Glps are probably the peptide with by far the best actual evidence they work to improve chronic pain, but she is already on those. The actual human clinical trial evidence is absent or minimal for the others unfortunately, except a study of ara-290 for neuropathy in humans, but you really need an accurate clinical diagnosis from an experienced expert clinician, if tests do not come up with answers, even if the best they can do is say is the symptoms or results are not specific enough to give a definitive diagnosis at this point in time, which is unfortunately sometimes the case. Trying to research or treat symptoms in the absence of a definite diagnosis is far from ideal.
A well prompted chatgpt being fed information from a medically experienced person can be surprisingly good at diagnosing less common conditions. Possibly better than doctors in some studies ( but it can get a bit carried away with some ideas, I had it tell me it thought my rash I was trying to diagnose was an early t-cell lymphoma for quite a while ) ,, but it can be very useful for coming up with things you had not thought of.
As a non peptide but with very unusually solid human clinical trial evidence for effectiveness for a supplement, and with good supporting preclinical data, PEA palmitoylethanolamide can be very useful for chronic pain, and has no known side effects which is very unusual, but be careful what you buy if you try it. Nearly all the ones on amazon US are majorly underdosed from what is advertised, one I bought advertised 1400mg of ingredients but weighed only 300mg per pill.
