Mct-10-0385 178.185

Preclinical Studies on the Mechanism of Action and the
Anti-Lymphoma Activity of the Novel Anti-CD20 Antibody
GA101
Stephane Dalle, Lina Reslan, Timothee Besseyre de Horts, et al.
2011;10:178-185. Published online January 10, 2011.
Mol Cancer Ther Access the most recent version of this article at: This article cites 39 articles, 24 of which you can access for free at: This article has been cited by 3 HighWire-hosted articles. Access the articles at: Reprints and
To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at To request permission to re-use all or part of this article, contact the AACR PublicationsDepartment at Preclinical Development Preclinical Studies on the Mechanism of Action and theAnti-Lymphoma Activity of the Novel Anti-CD20Antibody GA101 Stephane Dalle1,2, Lina Reslan1, Timothee Besseyre de Horts1, Stephanie Herveau1, Frank Herting3,Adriana Plesa2, Thomas Friess3, Pablo Umana4, Christian Klein3, and Charles Dumontet1,2 GA101 is a novel glycoengineered Type II CD20 monoclonal antibody. When compared with rituximab, it mediates less complement-dependent cytotoxicity (CDC). As expected for a Type II antibody, GA101 appearsnot to act through CDC and is more potent than the Type I antibody rituximab in inducing cell death vianonclassical induction of apoptosis cytotoxicity, with more direct cytotoxicity and more antibody-dependentcell-mediated cytotoxicity. We evaluated the antitumor activity of GA101 against the human-transformedfollicular lymphoma RL model in vivo in severe combined immunodeficient mice (SCID) mice. GA101induced stronger inhibition of tumor growth than rituximab. Combination of GA101 with cyclophosphamidein vivo confirmed the superiority of GA101 over rituximab. Neutralizing the complement system with cobravenom factor partially impaired the antitumor activity of rituximab, but had no impact on the efficacy ofGA101. In vitro GA101 more potently induced cell death of RL cells than rituximab. The expression of a limitednumber of genes was found to be induced by both antibodies after exposure in vitro. Among these, earlygrowth response 1 and activation transcription factor 3 were confirmed to be increased at the protein level,suggesting a possible role of these proteins in the apoptotic signalling of anti-CD20 antibodies. These dataimply that GA101 is superior to rituximab not only as a single agent, but also in combination withchemotherapy. These data suggest the presence of novel signalization pathways activated after exposureto anti-CD20 antibodies. Mol Cancer Ther; 10(1); 178–85.
2011 AACR.
despite its well-established clinical efficacy, a subpopu-lation of patients does not initially respond to rituximab Rituximab, directed against the CD20 antigen on B and most patients will relapse after rituximab therapy cells, was the first commercially available monoclonal (13, 14). Thus, there is still a need either for more antibody (mAb) for the treatment of lymphoma. It is the efficient rituximab combination therapies or for novel current treatment of choice for a variety of lymphopro- CD20-specific monoclonal antibodies with increased liferative disorders including low and high grade B-cell efficacy (15).
non-Hodgkin's lymphomas (NHL; refs 1–5). Follicular Various in vitro and in vivo experiments have shown lymphoma (FL) is the most common subtype of indolent that elimination of CD20þ lymphoma cells by rituximab lymphoma. Rituximab is now widely used either alone involves complement-dependent cytotoxicity (CDC; refs or in combination with multi-agent chemotherapy for 16–23), direct induction of apoptotic signalling (24–26), as the treatment of FL, either at diagnosis (6, 7), at relapse well as the recruitment of effector cells leading to anti- (8–10), or for maintenance therapy (2, 11, 12). However, body-dependent cell-mediated cytotoxicity (27). Never-theless, the in vivo mechanism of action of and resistanceto rituximab are not fully understood (28). Depending on Authors' Affiliations: 1Universit
e de Lyon and INSERM, U590; 2Hospices Civils de Lyon, Lyon, France; 3Discovery Oncology, Pharma Research, their ability to redistribute CD20 into lipid rafts and to Roche Diagnostics GmbH, Penzberg, Germany; and 4GlycArt Biotechnol- induce CDC, or to induce cell death and homotypic ogy AG, Schlieren, Switzerland adhesion, anti-CD20 mAb can be classified as Type I or Note: Supplementary material for this article is available at Molecular II CD20 antibodies. Schematically, Type I antibodies Cancer Therapeutics Online (http://mct.aacrjournals.org/).
(such as rituximab) induce CDC with redistribution of Corresponding Author: Charles Dumontet, INSERM U590, Laboratoire CD20 into lipid rafts, whereas type II mAB (such as the de Cytologie Analytique, Faculte de Medecine Rockefeller, UniversiteClaude Bernard Lyon I, Lyon, France. Phone: þ33478777236; Fax: murine antibody tositumomab, or the humanized anti- body GA101) are believed to act primarily through the direct induction of nonclassical cell death and exhibit low
2011 American Association for Cancer Research.
CDC activity (29).
