One of the most impressive aspects of the SNAP trial is its ability to expand. If you are going to conduct an international, prospective trial to get definitive answers about the treatment of Staphylococcus aureus bacteraemia, why not also try to start a new conversation about diagnosis of acute kidney injury in these patients, especially given the significance of AKI in the SNAP origin story (click here to read). We put this to the SNAP trial committee, and thankfully, they agreed.
Many of you are likely already aware about issues with current AKI diagnosis based on creatinine, it’s slow (GFR can decrease by 50% before creatinine rises, or after damage has been occurring for at least 48 hours), it’s affected by non-renal factors (eg diet, drugs, age), and the small absolute changes in creatinine to diagnose stage 1 AKI can be within the creatinine reference range.
So, what can we do and where does SNAP fit in? Firstly, we can look for creatinine alternatives. Urinary biomarkers represent a novel approach to AKI diagnosis; using specific molecules released from the kidney itself that can reveal information about renal injury, kidney function (glomerular filtration), tubule function (reabsorption of filtered molecules). The dream scenario is one where a test will identify sub-clinical AKI, so as to identify those that would go on to have an AKI diagnosed by current clinical criteria. In this situation, interventions can be instituted earlier, ie stopping nephrotoxic drugs, aiming to reduce the significant downstream complications of an AKI. In the SNAP trial, we plan to run a sub-study of patients (about 100) likely to experience AKI and collect their urine daily from platform entry for 5 days. Then we will observe what biomarkers are present in the urine and at what concentrations, especially comparing patients that did and did not have an AKI, and the timing of biomarker release compared to creatinine elevation for those that did have an AKI.
Even in 2021, it remains to be seen which is the best biomarker for this clinical scenario. There are many undergoing development, in many clinical contexts (eg septic patients in ICU, cardiothoracic surgery, so we have chosen a biomarker panel based on pre-clinical data and available for immediate clinical translation. Our proposed urinary panel is: KIM-1, NGAL, clusterin, glycocalyx, albumin and TIMP1*IGFBP7, all normalised to creatinine.
Kidney injury molecule-1 (KIM-1) is released from the proximal tubule into urine during kidney injury. KIM-1 has been shown in animal models to be the most sensitive predictor of nephrotoxicity from vancomycin, and can be detected early (ie day 1) in the treatment course. Because KIM-1 is released from the proximal tubule, where vancomycin-induced kidney damage is thought to occur, it is a promising candidate for use in early identification of vancomycin-induced kidney injury.
Urinary neutrophil gelatinase-associated lipocalin (NGAL) is released from the proximal and distal tubules and is elevated in ischemic or nephrotoxic kidney injury. In patients receiving vancomycin, NGAL detection was around 2 days earlier than creatinine elevation.
Clusterin (found in proximal and distal tubules) was also predictive of histopathological renal damage in an animal model of vancomycin-induced kidney injury. Clusterin is expressed in many physiologic fluids, but it is not freely filtered into the kidney so urinary concentrations reflect injury from renal tubule cells. Although clusterin data are derived from animal models, it has potential as an early marker of renal injury and may reflect more widespread injury compared with KIM-1.
TIMP1*IGFBP7 is a promising and already approved urinary biomarker that is expressed in renal tubular cells during periods of cellular stress or injury, released when the cell cycle is halted to prevent cell division during damage. While TIMP1*IGFBP7 is approved for adult ICU patients with cardiovascular and or respiratory compromise, it has not yet been studied for vancomycin induced kidney injury and methodologic reasons have limited pre-clinical study.
Albumin (to creatinine ratio) and glycocalyx components (predictive of AKI in malaria studies) will round out urinary biomarker panel.
So, watch this space, though it might be the beginning to the biomarker story in SAB, it will be one worth following – after all, with the SNAP study ur-in(e)-good hands!