immunotoxin-mediated inhibition of chronic lymphocytic leukemia

(CANCER RESEARCH 49, 3328-3332, June 15, 1989]

Immunotoxin-mediated Inhibition of Chronic Lymphocytic Leukemia CellProliferation in Humans1

Salvatore Siena,2 Marco Bregni, Anna Formosa, Bruno Brando, Paola Marenco, Douglas A. Lappi,

Gianni Bonadonna, and A. Massimo GianniCristina Candirti Bone Marrow Transplantation Unit, Division of Medical Oncology, Istituto Nazionale Tumori, Via Venezian 1, Milan 20133 [S. S., M. B., A. F.,G. B., A. M. G.]; Renal Transplantation Unit, Division ofNephrology [B. B.] and Division of Hematology [P. M.], Ospedale Niguarda-CÃ Gronda, Milan 20162; andImmunology Line, Farmitalia Carlo Erba, Milan 20146 [D. A. L.J, Italy

ABSTRACT

We evaluated the cytotoxicity of the immunotoxin OKT1-SAP onfresh B-chronic lymphocytic leukemia (B-CLL) cells from 31 consecutivepatients. OKT1-SAP comprised the OKT1 (CDS) monoclonal antibodydisilludi' linked to saporin-6 (SAP) ribosome-inactivating protein fromthe plant Saponaria officinali!. The effect of OKT1-SAP on target CD5-positive B-CLL cells was estimated using an in vitro proliferation inhibition assay in which control or OKTl-SAP-treated B-CLL cells wereinduced to proliferate by sequential stimulation with insolubilized anti-C3b receptor CB04 (CD35) antibody and low molecular weight B-cellgrowth factor. In 90% of patients, OKT1-SAP specifically suppressedB-CLL cell proliferation in a dose-related manner (50% inhibitory concentration, 4.0-6.8 UM). Taken together the findings reported in thisarticle provide information relevant to the clinical development of im-munotoxins because: (a) the in vitro conditions under which B-CLL cellproliferation is inhibited by OKT1-SAP are achievable in vivo withoutnonspecific toxicity according to our previous toxicology and pharmaco-kinetic studies in primates; and (ft) the B-CLL cell proliferation inhibition

assay described here provides a basis for future comparative studies.

INTRODUCTION

Immunotoxins, artificially constructed hybrid moleculescomposed of cell-specific antibodies coupled to bacterial orplant toxins, represent a new family of immunotherapeuticreagents potentially useful in cancer therapy and immunoreg-ulation (1). Initial clinical trials have utilized ricin A-chainimmunotoxins as systemic therapeutic agents in patients withgraft-vmMS-host disease (2), acute lymphoblastic leukemia, andB-chronic lymphocytic leukemia (3, 4). In these clinical trialsthe ability of ricin A-chain immunotoxins to eradicate targetcells in vivo has been unimpressive with the possible exceptionof graft- vOTiu-host disease (2). Interestingly, these clinical studies have allowed the understanding of a series of biologicalproblems encountered with the utilization of ricin A-chainimmunotoxins. First, ricin A-chain immunotoxins even at saturating concentrations in vitro as well as in vivo exhibit negligible cytotoxic activity against target malignant cells (5). Thislimitation has been obviated in the ex vivo setting for bonemarrow transplantation by devising optimal culture conditionsnonreproducible in vivo including an alkaline medium pH andthe addition of adjuvant molecules such as ammonium chloride,chloroquine, amantadine, or monensin which all raise the intra-lysosmal pH and thus most likely prevent the proteolytic inac-tivation of the immunotoxin (5, 6). Secondly, ricin A-chainimmunotoxins, once administered i.v. into a living individual,possess a short bioavailability mostly because some carbohydrate residues on their A-chain moiety mediate the specificrecognition and uptake by liver reticuloendothelial and paren-

Received 11/15/88; revised 3/6/89; accepted 3/22/89.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

' This study was partially supported by CNR Grant 87-01519-44 to A. M. G.2To whom requests for reprints should be addressed.

chymal cells (4, 7). Based on these studies, it is now evidentthat for the ultimate development of immunotoxins as therapeutic agents useful in vivo, it is necessary to search for newtypes of immunotoxins with increased cytotoxic activity andresistance to inactivation as well as better characteristics ofpharmacokinetics.

