Long-term suppression of alkaline phosphatase levels and clinical remission with bisphosphonate therapy in two patients with juvenile Paget disease

INTRODUCTION

Bakwin and Eiger first described juvenile Paget disease (JPD) in 1956.^[1] It is an extremely rare bone disease, with only around 80 cases having been reported since then.^[2] It is caused due to loss of osteoprotegerin (OPG),^[3] which acts as a decoy receptor for RANKL (receptor activator of nuclear factor kappa-B ligand) and prevents its interaction with RANK. Unopposed activating RANK signaling causes increased osteoclast number and activity.^[4] The majority of patients with JPD have mutations in the TNFRSF11B gene that codes for OPG.^[5] JPD results in fast remodeling and increased bone turnover, leading to painful inflammatory bone-like disease with bone expansion, frequent fractures, and deformities. This, as expected, can lead to significant morbidity in affected children.

On reviewing the literature, we found that the majority of published case reports on JPD lacked genetic testing to confirm the diagnosis as well as long-term clinical and laboratory follow-up.^[2] We also observed that most patients had an inadequate suppression of alkaline phosphatase (ALP) with bisphosphonate therapy. We report two cases of genetically proven JPD who experienced not only clinical remission but also sustained ALP suppression for over 2 years.

CASE REPORTS

Case 1

A 2.8-year-old girl was brought in with complaints of increased head size since early infancy, upper and lower-limb deformities, severe bone pain, and inability to walk. She had predominantly motor developmental delay. She was an only child to parents with 3^rd-degree consanguinity. She was born at term, with a birth weight of 2.5 kg. There was no significant family history. On examination, her weight (8.4 kg) and length (78 cm) were <3 standard deviation (SD) below the mean, and her occipitofrontal circumference (52 cm) was >3 SD. She had frontal bossing, hypertelorism, a rigid upper spine, and bowing of forearms and legs. On clinical examination, there were no extra-skeletal manifestations of JPD.

Her baseline biochemical levels are shown in Table 1. Investigations showed normal levels of serum calcium, phosphorus, and parathyroid hormone (PTH). 25-hydroxyvitamin D was 23.6 ng/mL. There was marked elevation of ALP, procollagen type I N-terminal propeptide, and beta-CrossLaps, suggestive of rapid bone turnover. Skeletal radiographs showed diffuse cortical thickening with periostitis, new bone formation, and medullary reaction with coarsened trabecular pattern in the long bones with Tc-99m methylene diphosphonate bone scan showing increased tracer activity in the skull, humerus, tibia, and femur [Figure 1]. Clinical exome sequencing identified a homozygous mutation in TNFRSF11B (TNFRSF11B(-)/Exon 2/c.344T>G(p.Ile115Arg)/Homozygous/Autosomal recessive/Uncertain significance), thus making the diagnosis of JPD highly probable. She was started on bisphosphonate therapy and received three cycles of pamidronate (7 mg/kg/year) in the 1^st year, followed by zoledronate (0.05 mg/kg/dose) at 6-monthly intervals. There was a significant decrease in pain and improved ability to weight bear within 3 weeks. There were no documented hypocalcemia episodes at any point. Serial monitoring of serum ALP levels showed a downward trend with age-appropriate levels after 2 years of therapy. This was followed by sustained remission till the latest follow-up. She also achieved a height velocity of 6 cm/year [Figure 2] and attained age-appropriate motor milestones. Since she was raised by a single parent, her height could not be compared to her mid-parental height. At her last clinical assessment at 6 years and 8 months of age, her height was 103 cm (–2.54 Z-score vs. —3.4 Z-score at baseline), and her weight was 14.9 kg (–1.92 Z-score vs. –3.3 Z-score at baseline). There was significant clinical improvement in her deformities, and she has been able to attend school along with her peers. Figure 3 shows the trend of ALP levels monitored for 3 years.

Table 1: Baseline biochemical characteristics of Cases 1 and 2 showing elevated bone turnover markers.

