Due to the increasing demands for higher data rate applications, also due to the actual spectrum crowd situation, DSA (Dynamic Spectrum Access) turned into an active research topic. In this paper, we analyze DSA in cellular networks context, where a CAB (Coordinated Access Band) is shared between RANs (Radio Access Networks). We propose an SMDP (Semi Markov Decision Process) approach to derive the optimal DSA policies in terms of operator reward. In order to overcome the limitations induced by optimal policy implementation, we also propose a simple, though sub-optimal, DSA heuristic. The achieved reward is shown to be very close to the optimal case and thus to significantly exceed the reward obtained with FSA (Fixed Spectrum Access). Higher rewards and better spectrum utilization with DSA optimal and heuristic methods are however obtained at the price of a reduced average user throughput.