An Independent Evaluation of the Validity of a DNA-Based Prognostic Test for Adolescent Idiopathic Scoliosis

Abstract

Background: ScoliScore is a DNA-based prognostic test, designed and used to help to predict the risk of curve progression in patients with adolescent idiopathic scoliosis. The role of this test in clinical practice remains unclear as the published results of the ScoliScore have not been validated independently. The purpose of this study was to determine if the ScoliScore effectively predicted the risk of curve progression in patients with mild and moderate adolescent idiopathic scoliosis in two urban academic medical centers.

Methods: One hundred and twenty-six patients with adolescent idiopathic scoliosis who met inclusion criteria at two centers were administered the ScoliScore test. Two groups were created: a progression group (those who had a Cobb angle of >40° or those who had undergone surgical fusion) and a non-progression group (those who had skeletal maturity without curve progression to 40°). ScoliScore values and risk levels were compared between the two groups. The negative predictive value was calculated for low-risk scores and the positive predictive value was calculated for high-risk scores.

Results: There was no significant difference (p = 0.706) in the mean ScoliScore (and standard deviation) between patients with curve progression (107 ± 55 points) and those without curve progression (102 ± 62 points). There was also no significant difference (p = 0.399) in curve progression between patients with high-risk scores (26.7%) and those with low-risk scores (12.9%). The positive predictive value of the test was 0.27 (95% confidence interval, 0.09 to 0.55), and the negative predictive value was 0.87 (95% confidence interval, 0.69 to 0.96). ScoliScores and rates of progression were not affected by brace-wear.

Conclusions: ScoliScores did not differ between patients with and without curve progression, and the negative and positive predictive values were lower in our study than in the previously published validation study by the developers of the test. This may be due to differences in our test population, genetic variability, or failure of patients in the non-progression group to follow up.