CANCER BIOTHERAPY & RADIOPHARMACEUTICALSVolume 14, Number 2, 1999Mary Ann Liebert, Inc.

Preclinical Evaluation of Chimeric L6 Antibody forthe Treatment of Kaposi's Sarcoma withRadioimmunotherapy

Bryan R. Leigh,1 Patricia A. Burke,1 Angela M. Hong,1 Robert T. O'Donnell,2 Lydia P. Howell,3Laird A. Miers,2 Gerald L. DeNardo2 and Sally J. DeNardo2'Department of Radiation Oncology, department of Internal Medicine, Division of Hematology/Oncology,Section of Radiodiagnosis and Therapy, department of Pathology, University of California, Davis,Medical Center, Sacramento, CA

L6 is a murine IgG2a monoclonal antibody with panadenocarcinoma reactivity. Chimeric L6 (ChL6), thevariable region ofmurine L6 combined with a human IgGl constant region, has been used in clinical trialsfor the delivery ofradioimmunotherapy to patients with breast cancer. AIDS-associated Kaposi 's sarcoma

(KS), a malignancy ofvascular endothelium, may be an excellent candidatefor systemic radioimmunotherapybecause KS is well vascularized and radioresponsive. Because ChL6 has been noted to bind vascularendothelium, our hypothesis was that ChL6 will recognize and bind KS tumors making this a potentially usefulantibodyfor the treatment ofKS with radioimmunotherapy. To test this hypothesis, 4 human KS spindle cellcultures establishedfrom cutaneous punch biopsy specimens (KS-MR, KS-NO, KS^JD and KS 6-3E) and one

well-characterized human KS cell line (KS Y-l) were assessedfor L6 immunoreactivity. All 5 cell cultureswere L6positive by immunohistochemistry. KS Y-l cells grown as nude mouse xenografis were alsp L6positiveby immunohistochemistry. Competitive binding assays performed on the KS Y-l and KS 6-3E cell culturesshowed high density and high affinity cell binding. Biodistrubition experiments performed on nude mice withKS Y-l xenografts demonstrate tumor targeting by ChL6. These findings indicate that ChL6 may be a usefulantibodyfor the radioimmunotherapy ofKS. Future experiements will assess the therapeutic efficacy ofradiolabeled ChL6 with and without concurrent systemic radiosensitizing chemotherapy.

INTRODUCTION

Kaposi's sarcoma (KS) is the most common cancer

associated with the acquired immunodeficiencysyndrome (AIDS). AIDS-associated KS is a malig¬nancy of vascular endothelium characterized byAddress reprint requests to Bryan R. Leigh, M.D., Depart¬ment ofRadiation Oncology, UC Davis Cancer Center, 4501XStreet, MG-126, Sacramento, CA 95817 Phone: (916)-734-8252. Fax: (916)-454-4614. E-maiL· bryan.leighfa ucdmc.ucavis.edu.

multifocal disfiguring skin lesions and life-threaten¬ing visceral tumors. Although isolated KS tumorsare treated effectively with local treatments includinglow dose external beam radiation therapy,1 more

effective and less toxic systemic treatment is neededfor disseminated KS. Currently, the standardtreatment for advanced KS is systemic chemotherapywhich suppresses bone marrow function and contrib¬utes to a decline of the immune system in patientswho are already immunocompromised.2

113

Radioimmunotherapy, a novel cancer treatment,delivers radiation therapy to tumors by way ofsystemically administered radiolabeled monoclonalantibodies that recognize tumor-associated antigens.Antibodies have been reported to contribute to tumorcell killing through complement dependent andantibody dependent cellular cytotoxicity,3 but theantitumor effect of radioimmunotherapy is primarilydue to the radiation from the radiolabeled antibody.Important determinants of radioimmunotherapyeffectiveness include the affinity and specificity ofthe antibody, the targeted antigen density, and thevascularity and radiosensitivity ofthe targeted tumor.KS is a highly vascular and markedly radio-responsive tumor making it an excellent candidatefor studies of radioimmunotherapy.

