Multi-access Edge Computing (MEC), an enhancement of 5G, processes data closer to its generation point, reducing latency and network load. However, the distributed and edge-based nature of 5G-MEC presents privacy and security challenges, including data exposure risks. Ensuring efficient manipulation and security of sensitive data at the edge is crucial. To address these challenges, we investigate the usage of threshold secret sharing in 5G-MEC storage, an approach that enhances both security and dependability. A (k,n) threshold secret sharing scheme splits and stores sensitive data among n nodes, requiring at least k nodes for reconstruction. The solution ensures confidentiality by protecting data against fewer than k colluding nodes and enhances availability by tolerating up to n-k failing nodes. This approach mitigates threats such as unauthorized access and node failures, whether accidental or intentional. We further discuss a method for selecting the convenient MEHs to store the shares, considering the MEHs' trustworthiness level as a main criterion. Although we define our proposal in the context of secret-shared data storage, it can be seen as an independent, standalone selection process for 5G-MEC trustworthy node selection in other scenarios too.