The Proton Exchange Membrane Fuel Cell is a promising energy converter for various fields of application: stationary, portable and mobile. However durability avoids its widespread deployment. Deterioration mechanisms are not all fully understood and that is the reason why the prognostic of such device is gaining attention. This helps determine the present and future state of health of Fuel Cell, to deduce the remaining life in order to take corrective actions. The work presented in this paper attempts to address this issue by proposing a method based on a degradation model. An observer, based on an Extended Kalman Filter, estimates the state of health and the dynamic of the degradations. This result is extrapolated until a threshold is reached and the residual life is deduced. This method allows estimating the lifespan with a single model, robust to uncertainties, whatever the operating conditions are. Simulations are conducted to validate the method. Finally, this framework is used on a set of experimental data from long term test on a 5-cell stack operated under a constant current solicitation.