Department of Medical Oncology, Subdivision of Clinical and Tumor Immunology, Erasmus Medical Center-Daniel, Rotterdam, The Netherlands

Correspondence to: Dr Cor H J Lamers, Department of Medical Oncology, Subdivision of Clinical and Tumor Immunology, Erasmus Medical Center-Daniel, PO Box 5201, 3008 AE Rotterdam, The Netherlands. E-mail: lamers@immh.azr.nl

In preparation of a clinical phase I/II study in renal cell carcinoma (RCC) patients, we developed a clinically applicable protocol that meets good clinical practice (GCP) criteria regarding the gene transduction and expansion of primary human T lymphocytes. We previously designed a transgene that encodes a single chain (sc) FvG250 antibody chimeric receptor (ch-Rec), specific for a RCC tumor-associated antigen (TAA), and that genetically programs human T lymphocytes with RCC immune specificity. Here we describe the conditions for activation, gene transduction, and proliferation for primary human T lymphocytes to yield: (a) optimal functional expression of the transgene; (b) ch-Rec-mediated cytokine production, and (c) cytolysis of G250-TAA^POS RCC by the T-lymphocyte transductants. Moreover, these parameters were tested at clinical scale, i.e., yielding up to 5-10´10⁹ T-cell transductants, defined as the treatment dose according to our clinical protocol. The following parameters were, for the first time, tested in an interactive way: (1) media compositions for production of virus by the stable PG13 packaging cell; (2) T-lymphocyte activation conditions and reagents (anti-CD3 mAb; anti-CD3+anti-CD28 mAbs; and PHA); (3) kinetics of T-lymphocyte activation prior to gene transduction; (4) (i) T-lymphocyte density, and (ii) volume of virus-containing supernatant per surface unit during gene transduction; and (5) medium composition for T-lymphocyte maintenance (i) in-between gene transduction cycles, and (ii) during in vitro T-lymphocyte expansion. Critical to gene transduction of human T lymphocytes at clinical scale appeared to be the use of the fibronectin fragment CH-296 (RetronectinÔ) as well as LifecellÒ X-foldÔ cell culture bags. In order to comply with GCP requirements, we used: (a) bovine serum-free human T-lymphocyte transduction system, i.e., media supplemented with autologous patients' plasma, and (b) a closed cell culture system for all lymphocyte processing. This clinical protocol routinely yields 30-65% scFvG250 ch-Rec^POS T lymphocytes in both healthy donors and RCC patients. Cancer Gene Therapy (2002) 9, 613-623 doi:10.1038/sj.cgt.7700477

clinical protocol; immunogene therapy; human T lymphocytes; single chain chimeric receptor; renal cell carcinoma

The successful clinical adoptive transfer of tumor-specific cytotoxic T lymphocytes (CTLs) for cancer treatment remains hampered by the difficulty to reproducibly isolate and expand such human T lymphocytes in vitro or in vivo.^1,2 In contrast, a library of monoclonal antibodies (mAbs) specific for tumor-associated antigens (TAA) is available. The tumor cell killing potential of any CTL canbetargeted to these TAA, when sensitized with bi-specificmAbs specific for both T lymphocyte and TAA.³ We and others^4,5,6,7,8 have indeed used such bi-specific mAbs-sensitized human T lymphocytes in clinical studies. In spite of objective clinical responses, its use is complicated by (a) limited accessibility of solid tumors to (bi-specific) mAb;⁹ (b) dissociation of bi-specific mAb from CTL;¹⁰ limited recycling capacity of those human T lymphocytes; (c) development of human anti-mouse antibody (HAMA) responses;¹¹ and (d) inactivation of the bi-specific mAb-redirected T lymphocytes following tumor target antigen binding.^3,10 Moreover, the clinical anti-tumor effects were locoregional, not systemic.⁸

Recently, human T lymphocytes can also be genetically programmed with mAb-dictated specificities.^12,13,14 We have chosen for renal cell carcinoma (RCC) patients because no adequate therapy is available for these patients.^15,16 TheG250 mAb recognizes the G250 TAA expressed on the cell membrane of over 90% of primary RCC and over 80% of RCC metastases,^17,18 and is RCC-selective: its ligand is not expressed on the membrane of normal renal cells (or cells derived from other tissues) in a density sufficient to elicit cellular or humoral immune responses.¹⁷ Moreover, the G250 mAb has extensively been tested in clinical studies.^18,19,20 We earlier reported on the genetic programming of human T lymphocytes yielding RCC killing human T cells following retroviral transduction with the transgene encoding for the single chain (sc) G250mAb (scFvG250) ch-Rec.^21,22,23,24 Further, studies from our laboratory have shown that scFvG250-CD4/^POS T lymphocytes that selectively and effectively kill G250-TAA^POS RCC also are specifically triggered by G250-TAA^POS RCC to produce cytokines. Moreover, the T-cell transductants can recycle their tumor cell killing capacity.²¹ Because the ch-Rec is mAb-based, TAA recognition and cytolysis mediated by scFvG250^POS Tlymphocytes are MHC-unrestricted. The scFvG250 ch-Rec does act in context with accessory molecules, such as CD11a/CD18, to allow lysis of even "low TAA^POS" tumor cells.^23,24

We succeeded to optimize the conditions for human T-lymphocyte activation, gene transduction, and expansion to yield: (a) optimal functional expression of the scFvG250-CD4/ transgene; (b) ch-Rec-mediated cytokine production; and (c) cytolysis of G250-TAA^POS RCC. Moreover, all these parameters were also tested at clinical scale, i.e., yielding up to 5-10´10⁹ T-cell transductants (>50-fold expansion in ±14 days), defined as the treatment dose in our clinical protocol for immunogene therapy of metastatic RCC patients. The defined cell processing protocols meet good clinical practice (GCP) criteria, and have recently been approved by the Dutch Regulatory Health Authorities.

Materials and methods

Cells and antibodies

The packaging cell line PG13²⁵ (ATCC, Rockville, MD) was cultured in DMEM medium containing 10% bovine calf serum (BCS; Hyclone Laboratories, Logan, UT), 2 mM l-glutamine, 1% MEM nonessential amino acids, 100U/mL penicillin, and 100 g/mL streptomycin (complete DMEM). Media and medium supplements were purchased from Biowhittaker (Verviers, Belgium), unless otherwise indicated. The pStitch-scFvG250-CD4/ construct was made as described²² and used for stable gene transduction into the PG13 cell line. The scFvG250-CD4/ ch-Rec^POS PG13 clone 1.2 was derived by limiting dilution of the transduced bulk population and routinely grown in DMEM medium supplemented with 10% "clinical grade" fetal calf serum (FCS; Summit Biotechnology, Ft. Collins, CO). The following cell lines were used as targets in cytotoxicity assays and lymphokine production assays: (a) the G250-TAA^POS RCC transfectant SKRC-17 MW1-clone 4 and (b) the G250-TAA^NEG RCC SKRC-17 PBJ-clone 1 (mock-transfectant), kindly provided by Dr E Oosterwijk (Nijmegen, the Netherlands); (c) the Daudi Burkitt's lymphoma cell line and (d) the OKT3 hybridoma cell line. Dr SO Warnaar (Wilex, München, Germany) kindly provided the renal TAA-specific mAb G250. Dr E Oosterwijk (Nijmegen, the Netherlands) kindly provided G250 anti-idiotype mAb NUH-82. Anti-CD3 mAb OKT3 (Orthoclone) was obtained from Cilag (Beerse, Belgium), anti-CD28 mAb (clone CLB-CD28/1) from the CLB (Amsterdam, the Netherlands), anti-CD8 mAb conjugated to phyco-erythrin (PE) from DAKO (Glostrup, Denmark), goat antimouse (GAM) IgG₁/PE from Southern Biotechnology (Birmingham, AL), and anti-CD4 mAb/FITC and isotype control conjugates mIgG₁/FITC and mIgG₁/PE from BDIS (San Jose, CA).

