Background: Therapeutic fasting improves joint pain in patients with osteoarthritis (OA), but the underlying mechanisms are unknown. Interactions of advanced glycation end products (AGEs) and their receptors (RAGE) play a role in OA pathogenesis. This study aimed to investigate whether the benefits of fasting in OA can be explained by changes in AGEs or RAGE. Patients and Methods: 37 patients with OA underwent fasting for 8 days. Serum levels of an AGE (N-ε-(carboxymethyl)-lysine; CML) and the soluble RAGE (sRAGE) as well as clinical outcome parameters such pain characteristics (measured by visual analogue scale; VAS), joint function (determined by the Western Ontario and McMaster Universities Arthritis Index; WOMAC), and quality of life (via the 36-Item Short-Form Health Survey (SF-36) questionnaire) were assessed. The variables were measured at baseline, the end of fasting, and at follow-up at 4 weeks. Results: The CML levels did not significantly change from baseline to the end of intervention (Δ = -25.6 ± 92.2 ng/ml; p = 0.10). In contrast, the sRAGE levels (Δ = -182.7 ± 171.4 ng/ml; p < 0.0001) and the sRAGE/CML ratio (Δ = -0.4 ± 0.6; p < 0.001) significantly decreased, but they returned to baseline levels 4 weeks after the end of fasting. The scores for pain, WOMAC, and the physical subscale of the SF-36 significantly improved during fasting. There was no correlation between the clinical outcomes and changes in serum levels of CML, sRAGE, or the sRAGE/CML ratio. Conclusions: Fasting resulted in a significant but non-sustained reduction of sRAGE levels and the sRAGE/CML ratio in OA, while the CML levels did not change. Improvement in clinical endpoints of OA does not correlate with changes in CML or sRAGE.