*1 “saving the tion that involves state half
computer as the second system,” Board step 5 requires.
claim Because the however, Allen, it considering
erred question. conclude
never reached this We Clouding argued, properly
that EMC dispute,
did not combination condition, Allen
Barnett and teaches likely problem [to
“when solution system]
computer cannot determined.”
Thus, secondary no claim of there is since
considerations, question remain-
ing for the Barnett and Board whether
Allen, combination, considered
suggest person skill ordinary “saving sys- the computer state tem,” We remand the Board answer question. AND REMANDED
VACATED
Costs parties shall costs. bear own their INC., formerly
NANTKWEST,
Conkwest, Inc., Plaintiff-
Appellant
v. LEE, Director,
Michelle K. Patent U.S. Deputy Office, Under
and Trademark
Secretary of Commerce Intellectu Property Deputy
al Director Patent and Trade United States Office, Defendant-Appellee
mark
2015-2095 of Appeals, Court
United States
Federal Circuit. 3,May
Decided:
Krause, Sarah E. Craven, Scott Weiden- FELLER. Prost,
Before Judge, Dyk, Chief and Stoll, Judges. Circuit Opinion by for the court filed Circuit Judge Prost, Dyk, Judge, Chief which joins.
Dissenting Opinion by filed Circuit Judge Stoll.
Dyk, Judge. Circuit The United States Patent and Trade- (“USPTO”) rejected Office mark claims 26, and 27 of Patent Application U.S. No. (“’955 10/008,955 patent application”) ground the claims would have been sought obvious. review court, pursuant district § 35 U.S.C. asserting and claims application were The nonobvious. district granted court the USPTO’s motion summary judgment of obviousness. Nant- appeals. Kwest We affirm.
Background assignee patent NantKwest is the of the issue, application by filed Hans Klin- gemann (“patent applicant”), directed to specific the use type immune cells for treating cancer. system
The immune can be into divided adaptive its innate responses. The in- nate response—which immune first of defense—comprises line immune cells (“NK”) like rapidly natural killer cells LLP, J. Irell & Manella Heinrich, Alan anything they foreign. attack sense as CA, Angeles, argued for plaintiff-ap- Los generally target- NK cells have limited Morgan pellant. represented by Also Chu, recognition specificity attack rather Gary Frischling, Drake; N. Lauren Nicole indiscriminately. adaptive The immune re- Beach, Newport CA. Haberny, Sandra sponse—which is the line of second de- Mary Solicitor, L. fense—comprises Office T immune like cells cells Kelly, specific antigens United States Patent and Trademark Of- foreign attack fice, Alexandria, VA, argued recognize. for defendant- been trained to represented appellee. adaptive response Also Thomas W. immune is thus slower target-specific. appeal dependent more cells and T J.A. 5. Also on but NK two claims. 26 teaches that proteins cell Claim meth- “[t]he have different surface treating od of a cancer target recep- described claim respond to certain wherein the route administration of differently. application tors *3 to is the cells the mammal intravenous and particular the use of a cell here concerns Id. the mammal human.” Claim 27 of cells—NK-92. line1 NK of- treating teaches that method a “[t]he Despite these differences between NK in claim 20 cancer described further com- cells, T throughout cells and the 1980s and prising step of administering to said 1990s, taught various references art cytokine promotes mammal a T capa- that both cells and NK cells were line.” Id. growth of cell said NK-92 lysing (destroying) of cells. ble cancer rejected The Examiner USPTO vitro, These references described ex claims at issue and found that “it would vivo vivo and experiments demonstrat- prima have been to person obvious a facie ability. By T ing this NK cells and ordinary in the art ... in April skill cells types were the of immune two to teachings combine the of Santoli recognize “to lyse cells known and tumor Gong and to arrive at the claimed method cells in vivo mammals.” J.A. 779-80. Gong because ... teaches specific prior Two are references cells, lyse cells to tumor while First, No. involved here. U.S. Patent teaches in vivo use of cytotoxic cell lines.” 