Successful Treatment of Canine Malignant Histiocytosis with the Human Major Histocompatibility Complex Nonrestricted Cytotoxic T-Cell Line TALL-1041

Vol. 3, 1789-1 797. October 1997 Clinical Cancer Research 1789

Sophie Visonneau, Alessandra Cesano, effectors and their therapeutic potential even in the most Thuy Tran, K. Ann Jeglum, and Daniela Santoli2 aggressive forms of the disease. The Wistar Institute, Philadelphia, Pennsylvania 19104 [S. V., A. C.. T. T., D. S.], and Veterinary Oncology Services and Research Center, INTRODUCTION West Chester, Pennsylvania 19382 [K. A. J.] MH3 in dogs, first reported in 1978 ( I ), is a tumor char- actenized by neoplastic proliferation of invasive atypical eryth- nophagocytic histiocytes in various tissues. The disease fre- ABSTRACT quently becomes manifest in the middle-age years and has been The human MHC nonrestricted cytotoxic T-cell line observed more frequently in males than in females (2). Bernese TALL-104 exerts potent antitumor effects in animal models mountain dogs are genetically prone to this type ofcancer (3, 4), with both induced and spontaneous cancers. The present but other breeds are also sporadically affected (5). Clinical report documents the ability of systemically delivered findings commonly include fever, generalized lymphoadenopa- TALL-104 cells to induce durable clinical remissions in four thy, and hepatosplenomegaly as well as concomitant anemia, of four dogs with malignant histiocytosis (MH). The animals leukopenia, and thrombocytopenia (I). Neoplastic histiocytes received multiple i.v injections oflethally irradiated (40 Gy) mainly infiltrate the spleen, liver, lungs, lymph nodes, bone TALL-iN cells at a dose of 108 cells/kg, with (two dogs) or marrow, and skin. The visceral form of this disease is rapidly without (two dogs) cyclosporin A, followed by monthly progressive, and the prognosis is poor (1). The diagnosis of MH boosts. No significant clinical or laboratory toxicities devel- presents a challenge to both clinicians and pathologists because oped during cell therapy; interestingly, a strong correlation the disease lies within a spectrum of histiocytic and hematolym- was found between the dogs’ clinical and immunological phoid disorders and may be confused with granulomatous in- responses. One dog with advanced disease (intrathoracic flammation. Recently, phenotypic similarities between malig- involvement) refractory to chemotherapy achieved a corn- nant fibrous histiocytoma and malignant histiocytosis have been plete remission (CR) within 2 months of the first TALL-104 reported (6). A diagnosis of MH is appropriate once the malig- cell infusion. This dog died 14 months later of unrelated nant nature and histiocytic differentiation of the process have causes: histological analysis of its organs postmortem re- been established. vealed no evidence of neoplasia, thus confirming the achieve- We have developed a new cell therapy approach based on ment of CR also at the pathological level. The other three the use of the human leukemic T-cell line TALL-104 (CD3/ dogs with MH that at diagnosis had multiple s.c. and cuta- TCRaf3, CD8, CDS6, and CDI6), which is endowed with neous lesions and lyrnphadenopathy, but no visceral involve- potent MHC nonrestricted tumoricidal activity, without lysing ment, were treated with TALL-104 cells as single agent (no cells from normal tissues (7, 8). Unlike patient-derived lympho- chemotherapy was administered). Two of these dogs kine-activated killer cells, TALL-l04 cells provide an unlimited achieved a CR soon after cell therapy, and the third dog had and reliable source of clonal effecton cells with stable cytotoxic two long-lasting partial responses; CR in this dog was later activity that is ideal for cell therapy approaches. Although achieved by combined administration of chemotherapy and dependent on recombinant human IL-2 for expression of cyto- cell therapy. None of the three dogs that received cell ther- toxicity and long-term survival in vitro, TALL-104 cells can apy at diagnosis developed visceral disease in the -9-22 exert antitumon effects in vivo without the concomitant admin- months of follow-up. The clinical responses experienced by istration of IL-2, thus eliminating the potential toxicity of this all four MH cases to TALL-104 cell therapy suggest the high cytokine. TALL-104 cells induce necrotic tumor cell death responsiveness of this canine tumor to these xenogeneic through secretory pathways involving perform and senine ester- ases and/or kill targets through the release of cytostatic/cytotoxic mediators such as TNF-a, TNF-3, IFN-y, or transforming growth factor 3 (7, 9). Despite the fact that TALL-104 cells Received 1 1/8/96; revised 5/24/97; accepted 7/1/97. would be rejected by a tumor-bearing MHC-incompatible host, 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 in accordance with I 8 U.S.C. Section 1734 solely to indicate this fact.

I Supported in part by American Cancer Society Grant RD391, the

Connelly Foundation, the Parker Hughes Trust, and the Butler Family 3 The abbreviations used are: MH, malignant histiocytosis; CsA, cyclos- Fund. porn A: PBMC, peripheral blood mononuclear cell: FACS. fluores- 2 To whom requests for reprints should be addressed, at The Wistar cence-activated cell-sorting: TNF, tumor necrosis factor: GM-CSF, Institute, 3601 Spruce Street, Philadelphia, PA 19104. Phone: (215) granulocyte macrophage-colony stimulating factor: CR, complete 898-3978: Fax: (215) 573-7919. response: PR, partial response: IL, interleukin: NK, natural killer.

