Decitabine (DCB) is very unstable in aqueous solutions because of the opening of the N-5 and C-6 aza-ring. The objective of this study was to investigate a stabilization mechanism of aqueous-labile decitabine (DCB) with three different materials by preparing binary spray dried powders (SDs): self-assembling fattigated gelatin-oleic acid conjugate (GOC), micellar polyethylene glycol hexadecyl ether (Brij® 58), nonionic surfactant and electrostatically interactive xanthan gum (XA), an anionic polysaccharide. GOC was synthesized by covalently conjugating oleic acid with gelatin. The degradation rate of intact DCB was highly dependent on the solution conditions, showing stability in the following decreasing order: pH 1.2 < pH 4.5 < pH 6.8 < pH 7.4 < deionized water (DW) < 50 % ethanol. DCB was the most unstable in the low pH 1.2 solution, with only 25.1 % remaining after 2 h. As the GOC content (DCB: GOC = 1:5, 1:10, 1:50) increased in the GOC SDs, the degradation rate of DCB decreased proportionally because GOC readily formed nanoparticles (NPs) with a critical micelle concentration (CMC) of 0.0416 mg/mL to protect DCB against degradation. This nano-forming behaviors of GOC was also visualized using field emission transmission electron microscope (FE-TEM) and field emission scanning electron microscope (FE-SEM). Brij® 58 SD (DCB: Brij® 58 = 1:10) slightly increased the DCB stability via micellization above CMC for a short period of time as compared to the DCB control. In contrast, XA had more favorable stabilizing effect at a low pH 4.5 solution via polyelectrolyte complex formation of positively charged DCB and negatively charged XA compared with other XA SD prepared in DW or pH 6.8 solution. Among the three stabilizing materials, GOC provided the highest stabilization capacity via the nanonization process compared with micellar Brij® 58 and electrostatically interactive XA. Furthermore, the molecular hydrogen bonding interactions and amorphousness of DCB with GOC, Brij® 58, and XA contributed to the stabilization of DCB according to instrumental analyses such as Fourier transform infrared spectroscopy (FT-IR), powder X-ray diffraction (PXRD), and differential scanning calorimetry (DSC). The in vitro release of GOC SDs was maintained at pH 1.2 and pH 6.8, while DCB alone was rapidly released and then degraded. The more GOC used, the more DCB release delayed. The release rate of Brij® 58 SD was released immediately but gradually decreased. However, XA SD also rapidly released DCB, which then degraded in a steeper manner. The release profiles of DCB were governed by the balanced effect of the stabilization capacity in the solution and loading contents of three stabilizing materials in SDs formulations.