5,272,082, (“Santoli”), by Santoli et al. (internal omitted). quotation J.A. 8 marks cells, taught specific that a cell line of T applicant appealed The then to the used TALL-104, can treat Appeals Board Patent and Interferences Second, Gong, Maki, Klingem- cancer. and (“Board”). The Board affirmed the Exam- (the applicant) published ann ’955 patent rejection ground iner’s on the that a per- study (“Gong”) taught specific that a ordinary son skill the art “would have NK-92, line, lyse NK can cell cells replace been the TALL-104 motivated efficacy. high question with cells Santoli’s method with NK-92 cells here is whether these references rendered Gong’s on disclosure that NK-92 application’s the ’955 claims obvious. spontaneously cells kill lym- and [leukemia 7, 2001, patent On appli- December phoma high efficiency.” cells with cancer] priority (internal omitted). cation was filed with date quotation J.A. 10 marks April independent 1997. Claim an § to 35 Pursuant U.S.C. NantKwest appeal here, provides claim on court, complaint then filed district A a cancer treating 20, 26, seeking judgment method that claims and comprising step mammal admin- application were nonobvi- istering to the mammal a com- medium ous. for summary The USPTO moved prising Depos- cell judgment. response, an NK-92 line ATCC argued NantKwest CRL-2407, it No. disputes wherein said cancer is this case involves factual recognized by lysed summary said issues that cannot be on resolved judgment, line. relying expert reports (and cancer) laboratory 1. nonstop Immune cells harvested in the hence cause be re- can lines,” from "cell which refer to derived patient moved from a nourished reproducing cancerous cells that continue laboratory reproduce more T or NK cells type. example, more of their own cell For subsequent experimental cells for use. produce tumor cells T or NK cells (“Miller”) success, pectation appeared and new references the court Dr. Miller § “one or proceeding. meaning consider “cancer” as submitted for the agree more cancer cells.” 22. granted summary judgment court J.A. We district that this is an incorrect con- genuine no material fac- NantKwest “because there is struction of “cancer.” The con- correct dispute tual as whether the invention “treating of the claim [patent application] ob- struction term in the claimed cells, art, “require[s] lysis many cancer” over the both vious found accomplish goal treating order the Examiner and the J.A. 15. Board.” cancer,” not merely lysing one or a jurisdic- appeals. We have few cancer cells. J.A. 722. 1295(a)(1). § tion under 28 U.S.C.
However, the court’s erroneous district *4 claim construction creates no basis for re- Discussion First, the district court’s versal. we review addressing de novo. In decision the issue This court reviews the district court’s obviousness, we will use the correct summary grant judgment or denial of de construction, which renders the district MicroStrategy Objects, novo. Inc. v. Bus. court’s erroneous construction harmless S.A., (Fed. 2005) 429 1349 F.3d Cir. Second, is no error. there assertion here (citations omitted). Summary judgment art that the relevant references only “if may granted the movant shows taught cancer lysed methods that one genuine dispute any that there is no as to lysed or insufficient num- cell otherwise 56(a). P. material fact.” Fed. R. Civ. Claim treating of cells cancer. bers an construction is issue law that we where, here, de novo is no review as there Ill relevant Teva Pharm. extrinsic evidence. —
USA, Sandoz, Inc., U.S. -, Inc. v. — - 831, 841, 135 L.Ed.2d S.Ct. § Under 35 U.S.C. (2015). applicant dissatisfied the deci- [a]n .., patent may A if “a sion of the [Board] is obvious skilled artisan Di- remedy by to action the have been motivated combine the civil would in court teachings court].... art rector [district the references to invention, may adjudge applicant such is enti- and that the that achieve claimed invention, for his artisan tled receive a skilled would have had reason of his expectation doing specified any able claims success so.” involved [Board], Procter Teva of the as the & Gamble Co. v. Pharm. the decision USA, Inc., may appear.... 566 994 F.3d facts the case (internal 2009) omitted). quotation marks proceedings, §In court “the district may presented consider new evidence”
II not applicant that was before the Kappos Hyatt, contends that v. we Board. U.S. (2012). 1690, 1696, 182 If grant summary judg should reverse the L.Ed.2d 704 S.Ct. fact, genuine ment because the district an issues of court used there are material Initially, claim “the must make de novo incorrect construction. district court findings “cancer” to factual that take account of district court construed mean a both “plurality multiple or cancer cells.” J.A. 16. the new evidence and the administrative However, addressing the reasonable the PTO.” Id. at 1701. ex- record before agree with the district court utilized clinical trials in