Table 1 Charactenis tics of the MH dogs treated with TALL- 104 cell therapy

Canine patient Sexlage (yr)fbreed Previous therapy Clinical stage at diagnosis

Dog I M/8/Scottish Terrier Five cycles of chemotherapy with Multiple cutaneous and s.c. lesions, doxorubicin, cyclophosphamide. metastatic lymphadenopathy, vincnistine, dacarbazine, and intrathoracic involvement L-asparaginase Dog 2 M/7IWest Highland Terrier None Multiple cutaneous and s.c. lesions, metastatic lymphadenopathy Dog 3 F/3/Bernese Mountain Dog None Cutaneous lesions on the muzzle and right axilla, metastatic lymphadenopathy Dog 4 F/l/Golden Retriever None Cutaneous and s.c. lesions on the muzzle; metastatic lymphadenopathy

we have used -1’-irradiation (40 Gy) as a precautionary measure three dogs were not given chemotherapy and were treated with to irreversibly arrest the growth of this leukemic cell line in the TALL-104 cells as the single agent. host tissues ( I 0). This treatment did not impair the antitumor Histological and Immunohistochemical Analysis. 5cc- effects of TALL-l04 cells in immunodeficient mice engrafted tions (6 jim) of paraffin blocks from the dogs’ tumor lesions with human neoplasms ( 1 1, 1 2) and in immunocompetent mice were stained with H&E for histological analysis. Immunohisto-

bearing syngeneic leukemia ( I 3), suggesting the feasibility of chemical staining was done using the avidin-biotin-peroxidase using -i-irradiated (nonproliferating) TALL-l04 cells in cancer complex method (iS): sections were deparaffinized, dehydrated

therapy. through graded alcohol, and washed in 0.01 M PBS (pH 7.4). To To determine the potential safety of TALL-l04 cells as an block nonspecific binding, sections were incubated with normal anticancer agent in a clinically relevant setting, we recently horse serum (1 :75; Vector Laboratories, Burlingame, CA) for 20 conducted a Phase I trial in pet dogs with spontaneous end-stage mm, followed by overnight incubation at 4#{176}Cwith primary tumors (14). Results of that study clearly showed not only the antibodies antilysozyme ( 1 :200), anti-a-i-trypsin (1:300; absence of significant adverse reactions in terminally ill canine DAKO, Carpinteria, CA), and anticathepsin B (1:300; ICN patients but also the manifestation of various levels of clinical Biochemicals, Costa Mesa, CA). Normal mouse serum was used responses including one CR and one PR in two dogs with MH. as control antibody. Sections were incubated in biotinylated The present report extends the follow-up for these two dogs and horse antimouse IgG (1 :200; Vector Laboratories) for 45 mm. describes the successful application of TALL-104 cell therapy All incubations were done in a humidified chamber at room to two more dogs with less-advanced MH. Results indicated temperature. 3-Amino-9-ethylcarbazole (Sigma Chemical Co., that: (a) this potentially fatal canine disease is extremely sensi- St. Louis, MO) was used as chromogen, and sections were tive to treatment with xenogeneic TALL-l04 cells; (b) durable counterstained with Mayer’s hematoxylin. CRs could be achieved even in dogs with advanced tumors, Large-Scale Expansion of TALL-104 Cells for Therapy. refractory to chemotherapy; and (c) TALL-l04 cells were very TALL-104 cells were grown in endotoxin-free Iscove’s modi- effective when used as a single agent in newly diagnosed cases fled Duibecco’s medium (Life Technologies, Inc., Grand Island, with systemic disease but no visceral involvement. NY) supplemented with 10% heat-inactivated fetal bovine serum (Atlanta Biologicals, Norcross, GA) and 100 units/ml re- MATERIALS AND METHODS combinant human IL-2 (Chiron Therapeutics, Emenyville, CA) Canine Patients. Four dogs diagnosed with MH were in a humidified incubator at 37#{176}Cwith 10% CO2 in T-175 treated with TALL-104 cells at the Veterinary Oncology Ser- vented cap flasks (Falcon, Franklin Lakes, NJ). Mycoplasma vices and Research Center (West Chester, PA), starting in contamination was monitored weekly on cell samples from 10 November 1994 (Tables 1 and 2). Complete work-up, including randomly chosen flasks using a commercial PCR kit (ATCC, complete blood counts, kidney and liver functions, chest X-rays, Rockville, MD). At the time of therapy, cells were harvested, abdominal ultrasounds, and bone marrow aspirate, was per- centrifuged in conical centrifuge tubes (Corning, New York, formed at diagnosis. The clinical characteristics of these patients NY), washed twice in saline (Abbott Laboratories, King of at the beginning of cell therapy are summarized in Table 1. Dog Prussia, PA), resuspended in 100 ml of saline, -1’-irradiated (40 I was enrolled in the study in 1994; at the time of diagnosis, he Gy) using a ‘37Cs source, and transferred to a blood transfer had advanced disease with skin, lymph nodes, and intrathonacic pack (Baxter Diagnostics, Inc., Glendale, PA) for systemic involvement (14). Although this dog had originally responded to administration within 2-4 h of irradiation. Cell aliquots were chemotherapy with a CR, progressive disease developed, and removed from the bag for determination of cytotoxic activity decreased cardiac contractibility precluded further treatment and sterility (quality control assays). with doxorubicin. The other three dogs (dogs 2, 3, and 4) that Protocols of TALL-104 Cell Administration. Table 2 entered the TALL-l04 cell study in 1995 and 1996 were newly summarizes the schedule of TALL-104 cell infusions received diagnosed cases with disease limited to skin and lymph nodes; by each dog in the study. Cells were rny-irradiated (40 Gy) and because of their less advanced stage and because of the dramatic administered at a dose of 108!kg in saline as an iv. bolus over response to TALL-104 cell therapy seen in dog 1 (14), these 30 mm. Cell boosts were given at the indicated intervals and

Table 2 Protocols of TALL-104 cell administration to dogs wit h MH Date of cell therapy Number of CsA Canine patient initiation Schedule boosts (10 mg/kg)

