The Japanese legacy encoding would support English. That requires use of the K functions, longer character columns and extra length for the SAS formats. The same type of SAS code changes have to be made for UTF-8. If you ever need to support an additional language, you would have to update the code again to use UTF-8. I also recommend that route.
The CVP engine can help you pad the character columns on existing SAS data sets. Here are links to some SAS Global Forum papers that provide additional information about making the necessary modification: