Discussion
Our study included a total of 56 patients and constitutes the largest number of cases collected in a single series. Previous studies presented a smaller number of cases. González-Rodríguez et al (7) collected 60 spontaneous vertebral fractures in 15 women with breast cancer who were undergoing treatment with aromatase inhibitors and in whom denosumab was discontinued. Fernández Fernández et al (14) described 49 vertebral fractures in 10 women and Florez et al (15) published a series of 7 women who had a median of 5 vertebral fractures. Another study collected the first 3 cases of hip fracture produced after abrupt DMAB discontinuation in the absence of other causes (10). Several systematic reviews have confirmed the magnitude of the problem (6,16-18). In this series, we publish the first two cases described in men.
The actual number of cases is probably much higher. The Spanish Agency for Medicines and Health Products (AEMPS), which collects adverse effects of drugs, described in 2019 a total of 64 patients with multiple vertebral fractures that were increased in a subsequent statement in 2020, 213 patients with multiple vertebral fractures and 50 hip fractures. There are several reasons that might explain why the magnitude of the problem is not preceived. We would mention: a) it is a complication not yet sufficiently known by the medical community in general, b) they are fractures that occur in patients who have osteoporosis, therefore, they can be attributed to the disease rather than to the suspension of the drug, c) given that the drug is administered every 6 months, it is possible to forget it, especially when the questioning is directed at drugs that are taken orally, and d) for scientific journals, the publication of new cases do not provide anything noteworthy. So, in recent years, the number of publications on the matter has decreased, while the number of fractures has not.
The mean age of our series was 68.1 years, somewhat older than those described in other series, such as that of Barcelona, where the median age was 65 years (15), and that of Madrid with a mean of 66.4 years (14 ). In the González-Rodríguez series (7), the mean age was lower, 62.3 years, but they were other types of patients, women with breast cancer and not postmenopausal osteoporosis. In a systematic review in which 24 cases were collected, the mean age was 64.1 years (6).
Our patients had received a median of 6 doses, with DMAB having been used a median of 30.5 months. These results coincide with those published in other series and reports of individual cases (6–8,10,14–16,19–23). In a “real world” study, the risk of fracture when discontinuing DMAB treatment has been calculated to increase markedly when the third injection is given (16). The time it takes for fractures to occur after the last dose of DMAB showed a median of 11 months in our study, which represents a 5-month delay, since the drug is administered every six months, although in one case it occurred after the delay of a month and a half. In different reported cases, this period ranges from 2 to 13 months (6,8,14,15).
Probably the appearance of fractures will depend on two factors, the severity of the disease and the withdrawal of the drug. The severity of the disease could be determined through the FRAX or by the presence of previous fractures. The 10-year risk of fracture calculated by the FRAX tool showed a median of 11% for major fracture and 3.9% for hip fracture. Although there is a debate on the optimal threshold to perform a therapeutic intervention (24–26), the high risk of fracture has been established at 20% for the major fracture and 3% for the hip fracture (27). In our study, the fracture risk at 10 years showed a median of 11% for the major fracture and 3.9% for the hip fracture. FRAX has rarely been estimated in the publications of other cases.
The other factor involved is the discontinuation of the drug. One of the reasons DMAB was discontinued came about after reported improvement in treatment with BMD, leading to the misconception that osteoporosis was cured. Following this line, the idea of the “treat to target” was developed according to which, when reaching a certain T-score value, the drug could be suspended, without verifying the results of this suspension (28–30).This led to the discontinuation of DMAB due to medical recommendation in 41.1% of cases. Closely related to this idea is the concept of therapeutic holidays wrongly applied to DMAB (31,32). On the other hand, given that the association between the use of denosumab and osteonecrosis of the jaws has been described (33–35), the suspension of denosumab was carried out by the dentist’s indication in 21.5% of the patients. Our results coincide with those reported in other series (7,8,22).
The deleterious effect of DMAB suppression is determined by the sudden increase in remodeling that can lead to a deterioration in bone strength and facilitate the appearance of fractures. This fact had been previously described, although an increase in fractures had not been observed. After discontinuing DMAB, beta-crosslaps increase significantly, from a median of 0.071 ng/mL to 0.520 ng/mL (p <0.001). To a lesser extent, but also significantly, the markers of bone formation increase, the P1NP that goes from 25.3 ng/mL to 101.2 ng/mL, p = 0.006 and osteocalcin from 10.7 ng/mL to 28.1 ng/mL. This indicates an increase in all bone remodeling in which osteoclastic activity clearly predominates, as has also been described in other series (20,36,37). We have not observed changes in serum levels of creatinine, calcium, phosphorus, total protein, vitamin D, measured as 25-hydroxyvitamin D, or in PTH.
Finally, we carried out a logistic regression analysis to try to identify which factors could be associated with the presence of a greater number of fractures, obtaining a statistically significant association with the time in which denosumab was previously used (p= 0.04).
Among the limitations of our study is the sample size, which is due to the difficulty in identifying these patients. On the other hand, since there is no control group, we have not been able to establish what the clinical, analytical or densitometric factors could be associated with the appearance of fractures. The strength of the study is determined by the high number of fractures associated with a full number of complementary tests.
To sum up, we present a series of 56 patients in which the abrupt discontinuation of DMAB caused a total of 192 vertebral fractures, the increase in bone removal probably being manifested through a considerable increase in biochemical markers of bone remodeling, especially those of resorption, which causes this effect.