Dog 1 1 1/28/94 Every other day for 2 weeks 2 (weekly) Yes Dog 2 8/21/95 Daily for 5 days, 2-week rest, daily for 5 days 20 (monthly) Yes Dog 3 1/16/96 Daily for 5 days 15 (monthly) No Dog 4 9/1/96 Daily for 5 days, weekend rest, daily for 5 days 8 (monthly) No

consisted of 2 consecutive days of TALL-104 cell injections fore each TALL-l04 cell injection, and during the follow-up (108/kg; iv.). The immunosuppressive drug CsA (Sandimmune, period) to monitor the possible development of TALL-l04- Sandoz, East Hanover, NJ) was administered to dog 1 and dog specific humoral and cellular immune responses, respectively. 2 at a dose of 10 mg/kg p.o. once a day, starting from the day Sera were diluted iO in FACS buffer (PBS without Ca2 and before the first TALL-104 cell administration throughout the 2 Mg2 , 0.1% NaN3 and 2% IgG-free horse serum) and trans- weeks of treatment. In the boosts phase, CsA was given to dog ferred to wells of a 96-well round-bottom plate (Falcon). FACS 1 and dog 2 only on the day before and on the same day of the buffer was used as negative control. TALL-104 cells were two TALL-104 cell injections. Because CsA did not prevent the washed once in FACS buffer and added to the plates at iO development of humoral responses to TALL-104 cells (see cells/SO p.1/well. Sera and cells were incubated for 1 h at room below), the dogs treated in 1996 (dogs 3 and 4) were not given temperature. After three more washes, FITC-conjugated rabbit CsA in association with cell therapy. Except for dog 1, which anticanine IgG (whole molecule; Sigma) was added at a dilution was part of a Phase I study (14) and received TALL-104 cells on of 2 X lO_2 for 1 h at 4#{176}C.At the end of the incubation, cells alternate days for 2 weeks after chemotherapy, the other three were washed, resuspended in 150 p.1 of FACS buffer, and dogs that were treated with single-agent TALL-104 cells re- analyzed by flow cytometry using an Ortho Cytofluorograph ceived a more aggressive cell treatment consisting of one or two cell sorter (14). cycles of cells administered for S consecutive days (Table 2). To monitor the development of TALL-l04-speciflc cellular Toxicity Monitoring. Clinical signs of acute toxicity immune response, PBMC samples were tested against 51Cr- (such as fever, chills, hypotension, diarrhea, and vomiting) were labeled TALL-104 cells in an 18-h 51Cr-release assay, as de- monitored during and after each cell injection. All dogs were scribed above. treated as outpatients during the course of TALL-l04 cell ther- PCR Analysis. PBMC samples from the dogs were ob- apy, and the owners were properly instructed to report on the tamed immediately before each TALL-l04 cell infusion and at well-being of their pets during therapy. Blood and serum sam- various intervals during and after cell therapy, as indicated in the ples (for complete blood counts, serum chemistry, cytotoxicity “Results”; DNA was extracted from the PBMCs and frozen assays, cytokine production, and immunological monitoring; see (14). The presence of circulating TALL-l04 cells in each cell below) were obtained from the patients before the study, imme- extract was evaluated by PCR analysis using two primers spe- diately before each TALL-104 cell administration, and through- ciflc for the human minisatellite region YNZ.22 ( 16). An oh- out the follow-up period. gonucleotide probe recognizing 24 nucleotides in the middle of Cytotoxicity Assays. PBMCs were isolated from hepa- the amplified sequence was used to demonstrate the specificity rinized blood samples by Accu-Prep (specific gravity, 1077 of the PCR products by Southern blot hybridization, as de- g!ml; Accurate Chemical, Westbury, NY) lymphocyte gradient scribed previously (16). centrifugation and tested as effectors in an 1 8-h ‘ Cr-release assay against the human leukemic target cell lines U937 and K562 (both NK-sensitive) and TALL-106 (NK-resistant). This RESULTS test was routinely used to monitor the possible development of Diagnosis of MH. The diagnosis of MH in all four dogs MHC nonrestricted cellular immune responses against human was made based on clinical and morphological criteria (histol- (xenogeneic) cells. Briefly, a fixed number of 5tCr-labeled ogy reports from veterinary pathologists) and confirmation of target cells (l0 cells/well) was tested against four effector cell the histiocytic lineage of tumor cells by immunohistochemical concentrations, as described previously (7, 8). The percentage of staining (positive staining for the histiocytic markers a-l-tryp- specific 51Cr-release was calculated from the mean of three sin, cathepsin B, and lysozyme; data not shown). replicates. Clinical Responses to TALL-104 Cell Therapy. Table Cytokine Assays. The presence of human IFN--y, 3 summarizes the clinical responses observed in the four dogs TNF-a, GM-CSF, and TNF-3 in serum samples collected pre- with MH to TALL-l04 cell therapy (updated to June 1997). At study and throughout cell therapy was tested using human the time of initiation of cell therapy, dog 1 had prefemoral and cytokine-specific ELISA kits (Endogen, Boston, MA) according left inguinal enlarged lymph nodes and right axillary lymphad- to the manufacturer’s instructions. The sensitivity of the assay enopathy, and its performance status was poor (modified was 20 pg!ml for IFN--y and TNF-a, 8 pg/mI for TNF-3, and 7.8 Karnofsky performance status = 2). After 2 weeks of systemic pg/mi for GM-CSF (14). TALL-104 cell therapy (six injections), the right axillary node Immunological Monitoring. Immunological studies completely regressed, but the other nodes remained unchanged. were performed on serum samples and PBMCs (prestudy, be- Two cell boosts, 1 week apart (see Table 2), were then admin-

Table 3 Clinical responses to TALL-104 cell treatment in dogs targets: with MH 80 : U937 Clinical Canine patient response Duration (mo) Outcome Q TALL-106 Dog 1 CR 14 Died of causes 60’ unrelated to cancer 4) Dog 2 First PR 2 (I) CD Second PR 6 4) 4) CR” 4+ NED” Dog3 CR 16+ NED 40 Dog4 CR 9+ NED C.) U) ‘, The CR in dog 2 was achieved by a combination of chemotherapy/cell therapy.

