Celsis: Federal Circuit Upholds Preliminary Injunction over Dissent

by Dennis Crouch

Celsis In Vitro v. CellzDirect and Invitrogen (Fed. Cir. 2012)

This case is important because it provides an important incremental analysis of when a validity challenge is sufficiently substantial to lead a court to reject a motion for preliminary injunctive relief.

Majority opinion by Chief Judge Rader (author) and Judge Prost. Dissent by Senior Judge Gajarsa.

Celsis owns Patent No. 7,604,929 — claiming as its invention a method and system of freezing and preserving human liver cells using multi-cryopreservation.  In a split opinion, the Federal Circuit affirmed the lower court's preliminary injunction order stopping the defendants from practicing the invention. 

When appealed, a preliminary injunction is reviewed for abuse of discretion that requires clear error, either in “judgment in weighing relevant factors” or in the exercise of “discretion based upon an error of law.”  A decision on preliminary relief is immediately appealable. 

Dissent on Substantial Question of Obviousness: Writing in dissent, Judge Gajarsa argued that this case is not amenable to the “extraordinary and drastic remedy of a preliminary injunction” because the defendants “raised a substantial question as to the validity” of the asserted patent.  Focusing on the burden of proof, Judge Gajarsa noted that a preponderance of the evidence is sufficient to prove the “substantial question” and that “it is unnecessary to prove a substantial question of invalidity by clear and convincing evidence.”

The the key dispute between the majority and dissent was on the issue of obviousness.  Judge Gajarsa wrote:

In my judgment, the district court abused its discretion in finding that Celsis had demonstrated a likelihood of success on the merits because the claimed invention is nothing more than a repetition of steps already known in the art.  Moreover, the majority perpetuates this error, and in so doing applies the wrong standard for obviousness and rationalizes the issuance of the preliminary injunction because it would prevent competition with a patented process which may be proven to be invalid. 

In particular, Judge Gajarsa argued that the district court and majority improperly rejected the obviousness argument by seeking some teaching or suggestion (TSM) in the prior art that would lead someone to attempt to combine the known elements of the claimed invention.

In particular, the majority wrote that the key prior art does not disclose “micro-cryopreservation” as claimed nor does it suggest this step as a solution to problems known in the art.  Chief Judge Rader wrote “This court has not seen [the Defendants] identify any teaching, suggestion, or motivation in the [prior art] that multiple rounds of freezing would somehow increase rather than decrease cell viability.”  The majority agreed with the lower court that the obviousness argument lacked credibility because "not a single one of that astonishingly large body of literature was devoted to the subject of multi-cryopreservation of hepatocytes."  Judge Gajarsa felt this fact was inconsequential because multi-cryopreservation simply involved repeating steps that were already known.  

DDC Comment: The TSM test for combining prior art was eliminated by the Supreme Court KSR v Teleflex in much the same way that the MOT test for patentable subject matter was eliminated in Bilski. In both cases, the strict tests remain probative, if inconclusive, tools.  However, in the context of a preliminary injunction a defendant's failure to provide a motivation to combine prior art may well be sufficient find a patent “likely nonobvious.” 

Although identified as a “factor”, the likelihood of success factor turns out to be a necessary element of a successful motion for preliminary relief.  In other words, if the plaintiff is unable to prove that it has a sufficient likelihood of success on the merits of the case then the district court will deny the preliminary injunction motion.  This case provides some assistance in determining when that threshold has been met.

Motivation versus Long Felt Need:  Chief Judge Rader is known for playing logical games with the secondary indicators of nonobviousness. Here, the accused infringers provided evidence of a “market need” for the solution and argued that those market forces would prompt the variation found in the patent.  That argument follows the statement in KSR that “market forces can prompt variations” of known design elements.  Judge Rader flipped the argument and instead noted that such market need is when “properly linked to the claimed invention, is actually probative of long felt need under objective criteria analysis and supportive of non-obviousness.”