Mol Cancer Ther; 10(1) January 2011 Preclinical Studies on the Mechanism of Action Rituximab as a chimeric mAb belongs to the first gen- ment inhibition was induced by weekly intraperitoneal eration of CD20 antibodies recognizing a Type I epitope.
injection of cobra venom factor (CVF; 2 mg/mouse, Qui- While the second generation of humanized (ocrelizumab, del Corporation).
veltuzumab, AME-133, Immu-106) or fully human (ofa- Combination studies were done with cyclophospha- tumumab) anti-CD20 antibodies recognize a Type I epi- mide. In these combination studies, the treatment was tope, GA101 represents a novel generation that in addition administrated weekly. The control group received vehi- to being humanized recognizes a type II epitope and is cle (NaCl 0.9%), whereas the treated groups received glycoengineered using GlycoMAb technology leading to rituximab (30 mg/kg), GA101 (30 mg/kg), rituximab bisected, afucosylated fragment crystallizable (Fc) region (30 mg/kg) þ GA101 (30 mg/kg), cyclophosphamide carbohydrates resulting in enhanced affinity for the (50 mg/kg), rituximab þ cyclophosphamide (50 mg/kg), human FcgRIIIa receptor on human effector cells such and GA101 þ cyclophosphamide (50 mg/kg). Rituximab as NK cells, macrophages, and dendritic cells (29). GA101 and GA101 (30 mg/kg) were provided by Roche, whereas was obtained by grafting CDR sequences from the murine cyclophosphamide was obtained from Baxter. The mice mAB B-ly1 on framework regions with fully human IgG1- were injected intravenously in the tail vein, once a week.
kappa germline sequences.
They were weighed and the tumor size was measured In this study, we compared the effect of GA101 and twice a week with an electronic calliper. The tumor volume rituximab on the human follicular RL lymphoma model, (TV) was estimated from two dimensional tumor measure- both in vitro and in vivo.
ments by the formula: tumor volume (mm3) ¼ length (mm)  width2/2. Median tumor growth inhibition (% TGI) was Materials and Methods calculated according to the NCI formula: 1  ([TVtreated(day 34  20)  100/TVcontrol (day 34  20)  100]).
Cell lines and culture The RL cell line, derived from a human transformed Flow cytometry analysis FL sample, was purchased from American Type Culture Cell surface antigen expression of RL cells was per- Collection within the 6 months before experimentation formed on a FACS Calibur flow cytometer (Becton and routinely characterized before and during the Dickinson). Analysis of the data was done with the Cell experimentation regarding the CD19; CD20; HLA DQ, Quest software program (Becton Dickinson). Mouse and HLA DR expression. Cells were maintained in fluorochrome-conjugated isotype control antibodies, culture medium consisting of RPMI-1640 (Life Technol- phycocyanin 5 (PC5)-coupled anti-CD19, phycoerythrin ogies), 10% of fetal calf serum (Integro), 100 units/mL of coupled APC anti-CD20, fluorescein isothiocyanate penicillin and 100 mg/mL of streptomycin (Life Tech- (FITC)-coupled anti-CD59, and PE-coupled anti-CD55 nologies). All cells were cultured at 37C in a 5% CO2 were purchased from Immunotech. FITC-coupled anti- CD46 and FITC active caspase-3 apoptosis kit werepurchased from Becton Dickinson. Mean fluorescence intensity (MFI) was determined by subtracting the sig- Six-week-old female CB17 severe combined immuno- nal of isotype-matched antibody staining from the stain- deficient mice (SCID) mice purchased from Charles River ing observed with the specific primary antibody. RL laboratories (l'Arbresle) were bred under pathogen-free exposed to rituximab, GA 101 and CVF were firstly conditions at the animal facility of our institute. Animals evaluated in vitro, but were also evaluated ex vivo, were treated in accordance with the European Union immediately after the extraction of tumor cells from guidelines and French laws for the laboratory animal animals. These experiments were done after exposure care and use. The animals were kept in conventional of 50,000 cells in 6-well culture plates in 2 mL of housing. Access to food and water was not restricted.
complete medium to rituximab or GA101 with or with- This study was approved by the local animal ethical out CVF. These assays were done after 1, 2, 4, 6, and 24 hours of culture.
