This paper presents a bilevel optimization framework that improves congestion management and economic efficiency in decentralized energy markets by coordinating hybrid Distribution System Operators (DSOs) and Virtual Power Plants (VPPs). The framework utilizes Distributed Locational Marginal Pricing (DLMP) to align technical constraints with market incentives and addresses challenges arising from high-penetration Distributed Energy Resources (DERs). Simulations on the modified KEPCO Active Distribution Test System (KADTS) show performance improvements compared to conventional methods, including profit increases of up to 3.35%, effective congestion mitigation, and enhanced operational flexibility enabled by adaptive DLMP mechanisms and strategic bidding. The framework is scalable, adaptable, and compatible with existing market structures. It offers a practical approach to modern energy systems and supports ongoing developments in smart grids and sustainable grid modernization.