Preliminary Injunction Standards: A preliminary injunction is decided on a four-factor test that is almost identical to the test outlined for permanent injunctions in eBay v. MercExchange.  Namely, a court must consider (1) likelihood of success on the merits, (2) irreparable harm, (3) balance of hardships, and (4) public interest.  In the context of permanent relief, the likelihood of success on the merits is no longer a factor because permanent relief is only determined after the patentee wins on the merits. In eBay, the Supreme Court included an additional factor –  that monetary damages are insufficient.  However, that factor is likely wholly subsumed by the “irreparable harm” factor.

Here irreparable harm was shown by expert testimony and evidence of "price erosion, damage to ongoing customer relationships, loss of customer goodwill (e.g., when an effort is later made to restore the original price), and loss of business opportunities." And, the public interest factor is normally presumed to be met in patent cases because the public interest favors enforcement of valid patents.

37 thoughts on “Celsis: Federal Circuit Upholds Preliminary Injunction over Dissent

  1. 36

    Max, I do not know if I agree with you. This is a process claim. To infringe, one must perform the steps. If as a result of the steps, one reaches the goal of 70% viability, then the process claim is sufficiently described, enabled and claimed.

    If, however, the steps to not inherently result in the 70% viability claimed, there is something radically wrong, either in the description, the steps required, or the like. That is my take. The 70% must be achieved by the process steps. The claim is not to a result, per se.

    From the facts in the case, I was drawn to the fact that the accused infringing process apparently had to perform a final grade in order to achieve 70% viability. If this was in fact true, they could not have infringed, for one. Second the claim is invalid under Section 112.

  2. 35

    Is your point that the necessary and sufficient function of a claim is to define an invention, while the complementary necessary and sufficient task of the specification is to enable the invention?

  3. 34

    You are wrong in passing off that what you are merely guessing.

    You did not read my posts closely enough. Had you prefaced your original post with something like “I am merely guessing…” then we could have avoided this back and forth (and that was the entire point of the moniker).

    BTW – you are still off on what is required in a claim as opposed to the enablement of that claim (call it the bane of “parameter claiming” or whatever you want to call it)..

  4. 33

    Gajarsa says that coming to the mere idea of freezing twice may very well be obvious, and that dressing it up with an impressive number like “70%” doesn’t make it any less so.

    I’m saying that I see his point. Resort to “parameter claims” is the standard way to get spurious subject matter to issue. Such claims are everywhere these days, an infernal nuisance to honest-to-goodness competing manufacturers, and the bane of every EPO opposition prosecutor’s life.

    I haven’t read all the evidence here. I’m guessing. Perhaps somebody who has read all the evidence will educate us both. Until then, I will guess that it is merely your guess, that I’m wrong.

  5. 32

    But I see in the claim no sign of the pains.

    Do you understand what has to be in the claims (as opposed to that which is to be supporting the claims in the spec)?

    And anyway, your logic of “And anyway, 80% is better than 70, and 90% is better than 80. So, the goal is 100%. So choosing, for want of any other number,70%, is indeed arbitrary” is fallacious as once again you are ASSuming that which you posit. Perhaps there is a reason why 80% could not be claimed. Perhaps not. You are merely ASSSuming and guessing.

  6. 31

    Ok, so the pains taken might get you over the threshold of 70%. But I see in the claim no sign of the pains. As it stands, the claim is just a free beer claim (claiming the result, without specifying what you have to do to get there). Obviousness is easier to prove, when all you claim is the desirable result.

    And anyway, 80% is better than 70, and 90% is better than 80. So, the goal is 100%. So choosing, for want of any other number,70%, is indeed arbitrary.

  7. 30

    Are you assuming that the number is arbitrary?

    Perhaps the inventor painstakingly refined his invention to reach that number (I don’t know – but my educated guess is that you don’t know either).

  8. 29

    I must say that I do enjoy these cases, where the panel gets a flawed set of papers, whereupon 2 out of 3 deliver the “right” result while the dissent says what is wrong with the majority’s reasoning. The panel of 3 can congratulate themselves on a job well done, and the patent bar can learn how to do a better job next time.