Baseline parameters	Case 1	Case 2
Serum calcium (mg/dL)	9.28	9.38
Serum phosphorus (mg/dL)	4.1	5.4
25-hydroxyvitamin D (ng/mL)	23.6	32.9
Serum parathyroid hormone (pg/mL)	33.1	25.7
Bone turnover markers	Case 1	Case 2
Alkaline phosphatase (U/L)	1419	2383
Procollagen type 1 N-propeptide (ng/mL) Reference range: Females (pre-menopause): 15.1–58.3 ng/mL	>1200	Not done
Beta-CrossLaps in serum (pg/mL) Reference range: Females (pre-menopause): 137–573 ng/mL	1070	Not done

Close

Figure 1:

Case 1: Plain radiographs of the (a) skull (lateral view), (b) bilateral hands and wrists (AP view), (c) pelvis (AP view), and (d) lower limbs (AP view) showing diffuse cortical thickening, coarsened trabecular pattern with periostitis; (e) Tc99m-methylene diphosphonate bone scan showing increased tracer uptake in skull, humerus, tibia, and femur. Case 2: Plain radiographs of (f) bilateral hands and wrists (AP view) and (g) left humerus (AP view) showing similar findings suggestive of rapid bone turnover and remodeling.

Export to PPT

Close

Figure 2:

Growth chart of Case 1 showing steady growth velocity following initiation of bisphosphonate therapy.

Export to PPT

Close

Figure 3:

Downward trend in alkaline phosphatase (ALP) (U/L) for both patients since initiation of treatment, with sustained remission to the normal range on follow-up.

Export to PPT

Case 2

A 2.5-year-old boy was taken to a local center with complaints of recurrent falls and bilateral genu valgum since 15 months of age. He was also an only child to parents with 3^rd-degree consanguinity. He was born at term, with a birth weight of 3 kg. He had no significant family history. Evaluation done elsewhere revealed normal serum calcium, serum phosphorus, and 25-hydroxyvitamin D, but an elevated ALP. Suspecting nutritional rickets, he was supplemented with 25-hydroxyvitamin D and calcium. However, there was no clinical improvement and he continued to have fractures to trivial trauma, including the right clavicle and left humerus at 26 months and 30 months of age, respectively. In view of persistent hyperphosphatasia, ALP was checked serially (2147 U/L, 1206 U/L, and 2485 U/L), and genetic analysis was done. Homozygous mutation in TNFRSF11B (TNFRSF11B(-)/Exon 5/c.997C>T(p.Arg333Ter)/Homozygous/Autosomal recessive/Pathogenic) was identified in the clinical exome, confirming the diagnosis of JPD. He was subsequently referred to our institution for further management.

On arrival, his height was 89 cm (–1.42 Z-score) with weight of 12.9 kg (–0.9 Z-score). Biochemical investigations [Table 1] showed normal levels of serum calcium, phosphorus, and PTH, along with sufficient 25-hydroxyvitamin D. Hyperphosphatasia with ALP of 2383 U/L was noted here as well. Skeletal radiographs [Figure 1] showed diffuse long bone expansion with marked cortical thickening and a coarsened trabecular pattern suggestive of rapid bone turnover and remodeling. He was started on 4-monthly pamidronate therapy and has received five cycles so far (8–9 mg/kg/year). There was significant subjective improvement in bone pain. On the last follow-up at 4 years and 9 months of age, he was ambulant and able to play outdoors. Figure 3 shows the trend of ALP levels monitored for 2 years following initiation of bisphosphonate therapy. He also has a good height velocity of 7 cm/year with no further fractures following initiation of therapy [Figure 4]. His height was 103 cm (–1.18 Z-score vs. –1.3 Z-score at baseline), and his weight was 14.9 kg (–0.46 Z-score vs. –0.95 Z-score at baseline). He was noted to have sustained remission from 9 months of initiation of treatment to the current follow-up.