,, NED, no evidence of disease. 20

istered; during this period of time, a significant improvement in the clinical well-being of dog 1 was documented (modified 11/28/94 1/11/95 1/25/95

Karnofsky performance status = 0), with complete regression of date of blood sampling the axillary lymph node and >50% decrease in the inguinal lymph nodes and in the neck mass. At this time, a thoracic Fig. I Cytotoxic activity of PBMCs from dog 1 against NK-sensitive radiograph was negative for lymphadenopathy and pulmonary (U937) and NK-resistant (TALL-lOb) target cells. PBMCs were sepa- infiltrates. TALL-104 cell therapy was ended at this point. At rated from blood samples obtained immediately before the first TALL- 104 cell infusion (November 28. 1994), on the day of regression of reexamination I week later, new s.c. nodules were found on the relapsed s.c. lesions (January 1 1. 1995), and 2 weeks later (January 25, dog’s left hind and left front legs. The nodules were confirmed 1995). E:T ratio, 50:1. to be a recurrence by cytological examination of needle aspi- rates. Surgical removal of the nodules followed by in vitro analysis of the susceptibility to TALL-l04 cell lysis was not possible, because 2 days later, the lesions had regressed by logical diagnosis of thromboembolic disease, which has been >50%. At reexamination 2 weeks later, no nodular lesions were described to be associated with renal failure ( 17, 18). palpable, and no signs of disease were detected. A state of Dog 2 was the second dog with MH treated with TALL- long-lasting CR was achieved, documented by the absence of 104 cells (and CsA; Table 3). This dog also had participated in clinical signs of relapse on repeated abdominal ultrasounds and our Phase I study (14). Before cell therapy, the dog had multiple thoracic X-nays throughout the rest ofthe dog’s life (14 months). cutaneous nodules between pads and along the metatarses of the Dog I developed recurrent bacterial cystitis approximately 6 right hind leg with metastatic popliteal lymph node. Within 3 months after its CR. A mild azotemia subsequently developed, weeks from the first S-day cycle of TALL-l04 cells, the most and abdominal ultrasound showed changes consistent with kid- proximal cutaneous lesions completely regressed, whereas the ney pyelonephritis. The disease was stabilized for a few months more distal mass had only partially regressed, and the popliteal using antibiotics and a low-protein diet as supportive therapy; lymph node had remained unchanged. A second 5-day cell cycle however, renal failure grew more severe, and dog I eventually was then administered. In the following month, dog 2 achieved died with a clinical picture of acute abdominal disease with a PR (>50% reduction of cutaneous masses and popliteal lymph ascites and bloody diarrhea. Necropsy revealed ischemic necro- node) that lasted for 2 months (Table 3). After the second sis of the intestine (ultimate cause of death), but no macroscopic month, both the interdigital masses and lymph node grew back. signs of MH. In particular, its spleen and liver, which are the A second PR was achieved after 2 monthly boostings with classic sites of systemic MH (1-5), seemed normal. Histological TALL-104 cells and lasted about 6 months with stabilization of analysis (H&E staining) was performed blindly by pathologists the cutaneous lesions. No signs of visceral disease were seen in at the University of Pennsylvania Veterinary School of Mcdi- this dog throughout cell therapy. When the lesions of the right cine on formalin-fixed, paraffin-embedded necnopsy samples of hind leg started to progress again (12 months from start of cell the dog’s kidneys, bladder, liver, spleen, stomach, intestine, therapy), the decision was made to suspend TALL-l04 cell heart, and lungs and of the mesenteric and mediastinal lymph treatment. Chemotherapy (doxorubicin and dacarbazine) given nodes. No infiltrating malignant histiocytes were identified in to this dog for S days induced a CR of the skin lesions and a PR any of these organs, and no evidence of other types of cancer of the popliteal lymph node. Aspiration cytology of the lymph was found at necropsy. On the other hand, severe chronic node did not reveal any evidence of intact histiocytes. Another lymphoplasmacytic tubulointerstitial ncphritis was found to- S-day TALL-104 cell cycle, given 2 weeks later as maintenance gethen with other organ changes, such as uremic pneumonitis therapy, resulted in a CR with complete disappearance of the and uremic gastnitis (data not shown), typically seen in severe lesions. At the time of this writing, dog 2 has no evidence of kidney failure leading to uremia. Moreover, signs of thrombo- cutaneous or visceral disease (4+ months; Table 3). embolism were detected in the liver and spleen and around the Dog 3 was diagnosed with MH in December 1995 and abdominal lymph nodes (not shown), consistent with a patho- started TALL-l04 cell treatment in the middle of January 1996

Dog#2 80

60’

4) Cl) CU 4) a) 40’ I-

I- C.)

Dog#3 Dog#4

0) 4) Cl) Cl) CU CU a, a) a) 4) 40

0 0

0 C’) It) C) It) (0 CD C) (“1 C’) C1 It) C) 0 It) C1 O (0 )‘.. (0 (“4 C’) r’.. C’) (‘4 -C #{149} #{149}C 1 1- 1 rsl ei CJ C’4 C’) ‘) #{149}D#{149} .(‘4 (“4 C’)

Time of blood sampling

Fig. 2 Cytotoxic activity of PBMCs from three dogs with MH on TALL-104 therapy. PBMC samples were harvested at the indicated times (d, day; w’ week) after initiation of cell therapy and tested for killer activity against K562 (0) and TALL-104 (0) cells in an 1 8-h 51Cr release assay. The E:T ratio was 50: 1 . Arrows in the Dog #2 panel indicate the times of relapse, documented clinically.

