Downloaded from
`
`jem.rupress.org
`
` on February 22, 2018
`
`Anti.tumor and Anti.metastatic Activity of Interleukin
`12 against Morine Tumors
`By Michael J. Brunda,* Leopoldo Luistro,* Rajeev R. Warrier,t
`Rosemary B. Wright,* Brian R. Hubbard,§ Molly Murphy,S
`Stanley F. Wolf,§ and Maurice K. Gatelyt
`
`From the Departments of *Oncology and *Inflammation/Autoimmune Diseases, Hoffmann(cid:173)
`IA Roche Inc., Nutley, New Jersey 07110; and SGenetics Institute Inc., Cambridge,
`Massachusetts 02140
`
`Summary
`It has recently been demonstrated that in vivo administration of murine interleukin 12 (IL-12)
`to mice results in augmentation of cytotoxic natural killer (NK)/lymphocyte-activated killer cell
`activity, enhancement of cytolytic T cell generation, and induction of interferon 'Y secretion.
`In this study, the in vivo activity of murine IL-12 against a number of murine tumors has been
`evaluated. Experimental pulmonary metastases or subcutaneous growth of the B16F10 melanoma
`were markedly reduced in mice treated intraperitoneally with IL-12, resulting in an increase in
`survival time. The therapeutic effectiveness of IL-12 was dose dependent and treatment of
`subcutaneous tumors could be initiated up to 14 d after injection of tumor cells. Likewise, established
`experimental hepatic metastases and established subcutaneous M5076 reticulum cell sarcoma and
`Renea renal cell adenocarcinoma tumors were effectively treated by IL-12 at doses which resulted
`in no gross toxicity. Local peritumoral injection of IL-12 into established subcutaneous Renea
`tumors resulted in regression and complete disappearance of these tumors. IL-12 was as effective
`in NK cell-deficient beige mice or in mice depleted ofNK cell activity by treatment with antiasialo
`GM1, suggesting that NK cells are not the primary cell type mediating the antitumor effects
`of this cytokine. However, the efficacy of IL-12 was greatly reduced in nude mice suggesting
`the involvement of T cells. Furthermore, depletion of CDS+ but not CD4 + T cells significantly
`reduced the efficacy of IL-12. These results demonstrate that IL-12 has potent in vivo antitumor
`and antimetastatic effects against murine tumors and demonstrate as well the critical role of
`CDS+ T cells in mediating the anti tumor effects against subcutaneous tumors.
`
`H uman IL-12 is a disulfide-bonded heterodirneric cytokine
`
`consisting of a 40- and a 35-kD subunit (1, 2). The
`genes for this cytokine have been cloned (3, 4) and purified
`recombinant protein has been produced. A number of bio(cid:173)
`logical properties of human IL-12 have been evaluated in vitro.
`Among its properties are the ability to act as a NK cell and
`T cell growth factor (5-7), to enhance NK/LAK cell cyto(cid:173)
`lytic activity (1, 7-9), to augment cytolytic T cell responses
`(9), and to induce secretion of cytokines, particularly IFN--y,
`from T and NK cells (1, 10). Since both T and NK cells have
`been implicated as antitumor effector cells (11) and IFN--y
`has been shown to have antitumor activity in animals (12,
`13), IL-12 has the potential to be used as an immunomodula(cid:173)
`tory cytokine in the therapy of malignancies.
`The ability to test the in vivo activities of IL-12 have been
`limited since human IL-12 is inactive on murine cells, but
`recently the genes for murine IL-12 have been cloned (14).
`Injection of mice with recombinant murine IL-12 augments
`
`NK activity, enhances allogeneic cytolytic T cell responses,
`and induces secretion of IFN--y, thus confirming the previ(cid:173)
`ously described in vitro activities of IL-12 (Gately, M. K.,
`R. R. Warrier, S. Honasoge, D. A. Faherty, S. E. Con(cid:173)
`naughton, T. D. Anderson, U. Sarmiento, B. R. Hubbard,
`and M. Murphy, manuscript submitted for publication). Based
`on these in vivo results, the antitumor and antimetastatic ac(cid:173)
`tivities of murine IL-12 against a number of murine malig(cid:173)
`nancies have been evaluated. We demonstrate in this study
`that systemic administration of IL-12 can inhibit the growth
`of both established subcutaneous tumors and experimental
`pulmonary or hepatic metastases, and that local peritumoral
`injections of IL-12 can result in regression of established sub(cid:173)
`cutaneous tumors. Based on results obtained using mice
`deficient in lymphocyte subsets and antibody depletion ex(cid:173)
`periments, the antitumor efficacy of IL-12 is mediated pri(cid:173)
`marily through CDS+ T cells.
