The immunosuppressive macrolide RAD inhibits
`growth of human Epstein–Barr virus-transformed
`B lymphocytes in vitro and in vivo: A potential
`approach to prevention and treatment of
`posttransplant lymphoproliferative disorders
`
`Miroslaw Majewski*, Magdalena Korecka*, Plamen Kossev*, Shiyong Li*, June Goldman*, Jonni Moore*,
`Leslie E. Silberstein*, Peter C. Nowell*, Walter Schuler†, Leslie M. Shaw*, and Mariusz A. Wasik*
`
`*Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA 19104; and †Novartis Pharma, Basel CH-4002, Switzerland
`
`MEDICALSCIENCES
`
`Herein, we report that RAD has potent inhibitory activity also
`on PTLD-like, human EBV⫹ lymphoblastoid B cell lines both
`in vitro and in vivo. RAD profoundly inhibited in vitro prolifer-
`ation of such cells and arrested their cell-cycle progression at the
`early G0兾G1 stage. In addition, the compound increased apo-
`ptotic rate of the EBV⫹ B cells. In vivo, it markedly delayed or
`completely inhibited growth of the EBV⫹ B cells xenotrans-
`planted into SCID mice, particularly when administered before
`the cell implantation. RAD was able to eradicate the established
`tumor in some instances; this effect seemed to be cell line specific
`and proportional to the drug effect seen in vitro. Clinical
`implications of these findings in regard to prevention and
`treatment of PTLDs are discussed.
`
`Materials and Methods
`Cell Lines. All B cell lines used in this study, with the exception
`of the BC-1 cell line, were lymphoblastoid B cell lines obtained
`by in vitro infection with EBV of peripheral blood mononuclear
`cells. Cell lines A1 and A2D6 were obtained from normal,
`healthy individuals. Cell lines 15A and 20A were obtained from
`two different patients with low-grade B cell lymphomas with
`monoclonal cold agglutinins (15). Both lines secreted cold
`agglutinins with the same specificity as the cold agglutinins
`found in the patients’ serum (15). Furthermore, the 20A cell line
`showed cytogenetic abnormalities seen in low-grade lymphomas:
`trisomy 3 and 12 (48, XX, ⫹3, ⫹12; ref. 15). The LCL EBV⫹ B
`cell line was obtained from peripheral blood mononuclear cells
`of a patient with a progressive cutaneous T cell lymphoprolif-
`erative disorder (16). BC-1 was derived from a primary effusion
`B cell lymphoma and, in addition to EBV, harbors HHSV8 virus
`(17). The other three cell lines, used as controls in the in vitro
`growth inhibition assay, were HTLV-I (⫹) T cell lines ATL-2 and
`C10MJ2 as well as HUT102B derived from patients with adult
`T cell leukemia兾lymphoma; these cell lines had been determined
`previously by us to be nonresponsive to RAD and rapamycin
`(18). All cell lines were maintained in humidified incubators at
`37°C with 5% CO2 in standard medium: RPMI medium 1640
`(GIBCO兾BRL) supplemented with 10% (vol兾vol) heat-
`inactivated FBS (BioWhittaker), 1% penicillin兾streptomycin兾
`
`Abbreviations: PTLD, posttransplant lymphoproliferative disorder; EBV, Epstein–Barr virus.
`‡To whom reprint requests should be addressed. E-mail: wasik@mail.med.upenn.edu.
`
`The publication costs of this article were defrayed in part by page charge payment. This
`article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C.
`§1734 solely to indicate this fact.
`Article published online before print: Proc. Natl. Acad. Sci. USA, 10.1073兾pnas.080068597.
`Article and publication date are at www.pnas.org兾cgi兾doi兾10.1073兾pnas.080068597
`
`PNAS 兩 April 11, 2000 兩 vol. 97 兩 no. 8 兩 4285– 4290
`
`Contributed by Peter C. Nowell, February 16, 2000
`
`Whereas the standard immunosuppressive agents foster develop-
`ment of posttransplant lymphoproliferative disorders (PTLDs), the
`impact of RAD, a macrolide with potent immunosuppressive prop-
`erties, and other immunosuppressive macrolides on these disor-
`ders remains undetermined. We found that RAD had a profound
`inhibitory effect on in vitro growth of six different PTLD-like
`Epstein–Barr virusⴙ lymphoblastoid B cell lines. Similar to normal
`T cells, RAD blocked cell-cycle progression in PTLD-like B cells in the
`early (G0兾G1) phase. Furthermore, RAD increased the apoptotic rate
`in such cells. The drug also had a profound inhibitory effect on the
`growth of PTLD-like Epstein–Barr virusⴙ B cells xenotransplanted
`s.c. into SCID mice. The degree of the RAD effect varied among the
`three B cell lines tested and was proportional to its effects on the
`cell
`lines in vitro.