Test media for retrovirus production, T-cell activation, T-cell maintenance in-between gene transduction cycles, and T-cell expansion

Various culture media and media supplements were tested for: (a) optimal production of retroviruses by the PG13 clone 1.2; (b) optimal T-lymphocyte activation for retroviral gene transduction; (c) T-lymphocyte maintenance in-between gene transduction cycles; and (d) posttransduction expansion of transduced T lymphocytes. The media tested were: (1) DMEM; (2) RPMI 1640; (3) AIM V (RPMI 1640-based; Life Technologies, Grand Island, NY); (4) Mix-Med (i.e., 80% RPMI 1640 and 20% AIM V),²⁶ and (5) X-VIVO 15 (DMEM-based). Serum supplements used were: (i) pooled AB serum (five donors); (ii) autologous or pooled human AB plasma; (iii) "clinical grade" FCS, and (iv) human serum albumin (HSA; CLB, Amsterdam, the Netherlands), and these were tested at final percentages ranging from 2 to 10. Other medium supplements used were HEPES (25 mM), heparin (2 IU/mL; Leo, Ballerup, Denmark), glutamine (2 mM), penicillin (100 U/mL), and streptomycin (100 g/mL). All T-lymphocyte expansion media were supplemented with 360 IU/mL recombinant interleukin 2 (IL-2; Chiron, Amsterdam, the Netherlands).

Isolation and activation procedures of human peripheral blood mononuclear cells (PBMCs)

PBMCs from healthy donors and RCC patients were isolated by Ficoll-Isopaque density centrifugation (density=1.077 g/cm³; Nycomed Pharma, Oslo, Norway) and cultured in Mix-Med culture medium supplemented with 2 IU/mL heparin, 2% human plasma, glutamine, and streptomycin (complete Mix-Med medium).²⁶ PBMCs were then activated at a density of 0.5-1.0´10⁶ cells/mL either with: (a) 10 ng/mL anti-CD3 mAb OKT3, (b) immobilized anti-CD3 mAb and anti-CD28 mAb, or (c) 1 g/mL phytohemagglutinin (PHA, HA16; Welcome, Dartford, UK). To immobilize mAbs, six-well non-tissue culture-treated plates (Falcon, Franklin Lakes, NJ) were coated with a mixture of anti-CD3 and CD28 mAbs of 4 g each in 4 mL of PBS per well, for 2 hours at 37°C. The mAb-coated plates were then saturated with 2% HSA in PBS for 20 minutes at 37°C, rinsed once with PBS, and then used for T-lymphocyte activation purposes. Tissue culture-treated plates or flasks (Greiner, Frickenhausen, Germany) were used for all T-lymphocyte activation procedures that did not involve immobilized antibodies.

Human T-lymphocyte activation and expansion: meeting GCP criteria

To meet GCP requirements, we evaluated the LifecellÒ X-foldÔ cell culture containers (Nexell Therapeutics, Deerfield, IL) providing a closed cell culture system to replace 24-well plates and tissue culture flasks. The cell culture containers (surface area range: 85-600 cm², maximum fill volume range: 65-1200 mL) are polyolefin gas-permeable bags with a protein-coatable polystyrene inner surface layer. The bags comprise an InterlinkÒ Injection Site for safe needleless access, a Sterile Connection Device (SCD; Terumo Medical, Elkton, MD)-compatible integral tubing set with spike and Luer Lock adapters, and two sampling pouches. The bags were gifts from Dr Lamond (Nexell International, Brussels, Belgium).

Retroviral transduction of the scFvG250-CD4/ transgene into primary human T lymphocytes followed by in vitro expansion of T-lymphocyte transductants

For polybrene-mediated gene transductions, activated Tlymphocytes were seeded at 10⁶ cells/well in tissue culture-treated 24-well cluster plates in 250 L of complete Mix-Med medium.²⁶ Subsequently, 750 L of RTVsup, supplemented with IL-2 (100 IU/mL final) and polybrene (8 L/mL final concentration; Sigma, St. Louis, MO), was added to the wells. After 24 hours of incubation at 37°C and 5% CO₂, the 750 L of medium was replaced with fresh RTVsup, or medium was refreshed with complete Mix-Med culture medium both supplemented with IL-2.

For retronectin-mediated gene transductions, we used recombinant fibronectin fragments CH-296 (RetronectinÔ). Takara Shuzo (Otsu, Shiga, Japan) kindly provided RetronectinÔ for the development of our clinical study. Non-tissue culture 24-well plates and LifecellÒ X-foldÔ cell culture bags were coated with 24 g/mL RetronectinÔ at a saturating concentration of 6 g/cm². Plates and bags were coated overnight at 4°C, subsequently blocked with 2% HSA for 30 minutes at 37°C, and washed once with PBS before use. Activated T lymphocytes were washed, resuspended in RTVsup supplemented with 100 IU/mL IL-2, and plated on RetronectinÔ-coated 24-well plates or cell culture bags. Gene transductions were performed using cell concentrations varying from 0.25 to 1.0´10⁶ cells/cm² and RTVsup volumes of 0.25-1.0 mL/cm². Plates and bags were incubated at 37°C and 5% CO₂ for 6 hours. Next, the transduction medium was replenished or completely replaced with various types of medium, again supplemented with 100 IU/mL IL-2; in-between gene transductions, Tlymphocytes are maintained in a volume of 750 L/cm². The following day, a second, identical cycle of gene transduction was performed.

Human T-lymphocyte transductants were then expanded by seeding at 0.5´10⁶ cells/mL in various test media. Lymphocytes were counted every 2-3 days and adjusted to 0.5´10⁶ cells/mL by adding fresh culture medium until day21.

Flow cytometric analysis of human gene-transduced T lymphocytes

The membrane expression of the scFvG250-CD4/ receptor on the PG13 clone 1.2 and on transduced T lymphocytes was determined by indirect immune fluorescence as described.²¹ In short, cells were stained with the G250 anti-idiotype mAb NUH-82, followed by GAM IgG₁/PE (Southern Biotechnology). Incubation of cells with second step only served asanegative control. Presence of CD4^POS and CD8^POS Tlymphocytes was analyzed using a mixture of anti-CD4 mAb/FITC and anti-CD8 mAb/PE. Mouse IgG₁/FITC and mIgG₁/PE served as isotype controls.

