For example, consider the strategic form game:
In this case, P2 never plays $R$ and P1 never plays $B$. $R$ is dominated by $C$ and $B$ is dominated by playing $T,M$ with respective probabilities $\frac{4}{5},\frac{1}{5}$. Is it always the case that if a strategy is never the best response in a simultaneous move game, then it is dominated?
In a finite game in strategic form, a strategy of a player is never a best response if and only if it is strictly dominated; see Lemma 60.1 in Osborne and Rubinstein, A course in Game Theory, 1994.
This lemma implies that an action that is weakly dominated - but not strictly dominated - is a best response to some belief.
(The question does not make clear which notion of dominance is intended.)