If you just want to strip all occurrences of a particular character you could just use 'tr'
However if the character occurs elsewhere and you only want to replace it in this particular context, you could try modifying this standard sed one-liner:
so that it joins lines when the following line does NOT begin with a digit, e.g.
# if a line begins with an equal sign, append it to the previous line # and replace the "=" with a single space
sed -e :a -e '$!N;s/\n=/ /;ta' -e 'P;D'
If the separating character really is something other than newline (\n) you should be able to specify it via its ASCII hex value e.g. for \n you could also write (substitute your \x value as appropriate)
sed -e :a -e '$!N;s/\n\([^0-9]\)/\1/;ta' -e 'P;D'
sed -e :a -e '$!N;s/\xA\([^0-9]\)/\1/;ta' -e 'P;D'