Severity modeling of propionic acidemia using clinical and laboratory biomarkers
Oleg A Shchelochkov, Irini Manoli, Paul Juneau, Jennifer L Sloan, Susan Ferry, Jennifer Myles, Megan Schoenfeld, Alexandra Pass, Samantha McCoy, Carol Van Ryzin, Olivia Wenger, Mark Levin, Wadih Zein, Laryssa Huryn, Joseph Snow, Colby Chlebowski, Audrey Thurm, Jeffrey B Kopp, Kong Y Chen, Charles P Venditti, Oleg A Shchelochkov, Irini Manoli, Paul Juneau, Jennifer L Sloan, Susan Ferry, Jennifer Myles, Megan Schoenfeld, Alexandra Pass, Samantha McCoy, Carol Van Ryzin, Olivia Wenger, Mark Levin, Wadih Zein, Laryssa Huryn, Joseph Snow, Colby Chlebowski, Audrey Thurm, Jeffrey B Kopp, Kong Y Chen, Charles P Venditti
Abstract
Purpose: To conduct a proof-of-principle study to identify subtypes of propionic acidemia (PA) and associated biomarkers.
Methods: Data from a clinically diverse PA patient population ( https://ichgcp.net/clinical-trials-registry/NCT02890342 ) were used to train and test machine learning models, identify PA-relevant biomarkers, and perform validation analysis using data from liver-transplanted participants. k-Means clustering was used to test for the existence of PA subtypes. Expert knowledge was used to define PA subtypes (mild and severe). Given expert classification, supervised machine learning (support vector machine with a polynomial kernel, svmPoly) performed dimensional reduction to define relevant features of each PA subtype.
Results: Forty participants enrolled in the study; five underwent liver transplant. Analysis with k-means clustering indicated that several PA subtypes may exist on the biochemical continuum. The conventional PA biomarkers, plasma total 2-methylctirate and propionylcarnitine, were not statistically significantly different between nontransplanted and transplanted participants motivating us to search for other biomarkers. Unbiased dimensional reduction using svmPoly revealed that plasma transthyretin, alanine:serine ratio, GDF15, FGF21, and in vivo 1-13C-propionate oxidation, play roles in defining PA subtypes.
Conclusion: Support vector machine prioritized biomarkers that helped classify propionic acidemia patients according to severity subtypes, with important ramifications for future clinical trials and management of PA.
Conflict of interest statement
The authors declare no competing interests.
© 2021. This is a U.S. government work and not under copyright protection in the U.S.; foreign copyright protection may apply.
Figures
References
- Shchelochkov OA, Carrillo N, Venditti C. Propionic Acidemia. University of Washington, Seattle; 2012.
- Kölker S, et al. The phenotypic spectrum of organic acidurias and urea cycle disorders. Part 1: the initial presentation. J. Inherit. Metab. Dis. 2015;38:1041–1057. doi: 10.1007/s10545-015-9839-3.
- Kölker S, et al. The phenotypic spectrum of organic acidurias and urea cycle disorders. Part 2: the evolving clinical phenotype. J. Inherit. Metab. Dis. 2015;38:1059–1074. doi: 10.1007/s10545-015-9840-x.
- Cappuccio G, et al. Expansion of the phenotypic spectrum of propionic acidemia with isolated elevated propionylcarnitine. JIMD Rep. 2017;35:33–37. doi: 10.1007/8904_2016_21.
- Kidd JR, Wolf B, Hsia E, Kidd KK. Genetics of propionic acidemia in a Mennonite Amish kindred. Am. J. Hum. Genet. 1980;32:236–245.
- Gravel RA, et al. Mutations participating in interallelic complementation in propionic acidemia. Am. J. Hum. Genet. 1994;55:51–58.
- Matern D, Tortorelli S, Oglesbee D, Gavrilov D, Rinaldo P. Reduction of the false-positive rate in newborn screening by implementation of MS/MS-based second-tier tests: the Mayo Clinic experience (2004-2007) J. Inherit. Metab. Dis. 2007;30:585–592. doi: 10.1007/s10545-007-0691-y.
- Lindner M, et al. Newborn screening for methylmalonic acidurias-optimization by statistical parameter combination. J. Inherit. Metab. Dis. 2008;31:379–385. doi: 10.1007/s10545-008-0892-z.
- Pillai NR, et al. Liver transplantation in propionic and methylmalonic acidemia: a single center study with literature review. Mol. Genet. Metab. 2019;128:431–443. doi: 10.1016/j.ymgme.2019.11.001.
- Vara R, et al. Liver transplantation for propionic acidemia in children. Liver Transpl. 2011;17:661–667. doi: 10.1002/lt.22279.
- Kasahara M, et al. Living-donor liver transplantation for propionic acidemia. Pediatr. Transpl. 2012;16:230–234. doi: 10.1111/j.1399-3046.2011.01607.x.
- Chandler RJ, et al. Adeno-associated virus serotype 8 gene transfer rescues a neonatal lethal murine model of propionic acidemia. Hum. Gene Ther. 2011;22:477–481. doi: 10.1089/hum.2010.164.