`
`1223
`
`J. Exp. Med. © The Rockefeller University Press • 0022-1007/93/10/1223/08 $2.00
`Volume 178 October 1993 1223-1230
`
`NOVARTIS EXHIBIT 2074
`Breckenridge v. Novartis, IPR 2017-01592
`Page 1 of 8
`
`

`

`pared in pristane-treated nude mice and lg purified by the proce(cid:173)
`dure of Reik et al. (21). Mice were injected intraperitoneally with
`1 mg oflg on the days indicated. Under these conditions, antibody
`treatment depleted >95% of the appropriate cell population in the
`spleen.
`
`Results
`We initially evaluated the effect of IL-12 on experimental
`pulmonary metastases of the B16F10 melanoma. After intra(cid:173)
`venous injection with tumor cells on day 0, C57BL/6 mice
`were treated intraperitoneally beginning on day 1 with varying
`doses of IL-12 five times per week for 3 wk. At the end of
`the treatment period, there was a dose-dependent inhibition
`of experimental metastases in the IL-12-treated mice (Fig.
`1 A). In the group receiving 1 µg per injection, the median
`number of metastases was reduced to 41 compared with 200
`in the diluent-treated mice (p <0.005). To determine if treat(cid:173)
`ment could be initiated at a time when metastases were al-
`
`200
`
`A
`
`V)
`
`uJ ; 150
`
`tj
`::Ii
`i...
`0
`a::
`uJ
`CD
`::Ii
`::::,
`z
`
`Downloaded from
`
`jem.rupress.org
`
` on February 22, 2018
`
`0
`
`0.1
`
`0.01
`
`0.001
`
`DOSE OF IL-12
`
`8
`
`100
`
`50
`
`0
`
`200
`
`100
`
`50
`
`uJ ; 150
`
`V)
`
`tj
`::Ii
`i...
`0
`a::
`uJ
`CD
`::Ii
`::::,
`z
`
`DILUENT
`
`IL-12
`
`TREATMENT
`
`Inhibition ofB16F10 metastases by IL12. (A) Mice were in(cid:173)
`Figure 1.
`jected intravenously with 2 x 105 B16F10 cells on day O and treatment
`with diluent or varying doses of IL 12 per injection was initiated in(cid:173)
`traperitoneally on day 1. Mice were treated five times per week until animals
`were killed on day 22. (B) Treatment with 1 µ,g per injection of IL 12
`five times per week was initiated on day 7 and mice were killed on day
`22. Data are represented as mean ± SE of 10 mice per group.
`
`Materials and Methods
`Mice. C57BL/6, C57BL/6 bglbg beige, C57BL/6 bgl+ het(cid:173)
`erozygous beige control and BALB/c mice, 6-8 weeks of age, were
`purchased from The Jackson Laboratories, (Bar Harbor, ME).
`BALB/c nude mice were obtained from Harlan-Sprague Dawley,
`Inc. (Madison, WI). Mice were routinely screened and found to
`be free of mycoplasma and selected murine viruses.
`Tumor Cell Lines. B16F10 malignant melanoma cells (15) and
`Renea renal cell adenocarcinoma cells (16), obtained from Dr. R.
`Wiltrout (Biological Response Modifiers Program, Frederick, MD),
`were maintained in vitro in RPMI 1640 medium supplemented
`with 10% fetal bovine serum, 100 IU/ml penicillin, 100 µg/ml
`streptomycin, 25 mM N-2-hydroxyethylpiperazine-N'-2-ethane(cid:173)
`sulfonic acid, 25 mM NaHCO3, and 60 µg/ml L-glutamine.
`M5076, a reticulum cell sarcoma (12, 17), was maintained as a sta(cid:173)
`tionary suspension culture in RPMI 1640 supplemented with 17%
`equine serum (Hyclone Laboratories, Inc., Logan, UT), 1 mM so(cid:173)
`dium pyruvate, 50 µg/ml gentamicin, glutamine, penicillin, and
`streptomycin. All reagents for cell culture were purchased from
`GIBCO BRL (Gaithersburg, MD) unless otherwise designated.
`All cell lines were found to be free of mycoplasma and viruses as
`above.
`Recombinant Murine IL12. Murine rIL-12 was expressed in CHO
`cells that had been stably transfected with the IL12 p40 and p35
`cDNAs and was purified as previously described (Gately, M. K.,
`et al., manuscript submitted for publication). SOS-PAGE analysis
`of the purified IL12 indicated it to be ;;i:95% pure with a small
`amount of contaminating IL12 p40 monomer. Monomeric IL12
`p40 does not bind to the IL12 receptor (18) and lacks biological
`activity (3, 4). Contamination by endotoxin, as assessed by the
`Limulus amebocyte assay, was <5 EU/mg IL12. The sp act of
`purified murine rlL 12, as determined by its ability to cause prolifer(cid:173)
`ation of human PHA blasts (18), was rv7 x 106 U/mg protein.