(Table 2). At the time of diagnosis, the dog had cutaneous size. However, 1 month later, just before the second cell lesions on the muzzle and right axilla, which were histolog- boosting was scheduled, new cutaneous lesions appeared ically confirmed to be cutaneous histiocytosis with a biopsy. along the muzzle. The dog received his second 2-day cell There were also regional metastatic lymph nodes. Chest boost, and in the following month, the new cutaneous lesions X-rays and abdominal ultrasound excluded intrathoracic and completely regressed: a long-lasting CR was then achieved abdominal involvement, respectively. One month after one with no signs of cutaneous or visceral disease recurrence, as S-day TALL-104 cell cycle, a decrease in the cutaneous documented in follow-up visits. Dog 3 remains disease-free lesions and axillary lymph nodes started to be noticeable. at the time of preparation of this manuscript (16+ months; After the first 2-day boost with TALL-l04 cells, the right Table 3). axihlary lymphadenopathy completely regressed, and the cu- Dog 4 was immediately treated with TALL-l04 cells at taneous lesions went down by more than 50% of their initial diagnosis in September 1996. At this time, the dog had multiple

Dog#1 Dog#2

1 00 - 000000000 100- 0000000000000

75 - 75.

50- 50-

Cl) 25 - 25- w 0 C.) ‘1. 0 00000 1 ‘I I I I I I I C I I I - ‘ IIIIIIIIIIIIII I 0 C’) It) ,- C”.l C’) ‘ CS) ‘‘ (0 (0 0 (“'1 ‘‘ ‘00’0 E EEE vv.o E E E E E” E E E E E E .1 ,- 1 #{149}- #{182}e I-. Dog#3 Dog#4

100- Q00000000000 100- 0 0 0 0 0 0 0 0 0

75. 75. C.)

0) 0

50’ 50-

25’ 25- 0

0 000

IIIIIIIIIIIIIII 0- 1 IIIIIIIIIII 0 C’) U) C”.I C’) - C”4 ‘ Ct) (0 0 (“1 CD (0 0 C’) LI) (“'I C’) . C”’) C’) ‘ci’ (0 (0 0) E E E E ..D.o E E E E E E E

Time of sera sampling

Fig. 3 Humoral immune responses against TALL- 104 cells mounted by the four dogs with MH on cell therapy. Sena of the treated dogs were harvested at the indicated times (d, day; w, week: m, month) after initiation of cell therapy, diluted at iO”3, and tested for the presence of TALL-104-spccific antibodies. The percentage of reactivity with TALL-104 cells stands for the percentage of TALL-104 cells binding immuno- globulin in the canine sera, as tested by FACS analysis.

s.c. nodules around the mouth and a skin lesion on the ventral Clinical Laboratory Changes and Toxicity Associated abdomen as well as enlarged submandibular, left pophiteal, and with TALL-104 Cell Therapy. Dogs 1 and 2 exhibited vom- prescapular lymph nodes. After the first S-day TALL-l04 cell iting (responsive to antiemetics) and diarrhea (treated with di- cycle, no changes were noted in any of the lesions, and a second phenoxylate/atropine) during cell administration and in the fol- cell cycle was administered (Table 2); within 1 week of the last lowing 24 h. These symptoms could be prevented in dog 2 by injection, all of the cutaneous lesions as well as the lymph nodes premedication with an antihistaminic. The other two dogs did started to regress, and a CR was achieved (Table 3). The dog is not show any signs of clinical toxicity associated with TALL- still in CR at the time of this writing (9+ months). 104 cell injections. Blood chemscreens, run regularly on sam-

Dog#1 2 3 6 7 8 and NK-resistant targets) and clinical disease (Figs. I and 2). In ______particular, the cytotoxic activity of dog l’s PBMCs against U937 (NK-sensitive) and TALL-l06 (NK-nesistant) cells was close to 0% before cell therapy (November 28, 1994), high just Dog#2 1 2 3 4 5 6 7 8 9 10 11 at the time when the tumor nodules appeared and then regressed (January 1 1, 1995), low 2 weeks later when CR was achieved

10 , (January 25, 1995; Fig. 1), and negative in the following months (data not shown). Dog#3 2345678 In the case of dog 2, the correlation between cell-mediated immunity and disease progressionlregression was even more evident because of the fluctuating course of his disease and the closer monitoring of his immunity (Fig. 2); three peaks of cytotoxic activity by dog 2’s PBMCs against K562 cells were

Dog #4 1 2 3 4 5 6 7 8 detected, the first during the initial PR, the second at the time of the first relapse and during the second PR, and the third at the time of the second relapse. In the periods of time between tumor : regression (when the disease was stable), the cytotoxic activity

