1824 Selective Expansion of Regulatory T Cells in Patients with Systemic Lupus Erythematosus by a Novel IL-2 Conjugate, NKTR-358 Christie Fanton1, Richard Furie2, Neha Dixit1, Cat Haglund1, Lin Lu1, Suresh Siddhanti1, Vishala Chindalore3, Robert Levin4, Islam Diab5, Brian Kotzin1, Jonathan Zalevsky1 1Nektar Therapeutics, Inc., San Francisco, CA; 2Division of Rheumatology, Northwell Health, Great Neck, NY; 3Pinnacle Research Group, Anniston, AL; 4Clinical Research of West Florida, Clearwater, FL; 5Paramount Medical Research and Consulting, Middleburg Heights, OH

BACKGROUND NKTR-358 led to sustained, dose-dependent increases in the number and proliferation of NKTR-358 led to the dose-dependent induction of associated with regulating immune bright CD25 Tregs processes and Treg function • Compared with recombinant human -2 (IL-2), PEG- • NKTR-358 has shown: Fold change in absolute number Proliferation (% Ki67+) conjugated NKTR-358 has:1 of CD25bright Tregs of CD25bright Tregs Pattern of differential expression Dose-dependent changes in two genes — Activity in animal models of systemic lupus erythematosus (SLE) 30 2 20 of genes at 24 μg/kg — An altered binding profile, with a lower binding affinity for IL-2Rβ and cutaneous hypersensitivity Placebo 25 25 and a different binding bias for IL-2Rα and IL-2Rβ — Selective stimulation of Tregs in a single-ascending dose (SAD) IL1R1 Color key Placebo study in healthy volunteers3 IL5RA — Selectivity for stimulating regulatory T cells (Tregs) over 15 3 µg/kg ID01 20 20 3 µg/kg conventional T cells (Tcons) 20 CD24 KLRC1 6 µg/kg 6 µg/kg NCAM1 1.5 15 15 Tregs (mean ± SEM) -1.5 -0.5 0.5 — An increased half-life PTPGDR2 12 µg/kg Log2 fold change 10 SMPD3 bright (mean ± SEM) 12 µg/kg 10 24 µg/kg + CCR3 10 NKTR-358: A PEG-conjugated rhIL-2 that selectively induces Tregs and their suppressive activity CD9 10 CSF1 Day of 24 µg/kg 5 5 Selectively 5 IL3RA administration IL-2Rα % Ki67 IL2RA NKTR-358 αβγ

Day of IL18R1 Fold change from D1 baseline expands Stimulates Tregs Fold change from D1 baseline Downregulates 0 0 Tregs Tregs with the potential administration SOCS1 proliferation of CD79A 0 2 4 6 828 30 32 34 36 38 0 2 4 6 828 30 32 34 36 38 for long-term control 0 0 1st dose 3rd dose 1st dose 3rd dose + conventional T cells 0 10 20 30 40 50 60 70 80 D1 D3 D5 D7 CD4 regulatory of the immune response 0 10 20 30 40 50 60 70 80 D29 D31 D34 D36 Day Day Fold change, CD25 T cells NanoString mRNA expression analysis RNA samples were collected after the 1st and 3rd administration. IL-2R Day Day (Pan-Cancer Immune Pro ling 770 panel). D1, day 1. SEM, standard error of the mean; Treg, regulatory . IL-2Rβ βγ Stimulates immune CTLs response to • NKTR-358 elicited an increase in the number and magnitude of differentially expressed genes associated with Treg regulation kill tumor cells • NKTR-358 (24 µg/kg) resulted in: • Some of the differentially expressed genes overlap with those reported previously in healthy volunteers CD8+ T cells bright — Maximum 12-fold mean peak increase over baseline in number of CD25 Tregs • and NK cells Repeat administration with NKTR-358 showed further induction in most dose-dependent genes — Treg levels peaked at Day 10 after the first NKTR-358 dose and remained above baseline for 20–30 days following the last dose CTL, cytotoxic T ; IL-2R, interleukin-2 ; NK, natural killer; rh, recombinant human; Treg, . — Maximum 5-fold mean peak increase over baseline in Ki67+ CD25bright Tregs METHODS • Treg increases observed with NKTR-358 in SLE patients were comparable with those seen at 28 μg/kg in healthy volunteers NKTR-358 demonstrated a dose-dependent reduction in CLASI-A score* • No overall changes in Tcon cell numbers were observed following NKTR-358 administration (data not shown) • One patient (24 µg/kg) experienced a reduction in Mean change in CLASI-A score from baseline with NKTR-358 CLASI-A score from 22 at baseline to 5 by Day 43 (2 (data shown are for patients [n=22] with a CLASI-A score of ≥4 at baseline) Study design Correlation of Treg number with FoxP3 demethylation at 24 µg/kg weeks after last dose) 1.0 0.25 Identification of NKTR-358-induced Tregs -0.25 A randomized, double-blind, multiple-ascending dose (MAD) Phase 1b study of subcutaneous NKTR-358 in patients with 250 • Seven of 18 patients had a ≥4-point reduction in CLASI-A 0.0 Placebo (n=4)

