Vandetanib is an important treatment option for advanced metastatic medullary thyroid cancer. The aims of this study were to evaluate the predictors of both a longer response to vandetanib and the outcome. Medical records of 79 medullary thyroid cancer patients treated with vandetanib at our center were analysed. Twenty-five patients were treated for <12 months, 54 were treated for ≥12 months and 24 of these latter were treated for ≥48 months (short-, long- and very long-term). The median progression free survival of the long and very long-term treated patients was significantly longer than in the ZETA trial. When comparing the groups of short - and long-term treated patients the only significant difference was that these latter were less frequently previously treated with a tyrosine kinase inhibitor. However, the long-term treated patients had a younger age, both at diagnosis and enrolment, which was statistically significant in the very long-term treated patients. In the long-term treated group, younger age, enrolment for symptoms and development of adverse events were significantly correlated with a better outcome. The enrolment for symptoms remained the only statistically significant predictor of a good outcome in the very long-term treated patients. In conclusion, early treatment with vandetanib, when patients are younger, with a good ECOG performance status and symptomatic disease, not necessarily progressing for RECIST, seem to be the best predictors of a longer and durable response. Further studies are needed to confirm these results