We been several is no dispute that there material 140. In patients.” separate Gong combination Santoli here used Klingemann et publication, al. cited produced persons invention and lyse fact that cells ... can because “NK-92 in the ordinary skill have been vitro,” the authors wanted [tumor cells] Gong. to combine motivated Santoli and NK-92 cells for ex suitability “[t]o test vivo purging.”3 appel- J.A. 344. While important an Santoli was advance be- these experiments lant correct it showed that T cells from a cell cause vivo allogeneic therapy, different line, belonging to a patient, can be patient’s they used the own NK patient produce into a administered cells, indisputably indicate skilled vivo therapeutic approach effects. Such an pursue artisans were motivated clinical “allogeneic” “adoptive or is called immu- applications for NK cells and the NK-92 specifically notherapy.” taught that cell line. cell line can be used in adoptive immunotherapy lysing Indeed, patent application the ’955 itself cancer cells.2 Gong superior referred describe . Gong taught cell line See, qualities of the NK-92 line. *5 showed in efficacy lysing in ¶ e.g., (“The patent ’955' application, 50 fact, In was to have cells. found by NK-92 cell line has been described “high efficiency” lysing in the same leu- ¶ (1994)”); id. at Gong (“NK-92 et al. 74 type kemia tumor cell as TALL-104 did. (1994)) (Gong cells from et al. were derived 128, 142. Both See J.A. TALL-104 and patient suffering from' a cells obtained lysed cancer cells via the same non-Hodgkin’s lymphoma.”). pat- from mechanism, i.e., in a “non-MHC-restrict- application highlighted ent “supe- NK-92’s manner, in require ed” which do not (as rior” in vitro efficacy well as the later- presentation antigens on the MHC cell determined in efficacy), compared proteins of the J.A. surface tumor cells. cells, against other immune as “activities 131,145. unexpected by a [that] are ... worker Id. at cytotherapy.” the of tumor publications by Prior art field patent ap- ¶ 121. patent application also plicant himself that there teaches indicated was that NK-92’s in vitro to applications “superi- motivation seek clinical for or example, Gong line. the NK-92 cell For those activities manifested ability preparations cytolytic noted that of their known nor- cells “[b]ecause cells, lyse malignant humans,” ... mally present suggests NK which reject argument alleged post-filing therapy 2. We NantKwest's that San- of TALL-104 failure toli does not "disclose a successful known human clinical trials—not at the 32, therapy Appellant for TALL-104.” A Br. application—is time of the See irrelevant. In study therapy dogs TALL-104 in vivo Vaeck, 1991) re F.2d eight responses in showed clinical out of nine ("[Expectation of success must founded in dogs study Another teen showed tested. added)). (emphasis art.” mice treated with TALL-104 cells remained months, cancer-free for at least 2 while un removing purging pa- 3. Ex vivo entails "a days. within 10 mice all died to 20 In treated fact, blood ... cells ... from [own] tient’s patent application itself acknowl ..., body, activating them] and then re- edged TALL-104 studies demon turning patient” them] back to the for thera- strated “antitumor in vivo ... to in py, stemming the effects the acti- spontaneous lymphomas remissions of duce cells. vated J.A. 7. 53; Cesano, dogs,” J.A. tbl. 3. The on for in vivo therapy. Through therapeutic utility NK-92’s been used fact, Id. In vitro data. the ’955 patent comparison, taught head-to-head Yan application concluded from persons art to data skilled combine the teachings of Santoli—using that “the NK-92 cells of the invention are cell line in surprisingly adoptive significantly immunotherapy—with more effec- the teach- ings Gong—the lysing patient-derived tive in tumor cells NK-92 cell ... line. than ... the cells from TALL- [the the field.” Id. 104] line[ ] known b
¶ 104.
We also
it
find that
been at
would have
taught
that “[t]here remains a
least obvious for
try
skilled
artisans
therapeutic
need
methods
teachings
combine the
Gong.
Santoli and
...
cytotoxic
cancers which can utilize
problem
When there
a ...
is
there
T
present
cell lines and
need ...
avoid
finite
identified, predict-
are a
number-of
for patient’s own killer cells.” J.A.
col.
solutions,
of ordinary
able
skill
person
provided
II.
thus
an ex
33-37. Santoli
good
pursue
has
reason to
the known
plicit suggestion
(alloge-
to use cell lines
options
If
....
this leads to the antici-
therapy)
neic
in cancer treatments because
success,
pated
it
likely
product
greater
availability.
fact,
their
In
but
ordinary
innovation
skill and
’955 patent application
recog
itself
common sense.
that instance the fact
prior investigators
nized that
turned
al-
that a combination
try
was obvious to
logeneic therapy
preference
using
might
it was
show that
obvious under
patient’s
own cells.
’955
applica
§ 103.
¶
tion,
(To
“major
overcome the
obsta
cle”
vivo”
“expanding]
Inc.,
v.
NK cells
KSR Int'l Co.
550 U.S.
Teleflex
*6
use, “many investigators
for clinical
have
398, 421,
1727,
127 S.Ct.
use
anticancer
(Fed.
