Individualized fracture risk assessment: progresses and challenges

Fracture risk prediction in old Chinese people-a narrative review

With aging, the burden of osteoporotic fracture (OF) increases substantially, while China is expected to carry the greatest part in the future. The risk of fracture varies greatly across racial groups and geographic regions, and systematically organized evidence on the potential predictors for fracture risk is needed for Chinese. This review briefly introduces the epidemiology of OF and expands on the predictors and predictive tools for the risk of OF, as well as the challenges for their potential translation in the old Chinese population. There are regional differences of fracture incidence among China. The fracture incidences in Hong Kong and Taiwan have decreased in recent years, while it is still increasing in mainland China. Although the application of dual-energy X-ray absorptiometry (DXA) is limited among old Chinese in the mainland, bone mineral density (BMD) by DXA has a predictive value similar to that worldwide. Other non-DXA modalities, especially heel QUS, are helpful in assessing bone health. The fracture risk assessment tool (FRAX) has a good discrimination ability for OFs, especially the FRAX with BMD. And some clinical factors have added value to FRAX, which has been verified in old Chinese. In addition, although the application of the osteoporosis self-assessment tool for Asians (OSTA) in Chinese needs further validation, it may help identify high-risk populations in areas with limited resources. Moreover, the translation use of the muscle quality and genetic or serum biomarkers in fracture prediction needs further works. More applicable and targeted fracture risk predictors and tools are still needed for the old Chinese population.

Can machine learning predict pharmacotherapy outcomes? An application study in osteoporosis

BACKGROUND AND OBJECTIVE

The specific aim of this study is to develop machine learning models as a clinical approach for personalized treatment of osteoporosis. The model performance on outcome prediction was compared between four machine learning algorithms.

METHODS

Retrospective, electronic clinical data for patients with suspected or confirmed osteoporosis treated at Wan Fang Hospital between 2011 to 2018 were used as inputs for building the following predictive machine learning models,i.e., artificial neural network (ANN), random forest (RF), support vector machine (SVM) and logistic regression (LR) models. The predicted outcome was defined as an increase/decrease in T-score after treatment. A genetic algorithm was employed to select relevant variables as input features for each model; the leave-one-out method was applied for model building and internal validation. The model with best performance was selected by a separate set of testing. Area under the receiver operating characteristic curve, accuracy, precision, sensitivity and F1 score were calculated to evaluate model performance. Main analysis for all the patients with subclinical or confirmed osteoporosis and subgroup analysis for the patients with confirmed osteoporosis (T score < -2.5) were carried out in this study.

RESULTS

A genetic algorithm was employed to select 12 to 18 features from all 33 variables for the four models. No difference was found in accuracy (ANN, 71.7%; LR, 70.0%; RF, 75.0%; SVM, 66.7%), precision (ANN, 80.0%; LR, 59.3%; RF, 70.0%; SVM, 63.6%), and AUC (ANN, 0.709; LR, 0.731; RF, 0.719; SVM, 0.702) among the ANN, LR, RF and SVM models. Main analysis in performance revealed significant recall in the LR model, as compared to ANN and SVM model; while subgroup revealed significant recall in ANN model, compared to LR and SVM model.

CONCLUSIONS

Machine learning-based models hold potential in forecasting the outcomes of treatment for osteoporosis via early initiation of first-line therapy for patients with subclinical disease; or a switch to second-line treatment for patients with a high risk of impending treatment failure. This convenient approach can assist clinicians in adjusting treatment tailored to individual patient for prevention of disease progression or ineffective therapy.

Osteoporosis and fractures in rheumatoid arthritis - risk factors

In this chapter, we emphasize among rheumatoid arthritis (RA) patients, whom and how to screen for osteoporosis. We highlight certain modalities, advancements in technology, secondary osteoporosis workup, and laboratory testing as well as their caveats. Finally, we discuss current guidance on how to direct the laboratory and radiology testing in the context of the individual patient with RA to guide and select from the osteoporosis treatment options currently available.

Assessing the clinical utility of genetic profiling in fracture risk prediction: a decision curve analysis

Using decision curve analysis on 2188 women and 1324 men, we found that an osteogenomic profile constructed from 62 genetic variants improved the clinical net benefit of fracture risk prediction over and above that of clinical risk factors and BMD.