Fig. 4 PCR amplification of the human minisatellite region YNZ.22 of this dog’s PBMCs was close to 0%. Similar curves were seen performed on the PBMCs from each of the four dogs with MH before in the case of dog 3 and dog 4 (Fig. 2), although their clinical and at different times after TALL-l04 cell injections. In all gels: Lane courses were much simpler (CR without clinically detectable 1, TALL-104 cells (positive control) and Lane 2, water (negative con- relapses). trol). In the gel for dog 1: Lane 3, pre-TALL-l04 cell therapy; Lanes 4 Immune Response Against TALL-104 Cells. No cellu- and 5, days 3 and 5, respectively, of the first TALL-l04 cell therapy cycle; Lanes 6-8, 1, 2, and 3 weeks after the last TALL-104 cell cycle. lar immune responses against TALL-104 cells could be demon- In the gel for dog 2: Lanes 3 and 4, days 3 and 5 of the first TALL-104 strated in PBMC samples from dog 1 at any interval during and cell cycle; Lanes 5 and 6, 12 and 36 days after the first TALL-104 cell after cell therapy (data not shown), whereas specific CTLs injection; Lane 7, day 2 of the first TALL-104 cell boost; Lanes 8 and against TALL-104 cells were present in the PBMCs of the other 9, 6 and 9 months later; Lane 10, day 2 ofthe 10th TALL-104 cell boost; Lane 11, 1 year after beginning ofTALL-l04 cell therapy. In the gel for three dogs (Fig. 2). dog 3: Lanes 3-6, 5, 6, and 7 months after the beginning of TALL-l04 Despite the immunosuppressive regimen with CsA, dogs 1 cell therapy: Lane 7, day 2 of the TALL-104 cell boost given 8 months and 2 developed antibodies against TALL-l04 cells between the after beginning of cell therapy; Lane 8, ninth boost. In the gel for dog second and third week of treatment (Fig. 3). The same was true 4: Lane 3, pre-TALL-104 therapy; Lane 4, day 4 ofthe 5-day TALL-l04 for dog 3 and dog 4 that did not receive any immunosuppressive cell therapy cycle; Lanes 5-8, days 1 1, 18, 22, and 28 after the first TALL-104 cell injection. PCR assays on samples from dogs 1 and 2 drug (Fig. 3). This humoral immune response in dogs 1, 2, and were run at different times (with new controls each time). PCR assays 3 was still present 1 year after beginning cell therapy (Fig. 3). on the samples from dogs 3 and 4 were performed at the same time using However, none of the TALL- I 04-reactive serum samples tested cryopreserved PBMCs. was able to inhibit TALL-104 cell tumoricidal activity, as de- termined by in vitro cytotoxicity assays against human tumor target cell lines (data not shown). Detection of Circulating TALL-104 Cells. PCR ampli- pies obtained during cell therapy, revealed no abnormalities. fication of the human minisatellite region YNZ.22 performed on Discrete leukocytosis with neutrophilia was noted in dog I the PBMCs of the TALL-l04-treated dogs documented the during the second week of treatment and after the two cell presence of TALL-104 cells in the circulation at various time boosts; the number of WBCs reached a maximum of 21.7 X points during therapy as well as their disappearance within S 109/liter with 98% granulocytes, but returned to baseline levels days after each injection (Fig. 4). The lack of long-term persist- 48 h after halting the therapy. No hematological changes were ence of TALL-l04 cells in the circulation was confirmed by seen in the other three dogs. PCR analysis in each dog months after the beginning of cell Correlation Between Laboratory Findings and Clinical therapy (Fig. 4). Responses. Serum samples obtained from the dogs at various times before, during, and posttherapy were evaluated for the presence of human cytokines (TNF-a, IFN--y, TNF-3, and GM- DISCUSSION CSF) possibly released by TALL-104 cells on tumor cell inter- Since its establishment from a child with T-cell leukemia action in vivo (19). Only human TNF-3 levels were significantly (7), the TALL-l04 cell line has been characterized extensively elevated in the serum of the dogs during TALL-104 treatment both in vitro, for its ability to selectively recognize and kill (data not shown). The lack of significant IFN--y, TNF-ct, or tumor cells in a MHC nonrestricted fashion, and in vivo, for its GM-CSF levels in dogs’ serum samples might be due to the antitumor efficacy in animal models with transplantable malig- lower production and/or metabolism of these cytokines or to nancies. Although the antigenic structures recognized by TALL- different production kinetics. 104 cells on the surface of tumor cells remain to be defined, An interesting observation in this study was the striking such structures seem to be phylogenetically well conserved, correlation between the dogs’ cell-mediated immune response because TALL-104 cells can kill tumors across species. More- (as measured in 51Cr-release assay against human NK-sensitive over, normal cells seem to escape TALL-l04 cell killing on

