Magnesium supplementation in the treatment of pseudoxanthoma elasticum: A randomized trial
To the Editor: Pseudoxanthoma elasticum (PXE) is a rare, autosomal recessive connective tissue disorder with significant systemic morbidity due to progressive elastic fiber mineralization; treatment is lacking.1-5 Research in humans and mouse models of PXE has linked increased magnesium levels to decreased calcification. Specifically, in an animal PXE model, magnesium prevented elastic tissue calcification, and in a prior study by LaRusso et al, there was a reduction in elastic fiber calcification in the magnesium-containing placebo arm.4,5
We conducted a randomized, double-blind, placebo-controlled prospective trial evaluating the effect of oral magnesium oxide (MgO) versus placebo on the skin and eyes in 44 PXE patients (approved by the Institutional Review Board at the Icahn School of Medicine at Mount Sinai; Clinicaltrials.gov listing: NCT01525875). In the first year (the double-blind, placebo-controlled phase), patients were randomized 1:1 to receive twice-daily 800 mg MgO (500 mg elemental magnesium) (treatment group) or placebo (control group). In the second year (the open-label phase), all patients received 2500 mg MgO (1500 mg elemental magnesium) divided over 2 doses daily.
A baseline target lesion of representative affected skin (usually neck, axilla, or antecubital fossa) was evaluated using a 10-point numeric grading score (Table I) and 4-mm punch biopsy. Patients had to have a clinical disease severity grade of $1 ( poorly defined, barely visible macules) at screening. These measures plus laboratory tests, electrocardiogram, bone mineral density scan, and ophthalmic examination were performed throughout the study.
The primary endpoint was calcification of skin elastic fibers. Biopsies were stained with Verrhoeff-van-Gieson and Von-Kossa stains. DensiteQuant image analyzing software quantified the area of calcification at the top and bottom of the sample. Changes in calcification were assessed with a mixed-effect model having location as a covariate. During the double-blind, placebo-controlled phase, the magnesium group had a nonsignificant decrease in calcification, whereas the placebo group was unchanged. However, during the open-label phase, both groups had increases in calcification (Fig 1, A).
Using a 30% decrease in elastic fiber calcification as a secondary endpoint, 36.36% of patients in the treatment group were responders compared with only 13.6% of patients in the placebo group (drop-outs are non-responders) (P ¼ .17). Considering only the patients completing the double-blind, placebo-controlled phase (N ¼ 40), the response rate in the treatment group was 38.09% versus 15.79% in the placebo group (P ¼ .29) (Fig 1, C ). Although not statistically significant, this trend might be clinically significant. After 2 years of magnesium treatment, 2 of the 8 responders in year 1 kept their response level, and 5 patients who were previously non-responders during the first year became responders in the second year.
|Score||Pseudoxanthoma elasticum severity|
|0||No evidence of pseudoxanthoma elasticum|
|1.0||Poorly defined, barely visible macules|
|2.0||Well defined, easily identified macules|
|3.0||Mostly macules with <5 papules|
|5.0||Patches consisting of confluent macules with <50% of target area covered by papules|
|6.0||Patches consisting of confluent macules with >=50% of target area covered by papules|
|8.0||Plaques with mild folds of skin|
|9.0||Plaques with redundant folds of skin|
Likewise, the treatment group had twice the reduction in target lesion score than the placebo group in the double-blind, placebo-controlled phase, albeit not statistically significant (Fig 1, B). In the magnesium group, the percentage of patients with a 1-point score reduction increased from 9.09% to 40.91% (P ¼ .0692.) across the 2 years. The scores of the patients in the placebo group significantly decreased when treated with magnesium (change e0.336, P\.0001) but not placebo (change e0.046). A sharp increase in clinical improvement was observed when these patients started the open-label phase (4.54%-31.82%, P ¼ .125). No significant treatment effect was observed in the ophthalmologic outcomes or bone mineral density Tscores. There was no clinically significant laboratory changes (including magnesium and calcium level alterations) during the study.
This study had limitations that likely contributed to the lack of statistical significance. It was a small study of a rare disease with tremendous patient variation. Individual patient changes could thus profoundly affect results. Likewise, the magnesium dose was limited because of concerns for side effects. However, magnesium was well tolerated. Also, there were not any ophthalmologic exclusion criteria. Many patients continued treatment with their ophthalmologists, causing their annual study visit to be merely a snapshot in their disease course.
Despite these limitations, we highlight a promising trend in the results showing calcification reduction of skin elastic fibers while on magnesium supplementation. Larger studies with higher magnesium dosage could be valuable to further elicit this beneficial trend.
Sharon Rose, MD,a Shelbi Jim On, MD,a Wayne Fuchs, MD,b Chen Chen, MD,a Robert Phelps, MD,a Davida Kornreich, MD,c Madelaine Haddican, MD,a Giselle Singer, BS,a Vicky Wong, BA,a Danielle Baum, RN,a Anjali Vekaria, MD,a Matthew Gagliotti, BS,a Ruiqi Huang, PhD,a Mayte Suárez-Fariñas, PhD,a Sharon F. Terry, MA,d and Mark Lebwohl, MDa
From the Department of Dermatology, Icahn School of Medicine at Mount Sinai, New York, New Yorka; Department of Ophthalmology, Mount Sinai Hospital, New York, New Yorkb; Department of Dermatology and Cutaneous Biology, Thomas Jefferson University Hospitals, Philadelphia, Pennsylvaniac; and PXE International, Washington, DCd
Conflicts of interest: None disclosed.
Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Reprint requests: Sharon Rose, MD, 1 Gustave L. Levy Place, Box 1048, New York, NY 10029
1. Ringpfeil F, Lebwohl MG, Christiano AM, Uitto J. Pseudoxanthoma elasticum: mutations in the MRP6 gene encoding a transmembrane ATP-binding cassette (ABC) transporter. Proc Natl Acad Sci U S A. 2000;97(11):6001-6006.