Supporting Online Material for

www.sciencemag.org/cgi/content/full/1129003/dc1 Supporting Online Material for Cancer Regression in Patients After Transfer of Genetically Engineered Lymphocytes Richard A. Morgan, Mark E. Dudley, John R. Wunderlich, Marybeth S. Hughes, James C. Yang, Richard M. Sherry, Richard E. Royal, Suzanne L. Topalian, Udai S. Kammula, Nicholas P. Restifo, Zhili Zheng, Azam Nahvi, Christiaan R. de Vries, Linda J. Rogers-Freezer, Sharon A. Mavroukakis, Steven A. Rosenberg* *To whom correspondence should be addressed. E-mail: SAR@mail.nih.gov This PDF file includes: Materials and Methods Tables S1 to S5 References Published 31 August 2006 on Science Express DOI: 10.1126/science.1129003

1 Supplemental On-line Material Methods Patient Peripheral Blood Mononuclear Cells (PBMC), and cell lines. All of the Peripheral Blood Lymphocytes (PBL) used in this study were cryopreserved PBMC obtained by leukapheresis of metastatic melanoma patients treated at the Surgery Branch, National Cancer Institute, NIH, Bethesda, MD. T2 is a lymphoblastoid cell line deficient in TAP function, whose HLA class I proteins can be easily loaded with exogenous peptides(1). Melanoma lines 938mel, 888mel, 624mel, and 526mel were generated at the Surgery Branch from resected tumor lesions. Other cell lines used were non-small cell lung cancer line H2087 (ATCC/CRL-5922), breast cancer line MDA 453 (ATCC/HTB-131), the human lymphoid cell line Sup T1 (ATCC CRL-1942), and PG13 gibbon ape leukemia virus packaging cell line (ATCC CRL 10,686). All the cell lines described above were maintained in R10 (RPMI1640, Invitrogen Corp, Grand Isle, NY) supplemented with 10% FCS (Biofluids, Inc., Gaithersburg, MD). Culture medium (CM) for established human T lymphocyte lines was RPMI1640 with 0.05mM mercaptoethanol, 300 IU/ml interleukin-2 (IL-2) (Chiron Corp. Emerville, CA) plus 10% human AB serum (Valley Biomedical Winchester, VA). Patient Treatment This human gene therapy protocol was reviewed and approved by the National Institutes of Health institutional biosafety committee, National Cancer Institute institutional review board,

2 recombinant DNA advisory committee of the Office of Biotechnology Activities, Office of the Director, National Institutes of Health, and the Center for Biologics Evaluation and Research of the Food and Drug Administration (all, Bethesda, MD). Patients eligible for treatment on this protocol were older than 18 years, HLA-A0201+, HIV infection, hepatitis B and C negative with a life expectancy greater than 3 months with good performance status. All patients signed institutional review board-approved consents and had their pathology confirmed by pathologists at the Clinical Center, National Institutes of Health in Bethesda, MD. All patients had measurable disease on computed tomography scan, magnetic resonance imaging or by physical exam. Tumor response was determined by RECIST criteria(2). Patients were required to have progressed after prior interleukin-2 therapy before enrollment and did not have tumor infiltrating lymphocyte cultures available for treatment. PBMC were isolated by leukapheresis, separated by centrifugation on a ficoll-hypaque cushion, washed in HBSS, then resuspended at a concentration of 1 X 10 6 /ml in T-cell culture medium AIM-V (Invitrogen Corp, Grand Isle, NY) with 300IU/ml IL-2, 100U/ml penicillin, 100µg/ml streptomycin, 1.25µg/ml amphotericin, 10µg/ml ciprofoloxicin, and 5% human AB serum supplemented with 50ng/ml OKT3. After 2 days of culture, cells were collected, resuspended in fresh T cell culture medium without OKT3. A retroviral vector (MSGV1AIB) was constructed(3) and optimized to express the MART-1 TCR alpha and beta chains using an Murine Stem Cell Virus (MSCV) long terminal repeat (LTR) (known to be relatively resistant to in vivo silencing), incorporating the splicing and optimized start codon of the MFG-design of vectors, and the use of a highly efficient internal ribosome entry site (IRES) derived from the human polio virus(4). After selection of high-titer producer cell