We have recently reported that conjugation of SAP3 to the

CDS-specific monoclonal antibodies OKT1 or SOT la yieldsimmunotoxins, referred to as OKT1-SAP and SOT la-SAP,respectively, which are capable of rapid, potent, and specifickilling of target CDS-positive human T-lymphocytes underphysiological culture conditions and also in the absence ofadjuvant molecules artificially added to the incubation medium(8, 9). In addition, we demonstrated that a single nontoxic doseof the immunotoxin OKT1-SAP administered i.v. to nonhumanprimates not only remains structurally intact into the circulation but also retains its biological activity and is not subject todegradation that could prevent its cytotoxic action on targetcells up to 48 h after administration (8). These distinctivefunctional characteristics of OKT1-SAP, likely to be relevantfor its in vivo utilization, are attributable to the SAP moietythat differs from ricin A-chain in that: (a) it is devoid ofcarbohydrate residues; (b) it is very resistant to proteolyticinactivation; and (c), as suggested by Thorpe et al. (10), underphysiological conditions it is positively charged (pi > 10) andthus gives rise together with the negatively charged antibody toa compact molecule which is more protected in vivo fromdegradation due to oxidation of the disulfide linkage.

In the perspective of a clinical application of OKT1-SAP, weelected to investigate the effect of this immunotoxin against B-CLL as the comparison with the recent clinical trials conductedwith anti-CD5 ricin A-chain immunotoxins for the same disease(3, 4) would allow the discernment of whether or not thepotentially favorable characteristics of OKT1-SAP would provereal in the clinical setting. B-CLL cells display the CDS determinant, a M, 65,000 glycoprotein antigen (11) against whichOKT1-SAP immunotoxin is directed. The recent availability ofa methodology for inducing the proliferation of quiescent B-CLL cells (12) prompted us to devise a preclinical proliferationinhibition assay to evaluate the effect of OKT1-SAP immunotoxin on target CDS-positive B-CLL cells. Herein, we demonstrate that in most of 31 consecutive patients the proliferationof B-CLL cells in vitro can be specifically inhibited by OKT1-SAP under conditions achievable in vivo and without the needof artificial potentiators. Taken together the findings reported

3The abbreviations used are: SAP, saporin-6 ribosome-inactivating protein;B-CLL, B-chronic lymphocytic leukemia; CB04-S, monoclonal antibody CB04(CD35) covalently linked to Sepharose CL-4B; CD, monoclonal antibody clusterof differentiation; CR1, C3b complement receptor; IC50, immunotoxin concentration inhibiting 50% of either DNA or protein synthesis, as specified; L-BCGF,low molecular weight B-cell growth factor; OKT1, murine anti-human T-lympho-cyte monoclonal antibody; OKT1-SAP, immunotoxin comprising the monoclonalantibody OKT1 (CDS) linked to SAP ribosome-inactivating protein; PBMC,peripheral blood mononuclear cells; T101-RTA, monoclonal antibody T101(CDS) linked to ricin A-chain ribosome-inactivating protein.

3328

on March 17, 2018. © 1989 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

IMMUNOTOXIN AGAINST CHRONIC LYMPHOCYTIC LEUKEMIA

in this article provide information relevant to the clinical development of immunotoxins because: (a) the in vitro conditionsunder which B-CLL cell proliferation is inhibited by OKT1-

SAP are achievable In vivo without nonspecific toxicity according to our previous toxicology and pharmacokinetic studies inprimates; and (b) the B-CLL proliferation inhibition assaydescribed here provides a basis for future comparativestudies.MATERIALS

ANDMETHODSPatients.

Table 1 summarizes the main clinical characteristics of the31 consecutive patients with B-CLL analyzed in this study. All patientsgave informed consent for research blood samplings. The diagnosis ofB-CLL was based on classic clinical and immunological criteria. Theimmunologie;!] phenotype was routinely determined by indirect im-munofluorescence using the following panel of monoclonal antibodies:OKT3 (CD3) (Ortho Systems, Raritan, NJ); OKT1 (CDS) (Ortho);Mol (CD lib) (Coulter, Hialeah, FL); HD37 (CD 19) (gift from Professor B. Dorken, Heidelberg, West Germany); Bl (CD20) (Coulter);and HD6 (CD22) (gift from Professor B. Dorken). Characteristically,B-CLL cells were CD3 negative, CDS positive, CD1 Ib negative, CD 19positive, CD20 positive, CD22 positive, and either Kor X light chainsurface immunoglobulin positive.

