Theranostics 2.0: Target-driven, artificial intelligence-enabled cancer therapy across tumor types.

Abstract

To date, prospective, pivotal clinical trials in the theranostic space have focused on specific disease states. Those include leveraging positron emission tomography (PET) findings to optimize the second-line use of ¹⁷⁷Lu-DOTATATE in patients with metastatic and progressive neuroendocrine tumors or the use of prostate-specific membrane antigen (PSMA) PET findings to select patients with metastatic castration-resistant prostate cancer for treatment with ¹⁷⁷Lu-PSMA-617. However, we are entering an era where a broader understanding of the expression patterns of a wide variety of targets is beginning to be understood. As an example, PSMA PET has shown potential utility in the imaging of clear cell renal cell carcinoma and other malignancies with high rates of neovascularity. Many of the theranostic agents in the pipeline are designed to bind against "pan-cancer" targets such as fibrinogen-activating protein or the C-X-C motif chemokine receptor 4, or targets that become over-expressed in a broad variety of cancer states, such as those that target carbonic anhydrase IX. Artificial intelligence methods will assist us in appropriately selecting patients and in delivering predictive and prognostic imaging biomarkers. Given the wide potential applicability of emerging theranostic agents, it will be incumbent upon the field to carefully design clinical trials that will lead to regulatory approvals that will, in turn, permit broad use across a number of cancer types. In our future clinical practices, we will evolve to leverage precision medicine to identify target expression and deliver appropriate theranostic agents that are not limited to a specific cell of origin or disease state.