Cytotoxicity assay

Cytotoxic activity of transduced T lymphocytes was routinely measured in our standard 4-hour ⁵¹chromium release assay,²⁷ at effector cell/target cell ratios (E/T) of 60, 40, 20, 10, 5, and 2.5: 1. In one experiment (Fig 3), the duration of the cytotoxicity assay was only 3 hours. Percentage specific cytolysis, i.e., specific ⁵¹Cr release, was calculated as follows: [(test counts-spontaneous counts)/(maximum counts-spontaneous counts)]´100%, with spontaneous and maximum ⁵¹Cr counts reflecting lysis of targets cells in medium only and detergent-induced lysis of target cells, respectively. In experiments aimed at specific blocking cytolytic activity, anti-G250 mAb (25 g/mL final concentration) was added to the targets cells 15 minutes before addition to the T-lymphocyte effector cells.

Measurement of cytokine production

To determine cytokine production by T-lymphocyte transductants by antigen specific stimulation, 6´10⁴ (non) transduced T lymphocytes were cultured for 24 hours in the presence of either 2´10⁴ adherent G250^POS or G250^NEG RCC stimulator cells in 0.2 mL of RPMI culture medium containing 360 IU/mL rIL-2, in 96-well plates. Plates were then centrifuged for 5 minutes at 1500 rpm, supernatant harvested, and levels of TNF- and IFN- measured by ELISA according to suppliers' specifications (CLB).

Results

Large-scale and stable production of retrovirusescontaining the scFvG250-CD4/ transgene meeting GCP criteria

Generation of a stable packaging PG13 clone 1.2: To establish stable, virus-producing cells, we selected the PG13 cell line because it allows production of MoMuLV particles with the Gibbon Ape Leukemia Virus (GaLV) envelope. These GaLV pseudotyped virus particles efficiently infect human T lymphocytes due to high levels of the GaLV envelope receptor on the lymphocyte membrane.²² The stable virus-producing PG13 scFvG250-CD4/ cell line was generated as described in the Materials and Methods section. By limiting dilution culture, the PG13 clone 1.2 was established, which stably expresses the scFvG250-CD4/ ch-Rec (Fig 1). Southern blot analysis confirmed stable integration of a single transgene copy per cell over a period of at least 50 passages (±6 months) of the clone.

Production of retrovirus comprising the scFvG250-CD4/ transgene: With the aim of clinical application, we chose bovine serum-free medium for the production of retrovirus-containing culture supernatant (RTVsup). Our final selection of virus production medium composition and virus production conditions at clinical scale was based on levels of functional transgene expression by human T lymphocytes, i.e., percentage scFvG250 ch-Rec^POS T lymphocytes and levels of immune specific ch-Rec-mediated cytotoxicity as well as lymphokine production.

Parameters tested are:

1. type of supporting medium: (a) DMEM, (b) RPMI 1640, and (c) synthetic serum-free media: AIM V as well as X-VIVO 15;
2. medium supplements: (a) 10% human serum, (b) 10% clinical grade FCS, and (c) 2% HSA;
3. temperature during virus production: 32°C vs 37°C;
4. RTVsup production volume per square surface: (a) 0.1 mL/cm², (b) 0.125 mL/cm², and (c) 0.15 mL/cm²;
5. RTVsup production time: (a) 24 hours, (b) 48 hours, and (c) 72 hours;
6. number of sequential RTVsup harvests: (a) one, (b) two and three harvests; and
7. type of production container: (a) Petri dish and (b) 162-cm² tissue culture flasks.

The RTVsup with the highest functional gene transduction efficiency of primary human T lymphocytes was obtained using RPMI 1640 supplemented with either 2% HSA or 10% FCS at 32°C (Fig 2). RTVsup produced in synthetic media, either serum-free or supplemented with FCS or HSA, showed two to eight times lower transduction efficiencies than those of RTVsup produced in RPMI medium (data not shown). RPMI/FCS-based RTVsup was less affected by an increase of the RTVsup production temperature from 32°C to 37°C than RPMI/HSA-based RTVsup (Fig 2). For the other RTVsup production parameters tested, the optimal conditions were as follows: (a) production volume: 0.125 mL medium/cm²; (b) production time: 24 hours; (c) number of sequential RTVsup harvests: single harvest; and (d) type of production container: 162-cm² tissue culture flasks (data notshown).

RTVsup produced according to the optimized clinical scale production conditions was used for optimizing the clinical scale T-lymphocyte gene transduction conditions (see below).

Gene transduction and expansion of primary human T lymphocytes meeting GCP criteria

Efficacy of FibronectinÔ versus polybrene in facilitating gene transduction of primary human T lymphocytes: Table 1 summarizes the results obtained in a large series of independent RTVsup gene transductions of primary human Tlymphocytes testing polybrene versus fibronectin fragment CH-296 (RetronectinÔ) as gene transduction-facilitating reagent. Median gene transduction efficiencies, determined as percentage scFvG250 ch-Rec^POS lymphocytes in flow cytometry, were 40% (range 23-67%) vs 8% (range 3-26%) for RetronectinÔ and polybrene, respectively (Table 1). The use of RetronectinÔ relative to polybrene not only resulted in a 5-fold increase of the percentage of ch-Rec^POS T lymphocytes, but also in a 2-fold increase inthe scFvG250-mediated RCC cytolytic capacity of the T-cell transductants.

Efficiencies of soluble anti-CD3 mAb versus immobilized anti-CD3+anti-CD28 mAb, and PHA to activate T lymphocytes prior to gene transduction

In a prior clinical trial, using adoptive transfer of bi-specific mAb-retargeted, autologous T lymphocytes, we used PHA for T-lymphocyte activation.^6,26 Other studies have used and compared either soluble anti-CD3 (sCD3) mAb, immobilized anti-CD3+anti-CD28 mAbs (iCD3/iCD28 mAbs), or PHA to activate lymphocytes prior to gene transduction.²⁸ Here we compared the efficiencies of T-cell activation by (a) sCD3 mAb, (b) iCD3/iCD28 mAbs, and (c) PHA, prior to gene transduction, yielding optimal transgene expression and ch-Rec-mediated immune-specific lysis of G250-TAA^POS RCC. The efficiency of RetronectinÔ-mediated T-lymphocyte transduction, performed at days 2 and 3 postactivation, was not dependent on the T-lymphocyte-activating reagent. Median gene transduction efficiencies ranged from 44% to 48% of scFvG250 ch-Rec^POS T cells for sCD3 mAb, iCD3/iCD28 mAbs, or PHA-activated T-cell transductants. However, relative to sCD3 mAb-activated T-cell transductants, the iCD3/iCD28 mAb-activated T-cell transductants tended to have a slightly lower level of G250-TAA-specific cytotoxic activity. Moreover, PHA-activated T-cell transductants showed higher levels of nonspecific activated kill activity when compared to sCD3 mAb- and iCD3/iCD28 mAb-activated T-cell transductants, respectively (Fig 3A).

