We conducted a retrospective analysis to compare
We conducted a retrospective analysis to compare a BMD-alone strategy, versus BMD+FRAX® and BMD+FRAX®+TBS screening strategies. To our knowledge, this constitutes the first study to assess the relative contributions of FRAX® and TBS. In our cohort, FRAX®+BMD identified 9 (9%) additional patients at increased risk for fractures at baseline and 2 (2%) after AI treatment compared to BMD alone, thus corroborating the current National Comprehensive Cancer Network and National Osteoporosis Foundation guidelines to conduct FRAX® assessment in osteopenic women who require AI . Likewise, the addition of TBS increased the numbers of at-risk patients detected both at baseline (an additional 16, 16%) and at follow-up (an additional 6, 6%) compared to BMD alone. Consistent with other reports [15,16], the effect of AI on TBS did not appear to correlate with changes in BMD, suggesting that these two risk assessments may be complementary rather than overlapping, and could potentially be combined to create a more sensitive measure of bone fragility. Our conclusions are corroborated by two small previous studies evaluating the influence of AI on TBS. In one study, 34 breast cancer patients treated with AI were evaluated with both BMD and TBS, and sustained a decrease of 5.9% and 2.1%, respectively, across a mean DXA interval of 2.1 years (corresponding to a rate of loss of 2.8% and 1% annually) . Another study evaluated the effect of exemestane on TBS in 19 patients, and showed a TBS decrease of 2.3% at 24 months (corresponding to a mean annual decrease of 1.2%) . These findings are generally consistent with our study, which demonstrated mean annual decreases of 1.5% for BMD and 0.5% for TBS. In our study, the decline in TBS during AIs treatment was not statistically significant. Changes in TBS after osteoporosis therapy are known to be of lesser magnitude than corresponding changes in BMD; conversely, it is possible that BMD losses in the setting of AI therapy occur in similar fashion . Additionally, longer follow up might have demonstrated the decline in TBS associated with relatively longer use of AIs (mean 2.1 years) that others have demonstrated . Furthermore, patients in our study with follow up DXA after >18 months of AI treatment were more likely to experience EFR than <18 months of follow up. Thus, serial TBS measurements appear to identify subjects with newly-acquired fracture risk following diindolylmethane of AI therapy with greater sensitivity following 18 months on AI. Future studies assessing TBS after longer follow up are needed to further evaluate AI-associated changes in trabecular microstructure over five to ten years. In summary, our analysis supports further evaluation of a screening approach of BMD+FRAX®+TBS, obtained at baseline and at 2 years following AI, with special attention paid to patients with osteopenia and non-degraded bone microarchitecture by TBS. An alternative approach could be the use of TBS-adjusted FRAX BMD as it has been recently proposed . This later approach would still have to be tested in this very specific population. Recently published European consensus guidelines suggest that the use of bisphosphonates should be considered for the prevention of cancer treatment-induced bone loss in all patients with a T-score of <−2.0 . According to this approach, 7 additional patients would have been identified at baseline in our population based on DXA alone, of these 2 would have been identified by the combined approach of osteopenia + low TBS score, 2 by osteopenia + high FRAX® score, and 1 by both osteopenia + low TBS score and osteopenia + high FRAX®. Only 28% of patients identified by the combined approach osteopenia + low TBS score would have been identified by lowering the T-score threshold. This result suggests that modification of T-score threshold would not hamper the usefulness of TBS. Our findings should be interpreted in the context of recently reported phase III studies, evaluating the potential utility of a universal BMA treatment approach for all post-menopausal women receiving AI. Recent large randomized studies in adjuvant AI treated breast cancer patients have demonstrated, respectively: an improvement in BMD with risedronate, as well as both an improvement in BMD plus a reduction in fragility fractures with denosumab . (Gnant et al., n.d.) These findings, while clearly demonstrating the clinical utility of BMAs, do not obviate the importance of finding better strategies to identify patients at risk of fragility fractures. An approach of universally treating all patients starting therapy with an AI with a BMA has not been adopted. Indeed, the majority of women will not experience a fragility fracture, and therefore a universal treatment approach would lead to overtreatment with BMAs, which may be costly and associated with toxicities. When considering extended treatment with BMAs, rare adverse effects such as atypical femur fractures and osteonecrosis of the jaw, likely related to duration of use, should be taken in consideration . An effective screening approach may serve to better identify patients who may benefit from BMAs.