Ex vivo activated memory t-lymphocytes as adoptive cellular therapy of human soft-tissue sarcoma targets with potentiation byCis-diamminedichloroplatinum (II)

Journal of Surgical Oncology 58:212-221 (1995)

Ex Vivo Activated Memory T-Lymphocytes as Adoptive Cellular Therapy of Human

Soft-Tissue Sarcoma Targets With Potentiation by Cis- Diamminedich loroplat hum( I I)

JAY E. GOLD, MD, PhD, TED R. MASTERSt, PhD, NORMAN D. BLOOM, MD,

MlCHAlL K. SHAFIR, MD, MICHAEL J . KLEIN, MD, PhD, SAMUEL KENAN, MD, AND

MICHAEL E. OSBAND, MD

From the Division of Hematology, Department of Medicine, and the Departments of Surgery, Orthopaedics, and Pathology O.E.G., T.R. M., N.D.B., M.K.S., M.I.K., S.K.), Mount

Sinai Hospital and Mount Sinai School of Medicine of the City University of New York, New York, New York; (M.E.O.) Division of Pediatric Hematology-Oncology, Department of

Pediatrics, Boston City Hospital and Boston University School of Medicine, Boston, Massachusetts; Cellcor, Newton, Massachusetts; Odyssey Therapeutics,

Cambridge, Massachusetts

Autolymphocyte therapy (ALT) is tumor-specific, adoptive cellular therapy of neoplastic disease using nonspecific ex vivo activation of autologous peripheral blood lymphocytes (PBL) , which are composed primarily of memory T-cells (ALT-cells) and are active in patients with metastatic renal cell carcinoma and melanoma. Ex vivo pretreatment of tumor target cells with certain chemotherapeutic agents can enhance susceptibility to lysis by antitumor lymphocytes. To determine if cis-diamminedichloroplatinum(I1) (CDDP) enhances ex vivo antitumor cytotoxicity of ALT- cells and if this lysis is mediated by T- and/or NK-cells and is human leukocyte antigen (HLA)-restricted, human soft tissue sarcoma (STS) target cells were derived from primary and metastatic surgical specimens and were incubated with and without CDDP. ALT-cells were prepared from autologous PBL obtained prior to surgery. Primary (PSTS) and metastatic (MSTS) target cells from each group were labelled with chromium 51 ("Cr) and used as targets for ALT-cells, CD45-depleted ALT-cells, CD56 (NK)-depleted ALT-cells, and PBL in a standard (4-hour) and delayed (1 8-hour) 'lCr release assay. Interferon-gamma (IFN-y) release was measured as an indication of antitumor effect and recognition by the noncy- tolytic lymphocytes in ALT-cells. Primary tumor target cells incubated in CDDP showed enhanced lysis as measured by the 51Cr release assay compared to non-CDDP-treated controls. Metastatic tumor target cells showed less lysis than the primary targets, although this was enhanced by pretreating metastatic tumor targets with CDDP. Lysis of all tumor targets was significantly greater when ALT-cells were used as the effector cells rather than PBL. Depletion of memory T-cells abrogated ex vivo lysis. Depletion of NK cells (CD56+) affected ex vivo lysis of autologous targets during the 4-hour but not the 18-hour assay. Ex vivo ALT-cell lysis and IFN-y release against only the autologous tumor targets confirmed tumor-

Accepted for publication November 15, 1994. Address reprint requests to Jay E. Gold, M.D., Mount Sinai Medical Center, 920 Park Avenue, New York, NY 10028. This work was presented at the 47th Annual Meeting of the Society of Surgical Oncology, March 17-20, 1994, Houston, Texas, 'Deceased.

0 1995 Wiley-Liss, Inc.

Adoptive T-cell Immunotherapy and CDDP of STS Targets 213

specificity in one patient. Restriction of ALT-cell lysis and IFN-y release against HLA-A2+ autologous and one allogeneic HLA-A2+ STS tumor target, but not other non-STS targets, was demonstrated in another patient. These data suggest that CDDP may help render STS susceptible to tumor- specific, immune-mediated attack and that the combination of ALT and CDDP may lead to effective tumor-specific chemoimmunotherapy in patients with metastatic STS. 0 1995 Wiley-Liss, Inc.

KEY WORDS: autolymphocyte therapy; immunotherapy, adoptive; T-cell, memory; sarcoma, soft-tissue; cis-diamminedichloroplatinum(I1)

INTRODUCTION Autolymphocyte therapy (ALT) is adoptive cellular

therapy of neoplastic disease using infusions of autologous peripheral blood lymphocytes (PBL) from tumor- bearing hosts (TBH), the PBL being activated ex vivo by low doses of anti-CD3 mitogenic monoclonal antibodies, with a mixture of previously prepared autologous cytokines (T3CS). We have demonstrated that nonspecific ex vivo activation of syngeneic or autologous PBL from murine or human TBH using T3CS results in the expansion of tumor-specific memory T-cells (autolympho- cytes; ALT-cells) [ 1-31. Monthly infusions of these memory T-cells and daily oral cimetidine (to reduce tumor-associated suppressor cell activity) as ALT have been shown to be active in human TBH with metastatic melanoma and renal cell carcinoma (RCC) [ M I . In a preliminary randomized trial, patients with metastatic RCC who were treated with ALT had a three- to fourfold prolonged survival with improved quality of life over those in the control group [6]. A recent update of the initial protocol and a new study with additional patients confirmed a 45% survival advantage as well as an 18% response rate of the ALT-treated RCC group [4]. Patients with advanced metastatic melanoma demonstrated complete and partial responses as well as increased survival compared to historical controls when matched for number of organ systems involved [5]. As previously described, this approach is not dependent upon infusions of exoge- nous IL-2 but rather upon the levels of IL-1, GM-CSF, IFN-y, IL-8, and IL-6 generated in the T3CS used for ex vivo activation [ 1,7]. Therefore, the toxicity of ALT is markedly less than other forms of IL-Zdependent adoptive cellular therapy, such as lymphokine-activated killer (LAK)-cell or tumor-infiltrating lymphocyte (TIL) ther-