Kinetics of the cytolytic potential of T-cell transductants was monitored in a reversed antibody-dependent cellular cytotoxicity (rev-ADCC) test, i.e., using the OKT3 hybridoma cell line as target cell as previously described.²⁶ Of note, the cytolytic potential of sCD3 mAb-activated scFvG250 ch-Rec^POS T lymphocytes increased faster than that of PHA or iCD3/iCD28 mAb-activated T-cell transductants, reaching a plateau from day 11 onwards following activation, in linewith previous findings for nongene-transduced Tcells.²⁶

Expansion kinetics of T lymphocytes was higher for iCD3/iCD28 versus sCD3 or PHA T-cell activation (Fig 6A). At day 7 following activation, >97% of the Tcells in culture was CD3^POS. The CD4/CD8 ratio of both the gene-transduced and nongene-transduced T cells declined more rapidly following sCD3 activation versus iCD3/iCD28 or PHA T-cell activation: the CD4/CD8 ratios for sCD3 mAb versus iCD3+CD28 mAbs, or PHA activated T-cell transductants was:

1. at culture day 11: 0.42, 1.0, and 0.90 respectively; and
2. at culture day 18: 0.17, 0.33, and 0.37 respectively.

Soluble CD3 mAb was chosen to activate T lymphocytes prior to gene transduction.

Kinetics of T-lymphocyte activation prior to gene transduction: Soluble-CD3 mAb-activated T lymphocytes were transduced with the scFvG250-CD4/ transgene for 1 or 2 days starting at day 1, 2, 3, or 4 following T-lymphocyte activation. T-cell transductions were performed with RTVsup for 6 hours on RetronectinÔ-coated plates. Lymphocytes transduced at days 2 and 3 after T-lymphocyte activation yielded the highest (46%) proportion of scFvG250 ch-Rec^POS T cells as well as the highest level of G250-TAA-specific renal tumor cell lysis (see Fig 3B). One transduction cycle yielded about 30%, two transduction cycles about 45%, and three transduction cycles about 48% of scFvG250 ch-Rec^POS T lymphocytes. Hence, the first transduction cycle accounted for about 70%, and the second for about another 30% T-lymphocyte transductants, and hence virtually no further increase during the third cycle. Two cycles of gene transduction were chosen as the standard.

RTVsup volume and lymphocyte density per surface unit affect gene transduction efficiency of human T lymphocytes: GLP criteria and cost-effectiveness favor minimization of the volume of RTVsup required for optimal gene transduction, and maximization of T-lymphocyte density per transduction cycle. We therefore titrated back the volume of RTVsup from 1.0 to 0.25 mL/cm², and titrated up the number of human Tlymphocytes from 0.25´10⁶ to 0.75´10⁶ cm². The minimal volume of RTVsup and the maximal number of Tlymphocytes that can be used per square centimeters without lowering gene transduction efficiencies (percentage scFvG250 ch-Rec^POS T lymphocytes) are 0.25 mL of RTVsup and 0.5´10⁶ T lymphocytes/cm², respectively (Fig 4).

Composition of medium for T lymphocytes between the two gene transduction cycles: To select for the optimal gene transduction protocol, we tested different media compositions at the different phases of T-lymphocyte activation/gene transduction and expansion because we speculated that optimal gene transduction may require medium supplements different from those required for optimal T-lymphocyte activation and expansion, as defined previously.²⁶ Complete Mix-Med medium,²⁶ containing 2% human plasma and 2 U/mL heparin and used to expand T lymphocytes for clinical use, was therefore also tested during the distinct phases, i.e., (a) T-lymphocyte expansion and (b) gene transductions. This approach was based on the fact that RetronectinÔ contains a heparin-binding domain that binds to retroviruses,²⁹ and hence, a heparin-containing medium such as Mix-Med, to be added following the 6-hour transduction period, was expected to negatively interfere with gene transduction efficiency of the second transduction cycle.

Indeed, the gene transduction efficiency appeared highest for T lymphocytes incubated with RPMI 1640+8% human serum without heparin following the 6-hour transduction period, and expectedly lowest in lymphocytes incubated in heparin-containing medium (see Fig 5A). When compared with RPMI 1640+8% human serum, the synthetic serum-free media AIM V and X-VIVO 15 performed less, the average percent ch-Rec^POS T lymphocytes being, i.e., 54%, 42%, and 31%, respectively (Fig 5A). As stated before, the medium selection was based on the operational criteria of scFvG250 ch-Rec membrane expression and RCC-specific immune reactivity of T-cell transductants. The levels of G250-TAA-specific cytolysis positively correlated with the percent ch-Rec^POS T lymphocytes (Fig 5B). Autologous serum and pooled serum (from five donors) as RPMI 1640 medium supplement yielded similar results. Replenishing medium versus replacing the medium in-between the gene transductions also made no difference. The latter finding is important because replenishment of medium relative to replacement of medium reduces the number of handlings during cell processing.

Selection of T-lymphocyte expansion medium following gene transduction: As for nontransduced T lymphocytes,²⁶ Mix-Med medium, which is supplemented with 2% autologous human plasma and 2 U/mL heparin, also proved optimal for expansion of human T-lymphocyte transductants. Mix-Med performed equal to RPMI 1640 medium plus 10% FCS regarding T-lymphocyte proliferation kinetics. Replacing FCS by human serum provides no alternative, as Mix-Med was superior to RPMI 1640 medium plus 10% human serum and superior to the synthetic serum-free media AIM-V as well as X-VIVO 15 (Fig 6B). Using Mix-Med at day 14 following activation, T-lymphocyte transductant cell numbers had increased from 1´10⁸ to 5-10´10⁹ (>50-fold expansion), defined as the treatment dose in our adoptive immune (gene) therapy protocol. At that time, the scFvG250 ch-Rec^POS T-lymphocyte transductants have high RCC-specific immune reactivity, i.e., cytolytic and cytokine production capacities (see Table 2).

Taken together, Mix-Med supports optimally the T-lymphocyte proliferation kinetics and meets the GCP criteria; hence, Mix-Med is the medium of choice.

Gene transduction and expansion of T lymphocytes meeting GCP: We have further developed a closed system cell culture technology throughout all phases of T-lymphocyte processing, i.e., during activation, gene transduction, and expansion of T lymphocytes using LifecellÒ X-foldÔ cell culture containers, which have a coatable polystyrene inner surface, e.g., for mAbs and/or RetronectinÔ. We, in parallel, compared this closed cell culture system with cluster wells/T flasks.

Results of 12 independent experiments are summarized in Figure 7 and show comparable transduction efficiencies, and hence, gene transduction is independent of the geometry of the system used, i.e., bags or plates. Median percentages scFvG250 ch-Rec^POS T lymphocytes in bags were 27% (range 14-34), comparable with 30% (range 13-38%) inplates.