    Here, Justice Gajarsa is 100% correct in his criticisms of the majority, isn’t he?

    On obviousness, the arbitrary inclusion of “70%” ought not to be the decisive point but one can’t deny it to drafters that peppering claims with arbitrary numerical values enhances their chances of getting through to issue, and beyond. Me, I think the more arbitrary numbers one finds in a claim, the more suspicious one should be of its patentability.

  9. 28

    I particularly enjoyed how Cellzdirect’s “market need” argument was characterized by Rader as “long-felt need” and used against them:

    “Without more, this reference to “market need,” properly linked to the claimed invention, is actually probative of long felt need under objective criteria analysis and supportive of non-obviousness.”


  10. 27

    Freeze, thaw, grade, refreeze: those are the steps. Perform them, you infringe.

    What if the result does not have >70% viable H’s. Have you infringed?

    There is the problem. It appears the defendant had to grade once more to get to 70%; but the evidence is not clear on this point.

    I think the wherein clause explains the result. If the process does not always result in the >70% viability without more, there is something wrong about the process claim. It does not include all the necessary steps, or the whole process in not enabled with enough information.

    On obviousness, I think the absence of any discussion in a crowded prior art is sufficient for me to indicate non obviousness. Enablement and infringement are the big issues.

  11. 25

    After reading about this case today I figured why not freeze the cells in the presence of a strong magnetic field and observe the results? Why not freeze/refreeze them in the presence of a strong magnetic field? Why not freeze them normal without strong magnetic field and then freeze them with a strong magnetic field the second time? And just see what happens?


    link to nextbigfuture.com

    However, I will note that I did correctly predict that the formation of the ice crystals would change in a free floating body of water to my officemate who thought there would be no change. See video confirming my speculation. And not only that, it turns out this procedure might be good for freezing foods without causing freezing damage so much. See the noodles they froze. And even saving removed wisdom teeth for later implantation when you need teeth when you get old and have no teeth or are missing a few.

    Beat me by a year :/

  12. 24


    I explained this all to you previously (your choice in refusing to understand and ascribe a “sockie” conspiracy viewpoint is, regretably, your choice).

    Regardless, whether you consider my pseudonym asinine or not is quite immaterial.

    Please stay on topic.


  13. 23

    simple sockie: MM, I have addressed your “sockie” concerns previously.

    Not really. You wouldn’t be posting under your current, asinine pseudonym if that was the case.

  14. 22


    MM, I have addressed your “sockie” concerns previously. Your condescension is unwarranted and unappreciated.

    Try to limit your postings to answering the questions.


  15. 21

    6 it is my perception that such issues usually get overlooked by those scrutinising the passing articles, on the conveyor belt of ex parte examination. But then, post-issue, when an opponent applies his finger to the delicate point, anybody with any feel for patent law is like “Oh yes. Good point. Wonder why nobody saw that point prior to issue” And, with that, the opposition succeeds.

  16. 19

    Perhaps there is an enablement argument hidding in there if you did all your homework on the case.

    I wonder also if the applicant enabled the full scope of these claims or if they were otherwise enabled at the time of filing. Those so called scope of enablement or “full scope doctrine” issues can really be devastating to applications and I would imagine patents also.

  17. 18

    What is truly amazing about such posts “with even less content” is that such still outshines the bulk of MM posts.


  18. 16


    You’ve outdone yourself, sockie. Until now, I thought your endless rants to the effect that my comments lack “substance” were the ultimate in meritless, unconvincing wanking. But this bit from you has even less content than those rants. I didn’t think that was possible.

  19. 15

    then the evidence supporting the argument, i.e., evidence showing what one skilled in the art would have expected at the time of filing, should be found in the prior art.


  20. 14

    I wonder if defendants facing preliminary injunction motions would benefit from filing SJ motions? Was there one here?

