The computational offloading from conventional cloud datacenter towards edge devices sprouted a new world of prospective applications in pervasive and Mobile Edge Computing (MEC) paradigm, leading to substantial gains in the form of increased availability, bandwidth with low latency. The MEC offers real-time computing and storage facility within the proximity of mobile user-access network, hence it is imperative to secure communication between end user and edge server. The existing schemes do not fulfill real time processing and efficiency requirements for using complex crypto-primitives. To this end, we propose a novel two-factor biometric authentication protocol for MEC enabling efficient and secure combination of Physically Unclonable Functions (PUFs) with user-oriented biometrics employing fuzzy extractor-based procedures. The performance analysis depicts that our scheme offers resistance to known attacks using lightweight operations and supports 30% more security features than comparative studies. Our scheme is provably secure under Real-or-Random (ROR) formal security analysis model.