Pharmacia, Inc. v. Frigitronics, Inc., 726 F. Supp. 876 (D. Mass. 1989)

James L. Sigel, Ropes & Gray, Boston, Mass., Hugh A. Chapin, Alan P. Bowes, Arthur D. Gray, Richard S. Gresalfi, Kenyon & Kenyon, New York City, for plaintiffs.

Marshall Tatun, David J. Brody, Stephen Kor Widett, Slater & Goldman, Boston, Mass., of counsel, Daniel Kane, Joseph T. Eisele, Gerald Levy, Kane, Dalsimer, Sullivan, Kurucz Levy, Eisele and Richard, New York City, Neil Jacobs, Hale & Dorr, Boston, Mass., for defendants.

KEETON, District Judge.

This is an action for patent infringement filed by Pharmacia, Inc. and Dr. Endre Balazs against Frigitronics, Inc., Precision-Cosmet, Inc. and Med-Chem Products, Inc. on June 20, 1984. The application for the patent at issue, United States Patent No. 4,141,973 (hereinafter "the '973 patent"), was originally filed on October 17, 1975 and covers ultrapure hyaluronic acid. In defense, defendant Med-Chem asserts inter alia that the patented product was "on sale" in the United States "more than one year prior to the date of the application for patent in the United States," and consequently, under 35 U.S.C. § 102(b), the patent which the defendants are alleged to have infringed is not valid.

Phase I of this trial, which was concerned solely with defendant Med-Chem's proffered "on-sale" defense, was held before the court from September 25 to September 28, 1989. Having heard and considered all of the evidence offered by both *877 parties, the court, pursuant to Fed.R.Civ.P. 52, makes the findings of fact and conclusions of law set out below. In addition, the court adopts as its findings of fact all of the parties' proposed findings of fact neither underlined nor bracketed in each party's critique of the opponent's proposed findings of fact (Docket Nos. 108 and 128).

In the early 1940's, Dr. Balazs, the owner of the '973 patent, began his research on the use of hyaluronic acid as a therapeutic agent. By 1958, it became apparent that hyaluronic acid could also be used in the eye. E. Balazs, M. Freeman, R. Klöti, G. Meyer-Schwickerath, F. Regnault, & D. Sweeney, Hyaluronic Acid and Replacement of Vitreous and Aqueous Humor, 10 Modern Problems of Ophthalmology 3, 3 (1972) (DTX-19 ("Defendants' Trial Exhibit")) (hereinafter "the Mod. Prob. article"). Through the 1960's and early 1970's, Drs. Balazs and David Swann (Chairman and Chief Executive Officer of defendant Med-Chem) were involved, first together and later separately, in the quest to isolate and purify hyaluronic acid.

Hyaluronic acid, a naturally-occurring, viscoelastic polymer composed of units of two sugars, glucuronic acid and N-acetyl glucosamine which is found in large quantities in rooster combs, human umbilical cords and other biological sources, is used extensively in veterinary medicine and eye surgery. In veterinary medicine, hyaluronic acid is used to improve the function of joints in traumatic arthritis by lubricating and protecting the internal surfaces of the joints. During eye surgery, hyaluronic acid protects sensitive cell layers and tissues (such as the corneal endothelium, iris and retina) from mechanical damage which might otherwise be caused by the surgery, helps to maintain a deep anterior chamber and helps to restore normal anatomical configuration to ocular tissues. Hyaluronic acid is inert, does not interfere with the healing process and normal function of eye tissues, and, most importantly for purposes of the '973 patent, does not cause inflammatory or immunological reactions in joints or in the eye.

Under the '973 patent, the noninflammatory nature of ultrapure hyaluronic acid is determined by a quantitative owl monkey eye test. For this test, a one percent solution of the hyaluronic acid's sodium salt, dissolved in a physiological buffer, is implanted in the eye of a Douroucoulis monkey (owl monkey) as a substitute for about one-half of the vitreous humor of the eye. The owl monkey is used because its eye structure is very similar to the structure of human eyes and is a sensitive reactor to inflammatory substances. The inflammation caused by the implanted hyaluronic acid can be measured by counting the number of white blood cells which pass from the blood stream into the vitreous and aqueous humor of the eye. If, forty-eight hours after the implantation, fewer than 200 white blood cells are found per mm3 of aqueous humor, then, under the '973 patent, the hyaluronic acid is deemed noninflammatory.

Before Dr. Balazs' invention of the quantitative owl monkey eye test in 1971, both Drs. Balazs and Swann used a qualitative owl monkey eye test to measure the degree of inflammation caused by hyaluronic acid. In the qualitative test, which also involved substituting a solution of hyaluronic acid for part of the vitreous humor of an owl monkey's eye, the inflammatory nature of the hyaluronic acid was determined not by extracting part of the aqueous humor and counting the number of white blood cells, but rather, by visually examining the eye with an ophthalmoscope or a slit-lamp and by estimating the degree of inflammation based upon the appearance of the eye. Because the presence of white blood cells in the eye is indicative of an inflammatory reaction, and because these cells tend to obscure the internal parts of the eye, the degree of inflammation could be estimated qualitatively, from the degree of interference with the appearance of the interior of the eye, and then ranked on a scale of 0 to 4+ or 5+. Although the calibration of the scale varied from clinician to clinician, the basic concept was the same. Dr. Ian Constable, who performed qualitative owl monkey eye tests for Med-Chem in the early 1970's, used a scale of 0 to 5+ where 0 *878 represented no reaction, 1+ represented a minimal reaction ("flare and cells"), and 5+ the most severe reaction ("gray reflex"). I. Constable & D. Swann, Biological Vitreous Substitutes, 88 Archives of Ophthalmology 544, 545 (1972) (DTX-174). Dr. Sheldon Buzney, who performed both qualitative and quantitative owl monkey eye tests for Med-Chem from 1983 to 1988, used a scale of 0 to 4+ where 0 represented no reaction and 4+ the most severe reaction ("dark reflex without obvious fundus details"). S. Buzney, Protocol for Assay of AMVISC 3 (Med-Chem Document, 1982) (DTX-207).