  21. 13

    sockie: Isn’t the inventor allowed to base his argument on hindsight?

    Why should that be the case? If your argument for non-obviousness rests in part on an assertion that a particular (e.g., claimed) result was unexpected as of the filing date (like the argument we’re discussing), then the evidence supporting the argument, i.e., evidence showing what one skilled in the art would have expected at the time of filing, should be found in the prior art.

    As I noted upthread, it’s certainly true that if you freeze and thaw a given population of cells, the percentage of viable cells in that population will go down. That’s been known for ages. This invention does not contradict that “expectation”. It’s also been well-known for ages that you can improve the percentage of viable cells in a population of thawed cells by gradient fractionation.

    Finally, it’s also been well-known for ages that if you re-freeze that selectively purified sub-population of cells and subsequently thaw them again, a certain percentage of that sub-population will die or fail to grow. Did skilled artisans reasonably expect that 100% of such a sub-population would die upon thawing? Of course not. Did they expect that 100% of them would live? Of course not. You could quite easily find ten or twenty scientists who have been doing such work for years and who are unaware of this case and ask them: based on your experience and knowledge, if you thawed a set of once-frozen hepatocyte cells and selected for viable cells by gradient fractionation, what percentage of that selected population would you expect to survive another round of freeze thawing. Isn’t Rader on the record pretty much extolling the integrity of scientific experts generally? It seems to me that the consensus answer to this question would necessarily carry a great deal of weight with the man.

  22. 12

    such an argument is just as much “based in hindsight”

    Isn’t the inventor allowed to base his argument on hindsight?

  23. 11

    As soon as you brought up the thing about the marathon runner I was like lulz, idk about that bro, you’re selecting for the cells that are good at surviving freezing. Like you said, the only theory here that seems incredible, if either of them, is the marathon runner one. I’m nowhere near a tiny bit of skill in the art and that much is apparent to me even from the tiny bit of reading I’ve done in that art.

    That said, perhaps back in the dark ages when this was filed the selection factor wasn’t as well known or comprehended? That I certainly can’t speak to all of my reading in this field has been very modern.

  24. 10

    Ned: The question is then, really, was the improvement using this process unexpected. You say no for the reasons you previously stated: those of ordinary skill would expect it.

    To be clear: based on what I read in the Federal Circuit opinion and my understanding of the technology, I would have decided the preliminary injunction issue differently.

    And also to be clear: the step of freezing thawed, purified cells a second time does not result in “improved” viability relative to using thawed, purified cells. It causes the death of roughly 30% of those cells which, according to the plaintiff’s expert, is less than one would have expected. Note that in the absence of any teaching in the prior art that one skilled in the art would have expected more than 30% of the cells to die after purification and re-freezing, such an argument is just as much “based in hindsight” than the arguments presented by the defense.

  25. 9

    Got ya.

    The question is then, really, was the improvement using this process unexpected.

    You say no for the reasons you previously stated: those of ordinary skill would expect it.

    I read the case earlier, but I do not have time to read it again right now. But I also recall there was a “teaching away” component to the argument as well.

  26. 8

    improvement patents should be banned

    The Reading Incomprehension Troll will be making its regular stops today. Please do not feed it. The strawman being attacked is for entertainment purposes only.

  27. 7

    in other words – improvement patents should be banned.

    next up, all patents – we will get you my pretties.

  28. 6

    Ned: Now, what about this “long felt need” argument? The need was established. It had been there for some time. Yet the solution was not reached until the inventors made the “invention.”

    A couple thoughts on this:

    First, as I noted above, I’m quite sure that “the solution” was “reached” before this application was filed. Researchers re-froze un-used gradient-purified cells a second time, thawed them. They just didn’t appreciate that (depending on their handling skill) the precise level of viability of the re-frozen cells. Or they didn’t appreciate that it was something that one could or should file a patent on. Can you imagine that? A scientist not thinking about patents all the time? Yes, it actually happens.