OKT1-SAP Ininninoti i\ in. OKT1 monoclonal antibody and SAPribosome-inactivating protein were purified and then conjugated bythe cross-linking agent A'-succinylmidyl-3-(2-pyridyldithio)proprionate(Pharmacia Fine Chemicals, Uppsala, Sweden) as previously reportedin detail (8). OKT1 reacts with the Tl (CDS) surface antigen, a M,65,000 glycoprotein expressed on 90 to 95% of human peripheral bloodT-lymphocytes, most T-cell-derived hematological malignancies, B-CLL, and some B-cells involved in the pathogenesis of autoimmunediseases. SAP is a type 1 ribosome-inactivating protein or hemotoxinisolated from the seeds of the plant Saponaria officinales (8, 9). Theresulting immunotoxin, referred to as OKT1-SAP (batchDAL-E-76),Table

1 Clinical characteristics of patients with chronic lymphocyticleukemiaPhenotype

ofAntileukemic therapy PBMC(%Rai's positive)

clinicalCaseUPN" Age Sex stage Previous Present CD3 CDS CD191

74-0115 75 M 2 Chi None 17 90 632 83-0043 74 M 0 Chi PDN PDN 8 96 873 83-0154 73 M 2 Chi Chi PDN 1 98 974 81-0916 51 F 4 Chi Chi 2 97 935 83-0275 55 F 4 Chi Chi 13 94 766 85-0783 62 M 1 Chi Chi 7 95 877 79-0374 72 F 0 Chi None 4 84 888 87-0019 66 M 2 Chi PDN Chi PDN 19 46 639 83-0154 73 M 2 Chi Chi PDN 0 89 97

10 87-0203 77 F 0 None None 4 92 8811 83-0111 75 M 4 Chi PDN None 9 96 8612 85-4274 63 M 0 None None 29 91 5013 84-7502 62 M 1 Chi PDN None 3 87 9414 85-9255 68 F 1 PDN PDN 5 ND ND15 85-4931 48 M 2 None None 25 93 6216 87-0479 67 M 2 None None 1 93 9017 85-0960 82 M 0 Chi PDN Chi PDN 5 97 8718 86-0370 59 M 2 None None 9 92 7819 71-0392 49 M 0 Chi PDN 3 95 9220 87-0780 81 F 4 Chi PDN Chi PDN 13 90 7621 82-0918 47 F 2 Chi PDN Chi PDN 3 96 9022 86-0745 72 M 1 None None 19 90 7723 87-0656 68 M 2 None None 8 83 8024 81-0656 47 M 1 TNI Chi CVP None 5 ND ND25 86-0618 64 M 0 None None 17 81 3426 86-1049 66 M 0 Chi Chi 6 95 8827 86-2214 51 F 2 Chi PDN None 14 95 8028 80-0633 74 M 0 Chi None 13 86 7529 79-0277 76 M 0 None None 6 87 7930 87-0098 59 M 0 None None 18 95 8131 87-1188 49 M 1 None None 13 95 84retained

the functions of its component parts: (a) inhibited proteinsynthesis in a cell-free reticulocyte assay (IC5<>0.12 nM); (b) bound to>90% of normal peripheral blood T-lymphocytes; and (c) specificallyinhibited the clonogenic growth and DNA synthesis of human T-

lymphocytes (ICso 0.30 nM), as characterized previously (8).Treatment of B-CLL Cells with OKT1-SAP Immunotoxin. PBMC

from patients with B-CLL were separated by density gradient centrif-ugation over Ficoll-Paque (Pharmacia). T-lymphocytes (CDS-positivecells) in excess of 5% were removed by rosetting with 2-aminoethyli-sothiouronium-bromide-treated sheep erythrocytes. In the majority ofexperiments (patients 1 to 23), PBMC were incubated with variousconcentrations of OKT1-SAP at IO7cells/ml for 2 h at 37Â°Cin a 5%

CO2 humidified atmosphere in RPMI 1640 (Gibco, Grand Island, NY),10% fetal calf serum (Hybri-Sure, Hazleton, Denver, PA), and thenwashed four times at 4Â°C.In a series of experiments (patients 24 to

31), 50 nM OKT1-SAP was added directly to fresh unprocessed wholeB-CLL peripheral blood. Following 2 h incubation at 37Â°Cin a 5%

CO2 humidified atmosphere, PBMC were isolated by centrifugationover Ficoll-Paque and then washed four times at 4Â°C.Isolated B-CLL

cells were assayed for: (a) number of viable cells by trypan blue dyeexclusion test; and (b) synthesis of DNA in a 6-day [3H]thymidine

incorporation assay.Detection of OKT1-SAP Binding to B-CLL Cells. The binding of

OKT1-SAP to the surface of B-CLL cells was assessed by an indirectimmunofluorescence technique with a FACSTAR cell sorter (BectonDickinson, Mountain View, CA), as described previously (8, 9). Briefly,cells were first incubated with OKT1-SAP, washed, and then incubatedwith fluoresceinated goat anti-mouse immunoglobulin reagent (TechnoGenetics, S. Mauro Torinese, Turin, Italy). For binding inhibitionstudies by dual-color fluorescence, B-CLL cells were first incubatedwith OKT1-SAP, washed, and then incubated with phycoerythrin-conjugated Leu-1 (CDS) antibody (Coulter).

