Antitumor Effects of Bortezomib (PS-341) on Primary Effusion Lymphomas

JAn1, Y Sun1, M Fisher1 and MB Rettig1,2

1VA Greater Los Angeles Healthcare System, Los Angeles, CA 90073, USA; and 2Department of Medicine, David Geffen School of Medicine at UCLA, Los Angeles, CA 90095, USA

Primary effusion lymphomas (PELs) are a rare type of non- treatment of refractory multiple myeloma, where it has an Hodgkin’s lymphoma that are resistant to cytotoxic chemother- acceptable toxicity profile.4–6 By inhibiting the proteasome, apy. PELs manifest constitutive activation of nuclear factor kappa B (NF-jB), and inhibition of NF-jB induces apoptosis of which degrades proteins that are marked by ubiquitination, PELs and sensitizes to tumor necrosis factor-related apoptosis- bortezomib modulates the cellular concentrations of hundreds inducing ligand (TRAIL)-induced death. Bortezomib (PS-341), a of proteins that are involved in a wide variety of functions, peptidyl boronic acid inhibitor of the proteasome, is a potent including cell cycle progression, signal transduction, and agent against a wide range of hematologic malignancies and apoptosis.7,8 For example, the proteasome regulates the activity has been shown to inhibit NF-jB. Thus, we examined the of the NF-kB pathway, in that the degradation of I kappa B (IkB), cytotoxic effects of bortezomib alone and in combination with the NF-kB inhibitory protein, is dependent upon the ubiquitin– various drugs. Bortezomib potently inhibited NF-jB in PEL cells 9 in a dose-dependent manner. In addition, bortezomib inhibited proteasome pathway. Given the important role or heightened 2 growth and induced apoptosis of PEL cell lines (IC50 values of NF-kB activity in promoting the survival of PEL cells, we 3.4–5.0 nM). Results of drug interactions between bortezomib postulated that bortezomib has potent cytotoxic effects on PELs. and chemotherapy (doxorubicin and Taxol) were schedule- Thus, we studied the cytotoxic effects of bortezomib alone and dependent: synergistic interactions were generally observed in combination with TRAIL as well as chemotherapy and when PEL cells were pretreated with bortezomib prior to chemotherapy, whereas additive or even antagonistic interac- dexamethasone on PELs. tions occurred with chemotherapy pretreatment or simultaneous treatment with bortezomib and chemotherapy. Most schedules of bortezomib and dexamethasone were synergistic, Materials and methods although pretreatment with dexamethasone resulted in additive interactions. Effects of combinations of bortezomib and TRAIL Cells and reagents were generally additive. Thus, bortezomib represents a promising potential therapy for the treatment of PEL. Leukemia (2004) 18, 1699–1704. doi:10.1038/sj.leu.2403460 The KS-1, BCBL-1, and BCP-1 PEL cell lines, which are infected Published online 2 September 2004 with KSHV but not other herpesviruses, were maintained in Keywords: proteasome inhibition; PS-341; primary effusion RPMI plus 20% FBS and antibiotics. Bortezomib (Millennium lymphoma; chemotherapy; TRAIL; NF-kB Pharmaceuticals, Cambridge, MA, USA) and recombinant, human TRAIL (Calbiochem, San Diego CA, USA) were dissolved in dimethyl sulfoxide (DMSO) and phosphate-buffered saline (PBS), respectively. Bay 11-7082 (Calbiochem, San Diego, CA, USA), a chemical inhibitor of the NF-kB pathway, Introduction was dissolved in DMSO. Doxorubicin and paclitaxel (Taxol) were purchased from Sigma (St Louis, MO, USA) and dissolved PELs, which