For xenograft experiments, 1  106 RL cells were injected subcutaneously on day 1. Mice were randomized Annexin V/Propidium iodide staining when a tumor became palpable in groups of 10 and To evaluate the induction of apoptosis and the reduc- treatment was initiated. In a first set of experiments, tion of cell viability in cells exposed to antibodies with or rituximab and GA101 were used as monotherapy at without CVF, 106 cells were resuspended in 300 mL of different dosages twice weekly. The 5 different groups human serum and 700 mL of culture medium with or of 10 mice were: control group receiving vehicle (NaCl without 100 mg/mL rituximab at 37C for 6, 15, and 24 0.9%), rituximab (30 mg/kg), GA101 (10 mg/kg), GA101 hours. Dead and viable cells were discriminated by (30 mg/kg), and GA101 (100 mg/kg). The treatment was Annexin V/propidium iodide (PI) staining using flow administered intravenously twice a week. The mice were cytometry. Briefly, the cells were washed and resus- closely monitored regarding weight and general status.
pended in binding buffer [10 mmol/L HEPES (pH 7.4), In experiments evaluating the role of CDC in rituximab 140 mmol/L NaCl, and 2.5 mmol/L CaCl2] containing inhibition of tumor growth in groups of 3 mice, comple- 1 mg/mL FITC-Annexin V PI (1.25 mg/mL) was also Mol Cancer Ther; 10(1) January 2011 Dalle et al.
added to the samples to distinguish between early apop- scanned using the Agilent model G2505B microarray scan- tosis and secondary necrosis.
ner, and data were extracted by Feature Extraction soft-ware, Western blot protein analysis photomultiplicator values (XDR high 100% and XDR Protein expression was determined by Western blot low 10%), were used to scan the microarrays. Data were analysis in rituximab-naive and rituximab-resistant normalized using the quantile normalization method (31).
tumors as previously described (30). Briefly, cell lysates Each sample was done in triplicate. Analyses of differen- were resolved by 12% SDS-PAGE, and transferred onto a tially expressed genes and Gene Ontology pathways (Gene polyvinylidene difluoride (PVDF) membrane (Hybond- Ontology Consortium, 2000, http://www.geneontology.
ECL). The blots were then incubated with the appropriate org/) were done using GeneSpring 7.0. Determination of dilution of primary antibody, followed by incubation differentially expressed genes was done using a parametric with peroxidase-conjugated secondary antibody. For this test, with a false discovery rate of 0.01. Quantitative RT- analysis, 107 cells were pelleted and proteins fractionated PCR confirmation of selected genes was done as previously by SDS-PAGE (12–15% gradient gels) and transferred to a described (30).
PVDF membrane using an electroblotting apparatus (Bio-Rad). The loading of equal amounts of protein was Statistical analysis verified by Ponceau staining of the PVDF membranes.
For the evaluation of tumor growth, calculations The membrane was blocked with 5% nonfat, dry milk for started at staging (day 34) until termination for the con- 1 hour and subsequently incubated with the primary trol group and the group receiving therapy. Values were antibody at a dilution of 1:1,000 for 1 hour at room documented as medians and standard deviations (SD).
temperature. Antibody directed against Bcl2 was pur- Median (%) TGI for volume (T/C) was calculated accord- chased from Dako (clone 124), YY1 from Active Motif, ing to the NCI formula:1 - ([TV_treated (day Y  100/ TV_control (day X  34)  100]) L (clone S18) from Santa Cruz; BAX from Santa Cruz (clone SC 493), BAK from Santa Cruz (clone SC Briefly, in a randomized 2-sample design the treat- 7873), BIM from Santa Cruz (clone SC 8265), CD59 from ment-to-control ratio: Serotec (Clone mem-43), CD55 from Abcam (MEM-118), CD20 from Abcam (clone L26), early growth response 1 (EGR1) and activation transcription factor 3 (ATF3) fromSanta Cruz, and caspase-3 from BD Biosciences (clone and its 2-sided nonparametric (1-a) confidence interval CPP32). Unbound antibody was removed by washing according to Fieller (1954)/Hothorn and Munzel (2000) with phosphate buffered saline (pH 7.2) containing 0.1% were estimated. The calculations were done with the Tween 20 and 5% nonfat, dry milk. The membrane was special SAS program TUMGRO (version 3) using version then incubated with the secondary antibody (anti 8.1 (SAS Inc. Cary, 2000).
mouse peroxydase-conjugated antibody [Sigma] at adilution of 1:6,000) for 1 hour at room temperature.
After extensive washing with phosphate bufferedsaline, proteins were detected after addition of the Inhibition of tumor growth in vivo by rituximab or staining substrates ECL (Amersham). The proteins were detected by chemiluminescence using Kodak film (East- The efficacy of GA101 and rituximab was compared in man Kodak Company) or using the Odyssey infrared the RL model that we recently described (30). In the first system (LI-COR Biotechnology). The Western blot study GA101 was administered i.v. twice weekly at 3 analyses were done for each animal from the different dosages (10, 30, and 100 mg/kg), whereas rituximab was groups of animals.
given at fixed dose of 30 mg/kg twice weekly (Fig. 1).
