Even though, proximal tibia is a common site of giant cell tumor and bone fractures, following tumor removal, nonetheless very little attention has been paid to affecting factors on the fracture risk. Here, nonlinear voxel-based finite element models based on computed tomography images were developed to predict bone fracture load with defects with different sizes, which were located in the medial, lateral, anterior, and posterior region of the proximal tibia. Critical defect size was identified using One-sample t-test to assess if the mean difference between the bone strength for a defect size was significantly different from the intact bone strength. Then, the defects larger than critical size were reconstructed with cement and the mechanics of the bone-cement interface (BCI) was investigated to find the regions prone to separation at BCI. A significant increase in fracture risk was observed for the defects larger than 20 mm, which were located in the medial, lateral and anterior regions, and defects larger than 25 mm for those located in the posterior region of the proximal tibia. Furthermore, it was found that the highest and lowest fracture risks were associated with defects located in the medial and posterior regions, respectively, highlighting the importance of selecting the initial location of a cortical window for tumor removal by the surgeon. The results of the BCI analysis showed that the location and size of the cement had a direct impact on the extent of damage and its distribution. Identification of critical regions susceptible to separation at BCI, can provide critical comments to surgeons in selecting the optimal cement augmentation technique, which may ultimately prevent unnecessary surgical intervention, such as using screws and pins.