damage (8, 10). The lack of TALL-104 cell toxicity to healthy effector cells for treatment of canine malignancies, it was crucial tissues has been shown in a variety of animal models, including in the present study to demonstrate not only the lack of acute mice, dogs, and nonhuman primates (20). The possibility of toxicity in these animals as an immediate consequence of administering TALL-104 cells as a single antitumor agent rep- TALL-l04 cell infusion (± CsA), but also the lack of chronic resents an advantage oven the use of the patient’s own lympho- tissue injury resulting from the release of toxic cytokines and/or kine-activated killer cells, which requires the concomitant infu- to a slow clearance of dying TALL-104 cells. Although labo- sion of toxic doses of IL-2 for efficacy (2 1 ). In addition, the ratory findings showed no direct evidence of acute kidney continuous availability of TALL-l04 cells, the possibility of toxicity during the course of cell therapy, the development in expanding them to therapeutic doses (lOs cells/kg) for repeated dog 1 of severe chronic pyelonephritis a few months after infusions, their high migratory activity to tumor sites (22), their systemic injection of TALL-104 cells raises the question of a antitumon efficacy even in the absence of cell proliferation (after causal relationship between CsA/cell therapy and the onset of lethal irradiation), and their ability to induce either necrotic on nephritis. In this regard, dog 1 had recurrent episodes of bacte- apoptotic death in tumors across MHC barriers (7-14) are all rial cystitis before the onset of evident renal disease, with no aspects pointing to the feasibility, safety, and efficacy of these alterations in kidney functions detectable during cell treatment. cells as an anticancer agent. The Phase I preclinical trial we Kidney abnormalities were also absent in all long-term survi- recently conducted in canine patients with advanced tumors (14), which included dog 1 and dog 2 of the present report, has vors enrolled in our Phase I clinical trial with TALL-l04 cells revealed important information on the possible use of TALL- with or without CsA (14) and in the other dogs with MH 104 cells in a clinical setting, specifically their ability to antag- described in this study. In addition, no significant changes in onize tumor growth even in an advanced disease setting of kidney function were observed in long-term toxicity studies refractory malignancies, without inducing significant adverse performed on healthy dogs and nonhuman primates that ne- reactions. ceived multiple systemic TALL-104 cell injections (20). On the The findings in the present report are particularly encour- other hand, a high frequency of renal disease has been reported aging, considering the aggressiveness and the poor prognosis of in a number of terrier breeds (23, 24). MH in most of the dogs and in humans. In this context, the The histopathological changes noted on dog 1 ‘s kidneys at achievement of a durable CR in a dog with very advanced the time of postmortem examination were not compatible with disease (dog 1) on systemic TALL-104 cell infusion was unex- those found with CsA-associated nephrotoxicity or with auto- pected, as was the regression of its relapsed nodules after the immune nephropathy. Renal biopsies of CsA-treated patients completion of cell therapy. The latter observation suggested the show an interstitial fibrosis with tubular atrophy (25). Toxic development of a state of specific antitumor immunity that tubulopathy, peritubular capillary congestion, arteriolopathy, protected this dog from further relapses throughout its life span. and a striped form of interstitial fibrosis with tubular atrophy Immunological studies of this dog’s blood samples before, dun- may also be present (25). On the other hand, immunologically ing, and after cell therapy showed the appearance of MHC induced nephropathy is usually characterized by primary alter- nonrestricted cell-mediated cytotoxicity against human tumor ations of the glomeruli (membrane-proliferative glomerular- cells (both NK-sensitive and -resistant) during TALL-104 then- nephritis; Ref. 26) that were absent in dog 1‘s kidneys. apy, with a peak response at the time of regression of the The remarkably long-lasting clinical responses associated relapsed nodules. The same pattern of responses was seen in the with activation of the immune system seen in this study on other three dogs treated. Unfortunately, no tumor specimens TALL-l04 cell therapy in dogs with histologically documented were available from the dogs during this study to test the development of specific antitumor immunity. However, exper- MH suggests the potential of these cells for treatment of this iments with other dogs in our Phase I trial (14) demonstrated the malignancy. To fully explore the feasibility of the TALL-104 ability of their PBMCs to recognize and kill their own tumor cell approach in an adjuvant setting, we are now conducting a cells in vitro during the course of TALL-lO4 cell therapy. The Phase Il/Ill clinical trial in dogs with minimal disease after induction of endogenous antitumor immunity on passive trans- remission induction for different types of cancer. These studies fer of TALL-104 cells was recently demonstrated in a murine will be crucial to the identification of tumors that are either model of syngeneic leukemia in which TALL-104-treated and resistant or highly susceptible to TALL-104 therapy and, ulti- cured mice were able to reject additional leukemia cell chal- mately, to the development of optimal treatment protocols lenges ( 1 3). That the tumor regression observed in dogs 1 , 3, and against both sets of tumors. These protocols envision the appli- 4 was a consequence of TALL- I 04 cell treatment was suggested cation of TALL-104 cells as either a single agent or in combi- by the facts that: (a) the disease in dog 1 was progressing under nation with chemotherapy, depending on the clinical responsive- chemotherapy; and (b) none of these three dogs received other ness of the tumor being treated. drugs (aside from CsA) during and/or after completion of cell therapy. It is also unusual that the disease in dog 2 did not systemically progress 22 months after the diagnosis. Moreover, ACKNOWLEDGMENTS to our knowledge, no spontaneous remissions have even been This work would not have been possible without the enthusiastic documented in dogs with MH; therefore, the tumor regressions participation, encouragement, and support of the dogs’ owners, to whom achieved upon single-agent TALL-104 cell therapy are to be we are deeply grateful. The histology, media, and flow cytometry considered significant. facilities and the editorial department of The Wistar Institute arc ac- Because of the novelty of infusing human (xenogeneic) knowledged.

