If you want to try to lower your Cts, you can also try to 

1) Increase RNA input for cDNA synthesis

2) Conduct an annealing temperature gradient. You may get 1-2 Cts but make sure you do not compromise efficiency. 

You may be able to report changes, it all depends on the variabilities you will encounter with late Cts and whether you can consistanly replicate a change when detected. Have a look at your technical triplicates, they will give you an idea whether the late Cts are effecting your results. If the triplicates are not tight, you will not get a measure of central tendency reliably. 

Dear Farzane

i will provide you with specific details on Monday when I return to work but in essence you can look at the difference in relative expression between treated and control samples using a method called the Livak method or delta delta Ct method. I can provide the principles and worked example of that method on Monday but in the absence of pertinent documentation - and before I Hoover the house !! - I will say this much for now:

1. You need to make sure that for your treated and untreated sample groups you have amplified a housekeeping gene in addition to your gene of interest from the same cDNA in each case

Here is a link to that method

http://www.gene-quantification.net/livak-2001.pdf

2. In essence you normalise the mean Ct for your treated and untreated samples by subtracting the ct of your housekeeping gene. You then take the 2 normalised values away from one another and then take that value and use as an exponent x to the integer 2 to yield the actual relative expression change; I.e if the expression of a particular gene in your untreated sample is 1 then your relative change in expression in your treated samples is 2 to the power of x; where x is derived from the normalised ct values as described 

3. I realise this is a very superficial description but I can provide an actual worked example Monday.  The link to the Livak paper will also fill in the detail

Thus: ( mean Ct GOI treated samples N=/>3 - mean Ct treated housekeeper samples N=/>3 ) - ( mean Ct GOI untreated sample N=/3 - mean Ct untreated housekeeper N =/3) = X

Actual relative expression of GOI in treated samples relative to control (=1) = 2 ( ^X) 

I will end for now by saying that ideally it is better not just to use one housekeeper to normalise your treated and untreated samples.  Rather to select the most stably expressed 2-3 housekeepers from about 5-6 housekeepers.  You can use a program like RefFinder to select the 3 most stable housekeepers 

http://fulxie.0fees.us/?type=reference

You then normalise your data using the geometric mean of the Cts for these 2-3 housekeepers in the above Livak or delta delta Ct anakysis

Once again I can provide full working details on Monday 

If you want to try to lower your Cts, you can also try to 

1) Increase RNA input for cDNA synthesis

2) Conduct an annealing temperature gradient. You may get 1-2 Cts but make sure you do not compromise efficiency. 

You may be able to report changes, it all depends on the variabilities you will encounter with late Cts and whether you can consistanly replicate a change when detected. Have a look at your technical triplicates, they will give you an idea whether the late Cts are effecting your results. If the triplicates are not tight, you will not get a measure of central tendency reliably. 

Dear Farzanne

this is an answer I provided recently to an identical question to your own; More specifically, apart from an overview of methodology I provide actual worked examples of the Livak method for enumerating relative expression

Hope this helps (in addition to above)

https://www.researchgate.net/post/How_can_I_Calculate_expression_fold_with_CT_values#view=578ca0a0ed99e13bab058f94