Ordered MCM-48 nanoporous silica (S-BET = 923(3) m(2).g(-1), VP = 0.63(2) cm(3).g(-1)) with cubic Ia3d symmetry was used as a support for drug delivery of anti-inflammatory poorly soluble drug indomethacin. The delivery from parent, unmodified MCM-48, and 3-aminopropyl modified silica carrier was studied into the simulated body fluids with the pH = 2 and pH = 7.4. The studied samples were characterized by thermal analysis (TG/DTG-DTA), N-2 adsorption/desorption, infrared spectroscopy (FT-IR), powder XRD, SEM, HRTEM methods, measurements of zeta potential (zeta) and dynamic light scattering (DLS). The determined content of indomethacin in pure MCM-48 was 21 wt.% and in the amine-modified silica MCM-48A-I the content was 45 wt.%. The release profile of the drug, in the time period up to 72 h, was monitored by TLC chromatographic method. It as shown, that by the modification of the surface, the drug release can be controlled. The slower release of indomethacin was observed from amino modified sample MCM-48A-I in the both types of studied simulated body fluids (slightly alkaline intravenous solution with pH = 7.4 and acidic gastric fluid with pH = 2), which was supported and explained by zeta potential and DLS measurements. The amount of the released indomethacin into the fluids with various pH was different. The maximum released amount of the drug was 97% for sample containing unmodified silica, MCM-48-I at pH = 7.4 and lowest released amount, 57%, for amine modified sample MCM-48A-I at pH = 2. To compare the indomethacin release profile four kinetic models were tested. Results showed, that that the drug release based on diffusion Higuchi model, mainly governs the release. (C) 2018 Elsevier B.V. All rights reserved.