`For administration to mice, rIL 12 was diluted in PBS containing
`100 µg/ml mouse serum albumin (Miles Scientific, Naperville, IL
`or Sigma Chemical Co., St. Louis, MO).
`Tumor Experiments. Exponentially growing tumor cells were
`harvested by brief trypsinization (B16F10 and Renea), washed, and
`injected subcutaneously or intravenously into groups of 10 mice
`on day 0, and intraperitoneal treatment with various doses of IL 12
`once per day five times per week was initiated between days 1 and
`28, depending on the experiment. Specific protocols are presented
`in the legends of individual experiments. The diameters of subcu(cid:173)
`taneous tumors were measured twice weekly with calipers, and
`volume was calculated by the formula: Vol = (longest diameter)
`x (shortest diameter)2 (Gately, M. K., et al., manuscript sub(cid:173)
`mitted for publication). For metastasis experiments, approximately
`3 wk after intravenous injection of tumor cells, mice were killed
`and the number of metastases enumerated as previously described
`(15). Statistical evaluation of the data was performed using the non(cid:173)
`parametric one-tailed Mann-Whitney U test (17). Experiments were
`repeated two to eight times, and data from representative experi(cid:173)
`ments are presented.
`Antibody Depletion Experiments. Antiasialo GM1 (anti-ASGM1)
`was purchased from Biochemical Diagnostics, Inc. (Edgewood,
`NY). Mice were injected intraperitoneally with 0.4 ml of 1:5 dilu(cid:173)
`tion of anti-ASGM1 or normal rabbit serum on the days indicated.
`Under these conditions of treatment with antiasialo GM1 splenic
`NK activity, is reduced to undetectable levels (17).
`Anti-CD4 (clone GK1.5, rat IgG2b) (19) and anti-CDS (clone
`2.43, rat IgG2b) (20) hybridomas were purchased from the Amer(cid:173)
`ican Type Culture Collection (Rockville, MD). Ascites were pre-
`
`1224
`
`Antitumor Activity of IL 12
`
`NOVARTIS EXHIBIT 2074
`Breckenridge v. Novartis, IPR 2017-01592
`Page 2 of 8
`
`

`

`Downloaded from
`
`jem.rupress.org
`
` on February 22, 2018
`
`ready firmly established, mice were similarly injected intra(cid:173)
`venously with tumor cells on day 0, but the initiation of
`treatment was delayed until day 7. As shown in Fig. 1 B,
`under these experimental conditions, treatment with 1 µ,g
`of IL-12 also significantly reduced the number of experimental
`
`A
`
`8000
`
`7000
`
`6000
`
`5000
`
`4000
`
`3000
`
`2000
`
`1000
`
`.:,
`E
`5
`L,J
`::I:
`3
`§?
`0::
`0
`::I:
`~
`
`0
`5
`
`10
`
`0
`
`T
`
`•
`
`0 ./'
`•/Q
`(cid:143)
`V
`-----•
`-•====
`- · i
`20
`15
`TIME (days)
`
`25
`
`metastases (.p <0.0005). It is clear from these results that IL-12
`has potent antitumor activity in this experimental metastasis
`model and that treatment of animals with IL-12 can be initi(cid:173)
`ated after metastases have already been established.
`A series of experiments was then performed to determine
`if IL-12 also has activity against subcutaneously growing
`B16F10 tumors. Initially, mice were injected subcutaneously
`with 106 tumor cells on day 0, and treatment with varying
`amounts of IL-12 was initiated on day 7 when small tumors
`were present in the animals. As previously demonstrated with
`experimental metastases, there was a dose-dependent inhibi(cid:173)
`tion of tumor growth with maximal effects observed at a
`dose of 1 µ,g per injection (Fig. 2 A). At all doses tested,
`no gross toxicity was evident and all animals survived the
`therapy, although at the 5 µ,g dose some lethargy was ob(cid:173)
`served. To determine if IL-12 could increase the survival of
`B16F10 tumor-bearing mice and to test if the effect of
`prolonged IL-12 treatment on tumor growth, tumor-bearing
`mice were treated with 1 µ,g per injection of IL-12 five times
`per week until day 50. As seen in Fig. 2 B, there was a large
`increase in survival time of IL-12-treated mice compared with
`
`i 70
`
`100
`90
`80
`
`60
`50
`40
`30
`20
`10
`0
`15
`
`:::,
`en
`!z L,J
`u
`0::
`L,J a..
`
`B
`
`--------'\
`
`' ' - - - - -
`
`' ' ' -' I
`
`20
`
`25
`
`30
`
`35
`40
`TIME (days)
`
`45
`
`50
`
`55
`
`60
`
`C
`
`20000
`
`,;, 16000
`E
`5
`~ 12000
`3
`§?
