Without a flattener an RC8 with a full frame sensor will show strong effects of field curvature in the corners of the images. A good example showing this is
here. With a tilt the things get even more complicated.
Unless you have proper tools (Tak scope) I would not touch the collimation until you can get rid of the tilt and get the focus right.
Pick a bright star in the middle of the image and adjust focus until the double diffraction pattern is gone.
Take an image of a bright globular cluster (Omega Cen is a good one at this time of the year). You should be able to see lots of stars all the way to the end of the image which can help diagnose the tilt.