B-CLL Cell Proliferation Inhibition Assay. After treatment withOKT1-SAP, washed B-CLL cells were specifically induced to proliferate as recently devised by Caligaris-Cappio et al. (12) with somemodifications. Briefly, 2 x 106/ml B-CLL cells were resuspended in U-bottomed 96-well tissue culture plates (Costar, Cambridge, MA) andcultured in triplicate in the presence or absence (background) of theanti-C3b receptor (CR1) monoclonal antibody CB04 (CD35) covalentlylinked to Sepharose CL-4B (CB04-S; gift from Professor F. Malavasi,Turin, Italy) (13). The culture medium was RPMI 1640 (Gibco) supplemented with 10% fetal calf serum, 10 mM 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid buffer, 100 units/ml penicillin, 100 ^g/ml streptomycin, and 2 mM L-glutamine. Cultures were maintained at37Â°Cin 5% CO2 humidified atmosphere. On day 3 of culture, L-BCGF

(Cellular Products, Inc., Buffalo, NY) at 10% final concentration (v/v)was added to CB04-S-stimulated cultures. On day 6, the synthesis ofDNA was determined by labeling the cultures with 1 /Â¿Ci/well[3H]

thymidine 18 h before harvesting onto glass fiber filters by a Skatronapparatus. The uptake of [3H]thymidine (DNA synthesis) was quanti

fied as cpm by a standard scintillation counting technique. The percentage of DNA synthesis inhibition induced by OKT1-SAP was calculatedasOKT1

-SAP-treated cpm - backgroundcpmControlcpm â€”¿�background cpm

= inhibition of DNA synthesis induced by OKT1-SAP (% ofcontrol)Statistical

analysis of the percentage of DNA synthesis inhibition ofisolated or whole blood OKT1 -SAP-treated B-CLL cells was performedusing the nonparametric Mann-Whitney U test (StatWorks; CricketSoftware, Philadelphia,PA).RESULTSrvtntnxirirv

nf OKT1-SAP Immiinntnvin on R-<~T,I, Tells. In" UPN, unique patient number; Chi, chlorambueil; PDN, prednisone; TNI,

total nodal irradiation; CVP, cyclophosphamide plus vincristine plus prednisone;ND, not determined.

an initial series of experiments (patients 1 to 23) the effect ofOKT1-SAP on resting B-CLL cells was evaluated by incubating

3329




isolated PBMC for 2 h with various immunotoxin concentrations, washing them to remove unbound immunotoxin, andthen activating B-CLL cells by sequential exposure to CB04-Sand L-BCGF. The effect of OKT1-SAP was estimated by measuring suppression of DNA synthesis during subsequent proliferation in response to CB04-S and L-BCGF. While cell viabilityassessed by trypan blue dye exclusion test at the end of a 2-hincubation was unaffected (data not shown), B-CLL cell DNAsynthesis was suppressed in a dose-related manner (Fig. 1,closed symbols). For the two representative patients reported inFig. 1, the OKT1-SAP concentration inhibiting 50% (IC50) ofB-CLL cell DNA synthesis comprised between 4.0 and 6.8 HM.Among 23 consecutive patients with B-CLL, 3 (cases 4, 13, and17) were not Ã©valuablebecause their malignant cells failed toproliferate in response to CB04-S and L-BCGF. For the re

maining 20 cases, the mean inhibition of DNA synthesis induced by 50 nM OKT1-SAP was 85.2% (median, 90.1%; range,14.3-100%) of medium-treated controls (Table 2). Controltreatment of B-CLL cells with a mixture of unconjugated OKT1and SAP had a negligible effect on B-CLL DNA synthesis (Fig.1, open symbols; Fig. 2a).