Both antibodies were administered as intravenous injec- Gene expression profiling tions, for a total of 5 injections. As shown in Figure 1, we To determine which genes were differentially expressed observed that the new CD20 antibody GA101 was more in cells exposed to rituximab or GA101, RL cells were active than rituximab administered at similar doses on exposed in vitro and in vivo to these antibodies then ana- established RL tumors. The antitumor effect of GA101 lysed by pangenomic profiling using Agilent 44K chips in against RL xenografts was dose dependent in terms of the Laboratoire de Caract
erisation Mol
eculaire des TGI. TGI was calculated using NCI formula at day 34 and Tumeurs (LCMT). Briefly, 1-color labeled cRNAs were showed values of 25, 75, and 85% for the 10, 30, and generated from 200 ng of total RNA using the Low RNA 100 mg/kg dosages of GA101, respectively, whereas the Input Amplification Kit (Agilent Technologies) according 30 mg/kg dose of rituximab induced a TGI of 43%. The to the instructions of the manufacturer. Labeled cRNA higher doses of 30 and 100 mg/kg of GA101 significantly were hybridized overnight to Whole Human Genome 4 inhibited the growth of RL tumors and resulted in some  44K microarrays (ref Agilent G4112F) containing 45,015 complete tumor remissions (10% and 30%, respectively), features representing 41,000 genes. Each probe is a 60-mer, whereas no complete tumor remissions were observed synthesized in situ. After washing, microarrays were in the rituximab group. Taken together, the antitumor Mol Cancer Ther; 10(1) January 2011 Molecular Cancer Therapeutics

Preclinical Studies on the Mechanism of Action Figure 2. Effect of combination therapy of GA101 or rituximab with Figure 1. Inhibition of tumor growth in vivo by rituximab or GA101. Mice cyclophosphamide. Mice-bearing established SC RL tumors were treated injected with RL cells subcutaneously were treated by IV infusion twice a weekly (on days 31, 38, 45 and 52 (black crosses). The control group week starting on day 17 and ending on day 31 (black crosses). The tumors received vehicle (NaCl 0.9%), whereas the treated groups received one of were measured twice a week. The mice were euthanized when the tumor the following: rituximab (30 mg/kg), GA101 (30 mg/kg), rituximab (30 mg/ volume reached 2 cm3. The difference between GA (30 mg/kg) and kg) þ GA101 (30 mg/kg), cyclophosphamide (CPM; 50 mg/kg), rituximab þ rituximab (30 mg/kg) was significant (P ¼ 8.105) CPM (50 mg/kg), GA101 þ CPM (50 mg/kg). The difference betweenrituximab þ CPM and GA101 þ CPM was significant (P ¼ 0.05) activity of rituximab against RL xenografts was inferior to tolerated with no toxic deaths, nor loss of body weight an equivalent dosing of GA101. Tolerability of GA101 greater than 10% (data not shown).
with these regimens was excellent and no significantmodification of body weight was observed. Since there Role of complement in the antitumor effect of was no significant difference between the 30 mg/kg and antibodies in vivo 100 mg/kg doses of GA101, the 30 mg/kg was used for Complement-dependent cytotoxicity appears to play a subsequent combination studies.
key role in the efficacy of rituximab in the RL model (30).
When GA101 or rituximab were administered in combi- Combination of cyclophosphamide with rituximab nation with CVF, we observed a significant loss of anti- tumor activity in the rituximab group, whereas we did In a separate series of experiments, rituximab 30 mg/kg not observe a loss of efficacy in the GA101 group (Fig. 3).
and GA101 30 mg/kg were administered once weekly i.
No difference was observed between the rituximab and v. for 4 weeks, either with or without cyclophospha- GA101 groups, we assumed that it was induced by the mide 50 mg/kg administered once weekly i.p. for 4 small number of mice in each group in these experiments.
weeks. As shown in Figure 2, this study confirmed the When RL cells were exposed to rituximab or GA101 previous finding that the new anti-CD20 antibody in vitro, the addition of 30% human serum as a source GA101 was more active against established RL tumors of complement increased the apoptotic fraction in the than rituximab administered at similar doses. TGI case of rituximab but not in the case of GA101 (data not values at day 42 were 79% for GA101, 35% for ritux- shown), thus supporting the lack of CDC in the case of imab, and 93% for cyclophosphamide administered as single agents when compared with untreated controls.