INTRODUCTION

Genetic profiling is a promising tool for assessing fracture risk. This study sought to use the decision curve analysis (DCA), a novel approach to determine the impact of genetic profiling on fracture risk prediction.

METHODS

The study involved 2188 women and 1324 men, aged 60 years and above, who were followed for up to 23 years. Bone mineral density (BMD) and clinical risk factors were obtained at baseline. The incidence of fracture and mortality were recorded. A weighted individual genetic risk score (GRS) was constructed from 62 BMD-associated genetic variants. Four models were considered: CRF (clinical risk factors); CRF + GRS; Garvan model (GFRC) including CRF and femoral neck BMD; and GFRC + GRS. The DCA was used to evaluate the clinical net benefit of predictive models at a range of clinically reasonable risk thresholds.

RESULTS

In both women and men, the full model GFRC + GRS achieved the highest net benefits. For 10-year risk threshold > 18% for women and > 15% for men, the GRS provided net benefit above those of the CRF models. At 20% risk threshold, adding the GRS could help to avoid 1 additional treatment per 81 women or 1 per 24 men compared with the Garvan model. At lower risk thresholds, there was no significant difference between the four models.

CONCLUSIONS

The addition of genetic profiling into the clinical risk factors can improve the net clinical benefit at higher risk thresholds of fracture. Although the contribution of genetic profiling was modest in the presence of BMD + CRF, it appeared to be able to replace BMD for fracture prediction.

Iron Overload Associated Endocrine Dysfunction Leading to Lower Bone Mineral Density in Thalassemia Major

CONTEXT

Patients with thalassemia major (TM) have a lower bone mineral density (BMD) and higher risk of fracture than the general population. The possible mechanisms include anemia, iron overload, malnutrition, and hormonal deficiency, but these have not been thoroughly investigated.

OBJECTIVE

To identify major mineral and hormonal factors related to BMD in adult TM patients to provide human evidence for the proposed mechanisms.

DESIGN

Retrospective study.

SETTING

Referral center.

PATIENTS

Twenty-nine patients with β-TM, aged 23 to 44 years who were followed-up during 2017 to 2018 were enrolled.

OUTCOME MEASUREMENTS

Endocrine profiles, including thyroid, parathyroid, and pituitary function, glucose, vitamin D, calcium, phosphate, and fibroblast growth factor 23 (FGF23) were obtained. The relationships among the above parameters, body height, fractures, and BMD were analyzed.

RESULTS

Abnormal BMD was observed in 42.9% of women and 23.1% of men. The mean final heights of women and men were 3.7 cm and 7.3 cm lower than the mean expected values, respectively. Fracture history was recorded in 26.7% of women and 35.7% of men. BMD was negatively correlated with parathyroid hormone, FGF23, thyrotropin, and glycated hemoglobin (HbA1c) levels, and positively correlated with testosterone, IGF-1, and corticotropin levels (all P < .05). Moreover, hypothyroidism was associated with lower BMD in both the lumbar spine (P = .024) and the femoral neck (P = .004). Patients with hypothyroidism had a higher percentage of abnormal BMD (P = .016).

CONCLUSION

Hypothyroidism, higher HbA1c, and lower adrenocorticotropin were predictors of abnormal BMD in patients with β-TM. Whether the correction of these factors improves BMD warrants further research.

Age-related periosteal expansion at femoral neck among elderly women may maintain bending stiffness, but not femoral strength

Periosteal expansion and bone loss have opposite effects on femur strength. Their combined effect has not been fully understood. Our investigation using a recently developed beam model suggested that periosteal expansion may maintain femur bending stiffness among elderly women, but not help preserve femoral strength and reduce hip fracture risk.

INTRODUCTION

Periosteal expansion and bone loss are two accompanying biological phenomena in old population. Their combined effect on bone stiffness, strength, and fracture risk is still not clear, because previous studies have reported contradictory results.

METHODS

A recently developed DXA (dual-energy X-ray absorptiometry)-based beam model was applied to study the effect at the femoral neck. We first made a theoretical analysis. Then, a clinical cohort consisting of 961 women (316 hip fractures and 645 controls, age of 75.9 ± 7.1) was used to investigate the associations quantitatively. We investigated (1) correlations of femoral-neck width and bone mineral density with femoral stiffness and strength; (2) correlations of femoral stiffness, strength, and hip fracture risk index with age; (3) associations of femoral stiffness, strength and fracture risk index with actual fracture status, measured by the area under the curve (AUC) and odds ratio (OR).