In 1968 and 1969, Dr. Balazs had produced what he believed at the time to be a noninflammatory hyaluronic acid. Accordingly, on October 28, 1968, Dr. Balazs filed a "Notice of Claimed Investigational Exemption for a New Drug," PTX-577 ("Plaintiffs' Trial Exhibit") (hereinafter "Balazs' IND Application"), which was approved on November 19, 1968. PTX-578. When Dr. Balazs later learned from clinicians that this hyaluronic acid was inflammatory, he resumed his search for a noninflammatory hyaluronic acid and continued his tests. By October 17, 1975, when he was satisfied that he had isolated and purified a noninflammatory hyaluronic acid, Dr. Balazs filed United States patent application serial no. 623,333 for "Ultrapure Hyaluronic Acid and the Use Thereof." On October 25, 1977, Dr. Balazs filed United States patent application serial no. 844,833, a continuation of his application serial no. 623,333. On February 27, 1979, this patent was issued as United States Patent No. 4,141,973. PTX-522.

35 U.S.C. § 102(b) provides:

A person shall be entitled to a patent unless ... the invention was ... in public use or on sale in this country, more than one year prior to the date of the application for patent in the United States....

Because Dr. Balazs first applied for the patent for ultrapure hyaluronic acid on October 17, 1975, the critical date for purposes of the on-sale defense is October 17, 1974. If defendant Med-Chem can show that anyonewhether it was Dr. Balazs, his company Biotrics, Pharmacia (the company to whom Dr. Balazs sold some of his patent rights), or Dr. Swann's company Med-Chemplaced the patented product on sale before October 17, 1974, then Dr. Balazs would not be entitled to his patent and would not be able to maintain this or any suit for infringement of the '973 patent.

As the Court of Appeals for the Federal Circuit has held, there need not be an actual sale to raise the "on-sale" bar to the validity of a patent; even a "single offer to sell is enough to bar patentability whether or not the offer is accepted." A.B. Chance Co. v. RTE Corp., 854 F.2d 1307, 1311 (Fed.Cir.1988). The offer to sell or sale, though, must be of the patented product. Thus, there must be an identity between the product offered for sale and the later-claimed invention in the sense that

the subject matter of the sale or offer to sell fully anticipated the claimed invention or would have rendered the claimed invention obvious by its addition to the prior art.... If the inventor had merely a conception or was working towards development of that conception, it can be said that there is not yet any `invention' which could be placed on sale. A sale made because the purchaser was participating in experimental testing creates no on-sale bar.

UMC Electronics Co. v. United States, 816 F.2d 647, 656-57 (Fed.Cir.1987), cert. denied, 484 U.S. 1025, 108 S. Ct. 748, 98 L. Ed. 2d 761 (1988).

For defendants to prevail on their on-sale defense, defendants must thus prove that the product offered for sale before the critical date was the same as the invention of the patent in the sense stated in UMC Electronics Co. To do so in this case, defendants must prove that the product sold or offered for sale met the 200 white blood cell limit of the quantitative owl monkey eye test.

Whether a product offered for sale is the same as the product of the patent can be difficult to discern. The Court of Appeals for the Federal Circuit "`has been careful to avoid erecting rigid standards for section *879 102(b).'" Id. at 653 (quoting Western Marine Electronics, Inc. v. Furuno Electric Co., 764 F.2d 840, 844 (Fed.Cir.1985)). Instead, the court has mandated a case-by-case approach, instructing the district court to consider "[a]ll of the circumstances surrounding the sale or offer to sell, including the stage of development of the invention and the nature of the invention," and to weigh these circumstances "against the policies underlying section 102(b)." Id. at 656. The specific factors relevant to this determination are discussed in further detail in Part III, infra.

In attacking the validity of a patent on the basis of § 102(b), the challenger bears the burden of proving the patent invalid: "A patent shall be presumed valid.... The burden of establishing invalidity of a patent ... shall rest on the party asserting such invalidity." 35 U.S.C. § 282. The Federal Circuit has determined that this burden is one of "`clear and convincing evidence,' ... an intermediate standard which lies somewhere between `beyond a reasonable doubt' and a `preponderance of the evidence.'" Buildex, Inc. v. Kason Industries, Inc., 849 F.2d 1461, 1463 (Fed.Cir.1988). Defendant Med-Chem thus bears the burden of proving by clear and convincing evidence that the ultrapure hyaluronic acid of the '973 patent was sold or offered for sale before October 17, 1974.

