Genetically Engineered Nanoparticles of Asymmetric Triblock Polypeptide with a Platinum(IV) Cargo Outperforms a Platinum(II) Analog and Free Drug in a Murine Cancer Model.

The development of platinum(Pt)-drugs for cancer therapy has stalled, as no new Pt-drugs have been approved in over a decade. Packaging small molecule drugs into nanoparticles is a way to enhance their therapeutic efficacy. To date, there has been no direct comparison of relative merits of the choice of Pt oxidation state in the same nanoparticle system that would allow its optimal design. To address this lacuna, we designed a recombinant asymmetric triblock polypeptide (ATBP) that self-assembles into rod-shaped micelles and chelates Pt(II) or enables covalent conjugation of Pt(IV) with similar morphology and stability. Both ATBP-Pt(II) and ATBP-Pt(IV) nanoparticles enhanced the half-life of Pt by ∼45-fold, but ATBP-Pt(IV) had superior tumor regression efficacy compared to ATBP-Pt(II) and cisplatin. These results suggest loading Pt(IV) into genetically engineered nanoparticles may yield a new generation of more effective platinum-drug nanoformulations.

Machine learning approaches in non-contact autofluorescence spectrum classification.

Manual surgical resection of soft tissue sarcoma tissue can involve many challenges, including the critical need for precise determination of tumor boundary with normal tissue and limitations of current surgical instrumentation, in addition to standard risks of infection or tissue healing difficulty. Substantial research has been conducted in the biomedical sensing landscape for development of non-human contact sensing devices. One such point-of-care platform, previously devised by our group, utilizes autofluorescence-based spectroscopic signatures to highlight important physiological differences in tumorous and healthy tissue. The following study builds on this work, implementing classification algorithms, including Artificial Neural Network, Support Vector Machine, Logistic Regression, and K-Nearest Neighbors, to diagnose freshly resected murine tissue as sarcoma or healthy. Classification accuracies of over 93% are achieved with Logistic Regression, and Area Under the Curve scores over 94% are achieved with Support Vector Machines, delineating a clear way to automate photonic diagnosis of ambiguous tissue in assistance of surgeons. These interpretable algorithms can also be linked to important physiological diagnostic indicators, unlike the black-box ANN architecture. This is the first known study to use machine learning to interpret data from a non-contact autofluorescence sensing device on sarcoma tissue, and has direct applications in rapid intraoperative sensing.