In murine TBH with solid or hematopoietic tumors, the combination of adoptively transferred T-cells and chemotherapeutic agents can result in enhanced cure of neoplastic disease compared to either modality alone [9-151. In addition, recent reports have demonstrated that pretreatment of murine and human tumor cells ex vivo with chemotherapeutic agents such as cis-diamminedichloro-

apy [81.

platinum(I1) (CDDP) can enhance their susceptibility to antitumor adoptively transferred T-cells both ex vivo and in vivo [14,16-211. Aside from their direct tumoricidal effects, these agents also possess potent immunomodula- tory properties both on the tumor cell itself as well as the adoptively transferred T-cells [9,10,17,18]. We and others have previously confirmed that incubation of human breast, colorectal, and RCC tumor targets with CDDP can either enhance ex vivo lysis of these tumor targets by immune effector cells or allow enhanced ex vivo tumor recognition as measured by interferon-gamma (IFN-y) release [ 19,20,21-241. For ALT-cells, this lysis was dependent on the presence of memory T-cells as depletion of the memory T-cell subset in murine and human TBH significantly reduced lysis of even CDDP-treated tumor targets [3,19,20]. Additionally, there have been reports demonstrating ex vivo lysis of sarcoma tumor targets by antitumor-specific T-cells derived from peripheral blood [25-271. Based on these studies as well as the results of a pilot protocol of adoptive chemoimmunotherapy showing in vivo synergy between ALT and cyclophosphamide in relapsed human non-RCC solid tumors [28], we sought to determine if: (1) CDDP could enhance the ex vivo antitumor cytotoxicity of ALT-cells against another chemotherapy-resistant tumor such as soft-tissue sarcoma (STS) by immunomodulation, (2) there was a difference in lysis of primary versus metastatic lesions and if this difference also could be effected by tumor target pretreatment with CDDP, (3) ex vivo lysis was dependent on T-cells and/or NK-cells, and (4) these ALT-cells demonstrated autologous tumor specificity by measuring ex vivo cytotoxicity and IFN-y release.

MATERIALS AND METHODS Tumor Specimens

PBL and STS specimens were obtained from 15 patients, including 9 men and 6 women. The median age was 52 years. Tumor specimens included 10 primary STS and five metastatic STS samples removed at the time of surgery. Among the STS were five leiomyosarcomas, one neurosarcoma, one endometrial stromal sarcoma, three malignant fibrous histiocytomas, one synovial sar-

214 Gold et al.

coma, one extraosseous osteogenic sarcoma, and three rhabdomyosarcomas. No patient had received any anti- cancer therapy prior to the time of surgery. Additional tumor targets used to study ex vivo cytotoxicity and specificity of STS-derived ALT-cells were the nonadherent myeloid leukemia line K562, the HLA-A2 malenoma HT144, the HLA-A2+ melanoma Malme-3 and pancre- atic carcinoma Capan- 1, and HLA-A2' osteogenic sarcoma lines U-2 0s and Saos-2 (all from American Type Culture Collection). The HLA-A2 status of the STS targets PSTS- 1, PSTS-4, and MSTS-5 was confirmed using the monoclonal antibody BB7.2 (HLA-A2; American Type Culture Collection) by indirect immunofluorescence.

Drugs CDDP was purchased from the Bristol-Myers Com-

pany (Lot #AOF 32A), diluted in normal saline, and then used for the pretreatment of STS and melanoma primary and metastatic tumor cell targets.

Preparation of Autologous Lymphokine Mixture (T3CS)

Peripheral blood was obtained from patients in the perioperative period. The blood was placed on 100% Ficoll-Hypaque (LSM-Litton Bionetics, Kensington, MD) and centrifuged at 400 g for 30 minutes. Mononu- clear cells were collected, washed in RPMI-1640, suspended in complete medium (CM), and divided into two equal portions. CM was composed of RPMI-1640, with 10% heat-inactivated human AB serum, 25 mmol/L HEPES, 2 mmol/L L-glutamine, 1 % nonessential amino acid, 100 U/ml penicillin, 100 kg/ml streptomycin, and 0.5 pg/ml Fungizone (all from Gibco). To generate T3CS, cells from one of the divided portions were incubated in the presence of OKT3, a mitogenic anti-CD3 monoclonal antibody directed at the antigen nonspecific CD3 portion of the T-cell receptor [29,30]. After 3 days, the OKT3 culture supernatant (T3CS) was harvested, ultrafiltered, and frozen at -70°C until required for use in cell cultures.

Preparation of ALT-Cells (CD45RO+ T-Cells) The other portion of mononuclear cells obtained above

were suspended in CM containing 25% T3CS, cimetidine (5 X mol/l), and indomethacin (lo-' mol/l) at a concentration of 2 X lo6 cells/ml. Cimetidine and indomethacin were added to reduce tumor-related suppressor cell activity [3 1,321. The cells received 50 cGy of gamma irradiation to reduce the concentration of radiosensitive suppressor T-cells [33,34]. The cells were cultured for 5 days in a moist-air incubator that contained 5% CO,, then washed extensively in RPMI-I640 and resuspended in CM.

Immunophenotyping of PBL and ALT-Cells Patients' PBL and ALT-cells were stained with a panel

of the following monoclonal antibodies: CD3 (Leu4), CD45RO (Leu45RO), CD25 (interleukin-2 receptor, IL- 2R), and CD45RA (Leul8) (Becton-Dickenson, Moun- tain View, CA). The monoclonal antibodies used in a direct immunofluorescence technique were conjugated with fluorescein isothiocyanate (FITC) or phycoerythrin (PE). The surface antigens were analyzed by single and/or two color flow cytometric study (Profile, Coulter Electronics, Hialeah, FL) .

Tumor Target Cells Primary and metastatic STS cells for use as targets

were separated from fresh surgical specimens. Blood and necrotic tumor was removed, the tissue was minced un- der sterile conditions, and placed in a sterile specimen container containing 20 ml of collagenase and 2 mg of DNase (both from Sigma Chemical Co., St Louis, MO). These fragments were then placed into flasks with 10 ml of 0.25% trypsin in Dulbecco's phosphate-buffered saline and placed in a shaker bath at room temperature for 1 hour. The tissue fragments were then decanted into flasks containing heat-inactivated human AB serum and the trypsinization was repeated. The cells in the total suspension were then pelletted and resuspended in 50 ml of CM. A discontinuous gradient of 10 ml each of 25, 15, and 10% Percoll in CM was made, and 10 ml of the cell suspension was layered on top. The gradient was centrifuged at room temperature for 7 minutes at 25 g. The top layer consisted of debris and unwanted cells and was discarded. The bottom layer consisted of tumor cells and following collection was washed twice with Hanks bal- anced salt solution, and resuspended in CM. A trypan blue-dye exclusion test confirmed >90% viability. Cyto- logic preparations that confirmed the presence of at least 75% STS cells were used as tumor target cells.