3 clones a clinical grade retroviral vector supernatant was produced and used in a solid-phase transduction protocol that resulted in highly efficient gene transfer without the use of any selection method. The transduction of up to 5 X 10 8 cells was performed by overnight culture on Retronectin (CH-296, GMP grade Retronectin purchased from Takara Bio. Inc, Japan) coated, MSGV1AIB vector-preloaded six well tissue culture plates, using 6 ml vector and up to 5 X 10 6 cells per well. In cohorts 1 and 3, PBL were transferred the following day to a second set of pre-coated Retronectin/retroviral vector tissue culture plates. Patients in cohort 2, received cells with only one cycle of transduction. Two days after the last transduction, the PBL cultures were assayed for the presence of Vβ12 TCR protein by FACS analysis and for activity by cytokine release assay. An initial cohort of three patients was treated with cells following an extended culture period of 19 days at which point they had cell doubling times ranging from 8.7 to 11.9 days (Table 1, cohort 1; patients 1, 2a, 3). In an effort to administer gene-modified lymphocytes that were in their active growth phase, we modified the culture conditions to limit the ex vivo culture time to 6 to 9 days after stimulation of cells with OKT-3. This modification was based on prior animal and human data indicating that the in vivo proliferation and persistence of the transferred cells was highly correlated with their ability to mediate cancer regression(5-9). Thus patient lymphocytes were stimulated with OKT-3, transduced and administered either on days 6, 7, or 8 at which time they had doubling times of two days or less (Table 1, cohort 2; 11 patients). To generate larger numbers of actively dividing cells for ACT, we cultured transduced cells for 17-18 days, exposing a portion of the culture to an OKT-3 based rapid expansion protocol(10) after 8-9 days (Table 1, cohort 3; 4 patients).

4 The doubling time of the cells in cohort 3 was 0.9 to 3.3 days. The mean number of cells infused in cohort 2 was 6.2 X 10 9 and in cohort 3 was 45.3 X 10 9. All patients received nonmyeloablative lymphodepleting chemotherapy, consisting of 2 days of cyclophosphamide (60mg/kg) and 5 days of fludarabine (25mg/m 2 ). One to four days after the final dose of chemotherapy, patients received MSGV1AIB (anti-mart-1 TCR) transduced lymphocytes by intravenous bolus infusion over thirty minutes. Patients then received high dose intravenous IL-2 (720,000 IU/kg every eight hours to tolerance or a maximum of 12 doses), and MART-1:27-36(27L) peptide vaccine (1 mg peptide, manufactured by the National Cancer Institute through a contract with Ben Venue Laboratories) emulsified in Montanide ISA-51 (Seppic, Inc., Fairchild, NJ) by intramuscular injection on days 1, 2, 3, 4, 5, 12, and 19 after cell infusion. Electroporation of in vitro transcribed mrna. mrna encoding the GFP was prepared from pgem4z/egfp/a64 (kindly provided by Dr, Eli Gilboa, Duke University Medical Center. Durham, NC). mrna encoding TCR α and β chains of MART-1, gp100, NY-ESO-1, and p53 were prepared from PCR products made using gene specific primer pairs containing the T7 RNA polymerase promoter sequence(11). mmassage mmachine High Yield Capped RNA transcription Kit (Ambion Inc. Austin TX)was utilized to generate in vitro transcribed (IVT) RNA. The IVT RNA was purified using an RNeasy Mini Kit (Qiagen, Valencia, CA) and purified RNA was resuspended in RNase free water at 1-0.5mg/ml. For the stimulation of the T cells, human PBLs were stimulated with IL-2 (300 IU/ml) plus 50 ng/ml OKT3 for 7 days, and enriched for CD8+

