ROR1 targeted immunoliposomal delivery of OSU-2S shows selective cytotoxicity in t(1;19)(q23; p13) translocated B-cell acute lymphoblastic leukemia

1. . Introduction

Acute lymphoblastic leukemia (ALL) comprises the overwhelming majority of acute leukemia diagnoses in children and approximately 20% of acute leukemia in adults [1]. While outcomes in children have improved since 1950′s, relapses still occur. Compared to pediatric ALL patients, survival outcomes in adolescents and young adults are worse due to high- risk genetic features and co-morbidities limiting the use of several agents. While overall 5-year survival has improved to ~70%, only 30–40% of adults with ALL achieve long term remission [1]. Cytogenetic and molecular studies are routinely used to risk-stratify ALL patients. One recurring cytogenetic abnormality in B-ALL is the t(1;19) (q23;p13) translocation seen in about 5% of pediatric and 2% of adult patients [2]. Its immunophenotype is characterized by lack of CD34 expression and presence of cytoplasmic immunoglobulin heavy-chain suggesting transformation in B-cell maturation arrest at an intermediate stage of B-lineage development. While prognosis in adults is not clearly defined, the presence of t(1;19) is favorable to intermediate risk in children [3,4], with a high risk of CNS relapse [5] and could benefit from improved therapies.

The receptor tyrosine kinase ROR1 is uniquely expressed on t(1;19)-positive ALL cells [2,6], but is mostly absent in normal adult tissues. Preclinical models of t(1;19) ALL have demonstrated a dependency of the leukemic cells on surface ROR1 for survival [2], rendering this an attractive therapeutic target. Here, we leverage the ROR1 tumor specific antigen to target drug delivery selectively to leukemic cells with the help of immunoliposomes. Liposomal nanoparticles, which are phospholipid bilayer vesicles, can be loaded with a variety of therapeutic payloads. Active targeting of liposomes by incorporating antibodies on the bilayer surface (immunoliposomes or ILPs) can selectively enhance drug payload to target cells by increasing affinity for target cells, mitigate off-target effects on normal cells and potentially overcome the drug efflux pumps [7–9].

FTY720, a sphingosine analog, shows potent preclinical activity against multiple hematologic malignancies including B-ALL [10,11]. However, immunosuppression and other side effects limit its therapeutic benefit, as with many clinical agents in B-ALL. OSU-2S, a synthetic FTY720 derivative, retains the anti-tumor activity of FTY720, while lacking the immunosuppressive nature [7], and demonstrates anti-neoplastic activities via modulating PP2A phosphatase activity to decrease oncogenic c-Myc and increase p21 in myeloid leukemias [12]. Here, we use ROR1 targeted immunoliposomal nanoparticles [7] to deliver the therapeutic payload OSU-2S, an FTY720 derivative [7,12], selectively to ROR1 + t(1;19) B-ALL cells, providing a precision medicine directed therapy for this aberration.

2. Materials and methods

2.1. Cells

Cell lines were obtained from American Type Culture Collection and tumor cells from bone marrow (BM) aspirates following written informed consent under an Institutional Review Board approved protocol. Cells were cultured in RPMI1640 (Life Technologies) with 10% fetal bovine serum (FBS), 2 mM L-glutamine, penicillin (100 U/ml), streptomycin (100 μg/ml) and cytokines (10 ng/ml IL7, FLT3-L and SCF; R&D Systems, Minneapolis, MN). Cell lines were cultured similarly without cytokines. OP9 murine stromal cells were expanded in DMEM with growth additives before co-culture with primary ALL blasts.

2.2. Chemicals and reagents

OSU-2S was synthesized as described [7] at The Ohio State University Medicinal Chemistry shared resource and verified by Nuclear Magnetic Resonance and Mass Spectrometry. DMSO (Invitrogen) was used to dissolve OSU-2S for in-vitro studies. Antibodies used include c-Myc, p21 and β-Actin (Cell Signaling, Danvers, MA), CD45-FITC, CD3-PE-Cy7, CD19-BV785 (BioLegend, San Diego, CA), murine CD45-PECF594 (BD Biosciences) and ROR1- PE (R&D Systems).

2.3. Cell viability, cell cycle and staining

Cell viability, cell cycle and staining: was carried out as previously described [7,12].