Tumor Target Cell Treatment With CDDP Tumor target cells at a concentration of 2 X lo5

cells/ml were incubated with CDDP at a concentration of 10 kg/ml for 3 hours in a moist-air incubator that contained 5% C02, at 37"C, based on the results of previous studies [16,19]. The tubes were then centrifuged and supernatants removed. Tumor targets were washed and suspended in CM. Trypan blue-dye test confirmed >90% viability after CDDP treatment.

Preparation of NK-Cell (CD56)-Depleted ALT-Cells PBL were incubated for 30 minutes at 4°C with PE-

labeled anti-Leu19 (CD56), washed, and sorted on a FACStar plus cell sorter (Becton-Dickenson) to obtain highly purified CD56-depleted PBL. The purity of the sorted CD56+ NK-cell population exceeded 98%. The


TABLE I, Mean Percent Memory and Naive T-cell Immunophenotype of PBL and ALT-Cells (n = 15)t

T-cell source CD3+ CD3+CD25 + CD3+CD45RO+ CD3+CD45RAi

PBL 72.2 2. I ALT-cells 75.1 14.2*

30.8 68.5*

23.5 4.5*

?Patients’ PBL and ALT-cells were stained with a panel of the monoclonal antibodies CD3 (Leu4), CD45RO (Leu45RO), CD25 (interleukin-2 receptor, IL-2R), and CD45RA (Leu1 8). The monoclonal antibodies used in a direct immunofluorescence technique were conjugated with fluorescein isothiocyanate (FITC) or phycoerythrin (PE). The surface antigens were analyzed by single and/or two-color flow cytometric study. *P < 0.001 in comparison to PBL controls.

CD56-depleted PBL were then prepared as CD56-depleted ALT-cells and used as effector cells against CDDP-treated and control STS cell targets.

Depletion of CD45RO+ T-Cells From the ALT-Cell Population

The ALT-cell population was incubated for 30 minutes at 4°C with PE-labeled anti-CD45R0, washed, and sorted on a FACStar plus cell sorter (Becton-Dickenson) to obtain highly purified CD45RO+ T-cells. The purity of the sorted CD45RO+ T-cell population exceeded 98%. The CD45RO-depleted ALT-cells were then used as effector cells against CDDP-treated and control STS cell targets.

Measurement of Cytotoxicity Detection of lytic activity was peformed using a 4-hour

and 18-hour chromium 51 (51Cr) release assay. ALT- cells, CD56-depleted ALT-cells, and CD45RO-depleted ALT-cells, and PBL were added to U-bottom, 96-well plates to achieve effector to target cell ratios (E:T) of 20:1. Prior to cytotoxicity assays, the target cells were labelled with 250 pCi of Na5’Cr04 (New England Nu- clear, Boston) washed three times, and then suspended in CM at a concentration of lo5 viable cells/ml. Thereafter, target cells (104/100 p1) were added to 96-well plates. The plates were incubated for 4 or 18 hours at 37°C in 5% COz, and the culture supernatants were harvested with the Skatron harvesting system and the amount of released 51Cr determined in a Beckman Gamma 4000 counter. Maximum 51Cr release was produced by incubation of the targets with 0.1 N HC1. Spontaneous release was measured in target cells to which medium alone was added. The percent specific lysis was calculated as:

Experimental cpm-

x 100% Spontaneous cpm Maximum cpm- Percent specific lysis =

Spontaneous cpm

Measurement of IFN-y Secretion ALT-cells were derived from three patients , co-cul-

tured with various tumor target cells, and IFN-y secretion was measured. ALT-cells were plated in 2 ml of CM at 5 X lo5 cells/well in 24-well tissue culture plates. ALT- cells from each patient were cultured alone or with lo6 control or CDDP-treated tumor stimulators in 100 pl of CM per well. ALT-cells were co-cultured with autologous primary or metastatic (CDDP-treated and control) as well as allogeneic STS and non-STS tumor targets for 24 hours. Cell-free supernatants were collected after centrifugation and stored at -70°C. IFN-y measurements were performed on thawed supematants in triplicate using a commercially available RIA kit (Centecor, Malvern, PA).

Statistics The Students t-test was used to determine differences

in the immunophenotypes of PBL and ALT-cells, as well as the difference in lysis between ALT-cell populations (whole, CD45R0, or CD56 depleted) and PBL and in lysis of control group tumor targets versus CDDP-treated tumor targets. A P value < 0.05 was considered statisti- cally significant.

RESULTS Immunophenotypic Analysis of PBL and ALT-Cells

Flow cytometric analysis of patients’ PBL and ALT- cells revealed an expansion of the CD45RO+ (memory) T-cell subset with a concomitant decrease of the CD45RA+ (naive) T-cell subset (Table I) in the ALT-cell population as previously reported [ 1,28,27]. Addition- ally, the number of T-cells bearing the activation marker CD25 also increased dramatically in the ALT-cell population compared to PBL. These data demonstrate that stimulation of PBL from human TBH with T3CS results in the preferential expansion of ex vivo activated memory T-cells.

CDDP Enhancement of Tumor Lysis by PBL, ALT-Cells, and CD45RO-Depleted ALT-Cells

All determinations were made in triplicate and the data reported as the mean.

