In your framework you have to read in the sheet and write it back out. Suppose you have an XLSM file (with macros, I presume) called "testXLSM2X.xlsm" containing one sheet with tabular columns of data. This will do the trick:
library(xlsx)
r <- read.xlsx("testXLSMtoX.xlsm", 1) # read the first sheet
# provides a data frame
# use the first column in the spreadsheet to create row names then delete that column from the data frame
# otherwise you will get an extra column of row index numbers in the first column
r2w<-data.frame(r[-1],row.names=r[,1])
w <- write.xlsx(r2w,"testXLSMtoX.xlsx") # write the sheet
The macros will be stripped out, of course.
That's an answer but I would question what you are trying to accomplish. In general it is easier to control R from Excel than Excel from R. I use REXCEL from http://rcom.univie.ac.at/, which is not open source but pretty robust.