Defendants contend that they need only show by clear and convincing evidence that offers or sales of the later-claimed invention occurred before the critical date, and that once it makes this prima facie showing, the burden shifts to the plaintiffs to prove by clear and convincing evidence that each and every one of those offers or sales was experimental and not commercial. This is a misreading of the applicable case law. In Hycor Corp. v. Schlueter Co., 740 F.2d 1529 (Fed.Cir.1984), relied upon by defendants, the Federal Circuit held only that, once the challenger to a patent (here the defendants) makes a prima facie case of debilitating sales, the defender of the patent has "the burden of going forward with convincing evidence to counter [the challenger's] prima facie showing of invalidating public use [or prior sale]." Id. at 1537 (emphasis added). More recently, the court held:

without question, the challenger has the burden of proving that there was a definite sale or offer to sell more than one year before the application for the subject patent, and that the subject matter of the sale or offer to sell fully anticipated the claimed invention or would have rendered the claimed invention obvious by its addition to the prior art. If these facts are established, the patent owner is called upon to come forward with an explanation of the circumstances surrounding what would otherwise appear to be commercialization outside the grace period.

UMC Electronics Co., 816 F.2d at 656 (citations omitted) (emphasis added). Thus, the burden of production shifts to the defender of the patent, but the "ultimate burden of persuasion of patent invalidity rests solely with the party attacking validity.... `That burden is constant and never changes.'" Id. at 1536 (quoting American Hoist & Derrick Co. v. Sowa & Sons, 725 F.2d 1350, 1360 (Fed.Cir.), cert. denied, 469 U.S. 821, 105 S. Ct. 95, 83 L. Ed. 2d 41 (1984)).

That the burden of proof does not shift is further demonstrated by TP Laboratories, Inc. v. Professional Positioners, Inc., 724 F.2d 965, 971 (Fed.Cir.), cert. denied, 469 U.S. 826, 105 S. Ct. 108, 83 L. Ed. 2d 51 (1984) (discussing the "public use" bar of § 102(b)). In that case, the Federal Circuit held that § 102(b) does not raise two issueswas there a prior sale (or public use), and was the sale (or use) experimental but rather, just one: was the product of the invention on sale before the critical date under § 102(b)? If the product was still in its experimental stage, then it was not the product of the invention and was not on sale under § 102(b).

Med-Chem's first claim of prior sales relates to its own conduct during the period before October 17, 1974. Med-Chem claims that it had developed a form of ultrapure hyaluronic acid which induced fewer than 200 white blood cells in the *880 quantitative owl monkey eye test and which it had placed on sale in this country under the name HYVISC as early as November 4, 1971, almost three years before the critical date. There is no evidence before the court that the quantitative owl monkey test was used by Med-Chem on this product before the critical date (October 17, 1974). Rather, Med-Chem's contention is that at least one batch of HYVISC, from which sales or offers to sell were made before October 17, 1974, was a form of ultrapure hyaluronic acid that was capable of satisfying this quantitative test. One critical fact question to be resolved, therefore, is whether Med-Chem has proved this contention by clear and convincing evidence.

Between September 1970 and October 17, 1974, Med-Chem produced fifteen identifiable batches of hyaluronic acid. Of those fifteen, Dr. Swann testified that, in his opinion, Batch C, which was manufactured in 1971, was noninflammatory and met the claims of the '973 patent. At the time, Batch C was only tested with the qualitative owl monkey eye test. Med-Chem has offered evidence, however, attempting to show that Batch C would have met the 200 white blood cell limit in the quantitative owl monkey eye test had Med-Chem been using that test at the time.

After weighing all the evidence offered at trial, the court finds, for reasons explained more fully below, that Batch C was not a form of ultrapure hyaluronic acid capable of satisfying the 200 white blood cell limit of the quantitative owl monkey test. In these circumstances, any dispute over who bears the burden of proof and by what standard is moot. Also, in these circumstances, it is unnecessary for the court to determine whether Batch C was "on sale" before the critical date. Med-Chem's contention that sales of or offers to sell Batch C were debilitating sales or offers fails because Batch C did not meet the test specified in the '973 patent and was not the patented product.

The defendants offered three independent lines of evidence to show that Batch C met the claims of the hyaluronic acid covered by the '973 patent: (A) Batch C is the same as the hyaluronic acid which defendant Med-Chem is now selling (AMVISC) and which plaintiffs allege is infringing on the '973 patent because it is made by the same process; (B) the qualitative owl monkey eye tests done on Batch C in the early 1970's yielded a 1+ reaction which correlates at a 98% degree of certainty to under 200 white blood cells in the quantitative test; and (C) a recently found and tested syringe of Batch C produced a reaction of 0 white blood cells which places this hyaluronic acid well within the criteria of the '973 patent.

Defendants first argue that if AMVISC infringes the '973 patent as plaintiffs claim (because it induces fewer than 200 white blood cells in the quantitative owl monkey eye test), then Batch C, which was produced by the same manufacturing process as AMVISC, must also induce fewer than 200 white blood cells in the quantitative test. Accordingly, defendants argue, Batch C was the same product as the hyaluronic acid of the '973 patent.

Because this argument is premised on the identity between the manufacturing process for Batch C and that for AMVISC, defendants must show by clear and convincing evidence that these two processes were exactly the same, or that any differences between them would have no material effect on the resulting product. Defendants have not made this showing even by a preponderance of the evidence.

