Hi Sam,
The simple answer is no. Most mass spectrometry based quantitation experiments are done label free. The use of the stable isotope labeled internal standards enables normalization to minimize the measured variance. If your goal is to demonstrate that the measured peptide signal is "differential" between groups then there is no need to assess the LLOQ or even demonstrate that the measurement is linearly related to the peptide quantity.

However, if the goal is to say that the quantity of the peptide has changed by a specific amount then it will be important to demonstrate that the response is linear within the range of measurements you are making. Most people are just interested in whether a measurement is differential and the exact amount of the fold change (i.e. is it 1.5x or 2x) is less important.

If knowing the accurate fold change is important then a useful experiment is to take the sample with the largest abundance and the sample with the smallest and to mix them in know ratios. You can mix them 100:0, 75:25, 50:50, 25:75, 0:100 and if the response between those five measurements is linear then your fold change measurements between your least and most abundant samples is probably linear.

I always like to think of the stable isotope labeled peptide just as a way to normalize the signal. If you are comparing sample A to sample B then the internal standard H amount will cancel out ... [A/H]/[B/H] = A/B. We use internal standards because A/H and B/H can be measured more precisely than A and B.

I hope this helps,
Mike