    Second, “long-felt need” arguments almost always distort the facts obscene ways. Of course people want to improve the viability of frozen hepatocytes. That will always be the case until someone comes up with a medium of freezing hepatocytes that results in 100% viability. Does that mean that for every improvement in hepatocyte viability that a “long felt need exists”? Or to put it another way, if someone files an improvement patent on Celsis’ technology, are they now precluded from presenting evidence of a “long felt need” because Celsis apparently satisfied that need fairly recently? Here on Earth, people have been working with frozen hepatocytes for a long time and accomplishing great things. During that time, I’m sure there have been numerous improvements made in improving the viability of those cells, many (nearly every one of which was simply dedicated to the public in the interest of science, just fyi).

  29. 5

    The ‘208 case is on the way to issue. I wonder if it will get opposed. I wonder if it will be asserted, as soon as it issues. Depends how wide its claims are, I suppose.

    There is, actually, another interesting aspect of these claims which is that they are rather “wide” in scope, at least with respect to the recited steps. The scope seems to be the sort of scope accorded to “pioneering” inventions. The breadth of the claim would seem more appropriate if applicants had discovered that two cycles of freeze-thawing actually improves the viability of hepatocytes relative to the viability of cells after one-cycle of freeze-thawing. But that’s not what was discovered.

    Merely carrying out the steps generically recited in the claim will not reproducibly achieve the “70% viability rate” which is trumpeted in the preamble and presented as a functional limitation in the body of the claim, and which was surely critical to the patentability of the claim. Perhaps the freezing, thawing, and gradient steps are defined with far greater specificity in the specification, providing additional critical limitations with respect to timing, temperature, resuspension/washing/storage media compositions, etc. I rather doubt that, however.

  30. 4

    Malcolm, good post.

    From Dennis’ summary of the case, I’d like to focus on the long felt need argument.

    Now, what about this “long felt need” argument? The need was established. It had been there for some time. Yet the solution was not reached until the inventors made the “invention.”

    Can you explain just how those of skill in the art could have missed this obvious solution for so long?

  31. 3

    In the “prioritised” thread, earlier today,, one “Ray” asked about the Celsis case. My reply was:

    “Ray for some discussion of patentability, you could read the EPO wrapper of 1891208 and its (now lapsed) div 2218777. The EPO files are full of evidence-based observations on patentability, emanating from an unnamed client of a prominent London patent attorney firm.

    The ‘208 case is on the way to issue. I wonder if it will get opposed. I wonder if it will be asserted, as soon as it issues. Depends how wide its claims are, I suppose.”

    So, I’m glad to see Dennis featuring obviousness. Malcolm, the EPO was told about the litigation in the US, had a close look, but then decided to issue a patent.

  32. 2

    Assuming the facts in the opinion are accurate, I agree with Gajarsa. The patent claims may ultimately be held to be valid, but the facts here don’t warrant a preliminary injunction because there are obviousness issues with the claims (see caveat below). For what it’s worth, like many experienced patent attorneys with biotechnology backgrounds, I’ve frozen, thawed, spun down and re-frozen plenty of cell preparations (way back when). All of the steps are routine.

    1. A method of producing a desired preparation of multi-cryopreserved hepatocytes, said hepatocytes, being capable of being frozen and thawed at least two times, and in which greater than 70% of the hepatocytes of said preparation are viable after the final thaw, said method comprising:

    (A) subjecting hepatocytes that have been
    frozen and thawed to density gradient fractionation to separate viable hepatocytes from non-viable hepatocytes,

    (B) recovering the separated viable hepatocytes, and

    (C) cryopreserving the recovered viable hepatocytes to thereby form said desired preparation of hepatocytes without requiring a density gradient step after thawing the hepatocytes for the second time, wherein the hepatocytes are not plated between the first and second cryopreservations, and wherein greater than 70% of the hepatocytes of said preparation are viable after the final thaw.

