Ex Parte Larsen et al

Patent Trial and Appeal Board

Oct 14, 2014

10766057 (P.T.A.B. Oct. 14, 2014)

UNITED STATES PATENT AND TRADEMARK OFFICE
__________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
__________

Ex parte ROY H. LARSEN and GJERMUND HENRIKSEN1
__________

Appeal 2012-005386
Application 10/766,057
Technology Center 1600
__________

Before DONALD E. ADAMS, ERIC B. GRIMES, and
ULRIKE W. JENKS, Administrative Patent Judges.

GRIMES, Administrative Patent Judge.

DECISION ON APPEAL
This is an appeal under 35 U.S.C. § 134 involving claims to a method
and conjugate for targeting malignant cells, which have been rejected as
obvious. We have jurisdiction under 35 U.S.C. § 6(b). We affirm.
STATEMENT OF THE CASE
The Specification states that the “use of folate and folate derivatives
to target tumours expressing folate binding protein (FBP) . . . has attracted
attention among researchers. As several types of human cancer cells have
been shown to overexpress FBP, this receptor may be a possible target for

1 According to Appellants, the Real Party in Interest is Algeta AS (Appeal
Br. 3).
Appeal 2012-005386
Application 10/766,057

2
delivery of therapeutic radioisotopes conjugated with folate.” (Spec. 1:8–14,
reference citations omitted.)
Claims 18 and 25–35 are on appeal. Claims 18 and 31 are the only
independent claims and read as follows:
18. A method of targeting a radionuclide to a malignant cell within a
subject, wherein said malignant cell expresses a tumor associated antigen
and expresses folate binding protein, said method comprising (i) coupling an
antibody, antibody fragment, or antibody construct having affinity for said
tumor associated antigen to at least one non-cytotoxic folate to form a dual
binding conjugate, (ii) coupling said radionuclide to said dual binding
conjugate, and (iii) administering said radionuclide coupled to said dual
binding conjugate to said subject.

31. A conjugate consisting of (i) a radionuclide, (ii) an antibody, antibody
fragment, or antibody construct, with affinity for a tumor associated antigen,
and (iii) at least one non-cytotoxic folate.

DISCUSSION
Issue
The Examiner has rejected claims 18 and 25–35 under 35 U.S.C.
§ 103(a) as obvious based on Wedeking2 and Sinkule3 (Ans. 4). The
Examiner finds that Wedeking discloses preparing a gadolinium-folate
conjugate and administering it to a mammal to target malignant cells
expressing folate binding protein (FBP) (Ans. 4). The Examiner finds that
Sinkule discloses a receptor-binding conjugate comprising a monoclonal
antibody, a radionuclide, and a chemotherapeutic agent such as a folate
analogue (id. at 5). The Examiner finds that the “antibodies are included in

2 Wedeking et al., US 6,093,382, patented July 25, 2000.
3 Sinkule et al., EP 0 282 057 A2, published Sept. 14, 1988.
Appeal 2012-005386
Application 10/766,057

3
[Sinkule’s] conjugate to target the conjugate to a desired tumor cell” (id. at
6) and concludes that it would have been obvious to include Sinkule’s
antibody in Wedeking’s gadolinium-folate conjugate because “FBP is
expressed in normal as well as different tumor cells and thus it would be
advantageous to attach an antibody to the conjugate of Wedeking et al. to
ensure site-specific targeting of the conjugate into the desired tumor cells
with enhanced affinity while minimizing the damage to normal cells” (id.).
Appellants contend that, as evidenced by Shinoda,4 a skilled artisan
would have expected that attaching an antibody to Wedeking’s conjugate
would interfere with the targeting by folate of cells expressing folate binding
protein (Appeal Br. 6–7, 10–11). Appellants also contend that the
Specification shows that, surprisingly, “both the antibody component and the
non-cytotoxic folate component of a conjugate can target the conjugate to a
malignant cell” (id. at 8).
The issues with respect to this rejection are (a) whether the evidence
of record supports the Examiner’s conclusion that it would have been
obvious to add a tumor-directed antibody to Wedeking’s conjugate and (b) if
so, whether the Specification provides evidence of unexpected results that
outweighs the evidence supporting a conclusion of obviousness.
Findings of Fact
1. Wedeking states that there are two major isoforms (α and β) of
folate-binding protein (FBP) (Wedeking 3:10–11, 25–26).