The effects of tumor cell pretreatment with CDDP and the ability of PBL, ALT-cells, and CD45RO-depleted

216 Gold et al.

TABLE 11. CDDP Enhancement of Ex Vivo Primary STS Target Lysis by PBL and ALT-Cells (n = lo)? Effector cells Tumor treatment 4-hour % lysis 18-hour % lysis

PBL ALT-cells PBL ALT-cells

None None CDDP CDDP

1.1 11.6** 2.2

31.7*

1.4 21.4**

I .7 48.3*

?Detection of lytic activity was performed using a 4-hour and 18-hour chromium 5 1 ("Cr) release assay. ALT-cells and PBL were added to U-bottom 96-well plates to achieve effector target ratios (E:T) of 20: 1. Prior to cytotoxicity assays, the primary STS target cells were labelled with 250 pCi of Na5'Cr04 (New England Nuclear), washed three times, and then suspended in CM at a concentration of lo5 viable cells/ml. Thereafter, primary STS target cells (104/100 p1) were added to 96-well plates. The plates were incubated for 4 or 18 hours at 37°C in 5% CO,, and the culture supernatants were harvested with the Skatron harvesting system and the amount of released 5'Cr determined in a Beckman Gamma 4000 counter. Maximum 51Cr release was produced by incubation of the targets with 0.1 N HCl. Spontaneous release was measured in target cells to which medium alone was added. The percent specific lysis was calculated as:

Experimental cpm-Spontaneous cpm Maximum cpm-Spontaneous cpm

Percent specific lysis = x 100%.

*P < 0.001 compared to PBL lysis of CDDP-treated tumor targets. **P < 0.001 compared to PBL lysis of control tumor targets.

TABLE 111. CDDP Enhancement of Ex Vivo Metastatic STS Lysis by PBL and ALT-Cells (n = S)?

Effector cells Tumor treatment 4-hour 70 lysis 18-hour % lysis

PBL None 1.3 I .8 ALT-cells None 3.4** 6.6** PBL CDDP 1.9 2.2 ALT-cells CDDP 14.1* 41.7*

?Detection of lytic activity was performed using a 4-hour and 18-hour chromium 51 ("Cr) release assay. ALT-cells and PBL were added to U-bottom 96-well plates to achieve effector target ratios (E:T) of 20: 1, Prior to cytotoxicity assays, the metastatic STS target cells were labelled with 250 pCi of NaS1CrO, (New England Nuclear), washed three times, and then suspended in CM at a concentration of lo5 viable cellsiml. Thereafter, metastatic STS target cells (104/100 p1) were added to 96-well plates. The plates were incubated for 4 or 18 hours at 37°C in 5% CO,, and the culture Supernatants were harvested with the Skatron harvesting system and the amount of released "Cr determined in a Beckman Gamma 4000 counter. Maximum "Cr release was produced by incubation of the targets with 0.1 N HCI. Spontaneous release was measured in target cells to which medium alone was added. The percent specific lysis was calculated as:

Experimental cpm-Spontaneous cpm Maximum cpm-Spontaneous cpm

Percent specific lysis = x 100%

*P < 0.001 compared to ALT-cell lysis of control targets. **P < 0.001 compared to PBL lysis of control target cells.

ALT-cells to lyse CDDP-treated STS cells and normal controls at an E:T of 20:l during a 4-hour and 18-hour

Cr release assay is shown in Tables 11-IV. ALT-cells were more effective at tumor lysis of the control group than PBL (Table 11). There was a marked enhancement of tumor lysis across all E:T tested when tumor cells were

pretreated with CDDP as opposed to the untreated control group. The lysis was significantly increased when ALT- cells rather than PBL were the anti-tumor effectors (Table 11). When CD45RO-depleted ALT-cells were used as effector cells, tumor target lysis was reduced to near-PBL control levels in primary and metastatic STS target cell

51


TABLE IV. No CDDP Enhancement of Ex Vivo STS Lysis by CD45RO-Depleted ALT-Cells (n = 15)*

Effector cells Tumorltreatment 4-hour % lysis 18-hour % lvsis ~~

CD45ROd,, ALT-cells Primary STS/None 1.3 1.9

CD4SROd,, ALT-cells Metastatic STSlNone 1.8 1.7 CD45ROd,, ALT-cells Metastatic STSKDDP 2.1 2.5

E:T = 20:l. *The ALT-cell population was incubated for 30 minutes at 4°C with PE-labeled anti-CD45R0, washed, and sorted on a FACStar plus cell sorter to obtain highly purified CD45RO+ T-cells. The purity of the sorted CD45RO+ T-cell population exceeded 98%. The CD45RO-depleted ALT-cells were then used as effector cells against CDDP-treated and control STS cell targets.

CD45ROd,, ALT-cells Primary STSKDDP 1.9 2.2

TABLE V. CDDP Enhancement of Ex Vivo STS Lysis by NK-Cell (CDS6)-Depleted ALT-Cells (n = 15H

Effector cells Tumoritreatment 4-hour % lysis 18-hour % lysis

CD56,,, ALT-cells Primary STSlNone 1.2 23.7** CD56,, ALT-cells Primary STS/CDDP 9.3* 53.2* CDS6,,, ALT-cells Metastatic STSlNone 2.7 7.1** CD56,,, ALT-cells Metastatic STS/CDDP 6.7* 43.3*

E:T = 20:l. tPBL were incubated for 30 minutes at 4°C with PE-labeled anti-Leu19 (CD56), washed, and sorted on a FACStar plus cell sorter to obtain highly purified CD56-depleted PBL. The purity of the sorted CD56+ NK-cell population exceeded 98%. The CDS6-depleted PBL were then prepared as CD56-depleted ALT-cells and used as effector cells against CDDP-treated and control STS cell targets. * P < 0.001 compared to PBL lysis of CDDP-treated tumor targets. **P < 0.001 compared to PBL lysis of control tumor targets.

groups (Table IV) during both the standard (4-hour) and delayed (18-hour) assay whether or not tumor targets were pretreated with CDDP.

CDDP Enhancement of Metastatic Tumor Target Lysis by PBL and ALT-Cells

As shown in Table 111, metastatic STS tumor cells were less sensitive to ex vivo lysis by both PBL and ALT-cells when compared to primary lesions. However, pretreatment of metastatic STS target cells with CDDP enhanced ex vivo lysis by ALT-cells to levels similar to that of CDDP-treated primary STS target cells.

