Exploratory Prognostic Biomarkers of Complement-Mediated Thrombotic Microangiopathy (CM-TMA) in Adults with Atypical Hemolytic Uremic Syndrome (aHUS): Analysis of a Phase III Study of Ravulizumab

Abstract

Background: Clinically validated biomarkers for monitoring of patients with complement-mediated thrombotic microangiopathy (CM-TMA) including atypical hemolytic uremic syndrome (aHUS) are unavailable. Improved characterization of biomarkers in patients with aHUS may inform treatment and monitoring for patients with CM-TMA.

Methods: This analysis used data collected from 55/56 (98.2 %) adult patients with aHUS enrolled in the global Phase III study of ravulizumab (NCT02949128). Baseline (pre-treatment) patient serum, plasma and urine biomarker levels were compared with the maximum observed levels in normal donors and evaluated for associations with pre-treatment plasma exchange/infusion and dialysis status. Biomarkers were also assessed for associations with key clinical measures at baseline and with changes at 26 and 52 weeks from treatment initiation via linear regression analyses.

Results: Complement-specific urine levels (factor Ba and sC5b-9) were elevated in >85 % of patients and are significantly associated with pre-treatment kidney dysfunction. Baseline levels of other evaluated biomarkers were elevated in >70 % of patients with aHUS, except for plasma sC5b-9 and serum sVCAM-1. Lower levels of urine complement markers at baseline are significantly associated with improvements in total urine protein and estimated glomerular filtration rate at 26 and 52 weeks of treatment. Clinical assessment of complement activation by a receiver operating characteristic analysis of Ba and sC5b-9 was more sensitive and specific in urine matrix than plasma.

Conclusion: This analysis identified a set of biomarkers that may show utility in the prognosis of CM-TMA, including their potential for measuring and predicting response to anti-C5 therapy. Further studies are required to enhance patient risk stratification and improve management of these vulnerable patients.

Clinical trials registration: NCT02949128, ClinicalTrials.gov.

Conflict of interest statement

All authors are employees and stockholders of Alexion, AstraZeneca Rare Disease, Boston, MA, USA.

© 2022. The Author(s).

Figures

Fig. 1

Complement cascade pathways. C complement protein, f factor

Fig. 2

Violin plots of biomarker observed levels over time to 52 weeks. A Biomarkers in blood. B Biomarkers in urine. Ravulizumab dose was determined by bodyweight and given at a loading dose at baseline, dose 2 at Day 15, maintenance dose at Day 71 and once every 8 weeks thereafter. Horizontal lines represent the 25 %, median and 75 % quartiles. P-values are calculated from a mixed model for repeated measures analysis with biomarker as dependent variable, and fixed categorical effect of visit and fixed continuous effect of baseline value as covariate. The null hypothesis that the mean change from baseline equals zero was tested against the alternative hypothesis that the mean change does not equal zero

Fig. 3

Line plot of eGFR and biomarker change from baseline over time by complete TMA response status at 52 weeks. A Biomarkers in blood. B Biomarkers in urine. Change from baseline for blood and urine biomarkers were compared to the clinical measure of eGFR out to 52 weeks. The p values are presented in figure for each of the biomarkers. P-value from a linear regression analysis with the 52-week change from baseline in eGFR as the dependent variable and log(2) of the baseline biomarker level and log(2) of the 52-week biomarker level as the independent variables. eGFR estimated glomerular filtration rate, TMA thrombotic microangiopathy

Fig. 4

Logistic regression analysis of complete TMA response at 52 weeks of treatment based on baseline biomarker levels. Odds ratios are derived from a logistic regression analysis with the response variable as the dependent variable and the log of the baseline biomarker level as the independent variable and represent the increased (or decreased) odds of achieving the efficacy response for every 2-fold increase in baseline biomarker. CI confidence interval, Cr creatinine, eGFR estimated glomerular filtration rate, LDH lactate dehydrogenase, TMA thrombotic microangiopathy

Fig. 5

Receiver operating characteristic curves of Ba and sC5b-9 levels in patients versus  NDs. A Dot plots for individual sC5b-9 and Ba levels in urine (ng/mg Cr) and plasma. B CombiROC curves for combined biomarkers in urine and plasma. C Violin plots of prediction probabilities for combined biomarkers in urine and plasma. D Summary table for combined biomarkers. Cr creatinine, FN false negative, FP false positive, ND normal donor, ROC receiver operating characteristic, TN true negative, TP true positive