Going through the quote:
Sequence homology of cytokeratins is very high,
There are many human cytokeratin genes and corresponding proteins, and they have very similar sequences of amino acids.
but when the proteins are digested into peptides by trypsin they generate many unique peptides.
The serine protease trypsin breaks the varieties of cytokeratins into many different polypeptide pieces. The fact that there are many resulting peptides could be due to the fact that the variety of original cytokeratins in fact have distinguishable differences near the cutting points that trypsin operates on, despite all the cytokeratins being from a protein family with high residue sequence homology. (Some of the unique peptides could also be due to there being multiple places in the consensus amino acid sequence of the cytokeratin family at which trypsin cuts the proteins, but the paper does not delve into that detail.)
In the amyloid deposits the cytokeratin peptides identified were mostly unique to one protein.
Given that there are many possible unique peptides resulting from trypsin digestion, it seems somewhat surprising that most of the peptides that show up in the amyloid deposits happened to come from an identifiable single type of cytokeratin. They are hinting that this particular cytokeratin likely has an important relationship to the process involved in causing the disease (and hence perhaps should be studied more closely than galectin-7, which has previously been thought to be involved).