Close

Figure 4:

Growth chart of Case 2 showing improved growth velocity following initiation of bisphosphonate therapy.

Export to PPT

DISCUSSION

JPD must be suspected in children with failure to thrive, skeletal deformities, bony pain, and low-velocity fractures. Rare, but well-documented extra-skeletal manifestations include hearing loss, retinopathy, vascular calcifications, internal carotid artery aneurysms, pulmonary complications, along with a host of endocrine manifestations such as hypogonadism, hypothyroidism, and growth hormone deficiency. Biochemical parameters reveal elevated bone turnover markers, and radiographs show bone expansion along with cortical thickening. Diagnosis is confirmed by mutation analysis of the TNFRSF11B gene. In view of the rarity of the disease, there are no established treatment guidelines. Hence, in resource-constrained settings, establishing a diagnosis and administering appropriate management can be extremely challenging.

Conventionally, calcitonin (both porcine and human derived) has been used to treat JPD.^[6,7] With calcitonin, patients have reported subjective improvement in bone pain and mobility, but it did not normalize ALP levels.^[6] More recently, recombinant OPG and denosumab—a human monoclonal neutralizing antibody against RANKL – have been found to be effective in selected patients since they prevent RANKL binding to RANK.^[8] In 2005, Cundy et al. described two adult patients with JPD treated with recombinant OPG, resulting in clinical improvement along with suppression of ALP levels to the normal range.^[9] Both required treatment for hypocalcemia due to hungry bone syndrome. At present, it is not available for routine clinical use and no study has been done in children yet. Grasemann et al., in 2013, used denosumab for the first time in an 8-year-old girl with JPD.^[8] Her bone pain was well controlled and there was sustained remission of bone turnover markers including ALP for a period of 5 months. However, following the first dose, she developed severe symptomatic hypocalcemia, requiring a prolonged hospital stay. Both these drugs can cause dangerous hypocalcemia, but more concerning is their exorbitant cost, making their widespread use difficult in lower-and middle-income countries (LMICs).

Bisphosphonates are analogs of pyrophosphate (PPi), which causes effective inhibition of osteoclast function with reduction of bone resorption, and have been widely used to treat various metabolic bone disorders. Bisphosphonate therapy, though more effective than calcitonin and more affordable than denosumab and recombinant OPG, is still considered inadequate in view of its lack of sustained control of ALP levels. Furthermore, hypocalcemia, hypophosphatemia, and secondary hyperparathyroidism have been found to complicate treatment.^[2] In the literature, various bisphosphonates have been used, in varying doses, including pamidronate (8 mg/kg daily PO or 0.75–1.75 mg/kg IV periodically), etidronate (20 mg/kg daily PO), risedronate (30 mg daily PO), alendronate (10 mg daily or 70 mg weekly PO), and ibandronate (5 mg IV periodically).^[2] However, zoledronate (3–5 mg IV periodically) may possibly be more efficient in normalizing ALP levels.^[10] Demir et al.^[11] and Tau et al.,^[12] in 2000 and 2004, respectively, showed normalization of ALP in children with chronic idiopathic hyperphosphatasia, however neither had genetically proven JPD. More recently, Prata et al.^[13] and Hoppner et al.^[14] have shown more promising results in three children with genetically proven JPD, with all of them being followed up till adulthood.

Our cases are therefore among the few cases in the literature where cyclical use of bisphosphonates resulted in normalization of ALP levels in genetically proven JPD. On follow-up of our patients, sustained remission, both clinically and in ALP levels were noted. During this period, they attained age-appropriate motor milestones, with no further deformities or fractures.

CONCLUSION

Cyclical bisphosphonate therapy showed favorable results by normalizing bone turnover markers and alleviating symptoms in both our patients with JPD, which persisted on long-term follow-up. Given the sparsity of literature on this disease and the lack of practice guidelines, our findings are encouraging for LMICs, where access and affordability issues make denosumab and recombinant OPG less feasible.