REFERENCES human MHC nonrestricted cytotoxic T-cell line (TALL-104) in dogs with advanced tumors. Cancer Res., 56: 3021-3029, 1996. 1. Schalm, 0. W. Histiocytic medullary reticulosis. Canine Pract., 5: 42-44, 1978. 15. Hsu, S. M., Raine, L., and Fanger, H. Use of avidin-biotin-perox- 2. Schmidt, M. L., Rutteman, G. R., van Niel, M. H. F., and idase complex (abc) in immunoperoxidase techniques: a comparison Wolvekamp, P. T. Clinical and radiographic manifestations of canine between abc and unlabeled antibody (pap) procedures. J. Histochem. malignant histiocytosis. Vet. Q., 15: 1 17-120, 1993. Cytochem. 29: 577-580, 1981. 3. Moore, P. F., and Rosin, A. Malignant histiocytosis of Bernese 16. Mackinnon, S., Barnctt, L., Bourhis, J. H., Black, P., Heller, G., and Mountain dogs. Vet. Pathol., 23: 1-10, 1986. O’Reilly, R. J. Mycloid and lymphoid chimenism after T-ccll-depletcd 4. Rosin, A., Moore, P. F., and Dubiclzig, R. Malignant histiocytosis in bone marrow transplantation: evaluation of conditioning regimens using Bernese Mountain dogs. J. Am. Vet. Med. Assoc., 188: 1041-1045, polymerase chain reaction to amplify human minisatellite regions of 1986. genomic DNA. Blood, 21: 3235-3241, 1992. 5. Wcllman, M. L., Davenport, D. J., Morton, D., and Jacobs, R. M. 17. Greene, R. A., and Kobel, A. L. Hypercoagulable state in three dogs Malignant histiocytosis in four dogs. J. Am. Vet. Med. Assoc., 187: with nephrotic syndrome: role of acquired antithrombin III deficiency. 919-921, 1985. J. Am. Vet. Med. Assoc., 181: 914-920, 1982. 6. Kerlin, R. L., and Hendrick, M. J. Malignant fibrous histiocytoma I 8. Greene, R. A. Hypoalbuminemia-nelated platelet hypersensitivity in and malignant histocytosis in the dog: convergent or divergent pheno- two dogs with nephrotic syndrome. J. Am. Vet. Mcd. Assoc., 186: typic differentiation? Vet. Pathol.. 33: 7 13-7 16, 1996. 485-496, 1985. 7. O’Connor, R., Cesano, A., Lange, B., Finan, J., Nowell, P. C., Clark, 19. Blay, J-Y., Favrot, M. C., Negrier, S., Combaret, V., Chouaib, S. C., Raimondi, S. C., Rovena, G., and Santoli, D. Growth factor S., Mercatello, A., Kaemmerlcn, P., Franks, C. R., and Philip, T. requirements of childhood acute T lymphoblastic leukemia: correlation Correlation between clinical response to interleukin 2 therapy and between presence of chnomosomal abnormalities and ability to grow sustained production of tumor necrosis factor. Cancer Res., 50: permanently in vitro. Blood, 77: 1534-1545, 1991. 2371-2374, 1990. 8. Cesano, A., and Santoli, D. Two unique leukemic T-cell lines en- 20. Cesano, A., Visonneau, S., Wolfe, J., Jeglum, A., Fernandez, J., dowed with stable cytotoxic function and different spectrum of target Gillio, A., O’Reilly, R., and Santoli, D. Toxicological and immunolog- reactivity: analysis and modulation of their mechanisms. In Vitro Cell. ical evaluation of the MHC nonrestricted cytotoxic T-cell line TALL- Dcv. Biol., 28A: 648-656, 1992. 104. Cancer Immunol. Immunother., 44: 125-136, 1997. 9. Cesano, A., and Santoli, D. Inducible expression of granulocyte- 21. Rosenberg, S. A., Lotze, M. T., Muul, L. M., Leitman, S., Chang. macrophage colony-stimulating factor, tumor necrosis factor-a, and A. E.. Ettinghausen, S. E., Matory, Y. L., Skibber, J. M., Shiloni, E., interferon--y in two human cytotoxic leukemic T cell lines. In Vitro Cell. Vetto, J. T., Scipp, C. A., Simpson, C., and Reichert, C. M. Observations Dcv. Biol., 28A: 657-662, 1992. on the systemic administration of autologous lymphokinc-activated 10. Cesano, A., Visonncau, S., Cio#{233},L.,Clark, S. C., and Santoli, D. killer cells and recombinant interlcukin-2 to patients with metastatic Effects of lethal irradiation and cyclosponin A treatment on the growth cancer. N. EngI. J. Med., 313: 1485-1492, 1985. and tumoricidal activity of a T-ccll clone potentially useful in cancer 22. Visonneau, S., Cesano, A., Torosian, M. H., and Santoli, D. Cell therapy. Cancer Immunol. Immunother.. 40: 139-151, 1995. therapy of a highly invasive human breast carcinoma implanted in 11. Cesano, A., Visonneau, S., Cio#{233},L.,Clark, S. C., Rovera, G., and immunodeficient (SCID) mice. Clin. Cancer Res., 3: 1491-1500, 1997. Santoli, D. Reversal of acute myclogenous leukemia in humanized SCID mice using a novel adoptive transfer approach. J. Clin. Invest., 94: 23. Eniksen, K., and Grondalen, J. Familial renal disease in softcoated 1076-1084, 1994. Wheaton terriers. J. Small Anim. Pract., 25: 489-501, 1984. 12. Cesano, A., Visonneau, S., and Santoli, D. Treatment of experi- 24. Picut, C. A., and Lewis, R. M. Hereditary kidney disease in dogs. mental glioblastoma with a human MHC nonrestricted cytotoxic T-cell Kal Kan Forum, 8: 2-8, 1989. line. Cancer Res., 55: 96-101, 1995. 25. Bennett, W. M. Basic mechanisms and pathophysiology of cyclos- 13. Cesano, A., Visonneau, S., Pasquini, S., Rovera, G., and Santoli, D. porn ncphrotoxicity. In: B. D. Kahan (cd), Cyclosporin, Diagnosis, and Antitumor efficacy of a human MHC nonrestricted cytotoxic T-ccll line Management of Associated Renal Injury, pp. 297-302. Orlando, FL: (TALL-l04) in immunocompetent mice bearing syngeneic leukemia. Grunc and Stratton, Inc. Publishing, 1985. Cancer Res., 56: 4444-4452, 1996. 26. Berkow, R., and Fletcher, A. J. Immunologically mediated renal 14. Cesano, A., Visonncau, S., Jeglum, K. A., Owen, J., Wilkinson, K., diseases. In: The Merck Manual of Diagnosis and Therapy, 16th Ed., pp. Caner, K., Reese, L., and Santoli, D. A Phase I clinical trial with a 1677-1683. Rahway, NJ: Merck Research Laboratories, 1992.

Downloaded from clincancerres.aacrjournals.org on September 29, 2021. © 1997 American Association for Cancer Research. Successful treatment of canine malignant histiocytosis with the human major histocompatibility complex nonrestricted cytotoxic T-cell line TALL-104.

S Visonneau, A Cesano, T Tran, et al.

Clin Cancer Res 1997;3:1789-1797.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/3/10/1789

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/3/10/1789. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.