2.4. 2A2-OSU-2S-ILP- preparation and characterization

a) Unencapsulated, non-targeted liposomes (Empty-LP), b) unencapsulated, control mouse IgG1 immunoliposomes (IgG-Empty-ILP), c) unencapsulated immunoliposomes targeting ROR1 using the 2A2 mouse anti-human ROR1 monoclonal antibody [7,13,14] (2A2-Empty-ILP), d) non-targeted liposomes encapsulating OSU-2S (OSU-2S-LP), e) control mouse IgG1 immunoliposomes encapsulating OSU-2S (IgG-OSU-2S-ILP) and f) ROR1 targeted immunoliposomes encapsulating OSU-2S (2A2-OSU-2S-ILP) were prepared and characterized as described [7].

2.5. In-vivo Xenograft study

1 × 10⁶ 697 cells were transplanted into NOD-scid IL2rγnull (NSG) mice by tail vein injection. Treatment with either IgG-OSU-2S-ILP or 2A2-OSU-2S-ILP (15 mg/kg equivalent of OSU-2S, 3x week) was started 3 days post engraftment. Mice were sacrificed 2 weeks post engraftment and tumor burden was assessed in bone marrow and spleen using flow cytometry. No animals were excluded from analysis and animal groups were not blinded. For assessing survival, mice engrafted with 1 × 105 697-GFP-Luciferase cells were treated with Empty-LP, 2A2-Empty-ILP, IgG-OSU-2S-ILP and 2A2-OSU-2S-ILP till end of study. Eight unengrafted mice were excluded from analysis (determined by bioluminescent imaging). All animal experiments were carried out under protocols approved by OSU Institutional Animal Care and Use Committee.

2.6. Statistics

Statistics analysis was conducted using SAS 9.4 (SAS Institute; Cary, NC). Cytotoxicity was analyzed by mixed effect model, incorporating repeated measures for each subject. Difference in survival between groups was compared using log-rank test. Multiplicity was adjusted by Holm’s method. Significance was indicated as ****P < 0.0001; ***P < 0.001; **P < 0.005,*P < 0.05.

3. Results and discussion

OSU-2S showed significant dose-dependent cytotoxicity in primary B-ALL cells with different cytogenetic aberrations including t(4;11), t (9;22), t(1;19), hyperdiploid and hypodiploid ALL (Fig. 1A). OSU-2S also showed significant cytotoxicity and cell cycle arrest at S-phase entry in B-ALL cell lines 697 [t(1;19)], RS4;11 [t(4;11)] and Nalm-6 [t (5;12)] (Fig. 1B,C), along with downregulation of proto-oncogene c-Myc and upregulation of cell cycle regulator p21 (Fig. 1D). Myc overexpression, often associated with poor prognosis, is characteristic of B-cell lymphomas and several other B-cell neoplasms and contributes to leukemogenesis. As such, OSU-2S mediated Myc targeting has possibilities across different B-cell malignancies. OSU-2S treatment also induced reactive oxygen species (ROS) (Fig. 1E) and treatment with ROS scavenger N-acetyl cysteine (NAC) partially rescued OSU-2S mediated cytotoxicity (Fig. 1F). Consistent with the role of stromal microenvironment in leukemic survival, co-culture of primary B-ALL blasts with OP-9 murine stromal cells significantly enhanced leukemic cell survival in-vitro, however, it did not affect OSU-2S mediated cytotoxicity (Fig. 1G). OSU-2S had no significant effect on viability of OP-9 cells alone (Fig. 1H).

Fig. 1. OSU-2S mediates cytotoxicity and cell cycle arrest in B-ALL.

