Empathetic response generation requires perceiving and un- derstanding the user’s emotion to deliver a suitable response. However, existing models generally remain oblivious of an interlocutor’s persona, which has been shown to play a vital role in expressing appropriate empathy to different users. To address this problem, we propose a novel Transformer-based architecture that incorporates retrieval-augmented prompt learning to generate persona-aware empathetic responses. Since personalized emotional resonance is subtle and un- controllable, we employ the dense passage retrieval to fetch exemplary responses that reflect specific persona and con- text characteristics to cue the generative model on signaling empathy. Extensive experiments on PEC, a newly-released dataset with persona-based empathetic conversations, confirm the effectiveness of our model.