Quantitative measurement of crystalline impurity in a pharmaceutical tablet by X-ray powder diffraction and method validation

Abstract

A powder X-ray diffraction method was developed and validated to measure the crystalline impurity 4-(5-cyclopentyloxy-carbonylamino-1-methyl-indol-3-ylmethyl)-3-methoxy-N-o-tolylsulfonylbenzamide hydrate in a pharmaceutical tablet ranging from 0.6 to 3% (w/w). The calibration plot was found to be linear with a correlation coefficient (r2) of 0.996, and was reproducible among operators. The detection limit was determined to be 0.6% with a signal-to-noise ratio of 3:1. The quantitation limit was determined to be 1% with a signal-to-noise ratio of 5:1. Instrument precision at the quantitation limit was 5.8%. Method precision was 6.1% at the quantitation limit and 7.4% at the detection limit. Intermediate precision at the quantitation limit was 7.3% during a 6-month study. Accuracy measurements using crystalline impurity standards prepared in an excipient mixture ranged from 89.3 to 105.5%. Accuracy measurements using tablets containing spiked quantities of crystalline impurity ranged from 72.0 to 92.7%. Accuracy measurements using spiked tablets were complicated because the crystalline impurity was lost during the manufacturing process and a correction factor was used. Ruggedness was assessed by evaluating repetitive assay, repetitive packing, sample packing, and sample stability. Repetitive assays show the exposure of standards to a relative humidity in excess of 57% caused displacement error because of an increase in sample volume and a peak-position shift. Repetitive-packing studies show the analyte was extracted from the sample at a low relative humidity because of a static-charge induction. Sample-packing studies show that two subjective packing techniques were equivalent, and that under- and over-packing samples cause changes in sample density which would not affect results within ±16%. Sample-stability studies show that the quantitation-limit standard was stable as long as the sample was exposed to a relative humidity below 57%.