Full suppression of neoclassical tearing modes (NTMs) using electron cyclotron current drive (ECCD) should be reached before mode locking (stop of rotation) makes suppression impossible. For an ITER scenario 2 plasma, the similar time scales for locking and island growth necessitate the combined modelling of the growth of the mode and its slow down due to wall induced drag. Using such a model, the maximum allowed latency between the seeding of the mode and the start of ECCD deposition and maximum deviation in the radial position are determined. The maximum allowed latency is determined for two limiting models for island growth; the polarization model with wmarg = 2 cm, representing the worst case, and the transport model with wmarg = 6 cm, representing the best case. NTMs with seed island widths up to 9.5 cm and 12 cm for the 2/1 and the 3/2 NTM, respectively, are suppressible. The maximum allowed latency is 1.05 s and 2.95 s for the 2/1 and 3/2 NTM, respectively, for the worst case model. Radial misalignment should not exceed 7–10mm for the 2/1 NTM and 5–16mm for the 3/2 NTM depending on the model for island growth. As long as the alignment suffices, it does not reduce the maximum allowed latency. Mode locking has serious implications for any real-time NTM control system on ITER that aims to suppress NTMs by ECCD.