occur in both human immunodeficiency virus- in water and methanol, respectively. The corticosteroid, dex- positive and -negative patients, represent a rare subtype of non- amethasone (Sigma) was dissolved in ethanol. For all experiments, Hodgkin’s lymphoma that is uniformly infected with the the final concentration of solvents was maintained at 0.1%. Kaposi’s sarcoma-associated herpesvirus, also known as human A kB responsive plasmid (p4x-kB-luc), in which four copies of herpesvirus 8. Unlike most non-Hodgkin’s lymphomas, PELs are the kB-response element (kB-RE) drives expression of firefly relatively resistant to standard cytotoxic chemotherapy, and luciferase, was purchased from Invitrogen (Carlsbad, CA, USA). virtually all patients succumb to the disease with a median The pRL-SV40 plasmid, in which Renilla luciferase is constitu- survival of approximately 60 days.1 One of the biochemical tively expressed under the regulation of the SV40 promoter/ hallmarks of PELs is constitutive NF-kB activation, which results enhancer, was purchased from Promega Corporation (Madison, in heightened expression of antiapoptotic genes and is critical WI, USA) and was used for normalization of firefly luciferase for PEL cell survival.2 In fact, drugs that inhibit NF-kB activation activity. have exhibited considerable activity against PELs both in vitro and in vivo.1–3 In addition, inhibition of NF-kB sensitizes PEL cells to tumor necrosis factor-related apoptosis-inducing ligand Transient transfections and reporter gene assays (TRAIL)-induced apoptosis.3 Recently, the proteasome inhibitor bortezomib (formerly PS- Cells were plated in 24-well plates the day prior to transfection. 341) has demonstrated potent cytotoxicity across a wide variety The plasmids were transfected with Lipofectamine Plus (Life of hematologic malignancies and solid tumors and received Technologies, Gaithersburg, MD, USA), according to the approval by the Food and Drug Administration (FDA) for the manufacturer’s instructions. Protein was extracted 48 h after transfection, and firefly and Renilla luciferase were measured on Correspondence: Dr MB Rettig, VAGLAHS-West LA, 11301 Wilshire a TD20/20 tube luminometer (Turner Designs, Sunnyvale, CA, Blvd, Building 304, Room E1-113, Los Angeles, CA 90073, USA; Fax: þ 1 310 268 4508; E-mail: [email protected] USA) using a Dual Luciferase Assay kit (Promega Corporation), Received 19 February 2004; accepted 10 June 2004; Published online according to the manufacturer’s instructions. Firefly luciferase 2 September 2004 activity was normalized to Renilla luciferase expression. Antitumor effects of bortezomib on PELs JAnet al 1700 Electrophoretic mobility shift assay (EMSA) and Western a BCBL-1 blotting 1.2 10 1 The EMSAs and Western blots were performed, as described. 0.8 0.6 Assessment of cytotoxicity RLU 0.4 0.2 0 Cells were seeded in 96-well plates the day prior to chemical DMSO 0.625 1.25 2.5 5 10 treatment. Various drug combinations were added to cells, and Bortezomib (nM) 48 h later, cell viability was assessed by the MTT (3,[4,5- dimethylthiazol-2-yl-] diphenyltetrazolium bromide) assay. In TNF b + total, 25 ml of MTT (5 mg/ml) was added to each well for 3 h at TNF Bortezomib (nM) cold comp. 371C. Subsequently, 120 ml of 10% sodium dodecyl sulfate/ − + 2.5 5 7.5 10 wt mut 0.01 N HCl was added overnight at 371C. Absorbance was measured at 570 nm on a microplate reader. All experiments NF-κB were performed in triplicate.

