Recently, I was studying the subtypes of Azheimer's disease, and learned that there are four major subtypes of it , and the categorization is mainly based on where the atrophy happens (if I understand correctly), e.g. A seminal study investigated 889 brain autopsies and identified 3 AD subtypes based on the distribution of neurofibrillary tangles (NFT),
typical AD, with balanced NFT counts in the hippocampus and association cortex;
limbic-predominant AD, with counts predominantly in hippocampus;
hippocampal-sparing AD, with counts predominantly in the association cortex
What puzzles me is that whether there is any evidence that this way of categorization is objective in the nature of the disease or it is only subjective categorization because we want to have some subtypes of the disease. For example, what if we have the counts predominantly in the association cortex and intermediately in the hippocampus? It sounds like a possible situation, and could be qualified vaguely as typical AD or hippocampal-sparing AD.
So, while subtyping is surly important to understand the disease, is there any evidence that the current subtyping of three of four subtypes can be better fine-grained? And what are the challenges that stop us from looking deeper?