Assessing Risk Factors for Disease Outcomes

Genetic variations identified in patients with rheumatological diseases could facilitate designing clinical interventions and developing therapeutic treatments, according to a recent existing literature review.

Known as Mendelian randomization, this technique uses genetic variants, called alleles, to test whether specific risk factors are associated with or cause certain disease outcomes. This strategy could help researchers better understand existing, unmeasured confounding factors that can correlate to both dependent and independent variables, particularly environmental exposures, impacting observational studies.

The basic Mendelian randomization principle is if an environmental exposure, such a urate levels, is linked to an outcome, such as cardiovascular disease, a genetic variation will be present. Identifying these genetic variants could be helpful with rheumatological diseases.

To better understand Mendelian randomization and how it touches rheumatological conditions, researchers, led by Philp C. Robinson from the University of Queensland School of Medicine in Australia, published a systematic literature review in Nature, analyzing existing results and how they can be applied to rheumatology.

“Mendelian randomization can shed new light on cause-effect relationships in rheumatology, helping to make the case for investment in confirmatory intervention trials and to generate new hypotheses regarding pathological mechanisms and aetiology,” investigators wrote. “The technique also enables experiments to be performed that would otherwise be unethical, logistically difficult or prohibitively expensive.”

Not only can Mendelian randomization studies potentially direct clinical practice, but it can also help prioritize causal pathway interventions and help avoid unnecessary clinical trials. It can be incorporated into randomized clinical trials that test therapies for common disease risk factors, potentially explaining why targeted-intervention benefits often are lower than expected, researchers said.

Understanding the Mendelian Method

Overall, successful observational studies rely on accurate confounding factor measurements. Without that, research is hampered. However, Mendelian randomization can potentially side-step the limitation by examining single-nucleotide polymorphisms. Single-nucleotide polymorphism generally explains less than 1 percent of any observed variance, so Mendelian randomization studies typically require 10,000+ individual data sets.

Any selected genetic variants shouldn’t be associated with confounders, and they shouldn’t have pleiotropic effects. For Mendelian randomization results to be valid, investigators must satisfy three criteria:

1.      Genetic variants used should be associated with – and explain between 0.5 percent to 1.0 percent of – exposure variance to give the study adequate power.

2.      Genetic variants shouldn’t be associated with any factors that confound the exposure-outcome relationship.

3.      Genetic variants should only influence outcome through exposure and shouldn’t have pleiotropic effects.

Mendelian Randomization & Rheumatology

To date, Mendelian randomization has been applied mainly to the cause-effect relationship between urate levels and cardiometabolic traits, including coronary heart disease, high blood pressure, diabetes, obesity, body mass index, and kidney function. These studies, which also test for reverse causality, have found genetic variants associated with both hyperuricemia and gout have causal relationships with urate levels. Other studies suggest a causal role for vitamin D levels in determining responses to rheumatoid arthritis therapies and disease outcomes, as well as a causal role for high body mass index and osteoarthritis risk.

An existing literature review also revealed causal relationship results based on Mendelian randomization.

N-glycosylation of IgG in Rheumatoid Arthritis:  Researchers tested 16 variants associated with IgG-N glycosylation with 14,361 rheumatoid arthritis patients and 43,923 healthy individuals. They didn’t find a causal association of any genetic variants with rheumatoid arthritis, but a lack of data meant the genetic variants weren’t directly associated with the trait. Consequently, via Mendelian randomization, IgG-N glycosylation doesn’t cause rheumatoid arthritis, but it is a disease biomarker.

IL-1 signalling in Rheumatoid Arthritis:  Investigators used Mendelian randomization to unearth cardiovascular risks of long-term inhibition of IL-1 signalling. The study used two genetic variants affecting the gene that encodes IL-1 receptor antagonist (IL-1RA) and included 453,411 participants. Results indicated alleles associated with elevated IL-1RA expression were also linked to protection from rheumatoid arthritis and a cardiovascular disease increase.