Functional gene transduction efficacies for RCC patient and healthy donor-derived T lymphocytes

Because we and others^30,31 reported that patients with advanced cancer may have compromised immune functions, the scFvG250 ch-Rec gene transduction into the RCC patient T lymphocytes may result in non- or decreased scFvG250 ch-Rec-mediated T-lymphocyte functioning. Cryopreserved PBL, selected from eight nephrectomized untreated metastatic RCC patients, complying with the entry criteria of the designed clinical protocol, and from two healthy donors were transduced in parallel with the scFvG250 ch-Rec transgene. Results show equal transduction efficiencies and RCC-specific immune functions for RCC patient and healthy donor T lymphocytes (Fig 8)

Discussion

We and others have shown that primary human and mouse Tlymphocytes can be genetically programmed via retroviral gene transduction to exert predefined antigenic immune specificities and functions. The transgenes constructed and used were either mAb-based,^{12,13,14,21,32,33,34,35} full-length two-chain TCR-based,^36,37,38 and, only recently sc, chimeric-TCR based.³⁷

For our immune gene treatment protocol, we used our RCC-specific scFv-Ig ch-Rec (derived from the G250 mAb¹⁸), and showed immune-specific scFvG250 ch-Rec-mediated cytokine production and cytolytic capacities by the primary human T-lymphocyte transductants.^21,22,23 The ch-Rec mAb-based, as well as TCR-based, T-lymphocyte transductants show effective anti-cancer activities in mouse model systems,^32,33,34 as well as the capacity to bypass T-cell tolerance.³⁹ In view of these data and of the reported clinical efficacy of virus and tumor-specific T lymphocytes in patients, this calls for a clinical protocol that allows efficient and rapid gene transduction, and expansion of human T lymphocytes at clinical scale, meeting GCP requirements. This requires: (a) production of a stable packaging cell line producing transgene-positive retroviruses; (b) reproducible functional gene transduction procedures for patient-derived T lymphocytes, e.g., RCC patients; and (c) rapid expansion of transduced T lymphocytes. Testparameters were evaluated by measuring levels of (1)scFvG250 ch-Rec expression by T-cell transductants, and (2) G250-TAA-specific cytolysis of RCC cells by scFvG250 ch-Rec^POS T-lymphocyte transductants.

We analyzed various medium compositions and conditions for optimal virus production, per unit volume, by the scFvG250-CD4/-positive PG13 packaging cell clone 1.2using percentage ch-Rec^POS T lymphocytes and T-lymphocytes immune-specific functions (cytolysis and cytokine production) as read-out criteria. We observed higher T-cell gene transduction efficiency by virus supernatant produced at 32°C vs 37°C, confirming previously reported data.⁴⁰ Further, we showed that virus supernatant produced in RPMI 1640 medium supplemented with 2% HSA (RPMI/HSA) was as potent as RTVsup produced in RPMI 1640 medium supplemented with 10% FCS or 10% human serum. For clinical application, RPMI/HSA is therefore the medium of choice because HSA is approved for clinical use.

Fibronectin fragment CH-296 (RetronectinÔ) facilitated a high functional transduction efficiency of the scFvG250 ch-Rec gene in human T lymphocytes. The optimal time point to perform gene transduction was at days 2 and 3 post T-cell activation and this coincided with optimal expression of VLA-4/5, the ligands for RetronectinÔ, on T lymphocytes⁴¹ (data not shown). Our finding that RetronectinÔ-mediated transduction efficiencies in T lymphocytes were similar following activation by sCD3 mAb, iCD3/iCD28 mAbs, and PHA contrasts with data reported by Pollok et al.^28,42 They reported highest transduction efficiency for murine B7.1 in iCD3/iCD28 mAb-activated human T lymphocytes. The discrepancy may be due to the use of different culture media. Indeed, we showed, for example, that the gene transduction efficiency of human Tcells is significantly affected by the composition of lymphocyte culture medium, in particular between two gene transduction cycles.

We and others⁴³ observed a faster T-lymphocyte proliferation following T-cell activation by iCD3/iCD28 mAbs, which is mainly caused by a more vigorous CD4⁺ T-cell expansion. However, this did not coincide with increased immune-specific scFvG250 ch-Rec-mediated cytotoxicity/cytokine production by the T cells, when compared with T cells activated by sCD3 mAb. In fact, sCD3 mAb-activated T cells yielded T-cell effector transductants with higher levels of specific immune functions, i.e., cytotoxicity and cytokine production, at culture days 14-18, the period over which the RCC patients receive the immune gene therapy. Moreover, sCD3 mAb-activated T-cell transductants showed only low levels of "nonspecific" activated killer cell activity. In view of clinical application, we consider these specific immune functions to represent the most critical criteria to determine the efficacy of T-cell activation and gene transduction procedures. For clinical application, we conclude that sCD3 mAb is the reagent of choice for T-lymphocyte activation, also because clinical grade anti-CD3 mAb is readily available, but at present, no clinical grade anti-CD28 mAb.

The percentage of scFvG250 ch-Rec^POS T cells was stable throughout culture days 9-18, i.e., days 6-15 post gene transduction. Although the density of ch-Rec expression on T transductants slightly decreases during this culture period, no loss in ch-Rec-mediated lytic capacity occurs. Of note, the lower density of the scFvG250 ch-Rec might result in a decreased recycling capacity of lytic activity.²⁴ Although scFvG250 ch-Rec expression by T transductants could be up-regulated by reactivation with iCD3/iCD28 mAb^28,42 (and own data, not shown), their murine nature does not allow clinical use shortly before reinfusion of the lymphocytes into patients because they stay bound to the T-cell surface and elicit a HAMA response.¹¹

The optimal cell processing protocol for T-cell gene transduction and expansion routinely yields 30-65% scFvG250 ch-Rec^POS T lymphocytes. This percentage remained stable during the 2 weeks of T-cell expansion phase and, hence, no enrichment of scFvG250^POS T cells is required to arrive at the protocol-defined treatment doses.

ScFvG250 ch-Rec^POS T-cell infusions are 50-90% CD8⁺ T lymphocytes, when harvested at culture days 14-18. This lymphocyte subset composition is similar to that of non-gene-transduced activated and expanded autologous T lymphocytes used for the treatment of ovarian carcinoma patients following sensitization with bi-specific mAb.^6,9,26 Upon G250-TAA-specific stimulation, the scFvG250 ch-Rec^POS T-lymphocyte transductants were highly cytolytic and also produce Th-1 cytokines TNF-, IFN-, and GM-CSF.^21,24

In summary, we established the following efficient and reproducible T-cell processing protocol for routine retroviral gene transduction and expansion of primary human T cells at clinical scale, using a mAb-based RCC-specific ch-Rec scFvG250-CD4/.

Step 1

Activate in vitro patient PBMC in Nexell culture bags in Mix-Med medium²⁶ (=80% RPMI-1640+20% AIMV), supplemented with 2 mM glutamine, 100 g/mL streptomycin, 2 IU/mL heparin, 2% autologous human plasma, and 10 ng/mL sCD3 mAb.

Step 2

Perform two transduction cycles at days 2 and 3 post T-lymphocyte activation using:

1. fibronectin fragment CH-296 (RetronectinÔ, 6 g/cm²), coated to
2. Nexell LifecellÒX-foldÔ culture bags at
3. 0.5´10⁶ lymphocytes per
4. 0.3 mL of RTVsup per square centimeters of culture bag per transduction.

