Musculoskeletal Features in Adults With X-linked Hypophosphatemia: An Analysis of Clinical Trial and Survey Data

Abstract

Context

Patients with X-linked hypophosphatemia (XLH) experience multiple musculoskeletal manifestations throughout adulthood.

Objective

To describe the burden of musculoskeletal features and associated surgeries across the lifespan of adults with XLH.

Methods

Three groups of adults were analyzed: subjects of a clinical trial, participants in an online survey, and a subgroup of the online survey participants considered comparable to the clinical trial subjects (according to Brief Pain Inventory worst pain scores of ≥ 4). In each group, the adults were categorized by age: 18-29, 30-39, 40-49, 50-59, and ≥ 60 years. Rates of 5 prespecified musculoskeletal features and associated surgeries were investigated across these age bands for the 3 groups.

Results

Data from 336 adults were analyzed. In all 3 groups, 43% to 47% had a history of fracture, with the proportions increasing with age. The overall prevalence of osteoarthritis was > 50% in all 3 groups, with a rate of 23% to 37% in the 18- to 29-year-old group, and increasing with age. Similar patterns were observed for osteophytes and enthesopathy. Hip and knee arthroplasty was reported even in adults in their 30s. Spinal stenosis was present at a low prevalence, increasing with age. The proportion of adults with ≥ 2 musculoskeletal features was 59.1%, 55.0%, and 61.3% in the clinical trial group, survey group, and survey pain subgroup, respectively.

Conclusion

This analysis confirmed high rates of multiple musculoskeletal features beginning as early as age 20 years among adults with XLH and gradually accumulating with age.

X-linked hypophosphatemia (XLH) is a rare, genetic, lifelong, phosphate-wasting disease caused by loss-of-function mutations in the phosphate-regulating endopeptidase homolog X-linked (PHEX) gene (1), resulting in elevation of circulating fibroblast growth factor 23 (FGF23) levels (2). Excess FGF23 reduces renal phosphate reabsorption and decreases production of active vitamin D (1,25[OH]₂D) (3), manifesting as chronic hypophosphatemia, impairing the mineralization of bones and teeth and reducing muscle strength (4).

The clinical presentation of XLH usually begins in early childhood and is characterized by rickets, skeletal deformities, short stature, and craniosynostosis (2, 5). Skeletal deformities and short stature in childhood typically remain after completion of growth, and even after surgical intervention. Moreover, hypophosphatemia persists in adulthood. Musculoskeletal features that dominate the clinical picture in adulthood include pseudofractures, fractures, enthesopathies, osteoarthritis (OA) with osteophytes, and spinal stenosis. These disease manifestations result in pain, stiffness, and limitations of physical function and mobility (2, 6, 7). Symptomatic hip and knee joint disease is also common, with joint replacement being reported in adults in their 40s (8).

A recent multinational online survey provided some estimates of skeletal disease burden in adults with XLH and pointed out the age-dependent increase in frequency (7). A more complete understanding of the evolution of these musculoskeletal features in adults with XLH will guide appropriate and timely management of this debilitating disease.

We hypothesize that prespecified musculoskeletal features associated with aging, as reported in the medical literature for the general population, begin earlier in adults with XLH and accumulate progressively with age.

Methods

We performed a series of analyses of data from 2 sources: baseline data of adults with XLH who participated in a clinical trial, and data collected from adult participants of a cross-sectional, multinational online survey.

Details of the clinical trial have been previously published (9). In summary, an international, randomized, double-blind, placebo-controlled phase 3 clinical trial assessed the efficacy and safety of burosumab in adults with XLH (NCT02526160) (9). Adults aged 18 to 65 years with a diagnosis of XLH supported by clinical and biochemical features consistent with the disease, and/or a confirmed PHEX mutation in the adult or a family member, were included. To be eligible, adults also needed to have skeletal pain, defined as a Brief Pain Inventory (BPI) worst pain score ≥ 4 in the last 24 hours prior to screening, attributed to XLH/osteomalacia (9). In this manuscript, the first of the 3 groups is referred to as the clinical trial group.