From 1970 to 1972, Mr. Bernard Sullivan, now Vice-President of Operations for Med-Chem, kept a lab notebook in which he recorded the steps which were performed in the manufacturing of the various hyaluronic acid batches. DTX-199 (hereinafter the "lab notebook"). From this lab notebook, Dr. Swann reconstructed the procedure that he used to make Batch C. A comparison of that process with Med-Chem's manufacturing process for AMVISC, which is set forth in Med-Chem's 1981 "Premarket Approval Application" for HYVISC, § XI (PTX-659 (the relevant *881 pages are also in evidence as DTX-200)), shows numerous differences. Some of these differences are highlighted below (see also chart comparing processes created by defendants, DTX-282, and copy of that chart with plaintiffs' notations, PTX-706):

(1) Stage I of Batch C began with the grinding of 30 pounds of wet rooster combs, and continued with three acetone extractions of hyaluronic acid. By contrast, batches of AMVISC are begun by grinding 30 pounds of frozen rooster combs which are then subjected to at least seven acetone extractions.

(2) Stage II of Batch C involved water extraction with a two-step filtration process, first through cheese cloth and then through a glass cinder with a small micron-size hole, at an unstated temperature. By contrast, AMVISC is subjected to water extraction in an ultraspeed centrifuge, first at a low-speed and then at highspeed, at 4 degrees centigrade. In addition, Batch C was treated with chloroform and AMVISC is treated with Thymol (a bacteriostatic agent which is less toxic than chloroform), although plaintiffs concede that this substitution would have no effect on the extraction process of Stage II. Plaintiffs' Critique of Defendant Med-Chem's Proposed Finding of Fact 206 (Docket No. 108).

(3) Stage III of Batch C involved stirring the mixture for an unspecified period of time at an unspecified temperature. By contrast, Stage III of AMVISC involves stirring the solution at 4 degrees centigrade for five to seven days.

(4) In Stage IV of Batch C, the hyaluronic acid was subjected to a pronase treatment with a "pinch" of calcium chloride; no instructions as to temperature or length of treatment were noted in the lab notebook. AMVISC, by contrast, is treated in pronase with "the correct amounts" of calcium chloride "(final concentration 0.004% W/V)" and then incubated at 37 degrees centigrade for 16 to 18 hours.

(5) Although the lab notebook is vague on the procedures followed at this stage, it appears that Batch C was not subjected to an ultracentrifuge at Stage V. The lab notebook does show, though, that the hyaluronic acid of Batch C was precipitated with ethanol. AMVISC, by contrast, is ultracentrifuged at this stage and is precipitated with denatured alcohol; chloroform is added as a bacteriostatic agent.

Drs. Swann and Eugene Davidson, Med-Chem's expert witness, testified that these differences, which merely take advantage of advances in technology, have no significant effect on the resulting hyaluronic acid: the two procedures were not different but rather equivalent. See, e.g., Swann test., 9/26/89 morn. sess., p. 6 ("In those days we had filtration as the clarification step [rather than the ultraspeed centrifugation of Stage II].... [W]e use the modern equivalent today"). In light of the many obvious differences between the manufacturing processes for Batch C and AMVISC, however, this testimony is not persuasive. The technological advances that the AMVISC process uses are likely to produce a better, purer, less inflammatory hyaluronic acid; that is one of the reasons they are technological advances. Furthermore, as discussed in Part II(B), infra, Med-Chem has failed to show that Batch C and AMVISC are the same. Because the court finds that the two products are different in the degree of inflammation that they induce, it is not credible for Med-Chem to suggest that the differences in the processes used to manufacture the two products are inconsequential and do not affect the final product.

The court thus finds that Med-Chem has failed to prove even by a preponderance of the evidence (and for stronger reasons has failed to prove by clear and convincing evidence) that the process it used to manufacture Batch C is the same as the process that it currently uses to manufacture AMVISC, the allegedly infringing hyaluronic acid. Nor, the court finds, is the current process an equivalent.

Med-Chem's second line of argument focuses on the inflammatory nature of the two products. Med-Chem asserts that Batch C received a reaction of 1+ in the *882 qualitative owl monkey eye test and that, although Med-Chem did not use the quantitative test in 1971, Batch C would have induced fewer than 200 white blood cells per mm3 of aqueous humor in the quantitative owl monkey eye test had Med-Chem been utilizing the quantitative test at the time. In support of this claim, Med-Chem offered expert testimony that a rating of 1+ in the qualitative test corresponds 98% of the time to fewer than 200 white blood cells in the quantitative test.

The court finds that Med-Chem's claim is not supported by a preponderance of the evidence, let alone by clear and convincing evidence. First, the evidence does not prove that Batch C would consistently rate a 1+ in the qualitative test. Second, Med-Chem's expert witness failed to show a reliable correlation between a rating of 1+ in the qualitative test and a rating of fewer than 200 white blood cells in the quantitative test.

In support of its claim that Batch C was noninflammatory and rated at 1+ when subjected to the qualitative owl monkey eye test, Med-Chem offered the November 11, 1971 entry from the lab notebook in which Mr. Sullivan recorded the contents of a telephone call from Dr. Constable. Dr. Constable, who was performing the qualitative owl monkey eye tests on Med-Chem's hyaluronic acid in the early 1970's, reported that sample 79 of Batch C, as Mr. Sullivan recorded it, was rated at "no rxvery good." Dr. Swann testified that this report was equivalent to a rating of 1+ or better on the qualitative test scale. Other than this record of Dr. Constable's phone call, defendants have no evidence, written or oral, contemporaneous or otherwise, of this test on sample 79.