To determine the activity of OKT1-SAP on unseparated B-CLL cells in whole blood, this immunotoxin was added tounprocessed heparinized peripheral blood from eight B-CLLpatients. After 2 h incubation, the mean inhibition of DNAsynthesis induced by OKT1-SAP was 81.2% (median, 97.5%;range, 0-100%) of medium-treated controls (Table 3). Thisresult was not statistically different from that obtained bytreating isolated B-CLL cells (P = 0.48). In two cases, the effectof OKT1-SAP on isolated and whole blood-treated B-CLL cellswas evaluated in parallel. The DNA synthesis of isolated andwhole blood-treated B-CLL cells of patient 23 was inhibited byOKT1-SAP to 90.2 and 85.6% of controls, respectively. TheDNA synthesis of isolated and whole blood-treated B-CLL cellsof patient 25 was inhibited by OKT1-SAP to 98.8 and 99.8%of controls, respectively. Thus, the cytotoxic activity of OKTl -SAP against B-CLL cells was not hampered by whole bloodcomponents.

120

Ã– 100

Â£oÂ§Â°Â°"in9f 40

i<

O20

-o Gasano-*- Casa 21-â€¢-CasalO-*- CasÂ«21

10-1 10' 10"

OKT1-SAPImmunotoxin(nM)Fig. l. Effect of OKTl -SAP immunotoxin on the DNA synthesis of B-chronic

lymphocytic leukemia cells. B-CLL cells were incubated with either OKT1-SAP(â€¢,A) or unconjugated OKTl and SAP (G. O) for 2 h in the absence of artificialpotentiators, washed, and cultured with CB04-S. On day 3 of culture, L-BCGFwas added to CB04-S stimulated cultures. Values represent median cpm of|'H|thymidine incorporation (DNA synthesis) from triplicate cultures assayed onday 6 after CB04-S stimulation as cpm. The percentage of DNA synthesisinhibition induced by OKTl-SAP was calculated as described in "Materials andMethods."

Table 2 Effect of OKTl-SAP immunotoxin on B-CLL cell proliferation:treatmentof isolatedB-CLL cells

Chronic lymphocytic leukemia cells were incubated with 50 nM OKT1-SAPimmunotoxin for 2 h, washed, and then cultured in the absence (background) orpresence of CB04-S. On day 3, L-BCGF was added to CB04-S-stimulated cultures.Values represent median cpm of [3H]thymidine incorporation (DNA synthesis)from triplicate cultures assayed on day 6 after CB04-S stimulation as cpm. Thepercentage of DNA synthesis inhibition induced by OKT1-SAP was calculated asdescribed in "Materials and Methods." NE, not Ã©valuablebecause of poor DNA

synthesis in medium control cultures.['HJThymidine incorporation (cpm)

Case1234567891011121314IS1617181920212223Background3461631,9122802,3254,8006294297,0633222088,818206233847788,2152056802,1205952994Medium70,4204,66477,64235041,54830,1488,75221,39258,3764,7292,70428,9336931,91831,84569,9006,5367,0656,27052,5454,20137,03411,848OKT1-SAPtreated11,45844566,78027611,57211,9767688,3654,8121,0002154,5305984004,5631,6805,4218201,4604,4804145,5662,147DNA

synthesis inhibitioninduced by OKT1 -SAP

(%)84.193.714.3NE82.771.698.262.110084.699.7100NE90.085.898.6NE91.086.095.391.485.090.2

100

80

M

=â€¢ 40

SI 2Â°

o o

â€¢¿�C80

I .

O 40

20

O

Fig. 2. Competitive inhibition of OKTl-SAP immunotoxin cytotoxicity byunconjugated OKT1 antibody. B-CLL cells from four different patients were firstincubated at 4Â°Cwith either unconjugated 3 JIMOKT1 antibody and SAP (Column

a), irrelevant 3 UM OKT11 antibody (Column b), or unconjugated 3 UM OKT1antibody (Column c); washed; and then treated for 2 h at 37Â°Cwith medium(Column a) or 50 nM OKT1-SAP (Columns b and c). Washed cells were thenassayed for DNA synthesis as described in "Materials and Methods." OKT 11antibody alone did not interfere with B-CLL cell DNA synthesis (data not shown).Results of incubation with 50 nM OKT1-SAP only are reported in Table 1.