CDDP Enhancement of Tumor Lysis by NK-Cell (CD56)-Depleted ALT-Cells

PBL were depleted of CD56+ NK-cells prior to the generation of ALT-cells to determine the relative contri- bution, if any, of NK-cells to the ex vivo lysis of autologous tumor targets by ALT-cells. As seen in Table V, ex vivo lysis of autologous tumor targets by CD56-depleted ALT-cells was reduced as measured by the 4-hour 5'Cr release assay, but was comparable to whole ALT-cell lysis at the 18-hour assay. Pretreatment of autologous tumor target cells with CDDP was able to potentiate ex vivo lysis as measured by the 5'Cr release assay at the

18-hour mark. This demonstrates that ex vivo lysis of autologous tumor targets is mediated mainly by CD45RO+ memory T-cells and that this effect can be potentiated by pre-treatment of tumor target cells with CDDP.

ALT-Cells Demonstrate Tumor Specificity ALT-cells were derived from three patients with STS

and ex vivo cytotoxicity against autologous and allogeneic STS (including PSTS-1, PSTS-4, MSTS-5, U-2 OS, and Saos-2), melanoma (HT144 and Malme-3), pancre- atic carcinoma (Capan-1), and leukemia (K562) CDDP- treated tumor targets at an E:T of 20: 1 were measured as described above. As shown in Table VI, Patient PSTS-1 (HLA-A2; leiomyosarcoma) demonstrated enhanced ex vivo lysis against autologous STS targets but not against HLA-A2+ allogeneic STS (PSTS-4), melanoma, pancre- atic carcinoma, or K562. Patient PSTS-4 (HLA-A2+ leiomyosarcoma) showed absolute autologous tumor-specificity with no cytotoxicity against other STS targets including those that were HLA-A2+. Using ALT-cells against a metastatic autologous STS, patient MSTS-5 (HLA-A2' extraosseous osteogenic sarcoma) showed cytotoxicity against autologous tumor targets with some lytic activity against Saos-2, and no lysis of melanoma,

218 Gold et al.

TABLE VI. Ex Vivo Tumor-Specificity of STS-Derived ALT-Cells (E:T = 201)*

PSTS-4b

MSTS-5

ALT-cell Tumor target source source 4-hour % lysis 18-hour % lysis IFN--y release

PSTS- I a PSTS- I 25.4 62.2 + - PSTS-4b 1.5 2.0 - MSTS-Sb 0.9 1.7

Sa0s-2~ 1.3 1.9 u-2 OSb 0.7 1.3 HT144 1.1 1.5

Malme-3 1.8 1.5 Capan- 1 0.5 1.1

K562 0.7 0.8 PSTS- 1 1.2 1.9 PSTS-4b 30.5 61.7 + MSTS-Sb 2.1 1.6 ~ a o s - 2 ~ 1.3 1.9 u-2 OSb 1.1 1.4 HT144 0.4 0.4

Malme-3 1.8 1.5 Capan- 1 0.7 1.3

K562 0.5 0.9 PSTS-1 2.1 1.9 PSTS-4 1.3 1.5 MSTS-5 21.2 57.7 + ~ a o s - 2 ~ 9.7 33.5 +

u-2 OSb 1.2 1.8 2.4 2.1 - HT144

Malme-3 0.7 1.3 - Capan- 1 0.4 0.3 -

K562 1.1 0.9 -

- - - - - - -

- - - - - - - - -

-

*ALT-cells were co-cultured with various tumor stimulators and IFN-7 secretion was measured. ALT-cells were plated in 2 ml of CM at 5 X lo5 cellsiwell in 24-well tissue culture plates. ALT-cells were cultured alone or with lo6 control or CDDP-treated tumor stimulators in 100 )*I of CM per well. ALT-cells were co-cultured with autologous primary or metastatic (CDDP-treated and control) as well as allogeneic STS and non-STS tumor targets, respectively, for 24 hours. Cell-free supernatants were collected after centrifugation and stored at -70°C. IFN--y measurements were performed on thawed supernatants in triplicate using a commercially available RIA kit (Centecor). aHLA-A2- STS . bHLA-A2+ STS.

carcinoma, leukemia, or other allogeneic STS targets. IFN--y release also followed a relative and absolute pat- tern of tumor-specificity similar to the 51Cr release assay. This demonstrates that relative and absolute ex vivo tumour specificity is possible with ALT-cells derived from patients with STS and that augmentation of lysis by CDDP preserves antitumour specificity of STS-derived ALT-cells.

DISCUSSION In addition to direct tumoricidal effects, it is well

known that many chemotherapeutic drugs possess immu- nomodulatory properties. Preclinical studies in murine TBH with hematopoietic or solid tumors have dernon- strated antitumor synergy between adoptively transferred T-cells (or T-cell subsets) and agents such as cyclophosphamide, dacarbazine, doxorubicin, and CDDP [9-211. The mechanisms of immunomodulation are diverse and depending on the drug may be secondary to inhibition of

suppressor mechanisms, or by alteration of the tumor cell itself and enhancing immunogenicity [9-15,17,18].

There have been preliminary studies of adoptive cellular therapy and chemotherapy or ACIT in human TBH. Takayama et al. [35] demonstrated enhanced therapeutic efficacy of adoptively transferred TIL when combined with mitomycin-C and doxorubicin. Rosenberg et al. [36] and Kohler et al. [37] combined CY with adoptive cellular therapy of metastatic melanoma and lymphoma, respectively. Barth et al. [38] reported on a preliminary trial of LAK-cells with sequential VP-16 and CDDP in patients with non-small cell lung cancer. In a pilot study of adoptive chernoimmunotherapy using ALT-cells and CY, Gold et al. [28] demonstrated responses in patients with a variety of relapsed solid tumors.