The multinational online survey, using a web-based questionnaire, aimed at adults and children with XLH was conducted from June 2014 to February 2016 (7), yielding participant-reported medical and surgical history responses. Adults were recruited via XLH patient advocacy networks and clinicians and are referred to as the survey group. Adults reporting use of burosumab were excluded.

Given that adults in the clinical trial group were selected based on the specific criterion of BPI worst pain score ≥4, while the adults in the survey group were self-selected, a subgroup of the survey participants was created for the purposes of benchmarking against the clinical trial group. This comparator group comprised those with a BPI worst pain score ≥4. The supposition was that with the similarity of worst pain score, this survey pain subgroup is likely to report comparable levels of musculoskeletal features to the clinical trial group. In this manuscript, this third group is referred to as the survey pain subgroup.

Data Sources

From both data sets, age, sex, age at diagnosis, height, weight, body mass index (BMI), and current XLH treatment were captured, as well as participant-reported prespecified musculoskeletal features and surgeries. Survey data were collected directly from participants and were not verified against medical records. As a conservative measure for all questions with a potential list of responses, if no option was ticked, it was assumed that the participant did not have the indicated feature.

Inclusion and exclusion criteria

To avoid overlap in the data from the clinical trial and survey, adults who self-reported as previously participating in a clinical trial of burosumab were excluded. Thus, no adults in this analysis had prior burosumab exposure. Thirty adults were excluded from the master data sets because of previous burosumab exposure: 7/134 from the clinical trial group and 23/232 from the survey group (Fig. 1).

Figure 1.

Disposition of subjects.

Outcomes

The prespecified participant-reported musculoskeletal features of interest were a history of fractures, OA, osteophytes, enthesopathies, and spinal stenosis, plus hip and knee arthroplasty and spinal surgeries (Table 1). Data on dental abscess were also collected. These features were described as having ever been present or as absent, but the age of diagnosis was not recorded.

Table 1.

Data sources for musculoskeletal features of XLH

Characteristic	Survey	Clinical trial
Fractures	Have you ever fractured a bone? If Yes, which bones and when?	Has the patient ever fractured a bone? If Yes, check all that apply.
Osteoarthritis	Has your physician ever diagnosed you with any of the following conditions? What symptoms of XLH have you experienced in the last year?	Has the patient ever been diagnosed with or treated for any of the following conditions? Please document ALL major illness diagnoses the subject has ever had.
Osteophytes
Enthesopathy
Spinal stenosis
Dental abscess
Hip replacement	Have you ever had any of the following surgeries? Other surgery details.	Has the patient ever had any of the following surgeries? Other medical history?
Knee replacement

Abbreviation: XLH, X-linked hypophosphatemia.

Statistical Analysis

Statistical analyses were performed using Statistical Analysis Software (SAS; SAS Institute Inc., Cary, NC, USA). Data were analyzed using descriptive statistics, according to means and SD or as medians (25^th to 75^th percentile). Adults were bracketed into 5 age bands (18–29, 30–39, 40–49, 50–59, and ≥60 years) to estimate age-dependent changes in prevalence. Independent t tests were used to assess differences in ages, comparing the clinical trial group with the survey group, as well as comparing the clinical trial group with the survey pain subgroup. Chi-squared tests of independence were performed to examine differences between groups based on the participants’ sex. In addition, Chi-squared tests of independence were conducted to explore between-group differences for reported musculoskeletal features. For each between-group Chi-squared test conducted, first a comparison between the clinical trial and survey group was made, followed by a second comparison between the clinical trial group and survey pain subgroup. A statistical significance level was set to 5% (P = 0.05). Analyses were conducted in accordance with the requirements of the European Network of Centres for Pharmacoepidemiology and Pharmacovigilance, and the International Society for Pharmacoepidemiology (European Union Post-Authorization Study Register Number: EUPAS32590).