A) Percent cell viability with OSU-2S normalized to DMSO Vehicle (Veh; 0μM) in patient derived B-ALL. Primary ALL cells [n = 11] were treated with increasing concentrations of OSU-2S followed by assessment of viability by Annexin/PI staining after 48 hrs. Cells negative for both Annexin and PI were considered viable. OSU-2S treatment showed significant dose dependent cytotoxicity (p < 0.0001 trend test). B) Percent cell viability with OSU-2S normalized to Veh in ALL cell lines 697, NALM-6 and RS4;11 [n = 4]. Cells were treated with increasing concentrations of OSU-2S followed by assessment of viability by Annexin/PI staining at 48 hrs (p = 0.0002 trend test). C) 697 and RS4;11 cells treated with Veh or OSU-2S (5μM, 16 hrs) showed significant increase in G0/G1 (697: n = 4, p < 0.0001, mean increase in G0/G1 = 28.42; RS4;11: n = 3, p = 0.002, mean increase in G0/G1 = 19.91) and significant decrease in S phase with OSU-2S (697: n = 4, p = 0.0003, mean decrease in S = 23.79; RS4;11: n = 3, p = 0.006, mean decrease in S= 13.38), as evidenced by PI staining followed by flow cytometry, indicating a cell cycle arrest at S phase entry. D) OSU-2S treatment (5μM, 16 hrs) results in reduction of c-Myc protein levels in and induction of p21 protein levels in 697 cells as detected by immunoblotting. E) OSU-2S treatment (5μM, 16 hrs) induces ROS in 697 cells as detected by DHE staining (n = 3, p = 0.012, mean increase in NAC 53.29), which is prevented by presence of NAC (10mM). Histogram shows representative DHE staining⋅H₂O₂ was used as a positive control. F) NAC (10mM) significantly reduces OSU-2S (5μM, 48 hrs) mediated cytotoxicity in 697 cells as detected by AnnexinV/PI staining (n = 3, p = 0.0052, mean rescue 29.95% with NAC). G) Live hCD19 + cell counts with OSU-2S in primary ALL cells (n = 4, ALL 8–11) in mono and co-culture, normalized to Veh treated ALL monoculture. OP9 stromal cells were co- cultured with patient derived ALL cells and treated with increasing concentrations of OSU-2S. Viability of ALL cells was assessed after 72 hrs. Presence of OP9 cells significantly increased live cell counts (n = 4, p < 0.0001, mean increase 232cells/ul with OP9) and viability (n = 4, p = 0.0125, mean increase 72.67% with OP9) of ALL cells as compared to monoculture in the Veh treated condition, mimicking a protective microenvironment. However, this protective effect was lost with OSU-2S treatment. OSU-2S induced significant cytotoxicity even in presence of supportive stromal cells [n = 4, Veh (+ OP9) vs 1.25μM (+ OP9) p = 0.0002, mean decrease 495.8cells/ul]. H) Percent relative viability as assessed by AnnexinV/PI staining of OP-9 cells with OSU-2S treatment (72 hrs). No significant change was observed.

Selective delivery of therapeutic payload to tumor cells provides the opportunity to further enhance the therapeutic index of pharmacological agents such as OSU-2S. Flow cytometry confirmed surface expression of ROR1 on t(1;19) primary ALL cells (Fig. 2A). B-ALL cells lacking the t(1;19) translocation did not show surface ROR1 cell surface expression. ROR1 expression was also detected on t(1;19) translocated 697 cells, but not on Nalm-6 or RS4:11 cells (Fig. 2B–C). OSU-2S encapsulated, anti-ROR1 (2A2) targeted immunoliposomes (2A2-OSU-2S-ILP) were synthesized (Fig. 2D) and characterized by nanoparticle tracking analysis (NTA) as previously described[7] (Fig. 2E). 2A2-OSU-2S-ILPs mediated significant cytotoxicity in 697 cells (Fig. 2F) and t(1;19) primary B-ALL cells (Fig. 2G) as compared to IgG-control immunoliposomes (IgG-OSU-2S-ILP), but had no effect on non-t(1;19) primary cells and cell lines that do not express surface ROR1 (Fig. 2H). Non-targeted OSU-2S immunoliposomes (OSU-2S-LP) and IgG-OSU-2S-ILP showed some non-significant cytotoxicity compared to non-targeted empty liposomes (Empty-LP), possibly due to nonspecific uptake of liposomes over extended periods in direct culture. IgG-Empty-ILP or 2A2-Empty-ILP liposomes had no effect on viability. Importantly, our previous studies had demonstrated that 2A2-OSU-2S-ILPs did not affect normal B-cells[7]. Further, we evaluated in-vivo anti-leukemic activity of 2A2-OSU-2S-ILP in ROR1 + 697 cell line derived xenograft murine model. 2A2-OSU-2S-ILP treatment for 14 days significantly reduced tumor burden in BM as assessed by analysis of human CD19 + cells compared to IgG-OSU-2S-ILP treated mice (Fig. 2I). No gross treatment related toxicities were observed in the mice. This prompted us to further evaluate the therapeutic effect of 2A2-OSU-2S-ILP on survival of 697 xenografted mice. 2A2-OSU-2S-ILP treatment significantly improved survival of leukemic mice as compared to control treatment cohorts 2A2-Empty-ILP and IgG-OSU-2S-ILP (Fig. 2J), providing in-vivo therapeutic benefit.

Fig. 2. 2A2 OSU-2S immunoliposomes selectively target ROR1+ B-ALL cells.

