Plasma C24:0- and C26:0-lysophosphatidylcholines are reliable biomarkers for the diagnosis of peroxisomal β-oxidation disorders

Abstract

The gold-standard diagnostic test for peroxisomal disorders (PDs) is plasma concentration analysis of very long-chain fatty acids (VLCFAs). However, this method's time-consuming nature and limitations in cases which present normal VLCFA levels necessitates alternative approaches. The analysis of C26:0-lysophosphatydylcholine (C26:0-LPC) in dried blood spot samples by tandem-mass spectrometry (MS/MS) has successfully been implemented in certain newborn screening programs to diagnose X-linked adrenoleukodystrophy (ALD). However, the diagnostic potential of very long-chain LPCs concentrations in plasma remains poorly understood. This study sought to evaluate the diagnostic performance of C26:0-LPC and other very long-chain LPCs, comparing them to VLCFA analysis in plasma. The study, which included 330 individuals affected by a peroxisomal β-oxidation deficiency and 407 control individuals, revealed that C26:0- and C24:0-LPC concentrations demonstrated the highest diagnostic accuracy (98.8% and 98.4%, respectively), outperforming VLCFA when C26:0/C22:0 and C24:0/C22:0 ratios were combined (98.1%). Combining C24:0- and C26:0-LPC gave the highest sensitivity (99.7%), with ALD females exhibiting notably higher sensitivity compared with the VLCFA ratio combination (98.7% vs. 93.5%, respectively). In contrast, C22:0-LPC exhibited suboptimal performance, primarily due to its low sensitivity (75%), but we identified a potential use to help distinguish between ALD and Zellweger spectrum disorders. In summary, MS/MS analysis of plasma C24:0- and C26:0-LPC concentrations represents a rapid and straightforward approach to diagnose PDs, demonstrating superior diagnostic accuracy, particularly in ALD females, compared with conventional VLCFA biomarkers. We strongly recommend integrating very-long chain LPC plasma analysis in the diagnostic evaluation of individuals suspected of having a PD.

Keywords: ALD female; Adrenoleukodystrophy; VLCFA; Zellweger spectrum disorders; Zellweger syndrome; fatty acid/transport; lipids; lipids/oxidation; lysophospholipid; tandem mass spectrometry.

Fig. 1

Plasma C24:0- and C26:0-LPC showed excellent diagnostic accuracy for peroxisomal β-oxidation deficiencies. Plasma concentrations of very long-chain LPCs were analyzed in 340 patients and 407 control individuals. Each graph corresponds to one analyte (C26:0-, C24:0-, and C22:0-LPC). Patients are presented both as a collective group and divided by disease category. Plasma LPCs concentrations are plotted in box-and-whisker format, with dotted lines indicating the cutoff values (C26:0-LPC ≤ 0.17 μmol/L, C24:0-LPC ≤ 0.27 μmol/L, and C22:0-LPC ≤ 0.19 μmol/L). Abbreviations are as follows, with the number of patients in each group shown in brackets: CTRL = controls (n = 407), ALL PT = total patients (n = 340), ♂ ALD = male adrenoleukodystrophy patients (n = 155), ♀ALD = female adrenoleukodystrophy patients (n = 77), ZSD = Zellweger spectrum disorder patients (n = 61), UnPD = Patients with a deficient peroxisomal β-oxidation by VLCFA analysis lacking a differential diagnosis (n = 47). Significantly higher concentrations of all three very long-chain LPCs were observed in all disease groups when compared with control individuals (P < 0.0001). The highest C24:0- and C26:0-LPC concentrations were observed in ZSD patients (ranging from 0.28 to 2.17 μmol/L and 0.16–3.68 μmol/L, respectively), whereas ALD males presented the highest C22:0-LPC values (ranging from 0.08 to 1.23 μmol/L). ALD males also exhibited significantly higher concentrations of C22:0-LPC (P < 0.0001), C24:0-LPC (P < 0.0001), and C26:0-LPC (P ≤ 0.001) than ALD females. Plasma C26:0-LPC concentrations were significantly higher in ZSD patients compared with ALD males (P < 0.0001), while an inverse relation was observed for plasma C22:0-LPC (P < 0.0001). No differences were observed for plasma C24:0-LPC between these two groups.

Fig. 2

C24:0- and C26:0-LPC showed higher specificity than VLCFA in plasma samples with elevated triglyceride concentrations. Plasma very long-chain LPCs and VLCFA were analyzed in samples collected from individuals with hypertriglyceridemia. Each data point on the scatter plot represents a plasma sample obtained from a single individual (n = 20; median triglyceride concentration = 12.08 mmol/L; range = 5.65–29.62 mmol/L; cutoff < 1.69 mmol/L). Dotted lines indicate the cutoff values for each parameter. Among these individuals, 14 out 20 (70.0%) presented an abnormal VLCFA profile. This included eight individuals with an increase in C26:0, one with an elevated C26:0/C22:0 ratio and five individuals with elevations in both C26:0 and C26:0/C22:0. In contrast, only one patient (H1) exhibited a slight increase in C24:0- and C26:0-LPC (0.30 and 0.23 μmol/L, respectively), and this patient also showed increases in C26:0 and the C26:0/C22:0 ratio.

Fig. 3

The C26:0-LPC/C22:0-LPC ratio is a valuable parameter to biochemically differentiate ZSD from ALD patients. A: The C26:0-LPC/C22:0-LPC ratios obtained for the different patient groups are shown. Abbreviations are as follows, with the number of patients in each group indicated in brackets: ZSD = Zellweger spectrum disorder patients (n = 61), ALD = total adrenoleukodystrophy patients including males and females (n = 232), ♂ ALD = male adrenoleukodystrophy patients (n = 155), ♀ ALD = female adrenoleukodystrophy patients (n = 77), ns = not statistically significant. Since differences in the C26:0-LPC/C22:0-LPC ratio between ALD males and ALD females were not statistically significant, all ALD individuals were considered as a single group for this analysis. B: ROC curve analysis demonstrated superior performance for the C26:0-LPC/C22:0-LPC ratio (red curve) to differentiate ZSD and ALD patients compared with individual plasma C26:0-LPC (blue curve) or C22:0-LPC (green curve) concentrations. The results of the ROC curve analyses are as follows (CI = confidence interval; AUC = area under the curve): C26:0-LPC/C22:0-LPC ratio: AUC = 0.91 (95% CI 0.86–0.96); C26:0-LPC: AUC = 0.83 (95% CI 0.76–0.90); C22:0-LPC: AUC = 0.72 (95% CI 0.64–0.80). ROC, receiver operating characteristic.