A Genetic Risk Factor for Periodic Limb Movements in Sleep

Related Article: Original Article, N Engl J Med 2007 :357 ;639 -647 .

To the Editor: Stefansson et al. (Aug. 16 issue)1 report the results of a genomewide association study of the restless legs syndrome (RLS) in an Icelandic series. The analysis revealed that one single-nucleotide polymorphism (SNP) in the BTB (POZ) domain-containing 9 (BTBD9 ) gene was associated with RLS. Replication in smaller samples from Iceland and the United States showed that the discovery originated from a subgroup of patients with RLS who presented with periodic limb movements in sleep, a common feature of RLS. Winkelmann et al.2 performed a similar analysis for patients with RLS who were of European descent, with replication in samples of European or French-Canadian origin, and highlighted SNP associations in four genes in three chromosomal loci: MEIS1 , BTBD9 , and MAP2K5 /LBXCOR1 .

We have genotyped 11 SNPs spanning all three reported loci for disease susceptibility in a U.S. series of 244 patients with RLS (121 with sporadic cases and 123 familial probands) and 497 controls. Patients with sporadic cases have no known family history of RLS, whereas all probands have affected first-degree relatives (97%) or affected second-degree relatives, with an average of 3.7 affected persons per family (range, 2 to 7). RLS was clinically characterized by means of standardized methods with a series of questionnaires3 ; the initial presentation, RLS symptoms, and duration of and response to treatment were similar in the sporadic and familial groups. Genotyping was performed by means of the TaqMan or Sequenom assay or both assays; all markers were in Hardy-Weinberg equilibrium. All minor allele frequencies for patients with RLS and control subjects are similar to those reported previously.1 ,2 Statistical analysis confirmed the association with BTBD9 and MEIS1 genes but showed only a trend for variants in MAP2K5 /LBXCOR1 and RLS. The odds ratio for the associated risk alleles ranged from 1.8 to 2.9 for BTBD9 and from 2.6 to 3.7 for MEIS1 (Table 1 ). MEIS1 haplotype analysis confirmed a more significant association for the AG haplotype (odds ratio, 4.6; 95% confidence interval, 2.9 to 7.2; P<10^-7 ), as previously reported.2

Given the magnitude of the effect, it is curious that MEIS1 was not detected in the Icelandic series1 ; whether this is because of the use of different genotyping platforms, detailed phenotyping of RLS versus a diagnosis based on a constellation of findings and symptoms, or population specificity should be addressed. Our study confirms that variants in BTBD9 and MEIS1 loci are involved in susceptibility to RLS susceptiband suggests that clinical differences between genetically defined groups may now be defined. The specific mutations and biological mechanism (or mechanisms) underlying familial RLS have yet to be elucidated, along with the environmental and stochastic factors that presumably play a greater role in sporadic RLS.

To the Editor: In their article, Stefansson et al. say that we question the authenticity of RLS. This is a mischaracterization.

We described how GlaxoSmithKline generated -- and the media, like Stefansson et al., uncritically repeated -- exaggerated prevalence estimates.1 The typical high-end U.S. prevalence claim ("10% prevalence") comes from a one-state survey using one unvalidated question.2 The editorialist's claim that 3% of adults have frequent, bothersome symptoms is also suspect.3 The telephone survey he cites used four validated questions but reported an implausible 98% response rate; selection bias undoubtedly has resulted in an exaggerated estimate of 3%.1

RLS -- like many disorders -- exists along a spectrum: from debilitating to ordinary experience. The question is where to draw the line identifying those who are "sick." This matters: diagnosis leads to treatments with important side effects (nausea with ropinirole as compared with placebo, 40% vs. 8%4 ) and unknown long-term consequences.

We are concerned that the $100 million spent in 2007 on the marketing of RLS drugs5 is distorting what is known about restless legs.

References

1. Stefansson H, Rye DB, Hicks A, A genetic risk factor for periodic limb movements in sleep. N Engl J Med 2007;357:639-647

2. Winkelmann J, Schormair B, Lichtner P, Genome-wide association study of restless legs syndrome identifies common variants in three genomic regions. Nat Genet 2007;39:1000-1006

3. Allen RP, Kushida CA, Atkinson MJ Factor analysis of the International Restless Legs Syndrome Study Group's scale for restless legs severity. Sleep Med 2003;4:133-135

Illustration

Table 1:

SNPs, Genotype Frequencies, and Odd Ratios for Each RLS Susceptibility Locus.

[Image Omitted: See PDF]

References

1. Woloshin S, Schwartz L Giving legs to restless legs:a case study of how the media helps make people sick. PLoS Med 2006;3:e170-e170

2. Phillips B, Young T, Finn L, Asher K, Hening WA, Purvis C Epidemiology of restless legs symptoms in adults. Arch Intern Med 2000;160:2137-2141

3. Winkelman JW Periodic limb movements in sleep--endophenotype for restless legs syndrome? N Engl J Med 2007;357:703-705

4. Requip (ropinirole hydrochloride) prescribing information. GlaxoSmithKline, 2005. (Available at http://www.fda.gov/cder/foi/label/2005/020658s013lbl.pdf.)

5. Nielsen Monitor-Plus. New York:The Nielsen Company, 2007. (Available at http://www.NielsenMedia.com.)

Carles Vilariño-Güell Ph.D.

Matthew J. Farrer Ph.D.

Siong-Chi Lin M.D.

Steven Woloshin M.D.

Lisa M. Schwartz M.D.

Mayo Clinic, Jacksonville, FL 32224 [email protected]

Veterans Affairs Outcomes Group, White River Junction, VT 05009 [email protected]