I have used the term probably hundreds of times in my thesis as a number of the models that exist in the literature and that I am comparing in my work are 'a priori' models, i.e. they make predictions entirely from scratch, using mathematical concepts and no in vivo data. Therefore I am often writing things like 'many methods for predicting drug distribution a priori exist in the literature' or 'predicting drug distribution a priori is of importance because...'. However, I agree that if it doesn't easily fit into your thesis, I wouldn't bother trying to shoehorn it in! It is a commonly used term in my field and it would be weird for me NOT to use it.