5 cells before electroporation. The stimulated CD8+ PBLs subjected to electroporation following resuspension in OPTI-MEM (Invitrogen) medium at the final concentration of 25X10 6 /ml. Cells and cuvettes were pre-chilled by putting them on ice for >5 min. For electroporation 0.2 ml of the cells were mixed with 2 µg/1x10 6 T cells of IVT RNA and electroporated in a 2 mm cuvette (Harvard Apparatus BTX, Holliston, MA), using ECM830 Electro Square Wave Porator (Harvard Apparatus BTX, Holliston, MA) at 500V/500µs. Immediately after electroporation, the cells were transferred to fresh CM with 300 IU/ml IL- 2 and incubated at 37 C until use. FACS analysis Cell surface expression of TCRVβ12, TCRVβ8, CD3, CD4, and CD8 molecules on PBL were measured using FITC, PE or APC-conjugated antibodies according to the manufacturer s instructions. Anti-TCR Vβ8 and Vβ12 were purchased from Immunotech (Westbrook, ME) and the other antibodies were from Becton Dickinson (San Jose, CA). MART-1 specific TCR were stained with a tetrameric HLA-A2/MART-1:27-36(27L) fluorochrome conjugate (itag MHC Tetramer, Beckman Coulter, Fullerton, CA) according to the manufacturer s recommendations. Cells were stained in a FACS buffer made of PBS (BioWhitaker, Walkersville, MD) and 1% FCS. Prior to staining of peripheral blood lymphocytes, Fc receptors were blocked with normal mouse IgG (Caltag Labs, Burlingame, CA). Immunofluorescence, was measured using a FACscan or FACScaliber flow cytometer (Becton Dickinson). A combination of forward angle light scatter and propidium iodide staining was used to gate out dead cells. Approximately 1 x 10 5 events were acquired. Staining was analyzed using Cell Quest (Becton Dickson) or FlowJo (Treestar, Ashland OR)

6 software. The Vβ12 intermediate cells were identified based on the Vβ12 profile of the preinfusion cells using markers set to exclude both CD8 + Vβ12 high and CD8 + Vβ12 negative events. Other markers were set by comparison to fluorochrome-conjugated isotype control antibodies or HLA-A2 tetramer containing an irrelevant control peptide. Cytokine release assays PBL cultures were tested for specificity and tumor reactivity in cytokine release assays. For these assays, 1x10 5 responder cells and 1x10 5 stimulator cells (T2 or human tumor lines) were incubated for 18 to 24 h in a 0.2-ml culture volume in individual wells of 96-well plates. T2 cells were pulsed with peptide (1 µg/ml) in medium for 2 h at 37 C, followed by washing (three times) before initiation of co-cultures. Ex vivo lymphocyte activity was measured after overnight culture in T cell medium supplemented with 300 IU/ml IL-2. Cytokine secretion was measured in culture supernatants diluted as to be in the linear range of the assay using commercially available ELISA kits (IFN-γ, Endogen, Cambridge, MA) according to the manufacturer s recommendations. Analysis of gene marked cells. The persistence of the gene-marked cells was determined from patient PBMC. Genomic DNA was isolated using QIAamp DNA Blood Midi Kit (Qiagen, Valencia, CA) according to the manufacture s instruction. 100ng of each DNA was used for the Real-time quantitative PCR assay (TaqMan, Applied Biosystems Inc, Foster City, CA). Total RNA was isolated from PBL using RNeasy Mini Kit (Qiagen). 1µg of total RNA was used in the first strand of cdna synthesis reaction using ThermoScript RT-PCR System (Invitrogen,

7 Carlsbad, CA) using random hexamers and diluted 10-fold with RNA-free water after the reaction. One tenth of the diluted reaction mix was used later for the real-time quantitative PCR (TaqMan). All PCR reactions were performed using an ABI 7500 Fast Real-time PCR System instrument(applied Biosystems Inc,). The TaqMan gene specific assay was designed by ABI Assays-by-Design SM software (Applied Biosytems Inc.,) using as target, the sequence between TCR α and IRES genes. The assay primers/probe were: 5 - GCCGGGTTTAATCTGCTCATG-3 (AIBAI-forward primer); 5 -GCCGGGTTTAATCTGCTCATG -3 (AIBAI-reverse primer), and 5 -FAM-TCCAGCTGAACTAGAACTA-3 (AIBAI- FAM labeled probe). The reference standard curve was established using the genomic DNA from a MSGV1AIB transduced SupT1 cell clone by mixing known ratios of genomic DNA into untransduced SupT1 cell DNA. TaqMan β-actin control reagents kit (Applied Biosytems Inc.,) was used to normalize reactions to input RNA/DNA amounts.