L6 is a mouse IgG2a monoclonal antibodyoriginally raised against a human lung adeno-carcinoma.4 L6 targets a 202 amino acid hydropho-bic, cysteine-rich polypeptide related to a family ofintegral membrane proteins controlling cellularproliferation.5 The same antigen target is found on

human adenocarcinomas of the lung, breast, colon,prostate and ovary. Chimeric L6 (ChL6), the vari¬able region of mouse L6 combined with a humanIgGl constant region, has been used successfully inclinical trials for the delivery of systemicradioimmunotherapy to patients with breast cancer.6"8ChL6 has been noted to bind normal vascularendothelium. However, 150-200 mg ofChL6 coversvascular targets allowing a subsequent injection ofradiolabeled ChL6 to localize in extravascular tumortissue. Because KS is a tumor ofvascular endothe¬lium, we were prompted to assess the potential ofChL6 for the radioimmunotherapy of KS. Thefollowing in vitro and in vivo experiments suggestthat ChL6 may be an excellent antibody for thedelivery of targeted systemic radiotherapy to treatadvanced KS.

MATERIALS AND METHODS

Cell Cultures

Primary KS cell cultures (KS-JD, KS-NO and KS-MR) were derived from punch biopsies of cutaneousKS tumors belonging to UC Davis AIDS OncologyClinic patients. Informed consent was obtained fromall KS tissue donors prior to biopsies in accordance

with the UC Davis Medical Center Human SubjectsReview Committee. Biopsies were finely minced andplated on 2% gelatin coated flasks in RPMI 1640(Gioco, New York, NY) with 15% fetal calf serum

(FCS), 4 mM L-glutamine, essential and nonessentialamino acids and 1% Nutridoma HU (BoehringerMannheim, Indianapolis, IN). After 3 weeks,colonies of spindle shaped cells appeared. Thecolonies were trypsinized and passed in the samemedium to gelatin coated flasks. KS 6-3E spindlecells were grown under the same conditions as KS-JD, KS-NO and KS-MR. KS Y-l is a well charac¬terized human KS cell line capable of growingxenografts in nude mice.9 KS Y-l cells were grownin RPMI 1640 with 5% FCS. Both KS Y-l and KS6-3E were kindly provided by the laboratory of Dr.Parkash Gill (University of Southern California, LosAngeles, CA).

Antibodies

The murine monoclonal antibody L6 recognizes a

cell surface antigen highly expressed on lung, breast,colon, and ovarian carcinomas and human vascularendothelial cells.4 L6 produces antitumor effects invitro by antibody-dependent cellular cytotoxicity andcomplement-dependent cytotoxicity.10 Chimeric L6(ChL6; Bristol-Myers Squibb PharmaceuticalResearch Institute, Seattle) is an antibody chimeraconsisting ofa human IgGl constant region and thevariable region of mouse L6.11

ImmunohistochemistryImmunohistochemical staining was performed as

previously described.12 Cell culture specimens were

grown on glass microscope slides. Mouse xenograftswere embedded in OCT frozen section gel, snap-frozen in liquid nitrogen, and cut into 4 micrometersections. After fixing for 5 minutes in acetone,samples were stained with a 1:200 dilution of L6 or

a 1:100 dilution of 155 (negative control, murineIgG2a anti-mucin antibody) by a routine avidin-biotin immunoperoxidase staining method. Cellcultures were stained using reagents from VectorLaboratories (Burlingame, CA) with diamino-benzidine as a substrate. Nude mouse xenograftswere stained with reagents from Innovex (OverlandPark, KS) using the Stat-Q kit with "background

114

buster" to decrease background staining. Theprocedure as suggested by the manufacturers was

modified only by the application of the 0.003%hydrogen peroxide and methanol blocking step afterthe addition ofthe secondary antibody, since it was

found that the routine method inhibited the L6staining reaction. A hematoxylin counterstain wasalso performed.Radioiodination

ChL6 was labeled with 1-125 using the cMoramine-Tmethod as previously described.13,14 The molar ratioof 1-125 to antibody and the mass ratio ofcUoramine-T to antibody were 1:10. Free 1-125 was

separated from antibody bound 1-125 by using a

Sephadex G-50-80 column (Sigma Chemical Co.,St. Louis, MO). HPLC of radioiodinated ChL6confirmed that the preparation contained greater than90%monomeric antibody. No aggregates, denaturedantibody, or free iodine was detected.

