A little over a year ago we caught considerable flak for our posts about Abigail Alliance for Better Access to Developmental Drugs v. von Eschenbach, 495 F.3d 695 (D.C. Cir. 2007) (en banc), the case that ruled against any constitutional right of patient access to unapproved, experimental, but potentially lifesaving drugs. We said then that we were on the fence about the constitutional right itself, but that we agreed with the result for practical reasons – because the inevitable next step would be suits brought against the pharma companies themselves. After all, drugs don’t come from the FDA – only from drugmakers.
A number of folks, especially over at the Volokh Conspiracy, thought we were full of it. Some of the critics thought that we were either ignorant or intentionally playing dumb about whether drug companies could be “state actors” for constitutional purposes. We readily conceded that we’re not constitutional scholars – but we stoutly maintained, then and now, that we don’t play on this blog. What you see is what you get.
We ended our part in that discussion at basically the same point we started it:

Our concern was that terminally ill patients who are permitted to take unapproved drugs must get the drugs from somewhere. The only place to get the drugs is from the people who have them – the pharmaceutical companies. Terminally ill patients will thus predictably turn handstands to try to force pharmaceutical companies to provide those drugs. And the emotional appeal of lawsuits filed to obtain access to those drugs will be truly compelling. Plaintiffs will be telling judges that they are terminally ill and have no hope for survival unless they are permitted to take an experimental drug. In that context, shouldn’t the drug manufacturer provide the drug, so the patient can have a chance to live?
The scholars at Volokh criticized our thinking in two ways. First, they insist that drug companies are not “state actors,” so no court would ever find that the fundamental constitutional right alleged in Abigail Alliance applies to them. As we said in our previous post, we’re far less sanguine about that than the academics are. The FDA regulates drug companies awfully closely, and judges will bend over backwards to find in favor of plaintiffs in the emotionally-charged situation we’ve just described. In that situation, we would not be surprised to find a judge (wrongly) ordering a company to provide an experimental drug.

Well, we hate to say “we told you so….”
Actually, let’s be honest about it – we love to say “we told you so.” It’s a natural human instinct, and as lawyers it’s a large part of what we do (come to think of it, in one form or another, that’s virtually all our adversaries on the other side of the “v.” do).
What the Internet allows us to do is say “I told you so” to a larger audience.
Well, even without a constitutional right, a terminally ill plaintiff very recently succeeded in getting a judge to issue an injunction ordering a drug company to provide him with an unapproved, experimental drug. What’s worse, it’s on an ad hoc, extra-regulatory basis – a tort suit – precisely the sort of thing that will encourage everyone else in a similar situation to start trying to build their own cases.
Our practical concerns, it appears, weren’t just the crocodile tears of big pharma lawyers after all.
The opinion, Gunvalson v. PTC Therapeutics, Inc., has already received some comment on the Net – the New York Times ran a story, and the usual suspects over at Pharmalot picked up on it, as did a couple other denizens of the blogosphere. Oddly, nobody seems to have provided a link to the Gunvalson opinion itself, so we’ll rectify that oversight.
Something else nobody seems to have done is examine the practical implications of judges taking it upon themselves to issue this kind of an order. As usual, we’re taking it upon ourselves to fill this much needed gap in the literature.
So what was Gunvalson about? As we predicted (that’s another way of saying “we told you so”), it’s a case with “truly compelling” “emotional appeal.” The plaintiff (not technically, but in reality) is a young boy with a nasty, invariably fatal condition – Duchenne muscular dystrophy. If he had an average case of the disease, he’d likely be dead in less than a decade. And the plaintiff’s case seems to be worse than average, which is a big part of the legal problem.
The defendant has a drug in development (something called “PTC124,” but the name isn’t important) that’s showed promise in treating Duchenne. That’s a really big deal, because right now, there isn’t any proven cure at all. In short, a Duchenne diagnosis is effectively a death sentence.
Most people will do just about anything to avoid death, so the plaintiff quickly entered the world of experimental drug treatments and ongoing clinical trials. That brought the plaintiff into contact with the defendant, as well as with at least one other potentially useful – but clinically unproven – treatment for Duchenne.
That also made for some tough medical choices. The first thing this plaintiff tried was something called Gentamicin, another experimental Duchenne treatment apparently sponsored by somebody other than defendant PTC (we’re assuming, anyway, although that’s not entirely clear from the opinion). Slip op. at 2.
The problem was, that to enter PTC’s clinical trial, plaintiff would have to stop taking the other drug (another drug would create a confounding factor, so presumably the protocol for the trial excluded polypharmacy (a fancy term for taking more than one drug at once)). A muscle biopsy (a procedure where a small chunk of muscle is surgically removed) was also required. Id.
At this time, plaintiff (or more properly plaintiff’s mother and presumably his doctor) had some direct communications with PTC that ultimately formed the basis of their lawsuit. Slip op. at 2-3.