`o:: 8000
`0
`::I:
`¢ 4000
`
`0
`
`r/
`0 T/1
`0 /'
`t•
`
`i'/ ---•-·-·/
`15
`20
`25
`30
`TIME (days)
`
`/
`•-•-·-·
`
`35
`
`40
`
`45
`
`50
`
`Figure 2.
`Inhibition of subcutaneous growth of B16F10 tumors. (A)
`Mice were injected subcutaneously with 1()6 B16F10 cells on day 0 and
`intraperitoneal treatment with diluent (0---0) or 5 µg (......,.), 1 µg
`(tr-A), 0.1 µg (......_.), or 0.01 µg (D---0) of IL-12 was initiated on
`day 7. Mice were treated five times per week until animals were killed
`on day 22. (B) Survival of mice treated intraperitoneally with diluent (solid
`line) or 1 µg ofIL-12 (dashed line) per injection five times per week through
`day 50. (q Treatment of mice was initiated on day 14. Mice were treated
`intraperitoneally with diluent (0---0) or 1 µg of IL-12 (......,.) per in(cid:173)
`jection five times per week for 3 wk. Data are represented as mean :±:
`SE of 10 mice per group.
`
`1225
`
`Brunda et al.
`
`i :::,
`
`en
`I-z
`L,J u
`0::
`L,J a..
`
`.:,
`E
`_§,
`"" ::I:
`::,
`...J g
`0::
`0
`:::E
`~
`
`100
`90
`80
`70
`60
`50
`40
`30
`20
`10
`0
`15
`
`--,
`\
`I
`I
`I
`I
`
`' \
`\
`
`20
`
`25
`
`30
`
`35
`40
`TIME (days)
`
`45
`
`B
`
`12000
`
`50
`
`55
`
`lo
`
`0
`
`8000
`
`4000
`
`·- -
`
`/
`T .....-o
`r_o
`. /9 l
`-.......-9
`•
`/0-9
`•-•---- !
`0
`•
`· - · - • - • - ·
`0 +---+---+---+--+--------1---t------t
`30
`25
`35
`40
`45
`50
`55
`60
`TIME (days)
`
`Figure 3.
`Inhibition of M5076 experimental hepatic metastases and sub(cid:173)
`cutaneous tumor growth by IL-12. (A) Survival of mice injected intrave(cid:173)
`nously on day 0 with 105 M5076 tumor cell and treated intraperitoneally
`with diluent (solid line) or 1 µg ofIL-12 (dashed line) per injection five times
`per week for 3 wk was initiated on day 1. (B) Mice were injected sub(cid:173)
`cutaneously with 1()6 M5076 tumor cells on day 0 and intraperitoneal
`treatment with diluent (0---0) or 1 µg of IL-12 (......,.) per injection
`was initiated on day 28. Animals were treated five times per week for
`4 wk. Data represented as mean ± SE of 10 mice per group.
`
`NOVARTIS EXHIBIT 2074
`Breckenridge v. Novartis, IPR 2017-01592
`Page 3 of 8
`
`

`

`Downloaded from
`
`jem.rupress.org
`
` on February 22, 2018
`
`diluent-treated control mice. Although tumor growth was
`suppressed for a prolonged time by IL-12 treatment, no "cures"
`were obtained, and tumors eventually grew, resulting in death
`of the animals. To evaluate the effect of IL-12 against larger
`B16F10 tumors, treatment with 1 µg of IL-12 per injection
`was delayed until day 14. As seen in Fig. 2 C, tumors grew
`progressively in control animals through day 29, at which
`point all animals had died. In contrast, the growth of tumors
`in mice treated with IL-12 was greatly diminished. These
`results demonstrate that IL-12 has marked antitumor activity
`against subcutaneous B16F10 tumors.
`Since antitumor activity of IL-12 had only been demon(cid:173)
`strated against the B16F10 melanoma, the antitumor efficacy
`of IL-12 was next evaluated against additional tumors. After
`intravenous injection of M5076 reticulum cell sarcoma, ex(cid:173)
`perimental hepatic metastases are formed (17). As seen in Fig.
`3 A, diluent-treated mice have a median survival of 21 d with
`all animals dead by day 26, but treatment of mice with IL-12
`results in an increase in median survival time to 38 d. After
`subcutaneous injection of M5076 tumor cells, tumors develop
`much more slowly than in mice injected with B16F10 mela(cid:173)
`noma cells. Thus, intraperitoneal treatment of mice with IL-12
`was begun 4 wk after implantation of tumor cells. Under
`these conditions, IL-12 markedly reduced tumor growth (Fig.
`3 B). In these IL-12-treated animals, spontaneous hepatic
`metastases are also greatly reduced (Brunda, M. J., and T. D.
`Anderson, unpublished observation). Therefore, the antitumor
`activity of IL-12 is evident against both subcutaneous tumor
`growth and experimental hepatic metastases of the M5076
`reticulum cell sarcoma.