4 Shinoda et al., In Vivo Fate of Folate–BSA in Non-Tumor- and Tumor-
Bearing Mice, 87 J. PHARM. SCI. 1521–1526 (1998).
Appeal 2012-005386
Application 10/766,057

4
2. Wedeking states that the α isoform of FBP (“FR-α”) is “expressed
in normal epithelial cells and is frequently strikingly elevated in a variety of
carcinomas” (id. at 3:31–32).
3. Wedeking states that “[o]verexpression of FBP by a number of
different tumors has led a number of investigators to explore its potential as
a delivery system for toxins or poorly permeable compounds coupled to folic
acid and as a means to increase selective delivery of antifolate drugs such as
methotrexate to tumors” (id. at 4:37–41).
4. Wedeking states that
FBP levels are low in many normal tissue types while, in
comparison, FBP levels are high in many tumor cells. This
difference between the folate receptor levels allows selective
concentration of pharmaceutical or diagnostic agents in tumor
cells relative to normal cells, thereby facilitating treatment or
visualization of tumor cells.
(Id. at 5:1–7.)
5. Wedeking discloses a “radiotherapeutic agent comprising a
chelated radioactive metal complexed with a folate receptor-binding ligand.
. . . A suitable radiotherapeutic composition according to the invention
comprises, as the active ingredient, a folate-metal chelate derivative that
bears an alpha- or beta-emitter that is suitable for radiotherapy.” (Id. at
8:29–40.)
6. Wedeking discloses a composition for radiotherapy comprising
“ligands chelated to radioactive alpha or beta-emitting metals that are
coupled to folic acid receptor binding ligands” (id. at 8:15–19).
Appeal 2012-005386
Application 10/766,057

5
7. Wedeking discloses that “[b]oth derivatives of folic acid and of
methotrexate (MTX) are included for use in the composition” (id. at 8:25–
27).
8. Wedeking discloses “a method for radiotherapy by neutron capture
techniques, comprising administering to a patient said composition, wherein
the metal is gadolinium” (id. at 8:53–56).
9. Wedeking discloses “administering to a mammal a composition
comprising a radioactive metal complexed with a chelating ligand coupled to
a folic acid analog contained in a pharmaceutically acceptable carrier for
radiotherapeutic treatment” (id. at 68:54–58).
10. Sinkule discloses “a chemo-radio-immuno-conjugate. . . . The
radioactive component is used to potentiate the therapeutic effect of the
chemotherapeutic agent of the conjugate although it can, in some cases, also
be used to monitor the distribution of the conjugate in vivo.” (Sinkule 2:15–
20.)
11. Sinkule discloses that among the preferred chemotherapeutic
agents are folic acid analogs (id. at 2:26–30).
12. Sinkule states that “[t]he therapeutic activity of the conjugate in
vivo is localized by the antibody, which is selected for its specificity for the
target cell” (id. at 5:13–15).
13. Sinkule states that “[m]onoclonal antibodies have already been
developed for a wide variety of target antigens, including tumor cells” (id. at
4:9–11).
Appeal 2012-005386
Application 10/766,057

6
Analysis
Wedeking discloses a conjugate consisting of a radionuclide and a
non-cytotoxic folate (FFs 5–7). Wedeking also discloses coupling a
radionuclide to a non-cytotoxic folate (FFs 6, 7) and administering the
resulting conjugate to a subject (FF 9). Wedeking does not specifically
describe this method as targeting the conjugate to a tumor. However,
Wedeking discloses that folate binding protein (FBP) levels are low in most
normal tissue types (except epithelial cells; FF 2) but high in many tumor
cells (FF 4) and “frequently strikingly elevated in a variety of carcinomas”
(FF 2). Wedeking also discloses that this overexpression of FBP by tumors
has led investigators to couple toxins to folic acid in order to increase
selective delivery to tumors (FF 3). Therefore, Wedeking would at least
have made obvious a method of targeting a radionuclide to a malignant cell
in a subject by coupling it to a non-cytotoxic folate and administering the
conjugate to a subject.
Wedeking does not disclose a conjugate that also includes an antibody
having affinity to a tumor-associated antigen. Sinkule discloses a conjugate
comprising a radioactive component, a chemotherapeutic agent, and an
antibody (FF 10). Sinkule discloses that the chemotherapeutic agent can be
a folic acid analog (FF 11) and that the antibody has specificity for the target
cell to which the therapeutic activity of the conjugate is directed (FF 12).
Sinkule also discloses that monoclonal antibodies for tumor cells have been
developed (FF 13).
We agree with the Examiner that these disclosures would have made
obvious the conjugate of claim 31, and the method of claim 18, on appeal.
Appeal 2012-005386
Application 10/766,057

