Thanks to an AI-based algorithm, a decrease in kidney volume can be detected on CT scans months before clinical symptoms appear. This subtle change may serve as an early biomarker for impending organ damage. A research team at the Technical University of Munich (TUM) has developed this promising approach to help predict declining kidney function caused by certain cancer therapies.
The current study focused on 121 patients with prostate cancer who were treated with the radioligand therapy Lutetium-177 PSMA. This advanced form of nuclear medicine is considered highly promising for specific tumors but can potentially impair kidney function.
“In an earlier study, we found that patients whose kidney function worsened after Lutetium-177 PSMA therapy showed changes in kidney tissue,” explains Dr. Lisa Steinhelfer, lead author of the study. “Since it's not feasible to routinely take tissue samples, we wanted to explore whether these changes could be detected using less invasive methods.”
The researchers chose a non-invasive approach based on routinely collected data during cancer therapy: CT imaging and blood tests. Their analysis revealed that a reduction in kidney volume was a particularly reliable early indicator. Specifically, if kidney volume had decreased by ten percent or more within six months of starting treatment, patients were highly likely to show significantly impaired kidney function six months later.
“These changes in kidney volume are very subtle. They can easily be missed during routine image assessments because clinicians are mainly focused on tracking tumors and other critical findings,” says Prof. Matthias Eiber, who co-authored the study with Prof. Rickmer Braren. “I
In contrast, image analysis algorithms – if properly trained – can reliably detect even these minor changes,” adds Dr. Friederike Jungmann, who, like Dr. Lisa Steinhelfer, is a first author of the study.