Quantitative Immunoreactivity Assay

Immunoreactivity of I-125-ChL6 was assessed bya cell binding radioimmunoreactive assay as previ¬ously described.1516 I-125-ChL6 (10 ng) was addedto tubes œntaining increasing amounts of unlabeledChL6 (0-1000 ng) to compete for the antigenic sites.1.0 106 KS 6-3E cells or 1.7 IO6 KS Y-l cellswere added to each tube for a 30 minute incubationat room temperature. Individual tubes were countedfor total radioactivity in a gamma well counter, thencentrifuged at 800 g for 5 minutes and the super¬natant carefully removed without disturbing the cellpellet. The radioactivity ofthe resulting pellets andsupernatants were counted and the percent of 1-125-ChL6 binding to the cells was calculated. Allsamples were performed in triplicate. The bindingaffinity constant and number of antibodies bindingper cell were calculated by means of Scatchardanalysis.17 Raji lymphoma cells were used as a

negative control and HBT3477 breast cancer cellswere used as a positive control.

Mouse Biodistribution

Female athymic BALB/c nu/nu mice (HarlanSprague-Dawley), 6-10 weeks of age, were main¬tained according to University of California animal

care guidelines on a normal diet ad libitum and underpathogen free conditions. KS Y-l cells were har¬vested in logarithmic growth phase and 3.5 IO6cells were injected subcutaneously into both ventralflanks ofeach mouse. When the tumors reached 50-300 mg (12 days), I-125-ChL6 (25pCi/100pg/ µ ) was injected intravenously into the tail veinof each mouse. The radioactive dose injected intoeach mouse was measured with a Capintec CRC-12radionuclide dose calibrator (Capintec, Inc., Pitts-burg, PA) and confirmed by counting the mouse

immediately after injection using two opposedsodium iodide detectors (Picker Nuclear, WhitePlains, NY) calibrated against appropriate standardsfor radioiodine, volume and geometry.

To determine the whole-body clearance of1-125,each mouse was counted immediately, 4 hr and 1,2,3,4 and 5 days after injection usng the dual-probesystem. The counts were decay-corrected andexpressed as percent injected dose (%JD). The bloodclearance was determined by collecting 2µ1 bloodsamples from the tail vein of each mouse at 5 min,1 hr, and 1, 2, 3, 4 and 5 days after injection andcounting the samples in a sodium iodide gamma wellcounter (Pharmacia LKB Nuclear, Inc., Gaithers-burg, MD). Decay-corrected radioactivity in theblood was expressed as %ID/ml.

Pharmocokinetic data were obtained by sacrific¬ing 3 groups of 5 mice with 2 tumors each at 24, 72,and 120 hours after injection, removing and weighingthe tumors and organs and counting them in the same

gamma well counter. The uptake for the tumor andeach organ at each time point was calculated andexpressed as %ID/g +/- standard deviation.

RESULTS

ImmunohistochemistryAll five KS spindle cell cultures demonstrated similarstaining patterns for L6 by immunohistochemistry

Antibody uptake appeared to involve the KScytoplasm and cell membrane (Figure 1). Frozensections ofKS Y-l xenografts from nude mice alsostained for L6 with similar pattern and intensity.

Quantitative Immunoreactivity AssayThe Scatchard analysis revealed that KS 6-3E cellsbind 2.2 105 antibodies per cell with an affinity

115

Figure 1. Immunohisto-chemical analysis of KS Y-lcells shows (A) heavy mem¬brane and cytoplasmic stainingwith the L6 antibody and (B)little or no staining with the155 (negative control) anti¬body (magnification 325x).All cell cultures and xenograftsstudied showed a similar stain¬ing pattern and intensity.

Figure 2. Biodistribution of l"l-ChL6 in KS Y-l tumor-bearing nudemice at 24, 72, and 120 hrs (mean+/- standard deviation). Five miceand 10 tumors were studied at eachtime point. Tumor levels increasedbetween 24 and 72 hours after injec¬tion, while organ levels decreased ata rate similar to the blood clearance.

