Continuous improvement of bone mineral density two years post zoledronic acid discontinuation in patients with thalassemia-induced osteoporosis: long-term follow-up of a randomized, placebo-controlled trial

β-thalassemia major (TM) is often accompanied by osteopenia or osteoporosis.1,2 Bisphosphonates have been used in the management of TM-induced bone loss.3–7 We report here the results of the long-term follow-up of our TM patients with osteoporosis who participated in a randomized, placebo-controlled trial with zoledronic acid (ZOL), the results of which at 12 months had been previously reported.5 According to that study, 66 patients with TM-induced osteoporosis (21M/45F; median age 35.5 years) were studied. Patients were blindly randomized to receive 4 mg ZOL, iv, in 15 min. infusion, every six months (group A, n=23) or every three months (group B, n=21), or to receive placebo every three months (group C, n=22), for a period of one year.

Patients of groups A and B then discontinued ZOL treatment, while patients of group C were given 4 mg ZOL, every three months, for 12 months, and subsequently stopped ZOL therapy.5 Informed consent was obtained from all patients prior to inclusion in the study. The study was conducted with the approval of the Greek National Drug Organization Committee (Ref. No Aα-KΔ-79/01/03) in keeping with the guidelines of Helsinki. Bone Mineral Density of the lumbar spine (L1–L4), femoral neck (FN) and wrist was evaluated in all patients prior to starting therapy and then after 12 and 36 months, using DEXA.5 Serum markers of bone remodeling [C-telopeptide of collagen type-I (CTX) and bone-specific alkaline phosphatase (bALP)] were measured in patients and in a control group of 40, age and gender-matched, healthy blood donors, as previously described.5 Patients were asked to quantify their degree of bone pain on Huskisson’s visual analog scale (VAS) and the McGill–Melzack scoring system before entering the trial, and then every six months for 36 months. Statistical analysis was described in our initial report.5

Patients of groups A and C showed no differences in BMD of all evaluated sites at the 12 month, while patients of group B achieved a significant increase only in their L1–L4 BMD5 (Figure 1A-1C). ZOL also reduced bone pain in groups A+B, while there was no pain relief in the placebo group after 12 months of therapy (Table 1). Similar results have been described in two subsequent trials where ZOL was given in TM patients with osteoporosis for 12 months6,7 confirming the beneficial effect of ZOL in this setting. However, the treatment duration has not been clarified in any trial.

Our patients were followed for 24 months after discontinuation of ZOL for groups A and B, and for 12 months for group C. Interestingly, at the 36 month, patients of both groups A and B showed a significant increase in BMD of all studied sites compared with baseline values (p<0.01) (Figure 1A-1B). Patients of group C, who had received ZOL for 12 months after the placebo period, also increased their BMD of all studied sites at the 36 month (p<0.01; Figure 1C), while they also reduced bone pain scores (Table 1). Furthermore, there was no more difference in BMD T-values of L1–L4 and forearm between patients of groups A, B, and C, while FN BMD continued to be higher in group B (either as a T-score absolute value or as a T-score percentage change; p=0.01). The BMD elevation observed in groups A and C was accompanied by a comparable reduction in CTX at the 36 month, which had not been reported at the 12 month (Table 1). These observations suggest that ZOL continues to act after its discontinuation. ZOL may need more time before its efficacy becomes evident in patients who receive it every six months. Pharmacokinetic studies showed that bisphosphonates remain in bone matrix for many years, normalize bone turnover rates within weeks and no further suppression is seen during long-term use.8 This indicates that long-term treatment may not be very different from short-term treatment.8

Figure 1.(A) Patients of group A showed a slight, but not significant, increase of T-score in BMD of all studied sites at 12 months, which dramatically increased at month 36 (mean ± SD): L1–L4: from −3.3 (±0.9) at baseline, to −2.8 (±1.19) at 12^th month and −2.2 (±1.0) at 36^th month; Femoral neck: from -1.9 (±1.5) at baseline, to −1.8 (±0.8) at 12^th month and −1.6 (±0.7) at 36^th month Wrist: from −4.0 (±1.7) at baseline, to −3.7 (±2.0) at 12^th month and −2.8 (±1.1) at 36^th month. (B) Patients of group B showed a statistically significant increase of L1–L4 BMD at 12 months which was continued at 36^th month: from −2.9 (±0.9) at baseline, to −2.2 (±1.0) at 12^th month and −1.7 (±0.7) at 36^th month. They also showed a borderline increase in FN BMD at 12 months [from −1.7 (±0.8) to −1.4 (±0.6)], which became significant at 36^th month (−0.9±0.9). The respective values for T-score of the wrist were: −3.1 (±1.1) at baseline, −2.7 (±1.0) at 12^th month and −2.3 (±0.9) at 36^th month. (C) Finally, patients of group C showed no significant alterations of BMD in all studied sites at 12^th month that improved at 36^th month after zoledronic acid administration: L1–L4: from −2.5 (±0.7) at baseline, to −2.5 (±0.9) at 12^th month and to −1.9 (±0.9) at 36^th month; Femoral neck: from −2.0 (±0.7) at baseline, to −2.1 (±0.8) at 12^th month and −1.6 (±0.6) at 36^th month Wrist: from −3.8 (±1.5) at baseline, to −3.9 (±1.8) at 12^th month and −2.5 (±1.1) at 36^th month.

Table 1.Pain scores and markers of bone remodeling during study period.

Thus some of the skeletal effects of bisphosphonates may last for years after treatment stops. Bisphosphonates bind strongly to hydroxyapatite. When bone containing bisphosphonate is resorbed, some of the bisphosphonate released recirculates locally and systemically, and binds again to bone surfaces. Therefore, when long-term treatment with bisphosphonates stops, the residual and recirculating bisphosphonate continues to inhibit bone resorption, although to a lesser extent than continued treatment.9,10

The retention of ZOL in bone may justify the continuous effect after its discontinuation observed in our study. In addition, recent pre-clinical reports established that the bone protection of zoledronic acid was dose-dependent and lasted for up to 32 weeks,11 thus supporting the rationale for an annual intravenous dosing regimen of ZOL for treatment of postmenopausal osteoporosis and confirming our results for the prolonged effect of zoledronic acid in our patients. In a very interesting study on the effect of alendronate on BMD and markers of bone remodeling, authors compared patients who continuously received alendronate with those who received alendronate and then stopped for five years.12 After five years, the cumulative risk of non-vertebral fractures was not significantly different between those continuing and discontinuing alendronate. Moreover, BMD continued to be elevated and CTX to be reduced compared to baseline after five years of alendronate discontinuation.12 Comparable results in terms of vertebral fractures have been reported with pamidronate in postmenopausal osteoporosis.10 In our study, we observed no new fractures during the study period although ZOL acid had been stopped. Furthermore, the changes in bone markers after the discontinuation of ZOL in groups A and B also suggest some residual effect of the drug for 24 months after its discontinuation.

No significant side-effects, including osteonecrosis of the jaw, were observed during this study. In conclusion, our study shows for the first time in the literature that ZOL continues to improve BMD, bone metabolism and pain scores 24 months after its discontinuation in TM patients with osteoporosis. We also confirm the safety of ZOL for up to 24 months after treatment. However, well-controlled studies with longer follow-up periods are needed to determine the best dose and treatment schedule of ZOL in the management of TM-induced osteoporosis.