Nonlinear control strategies are suitable for processes presenting important nonlinearities. In the present polymerization process, a wide domain of monomer conversion is reachable and the gel effect is considered, imposing large variations in the model. A precise dynamic model is thus required. In this study, an adaptive control for a multi-input multi-output linearizable system, together with an extended Kalman filter, is applied to a simulated continuous polymerization reactor to follow two output set-points (monomer conversion and reactor temperature) in the presence of model parameter uncertainties. Simulation results show that this technique is robust and the output performance can be ensured even in the presence of large model parameter errors or variations and disturbances.