`In this in vivo xenotransplant model, RAD
`markedly delayed growth or induced regression of the established
`tumors. In one line, it was able to eradicate the tumor in four of
`eight mice. When RAD treatment was initiated before tumor cell
`injection, a marked inhibition of tumor growth was seen in all three
`lines. In two of them, the drug prevented tumor establishment in
`approximately 50% of mice (5兾11 and 5兾8). In summary, RAD is a
`potent inhibitor of PTLD-like cells in vitro and in vivo. These
`findings indicate that, in contrast to the standard immunosuppres-
`sive agents, macrolides such as RAD may be effective in prevention
`and treatment of PTLDs.
`
`Posttransplant lymphoproliferative disorders (PTLDs), which
`
`usually represent expansion of B lymphocytes infected with
`Epstein–Barr virus (EBV), are a life-threatening complication of
`the immunosuppressive therapy necessary to prevent graft re-
`jection (1, 2). PTLDs comprise a whole spectrum of lympho-
`proliferative disorders ranging from a polyclonal atypical lym-
`phoid hyperplasia to a monoclonal, overtly malignant B cell
`lymphoma (1–4). Less advanced forms of PTLDs respond to a
`decrease in the dose of the immunosuppressive agents (1, 5).
`However, lowering the drug dose jeopardizes survival of the
`graft. Furthermore, it is not effective in the more malignant,
`lymphoma-type cases of PTLDs, which are usually fatal for the
`graft recipient. Clearly, more effective therapeutic and preven-
`tive measures are required to limit the severity and frequency of
`PTLDs.
`RAD (SDZ RAD) is a macrocyclic lactone with potent
`immunosuppressive and antiproliferative properties (6–10).
`Like rapamycin, a compound from which RAD was derived by
`chemical derivation, RAD inhibits growth factor-induced pro-
`liferation of hematopoietic as well as nonhematopoietic cells (7).
`It has been shown earlier for rapamycin that this compound
`inhibits intracellular signaling events downstream of the receptor
`for IL-2 and also other cytokine receptors (11, 12) and arrests the
`cell-cycle progression at the early G1 phase (13, 14).
`
`West-Ward Exhibit 1075
`Majewski 2000 - Page 001
`
`

`

`fungizone mixture (GIBCO兾BRL), and 2 mmol兾liter L-
`glutamine (GIBCO兾BRL).
`
`Inhibition of inVitroCell Growth by RAD. The assay was performed
`as described (16, 18). Briefly, cell lines were cultured for 32 h in
`triplicate at 2 ⫻ 104 cells per well in the presence of various
`concentrations of RAD (Novartis Pharma). After pulse with
`0.5 ␮Ci [3H]thymidine (New England Nuclear) and culture for
`the next 18 h, isotope incorporation to the cells was measured.
`The results of proliferation assays were expressed as the mean
`radioactivity of triplicate cultures. Standard deviation within the
`triplicates was ⬍15%.
`
`Detection of Cell-Cycle Inhibition and Apoptosis. The cell lines to be
`examined were cultured with several concentrations of RAD (0
`to 10 nM) for 24–48 h. The cells were washed with Dulbecco’s
`PBS and stain solution (pH 7.2) containing 3% (wt兾vol) poly-
`ethylene glycol (molecular weight ⫽ 6,000), 50 ␮g兾ml DNA
`fluorochrome propidium iodide (Calbiochem), annexin V-FITC
`as described (20), 0.1% Triton X-100 (Sigma), 4 mM citrate
`buffer (pH 7.8), and 360 units兾ml RNase A (Worthington) for
`30 min at 37°C. Next, salt solution [pH 7.2; 3% (wt/vol) poly-
`ethylene glycol (molecular weight ⫽ 6,000)兾50 ␮g/ml propidium
`iodide兾0.1% Triton X-100兾0.4 M NaCl] was added, and the cells
`were incubated at 4°C in the dark for 1 h before flow cytometry
`analysis (19–21).
`
`Mice. Immunodeficient 5- to 7-week-old SCID mice (C.B-17 and
`ICR) were purchased from Taconic Farms, housed at the
`University of Pennsylvania Animal Facility under pathogen-free
`conditions in a laminar air flow unit, and supplied with sterile
`food and water. In the drug tolerability studies, 5- to 7-week-old
`inbred BALB兾c mice (Taconic Farms) were used in addition to
`the SCID mice.