Specificity of OKT1-SAP. Competitive binding-inhibitionstudies were also done to demonstrate the specificity of OKTl -SAP. As shown in Fig. 2, preincubation of B-CLL cells withunconjugated 3 /Â¿MOKT1 (CDS) antibody (Fig. 2c) preventedmost or all of the cytotoxic effect of 50 nM OKT 1-SAP. Incontrast, preincubation of the same cells with irrelevant 3 /Â¿MOKT 11 (CD2) antibody of the same IgGl isotype did notprevent the cytotoxicity of OKT 1-SAP against B-CLL cells(Fig. 2b). In addition, dual fluorescence analysis of B-CLL cells

3330




Table 3 Effect ofOKTl-SAPimmunotoxin on B-CLL cell proliferation:treatment of fresh whole B-CLL blood

Fresh unprocessed peripheral blood from B-CLL patients was incubated with50 HM OKT1-SAP immunotoxin at 37'C in 5% CO2 atmosphere. After 2 h,

mononuclear cells were isolated, washed, and then cultured in the absence(background) or presence of CB04-S. On day 3, L-BCGF was added to CB04-S-stimulated cultures. Values represent median cpm of ['I I |th>mÂ¡dineincorporation

(DNA synthesis) from triplicate cultures assayed on day 6 after CB04-S stimulation. The percentage of DNA synthesis inhibition induced by OKT1-SAP wascalculated as described in "Materials and Methods."

[3H]Thymidine incorporation (cpm)

Case24

25262728293031Background2,750

3,61569

226333494

5,50050Medium13,731

21,70047,337

5,7111,4948,713

24,21120,589OKTl-SAP

treated5,616

3,82453,604

36846

2911,5571,392DNA

synthesis inhibitioninduced byOKTl-SAP73.7

98.90

97.588.897.6

10093.5

(patient 10) preincubated with OKTl-SAP showed that thesaturation of CDS cell surface determinants by the antibodymoiety of this immunotoxin completely prevented the subsequent binding of phycoerythrin-conjugated Leu-1 (CDS) antibody (data not shown). Taken together these data demonstratethat the anti-B-CLL effect of OKTl-SAP is mediated by specificbinding of this immunotoxin to target cell surface CDS antigen.

DISCUSSION

In this article we demonstrate that the CDS-restricted immunotoxin OKTl-SAP can specifically inhibit in vitro theproliferation of CDS-expressing B-CLL cells. The experimentalconditions under which the inhibition of leukemia cell proliferation is accomplished imply brief incubation of either isolatedor whole blood B-CLL cells with 10-100 nM OKTl-SAP atphysiological pH and in the absence of artificial potentiators.Taking into consideration our previous toxicology and phar-macokinetic studies in primates which demonstrated that theseconditions are achievable in vivo after a single nontoxic i.v.OKTl-SAP injection (8), it is reasonable to speculate that thisimmunotoxin may prove clinically active in vivo as it does invitro.

Present clinical trials usually rely on preclinical studies inwhich the efficacy of immunotoxins is evaluated against in vitroestablished tumor cell lines. However, the heterogeneity of theresponse of different cell lines sharing the same target antigento a given immunotoxin strongly suggests an individual susceptibility of malignant cells to several possible factors such as themean density of target antigen, the proportion of cells that donot express the antigen on cell membrane, and metabolic stateof those cells (1,14,15). Thus, one would expect that in clinicalconditions the sensitivity of leukemia cells to immunotoxinswould vary from one patient to another. Predictive determination of the sensitivity of malignant cells to immunotoxins iswarranted for the selection of patients who might benefit fromtreatment with these reagents. For B-CLL the standardizedassay for the measurement of protein synthesis in fresh leukemia cells is not the most appropriate because of the low levelof protein synthesis of these cells. We therefore evaluated thecytotoxic activity of OKTl-SAP on fresh leukemia cells usinga proliferation inhibition assay specifically devised for B-CLLcells (12). In this assay malignant B-CLL cells are treated withOKTl-SAP, washed to remove unbound immunotoxin, andthen activated by binding their surface complement C3b receptor (CR1) to an insolubili/ed anti-CRl monoclonal antibody

(CD35) referred to as CB04-S (14). The binding of CB04-Sprovides a progression signal to B-CLL cells which then becomeresponsive to L-BCGF and actively proliferate in the presence

of this cytokine. After 6 days of culture, the measurement ofDNA synthesis in immunotoxin-treated cells in comparison tomedium controls gives an estimate of the cytotoxic potential ofthe tested immunotoxin.

In our series, only 3 of 31 consecutive B-CLL patients couldnot be evaluated by the proliferation inhibition assay becausetheir cells failed to proliferate in response to in vitro stimulation.Among the remaining 28 patients, regardless of the concurrentantileukemic therapy or stage of disease, 26 exhibited a strongin vitro sensitivity to OKTl-SAP, and 2 (cases 3 and 26) wereresistant to this immunotoxin. In comparison to sensitive patients, the latter two did not possess any peculiar clinical orimmunological feature that could allow the prediction of resistant cases. As immunofluorescence studies demonstrated that inboth cases OKTl-SAP bound to B-CLL cells, it is likely thatthe immunotoxin resistance depends on a singular mechanismof internalization or of intracellular processing rather than ona surface binding defect.