Assessment of apoptosis

Cells were plated in 10 cm2 dishes. The next day, cells were treated with various concentrations of bortezomib or vehicle c control for 48 h. Cells were harvested and then stained with an Bortezomib (nM) 0 2.5 5 7.5 10 annexin V-FITC kit (BD Biosciences, Palo Alto, CA, USA), κ α according to the manufacturer’s instructions. Cells were I B analyzed on a Becton Dickenson FACS Caliber flow cytometer Actin with CellQuest software (BD Biosciences). Figure 1 Bortezomib inhibits constitutive NF-kB activation in Median effect/combination index (CI) isobologram PELs. (a) Bortezomib blocks NF-kB-driven reporter gene expression. method for multiple drug effect analysis The pNF-kB-luc (1 mg) and pRL-SV40 (1 ng) reporter plasmids were transfected into 1 Â 105 BCBL-1 cells in 24-well plates. At 5 h after transfection, bortezomib was added at the indicated concentrations for The effect of drug combinations on cytotoxicity was performed an additional 48 h. Firefly luciferase is reported and was normalized to by the median effect/combination index (CI) method using Renilla luciferase activity. Results represent median of three experi- Calcusyn software, version 1.1.1 (Biosoft, Ferguson, MO, USA) ments and are reported as relative luminescence units (RLU)7s.d. 11 as were the IC50 values, as previously described. CI and IC50 Similar results were obtained for the BCP-1 and KS-1 cell lines (data values were calculated from median results of the MTT not shown). (b) Bortezomib inhibits NF-kB nuclear localization. BCBL- 1 cells (1 Â 107 cells) were plated in 10 cm2 dishes and treated with assays. Fixed drug ratios were maintained for CI analysis. CI bortezomib for 60 min or TNFa (20 ng/ml) for 20 min. Nuclear protein values significantly greater than 1 indicate drug antagonism, was extracted for EMSA. Cold competition with excess wild-type CI values significantly less than 1 are indicative of synergy, and probe abrogated the gel shifted bands, whereas cold mutant probe had CI values not significantly different than 1 indicate an additive no effect (rightmost two lanes). (c) Bortezomib blocks the degradation 5 drug effect. Linear regression correlation coefficients of the of IkBa. BCBL-1 cells (5 Â 10 cells/well) were plated in six-well plates median effects plots were required to be 40.90 to demonstrate and treated with bortezomib for 90 min. In total, 10 mg of cellular protein was subjected to immunoblotting for IkBa (top panel). Bottom that the effects of the drugs follow the law of mass action, which panel: Immunoblotting for actin confirms equivalent protein loading. is required for a median effect analysis.

Results tively.2 As further evidence of NF-kB inhibitory effects of bortezomib, we demonstrated that the drug blocked the Bortezomib inhibits constitutive NF-kB activation degradation of IkBa (Figure 1c), which is consistent with its in PELs function as a proteasome inhibitor. The degree of NF-kB inhibition is less pronounced in the Inhibition of NF-kB activity by bortezomib is considered to play EMSA compared to the reporter assay because of the shorter a key role in its pro-apoptotic effects.5,12 Given the importance duration of exposure to bortezomib (60 min) employed in the of NF-kB activation in the survival of B-cell malignancies EMSA to avoid the nonspeciﬁc effects on intracellular protein including PELs,1,2,13,14 we postulated that bortezomib would levels and cellular health observed with proteasome inhibition. have potent apoptotic effects on PEL cells. As a prelude to study In contrast, the reporter assay (Figure 1a) allows for normal- the cytotoxic effects of bortezomib, we tested the ability of ization for factors other than NF-kB transcriptional activity that bortezomib to block constitutive NF-kB activation in PELs. may be affected by proteasome inhibition and, therefore, can be Bortezomib inhibited an NF-kB driven reporter in the BCBL-1 performed over a more prolonged duration. PEL cell lines in a dose-dependent manner (Figure 1a). The IC50 for NF-kB inhibition was 2.8 nM. Similar results were obtained for the KS-1 and BCP-1 cell lines (data not shown). In addition, Bortezomib potently inhibits growth and induces bortezomib blocked the nuclear localization of NF-kB tran- apoptosis of PELs scription factors in BCBL-1 cells as demonstrated by EMSA (Figure 1b). Previous studies have shown that the top and bottom The cytotoxicity of bortezomib was measured on the BCP-1, NF-kB bands represent p65–p50 and p50–p50 dimers, respec- BCBL-1, and KS-1 cell lines by the MTT assay. Bortezomib