- Ghosh S, Dasgupta A, Swetapadma A. A Study on Support Vector Machine based Linear and Non-Linear Pattern Classification. 2019 International Conference on Intelligent Sustainable Systems (ICISS) 2019;21–22:24–28. doi: 10.1109/ISS1.2019.8908018.
- Donoho D, Stodden V. Breakdown Point of Model Selection When the Number of Variables Exceeds the Number of Observations. The 2006 IEEE International Joint Conference on Neural Network Proceedings. 2006;16–21:1916–1921.
- Hastie T, Tibshirani R. Efficient quadratic regularization for expression arrays. Biostatistics. 5, 329–40. 10.1093/biostatistics/5.3.329 (2004).
- Baumgartner, et al. Supervised machine learning techniques for the classification of metabolic disorders in newborns. Bioinformatics20, 2985–2996 (2004).
- Baumgartner, C., Böhm, C. & Baumgartner, D. Modelling of classification rules on metabolic patterns including machine learning and expert knowledge. J. Biomed. Informatics. 38, 89–98 (2005).
- Chen W-H, et al. Web-based newborn screening system for metabolic diseases: machine learning versus clinicians. J. Med. Internet Res. 2013;15:e98–e98. doi: 10.2196/jmir.2495.
- Hauser NS, Manoli I, Graf JC, Sloan J, Venditti P. Variable dietary management of methylmalonic acidemia: metabolic and energetic correlations. Am. J. Clin. Nutr. 2011;93:47–56. doi: 10.3945/ajcn.110.004341.
- Barshop BA, et al. Metabolism of 1-13C-propionate in vivo in patients with disorders of propionate metabolism. Pediatr. Res. 1991;30:15–22. doi: 10.1203/00006450-199107000-00004.
- R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2013.
- Kowarik A, Templ M. Imputation with the R Package VIM. J. Stat. Softw. 2016;74:16. doi: 10.18637/jss.v074.i07.
- Max Kuhn, Contributions from Jed Wing, Steve Weston A. W., et al. caret: Classification and Regression Training. R package version caret: Classification and Regression Training. R package version 6.0-84. (2016).
- Thompson GN, et al. In vivo propionate oxidation as a prognostic indicator in disorders of propionate metabolism. Eur. J. Pediatr. 1990;149:408–411. doi: 10.1007/BF02009661.
- Walter JH, et al. Chronic renal failure in methylmalonic acidaemia. Eur. J. Pediatr. 1989;148:344–348. doi: 10.1007/BF00444131.
- Beretta L, Santaniello A. Nearest neighbor imputation algorithms: a critical evaluation. BMC Med. Inform. Decis. Mak. 2016;16:74. doi: 10.1186/s12911-016-0318-z.
- Manoli, I., Pass, A.R., Harrington, E.A. et al. 1-13C-propionate breath testing as a surrogate endpoint to assess efficacy of liver-directed therapies in methylmalonic acidemia (MMA). Genet Med. 10.1038/s41436-021-01143-8 (2021).
- Shchelochkov OA, et al. Chronic kidney disease in propionic acidemia. Genet. Med. 2019;21:2830–2835. doi: 10.1038/s41436-019-0593-z.
- Fujita Y, et al. GDF15 is a novel biomarker to evaluate efficacy of pyruvate therapy for mitochondrial diseases. Mitochondrion. 2015;20:34–42. doi: 10.1016/j.mito.2014.10.006.
- Ferreira R, et al. Deoxysphingolipid precursors indicate abnormal sphingolipid metabolism in individuals with primary and secondary disturbances of serine availability. Mol. Genet. Metab. 2018;124:204–209. doi: 10.1016/j.ymgme.2018.05.001.
- Kakkis ED, et al. Recommendations for the development of rare disease drugs using the accelerated approval pathway and for qualifying biomarkers as primary endpoints. Orphanet J. Rare Dis. 2015;10:16. doi: 10.1186/s13023-014-0195-4.
- Yannicelli S. Nutrition therapy of organic acidaemias with amino acid-based formulas: emphasis on methylmalonic and propionic acidaemia. J. Inherit. Metab. Dis. 2006;29:281–287. doi: 10.1007/s10545-006-0267-2.
- Kirmse, B. et al. Plasma fibroblast growth factor-21 levels in patients with inborn errors of metabolism. Mol. Genet. Metab. Rep. 13, 52–54 (2017).
- Manoli I, et al. FGF21 underlies a hormetic response to metabolic stress in methylmalonic acidemia. JCI Insight. 2018;3:e124351. doi: 10.1172/jci.insight.124351.
- Molema F, et al. Fibroblast growth factor 21 as a biomarker for long-term complications in organic acidemias. J. Inherit. Metab. Dis. 2018;41:1179–1187. doi: 10.1007/s10545-018-0244-6.
- An D, et al. Systemic messenger RNA therapy as a treatment for methylmalonic acidemia. Cell Rep. 2017;21:3548–3558. doi: 10.1016/j.celrep.2017.11.081.
- Jiang L, et al. Dual mRNA therapy restores metabolic function in long-term studies in mice with propionic acidemia. Nat Commun. 2020;21:11.
Source: PubMed