In ex vivo studies of CDDP and LAK-cells on human bladder cancer by Mituzani et al. [16], CDD preincubation of tumor cell targets resulted in enhanced tumor cell lysis by both LAK-cells and PBL over the non-CDDP-


treated control group. Gold et al. [19,20] were also able to demonstrate that pre-incubation of RCC tumor targets with CDDP, markedly enhanced ex vivo lysis of these tumor targets by ALT-cells. This ex vivo lysis by ALT- cells was dependent on the presence of memory T-cells as depletion of the CD45RO+ T-cell pouplation dramatically diminished ex vivo lysis [19]. Here again, we demonstrate that memory T-cells are the primary lymphocyte population in ALT-cell that contribute to the ex vivo cytotoxicity. Depletion of the CD56+ (NK)-cells had an effect on the 4-hour but not the 18-hour 5'Cr release assay. It is not known if the memory T-cell population is directly responsible for tumor lysis and/or the initiator of tumor target lysis. Finke et al. [27] have reported that separation and comparison of TCC TIL into CD4+, CD8+, and CD56+ populations revealed minimal autologous tumor target lysis by the CD4+ and CD8+ T-cells in a standard 4-hour 51Cr release assay. However, it was found that the CD3-CD56+ (NK)-cells did mediate significant lysis in a 4-hour 51Cr release assay with CD4+ and CD8+ T-cells showing substantial lysis in a delayed (18- and 72-hour) assay. It was also noted that similar to ALT-cells, the T-cells in the TIL population were CD45ROf, implying that many of these cells were memory T-cells 1391. It would then appear that memory T-cells are primarily effective on a delayed rather than immediate basis, an observation confirmed by Dye and North in murine studies [40].

We have demonstrated that preincubation with CDDP is capable of potentiating lysis of human STS targets by autologous ALT-cells (Tables 11, 111). Metastatic tumors showed less ex vivo lysis by ALT-cells than primary tumors (Tables IV, V). However, pretreatment of metastatic STS tumor target cells with CDDP enhanced ex vivo lysis to values that were similar to the CDDP-treated primary STS tumor targets. The mechanism of this synergistic effect by CDDP is not clear but may be secondary to drug-induced alteration of the tumor target cell membrane, rendering it more susceptible to effector cell attack. It has been postulated that one mechanism of development of tumor metastases may be immune-mediated escape or nonrecognition of tumor targets by host immune cells [41,42]. Indeed, an elegant study by Alex- ander et al. [43] showed that advanced but not early stage melanoma had defective antigen-presentation capability. Based on the results of this study, treatment with CDDP may: (1) restore the ability of immune cells to recognize metastatic lesions as well as the primary targets, or (2) increase tumor cell membrane sensitivity to lysis by immune cells. Interestingly, the mechanism may not be immunologic in nature as Mizutani et al. [44] have shown that surface antigen expression is not altered by CDDP. Direct cytotoxicity of CDDP on tumor targets does not appear to contribute to the synergy between ALT-cells and CDDP, as trypan blue-dye exclusion studies con-

firmed that the viability of the CDDP-treated cells were identical to the control group. In addition, as shown in Table VI, despite treatment with CDDP, tumor-specificity is preserved as ex vivo lysis and IFW-y release against irrelevant tumor targets was not demonstrated, although Mizutani et al. [44] did show that CDDP-treated and K562 target cells were more susceptible to lysis by nonspecific LAK-cells. We [21] and others [22-241 have shown that human breast, colorectal, and RCC tumors also can cause tumor-specific release of IFN-y and other cytokines by T-cells and T-cell subsets and that IFN-y release is a valuable measure of tumor-specific immune recognition by effector cells.

As previously demonstrated by Vanky et al. [45] and Slovin et al. [46], autologous T-cells are able to cause ex vivo lysis of STS targets. Balsari et al. [47] have also shown that ex vivo activation of PBL from human TBH with melanoma results in the generation of T-cells with autologous tumor specificity. Vanky et al. [48] also noted that the ability of T-cells from TBH with STS to cause ex vivo lysis of autologous tumors appears to correlate with increased survival over those TBH whose T-cells cannot lyse their tumors. Therefore, immunologic recognition of tumor targets may be an important mechanism in the prevention of metastases. In agreement with previous reports and as shown here, patients are able to generate absolute (autologous STS) or relative (autologous and allogeneic STS) tumor specificity via nonspecific ex vivo activation of PBL using T3CS from tumor-bearing patients with STS. In addition, the ex vivo lysis of STS targets by T-cells appears to depend on the use of the major histocompatibility complex (MHC) class I HLA-A2 antigen to recognize shared tumor determinants on STS [45,49]. In this report, tumors from patients PSTS-4 and MSTS-5 were HLA-A2+, but not PSTS- 1. This may explain why PBL from MSTS-5 demonstrated some lysis against PSTS-4. Down-regulation of MHC class I antigens on sarcomas may allow tumors to immu- nologically evade host defenses [41,42]. Some patients with HLA-A2+ disseminated melanomas that were suc- cessfully treated with ALT-cells and CY were able to demonstrate ex vivo lysis of allogeneic HLA-A2+ melanomas but not HLA-A2- melanomas [5].

We have previously demonstrated that ALT is active in the treatment of human TBH with metastatic melanoma and RCC [MI. In previous murine and human studies, this antitumor activity is secondary to the generation of memory T-cells as part of the ALT-cell population [ 1,2]. The data in this study has again confirmed that nonspecific ex vivo activation of PBL from TBH using T3CS results in the generation of tumor-specific ALT-cells with an increase in the CD45RO+ T-cell population and a concomitant decrease in the CD45RA+ (naive) T-cell population (Tables I, VI). Depletion of CD45RO+ cells but not NK-cells from the ALT-cell population abrogated

220 Gold et al.

the lytic effect of the ALT-cells against STS targets to a level similar to that of the patients' unstimulated PBL (Tables 11-V). In addition, the results obtained in this study suggest that CDDP may somehow alter the physical properties of tumor cells rendering them discernible and susceptible to immune-mediated attack and that the combination of ALT and CDDP may lead to effective tumor- specific chemoimmunotherapy in patients with metastatic STS. Further studies using CDDP potentiation of adoptively transferred memory T-cells in STS as well as other solid tumor types are warranted.

ACKNOWLEDGMENTS The authors thank Peter Harpel, M.D., Lloyd Mayer,

M.D., Susan F. Slovin, M.D., Ph.D., and Stephen C. Malamud, M.D., for their suggestions and comments. This work was sponsored by grants from Cellcor and the Jennifer Turner Cancer Research Foundation.

REFERENCES 1. Celis E, Bolwerk A, Clarke J, Goodwin J, Krane RJ, Oshand ME:

The immunologic mechanism of autolymphocyte therapy in the successful treatment of renal cell carcinoma (RCC) is the infusion of activated memory T-cells. J Urol 145:339A, 1991 (ahst).

