Detecting drug-induced nephrotoxicity using simultaneous label-free autofluorescence multiharmonic microscopy

Aneesh Alex1,2, Jindou Shi1,3,4, Eric J Chaney1,3

  • 1GSK Center for Optical Molecular Imaging, Beckman Institute for Advanced Science and Technology, University of Illinois Urbana-Champaign, Urbana, IL 61801, United States.

Insights

Simultaneous Label-free Autofluorescence Multiharmonic (SLAM) microscopy offers a rapid, label-free method to detect drug-induced nephrotoxicity. This technique reveals region-specific kidney injury patterns in preclinical studies.

Area of Science:

  • Biomedical Imaging
  • Toxicology
  • Drug Development

Background:

  • Drug-induced nephrotoxicity (DIN) is a significant hurdle in drug development, often detected late due to limitations in conventional methods.
  • Early and region-specific detection of kidney injury is crucial for accurate safety assessment.

Purpose of the Study:

  • To evaluate Simultaneous Label-free Autofluorescence Multiharmonic (SLAM) microscopy for quantifying kidney microstructural and metabolic changes in drug-induced nephrotoxicity.
  • To assess the sensitivity and specificity of SLAM for detecting region-specific kidney injury.

Main Methods:

  • SLAM microscopy was used to image rat kidneys post-cisplatin administration, capturing autofluorescence (NAD(P)H, FAD) and harmonic signals.
  • Images were acquired from the cortex and outer medulla (OM) at Days 1, 6, and 29.
  • Feature-based classification models were developed to analyze spatial and temporal injury patterns.

Main Results:

  • SLAM imaging identified region-specific nephrotoxicity, with the OM being more sensitive than the cortex.
  • Key injury markers like tubular degeneration and hyaline casts peaked at Day 6, while tubular dilation and fibrosis persisted.
  • Classification models achieved high balanced accuracy (OM: 0.944, Cortex: 0.860), driven by granularity and inter-channel correlations.

Conclusions:

  • SLAM microscopy provides a sensitive, label-free approach for early, region-specific detection and characterization of drug-induced nephrotoxicity.
  • The multi-channel data from SLAM is valuable for identifying distinct injury patterns in preclinical safety evaluations.

Related Concept Videos