`
`Establishment of the PTLD-Like Tumors in SCID Mice. Establishment
`and passaging of the xenotransplanted lymphoma tumors was
`performed as described (22–24). To deplete macrophages and
`natural killer cells and to enhance tumor engraftment, SCID
`mice were injected i.p. with 30–45 mg兾kg of etoposide (Bedford
`Laboratories, Bedford, OH) 4 days before implantation of the
`human EBV⫹ B cell lines (24). Cells (n ⫽ 10 million) of each line
`(see Results) were inoculated into mice either i.p. or s.c. in
`200 ␮l of Dulbecco’s PBS (BioWhittaker). Ascites or palpable
`s.c. tumors developed 3–5 weeks after cell injection. The tumor
`treatment and growth prevention experiments were performed
`by using fragments of the established s.c. tumors (22–24). For
`this purpose, mice were anesthetized with Ketalar (ketamine,
`Parke-Davis) by i.p. injection of 100 mg兾kg. Next, the primary
`tumor was aseptically removed and freed from necrotic, fatty,
`and connective tissue and divided into small pieces of roughly
`equal size, and three or four pieces per mouse were injected s.c.
`Treatment with RAD of established tumors was started when the
`tumors reached 5 mm in diameter. In growth prevention exper-
`iments, the treatment was initiated 3 days before tumor implan-
`tation. Tumor volume in all experiments was determined from
`the equation, volume ⫽ 0.4ab2, where a and b designate,
`respectively, long and short diameters of the tumor. The trans-
`planted mice were monitored for tumor growth for a period of
`up to 2 months, and 5 mg兾kg of RAD was given once a day by
`gavage as described (6, 7).
`
`RAD-mediated inhibition of in vitro proliferation of PTLD-like EBV⫹
`Fig. 1.
`B cells. BC-1 is an EBV⫹兾HSV8⫹ B cell line derived from primary effusion
`lymphoma. The other cell lines are in vitro EBV-transformed B cell lines derived
`from patients with low-grade B cell lymphoma (15A and 20A), a patient with
`T cell lymphoma (LCL), or healthy individuals (A1 and A2D6). HTLV-I⫹ malig-
`nant T cell lines HUT-102, C10MJ, and ATL-2 served as controls. The cell lines
`were pulsed for 18 h with tritiated thymidine (TdR) after 32 h culture with
`0 –10 nM of RAD.
`
`weakness, or lethargy appeared. A complete autopsy was per-
`formed on all mice at the end of the study regardless of their
`appearance. Tumor and internal organs (spleen, liver, lung,
`heart, kidney, small and large intestines, and femoral bone for
`bone marrow) were fixed in 10% (vol兾vol) formalin, paraffin
`embedded, cut into 0.4-␮m sections, transferred to glass slides,
`and stained with hematoxylin and eosin. Representative tumor
`fragments were stained immunohistochemically by the standard
`streptavidin-biotin complex technique with commercially avail-
`able reagents (Research Genetics, Huntsville, AL) with the
`following antibodies: anti-CD20 (L-26) and LMP-1 (both from
`Dako) and Ki-67 (mib1; Immunotech, Westbrook, ME). EBV-
`encoded RNA was detected with commercially available re-
`agents (Dako).
`
`Results
`RAD Inhibits Growth of EBVⴙ B Cells inVitro. To determine whether
`RAD can inhibit proliferation of cells mimicking PTLDs, we
`cultured six different EBV⫹ B cell lines in the presence of the
`drug at various concentrations. Three HTLV-I⫹ T cell lines
`resistant to RAD (18) were used as controls. As shown in Fig.
`1, all PTLD-like EBV⫹ B cell lines were very sensitive to RAD.
`A dose of RAD as small as 1 nM produced 60–95% inhibition
`of growth in all cell lines. This result was comparable to the
`inhibition of stimulated normal T lymphocytes (7, 25). Some
`subtle differences in the degree of response were noted among
`the EBV⫹ B cell lines. The lines derived from patients with B
`cell lymphomas (15A, 20A, and BC-1; see Materials and Meth-
`ods), and thus resembling the more advanced forms of PTLD,
`tended to show a lower degree of inhibition (80–90%), whereas
`the lines obtained from normal B cells, and thus mimicking the
`less advanced types of PTLD, were inhibited more profoundly
`(90–100%).