Stop right there – that’s problem number one – If direct company-patient communications can be the basis for expensive and disruptive litigation, those of us representing companies will advise our clients either not to engage in such communication at all, or to do so only through formal, recorded procedures. As a matter of medical practice and more than that, simple human compassion, we don’t think it’s a good idea to inhibit such communication between desperate, terminally ill people, and companies that might just be in a position to help. But as lawyers, our first duty is to our clients.
So that’s our first objection to what happened in Gunvalson – if lawyers insist upon turning companies like this into defendants, they will necessarily act defensively.
Anyway, what was said during these conversations is now disputed. Our bias is to believe the defendant, of course, but readers can compare the back and forth in the opinion and draw their own conclusions. We’re more concerned about how it was said.
In a nutshell, what seems to have occurred is that the course of Duchenne (like a lot of serious diseases) isn’t exactly predictable. The plaintiff had to make a choice among treatments. He wasn’t doing too badly with his current drug regimen, so staying on it was a viable option. At that time he also wasn’t doing too badly – for a Duchenne sufferer – from a general medical standpoint, so another option seemed to be to join a later phase of the defendant’s clinical trial. Slip op. at 2-3.
But things changed. For one thing, the defendant’s drug turned out to work pretty well against Duchenne. Slip op. at 3 (“[t]he trial was a success”). Thus plaintiff had every incentive to change his mind and enroll in one of the defendant’s studies. But that wasn’t possible because of a second change – his Duchenne apparently took an abrupt turn for the worse, and by the time he came back to PTC he was too sick to participate in its trial – given the protocol for the study. Id. (plaintiff’s “condition had severely deteriorated by this point”).
“[G]rowing desperate,” id., plaintiff tried to get the defendant to give him the drug through the FDA’s “compassionate use” regulations. PTC said “no.” While we don’t claim to be experts in “compassionate use,” we took a quick look at the relevant regulation, 21 C.F.R. §312.34. We note in particular that anybody implementing a “compassionate use” program has to comply with a bunch of other regulations “including the regulations governing informed consent and institutional review boards and. . .applicable provisions. . .including distribution of the drug through qualified experts, maintenance of adequate manufacturing facilities, and submission of IND safety reports.” 21 C.F.R. §312.34(d). In other words, it takes a lot of money and effort to offer a drug for compassionate use. Also, (assuming the discussion in the initial Abigail Alliance opinion is right) “compassionate use” drugs, can only be supplied “at cost.” 445 F.3d at 474.
Basically, a drug company would be crazy to open a compassionate use program just for one person – nor could one person afford to pay the “cost” of all the overhead compassionate use required.
The court dismissed these cost considerations as “trivial.” Slip op. at 8. They’re not. Otherwise the compassionate use regulation would not be, as the Abigail Alliance plaintiffs asserted, “available only to a fraction of those in desperate need.” 445 F.3d at 474. And in Gunvalson the plaintiff sure wasn’t offering to foot the bill (and in all likelihood couldn’t conceivably afford to) for the defendant’s compassionate use obligations in return for receiving the drug.
But a one-person compassionate use regime (as least by analogy) was precisely what plaintiff in Gunvalson demanded when it came to litigation. Being, as the court said, “desperate,” plaintiff took the position that whatever the defendant had said about later participation in its trials amounted to a promise to provide him with its experimental drug no matter what, even if he no longer qualified for any research protocol that would provide any data that would be useful to the defendant’s ongoing research.
Thus, the defendant in Gunvalson got nothing at all of value to it, and everything it’s required to do comes off its bottom line – making it more difficult and expensive to bring its lifesaving product to market for other Duchenne sufferers.
Plaintiff invoked theories of “promissory estoppel, fraudulent misrepresentation and negligent misrepresentation,” slip op. at 3, to demand essentially that PTC should be compelled to give him its drug for nothing. And because plaintiff was seeking a preliminary injunction, he didn’t even have to win – all he had to show was that it was “reasonably likely” to succeed, a standard that “need not show anything approaching a certainty of success.” Slip op. at 4.
So Gunvalson presented the difficult combination of a deathly ill plaintiff suing under a quite malleable legal standard. It’s a classic example of bad facts making bad law, and vice versa.
What did the court do to evaluate the chances of success? Just what no sane drug company would ever do. It applied the FDA’s standards for compassionate use to an individual claim. Slip op. at 4-5. Never mind that the FDA’s regulations are expressly voluntary (“may be”). And never mind that there is no private right of action under the FDCA. And finally, never mind that, under the law of every state that we know of (press reports indicate that the plaintiff is from Minnesota) a government enactment cannot be used as a legal standard in a tort case (negligence per se) unless that enactment is mandatory in its own right. E.g., Mervin v. Magney Construction Co., 416 N.W.2d 121, 123 (Minn. 1987).