Results

Participant Demographics and Characteristics

In total, 127 of the 134 adults from the clinical trial group and 209 of the 232 adults from the survey group were eligible for inclusion in this analysis; in addition, 136 adults from the survey group were included in the survey pain subgroup (Fig. 1). The characteristics of the adults in each of the 3 groups (the clinical trial group, survey group, and survey pain subgroup) are shown in Table 2. The clinical trial group were younger than both the survey group (P = 0.0003) and survey pain subgroup (P = 0.004). While the majority of participants were women, the ratio of female to male participants was lower in the clinical trial group than in the survey group (P = 0.04) and survey pain subgroup (P = 0.004). Approximately half of the adults (54.3%) in the clinical trial group reported receiving oral phosphate and active vitamin D in the past 2 years. In the survey group, 40.7% reported currently taking oral phosphate and active vitamin D. Similar rates of oral phosphate and active vitamin D use was found in the survey pain subgroup (Table 2). Mean (SD) BPI worst pain score was 6.8 (1.7), 4.2 (2.1), and 5.4 (1.2) in the clinical trial group, survey group, and survey pain subgroup, respectively.

Table 2.

Patient demographics and characteristics

Characteristic	Clinical trial group (N = 127)	Survey group (N = 209)	Survey pain subgroup (N = 137)
Sex, n (%)
Female	84 (66.1)	160 (76.6)	112 (81.8)
Male	43 (33.9)	49 (23.4)	25 (18.2)
Region, n (%)
USA	66 (52.0)	132 (63.2)	90 (65.7)
Europe	47 (37.0)	50 (23.9)	36 (26.3)
Rest of world	14 (11.0)	27 (12.9)	11 (8.0)
Age, years
Mean ± SD	39.8 ± 12.3	45.5 ± 13.0	45.4 ± 12.3
Median	40.7	44.9	45.0
Range	18.5–65.5	18.3–74.5	18.3–74.5
Adults in each age band, n (%)
18–29 years	35 (27.6)	26 (12.4)	15 (10.2)
30–39 years	27 (21.3)	45 (21.5)	27 (20.4)
40–49 years	39 (30.7)	61 (29.2)	48 (34.3)
50–59 years	19 (15.0)	46 (22.0)	32 (24.1)
≥60 years	7 (5.5)	31 (14.8)	15 (10.9)
Age at diagnosis, years
Mean ± SD	9.4 ± 14.4	9.6 ± 11.8	10.5 ± 14.3
Median	2.4	3.0	3.0
Range	0.0–54.7	0.0–59.0	0.0–55.0
BMI
Mean ± SD	30.4 ± 7.7	30.2 ± 8.1	30.2 ± 7.6
Median	30.0	28.7	29.1
Range	17.4–64.6	15.1–72.0	15.1–63.9
Weight, kg
Mean ± SD	71.2 ± 18.9	69.4 ± 18.7	68.9 ± 18.0
Median	69.0	68.0	67.1
Range	36.1–139.6	28.0–162.0	28.0–124.7
Height, cm
Mean ± SD	152.8 ± 10.5	151.8 ± 10.9	150.7 ± 10.9
Median	152.0	152.4	152.0
Range	120.6–176.0	115.0–182.9	115.0–180.3
XLH treatment, n (%) a , b
Oral phosphates only	6 (4.7)	4 (1.9)	2 (1.5)
Active vitamin D only	13 (10.2)	24 (11.5)	11 (8.0)
Phosphates and active vitamin D	69 (54.3)	85 (40.7)	53 (38.7)
Phosphates and vitamin D (status unknown)	0	13 (6.2)	11 (8.0)
Vitamin D only (status unknown)	0	15 (7.2)	13 (9.5)
Neither phosphates nor active vitamin D	39 (30.7)	68 (32.5)	47 (34.3)

Abbreviations: BMI, body mass index; XLH, X-linked hypophosphatemia.

^aIn the clinical trial group, current treatment was assessed by medical data and refers to any treatment taken for XLH within the past 2 years.

^bIn the survey, current treatment was assessed by a participant-reported questionnaire and refers to any treatment taken for XLH that the participant is currently receiving.