`
`Macroscopic and Microscopic Evaluation of Organs and Xenotrans-
`planted Tumors. Mice were killed by exposure to forane (isoflu-
`rane, Ohmeda, Liberty Place, NJ) on day 29 in the drug toxicity
`study, on day 40–55 in the tumor growth inhibition study, or
`when tumors achieved approximately 2 cm in diameter or when
`ulceration of the skin, signs of severe respiratory distress,
`
`RAD Blocks Cell-Cycle Progression in EBVⴙ B Cells. To elucidate at
`which stage of the cell cycle RAD inhibits the proliferation of
`PTLD-like B cells, we analyzed the DNA content of the cultured
`cells by means of flow cytometry. As shown in Fig. 2, treatment
`with RAD led to an increase in the number of cells in the G0兾G1
`phase of the cell cycle, indicating an arrest of cell-cycle progres-
`
`4286 兩 www.pnas.org
`
`Majewski et al.
`
`West-Ward Exhibit 1075
`Majewski 2000 - Page 002
`
`

`

`MEDICALSCIENCES
`
`cell-cycle inhibition data. In the most sensitive line, A1, RAD
`resulted in an increase in the number of apoptotic cells from 10
`to 16%. RAD led to an a rather moderate increase from 11 to
`13% in the A2D6 line and from to 6 to 8% in the 20A line. At
`the highest dose of RAD (10 nM), 24% of A1 cells, 16% of A2D6
`cells, and only 9% of 20A cells were apoptotic.
`
`RAD Tolerability Study. Previous studies have shown that a min-
`imal effective immunosuppressive dose of RAD was 5 mg兾kg兾
`day and ⬎5 mg兾kg兾day in the rat kidney and heart allotransplant
`models, respectively, when RAD was used as a single immuno-
`suppressive agent (6, 7). Because there was only very limited
`experience with RAD in mice, we tested how prolonged expo-
`sure to the compound was tolerated by immunocompetent and
`immunodeficient mice. To this end, we treated a cohort of
`normal BALB兾c mice with a dose of 5 mg兾kg兾day RAD for 28
`days; 10 treated and 5 control untreated mice were used in this
`study. Subsequently, we also analyzed SCID mice that had been
`transplanted with human PTLD-like lymphomas and treated
`with the same dose of RAD for up to 55 days (see below). Seven
`treated and five control untreated SCID mice were evaluated in
`this experiment. After the last dose of the drug, all mice were
`killed, and the following organs were harvested for histopatho-
`logic evaluation of possible toxic effects: liver, spleen, kidney,
`small intestine, large intestine, heart, lung, and femoral bone for
`bone marrow. None of the treated mice showed any visible signs
`of drug toxicity. Growth and weight gain were the same for the
`RAD-treated and the control group. Microscopic evaluation of
`the organs of all treated and untreated mice revealed no
`pathologic changes that could be attributed to the drug. We
`conclude, therefore, that a prolonged exposure to RAD at the
`dose of 5 mg兾kg兾day has no adverse effect in the treated mice
`regardless of their immune status.
`
`Establishment of the Xenotransplant Model of Human PTLD-Like
`Lymphoma in SCID Mice. In these experiments, we injected mice
`with EBV⫹ B cell lines via two different routes: i.p. and s.c. In
`the i.p. tumor model, SCID mice (five per group) were inocu-
`lated i.p. with 107 cells per line from four cell lines: 15A, 20A,
`A2D6, and BC-1. After 21–35 days, all mice developed fatal
`disease with ascites and symptoms of weakness or lethargy.
`Autopsy revealed extensive tumor infiltrates involving the peri-
`toneal wall, liver, spleen, and kidneys. Microscopic examination
`of the tumors showed a large cell lymphoma with an infiltrative
`growth pattern, high mitotic rate, focal and single-cell necrosis,
`and some degree of plasmacytoid differentiation. Foci of the
`lymphoma were also present in distant organs such as lungs and
`bone marrow, indicating hematogenous spread. In the s.c. model,
`mice (also five per group) were injected with 107 cells from the
`15A, 20A, and A1 lines. After 21–32 days, all mice developed
`tumors at the site of implantation. Autopsy revealed s.c. tumors
`invading adjacent skeletal muscle and skin. There was no gross
`evidence of distant spread of the tumor. Microscopic examina-
`tion revealed the same type of high-grade lymphoma as in the i.p.
`model. Distant internal organs, mainly liver and lungs, occasion-
`ally had small tumor foci. These foci occurred only when the s.c.
`lymphoma reached a size of at least 1.5 cm in diameter.
