Radiographic Response Assessment Strategies for Early-Phase Brain Trials in Complex Tumor Types and Drug Combinations: from Digital “Flipbooks” to Control Systems Theory

• Published in: Neurotherapeutics 2022-10, Vol.19 (6), p.1855-1868
• Main Authors: Ellingson, Benjamin M., Levin, Victor A., Cloughesy, Timothy F.
• Format: Article
• Language: English
• Subjects: Biomedical and Life Sciences; Biomedicine; Brain - pathology; Brain cancer; Brain Neoplasms - diagnostic imaging; Brain Neoplasms - drug therapy; Brain Neoplasms - pathology; Brain tumors; Clinical trials; Drug Combinations; Drug development; Glioma; Glioma - diagnostic imaging; Glioma - drug therapy; Glioma - pathology; Growth rate; Humans; Magnetic Resonance Imaging; Neurobiology; Neurology; Neurosciences; Neurosurgery; Review; System theory; Systems Theory; Tumors
• ISSN: 1933-7213; 1878-7479; 1878-7479
• DOI: 10.1007/s13311-022-01241-8

There is an urgent need for drug development in brain tumors. While current radiographic response assessment provides instructions for identifying large treatment effects in simple high- and low-grade gliomas, there remains a void of strategies to evaluate complex or difficult to measure tumors or tumors of mixed grade with enhancing and non-enhancing components. Furthermore, most patients exhibit some period of alteration in tumor growth after starting a new therapy, but simple response categorization (e.g., stable disease, progressive disease) fails to provide any meaningful insight into the depth or degree of potential “subclinical” therapeutic response. We propose a creative solution to these issues based on a tiered strategy meant to increase confidence in identifying therapeutic effects even in the most challenging tumor types, while also providing a framework for complex evaluation of combination and sequential treatment schemes. Specifically, we demonstrate the utility of digital “flipbooks” to quickly identify subtle changes in complex tumors. We show how a modified Levin criteria can be used to quantify the degree of visual changes, while establishing estimates of the association between tumor volume and visual inspection. Lastly, we introduce the concept of quantifying therapeutic response using control systems theory. We propose measuring changes in volume (proportional), the area under the volume vs. time curve (integral) and changes in growth rates (derivative) to utilize a “PID” controller model of single or combination therapeutic activity.