1989)
possibilities
Cir.
ed number
hood success -with fact that in vitro success does disagree. Santoli. We always not into in vivo translate success summary judgment. cannot recognize Our cases there is no defeat “[Obvi general simply by rule that skilled artisan cannot cannot be avoided ousness reasonably extrapolate in vivo showing degree unpredictability success of some from in vitro results. “Obviousness does long in the art so as there was a reason require predictability absolute of suc probability Pfizer, v. able success.” Inc. Indeed, many cess. inventions that Inc., (Fed. Apotex, 480 F.3d obvious, is no quite seem there absolute 2007). Indeed, Cir. NantKwest itself sim predictability success until the invention ply argues “[pjositive results in O’Farrell, In re practice.” is reduced necessarily do not establish reasonable 1988). “[pro 894, 903 853 F.2d therapeutic of success for probability viding proof justify conducting sufficient to drug Appellant of that in vivo.” Br. 43 humans, procedures while use added). here, But (emphasis as we have ful, patentability.” is not a test of Phar 'above, over-whelming there is discussed ViaCell, Therapeutics, Inc. v. maStem specific that a evidence skilled artisan *7 2007). Inc., 491 F.3d reasonably extrapolate could from the in
Rather, respect to our that vitro data with TALL-104 and cases hold whether extrapolate reasonably skilled artisan can in vivo suc NK-92 that would teach their fact- highly from in vitro results is successful substitution in vivo.4 cess argues Vujanovic reading. Vujanovic et al. A- 4. NantKwest that text indeed teaches that ("Vujanovic”) "al- reference teaches that NK and NA-NK have similar in vitro though against gastric NA-NK the A-NK and cells used for cancer cells. J.A. tbl. I. therapy cytotoxicity Vujanovic against gastric showed similar levels of that also teaches vivo, against [gastric undergoing cells tested mice NA-NK cancer] [in HR cancer cells in (Table I), develop A-NK cells dra- cancer vitro] demonstrated treatment 20% the greater matically significantly develop in antitumor metastases that would mice, untreated (Table V, undergoing than NA-NK cells in vivo while mice A-NK treatment this, 4).” Fig. develop only 960. From J.A. NantKwest con- the cancer metastases that 6% "difficulty predicting develop in mice. [in] cludes that there is would untreated sure, signifi- statistically is efficacious in vivo cancer treatments from in tbl. V. To be this a assays,” particular Ap- for NK the NA-NK A-NK cells. cant difference between However, pellant Rep. reject 5. We this out-of-con- treatments. what NantKwest fails to Br. Second, However, spe- argues five appellant asserts NantKwest Miller teaching away shows that from disputes cific material as to whether there is translate into vitro success here would using allogeneic therapy NK-92 because vivo success. not well at the “[i]t was understood time allogeneic the invention whether NK cells First, argues that Miller subject foreign the] would be [attack teaching away shows that there is a from host or [the whether administered NK using unmodified NK-92 cells in vivo be- indiscriminate[ly] cells would] kill[ ] cause used TALL-104 cells modi- host cells.” J.A. 411. gene.” Appellant fied to contain a “suicide “strongly discourages Br. 38. Santoli thus general This was a broad and concern attempting from to in- the skilled artisan fact, known to skilled artisans. USPTO into troduce unmodified cells [immune] expert agreed Dr. Lanier that the “use genetic host” this modifica- new without foreign allogeneic immune cells thera- such, tion, may cells because cause cancer. may pies be associated with” reactions J.A. 392. foreign from the host and the cells It necessary each other. J.A. 573. would be during argu-
NantKwest conceded oral “to reactivity cells for test immune language ment that claim here does cytotoxicity against allogeneic normal host require administering “unmodified” cells; is, developing administering treat- modi- in vivo.” Id. There was no encompassed by ments ... fied NK-92 cells is testimony here that such an adverse reac- Arg. claimed invention. Oral 33:40-56. Therefore, context, tion away likely teaches was this and there whether Santoli well-known, using testimony from is was no this unmodified NK-92 cells irrel- (that Furthermore, general phenomenon administering evant. art had dis- foreign “le- cells into host could cause unto- closed that cells could be reactions) irradiated,” thally prevent a skilled so that became ward cells trying “non-proliferating,” sacrificing without artisan lyse undertaking necessary and ability TALL-104’s cancer cells. while precautions. If the for such dispute J.A. 183. There is no there known need prevent se- pre-administration trials could would be irradiate cell lines motivation Thus, problem curing patent, if it then no this arose.5 address therapy im- before clinical safely using genetically would ever issue unmodified taught complete. appears trials were already mune cell While line view,6 been Dr. Miller’s that view prior art. own difference, prevent proliferation.” Appellant highlight despite both uncontrolled Rep. produced very Br. 17 n.2. NA-NK and A-NK treatments