`Immunohistochemical staining confirmed that the tumors were
`derived from the implanted human EBV⫹ B lymphocytes. The
`images of the 20A line shown in Fig. 4 A–D are representative
`for all three cell lines. Virtually all lymphoma cells were positive
`for the human B cell marker CD20. Most (50–80%) were
`positive for cell-cycle related Ki-67 antigen consistent with their
`high proliferative rate. Staining for EBV-related antigen EBV-
`encoded RNA1 (EBER1) was universally positive (100% of
`cells); 20–50% of cells expressed EBV-associated latent mem-
`brane protein 1 (LMP-1; not shown). These results confirm that
`
`Fig. 2. RAD-mediated inhibition of cell-cycle progression in PTLD-like B cells.
`Four EBV⫹ B cell lines were cultured for 48 h with 0 –10 nM of RAD, labeled
`with propidium iodine, and analyzed by flow cytometry.
`
`sion at the early G0兾G1 phase. This arrest was observed in all four
`PTLD-like B cell lines investigated: 20A, A1, A2D6, and LCL.
`The effect was drug-dose and cell-line-type dependent. Whereas
`the low RAD doses (1–2 nM) increased the percentage of cells
`in G0兾G1 by 5–25%, the highest dose tested (10 nM) resulted in
`a 10–70% increase. Percentages of cells in the later phases of the
`cell cycle (G2兾M and S) were diminished proportionally (data
`not shown). A1 and, to a lesser degree, the A2D6 cell line were
`particularly sensitive to RAD.
`
`RAD Increases Apoptosis in EBVⴙ B Cells. Previous studies have
`shown that immunosuppressive macrolides as represented by
`rapamycin can induce or enhance apoptosis stimulated by other
`agents (26–29). To determine whether RAD affects the apo-
`ptotic rate of PTLD-like cells, we tested by flow cytometry the
`apoptosis-induced cell membrane binding of annexin V in the
`20A, A1, and A2D6 cell lines (Fig. 3). Even the lowest RAD dose
`(1 nM) increased the percentage of apoptotic cells in all cell
`lines, with some differences resembling the cell growth and
`
`RAD-mediated increase in apoptotic rate of PTLD-like B cells. Three
`Fig. 3.
`EBV⫹ B cell lines were cultured for 24 h with 0 –10 nM of RAD, labeled
`with propidium iodine and anti-annexin V antibody, and analyzed by flow
`cytometry.
`
`Majewski et al.
`
`PNAS 兩 April 11, 2000 兩 vol. 97 兩 no. 8 兩 4287
`
`West-Ward Exhibit 1075
`Majewski 2000 - Page 003
`
`

`

`sions (22–24). Such implantation resulted in a fast establishment
`of tumors with very similar growth characteristics in virtually all
`recipient mice (22–24). For these reasons, we selected the s.c.
`model for further studies.
`
`Treatment of Established EBVⴙ B Cell Tumors. To determine the in
`vivo effect of RAD on the PTLD-like cells, 15A, 20A, and A1 cell
`line tumors were implanted into 7–16 SCID mice per cell line.
`Treatment was initiated once the tumors reached 5 mm in
`diameter, which corresponds to a volume of 50 mm3. RAD was
`administered by daily gavage at 5 mg兾kg, which was a well
`tolerated (see above) and effective immunosuppressive (6, 7)
`dose. As shown in Fig. 5, RAD had a profound inhibitory effect
`on growth of the xenotransplanted PTLD-like tumors, with
`visible differences in the degree of response among the three
`tumors. In mice implanted with 15A tumors, there was a marked
`drug-induced delay of the tumor growth but no absolute tumor
`growth inhibition or regression. On day 21, the median tumor
`volume was approximately 240 mm3 in the treated mice com-
`pared with 2,720 mm3 in the control untreated mice. As late as
`day 38, 15A tumors in the treated mice reached only around
`1,000 mm3.
`The effect of RAD on the 20A tumors was even more striking.
`Whereas on day 19, the control 20A tumors had a median volume
`of 1,440 mm3, the average treated tumor measured only 50 mm3,
`which was equal to the initial tumor size. Furthermore, a
`significant regression in the tumor volume was seen in 6 of 10
`treated mice. Accordingly, the median tumor volume on day 53
`was only ⬍5 mm3 for this subset.
`RAD proved to be the most effective against the A1 cell line.
`On day 21, the median volume of the treated tumors was 50 mm3,
`with none of the eight tumors showing any evidence of growth.
`Mean volume of the untreated tumors on that day was approx-
`imately 1,800 mm3. Further treatment resulted in a steady
`regression in all eight mice. On day 53, the mean tumor volume
`decreased to ⬍5 mm3, and no lymphoma could be detected in
`four mice, indicating total tumor eradication. The other four
`mice had microscopically confirmed residual lymphoma. It is
`noteworthy that the differences in the in vivo effectiveness of
`RAD against the 15A, 20A, and A1 tumors paralleled the
`differences seen in the proliferation (Fig. 1) and the other in vitro
`assays (Figs. 2 and 3). This observation suggests that cell analysis
`in vitro may be predictive of the response to RAD in vivo.