Leukemia Antitumor effects of bortezomib on PELs JAnet al 1701 exhibited potent, dose-dependent cytotoxic effects on all cell prior to exposure to either doxorubicin or Taxol generally lines after 48 h of drug exposure (Figure 2a). The IC50 values resulted in synergistic drug interactions with CI values less than ranged from 3.4 to 5.0 nM. Annexin V staining of bortezomib- 1 across most concentrations tested (Figure 3). In contrast, treated cells demonstrated that growth inhibition was predomi- pretreatment with chemotherapy for 24 h prior to bortezomib nantly due to apoptosis (Figure 2b). In addition, the concentra- treatment or simultaneous exposure to both chemotherapy and tions of bortezomib that inhibited NF-kB driven reporter gene bortezomib led to only additive or even antagonistic effects expression correlated with those required to induce cytotoxicity (Figure 3). When we tested different ratios of concentrations of and apoptosis (compare Figure 1a to Figure 2a, b). For example, bortezomib to chemotherapy other than the ratio used in in BCBL-1 cells, the NF-kB inhibitory IC50 was 2.8 nM and the Figure 3 (bortezomib:chemotherapy ¼ 1:10 000), the overall IC50 for growth was 4.4 nM. In an attempt to identify results of median effect/CI analysis remained the same (data antiapoptotic proteins that may be affected by bortezomib not shown). Thus, the results of drug interactions between treatment, we performed Western blots for NF-kB-regulated bortezomib and chemotherapy are schedule-dependent. targets, cFLIP, XIAP, and BCL-XL. Interestingly, bortezomib did We also tested the growth inhibitory effects of bortezomib in not affect the protein levels of any of these proteins (Supple- combination with the corticosteroid, dexamethasone. Similar to mentary Figure 1). chemotherapy, dexamethasone had minimal effects as a single agent (data not shown). The combination of dexamethasone and bortezomib, when used simultaneously or with bortezomib pretreatment was synergistic for all three cell lines tested Drug interactions between bortezomib and (Figure 4). In contrast, dexamethasone pretreatment followed chemotherapy/dexamethasone by bortezomib was generally additive, although the interaction was synergistic for BCBL-1 cells (Figure 4). The drug interactions between bortezomib and chemotherapy have been studied in other tumor models, and results have been variable. Whereas bortezomib and chemotherapy are synergistic in colon cancer,15 they are additive in prostate cancer.16 Drug interactions between bortezomib and TRAIL Using the BCBL-1, BCP-1 and KS-1 PEL cell lines, we studied the drug interactions between bortezomib and doxorubicin or Taxol Recent studies have demonstrated that interferon alpha (IFN-a) (paclitaxel), two commonly used chemotherapeutic agents in plus azidothymidine (AZT) results in marked apoptosis of the treatment of lymphomas. As single agents, doxorubicin and PELs,3,4 which is mediated by IFN-a induction of high levels Taxol were minimally active in all three cell lines tested with of TRAIL expression.3,17,18 The pro-apoptotic effects of TRAIL, B IC50 values that were in the 100–500 mM range (Supplemen- which ligates death receptors and thereby activates the extrinsic tary Table 1). When we assessed the effects of drug combina- pathway of apoptosis, can be abrogated by its simultaneous tions of bortezomib and either doxorubicin or Taxol by median activation of the antiapoptotic NF-kB pathway.19,20 Thus, it was effect/CI analysis, we found variable results depending upon further demonstrated that AZT sensitized to TRAIL by blocking drug scheduling. Thus, pretreatment with bortezomib for 24 h NF-kB activation.3,4,17,18 As we documented that bortezomib

IC50 Values for Bortezomib a 1.2 BCBL-1 KS-1 Bortezomib 1 (nM) BCP-1 0.8 BCBL-1 4.4 0.6 BCP-1 3.4 RCV 0.4 KS-1 5 0.2 0 0.625 1.25 2.5 5 10 Bortezomib (nM)

Untreated 1.25 nM 2.5 nM 5 nM 10 nM 0.18% 32.1% 40.7% 84.7% 88.9% PI PI PI PI PI

Annexin V Annexin V Annexin V Annexin V Annexin V

Figure 2 Effects of bortezomib on PEL growth and apoptosis. (a) Cytotoxicity (MTT) assay. BCP-1, KS-1 and BCBL-1 cells were plated in a 96- well format (2 Â 104 cells/well) and treated with the indicated concentrations of bortezomib for 48 h. The MTT assay was performed in triplicate, 7 and relative cell viability (RCV) values s.d. were normalized to that of vehicle control. At right are the IC50 values for bortezomib. (b) Bortezomib- induced apoptosis of PELs. BCBL-1 cells (1 Â 107 cells) were plated in a 10 cm2 dish, treated with the indicated concentrations of bortezomib for 48 h, and assayed for apoptosis using an annexin V-FITC kit by ﬂow cytometry. Percentages indicate apoptotic cells.

