Undaunted by
this fact, the court handled this obstacle by deciding that even if this
prescriber was considered a learned intermediary, a heeding presumption would
apply (again, that’s after admitting that there was no controlling Illinois
authority on this point).  Id. 
To support this conclusion, the court relied on a decision from the
Illinois Appeals Court, which applied Texas law and found a heeding
presumption applies to learned intermediaries. 
Bexis undoubtedly had an aneurism when he read that, given Erie, predicting expansions of state law, and all that.
 
See, the problem with the Rutz court’s reliance on the Illinois appellate
opinion was that the appellate court not only didn’t apply Illinois law, but got Texas law wrong.   Readers of the blog will remember that in Ackermann v. Wyeth, the Fifth Circuit
found that the heeding presumption does not apply to learned intermediaries,
and that even if it did, it would mean only that the doctor “would have
incorporated the additional risk into his [or her] decisional calculus.”  However, relying on the incorrect
statement of Texas law and one decision from a sister court (the Northern
District of Illinois), the Rutz court
proclaimed that “the heeding presumption is a natural result of or corollary to
the learned intermediary doctrine.”  Id. at *20.  Thus, if Mrs. Rutz’s prescriber was a learned
intermediary, there was a presumption that he would have heeded an adequate
warning.   While this seems dangerously close to circular
logic to us (and not at all natural), it’s also contrary to the law in a number
of jurisdictions.  As the blog has
discussed in the past, Thomas v. Hoffman
LaRoche
, 949 F.2d 806 (5th Cir. 1992), contains a good
substantive discussion of why the heeding presumption should not apply in
learned intermediary cases. 

Continuing with the theme of things that don’t exist
in Illinois law . . . . Illinois does not have a statutory presumption of
adequacy for FDA-approved drugs, so as you might have guessed, there was no
chance the defendant was getting judgment as a matter of law on that
basis.  But the court was unwilling to
consider the adequacy of the warning as a matter of law in general, based on
what it called the law-of-the-case doctrine.  
For this argument, however, the court relied not on rulings from the Rutz case, but on another Zometa case in
which the MDL court – applying Florida law, no less – found that the adequacy
of the warning was a question for the jury. 

Based on the warnings analysis, a discussion of the
court’s treatment of the specific causation argument may be unnecessary, but
humor us.  In his deposition, the
plaintiff’s expert had admitted that Rutz’s condition did not meet the accepted
definition of BRONJ developed by the medical community.  Id.
at *7.  The court found several ways to
get around this troubling admission. 
First, it noted that the definition had not been developed until after
the decedent was diagnosed.  Then, the
court rejected the definition itself. 
According to the criteria in the definition, in order to support a
diagnosis of BRONJ, the patient must have at least eight weeks of exposed bone.  However, the Rutz court accepted plaintiff’s expert’s opinion that BRONJ could
be present with only 4-5 weeks of exposed bone. 
The problem with this opinion (besides the fact that it goes against the
thinking of the rest of the medical community)?   This decedent didn’t have a single
incident
of exposed bone documented in her medical records.  Remarkably, the court handled this
inconvenient fact by holding:

Given that Rutz may have had exposed bone which was
not recorded, it would be unfair to permit [defendant] to use the absence of a
record to conclude that BRONJ was not present . . . . [a] jury could determine
that other indicia of BRONJ – such as non-healing wounds – made it likely that
exposed bone was present.” 

Id. at *10-11. 
The court also glossed over the defendant’s argument that this decedent
had other disease processes that could cause ONJ.  Id. at
*7.   Where did the burden of proof go?  The record has nothing to support causation.  Absence of evidence is just that … absence of evidence.

Let’s hope this opinion, with its many missteps,
doesn’t haunt defendants in drug cases like some sort of ghost of Christmases
past.