The experimental data presented here favorably contrast withthose of Hertler et al. (4) who studied the sensitivity of B-CLLto T101-RTA, another anti-CDS immunotoxin containing ricinA-chain instead of SAP as the toxic moiety. These authors,using a tritiated leucine incorporation assay, found that B-CLLcells in vitro as well as in vivo were primarily resistant to T101-RTA. Accordingly, they could not demonstrate any sustainedclinical benefit of i.v. administered T101-RTA in four B-CLLpatients despite complete immunotoxin saturation of circulating target leukemia cells (4). The doses of T101-RTA used byHertler et al. (4) were relatively small and might not have beenable to saturate B-CLL cells in lymph nodes and other tissuesites. In spite of this possibility, the different cytotoxic activityin vitro of OKTl-SAP in comparison to T101-RTA, i.e., immunotoxins with the same target cell specificity but containingdifferent toxins, remains to be elucidated. This difference islikely to reflect the biophysical properties of the toxic and/orligand component of the immunotoxin molecule. The isotypeof the antibody moiety of OKTl-SAP (IgGl) differs from thatof T101-RTA (IgG2) (4, 6, 8). However, when the IgG2 antibody T101 is conjugated with SAP, the resulting T101-SAPimmunotoxin exhibits the same characteristics of cytotoxicityagainst human T-lymphocytes as OKTl-SAP does (8). Thesedata indicate that for the above CDS-specific immunotoxinsthe toxic component, more than the ligand, may affect theircytotoxicity. At the molecular level evaluated on isolated liverribosomes, ricin A-chain and SAP share the same mechanismof action consisting in the inactivation of eukaryotic ribosomesby modifying both or either of two nucleoside residues in 28SrRNA (16). In contrast, unlike ricin A-chain, SAP has anisoelectric point >10, is extremely resistant to physical andproteolytic inactivation, and does not contain carbohydrateresidues (8, 9). How these biophysical differences may affectthe ligand-antibody-mediated ribosome inactivation of toxinsagainst eukaryotic cells is beyond the purpose of this study.However, the observation of Hertler et al. (4) that the resistanceof B-CLL cells to T101-RTA can be reversed in vitro by thelysosomotropic amine ammonium chloride suggests again thatthe modalities of intracellular processing probably peculiar foreach toxin and cell type are the crucial events determining thepreservation or abolition of the ribosome-inactivating activity.

We conclude that preclinical data presented in this articletogether with the previously reported OKTl-SAP toxicology

3331




and pharmacokinetics studies (8) warrant clinical phase I andII trials aimed at defining: (a) the potential advantages of usingOKT1-SAP immunotoxin for treatment of refractory B-CLLpatients; and (/>)the value of the B-CLL proliferation inhibitionassay described here to predict the clinical response of B-CLLto an immunotoxin administered /// vivo.

ACKNOWLEDGMENTS

We are grateful to Fabrizio Marcucci, Marco Soria, and DomenicoTrizio for their contributions to the immunoconjugates project; FabioMalavasi for providing CB04-S antibody; Bernd Dorken for providingHD6 and HD37 antibodies; Federico Caligaris-Cappio for interestingdiscussion; and Anna Basile for coordinating the supply of bloodspecimens.

REFERENCES

1. Vitella, E. S., Fullon, R. J., May, R. D., Till, M., and Uhr, J. W. Redesigningnature's poisons lo create anti-tumor reagents. Science (Wash. DC), 238:1098-1104,1987.

2. Byers, V., Henslee, J., Antin, J., Kernan, V. Blazar, B.. Gingrich, R., Phillips,Ct., O'Reilly, R., and Scannen, P. Safety and therapeutic response to a pan-T-lymphocyte monoclonal antibody ricin A chain immunoloxin in graftversus host disease (GVHD). Proc. Am. Soc. Clin. Oncol., 7:190, 1988.

3. Laurent, G., Pris, J., Parcel, J. P., Carayon, P., Blythman, H., Casellas, P.,Poncelet, P., and Jansen, F. K. Effects of therapy with T101 ricin A-chainimmunoloxin in two leukemia patients. Blood, 67: 1680-1687, 1986.