Leukemia Antitumor effects of bortezomib on PELs JAnet al 1702 a BCBL1 - DOX + Bortezomib BCBL1 - TAX + Bortezomib 1 2.5 1 2.5 0.8 2 0.8 2

0.6 1.5 0.6 1.5 CI CI FA 0.4 1 FA 0.4 1

0.2 0.5 0.2 0.5

0 0 0 0 0.1 0.63 1.25 2.5 5 0.1 0.63 1.25 2.5 5 Bortezomib (nM) Bortezomib (nM)

b BCP-1 - DOX + Bortezomib BCP - TAX + Bortezomib 1 4 1 1.5

0.8 3 0.8 1 0.6 0.6 2 CI CI FA FA 0.4 0.4 0.5 1 0.2 0.2 0 0 0 0 0.1 0.63 1.25 2.5 5 0.1 0.63 1.25 2.5 5 Bortezomib (nM) Bortezomib (nM)

c KS1 - DOX + Bortezomib KS1 - TAX + Bortezomib 1 2 1 4

0.8 1.5 0.8 3 0.6 0.6 1 2 CI CI FA FA 0.4 0.4 0.5 1 0.2 0.2 0 0 0 0 0.1 0.63 1.25 2.5 5 0.1 0.63 1.25 2.5 5 Bortezomib (nM) Bortezomib (nM)

FA – Simultaneous CI – Simultaneous FA – Chemo pre-treatment CI – Chemo pre-treatment FA – Bortezomib pre-treatment CI – Bortezomib pre-treatment

Figure 3 Median effect/CI analysis of bortezomib and chemotherapy combinations. PEL cells were plated in 96-well format (2 Â 104 cells/well) and then treated with combinations bortezomib and chemotherapy. Three different schedules of drug treatment were used: (1) bortezomib pretreatment for 24 h prior to addition of chemotherapy for an additional 24 h, (2) pretreatment with chemotherapy for 24 h prior to the addition of bortezomib for another 24 h, and (3) simultaneous treatment of bortezomib and chemotherapy for 48 h. Chemotherapy ¼ doxorubicin (DOX) and Taxol (TAX). The ratio of concentrations of bortezomib (nM):chemotherapy (nM) was held constant (1:10 000) for the purpose of median effect/ combination index (CI) method analysis of drug interactions (see Material and methods for details). CI values 41 indicate antagonistic effect, CI values o1 indicate synergy, and CI valuesD1 indicate an additive effect. Results were obtained from MTT data, which was performed in triplicate for each concentration. Vertical axis on the left represents fraction of cells affected (FA), and vertical axis on the right represents the CI values. (a) BCBL-1, (b) BCP-1, (c) KS-1.

blocks NF-kB activation in PELs and TRAIL represents a agent, as previously reported (Supplementary Table 2).2 Thus, potential cancer therapy that has been extensively studied in Bay 11-7082 and TRAIL combinations generally resulted in preclinical models,21–24 we studied the drug interactions additive effects, which were schedule-independent. between bortezomib and TRAIL. Consistent with prior work,17,18 TRAIL had minimal cytotoxic effects on its own with approximately 10% reduction in PEL cell Discussion viability at 48 h at a concentration up to 200 ng/ml (data not shown). Unlike the synergistic interactions between IFN-a and We have shown that bortezomib potently inhibits the growth TRAIL, when we treated PELs with bortezomib and TRAIL in and induces the apoptosis of PEL cells in vitro at pharmacolo- combination, median effect/CI analysis demonstrated additive gically achievable doses. The IC50 values for bortezomib were in interactions across all concentrations and all schedules of both the low nanomolar range, which closely correspond to the IC50 bortezomib and TRAIL (Supplementary Figure 2). When we values for multiple myeloma, a disease for which bortezomib tested the drug interactions between Bay 11-7082, a speciﬁc has shown signiﬁcant clinical activity and has received FDA inhibitor of the NF-kB pathway, and TRAIL, we found similar approval.5,6 In addition, bortezomib potently inhibits constitu- results as those obtained between bortezomib and TRAIL tive NF-kB activity. As selective inhibition of NF-kB can induce 2 (Supplementary Figure 2). Bay 11-7082 was active as a single apoptosis of PELs and our IC50 values for growth and NF-kB

