Rapid Communication The Effects of Fillers and Binders on the Accuracy of Tablet Subdivision Gessyka Rayana S. Pereira, 1 Stephania F. Taveira, 1 Marcilio Cunha-Filho, 2 and Ricardo Neves Marreto 1,3,4 Received 9 June 2018; accepted 6 August 2018 Abstract. The effects of excipients on the accuracy of tablet subdivision are severely underinvestigated. In this study, placebo tablets were prepared using a combined mixture design of fillers and binders to evaluate the effect of these excipients on subdivision accuracy. The responses assessed were mass loss, mass variation, tablet fragmentation, and increased friability. Dicalcium phosphate dihydrate (DCP) gave rise to more uniform and denser tablets than microcrystalline cellulose (MCC), thus resulting in greater subdivision accuracy. The binder type, hydroxypropylcellulose (HPC) or polyvinylpyrrolidone (PVP), did not affect the subdivision of DCP tablets. On the contrary, the structural similarity between HPC and MCC led to improved subdivision accuracy for MCC tablets. A less accurate subdivision was observed in tablets prepared with a DCP–MCC combination; this finding could be attributed to irregular binder distribution in this matrix. An optimized response was built using desirability analysis. This study helps to illuminate the relationship between fillers and binders to guide formulation scientists in the development of tablets with better subdivision performance. KEY WORDS: tablet splitting; binder distribution; cellulose excipients; matrix homogeneity. INTRODUCTION Tablets are frequently divided by patients or health professionals for dose adjustment, swallowing facilitation, and cost reduction [1,2]. This procedure is often related to the wide variations in drug dosage and can result in irregular plasma drug concentrations and unpredicted clinical outcomes [1,3]. Most published studies on this subject evaluate the accuracy of tablet subdivision by using drug products that are available on the market [1,3]. The subdivision patterns obtained from these studies reveal that certain characteristics of the tablet, such as oblong format, coating, and scored lines, lead to better subdivision results [1]. Nevertheless, consensus shows that tablet composition is the key to achieving successful subdivision and that excipients play a main role in determining the accuracy of tablet subdivision [4]. Despite this consensus, studies on the effect of pharma- ceutical excipients on tablet subdivision are still scarce [4–7]. Until now, the effects of pharmaceutical excipients on tablet subdivision had been described in terms of matrix density, matrix homogeneity, and tablet hardness. A more uniform and denser tablet structure favors subdivision accuracy [4–7]. An appropriate level of tablet hardness can also determine the success of tablet subdivision [4]. What remains to be understood is the role of excipient interaction on subdivision accuracy. To clarify this point, it is important to control the highest number of physical properties of the tablet. In the present study, tablets with different compositions were prepared with the same hardness and thickness by using the wet granulation method. A combined mixture design was applied to determine the effect of fillers and binders on the subdivision accuracy to understand the possible interactions between these materials and their repercussions on the subdivision process. Nonscore tablets were prepared to better discriminate the effects and interactions of excipients on tablet subdivision allowing for their rational use in the development of formulations that are designed to be split. MATERIAL AND METHODS Materials Klucel ® ELF (hydroxypropylcellulose (HPC), 40,000 Da) and polyvinylpyrrolidone (PVP)K-30 ® (58,000 Da) were kindly donated by Ashland (São Paulo, Brazil). Microcrystalline cellulose (MCC) pH 101 and dicalcium phosphate dihydrate (DCP) were obtained from Mingtai (Taiwan). Colloidal silicon dioxide, magnesium stearate, and sodium croscarmellose were purchased from Evonik Degussa GmbH (Essen, Germany) and ValdequímicaProdutosQuímicosLtda (São Paulo, Brazil), respectively. 1 Laboratory of Nanosystems and Drug Delivery Devices (NanoSYS), School of Pharmacy, Universidade Federal de Goiás, Goiânia, GO, Brazil. 2 Laboratory of Food, Drug and Cosmetics (LTMAC), School of Health Sciences, University of Brasília, Brasília, DF, Brazil. 3 Faculdade de Farmácia UFG, Rua 240, esquina com 5° Avenida, s/n, Setor Leste Universitário, Goiânia, GO 74 605-170, Brazil. 4 To whom correspondence should be addressed. (e–mail: ricardomarreto@ufg.br) AAPS PharmSciTech ( # 2018) DOI: 10.1208/s12249-018-1144-7 1530-9932/18/0000-0001/0 # 2018 American Association of Pharmaceutical Scientists