AS402 - Prospective study of biomarkers and risk factors for ALS incidence and progression

Investigator Names and Contact Information

Alberto Ascherio (aascheri@hsph.harvard.edu)

Introduction/Intent

Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease characterized by a loss of the neurons that control voluntary movements. None of over 30 new drugs tested in ALS has been found to be effective, and median survival from diagnosis remains at 2-3 years. Because ALS is a relatively rare disease with rapid fatality, epidemiological study has been limited. Our aim is to amass a sufficient number of participants with ALS from prospective studies including WHI who gave blood prior to disease onset and examine several biomarkers for epidemiological clues to the disease etiology. In previous work based on questionnaire data we found that ALS risk was inversely associated with long term intakes of vitamin E and other antioxidants, and positively associated with cigarette smoking. In parallel, preliminary evidence emerged that blood levels of urate, a potent antioxidant, are inversely related to the rate of ALS progression – this is encouraging in view of urate's strong antioxidant properties, its robust inverse associations with both risk and progression of Parkinson disease, and the fact that urate concentrations can be increased by administration of inosine (a urate precursor) or other available drugs. Aims: This proposal will address whether plasma levels of urate in apparently healthy individuals predicts ALS risk. Additionally, we will conduct a discovery study for novel risk markers using a metabolomic approach. Methods: WHI OS and CT combined would add 121 participants with blood samples who died from ALS to 161 cases contributed by the Nurses' Health Study, Health Professional Health Study and Cancer Prevention Study II Nutrition Cohort. For each case, two controls will be matched by exact year of birth. Controls will be randomly selected among cohort participants at risk of developing ALS at the approximate time of diagnosis of the matched case. Comparison of geometric means between cases and controls will be conducted using generalized linear models that take into account the matched design of the study. Conditional logistic regression will be used to estimate effects across quartiles. Specimen requested: 150 microL of baseline plasma is required to measure urate and potential confounders of a urate-ALS association (e.g., LDL, HDL cholesterol) and 200 microL of baseline plasma is required to measure the metabolomics profile. ​

The proposed aims of our study stand to identify or substantiate major new links to ALS, with realistic prospects for mechanistic insight and therapeutic impact. The key strength of our proposal is the availability for the first time of blood samples collected before the onset of ALS.

Specific Aims

Aim #1. Plasma urate and ALS risk. ALS risk decreases monotonically with increasing plasma urate concentration, and this association is independent from body mass index (BMI), plasma lipids (LDL, HDL, total cholesterol, and triglycerides), or use of cholesterol lowering drugs. In contrast, neither BMI nor cholesterol concentrations are independently related to ALS risk when urate levels are considered.

Aim #2. Metabolic markers of ALS risk 2.1 We hypothesize that, in addition to low urate levels, there are metabolic alterations in plasma of individuals with ALS that precede the onset of neurological symptoms. Specifically, we hypothesize that these pre-diagnostic markers include increased levels of branched chain amino acids (BCAA), glutamate, 2-hydroxybutyrate, creatine, pyruvate, carnitine, stearoyl sphingomyelin, and pantothenate, and reduced levels of creatinine and of tryptophan and its breakdown products. 2.2 We will use a metabolomics approach to identify a signature of ALS risk in blood collected before the onset of neurological symptoms.​

Additional Specific Aims for Modification #1:

1. We will test the hypothesis that ALS risk decreases with plasma n-3 PUFA concentration and this association is independent from body mass index (BMI), plasma lipids (LDL, HDL, total cholesterol and triglycerides) or use of cholesterol lowering drug. We also hypothesize that among the individual fatty acids, long chain n-3 PUFAs, especially DHA, are the most strongly associated with decreases in ALS risk in multivariate analyses.

2. We will explore whether the activity of desaturase enzymes, estimated from the product-to-precursor ratios, are associated with ALS risk. Ratios considered will include 16:1n-7/16:0 and 18:1n-9/18:0 (stearoyl coenzyme A desaturase), 18:3n-6/18:2n-6 (delta-6 desaturase) and 20:4n-6/20:3n-6 (delta-5 desaturase)