Leukemia Antitumor effects of bortezomib on PELs JAnet al 1703 BCBL-1 BCP-1 KS-1

1 1.2 1 3.5 0.9 2 0.9 0.9 1.8 1 3 0.8 0.8 0.8 0.7 1.6 0.7 0.8 0.7 2.5 1.4 0.6 0.6 0.6 1.2 2 0.5 CI 0.6 CI

0.5 0.5 1 CI FA FA 0.4 0.4 1.5 FA 0.4 0.8 0.3 0.4 0.3 1 0.3 0.6 0.2 0.2 0.2 0.2 0.5 0.4 0.1 0.1 0.1 0.2 0 0 0 0 0 0 0.625 1.25 2.5 5 0.625 1.25 2.5 5 0.625 1.25 2.5 5 Bortezomib (nM) Bortezomib (nM) Bortezomib (nM)

FA – DEX pre-treatment CI – DEX pre-treatment FA – Simultaneous CI – Simultaneous FA – Bortezomib pre-treatment CI – Bortezomib pre-treatment

Figure 4 Median effect/CI analysis of dexamethasone (DEX) and bortezomib. PEL cells were plated and treated, and the results were analyzed in a similar manner as described in Figure 3. PEL cells were either pretreated with bortezomib for 24 h prior to an additional 24 h of DEX exposure or vice versa or cells were simultaneously treated with both drugs for 48 h. The ratio of bortezomib (nM):DEX (mM) was held at 1:5.

inhibition in PELs closely correlated, we postulate that inhibition These studies involved combinations of bortezomib and TRAIL of NF-kB plays a central role in the mechanism of bortezomib- that induced a wide range of cytotoxicity. Thus, we cannot induced death. When we evaluated the effects of bortezomib on attribute the lack of synergy to the use of concentrations of the levels various antiapoptotic proteins (ie XIAP, cFLIP, BCL- bortezomib that induced such high levels of cytotoxicity that XL), we observed no effect, indicating that bortezomib-induced they obscured a potential synergistic effect with TRAIL. These growth inhibition functions through alternative mechanisms. results contrast the synergistic interaction observed between Standard chemotherapy is ineffective in patients with PEL, and AZT and TRAIL.3 Interestingly, AZT sensitizes to TRAIL-induced median life expectancy is approximately 2 months.1 Our results apoptosis by blocking NF-kB activation.3 Thus, although confirm that the chemotherapeutic agents, doxorubicin and bortezomib and AZT effectively inhibit NF-kB, only AZT Taxol, are ineffective single agents against PEL cell lines in vitro. sensitizes PEL cells to TRAIL-induced death. This difference In addition, we rigorously studied the effects of drug interactions remains to be explained, although it may be attributed to effects between bortezomib and doxorubicin or Taxol. In all three PEL of these drugs unrelated to NF-kB inhibition. In support of this cell lines tested, we observed synergistic interactions when cells concept, Bay 11-7082, a specific inhibitor of NF-kB activation, were pretreated with bortezomib for 24 h prior to chemotherapy induces apoptosis of PELs,2 yet AZT as a monotherapy does exposure. In contrast, pretreatment with chemotherapy or not,3 suggesting that AZT may have heretofore undefined simultaneous exposure to bortezomib and chemotherapy antiapoptotic effects that counterbalance its NF-kB inhibitory resulted in additive or even antagonistic interactions. Thus, the properties. In fact, when we tested drug interactions between results of drug interactions of bortezomib and chemotherapy are Bay 11-7082 and TRAIL, very similar effects as those between schedule-dependent. The mechanism that underlies the sche- bortezomib and TRAIL were observed, suggesting that inhibition dule dependence is unknown, but previous work in colon of NF-kB represents a predominant mechanism of action of cancer has shown that resistance to topoisomerase II inhibitors, bortezomib. such as doxorubicin, may been overcome by proteasome Synergy between bortezomib and TRAIL has been observed in inhibition prior to chemotherapy exposure and that this other tumor models, including murine kidney cancer and sensitization to chemotherapy by proteasome inhibition may murine myeloid leukemia. In these murine models, bortezomib be mediated by modulation of topoisomerase II protein levels.25 sensitized to TRAIL by reducing expression of cFLIP, an Sensitization to Taxol by bortezomib has been demonstrated in antiapoptotic protein that inhibits caspase 8 activation within pancreatic cancer and is mediated by inhibition of NF-kB the proximal portion of the extrinsic pathway of apoptosis. activity, which is enhanced by Taxol exposure.26 In PEL, we Interestingly, we found that bortezomib had no effect on cFLIP hypothesize that pretreatment with bortezomib may be required levels, which may account for the lack of synergy between to downregulate NF-kB activity enough to overcome Taxol bortezomib and TRAIL and indicates that the mechanism of resistance. action and downstream effects of bortezomib are apt to differ Similar to the results with chemotherapy, interactions among various tumor models.27 between dexamethasone and bortezomib were also schedule- dependent, such that simultaneous exposure to both drugs or bortezomib pretreatment led to synergistic interactions. Initial Conclusion treatment with dexamethasone was additive, except for one cell line, BCBL-1, for which this schedule resulted in synergy. Thus, This study is the first to report that bortezomib is an active agent synergistic interactions between dexamethasone and bortezo- against PELs. It is effective as a monotherapy as well as in mib were observed with more schedules as compared to the schedule-dependent combination with chemotherapy and interactions between chemotherapy and bortezomib, for which dexamethasone. As has recently been shown for patients with synergy was only observed with pretreatment with chemother- refractory, relapsed multiple myeloma, bortezomib can be used apy prior to bortezomib exposure. Our drug interactions studies safely in patients and results in an acceptable toxicity profile.6 between bortezomib and TRAIL demonstrated additive effects. Thus, bortezomib represents a promising agent for PELs, a