Tumor Liver Lung Muscle Intestine Kidney Spleen Heart Stomach Bone

Organ

coefficient (Ka) of2.2 IO8, and KS Y-l cells bind1.6 104 antibodies per cell with a Ka of 2.9 IO8.

The whole body and blood clearances are shown inFigure 3.

Mouse Biodistribution DISCUSSION

The results ofthe I-125-ChL6 biodistribution studiesincluding tumor and organ uptake at 1, 3 and 5 dayspostinjection are summarized in Figure 2. Tumoruptake ofI-125-ChL6 was higher than normal organsat all time points. The percent injected dose pergram oftumor was 9.4% +/- 3.2% at 24 hrs, 10.0%+/-1.5% at 72 hrs, and 6.0% +/- 2.3% at 120 hrs.

AIDS-associated KS is a significant clinical chal¬lenge. Current treatment with external beam radia¬tion therapy, chemotherapy, and biological responsemodifiers have been moderately successful, but poortolerance and limited efficacy remain significantproblems. Novel effective treatments with lesstoxicity could benefit large numbers ofpatients with

116

100

oQ V

- -24 48

- -72 96

BodyBlood

120

Time (hrs)

Figure 3. Whole body clearance and blood clearance graphs showeda monoexponential and biexponential decline in 125I-ChL6 levels,respectively, in KS Y-l tumored nude mice (mean +/- standard deviation).

KS. Because KS is a highly vascular multifocaltumor which is markedly radioresponsive,radioimmunotherapy may be an excellent treatmentoption. In fact, the complete response rate to only20 Gy external beam radiation therapy was 79% inone study.18

In the current study, a human/mouse chimericantibody, ChL6, was assessed for its ability to targetKS as a therapeutic antibody. First, the L6 antigenwas determined to be present at a high density on

the KS cell surface by immunohistochemistry andScatchard analysis. Then, biodistribution studiesconfirmed the ability ofradiolabeled ChL6 to localizein KS xenografts, proving in vivo immunoreactivity.The I-125-ChL6 bicdistribution results for KS Y-lxenografts in nude mice are similar to those reportedfor other I-125-monoclonal antibodies targeting nudemouse xenografts19,20 including ChL6 and humanbreast tumor xenografts.14 I-125-ChL6 graduallyclears from the blood pool while tumor uptakeremains stable for up to 72 hours.

L6, ChL6, 1-131-L6 and I-131-ChL6 haveproven to be safe and well tolerated in phase Iclinical trials.6,8,21·22 Reported toxicities included thefrequent occurrence of a self-limited vascular leaksyndrome and infrequent mild headaches, urticariaand fever. Unlabeled L6, with or without concurrentinterleukin 2, has induced partial and completeresponses in patients with advanced colorectal and

breast cancer.21,22 1-13 l-ChL6 radioimmunotherapyhas induced therapeutic responses in patients withchemotherapy-refractory metastatic breast cancer.6"8,23 Our results suggest that ChL6 immunotherapy,with or without radioimmunotherapy, may also beuseful for the treatment of KS.

Paclitaxel, a cytotoxic chemotherapy agent withactivity against a wide range ofmalignancies includ¬ing KS, has shown promise for improving theantitumor effects of radioimmunotherapy. Pachtaxelstabilizes microtubules resulting in mitotic block andactivation ofapoptosis.24 Phase clinical trials haveshown that paclitaxel produces a 71% response ratein patients with advanced AIDS-associated KS.25The National Cancer Institute is currently sponsoringa phase III clinical trial comparing paclitaxel toliposomal doxorubicin as primary treatment forpatients with advanced AIDS-associated KS.

Preclinical research in our laboratory hasdemonstrated that paclitaxel can dramaticallyenhance the therapeutic efficacy of radiolabeledChL6 against breast cancer xenografts.26 Theenhancement was striking when subtherapeutic dosesofpaclitaxel (300-600 pg/mouse) were administered24 hrs after radiolabeled ChL6 resulting in a 100%response rate and a 88% cure rate.27 Clinical trialsare in progress to determine if the synergistic effectofpaclitaxel and radioimmunotherapy will benefitpatients with adenocarcinoma. Future studies willmeasure the antitumor effect ofChL6 radioimmuno¬therapy against KS xenografts with and withoutpaclitaxel. These experiments will help determineifour success against breast cancer can be appliedto other tumors including KS.