Step 3

Between the two transduction cycles, maintain T lymphocytes in RPMI 1640 medium supplemented with 8% human serum (at 0.4 mL/cm²).

Step 4

Expand transduced T lymphocytes in Nexell culture bags in Mix-Med medium for >12 days.

IL-2 is added during the gene transduction period at 100IU/mL IL-2, and during the post gene transduction T-lymphocyte expansion phase at 360 IU/mL IL-2.

Only 20 mL of autologous patient serum is needed to supplement media to conduct gene transduction of patient Tlymphocytes and only 300 mL of autologous patient plasma to supplement media for T-lymphocyte activation and expansion. The volumes are readily collected at the time of apheresis as part of the clinical protocol. Nexell bags, a closed cell culture technology, meet all GCP and Good Manufacturing Practice requirements and will be used throughout all phases of T-lymphocyte processing: (a) lymphocyte activation, (b) transduction, and (c) expansion using clinical grade bioproducts and autologous protein sources as medium supplements. The defined transduction protocol will be applied in our phase I/II clinical study of adoptive immunogene therapy of metastatic RCC with scFvG250 ch-Rec-transduced autologous T lymphocytes involving 24 patients.

Acknowledgements

The authors thank Brigitte van Krimpen and Pascal van Elzakker for their technical assistance; Nexell International for kindly providing LifecellÒ X-foldÔ culture bags; and Takara Shuzo (Otsu, Shiga, Japan) for kindly providing fibronectin fragment CH-296 (RetronectinÔ). The work was supported by the Dutch Technology Foundation STW (project RGN.3498), the Dutch Cancer Society (Nederlandse Kankerbestrijding project DDHK99-1865), the Cancer Research Institute (New York, NY, USA), and the EU Grant QLK3-1999-01-262.

1 Rosenberg SA, Yannelli J, Yang J et al. Treatment of patients with metastatic melanoma with autologous tumor-infiltrating lymphocytes and interleukin 2. J Natl Cancer Inst 1994; 86: 1159-1166. MEDLINE

2 Falkenburg JH, Wafelman AR, Joosten P et al. Complete remission of accelerated phase chronic myeloid leukemia by treatment with leukemia-reactive cytotoxic T lymphocytes. Blood 1999; 94: 1201-1208. MEDLINE

3 Lanzavecchia A, Scheidegger D. The use of hybrid hybridomas to target human cytotoxic T lymphocytes. Eur J Immunol 1987; 17: 105-111. MEDLINE

4 Van Dijk J, Warnaar SO, Van Eendenburg JDH et al. Induction of tumour cell lysis by bi-specific monoclonal antibodies recognizing renal cell carcinoma and CD3 antigen. Int J Cancer 1989; 43: 344-349. MEDLINE

5 Mezzanzanica D, Garrido MA, Neblock DS et al. Human T lymphocytes targeted against an established human ovarian carcinoma with a bi-specific F(ab')2 antibody prolong host survival in a murine xenograft model. Cancer Res 1991; 5: 5716-5721.

6 Bolhuis RLH, Lamers CHJ, Goey SH et al. Adoptive immunotherapy of ovarian carcinoma with Bs-Mab-targeted lymphocytes: a multi-center study. Int J Cancer 1992; 7: 78-81.

7 Renner C, Jung W, Sahin U et al. Cure of xenografted human tumours by bi-specific antibodies and human T cells. Science 1994; 264: 833-835. MEDLINE

8 Canevari S, Stoter G, Arienti F et al. Regression of advanced ovarian carcinoma by intra-peritoneal treatment with autologous T lymphocytes retargeted by a bi-specific monoclonal antibody. J Natl Cancer Inst 1995; 19: 1463-1469.

9 Jain RK. Delivery of novel therapeutic agents in tumours: physiological barriers and strategies. J Natl Cancer Inst 1989; 81: 570-576. MEDLINE

10 Blank-Voorthuis CJAC, Braakman E, Ronteltap CPM et al. Clustered CD3/TCR complexes do not transduce activation signals following bi-specific monoclonal antibody triggered lysis by CTL via CD3. J Immunol 1993; 151: 2904-2914. MEDLINE

11 Lamers CHJ, Gratama JW, Warnaar SO et al. Inhibition of bi-specific monoclonal antibody (bs-Ab)-targeted cytolysis by human anti-mouse antibodies in ovarian carcinoma patients treated with bsAb-targeted activated lymphocytes. Int J Cancer 1995; 60: 450-457. MEDLINE

12 Eshhar Z, Waks T, Gross G, Schindler DG. Specific activation and targeting of cytotoxic lymphocytes through chimeric single chains consisting of antibody-binding domains and the g or z subunits of the immunoglobulin and T-cell receptors. Proc Natl Acad Sci USA 1993; 90: 720-724. MEDLINE

13 Hwu P, Shafer GE, Treisman J et al. Lysis of ovarian cancer cells by human lymphocytes redirected with a chimeric gene composed of an antibody variable region and the Fc receptor chain. J Exp Med 1993; 17: 361-366.

14 Roberts MR, Qin L, Zhang D et al. Targeting of human immune deficiency virus-infected cells by CD8⁺ T lymphocytes armed with universal T-cell receptors. Blood 1994; 84: 2878-2889. MEDLINE

15 Harris DT. Hormonal therapy and chemotherapy of renal-cell carcinoma. Semin Oncol 1983; 10: 422-430. MEDLINE

16 Kruit WHJ, Goey SH, Lamers CHJ et al. High-dose regimen of interleukin-2 (IL2) and interferon-alpha (IFN) in combination with lymphokine-activated killer cells (LAK) in patients with metastatic renal cell cancer. J Immunother 1997; 20: 312-322. MEDLINE

17 Oosterwijk E, Ruiter DJ, Hoedemaeker PHJ et al. Monoclonal antibody G250 recognizes a determinant present in renal cell carcinoma and absent from normal kidney. Int J Cancer 1986; 38: 489-494. MEDLINE

18 Oosterwijk E, Debruyne FMJ, Schalken JA. The use of monoclonal antibody G250 in the therapy of renal-cell carcinoma. Semin Oncol 1995; 2: 34-41.

19 Steffens MG, Boerman OC, Oosterwijk-Wakka JC et al. Targeting of renal cell carcinoma with Iodine-131-labeled chimeric monoclonal antibody G250. J Clin Oncol 1997; 15: 1529-1537. MEDLINE

20 Steffens MG, Boerman OC, Oyen WJG et al. Intratumoral distribution of two consecutive injections of chimeric antibody G250 in primary renal cell carcinoma: implications for fractionated dose radioimmunotherapy. Cancer Res 1999; 59: 1615-1619. MEDLINE

21 Weijtens MEM, Willemsen RA, Valerio D, Stam K, Bolhuis RLH. Single chain Ig/ gene-redirected human T lymphocytes produce cytokines, specifically lyse tumor cells, and recycle lytic capacity. J Immunol 1996; 157: 836-843. MEDLINE

22 Weijtens ME, Willemsen RA, Hart E, Bolhuis RLH. A retroviral vector system "Stitch" in combination with an optimized single chain antibody chimeric receptor gene structure allows efficient transduction and expression in human T lymphocytes. Gene Ther 1998; 5: 1195-1203. MEDLINE

23 Weijtens MEM, Willemsen RA, Van Krimpen BA, Bolhuis RLH. Chimeric scFv/gamma receptor-mediated T cell lysis of tumor cells is co-regulated by adhesion and accessory molecules. Int J Cancer 1998; 77: 181-187. Article MEDLINE

24 Weijtens MEM, Hart EH, Bolhuis RLH. Functional balance between T cell chimeric receptor density and tumor-associated antigen density: CTL mediated cytolysis and lymphokine production. Gene Ther 2000; 7: 35-42. MEDLINE

25 Miller AD, Garcia JV, Suhr Von N, Lynch CM, Wilson C, Eiden MV. Construction and properties of retrovirus packaging cells based on Gibbon Ape Leukemia Virus. J Virol 1991; 2220-2224.