) 3 µg/kg (n=5)

n -1.40 mild-to-moderate SLE (NCT03556007) supported by correlation with demethylation L score from baseline by Day 43 -1.0 µ -1.80

i o 200 / t 6 µg/kg (n=5) l a NKTR-358 24.0 µg/kg (n=9) status of FoxP3 gene l s • No observed changes in SLEDAI or joint scores were -2.0 -1.60 y e -2.00 12 µg/kg (n=4) Placebo (n=3) h t c 150 noted due to the short treatment duration in this study -3.0

3 doses at 2-week intervals e g -2.75 Eligibility criteria Cohort 4 Endpoints

• e 24 µg/kg (n=4) m

An epigenetically active (demethylated) FoxP3 gene is observed r

• SLE diagnosis within 6 months • Primary: Safety and tolerability + e -4.0 -3.50 4 T 100 + NKTR-358 12.0 µg/kg (n=9) solely in Tregs and not in activated, (non-Treg) CD4 Tcons d Spearman r=0.716 -4.00 • Mild-to-moderate disease activity Placebo (n=3) • Secondary: 3 -5.0 P Change from baseline 3 doses at 2-week intervals D R p<0.0001 in CLASI-A score (mean) Cohort 3 • Stable concomitant medications for – Time course and change in number and activity • x -5.00 After treatment with ≥6 μg/kg NKTR-358, a significant correlation S 50 *In this small subset of patients, primarily with mild disease and short treatment duration. o

T -6.0 ( ≥8 weeks; prednisone ≤10 µg/day of Tregs, Tcons, NK cells, and subsets was observed between NKTR-358-induced Tregs identified by F 0 28 43 NKTR-358 6.0 µg/kg (n=9) – Change in levels, peripheral blood cell + + Placebo (n=3) flow cytometry (CD4+ CD25 FoxP3 cells) and Tregs identified 0 Day 3 doses at 2-week intervals populations, serum , and Cohort 2 0 50 100 150 200 250 – Change in disease activity based on SLEDAI, CLASI-A, by epigenetic analysis (percentage of demethylation of FoxP3 CLASI-A, cutaneous lupus erythematosus disease area and severity index–activity. N=48 CD4 FoxP3 CD25 cellsL NKTR-358 3.0 µg/kg (n=9) Each cohort was followed for and joint scores* TSDR) Placebo (n=3) 50 days after the last dose Each data point represents an individual blood sample from nine NKTR-358-treated patients 3 doses at 2-week intervals