`
`Fig. 4. Morphology and phenotype of the 20A cell line, xenotransplanted
`into a SCID mouse. (A) Hematoxylin-eosin stain showing large cell lymphoma
`with high mitotic rate. (B) Immunoperoxidase stain with antibody against
`human CD20 (B cell antigen) showing cell-membrane staining in all lymphoma
`cells. (C) Immunoperoxidase stain with an antibody against Ki-67 (cell prolif-
`eration-related antigen) showing nuclear staining in 50 – 80%. (D) In situ
`hybridization for EBV-encoded RNA1 (EBER1) showing nuclear positivity in all
`lymphoma cells.
`
`the tumors represent human EBV⫹ B cell lymphoma corre-
`sponding to the monomorphic type of PTLD.
`The s.c. lymphoma model had some advantages over the
`peritoneal model. First, even small lymphomas could be iden-
`tified easily. Second, the lymphomas of up to 1.5 cm in diameter
`remained localized, which permitted us to determine the total
`tumor volume with great accuracy (22). Finally, the s.c. tumor
`could be transferred simultaneously into several mice by im-
`planting tumor tissue fragments rather than single-cell suspen-
`
`RAD-mediated inhibition of in vivo growth of PTLD-like B cells; treatment of established tumors. Fragments of tumors derived from EBV⫹ B cell lines
`Fig. 5.
`15A, 20A, and A2D6 were implanted into recipient SCID mice. Treatment with 5 mg兾kg兾day of the drug was started when the tumors reached 5 mm in diameter
`(50 mm3 in volume). The numbers of mice per group are shown in parentheses; lines depicting results from some of the mice are superimposed, particularly in
`the RAD-treated mice.
`
`4288 兩 www.pnas.org
`
`Majewski et al.
`
`West-Ward Exhibit 1075
`Majewski 2000 - Page 004
`
`

`

`MEDICALSCIENCES
`
`RAD-mediated inhibition of in vivo growth of a malignant PTLD-like B cells; prevention of tumor growth. Fragments of tumors derived from EBV⫹ B
`Fig. 6.
`cell lines 15A, 20A, and A2D6 were implanted into recipient SCID mice. The treatment with 5 mg兾kg兾day of the drug was started 3 days before the tumor
`implantation. The numbers of mice per group are shown in parentheses; lines depicting results from some of the mice are superimposed.
`
`Prevention of Tumor Establishment. Because RAD as an immuno-
`suppressive agent is administered chronically to transplant pa-
`tients, its main therapeutic impact on PTLDs may be to inhibit
`their development rather than to treat clinically symptomatic
`cases. To test whether RAD can prevent establishment of the
`PTLD-like lymphomas, daily treatment with 5 mg兾kg of the drug
`was initiated 3 days before tumor implantation. As shown in Fig.
`6, RAD proved to be extremely effective in this model. For 15A
`tumors, which are the least sensitive to RAD (Fig. 5), treatment
`profoundly delayed tumor growth but was unable to prevent
`tumor establishment. On day 25, the treated tumors measured on
`average 38 mm3, and the untreated tumors measured approxi-
`mately 1,940 mm3. On day 45, the median volume of the treated
`tumors was 480 mm3.
`Of 11 20A tumors, 6 were nondetectable on day 25; the median
`volume of the remaining 5 was 20 mm3. The average untreated
`tumor measured 210 mm3 on that day. On day 45, none of the
`treated tumors exceeded 150 mm3, and the average untreated
`tumor measured approximately 4,200 mm3. Five treated mice
`showed no signs of tumor; microscopic evaluation of the im-
`plantation site revealed no malignant cells.
`Finally, in mice with A1 cells, no tumor could be detected as
`late as on day 29 in any of the eight treated mice; the average
`tumor in the untreated mice measured 580 mm3. On day 53, only
`three mice showed small (⬍5 mm3), histologically confirmed
`lymphomas. The remaining five mice showed no evidence of
`tumor.