26 Lamers CHJ, Van De Griend RJ, Braakman E et al. Optimization of culture conditions for activation and large-scale expansion of human T lymphocytes for bi-specific antibody-directed cellular immunotherapy. Int J Cancer 1992; 51: 973-979. MEDLINE

27 Lamers CHJ, Gratama JW, Van Putten WLJ, Stoter G, Bolhuis RLH. Exogenous IL-2 recruits in vitro LAK activity by in vivo activated lymphocytes. Cancer Res 1991; 51: 2324-2328. MEDLINE

28 Pollok KE, Hanenberg H, Noblitt TW et al. High-efficiency gene transfer into normal and adenosine deaminase-deficient T lymphocytes is mediated by transduction on recombinant fibronectin fragments. J Virol 1998; 72: 4882-4892. MEDLINE

29 Hanenberg H, Xiao XL, Dilloo D, Hashino K, Kato I, Williams DA. Co-localization of retrovirus and target cells on specific fibronectin fragments increases genetic transduction of mammalian cells. Nature Med 1996; 2: 876-882. MEDLINE

30 Gratama JW, Zea AH, Bolhuis RLH, Ochoa AC. Restoration of expression of signal-transduction molecules in lymphocytes from patients with metastatic renal cell cancer after combination immunotherapy. Cancer Immunol Immunother 1999; 48: 263-269. MEDLINE

31 Finke JH, Zea AH, Stanley J et al. Loss of T-cell receptor chain and p56^lck in T-cells infiltrating human renal cell carcinoma. Cancer Res 1993; 53: 5613-5618. MEDLINE

32 Hwu P, Yang JC, Cowherd R et al. In vivo antitumor activity of T cells redirected with chimeric antibody/T-cell receptor genes. Cancer Res 1995; 55: 3369-3373. MEDLINE

33 Altenschmidt U, Klundt E, Groner B. Adoptive transfer of in vitro-targeted, activated T lymphocytes results in total tumor regression. J Immunol 1997; 159: 5509-5515. MEDLINE

34 Wang G, Chopra RK, Royal RE, Yang JC, Rosenberg SA, Hwu P. A T cell-independent antitumor response in mice with bone marrow cells retrovirally transduced with an antibody/Fc- chain chimeric receptor gene recognizing a human ovarian cancer antigen. Nat Med 1998; 4: 168-172. MEDLINE

35 McGuinness PR, Ge Y, Patel SD et al. Anti-tumor activity of human T cells expressing the CC49- chimeric immune receptor. Hum Gene Ther 1999; 10: 165-173. Article MEDLINE

36 Clay TM, Custer MC, Sachs J, Hwu P, Rosenberg SA, Nishimura MI. Efficient transfer of a tumor-reactive TCR to human peripheral blood lymphocytes confers anti-tumor reactivity. J Immunol 1999; 163: 507-513. MEDLINE

37 Willemsen RA, Weijtens MEM, Ronteltap C et al. Genetic programming of primary human T lymphocytes with cancer specific chimeric single chain and two chain TCR. Gene Ther 2000; 7: 1369-1377. MEDLINE

38 Cooper LJN, Kalos M, Lewinsohn DA, Riddell SR, Greenberg PD. Transfer of specificity for human immunodeficiency virus type 1 into primary human T lymphocytes by introduction of T-cell receptor genes. J Virol 2000; 74: 8207-8212. MEDLINE

39 Stanislawski T, Voss RV, Lotz C et al. Circumventing tolerance to a human MDM2-derived tumor antigen by TCR gene transfer. Nat Immunol 2001; 2: 962-970. Article MEDLINE

40 Bunnell BA, Muul LM, Donahue RE, Blaese RM, Morgan RA. High-efficiency retroviral-mediated gene transfer into human and non-human primate peripheral blood lymphocytes. Proc Natl Acad Sci USA 1995; 92: 7739-7743. MEDLINE

41 Shimizu Y, Van Seventer GA, Horgan KJ, Shaw S. Regulation of expression and binding of three VLA (1) integrin receptors on T cells. Nature 1990; 345: 250-253. MEDLINE

42 Pollok KE, Van Der Loo JCM, Cooper RJ, Kennedy L, Williams DA. Co-stimulation of transduced T lymphocytes via T cell receptor-CD3 complex and CD28 leads to increased transcription of integrated retrovirus. Hum Gene Ther 1999; 10: 2221-2236. Article MEDLINE

43 Garlie NK, Siebenlist RE, Lefever AV. T cells activated in vitro as immunotherapy for renal cell carcinoma: characterization of 2 effector T-cell populations. J Urol 2001; 166: 299-303. MEDLINE

Figure 1 PG13 clone 1.2 stably transduced with the scFvG250-CD4/ transgene expresses the scFvG250 chimeric receptor at high levels. PG13 clone 1.2 cells are stained by indirect immunofluorescence with (A) the anti-G250 idiotype mAb NUH-82 followed by incubation with GAM-IgG₁/PE (bold line) or (B) GAM-IgG₁/PE (thin line). Histogram analysis shows relative cell number (y-axis) plotted versus relative fluorescence intensity (x-axis).

Figure 2 Clinical production of retroviruses by PG13 clone 1.2 is optimal at 32°C in RPMI 1640 medium supplemented with 2% HSA. PG13 clone 1.2 producer cells were seeded at 4´10⁴ cells/cm² in culture medium (DMEM+10% FCS); after 24 hour-culture at 37°C, the culture medium was replaced by "production medium" and the RTVsup is produced for 24 hours at 32°C or 37°C in RPMI 1640 medium supplemented with 10% FCS or 2% HSA. At RTVsup harvest, PG13 clone 1.2 cell density varied between 1.4 and 1.6´10⁵/cm². Potency of the RTVsup is determined in a RetronectinÔ-mediated transduction assay on primary human T lymphocytes. Shown is the proportion of scFvG250 ch-Rec membrane expressing lymphocytes as determined by flow cytometry at day 11 postactivation using soluble anti-CD3 mAb (see Materials and Methods). All other experimental RTVsup production media (see Materials and Methods section) performed less than the presented media. The G250-TAA-specific cytolytic activity data of the transduced lymphocytes (not shown) are in line with the flow cytometry data. Individual results in two donors (solid versus hatched bar) are shown; four donors were tested in two experiments.

