AI-Driven CT Scan Analysis Detects Early Kidney Damage From Cancer Treatments

Reviewed by Lexie Corner on May 1, 2025

Scientists from the Technical University of Munich (TUM) have created a technique to spot initial indications of kidney harm triggered by specific cancer therapies. By employing an AI-driven algorithm to examine CT scans, the researchers managed to discern patterns linked with treatment-induced renal strain.

The research team likewise noted comparable alterations in the spleen. This evidence might aid in making timely modifications to therapy protocols down the line, possibly decreasing the likelihood of harm to organs.

The research, carried out by groups from the hospital's imaging and nuclear medicine sections, examined information gathered from 121 individuals receiving lutetium-177 PSMA treatment for prostate cancer.

This fairly recent type of focused nuclear medication, referred to as radioligand therapy, holds potential for addressing particular kinds of tumors. Nonetheless, a conceivable adverse outcome could be a reduction in kidney performance throughout the course of treatment.

In a previous investigation, we discovered that patients experiencing declining kidney function following lutetium-177 PSMA treatment exhibited alterations in their kidney architecture. Given that it’s impractical to regularly obtain biopsy specimens, we aimed to determine if such modifications could be identified through more non-invasive techniques. .

Lead Researcher Dr. Lisa Steinhelfer, Technical University of Munich

Kidney volume could act as a biomarker.

Related Stories

Dr. Steinhelfer and her group chose a method designed with patient comfort in mind, avoiding extra steps. As blood tests and CT scans are integral parts of typical cancer treatment, the research team concentrated on detecting initial indications of kidney harm using this regularly gathered information.

They discovered that alterations in kidney size were a robust indicator of subsequent harm. Despite unreliability associated with aspects such as kidney length or the age of patients, a reduction in kidney size of 10% or greater during the initial half-year of therapy was closely linked to a considerable deterioration in renal performance throughout the ensuing six-month period.

“ 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 ," clarified Matthias Eiber, Professor and one of the study’s lead authors, along with Professor Rickmer Braren.

“ On the other hand, image analysis algorithms, when adequately trained, can consistently identify even these subtle alterations. ", noted Dr. Friederike Jungmann, who co-authored the paper as the first author alongside Dr. Steinhelfer.

This Method Might Be Beneficial for Various Cancer Treatments

Should it become evident that a patient faces an elevated risk of kidney impairment following six months of treatment, the frequency of therapy sessions as well as the dose may be tailored specifically to them. Doing so enables a more customized therapeutic strategy. .

Lead Author Dr. Lisa Steinhelfer, Technical University of Munich

Two ongoing observational studies at TUM University Hospital aim to further assess this method.

Previously, Dr. Steinhelfer’s group showed that alterations in the size of the spleen could be an early sign of disturbances in blood cell creation.

A number of cancer treatments can impact liver function or the blood cell production system. In my view, our method has the potential to detect a broad spectrum of treatment-induced side effects considerably sooner than what is presently feasible. .

Dr. Lisa Steinhelfer, Lead Author of the Study, Technical University of Munich

Journal Reference:

‌Steinhelfer, L., et al . (2025) Automated CT Measurement of Total Kidney Volume for Predicting Renal Function Decline after ¹⁷⁷ Lu Prostate-specific Membrane Antigen–I&T Radioligand Therapy. Radiology . doi.org/10.1148/radiol.240427 .

Technical University of Munich