As opposed to this single, somewhat vague datum, the evidence before the court is that other qualitative owl monkey eye tests performed on samples of hyaluronic acid from Batch C yielded disparate results:

(1) The lab notebook entry of September 3, 1971 shows that samples 69 and 70 from Batch C yielded a 4+ reaction.

(2) The lab notebook entry of February 17, 1982 shows that samples of hyaluronic acid from Batch C, which were packaged for horses, yielded a result of 4+. Dr. Swann "surmise[d] that the ["very severe"] 4+ grade was obtained because the test syringe was contaminated during packaging or the eye in which it was tested was very damaged." Swann Affdvt., ¶ 9, p. 11 (Docket No. 119). His speculations, however, are not supported by any contemporaneous records.

(3) A study submitted to the United States Food and Drug Administration ("FDA") in June 1973 by The Retina Foundation, PTX-555 (hereinafter "The Retina Foundation report"), reported that sample 79, which Dr. Constable rated at "no rx very good," was rated at 2+, 2+, and 5+. Although Dr. Swann testified that this sample 79 was not the same as the sample 79 on which Dr. Constable reported, his testimony on this point is not persuasive. Dr. Swann offered no explanation for the use of the same number and did not raise the same objection to the other numbered samples in The Retina Foundation report which appear to correlate well with the sample numbers recorded in the lab notebook.

(4) The Retina Foundation report also shows that Batch C sample 75 was tested at 2+. Sample 75 was referred to as the "best yet" in an October 5, 1971 entry in the lab notebook.

(5) The Retina Foundation report also shows that Batch C sample 235, packaged and tested on June 7, 1972 according to the lab notebook, tested at 2+, 2+, and 3+. Dr. Swann again speculated that these reactions were caused by the later packaging rather than by the hyaluronic acid. Swann test., 9/25/89 morn. sess., p. 78.

(6) The Retina Foundation report also shows that Batch C sample 233 was tested at 2+. Contrary to Dr. Swann's assertion that some of the severe reactions were due to contamination during "later packaging," sample 233 was packaged at the same time as sample 79 which Dr. Constable allegedly reported resulted in no reaction. Lab notebook at June 7, 1972. Considering all the *883 evidence, the court finds that Dr. Swann's speculation that these reactions were caused by later packaging is without merit.

(7) Finally, Med-Chem's 1972 "Notice of Claimed Investigational Exemption for a New Drug" (PTX-556) (hereinafter "Med-Chem's IND Application"), submitted in support of HYVISC, reports a rating of "2+ to 3+ for the reaction in the eye injected with `HYVISC.'"

Defendants have thus failed to prove, even by a preponderance of the evidence, that Batch C would have tested at 1+ on the qualitative scale. Indeed, the court finds that the data submitted to the FDA in The Retina Foundation report and in Med-Chem's IND Application, showing a range of reactions from 2+ to 5+, more reliably indicate Batch C's inflammatory nature than the single "no rx" notation made in the lab notebook by Mr. Sullivan based upon a single telephone call from Dr. Constable.

Moreover, even if it were assumed that Batch C rated at 1+ on the qualitative scale, defendants have not shown by clear and convincing evidence that a rating of 1+ or "no rx" correlates to fewer than 200 white blood cells in the quantitative test.

Defendants offered the testimony of Dr. Buzney for the proposition that a qualitative rating of 1+ or better correlates with 98.3% certainty to fewer than 200 white blood cells in the quantitative test. Dr. Buzney's study, however, was contaminated by unsupported assumptions and arbitrarily chosen data.

Between October 1983 and November 1988, while serving as director of a research program at The Retina Foundation, Dr. Buzney conducted over 420 owl monkey eye tests ("assays") on Med-Chem's AMVISC hyaluronic acid. Significantly, Dr. Buzney used both the qualitative and quantitative methods of owl monkey eye testing in his assays, although, for the quantitative test, Dr. Buzney measured the number of cells in the vitreous humor, as per Med-Chem's standard policy, rather than the number of cells in the aqueous humor, as is mandated by the '973 patent. Dr. Buzney selected 290 of these assays, all of which rated at 1+ or better on the qualitative test, to determine the correlation between a qualitative test report of 1+ and the number of white blood cells induced per mm3 of vitreous humor in the quantitative owl monkey eye test. He testified that 285 of the 290 assays had fewer than 200 white blood cells per mm3 of vitreous humor; that only 5 of the 290 tests which had a qualitative reaction of 1+ also had a quantitative result of over 200 white blood cells. Thus, Dr. Buzney concluded, a qualitative owl monkey eye test result of 1+ correlates in 98.3% of the cases to a quantitative owl monkey eye test result of fewer than 200 white blood cells per mm3 of vitreous humor.