statistically significant reductions in *8 following during depo- Dr. Miller 6. stated compared to untreated mice. metastases sition: support argument Vujanovic thus does not Q. NK that cells' antitumor in vitro anything than an Is there less extrapolated be to vivo suc- cannot study using NK-92 cells in mammals cess; most, only suggests it that the exact at expectation provide reasonable of could magnitude may predict- hot be that success method? success for the claimed able. any predict- I’m not sure that there’s A.... with, ability be comfortable that would experiments doing types of ultimately those It determined that' NK-92 short 5. was cell present with the NK-92 line. cells did not this risk because 776-77. not need to be modified or irradiated to J.A. "do from cell lines to correspond existing stan- tumors derived tumors does not to the PharmaStem, directly from patentability. patients. See dard derived 1364; Pfizer, at 480 F.3d at F.3d shows, ap- And as this caveat the data (“[A] equating unpredictability rule of law TALL-104, only to not NK-92.7 plies but any patentability to ... would mean above, lysed As TALL-104 three discussed [drug] pat- separately ... would be new types out four cells derived properties its] ... ... [are] [after entable lines, laboratory only one out from cell but through testing.”). verified types of of nine cancer cells derived direct- Third, argues that Yan NantKwest ly patients. comparison, from In the same taught away substituting from NK-92 lysed all types of the four can- it TALL- showed that TALL-104 because from cers cell lines and all derived cytolytic 104 and NK-92 activi- “differed directly of cancers from types nine derived ty against” Appellant the tumors tested. words, no In other patients. showed However, appellant Br. 34 n.6. failed to efficacy against in lysis labora- difference actually mention that this difference was line tory cell and patient-derived cancers [cell the fact NK92 line] that “[t]he Therefore, TALL-104, unlike cancers. cytotoxic all” highly towards of the tumors differently lyses which cancer cells de- tested, lysed four out while source, pending on their NK-92’s efficacy J.A. of the thirteen tumors tested. 226. context-dependent. not appears to be Therefore, NK-92 was much more effica- Yan, 226. J.A. TALL-104, cious which would teach than Fifth, NantKwest makes much toward the substitution. skilled artisan T cells’ and cells’ between NK differences Fourth, argues that Yan NantKwest receptors, argue to cell-surface these provide does not a motivation combine made it differences would have difficult to extrapolating it im- because cautioned extrapolate NK-92 behavior from TALL- context to munotherapy results from one However, high- 104’s. while Miller indeed unpredictable. Specifically, another may be differences, lights he conceded that these suggest that Yan stated that “our studies killing killing “NK-92 and T-ALL can- [of cytotoxic activity [immune cells’] towards not, my knowledge, cer cells] has laboratory] from [cancer derived [1997], specific the[ir] been associated with extrapolated [cancer] lines cannot receptor pattern.” Deposition, Miller EOF directly patients. ... cells derived 59-1, Therefore, cell-surface 34. these .., immunotherapy must be [S]uch not receptor differences are material for accompanied by study careful lyse ability the two cell lines’ similar unique patterns activity” antitu- cancer cells. Contrary mor cells used. 226. immune J.A. “ad- Miller’s assertion that caveat V relying against on in tests monishes against argues that even activity,” Yan obviousness, it prima case had comparing fact facie fact, presented secondary Yan no in vivo considerations efficacies. contains Rather, experiments. cautioning “may genuine Yan is raise a nonobviousness lysis against sum- extrapolating precludes results issue of material fact *9 caveats, Yan’s, points apply only 7. also out a similar like to TALL- NantKwest also that publications, is See caveat in two found other 104. J.A, 46-47; 953; Appellant Br. 961. Those mary judgment.” Appellant 61. the Br. that “[E]vi- cell line surprisingly was secondary dence ... ... considerations safer than the TALL-104 therapy. line always present ¶ must Miller, ¶ when be considered J.A. 393-94 J.A. 396 53. to a en determination of route obvious- We the conclude that district prop- court In re Cyclobenzaprine Hydrochlo- ness.” erly granted summary judgment that ride Capsule Extended-Release Patent 20, 26, claims 27 were and invalid as obvi- (Fed. Litigation, 676 F.3d Cir. ous. 2012) (quotation marks omit- and citation ted). However, ... secondary “[flor con- VI siderations to be accorded substantial weight, proponent its The to agree nex- dissent appears must there establish us between the evidence the merits of is supporting and substantial the evidence dis- Moreover, the claimed invention.... sec- trict finding court’s obviousness. Howev- ondary er, ... cannot considerations over- the dissent concludes that NantKwest strong prima come a facie case obvious- evidence contrary submitted raises Wyers Co., ness.” v. Master Lock 616 F.3d genuine issues of material fact. 2010) (quotation The cited two cases dissent omitted).
