Biomarkers to predict disease progression and therapeutic response in isolated methylmalonic acidemia

Abstract

Methylmalonic Acidemia (MMA) is a heterogenous group of inborn errors of metabolism caused by a defect in the methylmalonyl-CoA mutase (MMUT) enzyme or the synthesis and transport of its cofactor, 5'-deoxy-adenosylcobalamin. It is characterized by life-threatening episodes of ketoacidosis, chronic kidney disease, and other multiorgan complications. Liver transplantation can improve patient stability and survival and thus provides clinical and biochemical benchmarks for the development of hepatocyte-targeted genomic therapies. Data are presented from a US natural history protocol that evaluated subjects with different types of MMA including mut-type (N = 91), cblB-type (15), and cblA-type MMA (17), as well as from an Italian cohort of mut-type (N = 19) and cblB-type MMA (N = 2) subjects, including data before and after organ transplantation in both cohorts. Canonical metabolic markers, such as serum methylmalonic acid and propionylcarnitine, are variable and affected by dietary intake and renal function. We have therefore explored the use of the 1-¹³ C-propionate oxidation breath test (POBT) to measure metabolic capacity and the changes in circulating proteins to assess mitochondrial dysfunction (fibroblast growth factor 21 [FGF21] and growth differentiation factor 15 [GDF15]) and kidney injury (lipocalin-2 [LCN2]). Biomarker concentrations are higher in patients with the severe mut⁰ -type and cblB-type MMA, correlate with a decreased POBT, and show a significant response postliver transplant. Additional circulating and imaging markers to assess disease burden are necessary to monitor disease progression. A combination of biomarkers reflecting disease severity and multisystem involvement will be needed to help stratify patients for clinical trials and assess the efficacy of new therapies for MMA.

Trial registration: ClinicalTrials.gov NCT00078078.

Keywords: FGF21; GDF15; biomarker; clinical trial endpoint; methylmalonic acidemia; propionate oxidation; stable isotope.

Figure 1:. The main biochemical pathways involved in isolated methylmalonic acidemia.

Pathways surrounding the metabolic block in the mitochondrial enzyme methylmalonyl-CoA mutase (encoded by the MMUT gene), including the Krebs cycle, urea cycle, electron transport chain and glycine cleavage system. Yellow boxes contain circulating biomarkers covered in this review. The 1-¹³C-propionate oxidation breath test (POBT) is shown in gray. 2-MCA: 2-methylcitrate; FGF21: Fibroblast growth factor-21; GDF15: Growth differentiation factor-15; 3-OHP: 3-hydroxypropionate; C3: propionylcarnitine; sMMA: serum MMA; C4DC: C4-dicarboxylic or methylmalonic/succinylcarnitine; NH3: ammonia; ROS: reactive oxygen species. Created with BioRender.com

Figure 2:. Main circulating biomarkers by MMA subtype (NIH cohort).

