ABSTRACT Background Limb girdle muscular dystrophy type R9 (LGMDR9) results from biallelic variants in FKRP . There is limited data to predict loss of ambulation (LOA) among those with LGMDR9. Methods Participants in an ongoing dystroglycanopathy natural history study (NCT00313677) with FKRP variants who had achieved ambulation and were more than 3 years old were included ( n = 97). LOA was defined as self‐reported full‐time wheelchair use, weakness preventing completion of the 10‐m walk‐run test (10MWT) or 10MWT time > 30 s. Interval‐censored time‐to‐event analysis was used to determine median age at LOA. Receiver operating characteristic curves were used to examine the ability of 10MWT and 4‐stair climb (4SC) times to predict LOA. Results Of 97 participants, 55 (57%) were homozygous for the c.826C>A founder variant. Thirty‐one participants lost ambulation; 15 (49%) were homozygous for c.826C>A. Earliest age at LOA was 9 years (non‐homozygous for c.826C>A). Median age at LOA for the cohort was 46.0 years. Performances on 10MWT and 4SC were highly predictive of LOA within 3 years, with areas under the ROC curve of 0.89 (10MWT) and 0.87 (4SC) when genotype was included in analysis. Optimal cutoffs for predicting LOA within 3 years differed by genotype and had acceptable sensitivity and specificity. Discussion LOA among those with LGMDR9 is strongly predicted by performance on 10MWT and 4SC. These results demonstrate the real‐world significance of standardized motor function tests used in LGMDR9 clinical trials and aid in anticipatory guidance.
Miller et al. (Sun,) studied this question.