An examination of Dr. Buzney's underlying data (see PTX-708 through PTX-711) demonstrates that his study does not reflect an accurate correlation between a 1+ qualitative rating (which, Dr. Swann asserts, is the equivalent of Dr. Constable's report of "no rx" on sample 79) and the number of white blood cells in the aqueous humor. First, Dr. Buzney's data included a large number of 0, "clear," and "crystal clear" ratings which represent reactions that are less serious than 1+ reactions. A 0 rating represents an eye with no reaction and no inflammation, whereas a 1+ rating represents an eye with a very minimal inflammation. Because the number of white blood cells in the vitreous and aqueous humor increases as the severity of the inflammation increases, an eye that tested at 0 will naturally tend to have fewer white blood cells in the vitreous humor than an eye that tested at 1+. The inclusion of the 0 reactions as well as the 1+ reactions in the study biased the results in favor of a high correlation. To avoid this bias (that is, to isolate the correlation between the number of white blood cells and a 1+ qualitative reaction), one would need to consider the results of examining only 1+ ratings and not ratings of 0, which would tend to have fewer white blood cells per mm3 of vitreous humor and thus be better than 1+.

Second, Dr. Buzney excluded from his study several reactions of 1+, listed in *884 PTX-711, which, had they been included, would have increased the number of white blood cells in the correlation. Indeed, of the eight 1+ reactions documented in PTX-711, none was included in his chosen 290 and none had fewer than 200 white blood cells in the vitreous humor; one of these assays rated at 3200 white blood cells.

Finally, as noted above, Dr. Buzney's assays were tests of the vitreous humor rather than of the aqueous humor. Defendants offered no evidence of the correlation between the number of white blood cells per mm3 of aqueous humor and the number of white blood cells per mm3 of vitreous humor. Given that Dr. Buzney's underlying data failed to include only 1+ reactions and failed to include all of the 1+ reactions, and given defendants' failure to offer any evidence on the correlation between the number of white blood cells in the aqueous humor and the number of white blood cells in the vitreous humor, this evidence is unpersuasive as proof of the claimed correlation.

The conclusion that defendants' suggested correlation is not reliable is reinforced by evidence of Dr. Balazs' experiences. Dr. Balazs testified that he produced some batches of hyaluronic acid by the late 1960's, particularly batches 1020, 1030, 1070, 1080, 1090, 1100A, and 1100B, which rated at "clear" or "crystal clear" in the qualitative eye test, a rating that he believed at the time to mean that the hyaluronic acid in question was noninflammatory. Based on these encouraging results, Dr. Balazs submitted his 1968 IND Application, published the Mod. Prob. article in 1972 discussing these findings, and sent samples of his hyaluronic acid to clinicians for human testing.

Shortly thereafter, Dr. Balazs testified, the clinicians began to report back to him that these batches of hyaluronic acid, while helpful in hopeless cases of retinal detachments, were in fact inflammatory. See, e.g., Dr. Balazs' letters to Drs. François Regnault and Gerd Meyer-Schwickerath, PTX-703 and PTX-596, Dr. Balazs' memo regarding his conversation with Dr. Rudolf Klöti, PTX-618, and Dr. Lorimer Fison's letter to Dr. Balazs, PTX-619. Dr. Balazs thus determined that the qualitative test was not a good indicator of the inflammatory nature of the hyaluronic acid and that a better test was needed. By April 1971, in conjunction with Dr. Melvin Freeman, Dr. Balazs developed the quantitative owl monkey eye test discussed in the '973 patent.

Dr. Balazs' first quantitative test, performed on batch 1100, yielded a reaction of 300 white blood cells. Subsequent tests on batches of hyaluronic acid which were made by the same process as batches 1020, 1030, 1070, 1080, 1090, 1100A and 1100B all yielded reactions of over 200 white blood cells, despite the fact that the earlier batches had rated at clear and crystal clear in the qualitative test. Dr. Freeman reached the same conclusion: "[e]ven though some of these batches tested `clear' or `crystal clear' in the qualitative owl monkey eye test, they nevertheless caused cell counts above 200 cells per mm3 when tested in the cell count [quantitative] owl monkey eye test." Freeman Test., Tab 1, ¶ 6, pp. 2-3 (Docket No. 104). Dr. Freeman's deposition testimony is corroborated by his contemporaneous notes of his conversation with Dr. Balazs on September 1, 1971, PTX-592, in which he reported that Batch 1146 had "1+ flare and rare cells" in the qualitative test, but yielded 268 white blood cells in the quantitative test.

Thus, even if Med-Chem had proved that Batch C rated at 1+ in the qualitative owl monkey eye test (which the court finds it did not), Med-Chem would still fail to prevail on this claim because it has not proved by clear and convincing evidence that a 1+ rating would correlate with a rating of fewer than 200 white blood cells per mm3 of aqueous humor in the quantitative owl monkey eye test. Accordingly, the court finds that Med-Chem has failed to prove by clear and convincing evidence that Batch C was the same product as the hyaluronic acid of the '973 patent. Indeed, quite apart from the "clear and convincing standard," the court finds that in fact Batch C was not the same product as the hyaluronic acid of the '973 patent.

Med-Chem's final attempt to prove that Batch C was the same noninflammatory hyaluronic acid as that covered by the '973 patent concerns its recent discovery and testing of a syringe of material alleged to be Batch C hyaluronic acid found in Dr. Swann's basement. Although the court admitted evidence of this newly discovered syringe, the court finds that it is worthy of very little weight and probative value. The syringe was found in an unsealed envelope in a basement under an outdoor porch where it had been subjected to variable temperature and humidity over 18 years. The material in the syringe had dried out and had to be reconstituted with saline solution; because only the water and not the salt had evaporated, the resulting solution contained more salt that the original Batch C hyaluronic acid. The defendants have not shown by clear and convincing evidence that the material in the syringe was Batch C hyaluronic acid, nor have defendants shown that hyaluronic acid made in 1971 and stored in a basement for 18 years, under variable climatic conditions, and subsequently reconstituted with additional saline solution, would retain all of its characteristics as hyaluronic acid. Accordingly, any tests performed on this newly discovered material are highly suspect and do not offer clear and convincing evidence of any sort.

