By the end of the incubation period, the control (cells only) should be about 70-80%, so you do need to optimize your cell numbers. It should not be confluent because it might affect your result as treated cells might be able to recover and you would not be able to differentiate confluent due to no effect or cells recover. Unless you do some observation using x-celligence. Usually after the cells attached then you add your drug. Some drugs might affect attachment process if you add too early. I usually add 4-6 hours after the cells settled down. If you add early, the cells might not attach and might give you inaccurate result. Some researchers add after 24 hours before adding drug. This is just my opinion. Wait for the expert to give their comments.

Hello Heba,

It depends on the doubling time of your cells, the cytotoxicity of your treatment and most of all, it depends on your question.

For example, if you use MTT assay to measure the cytoxicity of your treatment, the signal must not be saturated. So you have to be sure that the control cells will not be over confluent at the end of your experiment. Same for immunofluorescence.

What is your applcation?

Hello Heba,

As mentioned by previous respondents, different cell lines will reach confluency in different times depending on their cell-cycle time. You need to optimise your seeding rates to achieve the desired level of confluency within a convenient period, normally 24 to 48 hours. Remember that the cytotoxicity of your drug may be profoundly influenced by the cell density.  Generally the greater the confluency the lower the toxicity. This may be either because the dose per cell is decreased when there are more cells/well, and/or because cells may cooperate in detoxification of the drug via inter-cellular connections. So you must optimize cell numbers and then standardize if you wish to compare data between experiments. You could titrate a fixed concentration of drug against different levels of confluency to determine the optimum cell number.

I recently started cytotoxic assay on some tumor cells and what I noticed is that the confluence of seeding prior to drug addition greatly influences the result outcome. Like Peter Jenkinson suggested, a good way to have a consistent and unbiased result is to perform cytotoxicity of a fix drug concentration on difference cell confluence to obtain the best match. One can then proceed to use this best confluence for subsequent experiments.