Enhancing road safety and offering infotainment services are of utmost importance to today’s Intelligent Transport System (ITS). Technically speaking, an ITS is based on self-organizing wireless networks, known as vehicular ad hoc networks (VANETs). In the latter, mobile vehicles can track, compile and share reports on-road travel, driving conditions, and threats, for example, road accidents, in real-time. The spread of such data in VANET would actually improve road safety and driving comfort. In this regard, designing efficient data dissemination protocols in VANET has become a pivotal issue, attracting the interest of academic researchers, networking experts, and automotive companies. This paper introduces a new adaptive geocast data dissemination protocol called Markov Chain-based data Dissemination Protocol (MCDP). MCDP relies on the forward-if-relevant principle to dynamically determine the Zone Of Relevance (ZOR) of the event. We consider the event worth enough for a vehicle whenever there is a high probability that it will meet the event location shortly. Indeed, we introduce a mathematical model to manage vehicle mobility in the road network. Besides, we design a new method for accurately assessing the event relevance to the receiving vehicle. Simulation results underscore MCDP outperforms the pioneering baseline protocols in terms of effectiveness and efficiency.