marks and citation
proposition
that “lesser evidence [than
For
secondary
considerations
presented
is
here was]
what
sufficient to
nonobviousness,
presents
rejections by
past”
reverse
the PTO in the
announcing
article
million in
news
clearly distinguishable.
are
Op.
Dissent
$48
vestment
above,
NantKwest as evidence
As
876-77.
discussed
both Carroll’s
success,
some
commercial
results
Gangadharam’s holdings
highly
NK-92’s Phase
clinical trials as evidence
Carroll,
fact-specific.
In
held that
court
results.
unexpected
art on the in
of an
vitro use
antibiotic did
render
vivo
its in
investment,
respect
to the
With
we
teaching
obvious because
away
there was
agree
court that
there
district
from in vivo
specific
for that
extrapolation
no direct
nexus between
million
$48
Carroll, 601
antibiotic. In re
1186.
F.2d at
purchase
stock
merits of
and the
extrapola-
Carroll does not discuss
vivo
claimed invention
commer-
demonstrate
success,
tion
generally. Gangadharam,
the court
report
cial
indicates that the
case,”
simply
“in
found that
the USP-
purchase
stock
for corporate
made
TO
failed
“carr[y]
proving
its burden
control purposes.
prima
case of obviousness” because
facie
“pre-
NantKwest also contends
it
“general
sole
made
ref-
sented
the clinical success
evidence
positive
erence ...
[to]
results
were
therapy
...
NK-92 in
has dem-
[that]
in an entirely
obtained ...
con-
different
unexpected, superior
onstrated
results in
text,
...
only] precatory,
[made
en-
I”
two recent
Phase
clinical trials.
couraging
relating
statements
to uncertain
Appellant Br. 57.
is no
There
evidence
investigations”
future
applica-
com-
unexpected
results were
Here,
tions. 1989
*1-2.
WL
pared
expected
what was
contrast,
teaching
there is
fact
towards
Indeed,
above,
in vitro data.
as discussed
the invention.
suggested
vitro data
the in
Vujanovic, Yan,
vivo trials would be successful.
re-
dissent also
With
reads
results,
spect
safety
to the Phase I
testimony
trials'
and Gesano—as well Miller’s
support
concerning
gen-
there
for the
creating
was no
conclusion
these studies—as
*10
provide support for Mil-
pas-
uine issues of material
The cited
therefore cannot
fact.
in vivo
argues.’
warning against
extrapolation.
sages do not show what Miller
ler’s
Third,
dissent cites Cesano for the
First,
Vujanovic
cites
for the
the dissent
sensitivity
teaching
dogs’
that “the
of the
suggest that
teaching that
standard
“[w]e
in vitro did
lysis
tumors to TALL-104 cell
cytotoxicity assays
target
in
with
vitro
appear
good
not
to be a
of clinical
indicator
suspensions
little relevance
This
responses.” J.A. 958.
sentence does
predicting
the in vivo
antitumor
and, as
not
discussed
address
added).
962 (emphasis
effector cells.” J.A.
above,
NK-92 was shown
be more effica-
merely stating that a certain type
This is
Therefore,
cious than TALL-104.
this
(“cells
experiment setup
suspension”
actually
would
teach a skilled
to-
artisan
in vivo
assay)
predicting
is not
for
suitable
(as
substituting NK-92 for TALL-104
ward
Indeed,
very next
results.
sentence
do).
suggested
Yan
assays
spheroids
and concludes
addresses
imply
further
results
“[o]ur
opposing parties tell two
“When
differ-
vitro assessment of effector cell functions
stories,
blatantly
one of
con-
ent
which
spheroids [assays]
with multicellular CA
record,
by the
... a court should
tradicted
suspensions [assays]
or
instead
CA
adopt
not
that version of
for
the facts
[assays] might
greater
monolayers
be
purposes
ruling
on a motion for sum-
Harris,
predicting
relevance
vivo thera-
mary judgment.” Scott v.
