Kidneys glow when they are injured, and the glow tracks treatment response

Design

Preclinical study across three mouse injury models with mechanistic work and ex vivo human tissue validation

Population

Mouse models of ureteral obstruction, folic acid nephropathy and ischaemia-reperfusion injury; 34 human nephrectomy samples

Primary outcome

Correlation of near-infrared autofluorescence with tubular epithelial injury, fibrosis and eGFR

Effect

Autofluorescence correlated with injury, oxidative stress and fibrosis in all three models; in human samples correlated inversely with eGFR and positively with PAS injury score

Assessing tubular injury means a biopsy, which cannot be repeated often enough to follow a disease or a drug. This study, published in June, tested whether near-infrared autofluorescence — light emitted by the tissue itself, with no injected probe — can do the job.

Across three mouse models (ureteral obstruction, folic acid nephropathy, ischaemia-reperfusion), autofluorescence tracked tubular epithelial injury, oxidative stress markers and fibrosis. Spectroscopy identified coproporphyrin III as the likely fluorophore, and the mechanism held together: coproporphyrinogen oxidase, the enzyme that metabolises its precursor, was reduced in injured kidneys and lost from damaged proximal tubular cells. Mice with a mutation reducing that enzyme had higher baseline signal. In slice culture, blocking mitochondrial complex III raised the signal and a mitochondrial antioxidant reversed it.

Two results reach towards the clinic. Finerenone treatment reduced autofluorescence in line with improved histology, showing the signal can track a therapeutic response. And in 34 human nephrectomy samples examined ex vivo, autofluorescence correlated inversely with eGFR and positively with injury score on periodic acid–Schiff staining. This is proof of concept, entirely preclinical, and years from a bedside device — but it is unusually complete as a mechanism.

• Nothing to change in clinic — this is preclinical work with ex vivo human tissue only.
• Note the direction of travel if you contribute to nephrology trials; repeatable non-invasive injury measures would change what endpoints are feasible.
• Keep biopsy as the standard for assessing tubulointerstitial injury.
• The mitochondrial oxidative stress mechanism is the interesting part and generalises beyond the imaging question.

Why it matters

Every nephrology trial is limited by the fact that the tissue of interest can only be sampled by needle.

Don't overread it

The human data are ex vivo tissue samples; no one has imaged a living human kidney this way.

The statistics, in plain English

The human component is 34 ex vivo samples from non-cancerous regions of nephrectomy specimens, correlated against eGFR and a histological score. Correlation in explanted tissue says nothing about whether the signal could be detected through skin and abdominal wall in a living patient, which is the entire clinical question. The mouse findings are consistent across three injury models, which is what makes the mechanism credible.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.