`
`Discussion
`New treatment modalities are needed to inhibit development
`and improve cure rate of PTLDs. An ideal anti-PTLD drug
`would play a double role of preventing graft rejection and, at the
`same time, inhibiting development and growth of PTLD. Should
`PTLD develop despite the treatment, an increase, rather than
`decrease in the drug dose as currently done with the standard
`immunosuppressive drugs, might be effective. This approach of
`increasing the drug dose would have an additional advantage of
`not jeopardizing survival of the graft. Herein, we present in vitro
`and in vivo data that indicate that RAD, a macrolide immuno-
`suppressant, may play such a double role as an antirejection and
`anti-PTLD drug. RAD profoundly suppressed in vitro prolifer-
`ation, arrested cell-cycle progression at the early G0兾G1 stage,
`and increased apoptotic rate in PTLD-like human EBV⫹ B cells.
`In SCID mice xenotransplanted with three different PTLD-like
`B cell tumors, RAD markedly inhibited growth of these tumors,
`
`particularly if given before the tumor implantation. With the A1
`tumors, total eradication of already established tumors was
`achieved in four of eight mice. In the other two lines tested (20A
`and A1), RAD completely prevented establishment of tumors in
`approximately 50% of mice (5兾11 and 5兾8, respectively). The
`anti-EBV⫹ B cell tumor effect of RAD contrasts with the
`standard immunosuppressive agents cyclosporin and tacrolimus,
`which have been suggested recently to enhance outgrowth of
`EBV⫹ B cells, not only indirectly by suppressing an immune
`response against such cells, but also directly by protecting them
`from the effects of proapoptotic signals (30).
`We consider our results particularly encouraging, because the
`15A and 20A lines were derived from lymphoma cells (15) and
`as such seem to correspond to advanced, clinically aggressive
`PTLD. It is plausible, therefore, that other, less malignant forms
`of PTLD, which comprise the majority of clinical cases, should
`be even more sensitive to RAD. The fact that the A1 cell line
`obtained from a normal individual was more sensitive than 15A
`and 20A to RAD supports this assumption. Our data also suggest
`that monotherapy with RAD may not be sufficient to eradicate
`some established, overtly malignant PTLD tumors, and combi-
`nation therapy with conventional chemotherapeutic drugs might
`need to be considered in such clinical settings (28, 31). Alter-
`natively, adoptive transfer of in vitro generated EBV-specific
`cytotoxic T cells, recently shown to be effective in patients at risk
`or with an overt PTLD (32, 33), may be used together with RAD.
`Although inhibition of cell growth as measured by thymidine
`incorporation (Fig. 1) and cell-cycle progression (Fig. 2) seems
`to be the main mode of RAD action on the EBV⫹ B cells, our
`finding that RAD also increases apoptotic rate in such cells (Fig.
`3) suggests that programmed cell death might also play a role in
`the antitumor activity of the drug. This proapoptotic effect may
`be particularly important in treatment of already established
`PTLD tumors where inhibition of tumor growth alone may not
`be sufficient to achieve complete tumor regression. Also, our
`finding that only a prolonged exposure to RAD led to marked
`regression or elimination of many tumors (Fig. 5) suggests that
`a similar extended treatment may be required to eradicate
`established PTLDs.
`Because the PTLD-like B cell lines sensitive to RAD were all
`EBV⫹, the potential role, if any, of the virus in mediating this
`sensitivity needs to be explored. EBV encodes or induces in the
`target cells several proteins capable of activating cytokine sig-
`naling pathways (34). The membrane-anchored viral LMP-1,
`which is the best characterized, uses the TRAF signaling pathway
`
`Majewski et al.
`
`PNAS 兩 April 11, 2000 兩 vol. 97 兩 no. 8 兩 4289
`
`West-Ward Exhibit 1075
`Majewski 2000 - Page 005
`
`

`

`of the tumor necrosis factor receptor family (35–37). It is
`interesting in this context, that signaling via CD40, which belongs
`to the family and shares several features with LMP-1 including
`signaling via TRAF3 protein (38, 39), has been shown recently
`to be inhibited by rapamycin (12). However, our immunohisto-
`chemical analysis indicates that LMP-1 is expressed only by a
`subset of EBV⫹ B cells, which suggests that LMP-1 may not be
`critical for growth of the PTLD-like cells and their sensitivity to
`RAD. EBV also encodes two other, related membrane proteins,
`LMP-2A and LMP-2B, but they also do not seem to be essential
`for in vivo growth of EBV⫹ B cells (40). Alternatively, RAD
`might inhibit signaling mediated by cytokines induced by EBV in
`
`1. Morrison, V., Dunn, D. L., Manivel, C., Gajl-Peczalska, K. J. & Peterson, B. A.
`(1994) Am. J. Med. 97, 14–24.
`2. Warnke, R. A., Weiss, L. M., Chan, J. K. C., Cleary, M. L. & Dorfman, R. F.