History of Fractures

In the clinical trial group, 43.3% of adults reported ever having a fracture, which was similar to the reported history of fractures in both the survey group (43.5%, P = 0.97) and the survey pain subgroup (47.4%, P = 0.50). Even in the 18- to 29-year age band, more than 25% of adults reported having a fracture, increasing to more than 65% of those aged over 60 years (Fig. 2). Across all age bands, the most common fracture site was either the femur or hands/feet, except for the 18- to 29-years and ≥60-years age bands in the survey pain subgroup, where the most common fracture sites were hip, femur, tibia/fibula, and hands/feet (Fig. 3). Spinal, rib, and upper limb fracture sites were less frequent than lower limb fracture sites in all groups (Fig. 3).

Figure 2.

Adults reporting a history of fractures by age band across the 3 groups.

Figure 3.

Distribution of fracture sites by age band across the 3 groups; (a) clinical trial group (n = 127), (b) survey group (n = 209), and (c) survey pain subgroup (n = 136). *Adults could report fractures at more than one site. In the survey group, one pelvic fracture was included in ‘hip’.

Osteoarthritis, Arthroplasty, and Osteophytes

The overall prevalence of OA was > 50% in all 3 groups, and this generally increased with age (Fig. 4). Rates of reported OA in the clinical trial group (64.6%) were similar to the survey group (55.0%, P = 0.09) and the survey pain subgroup (62.8%, P = 0.76). OA was reported in 23.1% to 37.1% of the youngest age band (18-29 years) and > 50% of adults aged ≥40 years old. The rates of osteophytes were similar across the groups (Fig. 5), being reported by 41.7% of adults in the clinical trial group compared to 44.5% (P = 0.62) in the survey group, and 50.4% (P = 0.16) in the survey pain subgroup. Rates of arthroplasty are shown in Fig. 6, with more than 10% of adults aged 40 to 49 years reporting having had a hip arthroplasty (Fig. 6A) and more than 10% of adults aged 50 to 59 years having had knee arthroplasty (Fig. 6B).

Figure 4.

Adults with osteoarthritis by age band across the 3 groups.

Figure 5.

Adults with osteophytes by age band across the 3 groups.

Figure 6.

Adults who reported (a) hip arthroplasty and (b) knee arthroplasty by age across the 3 groups.

Enthesopathies

The overall prevalence of reported enthesopathy was significantly higher in the clinical trial group (39.4%) compared with both the survey group (26.3%, P = 0.01) and the survey pain subgroup (27%, P = 0.03). This difference was greatest between the ages of 18 and 59 years, with a lower rate in the clinical trial group among those aged ≥60 years (Fig. 7). However, in all 3 groups, there was a progressive increase in the prevalence with age.

Figure 7.

Adults with enthesopathies by age across the 3 groups.

Spinal Stenosis

Rates of reported spinal stenosis were similar in the clinical trial group (19.7%), compared with the survey group (18.7%, P = 0.82) and the survey pain subgroup (21.2%, P = 0.76). Spinal stenosis was reported as early as 18 to 29 years of age in the survey group and 30 to 39 years of age in the survey pain subgroup and the clinical trial group, increasing with age (Fig. 8).

Figure 8.

Adults with spinal stenosis by age across the 3 groups.

Seven adults reported having a history of spinal surgery: 5 in the clinical trial group and 2 in the survey group. All 5 surgeries reported in the clinical trial group were in female participants: 3 (1 in the 18–20-years age band; 2 in the 40–49-years age band) reported spinal fusion surgery, 1 (in the 50–59-years age band) reported intervertebral disc operations, spinal laminectomy, and spinal fusion surgery, and 1 (in the 40–49-years age band) reported spinal decompression surgery. In the survey group, 1 male participant in the 50- to 59-years age band reported spinal cord injury during spinal cord decompression surgery as a result of T10-T11 spinal stenosis, and a female participant in the 40- to 49-years age band described undergoing 3 decompressive laminectomies for spinal stenosis.

