How Reliable Are the Reported Genetic Associations in Disc Degeneration?: The Influence of Phenotypes, Age, Population Size, and Inclusion Sequence in 809 Patients. Issue 21 (1st November 2016)

• Record Type: Journal Article
• Authors: Rajasekaran, S.; Kanna, Rishi Mugesh; Reddy, Ranjani Raja; Natesan, Senthil; Raveendran, Muthuraja; Cheung, Kenneth M.C.; Chan, Danny; Kao, Patrick Y.P.; Yee, Anita; Shetty, Ajoy Prasad
• Is Part Of: Spine. Volume 41:Issue 21(2016)
• Publication Date: 2016-11-01
• Subjects: candidate genes -- disc degeneration -- genetic association -- phenotype variations -- single nucleotide polymorphism; Spine -- Abnormalities -- Periodicals; Spine -- Diseases -- Periodicals; Spine -- Surgery -- Periodicals; 616.73005
• DOI: 10.1097/BRS.0000000000001847
• Languages: English
• ISSNs: 0362-2436

Abstract

Abstract : Study Design: Prospective genetic association study. Objective: The aim of this study was to document the variations in the genetic associations, when different magnetic resonance imaging (MRI) phenotypes, age stratification, cohort size, and sequence of cohort inclusion are varied in the same study population. Summary of Background Data: Genetic associations with disc degeneration have shown high inconsistency, generally attributed to hereditary factors and ethnic variations. However, the effect of different phenotypes, size of the study population, age of the cohort, etc have not been documented clearly. Methods: Seventy-one single-nucleotide polymorphisms (SNPs) of 41 candidate genes were correlated to six MRI markers of disc degeneration (annular tears, Pfirmann grading, Schmorl nodes, Modic changes, Total Endplate Damage score, and disc bulge) in 809 patients with back pain and/or sciatica. In the same study group, the correlations were then retested for different age groups, different sample, size and sequence of subject inclusion (first 404 and the second 405) and the differences documented. Results: The mean age of population (M: 455, F: 354) was 36.7 ± 10.8 years. Different genetic associations were found with different phenotypes: disc bulge with three SNPs of CILP ; annular tears with rs2249350 of ADAMTS5 and rs11247361 IGF1R ; modic changes with VDR and MMP20 ; Pfirmann grading with three SNPs of MMP20 and Schmorl node with SNPs of CALM1 and FN1 and none with Total End Plate Score. Subgroup analysis based on three age groups and dividing the total population into two groups also completely changed the associations for all the six radiographic parameters. Conclusion: In the same study population, SNP associations completely change with different phenotypes. Variations in age, inclusion sequence, and sample size resulted in change of genetic associations. Our study questions the validity of previous studies and necessitates the need for standardizing the description of disc degeneration, phenotype selection, study sample size, age, and other variables in future studies. Level of Evidence: 4