PURPOSETo understand the mechanism of nano-crystalline drug formation in Pluronic (i.e., poly(ethylene oxide-block-propylene oxide) triblock copolymers) based drug-polymer solid dispersions.MATERIALS AND METHODSFour polymers, Pluronic F127, F108, F68 and PEG 8000, which have different poly(ethylene oxide) (PEO) and poly(propylene oxide) (PPO) ratio and chain length, were co-spray dried with BMS-347070, a COX-2 inhibitor, to form 50/50 (w/w) drug-polymer solid dispersions. The solid dispersions were analyzed by powder X-ray diffraction (PXRD), modulated differential scanning calorimetry (mDSC), and hot-stage microscopy. Average size of drug crystallites in different polymers was calculated by the Scherrer equation based on peak-broadening effect in PXRD. Two other drug compounds, BMS-A and BMS-B, were also spray dried with Pluronic F127, and the solid dispersions were analyzed by PXRD and mDSC.RESULTSThe average size of BMS-347070 crystallites in PEG 8000, F127, F108 and F68 polymers was 69, 80, 98 and 136 nm, respectively, and the degree of BMS-347070 crystallinity is the lowest in PEG 8000. Hot-stage microscopy showed that 50/50 drug-polymer dispersions crystallized in a two-step process: a portion of the polymer crystallizes first (Step 1), followed by crystallization of drug and remaining polymer (Step 2). The T (g) value of the BMS-347070/Pluronic dispersions after Step 1 (i.e., T(g1)) was measured and/or calculated to be 15-26 degrees C, and that of BMS-347070/PEG 8000 was 60 degrees C. Solid dispersions of BMS-A and BMS-B in Pluronic F127 have T(g1) of 72 and 3 degrees C, respectively; and PXRD showed BMS-A remained amorphous after approximately 3 weeks under ambient condition, while BMS-B crystallized in F127 with an average crystallite size of 143 nm.CONCLUSIONSThe size of drug crystallites in the drug-polymer solid dispersions is independent of polymer topology, but is caused kinetically by a combined effect of nucleation rate and crystal growth rate. When drug-Pluronic solid dispersions crystallize at room temperature, that is close to the T(g1) of the systems, a fast nucleation rate and a relatively slow crystal growth rate of the drug synergistically produced small crystallite size. While the much higher T(g1) value of drug-PEG 8000 led to a slower nucleation rate and an even slower crystal growth rate at room temperature, therefore, small crystallite size and low drug crystallinity were observed. Results from BMS-A/Pluronic and BMS-B/Pluronic systems confirmed this kinetic theory.