Background and purpose: Patients with advanced cancer may develop painful bone metastases, potentially resulting in pathological fractures. Adequate fracture risk assessment is of key importance to prevent fracturing and maintain mobility. This study aims to validate the clinical reliability of axial cortical involvement with a 30 mm threshold on conventional radiographs to assess fracture risk in femoral bone metastases. Materials and methods: All patients with bone metastases who received radiotherapy for pain included in two multicentre prospective studies were selected. Conventional radiographs obtained at a maximum of two months prior to radiotherapy were collected. Three experts independently measured lesions and scored radiographic characteristics. Sensitivity, specificity, positive (PPV) and negative predictive value (NPV) were calculated. Results: Hundred patients were included with a median follow-up of 23.0 months (95%CI: 10.6–35.5). Two fractures occurred in lesions with axial cortical involvement <30 mm, and 12 in lesions ≥30 mm. Sensitivity, specificity, PPV and NPV of axial cortical involvement for predicting femoral fractures were 86%, 50%, 20% and 96%, respectively. Patients with lesions ≥30 mm had a 5.3 times higher fracture risk than patients with smaller lesions. Conclusion: Our validation study confirmed the use of 30 mm axial cortical involvement to assess fracture risk in femoral bone metastases. Until a more accurate and practically feasible method has been developed, this clinical parameter remains an easy method to assess femoral fracture risk to aid patients and clinicians to choose the optimal individual treatment modality.