`To evaluate further the antitumor activity of IL-12, mice
`were injected subcutaneously with the Renea renal cell ade(cid:173)
`nocarcinoma and treated with IL-12. The Renea renal cell
`adenocarcinoma is a murine tumor of spontaneous origin in
`BALB/c mice that mimics histologically renal cell carcinoma
`in humans (16) and is sensitive to various cytokines in vivo
`(16, 22). Beginning on day 14, intraperitoneal treatment of
`mice with IL-12 resulted in a dramatic inhibition of tumor
`growth (Fig. 4 A), although complete regression of tumors
`was not observed. Since it has previously been reported that
`regression of the Renea tumor occurs after transfection of
`IL-4 into this tumor cell (22), mice bearing day 14 subcuta(cid:173)
`neous Renea tumors were injected peritumorally with 1 µg
`of IL-12. As seen in Fig. 4 B, tumors were markedly inhibited
`and, in fact, tumor regression occurred with no tumor de(cid:173)
`tectable in 70% of treated mice after 60 d. Upon rechallenge
`with Renea cells alone, tumor growth is greatly inhibited
`in these survivors (data not shown). Thus, IL-12 has pro(cid:173)
`nounced antitumor activity against the Renea renal cell ade(cid:173)
`nocarcinoma and can, under certain conditions, result in regres·
`sion of established tumors.
`Since IL-12 has been demonstrated to have effects on both
`NK and T cells (1-10, 14, and Gately, M. K., et al., manu(cid:173)
`script submitted for publication), the antitumor activity of
`IL-12 might be mediated through either of these cell types.
`To determine if the antitumor efficacy of IL-12 was primarily
`mediated through NK cells, its activity was evaluated against
`
`1226
`
`Antitumor Activity of 11-12
`
`A
`
`~ 15000
`E
`_§,
`w
`::I! 10000
`3
`~
`ct::
`0
`::I!
`~
`
`5000
`
`0
`10
`
`15
`
`20
`
`30
`25
`TIME {days)
`
`35
`
`40
`
`45
`
`I')
`
`15000
`
`'"'
`E
`E
`'-'
`w
`::I!
`3 10000
`~
`ct::
`0
`::I!
`~
`
`5000
`
`0
`10
`
`15
`
`20
`
`25
`
`40
`35
`30
`TIME {days)
`
`45
`
`50
`
`55
`
`Inhibition of Renea tumors by 11-12. Mice were injected sub(cid:173)
`Figure 4.
`cutaneously on day O with 1()6 Renea tumor cells and treatment with
`diluent (o---0) or 1 µg of 11-12 (....._..) per injection five times per week
`for 4 wk was initiated on day 14. (A) Treatment by intraperitoneal route.
`(B) Treatment by peritumoral route. Data represented as mean ± SE of
`10 mice per group.
`
`B16F10 tumors in NK cell deficient bglbg beige (23). Treat(cid:173)
`ment of either bglbg beige or phenotypically normal bgl+
`heterozygotes with 1 µg per injection of IL-12 resulted in
`a comparable degree of tumor inhibition (Fig. 5). To confirm
`and extend these results, Renea tumor-bearing BALB/c mice
`were depleted ofNK cells using anti-ASGM1 and treated with
`IL-12. No loss ofIL-12-induced antitumor efficacy was ob(cid:173)
`served in these NK cell-depleted mice (Fig. 6). Taken together,
`these results strongly suggest that the antitumor efficacy of
`IL-12 is not mediated through NK cells.
`To evaluate the potential role of T cells on IL-12-induced
`antitumor efficacy, the antitumor effect of IL-12 was tested
`against B16F10 tumors in nude mice (Fig. 7 A). Although
`there was a slight delay in tumor growth, the effect of IL-12
`was markedly diminished compared with that observed in
`euthymic mice. The greatly reduced activity of IL-12 was
`also obtained in BALB/c nude mice bearing syngeneic Renea
`tumors (Fig. 7 B). To evaluate further the requirement for
`T cells in mediating IL-12-induced antitumor efficacy and
`to determine the role ofT cell subsets, Renea tumor-bearing
`mice were treated with anti-CD4 or anti-CDS antibodies.
`As seen in Fig. 8, the activity of IL-12 is not affected in mice
`depleted of CD4 + T cells but substantially, although not
`
`NOVARTIS EXHIBIT 2074
`Breckenridge v. Novartis, IPR 2017-01592
`Page 4 of 8
`
`

`

`Downloaded from
`
`jem.rupress.org
`
` on February 22, 2018
`
`40000
`
`uJ
`
`'130000
`5
`:::I: 3 20000
`~
`0::
`0
`~ 10000
`I-
`
`20000
`
`A
`
`,;;-' 16000
`E
`5
`~ 12000
`~ et: 8000
`~
`~ 4000
`
`0+-----1--------t-----+------1
`30
`25
`20
`15
`10
`TIME (days)
`
`05
`
`10
`
`15
`
`25
`20
`TIME (days)
`
`30
`
`35
`
`40
`
`Figure 5. Antitumor effects of IL-12 in NK cell-deficient beige mice.
