At the time of her initial presentation, Ms X was 26 years old with a history of narcolepsy and juvenile arthritis, characterised by pain, stiffness, erosive synovitis and enthesitis since age 13, with numerous immunosuppressants trialed in the past with limited success (etanercept, adalimumab, tocilizumab, methotrexate, leflunomide and abatacept). In the preceding two years, she had bilateral ulnar shortening osteotomies for painful ulnocarpal abutment, complicated by non-union requiring bone grafts. Ms X sought an endocrine opinion to assess her delayed bone healing, however shortly prior to initial review, she received an iron infusion that exacerbated mild pre-existing hypophosphatemia (nadir 0.35mmol/L) that in retrospect was first reported at age 19. For her juvenile arthritis, Ms X received intermittent courses of systemic steroids, however these were brief and insufficient to explain her poor bone health.
History revealed normal growth and development to achieve a height of 163cm and weight of 65kg, and Ms X was notably a competitive swimmer and medical graduate. She sustained no fractures as a child, however noted brittle dentition since her teens. There was no significant family history. In January 2020, investigations revealed a raised fractional excretion of phosphate of 25.65% (and low TmP/GFR 0.53), with an inappropriate FGF-23 of 78ng/L, associated with an unusually low bone turnover: alkaline phosphatase (39U/L), CTx 172ng/L and P1NP 21mcg/L (see Table 1). Screening for Fanconi’s syndrome was negative. There was no evidence of hyperparathyroidism or vitamin D deficiency and DXA was unremarkable (April 2019 Z-scores: lumbar spine 0.1, total hip 0.0, femoral neck -0.2). Osteomalacia was confirmed on tetracycline double-labelled bone biopsy performed in October 2020. There was increased osteoid, no osteoclasts were identified, osteoblasts were scant and there was decreased blurred double labelling.
Targeted whole exome sequencing including known genetic causes of hypophosphatemic osteomalacia, hypophosphatasia and osteogenesis imperfecta were negative (see Table 2), and GaTate PET did not identify any mesenchymal tumours that could be causing tumour-induced osteomalacia. There was no evidence of Chiari malformation on MRI brain.
Ms X was managed with phosphate, colecalciferol and calcitriol supplementation, gradually titrated up to the current doses: phosphate phebra 1.5-2g daily in 2-3 divided doses, cholecalciferol 1000units daily and calcitriol 0.5mcg BD. Her most recent biochemistry on that regimen revealed a low normal phosphate of 0.83mmol/L, and calcium was 2.33mmol/L corrected. Unfortunately since her initial consultation, Ms X has sustained a right sixth rib fracture, bilateral cuboid stress fractures, a fracture of the left fourth and fifth metacarpals and her right ulnar shortening osteotomy site that was treated with a bone graft has not yet healed. Bone and joint pain did not improve following normalisation of her phosphate.
In summary, Ms X has hypophosphatemic osteomalacia in the context of inflammatory arthritis that appears to be genetic rather than acquired, given her young age of onset and lack of evidence of acquired causes such as tumour-induced osteomalacia on imaging. The atypical feature is the persistently suppressed bone turnover markers and the consequent delayed fracture healing.
Ms X’s phenotype is most consistent with X-linked hypophosphatemia (XLH). XLH is the most common form of heritable rickets or osteomalacia, caused by loss of function mutations in the phosphate-regulating endopeptidase homolog, X-linked (PHEX) gene (Carpenter 2011). It is characterised by renal phosphate wasting and has a diverse presentation, varying from isolated hypophosphatemia to significant lower limb bowing and other musculoskeletal complaints.
While XLH frequently manifests in childhood, it may be identified in adulthood. Among other features, fractures, bone pain, abnormal dentition and enthesopathy may be present, all of which affected Ms X (Lambert 2019). Adults are also predisposed to secondary and tertiary hyperparathyroidism, in addition to metabolic disorders including obesity and cardiovascular disease (Lecoq 2020).
Traditional management includes supplementing phosphate and calcitriol. However, this often does not fully correct hypophosphatemia and bony manifestations. Burosumab (KRN23), a novel recombinant human antibody against FGF-23, seems to be a promising therapy. An open-label phase 2 trial of burosumab in children demonstrated improved renal tubular phosphate reabsorption, serum phosphate, linear growth, decreased pain and increased function (Carpenter 2018). Further, FGF-23 is affected by inflammation (Lang 2018) and its levels have been linked to the activity of inflammatory arthritis such as rheumatoid arthritis (Sato 2016), although further research into this relationship is required. This is pertinent to Ms X’s case, and it may be useful to correlate inflammatory markers and disease activity with FGF-23 levels.
Despite fitting the clinical phenotype for XLH, it is peculiar that Ms X’s gene testing was negative and her bone turnover markers have been persistently suppressed. This poses a diagnostic and management dilemma as genetic counselling for family members is limited and management is now guided by the clinical picture and consensus opinion. Ms X is on the waitlist for compassionate access to burosumab and receives regular surveillance and reviews in the meantime. The ultimate goal would be to achieve bone healing and prevent further fractures in the future.
Take home messages