Synthetic Immunomodulatory Gene Circuits for Treating Cancers
Cancer immunotherapies have demonstrated great potential but still face significant challenges when treating solid tumors. One major hurdle is that solid tumors create hard to penetrate tumor masses and an immunosuppressive tumor microenvironment (TME) that limits immune cell recruitment, infiltration, and activation. To enhance the specificity and efficacy, we aim to develop a system to rationally produce payloads that target cancer, a programmable Synthetic Transcription-factor Activity Responsive (STAR) Gene Circuits that enables tumor-localized therapeutic payload production, for recruitment and activation of immune cells.
Briefly, these gene circuits can be encoded in viruses and delivered into the body. Once the circuits enter the cells, they will sense several pre-defined intracellular tumor signatures and determine whether the cell is cancerous by Boolean-logic computation (to enhance specificity). If the cell is considered cancerous, the circuit will hijack and command the cell, to produce an effective tumor-localized combinatorial immunotherapy (to enhance safety and efficacy). For example, we have designed these circuits to sense the activities of two cancer-associated transcription factors (TFs) (cMyc and E2F1). Only when the activities of both TFs are high (Boolean AND logic), will the circuits trigger tumor-localized combinatorial immunotherapy (e.g. a combination of Surface T-cell Engager [STE], CCL21, IL-12, and anti-PD1 antibody) (Fig. 1), while potentially keeping normal cells unharmed (Fig. 2).
Importantly, we had demonstrated that STAR mediates robust therapeutic efficacy in vivo in solid tumors such as ovarian cancer, even when only a small fraction of tumor cells were delivered with the circuit (Fig. 3).
Furthermore, to readily identify cancer sensors for potentially targeting any cancer type, we have developed a machine-learning-based screening platform for high-throughput identification of synthetic promoters (cancer sensors for STAR) with enhanced cell-state specificity (SPECS). SPECS platform can be used to identify synthetic promoters for potentially any cell state of interest (Fig. 4A). These promoters exhibit more than 1000-fold activity difference between tumor cell lines and non-tumorigenic cell lines (Fig. 4B), whereas native tumor-specific promoters generally exhibit less than 10-fold activity difference. These synthetic promoters can directly serve as sensors for highly active tumor-associated transcription factors in various cancer types such as ovarian cancer, breast cancer, and adult glioblastoma.
To further accomplish the clinical translation of STAR, several advances are required: 1) identifying cancer sensors that efficiently detect highly heterogeneous patient tumors, to optimize tumor-targeting efficiency and specificity, 2) optimizing therapeutic output combinations, and 3) encoding the circuits into FDA-approved viral vehicles, for achieving optimal anti-tumor immunity and maximal efficacy.
In addition, we are also actively developing the next-generation circuit platforms that allow for the detection of other important tumor signatures, temporal control of payload production, and efficient delivery (Fig. 5). We believe that STAR gene circuit will be highly synergistic with the current FDA-approved immunotherapies and could serve as an integral part of future combination therapies.