Musculoskeletal Features

The proportion of adults with ≥ 2 of the musculoskeletal features was 59.1%, 55.0%, and 61.3% in the clinical trial group, survey group, and survey pain subgroup, respectively. The pattern of the prespecified musculoskeletal features varied by age band (Fig. 9). All of the musculoskeletal features were more common with aging, with the highest prevalence in OA and greatest relative increase with age seen in spinal stenosis. By the 40- to 49-years age band, all musculoskeletal features reported at rates ≥ 20% (Fig. 9C and 9D).

Figure 9.

Distribution of musculoskeletal features across the 3 groups by age (a) 18–29 years, (b) 30–39 years, (c) 40–49 years, and (d) ≥50 years.

The number of musculoskeletal features per adult increased with age. The proportion of adults with ≥ 2 musculoskeletal features in all groups increased with age from 29% (clinical trial group), 23% (survey group), and 33% (survey pain subgroup) in the 18- to 29-years age band to 74%, 62%, and 67%, respectively, in the 40- to 49-years age band. By the age of 40 to 49 years, 3% to 15% of participants reported all 5 of the prespecified musculoskeletal features (Fig. 10).

Figure 10.

Cumulative distribution of musculoskeletal features by age in the 3 groups.

There was no evidence of any correlation between short stature and the presence of musculoskeletal features in the clinical trial group (P = 0.3760), survey group (P = 0.9674), or survey pain subgroup (P = 0.7353); nor the number of musculoskeletal features in the clinical trial group (P = 0.3701), survey group (P = 0.4783), or survey pain group (P = 0.9091). There was no evidence of any correlation between BMI and the number of musculoskeletal features in the clinical trial group (P = 0.1020) or survey pain group (P = 0.619), but a significant correlation with BMI was observed in the survey group (P = 0.0007).

Dental Abscesses

Reported rates of a dental abscess history were high in all groups, with rates in the clinical trial group lower at 63.0% compared with the survey group (79.9%, P = 0.001) and survey pain group (81.8%, P = 0.001). High rates of dental abscess were seen across all age groups, even in the 18- to 29-years age group where > 55% of patients in the clinical trial, survey, and survey pain groups reported dental abscesses (Fig. 11).

Figure 11.

Adults with dental abscesses by age across the 3 groups.

Discussion

The findings from this study highlight the frequency of clinically important musculoskeletal features in adults with XLH and their early onset, illustrating a scenario of premature musculoskeletal morbidity compared with the general population. This is the first analysis to characterize the accumulation of XLH-related musculoskeletal features across the lifespan of adults with XLH. All the musculoskeletal features were more common with increasing age, with OA being the most prevalent and spinal stenosis showing the greatest relative increase with age. By the 40- to 49-years age band, the prevalence of these musculoskeletal features was over 20%. Unlike other features that accumulate with time, dental abscesses are often present during childhood in XLH; therefore, it is not surprising that dental abscesses were commonly reported across all age groups in this analysis.

We anticipated that with similar worst pain scores between the clinical trial group and the survey pain subgroup, both groups would be likely to report comparable frequencies of musculoskeletal features. The reported rates of musculoskeletal features were similar for OA, with lower rates in the broad survey group where adults had lower levels of self-reported pain. This finding suggests there may be an association between greater pain levels and the presence of OA. The reporting of fractures was less consistent among groups, an observation that may relate to the fact that fractures may have occurred many years previously, and prior, healed fractures would not be expected to cause significant pain at the time of later reporting. There were no similarities in the rates of reported osteophytes, enthesopathies, and spinal stenosis, as patients are often not aware of these particular diagnoses even when manifest in themselves, which may have affected the survey reporting.

When compared with the presence of musculoskeletal features in the general population, these features occurred at a higher prevalence and earlier age in adults with XLH. For fractures, the prevalence and sites of fracture are different from the pattern seen in the general population (10). The high prevalence of lower limb fractures in younger adults, including the hip and femur, contrasts with the usual age-related incidence of hip fractures, with the mean age of hip fracture being over 80 years of age (11). This highlights the different mechanism of bone fragility in XLH compared with the general population, namely the lifelong hypophosphatemia altering the microarchitecture of cortical bone causing osteomalacia in combination with skeletal deformities from childhood (12).