ACKNOWLEDGMENTS

The authors wish to express their appreciation to Dr.Parkash Gill for providing KS cell lines and advice.We also thank Sharon Malle, David Kukis, GaryMirick, and Linda Kroger for assistance with thismanuscript. This research was supported by an

Institutional Research Grant from the AmericanCancer Society and grants from the American CancerSociety (Institutional Research Grant, UC Davis)and the National Cancer Institute (PHS CAI6861andP01CA47829).

117

REFERENCES

1. Berson A, Quivey J, Harris J, and Wara W. Radiationtherapy for AïïDS-related Kaposi's sarcoma. Interna¬tionalJournal ofRadiation Oncology, Biology, Physics1990;19:569-575.

2. Morris A and Valley A. Overview of the managementof AIDS-related Kaposi's sarcoma. Annals ofPharmacotherapy 1996;30:1150-1163.

3. Hellstrom I, Beaumier P, and Hellstrom . Antitumoreffects ofL6, an IgG2a antibody that reacts with mosthuman carcinomas. Proceedings of the NationalAcademy ofSciences 1996;83:7059-7063.

4. Hellstrom I, Horn D, Linsley P, Brown J, BrankovanV, and Hellstrom . Monoclonal mouse antibodiesraised against human lung carcinoma. Cancer Re¬search 1986;46:3917-3923.

5. Marken J, Bajorath J, Edward C, Fair A, Schieven G,Hellstrom I, Hellstrom , and Aruffo A. Membranetopology of the L6 antigen and identification of theprotein epitope recognized by the L6 monoclonalantibody. Journal of Biological Chemistry1994;269:7397-7401.

6. DeNardo S, O'Grady L, Richman C, Goldstein D,O'Donnell R, DeNardo D, Kroger L, Lambom K,Hellstrom , Hellstrom I, and DeNardo G.Radioimmunotherapy for advanced breast cancer usingI-131-ChL6 antibody. Anticancer Research1997;17:1745-1751.

7. DeNardo S, Richman C, Goldstein D, Shen S, SalakoQ, Kukis D, Meares C, Yuan A, Welbom J, andDeNardo G. Yttrium-90/indium-l 11-DOTA-peptide-chimeric L6: pharmacokinetics, dosimetry and initialresults in patients with incurable breast cancer.

Anticancer Research 1997;17:1735-1744.8. DeNardo S, Mirick G, Kroger L, O'Grady L, Erickson

K, Yuan A Lambom K, Hellstrom I, Hellstrom , andDeNardo G The biologic window for chimeric L6radioimmunotherapy. Cancer 1994;73:1023-1032.

9. Lunardi-Iskandar Y, Gill P, Lam V, Zeman R, Mi¬chaels F, Mann D, Reitz M, Kaplan M, Berneman Z,Carter D, Bryant J, and Gallo R. Isolation and charac¬terization ofan immortal neoplastic cell line (KS Y-l)from AIDS-associated Kaposi's sarcoma. Journal ofthe National Cancer Institute 1995;87:974-981.

10. Hellstrom I, Garrigues U, Lavie E, and Hellstrom .Antibody-mediated killing of human tumor cells byattached effector cells. Cancer Research 1988;48:624-627.

11. Liu A, Robinson R, Hellstrom K, Murray E, Chang C,and Hellstrom I. Chimeric mouse-human IgG 1 antibodythat can mediate lysis of cancer cells. Proceedings ofthe NationalAcademy ofSciences 1987;84:3439-3443.

12. Howell L, DeNardo S, Levy N, Lund J, and DeNardoG Immunohistochemical staining of metastatic ductalcarcinomas ofthe breast by monoclonal antibodies used

in imaging and therapy: A comparative study. TheInternational Journal of Biological Markers1999;10:129-135.

13. Hunter Wand Greenwood F. Preparation of iodine-131labelled human growth hormone of high specificactivity. Nature 1962;194:495-496.

14. Adams G, DeNardo S, Amin A, Kroger L, DeNardoG, Hellstrom I, and Hellstrom K. Comparison of thepharmacokinetics in mice and the biological activityof murine L6 and human-mouse chimeric ch-L6antibody. Antibody, Immunoconjugates, andRadiopharmaceuticals 1992;5:81-95.

