In addition to comment above, serum starving cells also allows for synchronisation of cell cycle. Under cell culture conditions we expose cells to non-physiological levels of serum proteins, primarily to support growth. By serum starving cells, you achieve G0 in a appreciably large portion of your flask. This is turns helps reduce variability in subsequent assays and tests.
Many thanks for your kind answer!! I forgot to include in my question that serum starve is made (by many authors) for several hours prior to any treatment. We want to analyse the effect of a drug on a granulosa cell culture. In this case, it is absolutely necessary to starve serum before addition of the drug to the culture? From your answer, I understand that it could be a way to minimise effects that should not be due to the drug, but to factors present in serum. But it would be incorrect to measure the effects of the drug without previous serum starving, even if you made time-matched paired controls in the presence of the drug solvent?
We serum starve cells before inducing inflammation. The reason for that is that you want the cells to respond to the stimulant and the response not to be affected by media ingredients, e.g. TGF-b, GFs, etc.
Take in consideration that starvation acts on many pathways inside the cell. For example AKT pathway ecc. So if starve or not cells it depends from what effect you want to analyze.