7
Specifically, it would have been obvious to include an anti-tumor antibody,
as taught by Sinkule, in Wedeking’s radionuclide/folic acid conjugate, in
order to more precisely target the radionuclide to tumor cells rather than
healthy cells. Wedeking provides a reason to make this modification,
because it discloses that folate binding protein (FBP) is expressed in normal
epithelial cells (FF 2) and expressed at a low level in many normal tissues
(FF 4). A skilled worker would therefore expect that Wedeking’s
radionuclide/folic acid conjugate would affect epithelial cells and, at a low
level, cells in other normal tissues, and that addition of an antibody that
binds to a tumor-specific antigen would decrease the undesirable cytotoxic
effect of the conjugate on healthy FBP-expressing cells.
Appellants argue, however, that because “in Sinkule, the antibody is
targeting the complex, one simply cannot conclude that the presence of the
antibody does not interfere with folate targeting” (Appeal Br. 6). Appellants
argue that Shinoda “show[s] that adding a large molecule, such as bovine
serum albumin (BSA), to a folate-containing complex can interfere with the
targeting ability of the folate” (id.). Appellants argue that an antibody is
over twice as large as BSA (id.) and therefore an antibody would also have
been expected to interfere with targeting based on folate/FBP binding (id. at
6–7, 10–11).
We do not find the cited evidence to support Appellants’ position.
Shinoda describes experiments in which “the disposition of folate-
conjugated bovine serum albumin (folate-BSA) was examined as a model
system for drug targeting” (Shinoda 1521, abstract). More specifically,
Shinoda compared the in vivo rate of uptake of 111In-labeled folate-BSA
Appeal 2012-005386
Application 10/766,057

8
with that of 111In-labeled BSA in mice that had received transplants of KB
tumor cells (which overexpress folate receptors on their membranes) (id.).
Shinoda concludes that “the tumor uptake rate index for 111In-folate-BSA
(0.46 µL/min/g) was slightly higher, by a magnitude of 1.4, than that for
111In-BSA (0.32 µL/min/g)” (id. at 1525, left col.). Shinoda concludes that
“111In-folate-BSA could not be targeted to the tumor sufficiently in this
experiment” (id.).
Appellants argue that Shinoda’s results are relevant to what one
skilled in the art would expect if an antibody, rather than BSA, was coupled
to folate (Appeal Br. 6). Specifically, Appellants argue that Shinoda
found that there was a significant difference in
pharmacokinetics and biodistribution of non-folate compared to
folate labelled BSA. A high liver uptake and rapid blood
clearance indicated that the folate labelled version of 111In-BSA
was not particularly suitable for radionuclide delivery to tumour
cells expressing folate binding protein.
(Id., quoting Spec. 1:22-27.)
Shinoda, however, proposes two possible reasons for its results, and
Appellants have not persuasively explained why either reason would have
been expected to apply to a folate-antibody conjugate. One of the reasons
Shinoda proposes is that “this failure could be due to inhibition of folate-
mediated 111In-folate-BSA uptake by KB tumor cells from the blood”
because of free folate in the blood (Shinoda 1525, right col.). Shinoda
states, however, that a concentration of 3 nM folic acid in the blood (the
normal level in humans) should result in only a 10% decrease in folate-BSA
uptake by tumor cells (id. at 1525, right col.). That is, Shinoda’s explanation
does not propose any interference by BSA to the binding of folate by FBP; it
Appeal 2012-005386
Application 10/766,057

9
states that competition for FBP binding by free folate may decrease FBP
binding by the folate in its folate-BSA conjugates. Appellants have not
explained why this explanation would lead a skilled artisan to doubt the
success of an antibody/folate/radionuclide conjugate.
Shinoda’s other proposed reason is that the observed result “could be
due to tissue structure differences between liver and tumor tissues” (id. at
1525, left col.). That is, the liver has “sieve plates” that allow passage of
small particles but in tumors the capillary is lined with tumor cells, which
can either be leaky or can “exhibit a continuous type of structure where there
are formidable barriers to dispersion” (id.). Shinoda states that its results
“suggest[ ] that the low vascular permeability into solid tumor tissue of the
large 111In-folate-BSA molecule may be the rate-limiting factor of
accumulation” (id. at 1525, right col.).
We do not agree with Appellants that this proposed explanation would
have led a skilled worker to expect that attaching a tumor-specific antibody
to Wedeking’s conjugate would have interfered with its binding, via its folic
acid moiety, with FBP. Sinkule discloses that the folic acid analog
methotrexate in its conjugate retains the ability to bind to the enzyme
dihydrofolate reductase (DHFR) (Sinkule 7:31–34). A person of ordinary
skill in the art would expect, therefore, that a non-cytotoxic folate would
also retain the ability to bind proteins, including FBP.
Shinoda’s explanations for its results do not contradict this expectation.
At best, Shinoda’s statement that the accumulation of its 111In-folate-BSA
conjugate might be limited by “low vascular permeability into solid tumor
tissue” (Shinoda 1525, right col.) suggests that an antibody/folate/
Appeal 2012-005386
Application 10/766,057