2. Gold JE, Oshand ME: Autolymphocyte therapy: I. In vivo tumor- specific adoptive cellular therapy of advanced metastatic murine melanoma and carcinoma using ex vivo activated memory T-cells. Eur J Cancer 1994 30A:1871-1882, 1994.

3. Gold JE, Oshand ME: Autolymphocyte therapy: 11. Dependence of in vivo tumor specific adoptive cellular therapy against advanced metastatic murine melanoma and carcinoma on ex vivo activated donor memory T-lymphocytes. Clin Immunol Immuno- path 71:325-332, 1994.

4. Graham S , Bahayan RK, Lamm DL, Sawczuck I, Ross SD, Lavin PT, et al.: The use of ex vivo-activated memory T-cells (autolymphocyte therapy) in the treatment of metastatic renal cell carcinoma: Final results from a randomized, controlled, multisite study. Semin Urol 11:27-36, 1993.

5. Gold JE, Ross SD, Malamud SC, Oshand ME: Adoptive transfer of ex vivo activated memory T-cells with and without cyclophosphamide for advanced metastatic melanoma: Results in 36 patients. Eur J Cancer 1995 (in press).

6. Osband ME, Lavin PT, Bahayan RK, et al: Effect of autolymphocyte therapy on survival and quality of life in patients with metastatic renal cell carcinoma. Lancet 335:994998, 1990.

7. Zhang ZL, Kurtzberg K, Oshand M, Miesowicz F, Babhit B: Synergistic effects of autologous cytokines and OKT3 in the ex vivo activation of memory T lymphocytes utilized in adoptive immunotherapv of metastatic renal cell carcinoma. J Immunol 150:1247, 1993 (abst).

8. Sieeel JP. Pun RK: Interleukin-2 toxicitv. J Clin Oncol 9 : 6 9 4 704: 1991.

9. Awwad M, North RJ: Cyclophosphamide (Cy)-facilitated adoptive immunotherapy of a Cy-resistant tumour. Evidence that Cy permits the expression of adoptive T-cell mediated immunity by removing suppressor T-cells rather than reducing tumour burden. Immunology 65237-92, 1988.

10. North RJ: Cyclophosphamide facilitated adoptive immunotherapy of an established tumor depends on elimination of tumor-induced suppression of T-cells. J Exp Med 155:1063-1074, 1982.

11. Greenberg PD, Kern DE, Cheever MA: Therapy of disseminated murine leukemia with cyclophosphamide and immune Lyt 1+,2 T-cells: Tumor eradication does not require participation of cytotoxic T-cells. J Exp Med 161:1122-1134, 1985.

12. Greenberg PD: Therapy of murine leukemia with cyclophosphamide and immune Lyt-2' cells: Cytolytic T-cells can mediate eradication of disseminated leukemia. J Immunol 136:1917-1922, 1986.

13. Bookman MA, Swerdlow R, Matis LA: Adoptive chemoimmunotherapy of murine leukemia with helper T-lymphocyte clones. J Immunol 139:3 1 6 6 3 170, 1987.

14. Formelli F, Rossi C, Sensi ML, Parmian G: Potentiation of adoptive immunotherapy by cis-diamminedichloroplatinum (II), but not by doxorubicin, on a disseminated mouse lymphoma and its association with reduction of tumor burden. Int J Cancer 42:952- 957, 1988.

15. Salup RR, Back TC, Wiltrout RH: Successful treatment of advanced murine renal cell cancer by bicompartmental adoptive chemoimmunotherapy. J Immunol 138:641447, 1987.

16. Mituzani Y, Yoshinori N, Yoshida 0: Modulation by cis-diamminedichloroplatinum(1I) of the susceptih lined and freshly separated autologous urinary bladder transitional carcinoma cells to peripheral blood lymphocytes and lymphokine activated killer cells. J Urol 147:505-510, 1992.

17. Bianchi R, Romani L, Puccetti P, Fioretti MC: Induction of tumor suppression and delayed-type footpad reaction by transfer of lymphocytes sensitized to a xenogenized tumor variant. Int J Cancer

18. Puccetti P, Bianchi R, Romani L, Cenci E, Fioretti MC: Delayed- type hypersensitivity to tumor antigens co-expressed with immu- nogenic determinants induced by xenogenization. Int J Cancer

19. Gold JE, Masters TR, Oshand ME: Autolymphocyte therapy (ALT) of human renal cell carcinoma (RCC): Demonstration of antitumor cytotoxicity (ATC) by ex vivo activated memory T-cells and potentiation by cis-diamminedichloroplatinum(I1) (CDDP). Proc Am SOC Clin Oncol 12:740, 1993 (ahst).

20. Gold JE, Masters TR, Babbit B, et al: Ex vivo activated memory T-lymphocytes as adoptive cellular therapy of human renal cell tumour targets with potentiation by cis-diamminedichlorolatinu- m(I1). Br J Urol 1995 (in press).

21. Gold JE, Masters TR, Goldfarh AB, et al.: Adoptive cellular therapy of human breast and colorectal tumour targets using ex vivo activated memory T-lymphocytes with potentiation by cis- diamminedichloroplatinum(I1). J Surg Oncol 55:222-228, 1994.

22. Schwartzentruber DJ, Topalian SL, Mancini M, Rosenberg SA: Specific release of granulocyte-macrophage colony-stimulating factor, tumor necrosis factor-a, and IFN-71 by human tumor-infiltrating lymphocytes after autologous tumor stimulation. J Immu- no1 146:3672-3681, 1991.

23. Schwartzentruher DJ, Solomon D, Rosenberg SA, Topalian SL: Characterization of lymphocytes infiltrating human breast cancer: Specific immune reactivity detected by measuring cytokine secretion. J Immunother 12:l-12, 1992.

24. Finke JH, Rayman P, Hart L, et al.: Characterization of tumor- infiltrating lymphocyte subsets from human renal cell carcinoma: Specific reactivity defined by cytotoxicity, interferon-g secretion, and proliferation. J Immunother 15:91-104, 1994.

25. Vanky F, Willems J, Kreicbergs A, et al.: Correlation between lymphocyte-mediated auto-tumor reactivities and clinical course. I. Evaluation of 46 patients with sarcoma. Cancer Immunol Immu- nother 16:ll-16, 1983.

