Classic Epidemic Routing in Delay-Tolerant Networks (DTN) delivers high message delivery rates; however, it incurs high overhead and buffer overflow, making its effectiveness compromised in dynamic scenarios like disaster relief, remote sensing, and mobile ad hoc networks. In this paper, we introduce the Epidemic-PF Router, a hybrid protocol that efficiently improves Epidemic Routing by combining Probabilistic Forwarding and adaptive buffer management. Dynamically tuning forwarding probabilities based on network density, message hop count, and buffer usage, this approach optimizes message spreading while substantially reducing congestion. In simulations performed using the ONE Simulator, we show that the Epidemic-PF protocol substantially reduces overhead, improves resource efficiency, and maintains high delivery reliability, closely approximating classic Epidemic Routing while only experiencing a small increase in latency. The protocol encourages a more controlled and scalable message spreading strategy, making it particularly well-suited to a wide range of DTN applications.