Avasopasem manganese (GC4419) protects against cisplatin-induced chronic kidney disease: An exploratory analysis of renal metrics from a randomized phase 2b clinical trial in head and neck cancer patients

Abstract

Head and neck squamous cell carcinoma (HNSCC) patients treated with high-dose cisplatin concurrently with radiotherapy (hdCis-RT) commonly suffer kidney injury leading to acute and chronic kidney disease (AKD and CKD, respectively). We conducted a retrospective analysis of renal function and kidney injury-related plasma biomarkers in a subset of HNSCC subjects receiving hdCis-RT in a double-blinded, placebo-controlled clinical trial (NCT02508389) evaluating the superoxide dismutase mimetic, avasopasem manganese (AVA), an investigational new drug. We found that 90 mg AVA treatment prevented a significant reduction in estimated glomerular filtration rate (eGFR) three months as well as six and twelve months after treatment compared to 30 mg AVA and placebo. Moreover, AVA treatment may have allowed renal repair in the first 22 days following cisplatin treatment as evidenced by an increase in epithelial growth factor (EGF), known to aid in renal recovery. An upward trend was also observed in plasma iron homeostasis proteins including total iron (Fe-blood) and iron saturation (Fe-saturation) in the 90 mg AVA group versus placebo. These data support the hypothesis that treatment with 90 mg AVA mitigates cisplatin-induced CKD by inhibiting hdCis-induced renal changes and promoting renal recovery.

Keywords: Acute kidney disease; Avasopasem manganese; Chronic kidney disease; Cisplatin; Epithelial growth factor; GC4419; Iron; Kidney injury; Radiation; Superoxide dismutase mimetic; eGFR.

Graphical abstract

Illustration 1

Treatment Timeline and sample collection.

Fig. 1

Mean changes in renal metrics up to one year following hdCis-RT. Mean change in eGFR (1A.), Creatinine (1B.), and BUN (1C.) from pre-hdCis-RT (D0) and p-values for the comparison between treatment groups at noted time points (* = p < 0.05). Highlighted values indicate a statistically significant difference in mean change from D0 between treatment groups. Changes in renal metrics (eGFR, Creatinine, BUN) were calculated from D0 to each subsequent assessment time point (Day 22, Day 43, Month 3, Month 6, Month 12). Linear mixed-effects regression models were used to evaluate changes over time between treatment groups using SAS v9.4 (SAS Institute, Cary, NC). Random effects were included to account for the longitudinally correlated nature of repeated assessments at unequal time spacing between visits with a spatial power correlation structure.

Fig. 2

Changes in renal metrics in AKD and CKD phase of renal injury. (2A). Estimated glomerular Filtration Rate (eGFR), Creatinine, and Blood Urea Nitrogen (BUN) levels were evaluated pre-hdCis-RT (D0), 3 weeks (D22), and 6 weeks (D43), and 3 months (M3) post-RT. The gray lines indicate each individual patient's trajectory whereas, the red line indicates the estimated mean along with the 95% confidence interval (CI). (2B). Estimated glomerular Filtration Rate (eGFR), Creatinine, and Blood Urea Nitrogen (BUN) levels were evaluated pre-hdCis-RT (D0), 6 months (M6), and 12 months (M12) post-RT. The gray lines indicate each individual patient's trajectory whereas, the red line indicates the estimated mean along with the 95% confidence interval (CI). Changes in renal metrics (BUN, creatinine, eGFR) were calculated from D0 to each subsequent assessment time point (Day 22, Day 43, Month 3, Month 6, Month 12). Linear mixed-effects regression models were used to evaluate changes over time between treatment groups using SAS v9.4 (SAS Institute, Cary, NC). Random effects were included to account for the longitudinally correlated nature of repeated assessments at unequal time spacing between visits with a spatial power correlation structure. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 3

Expression of functional renal biomarkers in patients treated with placebo and 90 mg AVA serum samples. (3A.) Mean changes in Epidermal Growth Factor (EGF expression in Placebo vs. 90 mg AVA at D0 and D22. Mean changes in levels of kidney biomarkers using an MSD sandwich immunoassay with U-PLEX platform from Mesoscale including (3B.) Tumor Necrosis Factor Receptor 1 (TNFR1), (3C.) Tumor Necrosis Factor Receptor 2 (TNFR2), (3D.) Cystatin C, and (3E.) Kidney Injury Molecule 1 (KIM1) and (3F.) Neutrophil Gelatinase-Associated Lipocalin (NGAL). Although no baseline differences were observed in the different biomarkers (3G.), the mean changes in EGF levels (3A.) between the placebo vs. 90 mg AVA-treated group were significantly different with a p-value of <0.01 (3H.). Repeated measures ANOVAs were used to evaluate mean changes from D0 to D22 between treatment groups using SAS v9.4 (SAS Institute, Cary, NC).

Fig. 4

Expression of functional renal biomarkers in patients treated with placebo and 90 mg AVA serum samples. Changes in levels of kidney biomarkers using an MSD sandwich immunoassay with U-PLEX platform from Mesoscale including (4A.) Osteoactivin, (4B.) Osteopontin, and (4C.) Uromodulin. No statistically significant differences were observed (4D.). Repeated measures ANOVAs were used to evaluate mean changes from D0 to D22 between treatment groups using SAS v9.4 (SAS Institute, Cary, NC).

Fig. 5

Changes in oxidative damage endpoints. Oxidative markers were assessed in serum samples using a dot blot and were assayed for protein carbonyls (5A. and 5B.), 4HNE modified proteins (4-Hydroxynonenal) (5C-5E.), and 3-Nitrotyrosine (3NT) (5F–5H.). Standard curves were generated for each of the three damage endpoints using increasing concentrations of Carbonyl (5A.), 2 mg/mL Transferrin spiked with increasing concentrations of Peroxynitrite (5C.), and 2 mg/mL Transferrin spiked with increasing concentrations of pure 4-HNE (5F.). (P₁ – P₉) denotes the placebo-treated patients whereas, (90₁- 90₉) denotes 90 mg AVA-treated patient serums in panels (5C.) and (5F.). Each patient sample has a set of two bands with the first band representing the sample from D₀ and the second band representing the sample collected on D₂₂ (5C. and 5F.). Blots for 3-Nitrotyrosine and 4-HNE were stained with Ponceau as a protein loading control (5D.) and (5G.) respectively. Quantification for 3-Nitrotyrosine and 4-HNE normalized to the protein is depicted in panels (5E.) and (5H.) respectively. Repeated measures ANOVAs were used to evaluate mean changes from D0 to D22 between treatment groups using SAS v9.4 (SAS Institute, Cary, NC).

Fig. 6

Analysis of changes in proteins involved in iron metabolism in placebo vs. 90 mg AVA treated groups. Systemic changes in iron homeostasis were assessed by analyzing changes in levels of iron metabolic proteins including (6A.) Serum Transferrin, (6B.) Total Iron Binding Capacity (TIBC), (6C.) Total iron (Fe-Blood), (6D.) Iron Saturation levels (Fe-Saturation), and (6E.) Ferritin. No statistically significant differences between treatment groups were noted, but some statistically significant mean changes were noted within treatment groups (*p < 0.05) (6F.). Repeated measures ANOVAs were used to evaluate mean changes from D0 to D22 between treatment groups using SAS v9.4 (SAS Institute, Cary, NC).