RESULTS

The investigation results showed that (i) femoral-neck width had stronger correlation with femoral bending stiffness (r = 0.61-0.82, p < 0.001) than with the other stiffness components, while bone mineral density had stronger correlation with axial/shearing stiffness (r = 0.84-0.97, p < 0.001), strength (r = 0.85-0.92, p < 0.001), and fracture risk index (r = -0.61-0.62, p < 0.001) than with bending stiffness. (ii) The association between femoral bending stiffness and age was insignificant (r = - 0.06-0.05, r > 0.05); The associations of axial/shearing stiffness (r = - 0.27--0.20, p < 0.001), strength (r = - 0.28, p < 0.001), and fracture risk index (r = 0.38, p < 0.001) with age were significant. (iii) Fracture risk index had the strongest association with actual fracture status (AUC = 0.75, OR = 3.19), followed by strength (AUC = 0.74, OR = 2.84) and axial/shearing stiffness (AUC = 0.56-0.65, OR = 2.39-2.49). Femoral bending stiffness had the weakest association (AUC = 0.48-0.69, OR = 1.42-2.09).

CONCLUSION

We concluded that periosteal expansion may be adequate to maintain femoral bending stiffness among elderly women, but it may not help preserve strength and reduce hip fracture risk.

Overview of Fracture Prediction Tools

The characterization of risk factors for fracture that contribute significantly to fracture risk, over and above that provided by the bone mineral density, has stimulated the development of risk assessment tools. The more adequately evaluated tools, all available online, include the FRAX® tool, the Garvan fracture risk calculator and, in the United Kingdom only, QFracture®. Differences in the input variables, output, and model construct give rise to marked differences in the computed risks from each calculator. Reasons for the differences include the derivation of fracture probability (FRAX) rather than incidence (Garvan and QFracture), limited calibration (Garvan), and inappropriate source information (QFracture). These differences need to be taken into account in the evaluation of assessment guidelines.

Fragility fracture: recent developments in risk assessment

More than half of older women who sustain a fragility fracture do not have osteoporosis by World Health Organization (WHO) bone mineral density (BMD) criteria; and, while BMD has been used to assess fracture risk for over 30 years, a range of other skeletal and nonskeletal clinical risk factors (CRFs) for fracture have been recognized. More than 30 assessment tools using CRFs have been developed, some predicting fracture risk and others low BMD alone. Recent systematic reviews have reported that many tools have not been validated against fracture incidence, and that the complexity of tools and the number of CRFs included do not ensure best performance with poor assessment of (internal or comparative) validity. Internationally, FRAX® is the most commonly recommended tool, in addition to QFracture in the UK, The Canadian Association of Radiologists and Osteoporosis Canada (CAROC) tool in Canada and Garvan in Australia. All tools estimate standard 10-year risk of major osteoporotic and 10-year risk of hip fracture: FRAX® is able to estimate fracture risk either with or without BMD, but CAROC and Garvan both require BMD and QFracture does not. The best evidence for the utility of these tools is in case finding but there may be future prospects for the use of 10-year fracture risk as a common currency with reference to the benefits of treatment, whether pharmacological or lifestyle. The use of this metric is important in supporting health economic analyses. However, further calibration studies will be needed to prove that the tools are robust and that their estimates can be used in supporting treatment decisions, independent of BMD.

Secondary prevention and estimation of fracture risk

The key questions addressed in this chapter are: • How can individual risk of fracture be best estimated? • What is the best system to prevent a further fracture? • How to implement systems for preventing further fractures? Absolute fracture risk calculators (FRCs) provide a means to estimate an individual's future fracture risk. FRCs are widely available and provide clinicians and patients a platform to discuss the need for intervention to prevent fragility fractures. Despite availability of effective osteoporosis medicines for almost two decades, most patients presenting with new fragility fractures do not receive secondary preventive care. The Fracture Liaison Service (FLS) model has been shown in a number of countries to eliminate the care gap in a clinically and cost-effective manner. Leading international and national organisations have developed comprehensive resources and/or national strategy documents to provide guidance on implementation of FLS in local, regional and national health-care systems.