The trial enrolled adult patients scheduled for complex cardiac procedures who were considered at high risk of losing kidney function. To qualify, participants had to have pre-existing renal impairment, defined by an estimated glomerular filtration rate between 25 and 65 mL/min/1.73 m², calculated using the 2021 CKD-EPI formula without race correction. Eligible operations were those associated with prolonged exposure to the heart-lung machine, including coronary artery bypass grafting with three or more distal anastomoses, combined bypass and valve surgery, or aortic valve surgery with aortic root replacement. Patients whose kidney risk stemmed from other mechanisms, such as cardiogenic shock, sepsis, endocarditis or deep hypothermic circulatory arrest, were excluded, sharpening the focus on surgery-driven injury.

Between December 2023 and September 2025, 244 patients were randomized across European cardiothoracic centers. Of these, 204 received both study infusions and formed the per-protocol analysis set: 109 were treated with ilofotase alfa and 95 with placebo. Each patient received two intravenous doses of 128 milligrams, the first within six hours before the initial incision and the second within two hours after the sternum was closed, a strategy intended to have the enzyme circulating before and immediately after the ischemic and inflammatory insult. The trial was stratified by baseline kidney function and type of surgery, and blinding was strictly maintained, even prohibiting serum alkaline phosphatase measurements within fourteen days of dosing to prevent accidental unmasking.

The primary endpoint was the serum creatinine ratio, the highest creatinine value measured within five postoperative days divided by the preoperative baseline. The investigators chose this continuous measure deliberately: unlike the binary diagnosis of acute kidney injury, it can detect smaller shifts in renal function, accounts for each patient’s starting point, and correlates with longer-term outcomes. In the ilofotase alfa group the mean ratio was 1.21, against 1.27 for placebo, a numerical difference of roughly five percent that carried no statistical weight (p=0.31). Daily creatinine trajectories overlapped almost completely across the first five days after surgery, and no subgroup, examined by eGFR, sex, surgical type or bypass duration, showed a clear therapeutic signal.

The secondary endpoint, major adverse kidney events through day 60, or MAKE60, told the same story. This composite captured a drop of at least 25 percent in eGFR relative to baseline at day 60, initiation of renal replacement therapy, or death within sixty days. It occurred in 15.9 percent of ilofotase alfa patients and 15.4 percent of placebo patients, with an adjusted odds ratio of 0.94 (p=0.87). A win-ratio analysis weighting mortality, dialysis and eGFR decline in clinical priority order also came back null at 0.95. Exploratory KDIGO-based acute kidney injury rates were 29.4 percent versus 35.8 percent, again not significant, and median intensive care stays were identical at three days in both arms.

Safety, at least, was reassuring. Adverse event rates were comparable between groups, with 61.6 percent of treated patients experiencing at least one adverse event and 18.8 percent a serious event, versus 64.8 and 19.4 percent on placebo. None of the events were judged related to the drug, and no second dose was ever withheld for safety reasons. Anti-drug antibodies appeared in 21 subjects, predominantly at very low titers and without any discernible effect on pharmacokinetics, ruling out immunogenic interference. Measurement of ilofotase alfa concentrations revealed a median of 450 ng/mL on the first postoperative day, well above the 170 ng/mL threshold previously associated with clinical benefit, and exposure-response analyses showed no correlation between drug levels and the creatinine ratio.

Why did a therapy that protected kidneys in animal models of ischemia-reperfusion and in septic patients fail here? The investigators systematically ruled out the most obvious explanations. Underdosing seems unlikely given the plasma concentrations achieved. Mistimed treatment is equally implausible, because elective surgery offers a predictable window of injury and patients had therapeutic drug levels before the first incision. An assessment window of five days is arguably short for a lagging biomarker like creatinine, yet the absence of any signal in MAKE60 at two months argues against a delayed benefit that the study simply missed, even though it was not powered for that endpoint.

That leaves biology. Cardiac surgery-associated acute kidney injury is a notoriously multifactorial entity, driven by hemodynamic swings, hemolysis, oxidative stress, complement activation, nephrotoxin exposure and the systemic inflammatory storm triggered by the cardiopulmonary bypass machine. Neutralizing damage-associated and pathogen-associated molecular patterns, the mechanism through which alkaline phosphatase exerts its detoxifying effect, may simply be insufficient to prevent downstream organ damage when so many injurious pathways operate in parallel. The contrast with sepsis-associated kidney injury, where ilofotase alfa enhanced creatinine clearance recovery in earlier phase 2 and phase 3 work, suggests these are fundamentally distinct clinical syndromes whose treatments may not transfer across settings.

There is one tantalizing thread left for future research. In subgroup analysis, patients with longer aortic cross-clamp times, and therefore longer periods of renal ischemia, showed a tendency toward benefit that did not reach statistical significance. Whether the very highest-risk subsets, perhaps those with cross-clamp times well beyond the mean of roughly one hour seen in this cohort, could still derive meaningful protection remains an open question. Detecting the modest five percent effect actually observed would demand more than 2,000 patients, raising hard questions about whether such a trial would be feasible or clinically worthwhile in the absence of longer-term improvement.

The study also highlights a methodological innovation with broader implications: the serum creatinine ratio as a sensitive, continuous trial endpoint that captures mild renal impairment below the diagnostic threshold of acute kidney injury. Small creatinine rises are known to predict chronic kidney disease and mortality, making this a potentially valuable tool for future nephroprotection studies. For now, however, the clinical message is clear. With roughly 30 percent of cardiac surgery patients still developing acute kidney injury, mortality reaching 60 percent among those needing dialysis, and no approved pharmacological prevention anywhere on the horizon, clinicians are left with supportive measures, meticulous hemodynamics, nephrotoxin avoidance and careful fluid management, while researchers confront the uncomfortable truth that inflammation is only one thread in a tangled pathophysiological web.

Subject of Research: A phase 2 trial testing ilofotase alfa for prevention of acute kidney injury after open heart surgery

Article Title: A multicenter randomized, double-blinded placebo-controlled phase 2 trial to evaluate safety and efficacy of ilofotase alfa in patients at risk for kidney injury following open heart surgery

Article References: A multicenter randomized, double-blinded placebo-controlled phase 2 trial to evaluate safety and efficacy of ilofotase alfa in patients at risk for kidney injury following open heart surgery. (n.d.). https://doi.org/10.1007/s00134-026-08596-y

Image Credits: AI Generated

DOI: 10.1007/s00134-026-08596-y

Keywords: acute kidney injury, cardiac surgery, ilofotase alfa, alkaline phosphatase, cardiopulmonary bypass, clinical trial, phase 2, MAKE60, serum creatinine, renal protection, intensive care, nephrology