Leukemia Antitumor effects of bortezomib on PELs JAnet al 1704 disease for which standard cytotoxic chemotherapy has proven 12 Mitchell BS. The proteasome – an emerging therapeutic target in to be ineffective. cancer. N Engl J Med 2003; 348: 2597–2598. 13 Bargou RC, Emmerich F, Krappmann D, Bommert K, Mapara MY, Arnold W et al. Constitutive nuclear factor-kappaB-RelA activation Acknowledgements is required for proliferation and survival of Hodgkin’s disease tumor cells. J Clin Invest 1997; 100: 2961–2969. 14 Davis RE, Brown KD, Siebenlist U, Staudt LM. Constitutive nuclear Supported by grants from the National Institutes of Health factor kappaB activity is required for survival of activated B cell- (5R01CA80004) and the American Cancer Society (RPG-00-305- like diffuse large B cell lymphoma cells. J Exp Med 2001; 194: 01-MBC) to MBR. In addition to supplying bortezomib, Millen- 1861–1874. nium Pharmaceuticals, Incorporated, also provided ﬁnancial 15 Cusack Jr JC, Liu R, Houston M, Abendroth K, Elliott PJ, Adams J support for this study. et al. Enhanced chemosensitivity to CPT-11 with proteasome inhibitor PS-341: implications for systemic nuclear factor-kappaB inhibition. Cancer Res 2001; 61: 3535–3540. Supplementary Information 16 An J, Sun YP, Adams J, Fisher M, Belldegrun A, Rettig MB. Drug interactions between the proteasome inhibitor bortezomib and cytotoxic chemotherapy, tumor necrosis factor (TNF) alpha, and Supplementary Information accompanies the paper on Leuke- TNF-related apoptosis-inducing ligand in prostate cancer. Clin mia website (http://www.nature.com/leu). Cancer Res 2003; 9: 4537–4545. 17 Lee RK, Cai JP, Deyev V, Gill PS, Cabral L, Wood C et al. Azidothymidine and interferon-alpha induce apoptosis in References herpesvirus-associated lymphomas. Cancer Res 1999; 59: 5514–5520. 1 Komanduri KV, Luce JA, McGrath MS, Herndier BG, Ng VL. The 18 Toomey NL, Deyev VV, Wood C, Boise LH, Scott D, Liu LH et al. natural history and molecular heterogeneity of HIV-associated Induction of a TRAIL-mediated suicide program by interferon primary malignant lymphomatous effusions. J Acquir Immune alpha in primary effusion lymphoma. Oncogene 2001; 20: Deﬁc Syndr Hum Retrovirol 1996; 13: 215–226. 7029–7040. 2 Keller SA, Schattner EJ, Cesarman E. Inhibition of NF-kappaB 19 Oya M, Ohtsubo M, Takayanagi A, Tachibana M, Shimizu N, induces apoptosis of KSHV-infected primary effusion lymphoma Murai M. Constitutive activation of nuclear factor-kappaB prevents cells. Blood 2000; 96: 2537–2542. TRAIL-induced apoptosis in renal cancer cells. Oncogene 2001; 3 Ghosh SK, Wood C, Boise LH, Mian AM, Deyev VV, Feuer G et al. 20: 3888–3896. Potentiation of TRAIL-induced apoptosis in primary effusion 20 Ravi R, Bedi GC, Engstrom LW, Zeng Q, Mookerjee B, Gelinas C lymphoma through azidothymidine-mediated inhibition of NF- et al. Regulation of death receptor