When groups receiving combination therapy were Flow cytometry and Western blot analyses of tumor compared with the groups receiving the corresponding cells exposed to antibodies in vitro and in vivo single agent antibody, cyclophosphamide increased To assess the direct effect of GA101 on RL cells, we did antitumor efficacy with TGI values of 83% at day 42 PI/Annexin testing. We observed more apoptotic cells and 55% at day 66% for rituximab and 94% at day 42 (early and late apoptosis) in cells in vitro exposed to and 88% at day 66 for GA101, respectively. Taken GA101 than in cells exposed to rituximab. This difference together, the GA101-cyclophosphamide combination was seen after time exposure varying from 6 to 24 hours was significantly better than the rituximab-cyclopho- (Fig. 4), and disappeared after 48 and 72 hours (data not sphamide combination in this setting. Thus, when shown). For instance apoptotic cells (early and late apop- using a suboptimal dose of the classical antilymphoma tosis) represented 6.70%, 10.72%, 17.35%, respectively, in alkylating agent cyclophosphamide, the combination of untreated and rituximab- and GA101-treated cells after either antibody with cyclophosphamide was more 15 hours of exposure to the treatment. As a consequence, active than either agent alone, and the most active the percentage of live cells was more strongly reduced in combination was GA101 in combination with cyclopho- the GA101 group compared with the rituximab group sphamide. In all cases, the administration was well (Fig. 4). In agreement with this, procaspase-3 protein was Mol Cancer Ther; 10(1) January 2011 Dalle et al.
Figure 3. Effect of CVF on thein vivo antitumor efficacy ofantibodies. The mice (3 animalsper group) were treated by IVinfusion two times a week startingon day 17 with rituximab (30 mg/kg) and GA101 (30 mg/kg) with orwithout weekly injection of CVF(2 mg/mouse). The differencebetween rituximab and rituximabþ CVF was significant (P < 0.05),whereas the difference betweenGA101 and GA101 þ CVF was notstatistically significant.
found to be more strongly expressed in cells exposed to phase (6 hours), it became detectable on FACS analysis GA101 than in cells exposed to rituximab from 6 hours from 12 to 48 hours after exposure (Fig. 5). There were no after the treatment start (data not shown). Although differences concerning Bim, Bak, Bcl2, Bcl-xL, caspase-8, cleaved form of caspase-3 was not detected in the early and caspase-9 by western-blot analysis and CD20 Figure 4. Induction of apoptosis by rituximab and GA101 on RL cells. Induction of apoptosis by rituximab and GA101 on RL cells was evaluated by flowcytometry. Apoptosis was evaluated at various time points after exposure to rituximab or GA101 followed by staining with Annexin V and PI. The dot plotsrepresent the PI and Annexin V expression for untreated and rituximab- and GA101-treated cells after 15 hours of in vitro exposure. The tableshows the results observed in each groups (alive, early, and late apoptotic, necrotic cells) in percentage of 10,000 analyzed cells per condition.
Mol Cancer Ther; 10(1) January 2011 Molecular Cancer Therapeutics Preclinical Studies on the Mechanism of Action Figure 5. Effect of rituximab and GA101 on caspase-3 expression in RL cells in vitro. Cleaved caspase-3 expression was evaluated in vitro by FACSanalysis after 6, 24, and 48 hours of exposure to either GA101 or rituximab. Percentage represents the number of positive cells.
expression both by FACS and Western blot analysis (data imab and to GA101 were analysed by RT-PCR (data not not shown). When cells were co-incubated with antibo- shown). Among these, overexpression of EGR1 and ATF3 dies and CVF, the quantities of caspase-3 appeared to be were confirmed in 3 independent experiments. EGR1 was decreased in the case of rituximab-exposed cells but not increased up to 10-fold and ATF3 to 4-fold in RL cells in the case of GA101-exposed cells.
after exposure to antibodies (P < 0.01). Increased expres- The expression levels of CD19, as well as of comple- sion of the corresponding proteins was also documented ment inhibitors CD46, CD55, and CD59, on RL cells were by immunoblotting (Fig. 6). We did not observe increase studied after 1, 2, 4, 6, and 24 hours in vitro exposures to in Bax protein level expression.
GA101 or rituximab, in the presence or absence of CVF.
There was no change in the CD19 expression after expo-sure to rituximab or GA101 (data not shown). In addition,we did not observe any upregulation of the CD46, CD55,and CD59 antigens in tumor cells exposed in vivo to eitherantibody. Although procaspase-3 was found slightlyincreased by Western blot analysis in the GA101-treatedgroup in comparison with untreated or rituximab-treatedmice 12 hour after exposure, we did not observe cleavedform of caspase-3 expression following these in vivoexperiments (data not shown).
Gene expression arrays In RL cells exposed to mAb in vitro for 6 hours, a total of 867 genes were induced by rituximab and 664 by GA101, Figure 6. Effect of rituximab and GA101 on EGR1, Bax, and ATF3 including 152 genes induced by both antibodies (See expression in RL tumors. EGR1, Bax, and ATF3 protein levels were Supplementary Table 1). A series of genes found to be evaluated by Western blotting after exposure of RL cells to rituximab and significantly overexpressed after exposure both to ritux- GA101 in vitro.
Mol Cancer Ther; 10(1) January 2011 Dalle et al.
