We hypothesized that denosumab would be effective in preserving or increasing BMD in KTRs in routine clinical practice, even when there is a significant time gap between transplantation and the start of treatment.

Our study investigated the long-term changes in BMD in KTRs receiving denosumab for 4 years and compared them to the natural history of a matched untreated KTR cohort in a real-life scenario. The denosumab group exhibited significant increases in BMD from baseline at the LS and TH, whereas the untreated cohort experienced significant BMD loss at all sites. Although the data from the literature are limited, our findings are in line with those of previous reports [16,17,18] and corroborate the effectiveness of denosumab in increasing BMD in this special population.

In addition, the present study adds significant insights on the topic; for instance, the POSTOP study, a 12-month open-label randomized controlled trial [17], showed for the first-time the benefits of denosumab treatment in this special population. However, in that trial, the treatment course was started soon after the transplant surgery, a scenario that may not necessarily be generalizable to those patients whose bone health assessment might have been overlooked for several years. The average time gap between the transplant surgery and the first DXA evaluation in our study was almost 8 years and nevertheless denosumab showed significant BMD gains over the 4 years of follow-up. Furthermore, the treatment period of the POSTOP trial was limited to twelve months, a timeframe in which the untreated group BMD appeared to be stable.

In the present study, at the 4-year mark, a notable difference in the respective trajectories of the two groups becomes evident when considering the Z-score values. As is well-established, the Z-score represents the number of standard deviations that distinguish the patient's data from the mean of a healthy population of the same age, thereby accounting for the natural decline in BMD associated with advancing age over time. A significant decrease from the baseline of the Z-score marks a BMD decline inappropriate for aging and therefore emphasizes the decline in skeletal health of these patients, a well-documented multifactorial phenomenon [19]. Our study suggests that, in KTRs, denosumab may be not only able to stop this trend but even reverse it, as shown by the converging and intersecting lines of the Z-scores depicted in Fig. 2. As a consequence, at the end of the follow-up, several patients of the treated cohort reached or even exceeded the BMD of the untreated individuals.

In a recent retrospective study, McKee et al. compared the effects on BMD of forty-six patients treated with denosumab to thirty-nine receiving bisphosphonates [18] with 3.4 years of follow-up. With a comparable treatment duration, the Authors observed a mean BMD increase at the LS of roughly 7.5%, in line with ours (9.05%), and an increase at the femoral neck of 4.49% (with a calculated 95% confidence interval [CI] of 1.15 to 7.8%), slightly superior but largely overlapping with ours. Conversely, the overall effect size within the bisphosphonate group was significantly lower and without any BMD gain at the femoral neck.

Interestingly, in our study, the net difference in the BMD changes between the denosumab and untreated groups (12%) are very similar at the LS to those observed after 4 years in the FREEDOM extension trial (with a difference vs. placebo of roughly 12%) [20], and even greater at the femoral neck (8.95% vs. roughly 6% of the FREEDOM). In addition, in long-term non-nephropathic glucocorticoid users, a 2.2% BMD increase at the femoral neck was observed in the second year of treatment [21]. Given these premises, it seems that, apart from the chronic glucocorticoid treatment that nearly all KTRs receive, there may be specific kidney transplant-related factors that impede (or slow down) the effectiveness of RANKL inhibition when assessing its benefits in terms of BMD from baseline. However, when we take into account the natural history of this special population, marked by accelerated bone loss, the overall benefit may indeed be preserved (Fig. 2, panel a).

Finally, when denosumab was used in real-world populations, an increased risk of hypocalcaemia was observed, particularly in those with CKD [22]. However, we did not observe any concerning trend suggesting the development of this complication in our treated cohort, albeit the blood samples were taken as per the clinical schedule and not specifically to investigate the nadir of the serum calcium expected after the administration of denosumab (which is expected one to two weeks after its administration) [23].

The present data corroborate the effectiveness of denosumab in increasing BMD at all sites, without significant signals for safety concerns, especially in terms of the risk of hypocalcaemia. The use of denosumab in this patient population is supported by the fact that, unlike bisphosphonates, this molecule does not interfere with kidney function [24], and it has been associated with greater BMD increases also in cases of corticosteroid-induced osteoporosis [25].

Differently from bisphosphonates, denosumab can be associated with an unwanted rebound effect in the case of unsupervised discontinuation [26], a phenomenon also described in KTRs [27], albeit such events might not be so common, even in special populations [28].

The risk of this rebound makes it advisable to initiate bisphosphonate therapy after denosumab cessation [29]. Future studies should focus on developing strategies to mitigate the risk of a rebound effect, especially in those patients who cannot receive treatment with bisphosphonates.

As expected, we observed a decrease in ALP, a commonly adopted surrogate marker of bone turnover, especially in CKD patients. Interestingly, we did not observe significant changes in the levels of serum PTH, probably due to the achievement of adequate levels of 25OHD serum levels and calcium intake. The same measure could also help explain the absence of any episode of significant hypocalcaemia, even though a statistically significant interaction effect for treatment was observed. However, the magnitude of this effect was limited and plateaued at month six. In our opinion, this remark further emphasizes the importance of correct calcium and vitamin D supplementation in these subjects [30].

This study has several limitations. The retrospective design, even after adjusting for the variables of interest included, cannot account for all confounders or uncontrollable exposures. Of note, the two samples differed significantly in baseline BMD. Although we adjusted our analysis for this potential confounder, without a randomized design, we cannot entirely exclude a confounding effect. Furthermore, the limited sample size and single-center nature restrict its generalizability and preclude the drawing of strong conclusions regarding incident fractures and adverse events.

In conclusion, our study showed the benefit of 4 years of denosumab treatment in a real-life setting of KTRs. BMD gains were significant at the lumbar and hip sites, while the untreated population experienced BMD losses at all sites, especially relevant at the total hip and neck.