(A) Serum methylmalonic acid concentrations are depicted for isolated methylmalonic acidemia (MMA) subjects, per disease subtype, where N = subjects, n = data points [mut⁰ (N = 50, n = 219), mut⁻ (N = 15, n = 36), cblB (N = 11, n = 25), cblA (N = 16, n = 74) and controls (N=139 heterozygote parents)]. The graphs contain previously published data^,,, and 26% additional patients and 3.5 times more measurements. mut⁰ subjects had a wide range of sMMA values with a mean ± SD of 1587 ± 1792μmol/L (P<0.0001 compared to mut⁻ 130.5 ± 106.0, cblA 46.98 ± 32.91 and controls 0.195 ± 0.09, and P=NS to cblB subjects 647.1 ± 559.5. Reference values: <0.40μmol/L). The cblB sMMA values that were significantly elevated in comparison to the mut⁻ (P=0.0091) and cblA (P<0.0001). (B) The acyl/free carnitine ratio values are shown per disease subtype [mut⁰ (N = 49, n = 141), mut⁻ (N = 14, n = 33), cblB (N = 9, n = 17), cblA (N = 16, n = 58)]. Acyl/free carnitine ratio values were 3.740 ± 5.641 (mean ± SD) in mut⁰ subjects (P<0.0001 compared with mut⁻ 0.9718 ± 0.8206 and cblA 0.3589 ± 0.1825, and P=NS to cblB 2.758 ± 2.954. Reference values: 0.1–0.9). cblA patients’ values were significantly lower in comparison to the mut⁻ (P=0.0028) and cblB (P<0.0001). (C) The glycine levels are displayed per disease subtype [mut⁰ (N = 55, n = 229), mut⁻ (N = 15, n = 36), cblB (N = 10, n = 20), cblA (N = 16, n = 65)]. The mut⁰ patients had glycine levels of 606.4 ± 304.7 nmol/ml (mean ± SD, P<0.0001 compared with cblA 328.1 ± 112.0, P=0.0076 mut⁻ 443.5 ± 175.2, and P=NS to cblB 546.9 ± 310.1. Reference values (age 2–17 years): 149–417 nmol/ml). cblA patients had values that were significantly lower in comparison to the mut⁻ (P=0.0285) and cblB (P<0.0044). (D) The fibroblast growth factor-21 (FGF21) levels are provided for mut⁰ (N = 56, n = 92), mut⁻ (N = 9, n = 13), cblB (N = 10, n = 19), cblA (N = 14, n = 31) and Control subjects (N=12 heterozygote parents). The severe mut⁰ patients had a mean ± SD of 6987 ± 8134pg/ml (P<0.0001 compared with mut⁻ 1152 ± 1311, cblA 597.2 ± 508.1, controls 73.67 ± 58.9, and P=NS to cblB 3495 ± 4523. Reference range by manufacturer: ND-1012). The cblB patients had levels that were significantly elevated in comparison to cblA patients (P=0.0061). (E) The growth differentiation factor-15 (GDF15) levels are displayed per disease subtype [mut⁰ (N = 47, n = 65), mut⁻ (N = 9, n = 10), cblB (N = 10, n = 10), cblA (N = 14, n = 15)]. The mut⁰ patients had a mean ± SD of 3271 ± 2787pg/ml (P<0.0001 compared with cblA 598.7 ± 478.3, and P=NS to mut⁻ 1691 ± 2491 and cblB 2084 ± 1246. Reference range by the manufacturer: 278–1064). (F) Alanine concentrations are shown per disease subtype in same subjects as for glycine above, [mut0 patients had a mean ± SD of 495.8 ± 204.2nmol/ml, mut⁻ 513.6 ± 176.2, cblB 472.5 ± 147.2 and cblA 445.5 ± 154.4, Reference values (ages 2–17 years): 144–557 nmol/ml). (G) Cystatin-C is presented for [mut⁰ (N = 42, n = 106), mut⁻ (N = 12, n = 25), cblB (N = 9, n = 19), cblA (N = 16, n = 38). Mean values were 1.719 ± 0.75 for mut⁰, 1.16 ± 0.76 for mut⁻, 1.257 ± 0.6 for cblB and 1.138 ± 0.6 for cblA subjects. Reference values for >18 years old: 0.61–0.95 mg/L, reference ranges have not been established for pediatric subjects (H) The Lipocalin 2 (LCN2) levels are depicted per MMA subtype [mut⁰ (N = 53, n = 80), mut⁻ (N = 8, n = 10), cblB (N = 6, n = 12), cblA (N = 8, n = 16)]. The severe mut⁰ patients had a mean ± SD of 122.5 ± 129.1 ng/ml (P=NS compared with mut⁻ 67.84 ± 93.63, cblB 145.4 ± 103.7, and with cblA 99.41 ± 73.38). Significance is noted as follows: ***P<0.0001, **P<0.001, *P<0.01.

Figure 3:. Circulating biomarkers per MMA subtype (OPBG cohort).

Serum methylmalonic acid levels are depicted for isolated methylmalonic acidemia (MMA) subjects, per disease subtype, where N = subjects, n = data points [mut⁰ (N = 13, n = 239), mut⁻ (N = 6, n = 91), cblB (N = 2, n = 45)]. The graphs contain published data from 9 patients^, and 12 additional patients and over 200 more multiple measurements. (A) sMMA concentrations in the mut⁰ subgroup were 1525 ± 1506 μmol/L (mean ± SD, P<0.0001 compared to mut⁻ 40.76 ± 37.46 and P=NS compared to cblB 1619 ± 2053). (B) 2-Methylcitric acid (MCA) levels for the mut⁰ subgroup were 61.24 ± 46.16 μmol/L, compared to 8.151 ± 5.001 in the mut⁻ (P<0.0001) and 30.58 ± 18.75 in the cblB subjects (P=0.0057). (C) Glycine levels had a mean ± SD of 588.2 ± 329 U/L in the mut⁰ subjects, compared to 293.9 ± 69.23 in the mut⁻ (P<0.0001), and 401.5 ± 212.6 in cblB subjects (P<0.0001). (D) FGF21 concentrations had a wide range with a mean ± SD of 3271 ± 1378 pg/ml in mut⁰ subjects, 467.2 ± 427.5 in mut⁻ (P<0.0001) and 2522 ± 1187 in cblB subjects (P=NS). (E) GDF15 levels had a mean ± SD of 2617 ± 1072 pg/ml in mut⁰ and 596.2 ± 198.2 in mut⁻ subjects (P<0.0001). Lastly, (F) alanine concentration mean ± SD were 490.4 ± 179.7 in mut0, 394.9 ± 117.0 in mut- (P<0.0001) and 415.3 ± 130.9 in cblB subjects (P=0.06).

