RegEx

Instead of .* (greedy), you should almost always use .*? (lazy) to reduce the amount of backtracking needed.
If you want to escape a string for use as a literal pattern, just wrap it inside \Q and \E. In Java, you can call Pattern.quote(...).
In most languages, look-behinds and look-aheads must have an obvious length, i.e. + and * are off-limits.
Most metacharacters (e.g. . and *) lose their special meaning when inside a character class. Thus you don't have to escape them anymore. The only characters you have to escape there are [, ], - and ^. Even better, you don't have to escape - when at the very beginning or end of the character class, and ^ must only be escaped when it comes first in the class.

Lookarounds

Lookarounds are atomic groups, i.e. they are non-capturing.
Wrapped in parentheses and starting with a question mark, lookarounds have the basic form (?...) that is not complete yet.
Lookaheads add a = if they "want" a certain group to follow (positive) or a ! if they "don't want" a certain group to follow.
Lookbehinds work like lookaheads, but they add an additional < after the question mark and let you check what precedes, not follows.
The positive lookahead a(?=b) does only match a characters that are followed by a b.
The negative lookahead a(?!b) matches all a characters that are not followed by a b.
The positive lookbehind (?<=b)a does only match a characters that are preceded by a b.
The negative lookbehind (?<!b)a matches all a characters that are not preceded by a b.

If you want to match all characters including newlines, you have to use the negative empty selection ([^]) instead of the dot (.).
JavaScript doesn't know lookbehinds, neither positive (?<=) nor negative (?<!).