Phosphate and Bone Mineral Density: According to Mendelian randomization, high phosphate intake affects calcium metabolism. The study, involving fewer than 200 patients, tested the genetic variant FGF23 associated with phosphate levels to identify any causal relationship with bone mineral density in children. The result was null.

Serum urate levels: Although genetic variation in five genes involved with uric acid excretion account for 3 percent to 4 percent of variance in serum urate levels, Mendelian randomization studies don’t point to a causal role for serum urate levels in hyperuricemia phenotypes, including poor kidney function, ischemic heart disease, diabetes, high blood pressure, high cholesterol, high bone mineral density, and high body mass index.

However, researchers said, reverse causality studies identify a causal effect of body fat measurements on urate levels or triglyceride levels. Among 26 serum-urate modifying genetic variants, only four revealed weak associations with blood pressure.

Additionally, a study into five urate-associated variants identified a positive serum urate level-kidney function relationship.

Rheumatoid Arthritis and Cardiovascular Disease: Rheumatoid arthritis is associated with a higher cardiovascular disease risk, and LDL cholesterol-associated genetic variants could help determine which drives the other. Mendelian randomization could also inform the paradoxical relationship between low body mass index and high mortality with rheumatoid arthritis.  

Gout and Other Metabolic Diseases: Any causal relationships between gout and comorbidities are unknown. According to researchers, Mendelian randomization studies could be designed to focus on the disease’s inflammatory aspects to reveal existing links.

IL-6 Pathways and Rheumatoid Arthritis: Rheumatoid arthritis is characterized by high interleukin-6 levels with interleukin-6 inhibitors treating established cases. Mendelian randomization cardiovascular studies suggest interleukin-6 receptor inhibition is associated with cardiovascular-disease prevalence. Mendelian randomization can also address whether interleukin-6 or interleukin-6 receptor pathways contribute to cardiovascular disease development in rheumatoid arthritis patients.

Urate Levels and Neurological Disease: Research shows urate levels protect against neurodegenerative condition development, including Alzheimer’s disease, Parkinson’s disease, and multiple sclerosis. Using Mendelian randomization to determine whether this relationship is causal could play into how aggressively providers approach reducing urate in gout.

Alcohol Intake and Rheumatoid Arthritis: Large, existing epidemiological studies suggest moderate alcohol intake reduces rheumatoid arthritis risk. Since alcohol dependence carries a substantial genetic component – approximately 30 percent heritability based on analyzed single-nucleotide polymorphism – Mendelian randomization could further investigate if alcohol intake influences rheumatoid arthritis susceptibility.

Urate Levels and Cancer: The potential urate level and cancer risk link has been well reported and is assumed to be the result of urate’s oxidative intracellular action. However, some studies find those resuls inconclusive. It’s possible the serum urate level effect on cancer risk could be location and malignancy-type dependent.

Overall, researchers said, the urate, gout, and cancer association is complex. Existing studies found higher cancer rates in patients with gout versus those without the condition. Others found no association.

Limitations

Mendelian randomization isn’t a silver bullet for solving all observational-study problems, though. It’s difficult to verify the technique’s underlying assumptions aren’t being violated. And, using genetic risk scores based on multiple variants increases analysis power, but using individual variants provides for easier interpretations of biological function.

Additionally, genetic variants are mostly pulled from genome-wide association studies conducted on middle-aged and older adults. Consequently, any detected effects could arise from years of cumulative environmental exposures. As a result, investigators recommended returning to Mendelian randomization’s origins where any exposure effect on outcomes would be tested with linear or logistic regression only.

“This proposal also contains the suggestion,” researchers wrote, “that Mendelian randomization is particularly valuable in establishing null results in a large, well-powered sample sets with an established association between genetic variant and exposure but no association between genetic variant and outcomes, supporting the conclusion that little or no effect of exposure on outcome exists.”

References:

Robinson PC, Choi HK, Do R, Merriman TR.

"Insight into rheumatological cause and effect through the use of Mendelian randomization,"

Nature Reviews Rheumatology. July 14, 2016. DOI: 10.1038/nrrheum.2016.102.