Staining edge effect with Akoya Phenocycler Fusion but not IHC

Hello, wondering if anyone may have suggestions to a problem we’re having with our PhenoCycler-Fusion (CODEX) from Akoya Biosciences. We’ve been running it well for several months, but recently we’ve run into issues where certain antibodies that we’ve conjugated ourselves only stain the edge of tissue. During antibody validation, we did not see the same staining patterns via IHC (on the same tissue, serial sections).

We’ve tried a variety of things to solve this issue including changing out all buffers, trying multiple variations of antigen retrieval, staining and incubation, etc., but it hasn’t helped much. The most impactful change was to completely forgo using parafilm during staining, but it doesn’t seem to have resolved this affect for all Abs. I’m concerned it’s likely the tissue (it’s large pancreatic biopsies which were likely overfixed on the outside but underfixed on the inside). Any suggestions as to what to try next or how to “prove” it’s a tissue problem?

For example, p16:

Here is another example, PRSS2:

Meanwhile, many other markers look great (DAPI (blue), Vimentin (green), pancreatic endocrine proteins (cyan/pmagenta)):

We are considering to buy the Akoya Phenocycler-Fusion 2.0 or the Lunaphore COMET for Spatial proteomics services.
We can only choose one, so would like to know your experiences with these two platforms.