The rate of OA in this analysis of adults with XLH exceeds the rates in the general population, especially for younger ages. A diagnosis of OA was reported in > 50% of adults with XLH across all groups in the 40- to 49-year age band. By contrast, in an analysis of a general population aged ≥ 45 years, only 28% had radiographic knee OA and 16% symptomatic knee OA (13). In a similar analysis, 28% had radiographic hip OA and 10% symptomatic hip OA (14). Rates of symptomatic OA reported in adults with XLH in their 20s in this study were higher than those seen in adults > 45 years of age in the general population.

Arthroplasty is reserved for people with the most severe OA (15). In all 3 groups, some adults in their 30s and 40s reported a history of hip arthroplasty, while a number of adults in all age bands ≥ 40 years reported having had knee arthroplasty. We found a notably higher rate of hip and knee arthroplasty in young adults with XLH compared with the United States general population aged <50 years, which have been reported to be 0.11% and 0.08%, respectively (16). In the UK general population, the average age for these surgeries has been reported as 69 years for hip arthroplasty (17) and 70 years for knee arthroplasty (18, 19). In this analysis, hip and knee arthroplasty were reported at a substantially younger age in adults with XLH than in the UK general population, with hip arthroplasty reported from 30 to 39 years of age and knee arthroplasty from 40 to 49 years of age.

In the general population, a younger age of arthroplasty is associated with a poorer symptomatic benefit and a shorter time to joint revision. Additional work is required to ascertain the long-term clinical symptoms and implant survival in adults with XLH, as these could lead to a substantial impact on quality of life as well as health economic burden (20). Indeed, according to a recent thematic analysis by Ferizović et al, adults with XLH reported orthopedic surgery, pain, mobility impairment, fatigue, and dental problems as some of the most common sources of burden (21).

A United States retrospective analysis assessing people aged > 65 years between 2002 and 2007 reported the average age of surgery for spinal stenosis as ~75 years (22). Although the overall rates of spinal surgery were low in the present analysis, participants reported spinal stenosis in the youngest age band (18-29 years old); surgery for spinal stenosis included both decompressive and fusion surgery. Among these few cases, the need for multiple surgeries and complications from surgery were reported in line with the reported risks of spinal surgery (22).

This study has some limitations. First, it is possible that adults with a BPI worst pain score ≥4 participating in the survey could subsequently have been recruited to the clinical trial, and thereby may have been included in both data sets. However, we propose that the likelihood of participation in the survey and the clinical trial is low, with no instances of this occurrence brought to our attention by participating investigators. In addition, selection bias in the survey group cannot be ruled out or quantified, as survey respondents may have represented more severe and/or more engaged patients with XLH. Therefore, the overall prevalence or severity of prespecified features may have been overestimated by our study’s methodology. Finally, pain in XLH is multifactorial, and it is often difficult to assess whether pain is due to OA, osteomalacia, or other aspects of XLH. To minimize the effect of variable disease severity, we performed an analysis of a subset with a BPI score of ≥4. We also note that 60% of survey respondents had BPI pain scores ≥4, which is typical of the adult XLH population (23), and the combined cohort arising from this study represents the largest survey of adults with XLH. Together, these findings suggest the results are likely to be generalizable to the larger adult population with XLH.