Gene ontology pathways of the genes induced by a greater induction of early cell death as well as higher rituximab and GA101 are presented in Supplementary expression of procaspase-3 and Bad protein after expo- Tables 2 and 3. Interleukin 12 (IL-12) biosynthesis, sure to GA101 than after exposure to rituximab. Although glucose import, and ribosome biogenesis were among the activated form of caspase-3 was not detected in our the pathways most significantly enriched after exposure animal model euthanized 12 hours after last exposure to to rituximab. Response to biotic stimulus or parasites, treatment, we observed in additional in vitro kinetic immune activation, and IL-1 biosynthesis were among experiments that the cleaved-caspase-3 form was mainly the pathways most significantly enriched after exposure detectable after 24 and 48 hours of exposure.
to GA101. Several pathways were found to be enriched The signalization pathways involved in anti-CD20 with both antibodies, including IL-1, IL-6, IL-13 bio- antibody-mediated cytotoxicity have been explored synthesis as well as negative regulation of cell differ- by several groups. Bonavida and Jazirehi have reported entiation and osteoclast differentiation. It thus appears the importance of RKIP as a central regulator of ritux- that the pathways activated by these 2 anti-CD20 anti- imab-induced cytotoxicity, regulating BclX-L (35).
bodies partially overlap but differ significantly in other Leseux et al. have found that PKCz was involved in cytotoxicity of rituximab in the RL line (36). In thisstudy, we identified 2 novel proteins potentially involved in CD20-mediated cytotoxicity. ATF3 andEGR1 were overexpressed both after exposure to ritux- This preclinical study was initiated to compare the in imab and to GA101, suggesting that they may be vivo efficacy of rituximab, the first in class anti-CD20 involved in CD20-mediated signaling. EGR1 or early mAb, and GA101, a novel generation CD20 antibody, growth response 1 is a zinc finger protein that has been in a preclinical model of human NHL. The results show reported to possess both oncogenic and tumor suppres- the superior antitumor activity of GA101 in a model of sor properties. Zheng et al. have shown that NFkB- human FL RL grown as xenografts in SCID mice, either as induced EGR1 transcription allowed survival of pros- a single agent or in combination with cyclophosphamide, tate tumor cells, whereas other authors have suggested in comparison to rituximab. The effect of GA101 did not a proapoptotic role by induction of Bax (37, 38). ATF3, appear to involve complement, whereas the effect of or activating transcription factor 3, is a leucine zipper rituximab in this model was at least partially complement protein involved in cellular stress pathways. ATF3 has dependent. Tolerability of GA101 with these regimens been reported to behave as an oncogene in murine was excellent with no toxic deaths and no significant mammary carcinogenesis, but has also been found to modification of body weight.
be involved in apoptosis in prostate cancer (39, 40).
The role of complement in anti-CD20 treatment is Additional experiments are required to show whether now a matter of debate. CDC is clearly involved in the overexpression of these genes after exposure to the action of rituximab both in vitro and in preclinical anti-CD20 monoclonal antibodies is required for mouse models (21, 22, 32–34). Manches et al. have also reported a correlation between in vitro sensitivity to In conclusion, these results show that GA101 used as CDC of various lymphoma subtypes and the likelihood a single agent or in combination with cyclophospha- of response to rituximab in the clinic (17). Conversely, it mide is more active than rituximab on human lym- has been suggested that deposition of inactivated C3b phoma RL xenografts. As expected for a Type II on rituximab-coated cells could diminish the interaction antibody, GA101 appears not to act through CDC, between the Fc region and CD16 on NK cells. Our in vivo and is more potent than the Type I antibody rituximab results using rituximab corroborate the previously in inducing cell death via nonclassical induction of reported data and support the role of complement in rituximab-mediated antitumor activity. However, wedid not observe variations in the expression of the Disclosure of Potential Conflicts of Interest complement inhibitors CD46, CD55, or CD59 in ourmodel. These antigens have been reported to be altered S. Dalle received a research grant from Roche to Inserm.
in cells exposed to antibodies or to be correlated withresponse to anti-CD20 antibodies (20). In the case ofGA101, the role of CDC is clearly less important both in vitro and in vivo (29).
This work was funded in part by a grant support from Roche to In the SCID mice model, it is not expected that a major contribution to efficacy comes from the optimized inter- The costs of publication of this article were defrayed in part by the action of the glycoengineered Fc-part of GA101 with the payment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicate murine FcgRIV receptors expressed on macrophages/ monocytes. Thus, the direct induction of cell death islikely to play an important role in GA101-mediated Received April 23, 2010; revised October 15, 2010; accepted October cytotoxicity. In support of this hypothesis, we observed 27, 2010; published online January 10, 2011.