`Animals were injected subcutaneously with 1()6 B16F10 tumor cells on
`day O and intraperitoneal treatment with diluent (open symbols) or 1 µ.g
`of IL-12 (filled symbols) per injection five times per week until the end of
`the experiment was initiated on day 14 in bg I+ heterozygous mice ( 0 I e)
`or bg/bg beige mice (Al.&.). Data are represented as mean ± SE of 10
`mice per group.
`
`completely, lost in mice treated with anti-CDS. These results
`demonstrate that CDS+ T cells are critical for mediating the
`antitumor effects of IL-12.
`
`Discussion
`Using purified recombinant murine IL-12, we have demon(cid:173)
`strated that this cytokine has potent antitumor and an(cid:173)
`timetastatic activity against a number of murine tumors of
`various histological types. Therapeutic intervention by sys(cid:173)
`temic administration of IL-12 can be initiated when tumors
`or metastases are well established, up to day 28 after injec-
`
`8000
`
`,;;-' 6000
`E
`5
`
`4000
`
`2000
`
`0 +----1----+----+---l----+-----I
`10
`15
`20
`25
`30
`35
`40
`TIME (days)
`
`Figure 6. Antitumor effects ofIL-12 in mice treated with anti-ASGM1.
`Animals were injected subcutaneously with 1()6 Renea tumor cells on day
`0 and intraperitoneal treatment with diluent (0--0) or IL-12 ( ........ )
`was initiated on day 14 and continued five times per week for 4 wk. Other
`groups were treated with IL-12 beginning on day 14 and were injected
`intraperitoneally with either NRS (6----ll.) or anti-ASGM1 (..-.) on
`days 15, 17, 21, 24, 28, and 31. Data are represented as mean ± SE of
`10 mice per group.
`
`1227
`
`Brunda et al.
`
`B
`
`12000
`
`,;;-'
`E
`5
`uJ
`:::I:
`::::,
`_J
`~
`0::
`0
`::IE
`~
`
`8000
`
`4000
`
`0
`
`5
`
`I
`
`l/r
`
`T
`
`T ,,...-t:.
`/ t l __.-----o
`i
`0
`T e O
`T / 1 __.,.---e
`•
`1
`~~ / • - ! / .
`.~e-&==·..L·-•-•-·
`__.-----f
`~A
`/2~
`e
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`TIME (days)
`
`Figure 7. Reduced efficacy of IL-12 in tumor-bearing nude mice. Mice
`were treated intraperitoneally five times per week for 4 wk with diluent
`(open symbols) or 1 µ.g IL-12 (cfused symbols) beginning on day 7. Data are
`represented as mean ± SE of 10 mice per group. (A) B6 (Al .A.) or BALBlc
`nude mice (Ole) were injected subcutaneously with 1()6 B16F10 tumor
`cells on day 0. (B) BALBlc (Al.&.) or BALBlc nude mice (Ole) were
`injected subcutaneously with 1()6 Renea tumor cells on day 0.
`
`16000 . . - - - - - - - - - - - - - - - - - - - - , , - - - - - ,
`
`uJ
`
`~ ,...,E 12000
`5
`::IE 3
`~
`Cl:'.
`0
`::::;
`~
`
`8000
`
`4000
`
`0 +---+---+----+----+------1----+------lf--~
`10
`15
`20
`25
`30
`35
`40
`45
`TIME (days)
`Figure 8. Effect of depletion of CD4 + or CDS+ T cells on antitumor
`efficacy of IL-12. Mice were injected subcutaneously with 1()6 Renea tumor
`cells on day O and intraperitoneal treatment five times per week until the
`end of the experiment with diluent (0--0) or 1 µ.g IL-12 ( ........ ) was
`initiated on day 14. Other groups of mice were injected with IL-12 and
`either anti-CD4 (6----ll.) or anti-CDS (..-.) on days 14, 21, and 28.
`Data are represented as mean ± SE of 10 mice per group.
`
`NOVARTIS EXHIBIT 2074
`Breckenridge v. Novartis, IPR 2017-01592
`Page 5 of 8
`
`

`

`Downloaded from
`
`jem.rupress.org
`
` on February 22, 2018
`
`tion of tumor cells in the M5076 model, resulting in inhibi(cid:173)
`tion of tumor growth, reduction in the number of metastases
`and an increase in survival time. Furthermore, after local
`peritumoral injections of IL-12, established subcutaneous
`tumors regress, resulting in animals that are tumor free.