Figure 3 RetronectinÔ-mediated transduction of primary human T lymphocytes. A: The effect of T-lymphocyte preactivation with sCD3 mAb, PHA, or iCD3/iCD28 mAbS on gene transduction efficiencies. T lymphocytes were transduced with the scFvG250 ch-Rec at days 2+3 after activation by 10 ng/mL sCD3 mAb, 1 g/mL PHA, or 1 g/mL immobilized CD3/CD28 mAb. The G250-TAA-specific cytolytic activity of the transduced lymphocytes on G250-TAA^POS (black box) and G250-TAA^NEG (black triangle) RCC target cells was determined in a 3-hour ⁵¹Cr release assay at day 18 after lymphocyte activation. The open box depicts the lysis of the G250-TAA^POS RCC target cell in the presence of 25 g/mL blocking parental G250 mAb. No TD: nontransduced lymphocytes. Between brackets: percent scFvG250 ch-Rec^POS lymphocytes as determined by flow cytometry. Results of a representative donor are shown (four donors were tested in two experiments). B: Optimal gene transduction at days 2+3 post T-lymphocyte activation with sCD3 mAb. G250-TAA-specific cytolysis by T lymphocytes transduced with the scFvG250 ch-Rec at different days after activation by sCD3 mAb. The 3-hour ⁵¹Cr release assay is carried out at day 18 after lymphocyte activation, i.e., 13-16 days after lymphocyte transduction. Results of a representative donor are shown (four donors were tested in two experiments). See legend to (A) for further details. For proliferation data of transduced T lymphocytes following activation by sCD3 mAb, PHA, or iCD3/iCD28 mAb, see Figure 6A. The levels of specific ⁵¹Cr release are low when compared to the levels of specific ⁵¹Cr release presented in Figures 5,7 and 8 due to the shorter assay duration (3 versus 4 hours) and variation between selected donors.

Figure 4 Ratio between the number of T lymphocytes and volume of virus-containing medium determines efficiency of RetronectinÔ-mediated transduction of primary human T lymphocytes. T lymphocytes were transduced in wells (surface area 2 cm²) at days 2+3 after sCD3 mAb activation on 6 g/cm² RetronectinÔ using 0.25, 0.5, or 0.75´10⁶ cells and 0.25, 0.5, or 1.0 mL of virus-containing supernatant per square centimeters. The proportion of scFvG250^POS T lymphocytes as determined by flow cytometry at day 11 postactivation is shown. Individual results are shown in two donors (solid versus hatched bars).

Figure 5 Medium composition in-between transduction cycles clearly affects the efficiency of RetronectinÔ-mediated transduction of primary human T lymphocytes. T lymphocytes were transduced at days 2+3 after sCD3 activation on 6g/cm² RetronectinÔ using 0.5´10⁶ cells in 0.3 mL of virus containing RPMI+2% HSA medium per square centimeters (in 2-cm² well). After the 6-hour transduction, 0.8 mL of medium was added to the wells, followed by overnight incubation. In-between the two gene transductions, the following media were tested: RPMI 1640 supplemented with 10% FCS (10F), 2% HSA (2A), 2-8% human serum (2S-8S); Mix-Med supplemented with 2% human plasma (2P) and 2 U/mL heparin, or 2% Hu-S human serum (2S); and the serum-free media AIM V and X-VIVO 15. Shown are (A) percent scFvG250 ch-Rec^POS lymphocytes as determined in flow cytometry, and mean expression at days 7-15 postactivation in two individual donors (closed versus hatched bar); and (B) G250-TAA-specific cytolysis by a representative donor at day 14 postactivation, using target cells G250-TAA^POS RCC (closed bar) and G250-TAA^POS RCC in the presence of 25 g/mL blocking parental G250 mAb (open bar). Cytolysis is expressed as percent ⁵¹Cr release in a 4-hour assay at E:T ratio of 10:1. Six donors were tested in three experiments.

Figure 6 Kinetics of proliferation of scFvG250-CD4/-transduced T lymphocytes. A: Fresh healthy donor T lymphocytes were activated with PHA (--), sCD3 mAb (--), or iCD3/iCD28 mAb (--) in complete Mix-Med medium prior to gene transduction at days 2 and 3. Following gene transduction, lymphocytes were expanded in RPMI-1640 medium+10% Hu serum: Mean+SD of eight measurements; (¾) fold increase cell number, ´50. See also Figure 3, A and B. B: Cryopreserved/thawed healthy donor T lymphocytes were activated with sCD3 in complete Mix-Med medium prior to gene transduction at days 2 and 3. Following gene transduction, lymphocytes were expanded in complete Mix-Med, (--), RPMI-1640 medium+10% Hu-serum (--), or in X-VIVO 15 (--). Representative donor is shown; (¾) fold increase cell number, ´50. C: Cryopreserved/thawed T lymphocytes of RCC patients (--; n=8) or healthy donors (--; n=4) were activated with sCD3 in complete Mix-Med medium prior to gene transduction at days 2 and 3. Following gene transduction, lymphocytes were expanded in complete Mix-Med; (¾) fold increase cell number, ´50. See also Figure 8.

Figure 7 Efficient upscaling of the transduction of primary human T lymphocytes using culture bags. Donor lymphocytes (n=12) were activated, transduced, and expanded according the "optimal" conditions (see text) using either cluster well plates (surface area 2 cm²) or in the LifecellÒ X-foldÔ bags (surface area 85 cm²). Shown are the percent scFvG250^POS T lymphocytes (left panel) and the percent G250-TAA-specific cytolytic activity (right panel) determined at days 11-18 after lymphocyte activation. Cytolytic capacity is expressed as percent ⁵¹Cr release in a 4-hour assay at E:T ratio of 10:1. Results are expressed as mean+SD. At day 14 after lymphocyte activation, the cell expansion in LifecellÒ X-foldÔ and cluster well plates was 105-fold (median, range 40-374) and 200-fold (median, range 34-414), respectively (difference not significant). See also legends to Figure 5B) for details.

Figure 8 Efficient transduction of T lymphocytes from RCC patients with the scFvG250 chimeric receptor. PBL from eight nephrectomized but untreated metastatic RCC patients, and from two healthy donors were transduced according the "optimal" transduction conditions (see text). Shown are (A) percent scFvG250 ch-Rec expressing lymphocytes [mean+SD of five determinations at five different time points (days 7-21 postactivation), on the same batch of gene-transduced T lymphocytes], and (B) percent G250-TAA-specific cytolysis by T-lymphocyte transductants at day 14 postactivation, expressed as percent ⁵¹Cr release at E:T ratio 10:1. P1-P8: RCC patients; D: healthy donor (see also legends to Figure 5B for details). For proliferation data of RCC patient and healthy donor T-lymphocyte transductants following activation by sCD3 mAb, see Figure 6C).

Table 1 Retronection CH-296 is superior to polybrene as a transduction-mediating agent

Table 2 scFvG250 ch-Rec^POS T lymphocytes efficiently lyse G250 ligand-positive RCC and produce TNF- and IFN- upon G250 ligand-specific stimulation