Mol Cancer Ther; 10(1) January 2011 Molecular Cancer Therapeutics Preclinical Studies on the Mechanism of Action Dillman RO. Treatment of low-grade B-cell lymphoma with the mono- 21. Golay J, Cittera E, Di Gaetano N, et al. The role of complement in the clonal antibody rituximab. Semin Oncol 2003;30:434-47.
therapeutic activity of rituximab in a murine B lymphoma model Coiffier B. Rituximab therapy in malignant lymphoma. Oncogene homing in lymph nodes. Hematological 2006;91:176-83.
22. Di Gaetano N, Cittera E, Nota R, et al. Complement activation Hiddemann W, Kneba M, Dreyling M, Schmitz N, Lengfelder E, determines the therapeutic activity of rituximab in vivo. J Immunol Schmits R, et al. Frontline therapy with rituximab added to the combination of cyclophosphamide, doxorubicin, vincristine, and pre- 23. Weng WK, Levy R. Expression of complement inhibitors CD46, CD55, dnisone (CHOP) significantly improves the outcome for patients with and CD59 on tumor cells does not predict clinical outcome after advanced-stage follicular lymphoma compared with therapy with rituximab treatment in follicular non-Hodgkin lymphoma. Blood CHOP alone: results of a prospective randomized study of the Ger- man Low-Grade Lymphoma Study Group. Blood 2005;106:3725-32.
24. Shan D, Ledbetter JA, Press OW. Signaling events involved in anti- Lin TS, Lucas MS, Byrd JC. Rituximab in B-cell chronic lymphocytic CD20-induced apoptosis of malignant human B cells. Cancer Immu- leukemia. Semin Oncol 2003;30:483-92.
nol Immunother 2000;48:673-83.
Ghobrial IM, Gertz MA, Fonseca R. Waldenstrom macroglobulinae- 25. Chan HT, Hughes D, French RR, Tutt AL, Walshe CA, Teeling JL, et al.
mia. Lancet Oncol 2003;4:679-85.
CD20-induced lymphoma cell death is independent of both caspases Colombat P, Salles G, Brousse N, Eftekhari P, Soubeyran P, Delwail V, and its redistribution into triton X-100 insoluble membrane rafts.
et al. Rituximab (anti-CD20 monoclonal antibody) as single first-line Cancer Res 2003;63:5480-9.
therapy for patients with follicular lymphoma with a low tumor burden: 26. Beers SA, Chan CH, James S, French RR, Attfield KE, Brennan CM, clinical and molecular evaluation. Blood 2001;97:101-6.
et al. Type II (tositumomab) anti-CD20 monoclonal antibody out per- Hainsworth JD, Litchy S, Burris HA 3rd, Scullin DC Jr, Corso SW, Yardley forms Type I (rituximab-like) reagents in B-cell depletion regardless of DA, et al. Rituximab as first-line and maintenance therapy for patients complement activation. Blood 2008.
with indolent non-Hodgkin's lymphoma. J Clin Oncol 2002;20:4261-7.
27. Clynes RA, Towers TL, Presta LG, Ravetch JV. Inhibitory Fc receptors McLaughlin P, Grillo-L
opez AJ, Link BK, Levy R, Czuczman MS, modulate in vivo cytotoxicity against tumor targets. Nat Med 2000;6: Williams ME, et al. Rituximab chimeric anti-CD20 monoclonal anti- body therapy for relapsed indolent lymphoma: half of patients respond 28. Bonavida B. Rituximab-induced inhibition of antiapoptotic cell survi- to a four-dose treatment program. J Clin Oncol 1998;16:2825-33.
val pathways: implications in chemo/immunoresistance, rituximab Davis TA, Grillo-Lopez AJ, White CA, McLaughlin P, Czuczman MS, unresponsiveness, prognostic and novel therapeutic interventions.
Link BK, et al. Rituximab anti-CD20 monoclonal antibody therapy in non-Hodgkin's lymphoma: safety and efficacy of re-treatment. J Clin unker P, Moser S, Püntener U, Schmidt C, Herter S, et al. Increasing the efficacy of CD20 antibody therapy through the 10. Lemieux B, Bouafia F, Thieblemont C, Hequet O, Arnaud P, Tartas S, engineering of a new type II anti-CD20 antibody with enhanced direct- et al. Second treatment with rituximab in B-cell non-Hodgkin's lym- and immune effector cell-mediated B-cell cytotoxicity. Blood 2010; phoma: efficacy and toxicity on 41 patients treated at CHU-Lyon Sud.
Hematol J 2004;5:467-71.
30. Dalle S, Dupire S, Brunet-Manquat S, Reslan L, Plesa A, Dumontet C.
11. Ghielmini M, Schmitz SF, Cogliatti SB, Pichert G, Hummerjohann J, In vivo model of follicular lymphoma resistant to rituximab. Clin Cancer Waltzer U, et al. Prolonged treatment with rituximab in patients with Res 2009;15:851-7.
follicular lymphoma significantly increases event-free survival and 31. Bolstad BM, Irizarry RA, Astrand M, Speed TP. A comparison of response duration compared with the standard weekly  4 schedule.
normalization methods for high density oligonucleotide array data based on variance and bias. Bioinformatics 2003;19:185-93.
