Your ability to calculate 'Mutual Information' depends on whether you are using the term in the broad sense (e.g. the correlation between variables might be a way to assess information which two variables share) or if you are talking about the more specific way the term has been used. Specifically, Mutual Information is a term used to measure how similar the joint distribution p(X,Y) is to the products of a factored marginal distribution p(X)p(Y) for a pair of variables. If you are referring to this specific definition, I know of no Base SAS or SAS/STAT procedures which provide this calculation.
Note: You can specify Mutual Information as a weighting method inside of SAS Text Miner but this would not provide the information described above.
Cordially,
Doug