8 Supplemental References 1. R. D. Salter, D. N. Howell, P. Cresswell, Immunogenetics 21, 235 (1985). 2. P. Therasse et al., J Natl Cancer Inst 92, 205 (Feb 2, 2000). 3. M. S. Hughes et al., Hum Gene Ther 16, 457 (Apr, 2005). 4. R. A. Morgan et al., Nucleic Acids Res 20, 1293 (Mar 25, 1992). 5. J. Zhou et al., J Immunol 175, 7046 (Nov 15, 2005). 6. J. Zhou, M. E. Dudley, S. A. Rosenberg, P. F. Robbins, J Immunother 28, 53 (Jan- Feb, 2005). 7. P. F. Robbins et al., J Immunol 173, 7125 (Dec 15, 2004). 8. J. Huang et al., J Immunother 28, 258 (May-Jun, 2005). 9. L. Gattinoni et al., J Exp Med 202, 907 (Oct 3, 2005). 10. S. R. Riddell, P. D. Greenberg, J Immunol Methods 128, 189 (Apr 17, 1990). 11. Y. Zhao et al., Mol Ther 13, 151 (Jan, 2006).

9 Supplemental Table1. RNA electroporation of tumor antigen specific TCR. Effector TCR RNA Tumor GFP MART-1 gp100 NY-ESO-1 p53 938mel 23 68 35 47 38 526mel 61 1146 317 163 306 H2087 62 224 96 116 >1654 MDA 435S-A2 96 175 125 342 127 MDA 435S 130 65 73 75 65 CD8+ PBLs were electroporated with IVT GFP RNA or tumor antigen specific TCRs (MART-1, gp100, NY-ESO-1, p53). Two hours post electroporation, the cells were cocultured with tumor cell lines and after overnight co-culture, the supernatants were collected and subjected to ELISA detection of interferon-γ secretion (in pg/ml). The known phenotypes of these tumor cell line were; melanoma 938mel (HLA-A2-/MART- 1+/gp100+), melanoma 526mel (HLA-A2+/MART-1+/gp100+/p53+), non-small cell lung cancer H2087 (HLA-A2+/NY-ESO-1-/p53+), breast cancer MDA 453S-A2 (HLA- A2+/NY-ESO-1+/p53-), and breast cancer MDA 453S (HLA-A2-/NY-ESO-1+/p53-). Values demonstrating specific release of cytokine are in bold.

10 Supplemental Table 2, Gene transfer efficiency at time of infusion. Patient Number 1 2a 3 4 5 6 7 8 9 10 11 12 13 2b 14 15 16 17 Ave Vβ12 (%CD4) Tetramer (%CD4) Vβ12 (%CD8) Tetramer (%CD8) 58 63 21 n/d 18 36 45 38 43 52 51 66 43 66 58 47 18 41 45 8 15 1 n/d 2 10 15 8 15 21 22 33 15 15 7 4 10 7 12 49 65 63 24 17 32 23 29 43 42 43 59 44 72 51 51 36 24 42 16 28 3 23 3 10 6 5 13 15 19 30 16 33 17 19 11 15 17 Samples were taken from transduced T lymphocytes cultures at or near time of infusion. FACS analysis for CD4, CD8, Vβ12, and MART-1 Tetramer (M27L) was as described in methods. n/d, not determined.; Ave, average value.

11 Supplemental Table 3, Q-RT PCR Analysis Patient Number 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Infusion 5.8 2.2 14.5 10.7 21.5 20.8 4.0 10.3 9.6 11.2 4.0 5.1 24 11.1 Post Infusion 1.2 0.8 3.5 1.9 0.6 2.8 0.9 1.4 0.6 2.9 2.8 1.1 1.9 2.2 Amount of TCR vector-derived RNA normalized to cellular actin mrna. Using RNA derived from a stable transduced human T cell line (Sup T1) as reference; total RNA from infused T cells or patient PBMC post-infusion were subjected to Q-RT PCR and then normalized to the amount of actin mrna (values are vector/1 x 10 6 actin). Samples were 2-5 weeks post-infusion, with exception of patients 8, 9, and 11, which were 2 months, 3 months, and 2 weeks respectively.