12. Hainsworth JD, Litchy S, Shaffer DW, Lackey VL, Grimaldi M, Greco 32. Cragg MS, Morgan SM, Chan HT, et al. Complement-mediated lysis FA. Maximizing therapeutic benefit of rituximab: maintenance therapy by anti-CD20 mAb correlates with segregation into lipid rafts. Blood versus re-treatment at progression in patients with indolent non- Hodgkin's lymphoma—a randomized phase II trial of the Minnie Pearl 33. Daniel D, Yang B, Lawrence DA, Totpal K, Balter I, Lee WP, et al.
Cancer Research Network. J Clin Oncol 2005;23:1088-95.
Cooperation of the proapoptotic receptor agonist rhApo2L/TRAIL 13. Smith MR. Rituximab (monoclonal anti-CD20 antibody): mechanisms with the CD20 antibody rituximab against non-Hodgkin lymphoma of action and resistance. Oncogene 2003;22:7359-68.
xenografts. Blood 2007;110:4037-46.
14. Maloney DG. Immunotherapy for non-Hodgkin's lymphoma: mono- 34. Cittera E, Leidi M, Buracchi C, Pasqualini F, Sozzani S, Vecchi A, et al.
clonal antibodies and vaccines. J Clin Oncol 2005;23:6421-8.
The CCL3 family of chemokines and innate immunity cooperate in vivo in 15. Dalle S, Thieblemont C, Thomas L, Dumontet C. Monoclonal antibo- the eradication of an established lymphoma xenograft by rituximab. J dies in clinical oncology. Anticancer Agents Med Chem 2008;8:523-32.
16. Anderson DR, Grillo-Lopez A, Varns C, Chambers KS, Hanna N.
35. Jazirehi AR, Bonavida B. Cellular and molecular signal transduction Targeted anti-cancer therapy using rituximab, a chimaeric anti- pathways modulated by rituximab (rituxan, anti-CD20 mAb) in non- CD20 antibody (IDEC-C2B8) in the treatment of non-Hodgkin's B- Hodgkin's lymphoma: implications in chemosensitization and ther- cell lymphoma. Biochem Soc Trans 1997;25:705-8.
apeutic intervention. Oncogene 2005;24:2121-43.
17. Manches O, Lui G, Chaperot L, Gressin R, Molens JP, Jacob MC, et al.
36. Leseux L, Laurent G, Laurent C, Rigo M, Blanc A, Olive D, Bezombes In vitro mechanisms of action of rituximab on primary non-Hodgkin C. PKC zeta mTOR pathway: a newtarget for rituximab therapy in lymphomas. Blood 2003;101:949-54.
follicular lymphoma. Blood 2008;111:285–91.
18. Golay J, Lazzari M, Facchinetti V, Bernasconi S, Borleri G Barbui T, 37. Zheng C, Ren Z, Wang H, Zhang W, Kalvakolanu DV, Tian Z, et al.
et al. CD20 levels determine the in vitro susceptibility to rituximab and E2F1 Induces tumor cell survival via nuclear factor-kappaB-depen- complement of B-cell chronic lymphocytic leukemia: further regula- dent induction of EGR1 transcription in prostate cancer cells. Cancer tion by CD55 and CD59. Blood 2001;98:3383-9.
19. Bellosillo B, Villamor N, L
opez-Guillermo A, Marcé S, Esteve J, Campo 38. Zagurovskaya M, Shareef MM, Das A, et al. EGR-1 forms a complex E, et al. Complement-mediated cell death induced by rituximab in B- with YAP-1 and upregulates Bax expression in irradiated prostate cell lymphoproliferative disorders is mediated in vitro by a caspase- carcinoma cells. Oncogene 2009;28:1121-31.
independent mechanism involving the generation of reactive oxygen 39. Wang A, Arantes S, Yan L, Kiguchi K, McArthur MJ, Sahin A, et al. The species. Blood 2001;98:2771-7.
transcription factor ATF3 acts as an oncogene in mouse mammary 20. Golay J, Zaffaroni L, Vaccari T, et al. Biologic response of B lymphoma tumorigenesis. BMC Cancer 2008;8:268.
cells to anti-CD20 monoclonal antibody rituximab in vitro: CD55 and 40. Huang X, Li X, Guo B. KLF6 induces apoptosis in prostate cancer cells CD59 regulate complement-mediated cell lysis. Blood 2000;95:3900-8.
through up-regulation of ATF3. J Biol Chem 2008;283:29795-801.
Mol Cancer Ther; 10(1) January 2011

Source: http://www.antineo.fr/assets/2011-Dalle.pdf