In summary, for the reasons stated in Parts II(A), II(B), and II(C), supra, the court finds that Med-Chem has not proved in any sense, and much less by clear and convincing evidence, that it had any hyaluronic acid before October 17, 1974 which met the criteria of the '973 patent, and consequently, has not proved that the hyaluronic acid of the '973 patent was on sale before October 17, 1974.

Defendants claim that the '973 patent is also invalidated by Biotrics' own sales of ultrapure hyaluronic acid before to the critical date. Defendants support their claim with evidence of Biotrics' sales of, and offers to sell, hyaluronic acid to Drs. Freeman, Clare Kok-van Alphen, Wallace Foulds, and A.W. Kersjes, Pharmacia, Propan Ethicals, Ltd., the Boston Biomedical Research Institute, Miles Laboratories, Lindeburg Farma (for Dr. Edmund), the Department of Animal Genetics at the University of Connecticut, the Department of Ophthalmology at the University of Glasgow, and the Netherlands Government Purchasing Office (for "Afd. Ooogheelkunde van het Academisch Ziekenhuis"). See Defendant Med-Chem's Prehearing Memorandum, ex. B (Docket No. 117) for the relevant trial exhibit numbers. For defendants to show a prima facie case regarding Biotrics' prior sales, defendants must first show by clear and convincing evidence that Biotrics had available to sell the product of the invention.

The seminal law in this area was articulated by the Supreme Court in Elizabeth v. Pavement Co., 97 U.S. 126, 24 L. Ed. 1000 (1877). In that case, Samuel Nicholson, who invented a new method of road pavement using a checker-board pattern of wooden blocks, was prosecuting infringers of his patent. In defense, the defendants claimed that the pavement in question had been in public use on a Boston street for six years, thus rendering the patent invalid under the predecessor to 35 U.S.C. § 102(b). Although the "discussion [in Elizabeth] is addressed to public uses, not sales, ... the underlying principle is the same." In re Hamilton, 882 F.2d 1576, 1580-81 (Fed.Cir.1989) (available on Westlaw, CTAF database, at screen 15 of 23).

The Supreme Court rejected the public use defense, concluding that the paved street in Boston was an experiment and that the invention was not completed, and not subject to the public use bar, until after the experimentation was completeduntil after six years of traffic had tested the longevity of the new paving technique. "The use of an invention by the inventor himself, or of any other person under his direction, by way of experiment, and in order to bring the invention to perfection, has never been regarded as [a public] use." Elizabeth, 97 U.S. at 134. The use was deemed experimental, and thus not subject to the public use bar, despite the fact that *886 Nicholson made no changes in the pavement design after his extensive and lengthy experiment:

such use is not a public use, within the meaning of the statute, so long as the inventor is engaged, in good faith, in testing its operation. He may see cause to alter it and improve it, or not. His experiments will reveal the fact whether any and what alterations may be necessary. If durability is one of the qualities to be attained, a long period, perhaps years, may be necessary to enable the inventor to discover whether his purpose is accomplished. And though, during all that period, he may not find that any changes are necessary, yet he may be justly said to be using his [invention] only by way of experiment; and no one would say that such a use, pursued with a bona fide intent of testing the qualities of the [invention], would be a public use, within the meaning of the statute. So long as he does not voluntarily allow others to make it and use it, and so long as it is not on sale for general use, he keeps the invention under his own control, and does not lose his title to a patent.

Id. at 135.

This conclusion remains as valid today as it was over a century ago and serves as a cornerstone of the public policy underlying 35 U.S.C. § 102(b): to encourage "prompt and widespread disclosure of inventions to the public, while giving the inventor a reasonable amount of time (1 year, by statute) [from the completion of the invention] to determine whether a patent is worthwhile." Western Marine Electronics, 764 F.2d at 845. If further experimenting is required to ensure that the product will suit its intended purposes, then the product is still in its experimental stages and any sales or public uses of the product are by definition noncommercial and not subject to the § 102(b) bar.

As noted in Part I, infra, the Federal Circuit prefers a case-by-case analysis to the former, rigid, "reduced to practice" test. UMC Electronics Co., 816 F.2d at 656. In making this analysis, the court must consider the totality of the circumstances, "including the stage of development of the invention," id. at 656, and in particular, the following factors:

the necessity for public testing, the amount of control retained over the operation, the extent of public testing in relation to the nature of the invention, the length of the test period, whether any payment was made, whether there was a secrecy obligation, whether progress records were kept, who conducted the experiments, and the degree of commercial exploitation during the test in relation to the purpose of the experimentation.

Baker Oil Tools, Inc. v. Geo Vann, Inc., 828 F.2d 1558, 1564 (Fed.Cir.1987) (citations omitted).

The evidence demonstrates that Dr. Balazs was still experimenting with his hyaluronic acid during the period of the offers and sales that defendants assert were debilitating under § 102(b). In 1968 and 1969, Dr. Balazs thought that he had isolated a noninflammatory hyaluronic acid, but he subsequently learned that the hyaluronic acid was inflammatory. He continued to test different production methods until, in 1975, he was satisfied with batch C2114 and applied for the '973 patent.