Well, the defendant’s drug – like virtually any experimental drug possibly useful in the treatment of an otherwise fatal disease – satisfied the FDA’s voluntary compassionate use regulations. It wasn’t even a hard call. Slip op. at 4-5. Of course not, since the FDA’s regulation was expressly written to allow compassionate use for as wide a universe of experimental drugs as possible. But the court’s injunction converted a voluntary program into a mandatory legal duty.
The crux of the rationale for that legal legerdemain was a disputed oral promise that the plaintiff claimed entitled him to a life-long supply of the defendant’s experimental drug. The expansive doctrine of promissory estoppel makes even promises without “consideration” (a legal term for giving something to get something in return) legally enforceable if there’s been “detrimental reliance.” Slip op. at 5. The oral promise at issue, of course, was subject to considerable factual dispute. The opinion details the various disputes at considerable length. Id. at 6-8. For purposes of the injunction, all the court had to hold was that it was “reasonably likely” that such a promise had been made. Id. at 6-7.
We weren’t there, so we can’t say what really happened either, so we won’t go into that.
But another aspect of the opinion is clearer. The allegedly “detrimental” reliance was the plaintiff’s medical decision to continue with a treatment that seemed to be doing OK at the time, rather than switch to the alternative of the defendant’s drug (which, afterwards, clinical data showed to work better). Slip op. at 5, 7. This exercise of medical judgment may be “reliance” but it only proved “detrimental” with the benefit of 20/20 hindsight. Nobody knew at the time how the defendant’s ongoing clinical studies were going to turn out, nor was it apparent that the plaintiff’s Duchenne was going to deteriorate so rapidly. All the plaintiff essentially did was make the best informed medical judgment he could at the time. To allow legal liability to turn on such a choice seems, to us, a form of legal “heads, I win; tails you lose.” The defendant’s drug turned out to be a lifesaver. But it’s reward for developing the drug turns out to be an injunction requiring it to give its drug away.
To justify its result, the court cites (slip op. at n.3) as “similar” Dahl v. HEM Pharmaceuticals Corp., 7 F.3d 1399, 1401 (9th Cir. 2003). So we looked at Dahl. It did affirm an injunction mandating the supply of an experimental drug – but the facts were critically different in two respects. First, the plaintiffs in Dahl were already “enrolled” in the defendant’s clinical trial. 7 F.3d at 1401. That means a limited universe of plaintiffs. Second, and even more important, because of plaintiffs’ enrollment, their claim of entitlement to the drug was based on express language (“you will be offered [the drug]”) in the contract of enrollment, a contract drafted by the defendant. Id. at 1401-02. That means there was a written promise intentionally made by the defendant, for consideration (enrollment in the study). It further means that the defendant had the ability to eliminate this kind of liability simply by writing a different contract.
That’s the second big problem – Gunvalson is thus hugely different from Dahl. Any poor soul who, when staring death in the face, succeeds in opening communication with the maker of a promising experimental treatment for his/her condition can take a shot at a “promissory estoppel” lawsuit by alleging a consideration-less oral promise under the rationale in Gunvalson. The stark, unavoidable fact is that people at risk of imminent death literally have nothing to lose. The scope of potential liability under the legal theory permitted in Gunvalson essentially runs to anybody whom a defendant’s unapproved, experimental drug might conceivably help, and who does nothing more than gets the defendant to talk to him/her, not just the actual clinical trial participants in Dahl.
And, of course, the differences between Gunvalson and Dahl tie back to our first concern: the only way to avoid Gunvalson-type liability is for a defendant to restrict its oral communication with desperate people who might later turn into plaintiffs, as well as patients. We don’t think it’s a good idea to force manufacturers to go that route.
Finally, the court in Gunvalson dismisses the defendant’s “floodgates” concerns as “unpersuasive.” Slip op. at 8-9. The court goes out of its way to characterize the plaintiff’s mother’s “involvement” with the defendant as “unique” in terms of the “unusually close relationship.” Id.
We respectfully disagree. While it may be that, under Gunvalson few plaintiffs would actually succeed (and we have doubts about that), that doesn’t mean that there won’t be plenty of people desperate enough to try. Dahl, after all, involved seventeen people even though it drew from a much smaller universe of potential plaintiffs.
Gunvalson adopted a broadly applicable liability theory that allows injunctive relief where the plaintiff shows nothing more than a disputed assertion of an oral promise with no consideration – only “reliance” in the form of a medical decision that seemed perfectly rational at the time it was made.
One lesson we draw from our mass tort experience is: “If you build it, they will come.”
And we’re not dealing with totally rational actors here. The plaintiff in Gunvalson had no other straw to grasp at and his counsel worked pro bono (for free). Ask yourself these questions: Would you beg, plead, and cajole to get into an experimental drug trial if it was the only thing that might possibly save your life? If you wouldn’t do it for yourself, would you do it for your son or daughter? How about if by filing a lawsuit you could jump the drug trial queue altogether and get this potentially lifesaving treatment for free – and apparently forever?
A whole lot of lawsuits have been filed over a whole lot less.