4. Hertler, A. A., Schlossman, D. M., Borowitz, M. J., Laurent, G., Jansen, F.K., Schmidt, ( '.. and Frankel, A. E. A phase I study of T101-ricin A chain

immunotoxin in refractory chronic lymphocytic leukemia. J. Biol. ResponseModif., 7:97-113, 1988.

5. Casellas, P., and Jansen, F. K. Enhancement of A-chain immunotoxins bypharmacological agenls. In: C. W. Vogel (ed.), Immunoconjugales. AntibodyConjugates in Radioimaging and Therapy of Cancer, pp. 97-1 IS. New York:Oxford University Press, 1987.

6. Siena, S., Villa, S., Bregni, M., Bonadonna, G., and Gianni, A. M. Amantadme potentiates T-lymphocyte killing by an anti-pan-T cell (CDS) ricin A-chain immunotoxin. Blood, 69:345-348, 1987.

7. Bourne, B. J. P., Casellas, P., Blythman, H. E., and Jansen, F. K. Study ofthe plasma clearance of antibody-ricin-A chain immunoloxins. Evidence forspecific recognition siles on the A-chain lhat mediates rapid clearance of theimmunotoxin. Eur. J. Biochem., 755:1-10, 1986.

8. Siena, S., Lappi, D. A., Bregni, M., Formosa, A., Villa, S., Soria, M.,Bonadonna, G., and Gianni, A. M. Synthesis and characterizalion of anantihuman T-lymphocyte saporin immunotoxin (UKTI SAP) with in vivostability into nonhuman primates. Blood, 72: 756-765, 1988.

9. Siena, S., Bregni, M., Formosa, A., Martineau, D., Lappi, D. A., Bonadonna,G., and Gianni, A. M. Evaluation of antihuman T-lymphocyte saporinimmunotoxins potentially useful in human transplantation. Transplantation(Baltimore), 46: 747-753, 1988.

10. Thorpe, P. E., Brown, A. N., Bremmer, J. A. G., Jr., Foxwell, B. M. J., andStirpe, F. An immunotoxin composed of monoclonal anlibody anli Thy 1.1antibody and a ribosome-inaclivating protein from Saponaria officinali!:potent antitumor effects in vitro and in vivo. J. Nail. Cancer Insl., 75: 151-159, 1985.

11. Royston, I., Majda, J. A., Baird, S. M., Mcscrve, B. I,., and Griffins, J. C.Human T cell antigens defined by monoclonal antibodies: The 65,000 daltonantigen on T-cclls (T65) is also found on chronic lymphocyte cells bearingsurface immunoglobulin. J. Inimunol.. 125: 725-731,1980.

12. Caligaris-Cappio, F., Schena, M., Bergui, I,., Tesio, I.., Riva, M., Rege-Cambrin, G., Funaro, A., and Malavasi, F. C3b receptors mediate the growthfactor-induced proliferation of malignant B-chronic lymphocytic leukemialymphocytes. Leukemia, 1: 746-752, 1987.

13. Malavasi, F., Funaro, A., Bellone, G., Caligaris-Cappio, F., Berti, E., Tetta,C., Delia-Bona, P., de Maria, S., Campogrande, M., and Cappa, A. P. M.Functional and molecular characterization by the CB04 monoclonal antibodyof a cell surface structure exerting O complement receptor activity. J. Clin.Immunol., 5: 412-420, 1985.

14. Manske, J. M., Buchsbaum, D. J., Azemove, S. M., Hanna, D. E., andVallera, D. A. Antigenic modulation by anti-CD5 immunotoxins. J. Immunol., 136:4721-4728, 1986.

15. Leonard, J. E., Johnson, D. E., Shawler, D. I.., and Dillman, R. O. Inhibitionof human T-cell tumor growth by '1101-ricin A-chain in an athymic mouse

model. Cancer Res., 48:4862-4867, 1988.16. Stirpe, F., Bailey, S., Miller, S. P., and Bodley, J. W. Modification of

ribosomal RNA by ribosome-inactivating proteins from plants. NucÃ.AcidsRes., 16: 1349-1357, 1988.

3332



1989;49:3328-3332. Cancer Res Salvatore Siena, Marco Bregni, Anna Formosa, et al. Leukemia Cell Proliferation in HumansImmunotoxin-mediated Inhibition of Chronic Lymphocytic

Updated version

http://cancerres.aacrjournals.org/content/49/12/3328

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/49/12/3328To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



immunotoxin-mediated inhibition of chronic lymphocytic leukemia

Documents