That the sales before the critical date were not sales of the completed product is true even of the March 22, 1974 sale to Pharmacia of hyaluronic acid from lot C2114 (DTX-151), the very lot of hyaluronic acid which later formed the basis for Dr. Balazs' patent application. Because Dr. Balazs and Pharmacia were still experimenting to determine if the hyaluronic acid was noninflammatory, the court finds that this lot of hyaluronic acid was not the product of the '973 patent until the experiments were complete. Thus, in March 1974, C2114 was just an experimental batch of hyaluronic acid, like Nicholson's checker-board pavement during its six years of traffic-testing. In the face of the fact finding, which the court makes, that Dr. Balazs did not know at the time that lot C2114 was satisfactorily noninflammatory, defendants effort to retrospectively assert that Dr. *887 Balazs had the claimed invention in March 1974 fails. Baker Oil Tools, 828 F.2d at 1563.

Applying the Baker Oil Tools standard, the court finds that the hyaluronic acid sold before October 17, 1974 was not the product of the '973 patent. There was, for example, a clear need for extensive testing, both in animals and in human clinical trials, before the hyaluronic acid could be made available for use in human eye surgery. Dr. Balazs testified that he did not have the resources needed for this extensive testing, so he elicited the help of Pharmacia. Even as Pharmacia was conducting the tests, though, Dr. Balazs retained a measure of control over the experiments as Pharmacia reported back to him with results and questions.

Another relevant factor concerns the issue of payments. As the court held in Baker Oil Tools, for a sale to be experimental, it must be "substantially for the purpose of experiment" rather than "mainly for the purpose of trade or profit." 828 F.2d at 1563. Here, plaintiffs have offered ample evidence that Biotrics sold all of the hyaluronic acid in the period before the critical date at a "production cost" price, rather than for profit, and that there was no "commercial exploitation [of the hyaluronic acid] during the test" period. "Production cost" is an ambiguous phrase because of lack of specificity regarding which elements of cost, among all those that would be used in cost accounting for an ongoing commercial operation, are to be taken into account. The evidence in this case makes clear that the pricing of the hyaluronic acid sold by Biotrics before October 17, 1974 did not take into account major elements of cost accounting. For example, developmental research costs were not included as a factor in the price, nor was the reasonable value of the services of Dr. Balazs and his assistants in discovering the ultrapure hyaluronic acid and in negotiating and effectuating these sales. The court thus finds that the sales that occurred were "for the purpose of experiments."

Yet another factor weighing in favor of finding that the hyaluronic acid was still experimental and not yet the product of the invention is the secrecy provision of the contract between Biotrics and Pharmacia. DTX-64, p. 11. Another factor is the fact that many of the letters between Biotrics and Pharmacia refer to clinical trials, tests, or experiments. PTX-625, PTX-627-638, PTX-640-641, PTX-643-645, PTX-647, PTX-649, PTX-653-656. Indeed, one of these letters which refers to clinical trials, PTX-656, was written on October 2, 1974, just days before the critical date. This documentary evidence is corroborated by the testimony of Mrs. Marianne Granat, Pharmacia's project manager for the hyaluronic acid project: "Through the balance of 1974 and into 1975 we continued our trials and experimental work." Granat Affdvt., ¶ 27, p. 12 (Docket No. 101). Indeed, as late as August 1975, well after the critical date, Dr. Balazs was writing to Mrs. Granat with information on further testing of hyaluronic acid (DTX-23).

Against all of these factors, defendants argued that a footnote in the Mod. Prob. article, which states that noninflammatory hyaluronic acid is "available commercially from Biotrics," proves that the hyaluronic acid of the patent was on sale before the critical date. This argument is without merit. The court finds that Dr. Balazs used the word "commercial" in the sense that samples were available at a modest fee, but would not be provided free of charge; he did not use the term "commercial" in the sense of "for profit" or in the way that the term "commercial" is used for purposes of the on-sale bar of § 102(b). This interpretation of the footnote is supported by Dr. Balazs' contemporaneous letters to clinicians in which he stated that the hyaluronic acid is "not on the market" (PTX-601) and is "an experimental drug [which is available for] clinical trials" (PTX-614).

Defendants also argue, relying on Western Marine, that the term "nonrestricted chemical," found on some of the invoices sent with the hyaluronic acid (e.g., DTX-114, DTX-127, DTX-135, DTX-138-140, DTX-223, DTX-225, DTX-230, DTX-234), *888 demonstrates that the sales were unrestricted commercial sales. This argument, too, is without merit. As Dr. Balazs testified, the term "nonrestricted chemical" did not mean that the hyaluronic acid was nonrestricted for commercial sales purposes, but rather, was not restricted by United States Customs for export purposes.

In sum, based upon the totality of the circumstances, the court finds that the hyaluronic acid of the '973 patent was not on sale before the critical date. Rather, the evidence demonstrates that the hyaluronic acid was still in its experimental stage for a period of time beyond the critical date.

On the foregoing findings, and for the reasons stated, it is ORDERED:

(1) Defendants' "on-sale" defense is without merit.

(2) A conference will be held on October 23, 1989 at 3:30 p.m. to schedule proceedings regarding plaintiffs' pending motion for preliminary injunction (Docket No. 81) and any other related issues.