12 Supplemental Table 4, Elispot Analysis Patient Number 1 2a 3 4 5 6 8 9 13 2b 14 15 16 17 Infusion 1523 1783 215 4527 1367 1203 1352 460 3380 4877 2023 2938 1393 190 Pre 22 21 1 14 0 1 0 0 0 0 20 1 6 0 treatment Post infusion 13 45 0 41 7 1 215 20 700 13 530 35 92 108 Number of interferon-γ positive Eispots/100,000 cells. Post-infusion values derived from PBL taken at one to four weeks post-infusion and rested overnight in medium (without cytokine) prior to assay.

13 Supplemental Table 5. Antigen reactivity of ex vivo cultured PBL. Patient 4: Melanoma Cell Line A2- A2+ Peptide pulsed T2 cells 888 938 526 624 None gp100 MART-1 1.0 µm 10 µm 1.0 µm None A1,24 A1,24 A2,3 A2,3 Controls None 0 0 0 0 0 0 0 0 0 0 AK 1700-3 181 66 9 37 10 46 36 32 35 32 JKF6 0 22 28 10250 10680 1 0 >16910 >15430 7770 JR6C12 0 41 0 4190 4640 0 8000 0 8520 143 Patient 4 Pretreatment 130 204 117 98 133 521 421 422 462 520 Day 7 40 65 44 131 165 104 94 >1255 >1480 851 Day 8 87 110 75 92 78 122 190 957 893 537 Day 18 42 27 23 43 55 184 63 108 160 113 0.1 µm Anti-melanoma properties of genetically engineered lymphocytes determined for patient 4 PBL following overnight culture in IL-2 (300 IU/ml). The production of interferon-γ (pg/ml) following co-culture with peptide pulsed T2 cells (peptide reactivity) and anti-melanoma activity (tumor reactivity) for HLA-A2 + lines (526, 624) and HLA-A2- lines (888, 938). Controls included; non-peptide reactive TIL AK 1700-3, MART-1 reactive TIL JKF6, and gp100 reactive TIL JR6C12. JR6C12 recognition of 1.0µM MART-1 peptide was not reproducible in repeated assays. Values demonstrating specific release of cytokine are in bold.

14 Patient 14: Melanoma Cell Line A2- A2+ Peptide pulsed T2 cells 888 938 526 624 None gp100 MART-1 1.0 None A1,24 A1,24 A2,3 A2,3 µm 10 µm 1.0 µm 0.1 µm Controls None 0 0 43 0 0 0 23 70 0 18 AK 1700-3 35 53 33 21 26 52 43 94 73 105 15 11 66 6750 9160 37 337 >2168 >2168 >182 JKF6 8 JR6C12 0 0 0 4410 3760 0 >1510 23 15 61 Patient 14 Pretreatment 368 410 352 284 344 639 645 874 614 471 Day 5 66 53 67 110 170 134 96 918 716 634 Day 7 69 52 45 41 55 33 57 192 221 109 Day 10 79 91 0 80 46 33 52 245 214 108 Day 27 52 76 35 33 55 67 103 282 232 158 Day 88 26 50 31 0 52 75 41 52 43 31 Day 257 105 52 93 53 46 111 83 597 492 399 Anti-melanoma properties of genetically engineered lymphocytes determined for patient 14 PBL following overnight culture in IL-2 (300 IU/ml). The production of interferon-γ (pg/ml) following co-culture with peptide pulsed T2 cells (peptide reactivity) and anti-melanoma activity (tumor reactivity) for HLA-A2 + lines (526, 624) and HLA-A2- lines (888, 938). Controls included; non-peptide reactive TIL AK 1700-3, MART-1 reactive TIL JKF6, and gp100 reactive TIL JR6C12. Values demonstrating specific release of cytokine are in bold.