expression and TRAIL/ kappa B. Blood 2003; 101: 2321–2327. Apo2L-induced apoptosis by NF-kappaB. Nat Cell Biol 2001; 3: 4 Adams J, Palombella VJ, Sausville EA, Johnson J, Destree A, 409–416. Lazarus DD et al. Proteasome inhibitors: a novel class of potent 21 Ashkenazi A, Pai RC, Fong S, Leung S, Lawrence DA, Marsters SA and effective antitumor agents. Cancer Res 1999; 59: 2615–2622. et al. Safety and antitumor activity of recombinant soluble Apo2 5 Hideshima T, Richardson P, Chauhan D, Palombella VJ, Elliott PJ, ligand. J Clin Invest 1999; 104: 155–162. Adams J et al. The proteasome inhibitor PS-341 inhibits growth, 22 Lawrence D, Shahrokh Z, Marsters S, Achilles K, Shih D, Mounho induces apoptosis, and overcomes drug resistance in human B et al. Differential hepatocyte toxicity of recombinant Apo2L/ multiple myeloma cells. Cancer Res 2001; 61: 3071–3076. TRAIL versions. Nat Med 2001; 7: 383–385. 6 Richardson PG, Barlogie B, Berenson J, Singhal S, Jagannath S, 23 Munshi A, Pappas G, Honda T, McDonnell TJ, Younes A, Li Y et al. Irwin D et al. A phase 2 study of bortezomib in relapsed, refractory TRAIL (APO-2L) induces apoptosis in human prostate cancer cells myeloma. N Engl J Med 2003; 348: 2609–2617. that is inhibitable by Bcl-2. Oncogene 2001; 20: 3757–3765. 7 Adams J. Development of the proteasome inhibitor PS-341. 24 Zisman A, Ng CP, Pantuck AJ, Bonavida B, Belldegrun AS. Oncologist 2002; 7: 9–16. Actinomycin D and gemcitabine synergistically sensitize andro- 8 Pajonk F, McBride WH. The proteasome in cancer biology and gen-independent prostate cancer cells to Apo2L/TRAIL-mediated treatment. Radiat Res 2001; 156: 447–459. apoptosis. J Immunother 2001; 24: 459–471. 9 Palombella VJ, Rando OJ, Goldberg AL, Maniatis T. The ubiquitin– 25 Ogiso Y, Tomida A, Lei S, Omura S, Tsuruo T. Proteasome proteasome pathway is required for processing the NF-kappa B1 inhibition circumvents solid tumor resistance to topoisomerase II- precursor protein and the activation of NF-kappa B. Cell 1994; 78: directed drugs. Cancer Res 2000; 60: 2429–2434. 773–785. 26 Dong QG, Sclabas GM, Fujioka S, Schmidt C, Peng B, Wu T et al. 10 An J, Sun YP, Fisher M, Rettig MB. Maximal apoptosis of renal cell The function of multiple IkappaB: NF-kappaB complexes in the carcinoma by the proteasome inhibitor, bortezomib (VELCADETM, resistance of cancer cells to Taxol-induced apoptosis. Oncogene formerly known as PS-341), is NF-kB-dependent. Mol Cancer Ther 2002; 21: 6510–6519. 2004; 3: 727–736. 27 Sayers TJ, Brooks AD, Koh CY, Ma W, Seki N, Raziuddin A et al. 11 Chou TC, Talalay P. Quantitative analysis of dose–effect relation- The proteasome inhibitor PS-341 sensitizes neoplastic cells to ships: the combined effects of multiple drugs or enzyme inhibitors. TRAIL-mediated apoptosis by reducing levels of c-FLIP. Blood Adv Enzyme Regul 1984; 22: 27–55. 2003; 102: 303–310.

Leukemia