`The mechanism through which IL-12 exerts its antitumor
`activity is at present unknown, but based on its established
`in vivo activities several potential hypotheses can be suggested.
`IL-12 may have direct growth inhibitory effects on tumor
`cells. This is unlikely since addition of IL-12 in vitro to the
`murine tumor cells used in our studies resulted in no inhibi(cid:173)
`tion of tumor cell proliferation (Brunda, M. J., and L. Luistro,
`unpublished observation) and there is reduced activity of this
`cytokine against tumors in nude mice (Fig. 7). However, IL-12
`has been shown to induce production and secretion of IFN--y
`(1, 10, and Gately, M. K., et al., manuscript submitted for
`publication), which might then in turn inhibit proliferation
`of tumor cells (24). Such a proposed mechanism would be
`consistent with the lack of direct in vitro antiproliferative
`activity on tumor cells and the observed in vivo efficacy of
`IL-12. IFN--y has been previously shown to have activity in
`vivo against murine tumors, although not to the extent ob(cid:173)
`served with IL-12 (12, 13). Both T and NK cells can produce
`IFN--y in response to IL-12 in vitro (1, 10), but if this hypoth(cid:173)
`esis is correct, it would appear that T cell production of IFN--y
`is critical to the antitumor activity of IL-12 in vivo because
`the efficacy of IL-12 is greatly reduced in athymic nude mice
`(Fig. 7). It is unlikely that the antitumor efficacy of IL-12
`is directly an effect of the level of IFN--y induced since IL-12
`induces an approximately fivefold higher level of IFN--y in
`nude mice than in euthymic mice (Hendrzak, J., and M.
`Brunda, unpublished observation) but IL-12 has a substan(cid:173)
`tially reduced antitumor effect in nude mice (Fig. 7). IL-12
`retains some activity in nude mice, which may reflect the
`induction of IFN--y or other cytokines. Treatment of mice
`with antibodies to IFN--y may resolve some of these issues
`but in other studies (25, 26) this approach yielded conflicting
`results. Recently, IFN--y knock out mice have been devel(cid:173)
`oped (27) and the evaluation of the antitumor activity of IL-12
`in these animals could prove interesting.
`In addition to secretion of IFN--y, stimulation of human
`NK or T cells with human IL-12 results in secretion of tumor
`necrosis factor (6), which can have antiproliferative effects
`on tumor cells (28). Recent data using cells from SCID mice
`demonstrate that incubation of these cells with murine IL-12
`resulted in no secretion of TNF in vitro using conditions
`that induced IFN--y (29). Likewise, in preliminary experi(cid:173)
`ments, no TNF was induced from normal murine spleen cells
`stimulated by murine IL-12 in vitro (Hendrzak, J., and M.
`Brunda, unpublished observations). Thus, it appears that IL-12
`is a relatively poor inducer of TNF in mice. However, the
`contribution ofTNF or potentially other cytokines in medi-
`
`ating the antitumor effects of IL-12 in vivo needs to be ad(cid:173)
`dressed in future experiments.
`IL-12 can both enhance the growth (5-7) and augment
`the cytolytic activity (1, 7-9, and Gately, M. K., manuscript
`submitted for publication) of NK/LAK. and T cells. Thus,
`the antitumor efficacy of IL-12 might be mediated through
`stimulation of one or both these lymphocyte populations.
`It is unlikely that NK cells are primarily involved since the
`antitumor activity of IL-12 is maintained in both NK
`cell-deficient beige mice (Fig. 5) and in mice depleted of NK
`cells with anti-ASGM1 (Fig. 6). However, T cells appear to
`be necessary to obtain maximal effects since the efficacy of
`IL-12 treatment is greatly diminished in nude mice (Fig. 7).
`Under other experimental conditions, the antitumor activity
`of a number of cytokines has been reported to be dependent
`upon T cells (22, 30, 31). It has been suggested that IL-12
`promotes the development of Thl cells (32, 33), and these
`in turn may positively regulate the expansion and/or activa(cid:173)
`tion of other lymphoid cell populations. After depletion of
`T cell subsets with mAbs, the efficacy ofIL-12 was decreased
`in mice depleted of CDS+ T cells but remained intact in
`mice depleted of CD4 + T cells (Fig. 8). These results indi(cid:173)
`cate that CDS+ T cells are the critical cell type for medi(cid:173)
`ating the antitumor efficacy of IL-12. The efficacy of IL-12
`is not completely eliminated in CDS+ T cell-depleted mice.
`This finding may reflect the effect of IL-12 on the few re(cid:173)
`maining CDS+ T cells or on other cell types, such as mac(cid:173)
`rophages. It appears that CD4 + T cells are not necessary for
`mediating the anti tumor effects of IL-12, which is puzzling
`in light of the effects of IL-12 on CD4 + Thl cells (32, 33).