550 U.S.
peutic
potential of
ef- 372, 380,
antitumor
immune
127 S.Ct.
875 (Fed. 2012) (internal Drawing quotation cells. all reasonable inferences in Cir. marks NantKwest, must, omitted). favor of as we this evi and citations Whether a skilled genuine dispute of dence creates materi artisan would had a reasonable ex grant summary- al fact that bars the pectation question fact, is a success judgment. majority explains Because the Mylan Cumberland Pharm. Inc. v. Insti away giv NantKwest’s evidence instead of LLC, tutional 1213, (Fed. 846 1222 F.3d law, ing it weight required by I re 2017), contemporaneous Cir. evidence spectfully dissent. thought what skilled artisans at the time of the help invention can inform our
I. inquiry into expectation whether Carroll, In re matter, As an success was reasonable. initial the difference be- 1184, (C.C.P.A. 1979). vitro and testing tween is criti- F.2d 1186-87 When cal to understanding difficulty using determining person ordinary whether a results from to efficacy the former possessed expectation skill a reasonable the latter. In vitro experiments typically in predicting success occur the controlled environment of a on in vitro results, recog we have vivo experiments tube; petri dish or test “simply drug gives nized that because performed living organism. in a J.A. vitro, positive results it does not neces in vitro cannot account Experiments 886. sarily follow that there is a reasonable for the living variable environment probability of therapeutic success for organism replicate and cannot a cell line’s Gangadhar in vivo.” In re drug system, interaction with the host’s immune am, 1101, 127023, 889 F.2d WL *3 ¶ 384, among 24; things. other J.A. J.A. 1989) (non-precedential) (citing disparity testing 886. This environments Carroll, 1186). 601 F.2d at in vivo results. unpredicted can to lead majority person that a concludes example, For cell lines with encouraging ordinary skill in would have been cytotoxic activity in vitro can unexpectedly ¶ to in vivo. J.A. Gong motivated combine with all Santoli lose 93. The in vivo effi- system destroy host’s immune can even because Santoli demonstrated line, rendering ineffective cells, it cacy Gong for TALL-104 detailed vivo. Id. It is also possible the cell line NK-92’s in vitro lyse ability to trigger immune reactions in severe cells, prior art indicated desire produce complications. host serious applications clinical for the seek ¶ Therefore, J.A. 47. demonstrated Maj. Op. cell line. NantKwest’s 867-69. in vitro does not cytotoxic activity always arguments person ordinary skill that a translate to success in vivo. a reasonable in the art would have had combining
expectation of success II. ma- Gong persuasive with were not jority light “over-whelming specif- of the presence of each element claimed ic that a skilled artisan could evidence art is insufficient to render a the reasonably extrapolate Rather, claim obvious. there also must be respect and NK-92 data a motivation to combine the art and reasonably their success- that would teach ordinarily an skilled artisan must have substitution in vivo.” Maj. Op. 870. expectation had a in ful reasonable success conclusion, majority Concepts, so. Kinetic Inc. v. Smith doing reaching its failed Inc., Nephew, & weight it give 688 F.3d evidence the NantKwest’s of the J.A. all reason- ments at the time deserved and declined draw invention. ¶ ¶ 439-40, 48; 94. This creates a inferences in favor. able NantKwest’s dispute regarding rea- of material fact contemporaneous references Numerous *12 view, In expectation my sonable of success. by that in NantKwest warned identified Dr. majority’s willingness the to discredit predictive not vivo vitro of in results were understanding the in Miller’s of disclosures example, in For the this field. Vujanovic dispose and Yan not the does Vujanovic questioned reference the corre- genuine dispute Although of material fact. vivo lation between in vitro and in studies its view of the majority the believes sug- for the line: “We behavior NK-92 cell Miller’s, superior is to Dr. references gest cytotoxicity that in vitro standard analysis by its not supported citations assays target suspensions with cells in or expert report the USPTO’s the district little the predicting in relevance in Maj. Op. Dr. opinion. court 873-74. activity vivo antitumor cells.” effector hand, is opinion, Miller’s on the other illu- added). 