no need for rexeg ...
given your inputs
df[df['col1'].apply(lambda x: any(x.startswith(prefix) for prefix in prefixes))] col1 0 12345 2 54467899