Cohort 1 where data were available for both flow cytometry and demethylation assays. TSDR, Treg-specific demethylated region; Treg, regulatory T cell. CONCLUSIONS CLASI-A, cutaneous lupus erythematosus disease area and severity index–activity; NK, natural killer; SLE, systemic lupus erythematosus; SLEDAI, systemic lupus erythematosus disease activity index; Tcon, conventional NKTR-358 (≥12 µg/kg) elicited a dose-dependent induction of Treg activation markers T cell; Treg, regulatory T cell. *This Phase 1b study design, including small numbers of patients, low entry disease activity, and short treatment duration is unlikely to support adequate assessment of disease activity effect. Expression of CD25 on total Tregs Expression of Helios on total Tregs Expression of CTLA-4 on total Tregs

e e e n ) n 2.0 ) •

n 4 i 3 i

) NKTR-358 was safe and well tolerated, with a similar safety profile for single and repeat doses i l s l Placebo s

Assessments l s e e g g e g s s e e s e r r a a 3 µg/kg a • • r bright b

Immunophenotyping by multicolor flow cytometry was performed Epigenetic modifications were evaluated with the Epiontis ID b ( T b

( T •

A selective, dose-dependent expansion of CD25 Tregs was observed, which was

( T 3

m

m 6 µg/kg to quantify multiple immune cell subsets, using whole blood collected qPCR-based assay to monitor methylation status of the Treg-specific m 1.5 o o o E S F E S F 2 f r f r E S F f r

maintained through multiple NKTR-358 administrations

12 µg/kg

demethylation region (TSDR) of the FoxP3 gene, using whole M M e

at multiple time points pre- and post-NKTR-358 administration e

e M 2

g g 4 s g 5

n 24 µg/kg blood collected at multiple time points pre- and post-NKTR-358 n o A n i 2 a a l Treg induction was further supported by a correlation between the number of Tregs and the a bright + + + L — 1.0 h D h e h — CD25 Tregs: A CD4 FoxP3 CD25 Treg subpopulation with administration T Day of c c C c

H

C

1

d d n d l n

l 1 administration n the highest CD25 expression; expected to have the highest i l • i extent of demethylated FoxP3 i

Gene expression was measured with the NanoString platform, o o o F F suppressive capacity with whole blood collected at multiple time points pre- and post- F + + + 0.5 0 — Increases in Treg activation markers (CD25, Helios, and CTLA-4) and genes associated — CD4 T cells: CD3 CD4 Tcons NKTR-358 administration 0 10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70 80 + + + — CD8 T cells: CD3 CD8 Tcons • Post-hoc analyses were performed for change in cutaneous lupus Day Day Day with Treg regulation were observed with NKTR-358 – +++ – erythematosus disease area and severity index–activity (CLASI-A) — Natural killer (NK) cells: CD3 CD56 CD16 (bright), CTLA4, cytotoxic T-lymphocyte-associated 4; MESF, molecules of equivalent soluble uorochrome; Treg, T regulatory cell. • Low-level increases in NK cell numbers occurred in most patients at the highest NKTR-358 CD3– CD56+ CD16+ (dim), and CD3– CD56+ (total) score for individual patients following NKTR-358 treatment • A sustained induction of markers associated with increased suppressive Treg function was observed at 12 and 24 μg/kg NKTR-358 dose; the CD56bright population was more sensitive than the CD56dim population • The magnitude of increase was similar to the change observed in the SAD study in healthy volunteers RESULTS • A dose-dependent reduction in CLASI-A score was seen with NKTR-358 treatment, which NKTR-358 led to low-level increases in CD56+ NK cells in most patients at 24 μg/kg Baseline demographics and disease Safety and tolerability Fold change in frequency (% CD45) of CD56bright CD16– NK cells Fold change in frequency (% CD45) of CD56dim CD16+ NK cells warrants further exploration characteristics • No dose-limiting toxicities, deaths, or clinically significant vital sign, 50 6 • These data provide strong support for continued testing of NKTR-358 in patients with SLE electrocardiogram, or physical examination abnormalities were observed Placebo NKTR-358 (n=36) Placebo (n=12) 5 • Adverse events were primarily mild or moderate (Grade 1 or 2) 40 and other inflammatory diseases Age, years, mean (SD) 47.2 (12.5) 47.8 (8.3) 3 µg/kg injection-site reactions 4 Female, n (%) 34 (94.4) 12 (100) 6 µg/kg — A Phase 2 trial of NKTR-358 in patients with SLE is currently recruiting (NCT04433585) • One patient in the lowest dose cohort (3 µg/kg) experienced a 30 Body mass index, mean (SD) 26.9 (3.0) 26.7 (4.6) serious adverse event of migraine 3 weeks after the last dose of 3 12 µg/kg Disease duration, months, mean (SD) 9.5 (8.9) 14.3 (9.7) NKTR-358; it was not considered related to the study drug NK cells (% of CD45) 20