962 (emphasis As confirmed of a by background knowledge the minated Miller, by Dr. passage Vujanovic this from field, this I not skilled artisan in am against relying cyto- “caution[s] inon vitro in opinion support convinced that his lacks toxicity results, Gong, such those in genuine dispute the The result is record. ¶ predict in vivo behavior.” J.A. 94. is not material fact that believe suited Yan contains con- Yan warnings. similar summary judgment for resolution at the his comparison cluded NK-92 and stage. by noting: “[0]ur cells studies Indeed, found lesser evidence we have suggest cytotoxic activity towards rejections by sufficient the PTO to reverse leukemic cell extrapolated cannot be lines Carroll, past. for example, the directly patients. derived rejected Appeals PTO’s Board of as obvi- biologic or in reagents such vitro treating for patent ous claims M. immunotherapy purging vivo for or must acid based on paratuberculosis with lauric study accompanied by careful of the patentee’s master’s The thesis thesis. patterns activity....” unique J.A. 226 1) types disclosed two of studies: in vitro added). (emphasis Dr. Miller reiterated studies, reported patentee from which the against relying admonishes on “Yan completely lauric inhibited the acid vivo activity,” in vitro tests growth paratubercu- three strains M. was with “the common which consistent losis, 2) studies, from which understanding 439-40, in the art.” J.A. reported suitability of patentee ¶ Finally, 48. in Cesano’s describing article certain strain C57 black mice as labora- I Clinical Trial a Phase for TALL-104 tory studying animals the diseases that, dogs, acknowledged authors Carroll, paratuberculosis. caused M. “[sjurprisingly, sensitivity dogs’ of the field, expert An F.2d at 1185. tumors to TALL-104 in vitro lysis did Merkal, patentee’s Dr. the- discounted appear good to be a of clini- indicator because, publication sis the time of its added). (emphasis responses.” cal J.A. 963 reasons, among testing other “in vitro was supporting evidence NantKwest’s an unreliable indicator for the in vivo ef- expert testimony uncertainty bare the laid paten- Id. When the fectiveness.” at 1186. complex sought Miller discovery As Dr. ex- tee later field. his plained, the refer- can authors above that mammals be treated lauric predicting orally paratuberculosis, ences feel did not confident acid to treat M. activity experi- argued the PTO claims were based obvi-' Sandoz, Inc., 1341, 1350-52 his thesis ous because earlier disclosed 544 F.3d 2008) in vitro and that lauric (affirming acid’s mice Cir. non-obviousness studying suitable animals for the dis- were grant context district court’s of prelimi ease. Id. at 1185-86. Our court alia, predecessor because, inter injunction nary PTO, disagreed relying principal- with the patentee’s evidence “it demonstrated “contemporaneous ly on Dr. Merkal’s eval- not predictable” an how antibiotic appellant’s uation one thesis[] on perform given in this art skilled would have no experiments using closely related antibi weight findings reported to the therein.” otic).
Id. at 1186-87 added). (emphasis *13 As was the case with both Carroll Gangadharam, we held Gangadharam, Similarly, the record here contains Appeals Board of In that the Patent persons ordinary evidence that ample finding erred in terferences a reasonable skill in skeptical the art were about the expectation using CQQ success to ability of in vitro tests to in vivo (in vivo) treat tuberculosis in mammals efficacy for the two relevant cell lines. Gangadharam, on in vitro results. based Vujanovic, Yan, very and Cesano—in the (nonprecedentia]). *3 WL at same articles where discussed single relied a Board reference promising results for the NK-92 reject in part by applicant authored lines—proceeded and TALL-104 cell to ex Gangadharam the claims. It found pressly against inferring efficacy caution “very positive reference’s disclosure of in vivo A on these rea outcomes. vitro bactericidal CQQ reading sonable these sup references [pertinent] reported by bacteria Gan- ports the conclusion that one in the skilled the gadharam certainly favors expec have had a reasonable compound said in the treatment tu Santoli’s combining tation success in Id. at *1. berculosis But mammals.” testing Gong’s for TALL-104 with simply “[rjemarking positive NK-92 cell line. Because is the vitro results ‘favored’ use in vivo does not summary judgment non-movant at standard,” statutory meet ex we it is stage, a reasonable inference that we plained, and therefore PTO “fell woe Anderson, favor. in its must draw fully short of its burden” establish 255, 106 at U.S. S.Ct. 2505. Id. expectation reasonable success. at Gangadharam Importantly, *2. refer III. proviso ence contained further stud CQQ sug ies needed “before were can be grant for the Our standard review gested possible antimycobacterial summary judgment requires us to believe treating drug humans with [tuberculo of the non-movant and to the evidence sis],” article contemporaneous draw all in its favor. reasonable inferences vitro tests were neither warned majority neither in finding the did to, for, equivalent nor a substitute patent application claims NantKwest’s experiments. Id. at *2-3. “there Because Accordingly, respectfully obvious. dis- in this regarding [wa]s evidence record sent. in vivo from the noncorrelation efficacy generally respect and with tu
berculosis,” found that the PTO failed we expectation demonstrate reasonable *3; see also Labs. Id. Abbott v.
success.