A) ROR1 Mean Fluorescence Intensity (MFI) normalized to isotype (ΔMFI) of non t(1;19) and t(1;19) translocated B-ALL samples. Translocated patient derived cells show specific ROR1 expression (n = 3, p = 0.0002, mean difference in MFI 403.3 ± 48.04). Histograms show representative surface ROR1 expression in t(1:19) translocated and non t(1:19) B-ALL patient derived CD19 cells. B) ROR1 Mean Fluorescence Intensity (MFI) normalized to isotype (ΔMFI) of non t(1;19) (NALM-6, RS4;11) and t(1;19) translocated (697) B-ALL cell lines. t(1;19) translocated cells show specific ROR1 expression (n = 3, p = <0.0001, mean increase in MFI 697 vs NALM-6 = 2528, 697 vs RS4;11 = 2591). C) Representative histograms showing ROR1 expression on 697 cells but not non t(1;19) NALM-6 and RS4;11 cells as detected by flow cytometry. D) Scheme depicting preparation and expected action of ROR1 targeting immunoliposomes encapsulating OSU-2S (2A2-OSU-2S-ILP). E) Representative analysis of size and concentration of 2A2-OSU-2S-ILPs using Nanoparticle Tracking Analysis by NanoSight. Mean size was 186.9 +/− 0.8nm, mean concentration was 1.2×1013 +/− 1.05 × 1011 particles/ml. F) 2A2-OSU-2S-ILPs mediate selective cytotoxicity in ROR1 + 697 cells as compared to IgG-OSU-2S-ILP (n = 8, p < 0.0001, mean decrease in viability 61.62%). G) 2A2-OSU-2S-ILPs mediate selective cytotoxicity in ROR1 + t(1:19) primary ALL cells as compared to IgG-OSU-2S-ILP and Empty ILPs (n = 3, p = 0.0174, mean decrease in viability 35.14%). H) No significant relative cytotoxicity was observed with 2A2-OSU-2S-ILPs in ROR1- non t(1;19) cell line (NALM-6) and B-ALL primary cells. I) 2A2-OSU-2S-ILP significantly reduced tumor burden in the bone marrow of treated mice as compared to IgG control as detected by human CD45+/CD19+cells [p = 0.022, n = 5 (IgG-OSU-2S-ILP), n = 6 (2A2-OSU-2S-ILP), mean decrease = 1.751 ± 0.6372 × 106 cells. J) 2A2-OSU-2S-ILP (n = 9) treatment significantly improves survival in 697 CDX model as compared to (n = 6) IgG-OSU-2S-ILP (p = 0.013) and (n = 5) 2A2-Empty-ILP (p = 0.0044) treated mice.

Studies have shown that ROR1 is not expressed in most normal adult tissues [15], making it an attractive target for immunotherapy, antibody-drug conjugates (ADC) [16] and other targeted inhibitors. While later studies have detected ROR1 expression in certain normal tissues, several preclinical and phase I clinical studies have the established safety of ROR1-targeted therapies [17,18]. Several liposomes have been approved for use in solid tumors and hematological malignancies including breast and ovarian cancer, acute myeloid leukemia and ALL [8]. The functionalization of liposomes with antibodies to generate immunoliposomes allows for active targeting of tumor cells. While ADCs use a conceptually related strategy, immunoliposomes offer the potential advantages of the ability to deliver a combination of agents to overcome drug resistance. Early clinical trials have established the safety of EGFR based immunoliposomes in solid tumors [19], immunoliposomes are yet to be exploited in hematologic malignancies.

Free OSU-2S is effective against wider subgroups of B-ALL, and recent studies demonstrate favorable pharmacokinetic properties of OSU-2S [20]. However, OSU-2S is also toxic to normal B cells [7] and peripheral blood mononuclear cells at high OSU-2S concentrations [12]. On the other hand, 2A2-OSU-2S-ILP show significantly less toxicity to normal B-cells as compared to free OSU-2S [7]. In fact, liposomal drug delivery has shown advantage over traditional treatment strategies in terms of improved pharmacokinetics, permeability and retention, as well as reduced toxicities [8,9]. As such, immunoliposomal 2A2-OSU-2S-ILP, while specifically targeting t(1;19) B-ALL, are more suited to in-vivo therapeutic use due to minimal toxicity to normal cells and improved efficacy. Together, these findings characterize the potential therapeutic applicability of OSU-2S in B-ALL and provides a framework for immunoliposomal targeted delivery in ROR1 + malignancies.