`However, since IL-12 also stimulates cytokine secretion from
`NK cells (1, 10), the cytokine-secreting function of CD4 +
`T cells may be replaced by NK cells under the conditions
`used in these tumor therapy experiments. In addition, ex(cid:173)
`ogenously administered IL-12 may at least partially replace
`any requirement for endogenously produced IL-2, since the
`biological activities of IL-12 substantially overlap those of IL-2.
`A number of cytokines, including IFN-a and IL-2, have
`been demonstrated to be active in murine tumor models (11-13,
`15, 30) and, subsequently, to be useful in the treatment of
`human malignancies (11, 34, 35). In contrast, with other
`cytokines, such as IFN--y or TNF, the activity demonstrated
`in animal models (11-13, 31) has not translated to successful
`use of these proteins for therapy of human malignancies (11,
`34). The data presented in this study clearly establish IL-12
`as a cytokine that has potent antitumor and antimetastatic
`activities in several murine tumor models through an immune(cid:173)
`mediated, T cell-dependent mechanism. Future clinical trials
`with this cytokine will determine if the activity demonstrated
`in animals can be translated into efficacy against human malig(cid:173)
`nancies.
`
`Address correspondence to Dr. Michael]. Brunda, Department of Oncology, Hoffmann-La Roche Inc.,
`Nutley, NJ 07710.
`
`Received far publication 7 April 1993 and in revised farm 9 June 1993.
`
`1228
`
`Antitumor Activity of IL12
`
`NOVARTIS EXHIBIT 2074
`Breckenridge v. Novartis, IPR 2017-01592
`Page 6 of 8
`
`

`

`Downloaded from
`
`jem.rupress.org
`
` on February 22, 2018
`
`References
`1. Kobayashi, M., L. Fitz, M. Ryan, R.M. Hewick, S.C. Clark,
`S. Chan, R. Loudon, F. Sherman, B. Perussia, and G. Trin(cid:173)
`chieri. 1989. Identification and purification of natural killer
`cell stimulatory factor (NKSF), a cytokine with multiple bio(cid:173)
`logical effects on human lymphocytes.]. Exp. Med. 170:827.
`2. Stern, A.S., F.J. Podlaski, J.D. Hulmes, Y.E. Pan, P.M. Quinn,
`A.G. Wolitzky, P.C. Familletti, D.L. Stremlo, T. Truitt, R.
`Chizzonite, and M.K. Gately. 1990. Purification to homoge(cid:173)
`neity and partial characterization of cytotoxic lymphocyte matu(cid:173)
`ration factor from human B-lymphoblastoid cells. Proc. Natl.
`Acad. Sci. USA. 87:6808.
`3. Wolf, S.F., P.A. Temple, M. Kobayashi, D. Young, M. Dicig,
`L. Lowe, R. Dzialo, L. Fitz, D. Ferenz, R.M. Hewick, et al.
`1991. Cloning of cDNA for natural killer cell stimulatory factor,
`a heterodimeric cytokine with multiple biological effects on
`T and natural killer cells. J. Immunol. 146:3074.
`4. Gubler, U., A.O. Chua, D.S. Schoenhaut, C.M. Dwyer, W.
`McComas, R. Motyka, N. Nabavi, A.G. Wolitzky, P.M.
`Quinn, P.C. Familletti, and M.K. Gately. 1991. Coexpression
`of two distinct genes is required to generate secreted, bioac(cid:173)
`tive cytotoxic lymphocyte maturation factor. Proc. Natl. Acad.
`Sci. USA. 88:4143.
`5. Gately, M.K., B. Desai, A.G. Wolitzky, P.M. Quinn, C.M.
`Dwyer, F.J. Podlaski, P.C. Familletti, F. Sinigaglia, R. Chiz(cid:173)
`zonite, U. Gubler, and A.S. Stern. 1991. Regulation of human
`lymphocyte proliferation by a heterodimeric cytokine, IIA2
`(cytotoxic lymphocyte maturation factor). J. Immunol. 147:874.
`6. Perussia, B., S.H. Chan, A. D'Andres, K. Tsuji, D. Santoli,
`M. Pospisil, D. Young, S.F. Wolf, and G. 'Iiinchieri. 1992. Nat(cid:173)
`ural killer (NK) cell stimulatory factor or IL-12 has differential
`effects on the proliferation of TCR-afj+, TCR--yo+ T lym(cid:173)
`phocytes, and NK cells. J. Immunol. 149:3495.
`7. Robertson, M.J., R.J. Soiffer, S.F. Wolf, T.J. Manley, C.
`Donahue, D. Young, S.H. Herrmann, and]. Ritz. 1992. Re(cid:173)
`sponses of human natural killer (NK) cells to NK cell stimula(cid:173)
`tory factor