Reducing toxicity of 4–1BB costimulation: targeting 4–1BB ligands to the tumor stroma with bi-specific aptamer conjugates

Abstract

Systemic administration of immune modulatory antibodies to cancer patients is associated with autoimmune pathologies. We have developed a clinically feasible and broadly applicable approach to limit immune stimulation to disseminated tumor lesions using a bi-specific agonistic 4–1BB oligonucleotide aptamer targeted to a broadly expressed stromal product (e.g., VEGF or osteopontin). The stroma-targeted aptamer conjugates engendered potent antitumor immunity against unrelated tumors and exhibited a superior therapeutic index compared to non-targeted agonistic 4–1BB antibody.

Keywords:

• cancer immunotherapy
• immune stimulation
• autoimmune pathology
• 4–1BB
• tumor stroma
• VEGF
• tumor targeting

Dose limiting toxicities of cancer drugs, including immunostimulatory antibodies, are a major hurdle in developing effective treatments for cancer.¹ Arguably, enhancing the therapeutic index of immune potentiating drugs by reducing their toxicity without compromising their antitumor activity is paramount to fully exploiting their clinical potential. Enhancing TNFR superfamily, member 9 (TNFRSF9, best known as 4–1BB) costimulation by systemic administration of agonistic ligands is a case in point. 4–1BB (CD137) is a major costimulatory receptor promoting the survival and expansion of activated CD8⁺ T cells and their differentiation into memory cells. Preclinical studies in mice suggest that enhancing 4–1BB costimulation by systemic administration of a 4–1BB ligand could represent a useful modality to potentiate protective immunity in cancer patients.² Notwithstanding, systemic administration of agonistic 4–1BB antibodies to mice was also accompanied by immune anomalies, notably polyclonal activation of CD8⁺ T cells and a “cytokine storm” that affected the function of organs such as liver, spleen and bone marrow.³ Consistent with the preclinical murine studies, clinical trials using an agonistic anti-4–1BB antibody in patients with advanced cancers has been associated with high frequencies of objective responses, however, adverse effects became significant at the highest dose used causing liver toxicity that resulted in 2 fatalities.¹ Arguably, reducing the toxicity of such immune potentiating drugs without compromising their antitumor activity will be paramount for exploiting their clinical potential to the fullest extent.

We have previously shown that an agonistic 4–1BB oligonucleotide aptamer conjugated to a prostate-specific membrane antigen (PSMA) binding aptamer is efficiently targeted to PSMA-expressing tumors in vivo and potentiates vaccine-induced protective immunity.⁴ Oligonucleotide aptamer ligands represent a novel and emerging platform for drug and drug delivery that offer potentially important advantages compared to antibody ligands in terms of development, manufacture, cost, conjugation chemistry, and (lack or reduced) immunogenicity.⁵ Given that most ligand-engaged receptors, including PSMA, are internalized, and since the tumor targeted 4–1BB costimulatory ligands need to be displayed on the cell surface to engage the 4–1BB-expressing tumor-infiltrating T cells, a deletion was introduced in the cytoplasmic domain of PSMA to prevent its internalization upon aptamer engagement. Since this is not clinically feasible, the need to identify tumor-specific surface products that do not internalize upon interaction with the bispecific aptamers would significantly reduce the clinical applicability of this approach.

In our recent work, we tested the hypothesis that by targeting the costimulatory ligands to products secreted into the tumor stroma, tumor-infiltrating T cells will be costimulated prior to their engagement with the tumor cell, thereby obviating the need to target the costimulatory ligands to non-internalizing cell surface product expressed on the tumor cells (Fig. 1).⁶ To this end we conjugated an agonistic dimeric 4–1BB aptamer, that we have previously shown to costimulate activated CD8⁺ T cells as effectively as an agonistic 4–1BB antibody, to aptamers that bind either vascular endothelial growth factor (VEGF) or osteopontin (OPN/SPP1), 2 products that are secreted into the tumor stroma by progressing tumor lesions in many types of cancers. The systemically administered aptamer conjugates accumulated preferentially in tumors expressing high levels of VEGF or OPN. On a molar basis, 6–10 lower doses of VEGF-4–1BB aptamer conjugate were as effective as an agonistic 4–1BB antibody (the gold standard) or unconjugated 4–1BB aptamer at inhibiting tumor growth in tumor-bearing mice. Nevertheless and as previously described, the antibody-, but not aptamer conjugate-treated mice exhibited CD8⁺ T-cell hyperplasia and multi-organ inflammatory responses. Thus, the VEGF-targeted 4–1BB aptamer ligands exhibited a superior therapeutic index that was at least 10-fold higher than that of a 4–1BB antibody, the “gold standard." Interestingly, in this and 2 other models, unlike the 4–1BB antibody, the non-targeted free 4–1BB aptamer did not elicit CD8⁺ T-cell hyperplasia nor non-specific inflammation, suggesting that head-to-head the 4–1BB aptamer itself exhibited a superior therapeutic index compared to the antibody, though less than the stroma targeted 4–1BB aptamer. Consistent with the broad expression profile of tumor-secreted VEGF, systemic administration of the stroma targeted 4–1BB aptamer conjugate engendered potent antitumor immunity against multiple unrelated tumors in subcutaneously implanted tumor-bearing mice, as well as in more stringent and clinically relevant post-surgical breast carcinoma metastasis, carcinogen-induced fibrosarcoma, and autochthonous glioma tumor models. The therapeutic impact seen in the post-surgical⁷ and carcinogen-induced⁸ models appears to be significantly more pronounced than what has been reported before.

Figure 1. Tumor targeted immune stimulation with bi-specific aptamers. (A) Targeting to the tumor cell. A costimulatory aptamer (4–1BB is used in this model) is conjugated to an aptamer that binds a product expressed preferentially on cancer cells (prostate specific membrane antigen [PSMA]) in this example. Binding of the PSMA-4–1BB bi-specific aptamer conjugate to PSMA expressed on the tumor cell and 4–1BB on the tumor-infiltrating T cells will lead to 4–1BB costimulation, indicated by the red asterisk (signal 2). However, if the PSMA-4–1BB aptamer conjugate binds to the tumor cells prior to engagement of the T cells, the 4–1BB aptamer ligand may be prematurely internalized and the T cell will not be co-stimulated. (B) Targeting to the stroma. The 4–1BB aptamer is conjugated to another aptamer that binds to a stroma secreted product such as vascular endothelial growth factor (VEGF). The VEGF-bound 4–1BB bi-specific aptamer will, therefore, accumulate in the stroma and intercept and co-stimulate the incoming T cells prior to engaging cancer cells.

In summary, this study suggests that tumor stroma targeted costimulation with bi-specific aptamer ligands, not limited to 4–1BB co-stimulation, is a clinically feasible and broadly applicable approach to enhance tumor immunity that obviates the current limitations of using nontargeted administration of monoclonal antibodies.