    As was admitted, freezing and thawing the cells and subjecting them to a density gradient was old in the art. The density gradient step was well-known to enhance the viability of cells (it’s simply a purification step which separates viable cells from non-viable cells).

    The only additional step recited in the claim is re-freezing the cells. I was surprised to see that there wasn’t an inherent anticipation argument. Perhaps there was such a rejection raised during prosecution which would account for the strange negative limitation in the claim, i.e., “wherein the hepatocytes are not plated between the first and second cryopreservations.” That limitation has no effect on the viability of the cells being frozen but it does remove the ordinary researchers who, in the interest of saving money and material, chooses to re-freeze any un-used, thawed cells. Surely such researchers are/were numerous in the prior art (perhaps some prior practioners of this method will step forward now that the case has been publicized).

    The only issue, then, is the one which appears to have been battled over by the experts at the district court level: what is the expected percent viability of cells after the previously frozen and purified cells are thawed a second time? The plaintiff’s expert put forward the following theory:

    Imagine a runner who finishes one marathon and then immediately begins a second marathon. One would not expect the runner to perform the second in the same time as the first. More likely, the runner would not even finish the second mara- thon. Similarly… one would expect lower cell viability and a greater loss of cells after the second cryopreservation than after the first, thus teaching away from multi-cryopreservation.

    In many years in the lab and many subsequent years working with scientists, I have never heard anyone compare freezing a cell with “running a marathon.” More importantly, this theory completely ignores a key step in the process: the selection of cells which survived the first round of freezing. It’s certainly true that on average the viability of any given cell is expected to decrease after multiple rounds of freezing and thawing. But if one selects first for cells which tolerate freeze-thawing (using the well-known density gradient step), it’s entirely reasonable to expect a greater proportion of such selected cells to survive one additional cycle of freeze/thawing. Indeed, this was the theory proposed by the defense experts.

    And here is where the majority’s opinion weakest point lies. According to the opinion, the district court did not find the defense expert’s testimony “credible.” After making that statement, Rader cites a long paragraph of cases supporting the view that district court’s opinions about witness credibility are Teh Most Important Things Evah (funny how those cases seem to have been brushed aside when Therasense was decided!). What’s completely missing from Rader’s opinion, however, was any fact-based explanation as to why the defense experts’ testimony was deemed to be “not credible.” All that is offered is handwaving about “hindsight” which, as noted by Judge Gajarsa, applies equally to the patentee’s “theory” as to why skilled artisans allegedly believed that re-freezing gradient-purified cells would necessarily result in a preparation with less than 70% viability.

    In fact, there is nothing new about the reasoning provided by the defense’s experts. If anything, the theory that selecting for “endurance” leads to diminished endurance is more wanting. Can someone who has read the district court decision or who was at the district court hearings let us know: what was the demeanor of the defense experts on the stand? Did they seem like they were lying? Were they caught contradicting themselves? These are the sort of facts necessary to support a district court finding of “lack of credibility.” If in fact the defense experts were clearly full of shxt based on the incomprehensibility of their statements or their demeanor, then the finding of “lack of credibility” has legs and Gajarsa is simply out of luck. But in the absence of such findings in the record, I think the preliminary injunction should not have issued.

  33. 1

    “”price erosion, damage to ongoing customer relationships, loss of customer goodwill (e.g., when an effort is later made to restore the original price”

    I often wonder how this evidence should be persuasive. What they’re asking the court to do is stop someone already eroding the price and that will already cause customers to be POed when they try to jack up the price. The part of the damage that is irreparable has already been done it would seem.

    Indeed, what might hurt their customer relations even more is for their competitor to simply tell the customers that they no longer provide the service due to their competitor engaging in what they believe to be the assertion of an invalid patent in order to jack up the prices and the courts upholding such thus diminishing the good reputation of the courts. Try that customer/voter relation damage on for size. That would be much more interesting imo.

    Loss of business opportunities is likewise a little suspect at least in so far as they’re talking about more than simple competition in the market entails.

Comments are closed.