26. Vanky F, Klein E, Willems J, et al.: Lysis of autologous tumor cells by blood lymphocytes tested at the time of surgery: Correla- tion with the postsurgical clinical course. Cancer Immunol Immu- nother 21:69-76, 1986.

27. Finke JH, Rayman P, Alexander J, et al.: Characterization of the cytolytic activity of CD4+ and CD8+ tumor-infiltrating lymphocytes in human renal cell carcinoma. Cancer Res 50:2363-2370, 1990.

28. Gold JE, Malamud SC, LaRosa F, Seder R, Oshand ME: Adop- tive chemoimmunotherapy of relapsed and refractory solid tumors using ex vivo activated memory T-cells (autolymphocyte therapy) and cyclophosphamide. J Immunother 13:213-21, 1993.

29. Meuer SC, Hussey RE, Cantrell DA, Hodgson JC, Schlossman SF, Smith KA, Reinherz EL: Triggering of the T3-antigen recep-

42~71-75, 1988.

43:279-284, 1989.


tor complex results in clonal T-cell proliferation through an interleukin-2 dependent autocrine pathway. Proc Natl Acad Sci USA

30. Ashwell JD, Longo DL, Bridges SH: T-cell tumor elimination as a result of T-cell receptor-mediated activation. Science 237:6144, 1987.

31. Khan MM, Sansoni P, Engleman EG, Melmon KL: Pharmaco- logic efefcts of autocoids on subsets of T-cells: Regulation of expressiodfunction of histamine-2 receptors by a subset of suppressor cells. J Clin Invest 75:1578-1583, 1985.

32. Waymack JP, Guzman RF, Burleson DG, McManus AT, Mason AD, Pruitt BA: Effect of prostaglandin E in multiple experimental models. Prostaglandins 38:345-353, 1989.

33. Wasserman J , Petrini B, Blomgren H: Radiosensitivity of T-lymphocyte subpopulations. J Clin Lab Immunol7: 139-140, 1982.

34. Hersh EM, Patt YZ, Murphy SG, Dicke K, Zander A, Adegbite M, Goldman R: Radiosensitive, thymic-hormone sensitive peripheral blood suppressor cell activity in cancer patients. Cancer Res 40: 3 1 34-3 140, 1 980.

35. Takayama T, Makuuchi M, Sekin T, et al.: Distribution and therapeutic effect of intraartenally transferred tumor-infiltrating in hepatic malignancies. A preliminary report. Cancer 68:2391- 2396, 1991.

36. Kohler PC, Hank JA, Exten R, Minkoff DZ, Wilson DG, Sonde1 PM: Clinical response of a patient with diffuse histiocytic lymphoma to adoptive chemoimmunotherapy using cyclophosphamide and alloactivated haploidentical lymphocytes: A case report and phase I trial. Cancer 55:552-560, 1985.

37. Rosenberg SA, Packard BS, Aebersol PM, et a].: Use of tumor- infiltrating lymphocytes and interleukin-2 in the immunotherapy of patients with metastatic melanoma. N Engl J Med 319:167& 1680, 1988.

38. Barth N, West W, Oldham R, Hoogstarten B, Dillman R: A phase I1 trial of adoptive immunotherapy (AIT) and sequential chemotherapy with cisplatin, VP-16 in advanced non-small cell lung cancer. Proc Am SOC Clin Oncol9:910, 1990 (abst).

39. Bukowski RM, Sharfman W, Murthy S, et al.: Clinical results and characerization of tumor-infiltrating lymphocytes with or without

8 1 : 1509-15 16, 1984.

recombinant interleukin 2 in human metastatic renal cell carcinoma. Cancer Res 51:4199-4205, 1991.

40. Dye ES, North RJ: Adoptive immunization against an established tumor with cytolytic versus memory T-cells: Immediate versus delayed onset of regression. Transplantation 37:60&605, 1984.

41. Wang Z, Cao Y, Houghton AN, Ferrone S: Molecular ahnonnali- ties in the expression of HLA class-I antigens by melanoma cells. Int J Cancer S6:lOl-105, 1991.

42. Wang P, Vanky F, Li SL, Vegh Zs., Persson U, Klein E. Expres- sion of MHC-class-I antigens in human carcinomas and sarcomas analyzed by isoelectric focusing. Int J Cancer S6:10&116, 1991.

43. Alexander MA, Bennicelli J, Gueny D: Defective antigen presentation by human melanoma cell lines cultured from advanced but not biologically early, disease. J Immunol 142:4070-4078, 1989.

44. Mizutani Y, Bonavida B, Yoshinori N , Yoshida 0: Enhanced susceptibility of cis-diamminedichloroplatinum-treated K652 cells to lysis by peripheral blood lymphocytes and lymphokine activated killer cells. Cancer 71:1313-1322, 1993.

45. Slovin SF, Lackman RD, Ferrone S, Kiely PE, Mastrangelo MJ: Cellular immune response to human sarcomas: Cytotoxic T cell clones reactive with autologous sarcomas. I. Development, phe- notype, and specificity. J Immunol 137:3042-3048, 1986.

46. Vanky F, Willems J, Kreicbergs A, et al.: Correlation between lymphocyte-mediated auto-tumor reactivities and clinical course. I. Evaluation of 46 patients with sarcoma. Cancer Immunol Immu- nother 16:ll-16, 1983.

47. Balsari A, Fossati G, Taramelli D, Tona G, Delia D, Giardini R, Parmiani G: Allostimulation of patients' lymphocytes generates both T and NK-like cells cytotoxic for autologous melanoma. Br J Cancer 552:73-80, 1985.

48. Vanky F, Klein E, Willems J, et al.: Lysis of autologous tumor cells by blood lymphocytes tested at the time of surgery: Correla- tion with the postsurgical clinical course. Cancer Immunol Immu- nother 21:6%76, 1986.

49. Topalian SL, Hom SS, Kawakami Y, Mancini M, Schwartzentn- ber DJ, Zakut R, Rosenberg SA: Recognition of shared melanoma antigens by human tumor-infiltrating lymphocytes. J lmmunother 12:203-206, 1992.

Documents

Ex vivo activated memory t-lymphocytes as adoptive cellular therapy of human soft-tissue sarcoma targets with potentiation byCis-diamminedichloroplatinum (II)