10
radionuclide conjugate might have difficulty entering solid tumor tissue if
administered intravenously. However, Sinkule states that an antibody/folate
analog/radionuclide can be administered “by any conventional procedure,
topically, orally, intravenously, parenterally, and the like” (Sinkule 6:49–52).
Sinkule also the effective dose of such a conjugate “will depend, inter alia,
upon tumor burden, accessibility and other variables” (id. at 7:6–8).
Appellants have not explained why a person of ordinary skill in the art would
have required more than routine experimentation to adjust the route of
administration and/or the dosage of an antibody/folate/radionuclide conjugate
achieve accumulation of an effective dose at the site of a tumor.
Appellants also argue that they, “surprisingly, have shown that both
the antibody component and the non-cytotoxic folate component of a
conjugate can target the conjugate to a malignant cell” (Appeal Br. 8).
Appellants point to Examples 5 and 6 of the Specification as evidence of this
unexpected result (id.).
This argument is also unpersuasive. The Specification’s Example 5
describes an experiment comparing the binding of folated and unfolated
antibodies to a cell-surface antigen (Spec. 12:10–13). The Example states
that “although some immunoreactivity was lost, [folated] antibodies were
still reactive with the osteosarcoma associated antigen” (id. at 13:1–2).
The Specification’s Example 6 (id. at 13–14) compares binding of a
folated, radiolabeled antibody to OHS osteosarcoma cells, which express the
antigen to which the antibody binds (id. at 12:10–14), with binding of the
same antibody to HELA-S3 and OVCAR-3 Cells, which express folate
binding protein (but not the antigen). The Example concludes that “as
Appeal 2012-005386
Application 10/766,057

11
demonstrated by specific binding of folate-TP-3-IgG-125I to antigen-positive
OHS cells as well as FBP-positive HELA-S3 and OVCAR-3 cells, folate-
antibody-radionuclide conjugates can possess dual binding ability” (id. at
14:20–23).
However, Appellants have pointed to no evidence showing that this
result was surprising or unexpected. The Specification does not characterize
the dual-binding activity as surprising, even though it uses that adjective to
describe other results. (See Spec. 4:9–11 (“When we evaluated folate
labelled human IgG radiolabelled with astatine-211 and iodine-125 we
found, surprisingly, that folate labelled vs. non-folated radiolabelled IgG did
not show significant difference in biodistribution in mice.”)). In addition, as
discussed above, Sinkule states that the folic acid analog methotrexate in its
conjugate retains its ability to bind to DHFR (Sinkule 7:31–34). In short,
Appellants have not provided an evidentiary basis for concluding that the
dual-binding activity of the conjugate recited in the claims would not have
been expected based on what was known in the art. Although Appellants
argue that the activity was surprising, attorney argument is not evidence.
See Johnston v. IVAC Corp., 885 F.2d 1574, 1581 (Fed. Cir. 1989)
(“Attorneys’ argument is no substitute for evidence.”).
Conclusion of Law
The evidence of record supports the Examiner’s conclusion that it
would have been obvious to add a tumor-directed antibody to Wedeking’s
conjugate. The Specification does not provide evidence of unexpected
results that outweigh the evidence supporting a conclusion of obviousness.
Appeal 2012-005386
Application 10/766,057

12
SUMMARY
We affirm the rejection of claims 18 and 31 under 35 U.S.C. § 103(a)
based on Wedeking and Sinkule. Dependent claims 25–30 and 32–35 have
not been argued separately and therefore fall with claims 18 and 31.
37 C.F.R. § 41.37(c)(1)(vii).
TIME PERIOD FOR RESPONSE
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a).

AFFIRMED

cdc