Figure 4:. Main circulating biomarkers in transplanted patients (NIH cohort).

(A) The serum methylmalonic acid concentrations are depicted for the severe mut⁰ patients, per transplant status [liver, LT (N = 5, n = 15), liver and kidney, LKT (N = 13, n = 69), isolated kidney, KT (N = 5, n = 18)]. Patients with isolated KT had the highest sMMA values with a 2085 ± 3146μmol/L, which includes patients with progressive decline in transplanted kidney function due to episodes of rejection (mean ± SD, P<0.0001 compared to LT 271.7 ± 255.2 or LKT 215.1 ± 139.4). (B) The acyl/free carnitine ratio per transplant status [LT (N = 5, n = 10), LKT (N = 15, n = 53), KT (N = 4, n = 13)] was higher in patients with KT, 1.466 ± 1.021 (mean ± SD), as opposed to patients with LT, 1.092 ± 0.3626, and those with LKT 1.077 ± 0.7186 (P=NS). (C) The glycine levels are displayed for the severe mut⁰ patients, per transplant status [LT (N = 4, n = 9), LKT (N = 15, n = 73), KT (N = 6, n = 10)]. The patients with KT had a range of glycine levels with a mean ± SD 583.5 ± 252.2, patients with LT had 440.4 ± 181.7 and those with LKT 473.1 ± 191.0 (P=NS). (D) Fibroblast growth factor-21 (FGF21) levels are displayed for severe mut⁰ patients, per transplant status [LT (N = 4, n = 8), LKT (N = 14, n = 24), KT (N = 4, n = 13)]. The patients with KT had FGF21 concentrations averaging 4247 ± 3631pg/ml (P<0.0001 compared to LKT recipients, 528.6 ± 432.7 and P=0.001 compared to patients with isolated LT 376.0 ± 185.3). (E) The growth differentiation factor-15 (GDF15) levels [LT (N = 4, n = 6), LKT (N = 12, n = 25), KT (N = 4, n = 7)] were again higher in isolated KT recipients, 3031 ± 3065pg/ml (mean ± SD), than patients with LT, 1512 ± 674.3, or those with LKT 1324 ± 924.5 (P=NS). (F) Alanine concentrations mean ± SD were 510 ± 141.7 for LT, 456.5 ± 167.8 for LKT and 534.0 ± 225.6 for isolated KT recipients (P=NS). (G) Cystatin-C and (H) Lipocalin-2 (LCN2) concentrations were low in all transplant recipients. Cystatin-C levels were 1.3 ± 0.44 mg/L in LT, 1.33 ± 0.37 in LKT and 1.84 ± 0.80 in KT recipients (P=0.0057 compared to LKT); LCN2, was 85.28 ± 98.49 ng/ml in patients with LT, 86.96 ± 65.80 with LKT, and 108.7 ± 135.5 in those with KT(P=NS).

Figure 5:. Biomarkers pre and post liver transplant (NIH and OPBG cohort)

Changes in serum biomarkers following a liver or liver and kidney transplant with the latest available levels pre and at the earliest follow-up visit post transplantation, obtained under the NIH natural history protocol, are depicted (clinicaltrials.gov identifier: NCT00078078). Twelve patients received a combined liver and kidney transplant (black circles), while two patients received an isolated liver transplant (gray squares). (A) Serum MMA levels of mut⁰ patients decreased significantly following liver or liver and kidney combined transplantation (***P=0.0001). (B) Similarly, the serum acyl/free carnitine levels decreased significantly following liver or liver and kidney combined transplantation (***P=0.0005). (C) The glycine levels were not significantly different following liver or liver and kidney combined transplantation (P=NS). (D) Both, the FGF21 (***P=0.0001) and (E) GDF15 levels decreased significantly following liver or liver and kidney combined transplantation (***P=0.0005). (F) The cystatin-C levels of severe mut⁰ patients were not significantly different following liver or liver and kidney combined transplantation (P=NS). (G-K) Individual patient longitudinal sMMA, MCA and FGF21 trends following isolated liver, liver and kidney or isolated kidney transplantation in the OPBG study cohort are displayed in panels. The most significant and sustained decrease was observed in the combined LKT recipients (panel H). Significance is noted as follows: ***P<0.0001, **P<0.001, *P<0.01.

Figure 6:. Monitoring of organ system disease burden.

A depiction of multiorgan involvement and recommended laboratory testing and other assessments for long-term patient monitoring. Created with BioRender.com