`(1995) Armed Forces Inst. Pathol. Fascicle 14, 531–535.
`3. Curtis, R. E., Travis, L. B., Rowlings, P. A., Socie´, G., Douglas, W., Kingma,
`D. W., Banks, P. M., Jaffe, E. S., Sale, G. E., Horowitz, M. M., et al. (1994)
`Blood 94, 2208–2216.
`4. Harris, N. L., Ferry, J. A. & Swierdlow, S. H. (1997) Semin. Diagn. Pathol. 14, 8–14.
`5. Gummert, J. F., Ikonen, T. & Morris, R. E. (1999) J. Am. Soc. Nephrol. 10,
`1366–1380.
`6. Schuurman, H. J., Cottens, S., Fuchs, S., Joergensen, J., Meerloo, T., Sedrani,
`R., Tanner, M., Zenke, G. & Schuler, W. (1997) Transplantation 64, 32–35.
`7. Schuler, W., Sedrani, R., Cottens, S., Haberlin, B., Schulz, M., Schuurman,
`H. J., Zenke, G., Zerwes, H. G. & Schreier, M. H. (1997) Transplantation 64,
`36–42.
`8. Sedrani, R., Cottens, S., Kallen, J. & Schuler, W. (1998) Transplant. Proc. 30,
`2192–2194.
`9. Schuurman, H. J., Schuler, W., Ringers, J. & Jonker, M. (1998) Transplant.
`Proc. 30, 2198–2199.
`10. Hausen, B., Boeke, K., Berry, G. J., Segarra, I. T., Christians, U. & Morri, R. E.
`(1999) J. Heart Lung Transplant. 18, 150–159.
`11. Seghal, S. N. (1998) Clin. Biochem. 31, 335–340.
`12. Sakata, A., Kuwahara, K., Ohmura, T., Inui, S. & Sakaguchi, N. (1999)
`Immunol. Lett. 68, 301–319.
`13. Terada, N., Lucas, J. J., Szepesi, A., Franklin, R. A., Domenico, J. & Gelfand,
`E. W. (1993) J. Cell. Physiol. 154, 7–15.
`14. Flanagan, W. M. & Crabtree, G. R. (1993) Ann. N.Y. Acad. Sci. 696, 31–37.
`15. Silberstein, L. E., Jefferies, L. C., Goldman, J., Friedman, D., Moore, J. S.,
`Nowell, P. C., Roelcke, D., Pruzanski, W., Roudier, J. & Silverman, G. J. (1991)
`Blood 78, 2372–2386.
`16. Zhang, Q., Nowak, I., Vonderheid, E. C., Rook, A. H., Kadin, M. E., Nowell,
`P. C., Shaw, L. M. & Wasik, M. A. (1996) Proc. Natl. Acad. Sci. USA 93,
`9148–9153.
`17. Cesarman, E., Moore, P. S., Rao, P. H., Inghirami, G., Knowles, D. M. &
`Chang, Y. (1995) Blood 86, 2708–2714.
`18. Zhang, Q., Lee, B., Korecka, M., Li, G., Weyland, C., Eck, S., Gessain, A.,
`Arima, N., Shaw, L., Luger, S., et al. (1999) Leukemia Res. 23, 373–384.
`19. Kawamata, S., Sakaida, H., Hori, T., Maeda, M. & Uchiyama, T. (1998) Blood
`91, 561–569.
`20. Pepper, C., Thomas, A., Tucker, H., Hoy, T. & Bentley, P. (1998) Leuk. Res.
`22, 439–444.
`21. Douglas, R. S., Pletcher, C. H., Jr., Nowell, P. C. & Moore, J. S. (1998)
`Cytometry 32, 57–65.
`
`the target cells, such as tumor necrosis factor-␣ and tumor
`necrosis factor-␤ (34). However, an autocrine role for these or
`any other cytokine(s) in the EBV⫹ B cells remains to be
`established.
`In summary, our data show that RAD has a potent inhibitory
`effect on EBV⫹ B lymphocytes in vitro and in vivo. Therefore,
`it may be effective in treatment and prevention of PTLDs in
`transplant patients.
`
`The study was supported in part by a grant from Novartis Pharma and
`by National Cancer Institute Grant CA76627.
`
`22. Wasik, M. A., Sioutos, N., Tuttle, M., Butmarc, J. R., Kaplan, W. & Kadin,
`M. E. (1994) Am. J. Pathol. 144, 1089–1097.
`23. Pasqualucci, L., Wasik, M. A., Teicher, B., Flenghi, L., Bolognesi, A., Stirpe,
`F., Polito, L., Falini, B. &