The analysis reported here was based on a combination of physician diagnosis and self-reporting in the clinical trial and participant self-reporting in the survey, without radiographic imaging to confirm if features were present. No definitions of musculoskeletal features were provided to the participants or clinicians. It is therefore possible that some patients may have answered some survey questions without complete knowledge about the definition of certain medical terms or procedures. This may have resulted in erroneous reporting of certain features. Although not tested, we do anticipate that clinicians would be able to make these distinctions for patients within the trial. In addition, this analysis only assessed self-reported history of fractures; quantifying the prevalence of healed fractures and also pseudofractures requires radiographic diagnosis and was therefore not feasible in this analysis. Additional research should consider assessing radiographic data over the lifetime of patients with XLH to further elucidate the natural history of XLH and to compare the frequency of morbidities on imaging with participant-reported data. Furthermore, the cause of fracture was not assessed. However, patients with XLH typically sustain osteomalacic fractures at any time in adult life (24). That said, we cannot completely rule out osteoporosis as a cause in those aged >60 years, in the absence of specialized diagnostic testing such as peripheral quantitative computed tomography or transiliac bone biopsy.

The actual age of diagnosis for the musculoskeletal features in this report was not obtained in either of the data sets; only their presence/prior occurrence or absence in relationship to the patient’s age at the time the historical data were collected for the study. The data available did not allow for a robust investigation of the relationship between musculoskeletal features and the effect of time. Furthermore, if those in the older age bands developed the reported features at a younger age, the reported age distribution may reflect historical differences in management strategies. Further research should consider additional time-precise data (including date of diagnosis for features presented) to analyze the effect of time and historic management on prevalence. In addition, a country-specific review of musculoskeletal features in the general population with respect to age of onset would be useful to understand the impact of differences in populations and healthcare services.

There was no information on the severity of the musculoskeletal features. Although arthroplasty is usually reserved for patients with the most severe OA, a number of additional factors may result in some people with severe OA not having surgery in the general population (25, 26). Parameters specific to XLH such as limb deformity and previous surgeries may also increase the surgical complexity and risks. Understanding the severity and evolution of OA, for example, would enhance knowledge about the trajectory of OA in adults with XLH compared with the general population.

Finally, we were not able to explore all relevant lines of inquiry in full. For example, BMI was elevated in this population and a correlation was found in the survey group with number of musculoskeletal features. However, we had insufficient historical data to understand the relevant contribution of factors that might contribute to elevations in BMI, including family history of overweight or obesity, reduced physical activity (due to pain or otherwise), or increased caloric intake. It would be interesting in future studies to explore the basis for the increased BMI using total body fat mass and fat mass index by dual-energy x-ray absorptiometry in relationship with musculoskeletal XLH features.

Survey respondents were asked to confirm the diagnosis of XLH, including whether they had received genetic confirmation of the condition with a positive PHEX variant. In total, 39% of adults self-reported PHEX variants, but variants were not verified by medical records (7). Of note, 95% of the clinical trial population had confirmed PHEX variants, supporting the validity of the data from that group (9). Further research assessing radiographic data in individuals with confirmed PHEX mutations would help to verify the data from the contributing survey.

In conclusion, this analysis of the clinical trial and the survey group has demonstrated that adults with XLH bear a substantial burden of multiple musculoskeletal features, emerging decades earlier than in the general population, as early as in their 20s, and progressively accumulating with age.

Glossary

Abbreviations

1,25(OH)2D

1,25-dihydroxyvitamin D (active vitamin D)

BMI

body mass index

BPI

Brief Pain Inventory

FGF23

fibroblast growth factor 23

OA

osteoarthritis

XLH

X-linked hypophosphatemia

Grants or Fellowships

None.

Additional Information

Disclosures: The following authors served as clinical investigators for one or more studies, including this study, sponsored by Ultragenyx Pharmaceutical Inc. in partnership with Kyowa Kirin International plc: M.K.J., L.W., K.I., and E.A.I. In addition, L.W. and E.A.I. have also received honoraria for serving as an advisory board member or for lectures from Ultragenyx Pharmaceutical Inc. R.P. reports personal fees from Mereo Biopharma and Kyowa Kirin International plc. and grants from Mereo Biopharma, outside the scope of the submitted work. A.R. and A.W. are employees of Kyowa Kirin International plc.

Data Availability

Restrictions apply to the availability of all data generated or analyzed during this study to preserve patient confidentiality. The corresponding author can on request detail the restrictions and any conditions under which access to some data may be provided.