arXiv subjects
Publications and source records attributed to Ross Berbeco.
Comparison of click chemistry and sortagging grafting strategies for functionalizing AGuIX nanoparticles with nanobodies to develop a tri-functional technology combining MRI imaging, radiotherapy, and immunotherapy by inhibiting immune checkpoints.
Background: The introduction of magnetic resonance (MR)-guided radiation treatment planning has opened a new space for theranostic nanoparticles to reduce acute toxicity while improving local control. In this work, second-generation AGuIX nanoparticles (AGuIX-Bi) are synthesized and validated. AGuIX-Bi are shown to maintain MR positive contrast while further amplifying the radiation dose by the replacement of some Gd$^{3+}$ cations with higher Z Bi$^{3+}$. These next-generation nanoparticles are based on the AGuIX platform, which is currently being evaluated in multiple Phase II clinical trials in combination with radiotherapy. Methods: In this clinically scalable methodology, AGuIX is used as an initial chelation platform to exchange Gd$^{3+}$ for Bi$^{3+}$. AGuIX-Bi nanoparticles are synthesized with three ratios of Gd/Bi, each maintaining MR contrast while further amplifying radiation dose relative to Bi$^{3+}$. Safety, efficacy, and theranostic potential of the nanoparticles were evaluated in vitro and in vivo in a human non-small cell lung cancer model. Results: We demonstrated that increasing Bi$^{3+}$ in the nanoparticles is associated with more DNA damage and improves in vivo efficacy with a statistically significant delay in tumor growth and 33% complete regression for the largest Bi/Gd ratio tested. The addition of Bi$^{3+}$ by our synthetic method leads to nanoparticles that present slightly altered pharmacokinetics and lengthening of the period of high tumor accumulation with no observed evidence of toxicity. Conclusions: We confirmed the safety and enhanced efficacy of AGuIX-Bi with radiation therapy at the selected ratio of 30Gd/70Bi. These results provide crucial evidence towards patient translation.
Lung tumor tracking for radiotherapy requires real-time, multiple-step ahead forecasting of a quasi-periodic time series recording instantaneous tumor locations. We introduce a location-mixture autoregressive (LMAR) process that admits multimodal conditional distributions, fast approximate inference using the EM algorithm and accurate multiple-step ahead predictive distributions. LMAR outperforms several commonly used methods in terms of out-of-sample prediction accuracy using clinical data from lung tumor patients. With its superior predictive performance and real-time computation, the LMAR model could be effectively implemented for use in current tumor tracking systems.