NK cells (% of CD45) − 2 24 µg/kg SLEDAI score, mean (SD, min–max) 6.0 (2.8, 0–10) 5.2 (2.7, 2–10) • Three patients discontinued treatment (one patient [24 µg/kg] due + CLASI-A score, mean (SD, min-max) 4.1 (4.7, 0–22) 2.7 (3.2, 0–9) to elevated eosinophil levels, with no clinical sequelae; two patients Day of CD16

10

CD16 1 Baseline medication, n (%) for reasons unrelated to adverse events) administration REFERENCES

dim 1. Fanton C, et al. Poster presented at the European Congress of Rheumatology (EULAR), 2020: THU0054. bright Fold change from baseline in Prednisone 12 (33.3) 4 (33.3) • One patient in the highest dose cohort (24 µg/kg) demonstrated Fold change from baseline in 2. Langowski J, et al. Poster presented at the American College of Rheumatology Annual Meeting, 2017: 2715. 0 0 3. Fanton C, et al. Poster presented at the American College of Rheumatology Annual Meeting, 2019: 0098. CD5 6 Hydroxychloroquine 24 (66.7) 6 (50.0) transient and mild (Grade 1) flu-like symptoms after the second CD5 6 0 10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70 80 4. Baron U, et al. Eur J Immunol 2007;37:2378–2389. Copies of this poster are for personal and third doses that were considered related to the study drug; no ACKNOWLEDGMENTS Methotrexate 4 (11.1) 0 Day Day use only and may not be reproduced clinically relevant changes in hematology, chemistry, or cytokine The investigators would like to thank the patients, their families, and the clinical teams for their participation in this study. This study was sponsored without written permission from ACR Mycophenolate mofetil 1 (2.8) 2 (16.7) NK, natural killer. by Nektar Therapeutics. All authors contributed to and approved the poster. Editorial assistance was provided by BOLDSCIENCE Inc. and funded by Nektar Therapeutics. levels were associated with either episode, and both episodes + and the authors. Azathioprine 5 (13.9) 0 resolved within 24 hours without treatment • NKTR-358 treatment led to low-level increases in total CD56 NK cells in most patients at 24 μg/kg, with the increase driven by two outlier individuals DISCLOSURES bright dim CF, ND, CH, LL, SS, BK, JZ: shareholders of Nektar Therapeutics; employees of Nektar Therapeutics. VC: grant/research support from Nektar • • The CD56 population was more sensitive to NKTR-358 than the CD56 population, showing greater induction following NKTR-358 administration CLASI-A, cutaneous lupus erythematosus disease area and severity index–activity; SD, standard deviation; SLEDAI, systemic lupus No antidrug were detected throughout the entire 50 Therapeutics for conducted studies; speaker bureaus: >5 years ago. RL: grant/research support for industry-sponsored trials; consultant for Gilead, • bright dim Exagen, Myriad Rheumatology; speaker bureaus: Sanofi/Genzyme, Regeneron, Bristol-Myers Squibb, AbbVie.RF: grant/research support from erythematosus disease activity index. days of follow-up The ratio of CD56 to CD56 NK cells increased 16-fold over predose at 24 μg/kg Nektar Therapeutics to Northwell Rheumatology to conduct this study; consultant for Nektar Therapeutics. ID has no disclosures.

Poster presented at the ACR Convergence Virtual Meeting, November 5–9, 2020. Corresponding author: Jonathan Zalevsky: [email protected]