15. Wofsy C and Goldstein B. Interpretation of Scatchardplots for aggregating receptor systems. MathematicalBiosciences 1992;112:115-154.

16. Beaumier P, Neuzil D, Yang H, Noll E, Kishore R,Eary J, Krohn K, Nelp W, Hellstrom and HellstromI. Immunoreactivity assay for labeled anti-melanomamonoclonal antibodies. Journal ofNuclearMedicine199827:824-828.

17. Scatchard G. The attraction of proteins for smallmolecules and ions. Annals ofthe New York AcademyofSciences 1949;51:660-672.

18. Stelzer and Griffin T. A randomized prospective trialof radiation therapy for AIDS-associated Kaposi'ssarcoma. International Journal ofRadiation Oncol¬ogy, Biology, Physics 1993;27:1057-1061.

19. Hosono M Endo Hosono M Kobayashi H, ShiratoM, Sakahara H, Ueda R, and Konishi J. Treatment ofsmall-cell lung cancer xenografts with iodine-131 -anti-neural cell adhesion molecule monoclonal antibody andevaluation of absorbed dose in tissue. Journal ofNuclear Medicine 1994;35:296-300.

20. King D, Turner A, Famsworth A, Adair J, Owens R,Pedley R Baldock D, Proudfoot K, Lawson A, BeeleyN, Mllar Millican T, Boyce B, Aritoniw P, Moun¬tain A, Begent R, Shochat D, and Yarranton G. Im¬proved tumor targeting with chemically cross-linkedrecombinant antibody fragments. Cancer Research1994;54:6176-6185.

21. Goodman G, Hellstrom I, Brodzinsky L, Nicaise C,Kulander B, Hummel D, and Hellstrom K. Phase I trialof murine monoclonal antibody L6 in breast, colon,ovarian, and lung cancer. Journal ofClinical Oncol¬ogy 1990;8:1083-Í092.

22. Ziegler L, Palazzolo P, Cunningham J, Janus M, ItohK, Hayakawa K, Hellstrom I, Hellstrom K, Nicaise C,Dennin R, and Murray J. Phase I trial of murinemonoclonal antibody L6 in combination with subcuta¬neous interleukin-2 in patients with advanced carci¬noma of the breast, colorectum, and lung. Journal ofClinical Oncology 1992;10:1470-1478.

23. DeNardo S, O'Grady L, Richman C, and DeNardo G.Overview of radioimmunotherapy in advanced breastcancer using 1-131 chimeric L6. In: L. Ceriani (ed.),Antigen and Antibody Molecular Engineering in BreastCancer Diagnosis, pp. 203-211, New York: PlenumPress. 1994.

118

24. Wahl A, Donaldson , Fairchild C, Lee F, Foster S, 26.Demers G, and Galloway D. Loss of normal p53function confers sensitization to Taxol by increasingG2/M arrest and apoptosis. Nature Medicine1996;2:72-79.

25. Welles L, Seville MW, Lietzau J, Pluda JM, WyvillKM Fuerstein I, Figg WD, Lush R, Odom, J, Wilson 27.W, Fajardo, MT, Humphrey RW, Feigal E, Tuck D,Steinberg SM Broder S, and Yarchoan R. Phase trialwith dose titration of paclitaxel for the therapy ofhuman immunodeficiency virus-associated Kaposi'ssarcoma. JournalofClinicalOncology 1998;16:1112-1121.

DeNardo S, Kukis D, Kroger L, ODonnell R, LambomK, Miers L, DeNardo D, Meares C, and DeNardo G.Synergy of Taxol and radioimmunotherapy withyttrium-90-labeled chimeric L6 antibody: efficacy andtoxicity in breast cancer xenografts. Proceedings ofthe NationalAcademy ofSciences 1997;94:4000-4004.DeNardo S, Richman C, Kukis D, Shen S, LambomK, Miers L, Kroger L, Perez E, and DeNardo G.Synergistic therapy ofbreast cancer with -90-chimericL6 and paclitaxel in the xenografted mouse model:development of a clinical protocol. Anticancer Re·search 1999;18:4011-4018.

119