Consensus clustering integrates multiple partitions into a unified solution, enhancing stability across clustering algorithms. However, ensuring global consistency remains computationally demanding. Quadratic Unconstrained Binary Optimization (QUBO) formulations provide a natural framework for this task, making consensus clustering amenable to quantum optimization methods such as quantum annealing and the Quantum Approximate Optimization Algorithm (QAOA). We extend the established QUBO formulation for correlation clustering with a refinement protocol that allows for improvement of suboptimal solutions. To address current quantum hardware limitations, we implement a subproblem decomposition strategy that iteratively resolves constraint violations on manageable subsets. Validation on benchmark and real-world datasets demonstrates competitive accuracy compared to established methods.