Department of Pharmaceutical Sciences, School of Pharmacy, University of Puerto Rico, Medical Sciences Campus, San Juan, PR 00936, USA; Crystallization Design Institute, Molecular Sciences Research Center, University of Puerto Rico, San Juan, PR 00926, USA.
Department of Chemistry, University of Puerto Rico, Mayagüez Campus, Mayagüez, PR 00681, USA.
J Pharm Biomed Anal. 2023 Sep 5;233:115451. doi: 10.1016/j.jpba.2023.115451. Epub 2023 May 8.
Compact composite probes were identified as a priority to alleviate space constraints in miniaturized unit operations and pharmaceutical manufacturing platforms. Therefore, in this proof of principle study, a compact composite sensor array (CCSA) combining ultraviolet and near infrared features at four different wavelengths (280, 340, 600, 860 nm) in a 380 × 30 mm housing (length x diameter, 7 mm diameter at the probe head), was evaluated for its capabilities to monitor in situ concentration of solutions and suspensions via multivariate analysis using partial least squares (PLS) regression models. Four model active pharmaceutical ingredients (APIs): warfarin sodium isopropanol solvate (WS), lidocaine hydrochloride monohydrate (LID), 6-mercaptopurine monohydrate (6-MP), and acetaminophen (ACM) in their aqueous solution and suspension formulation were used for the assessment. The results demonstrate that PLS models can be applied for the CCSA prototype to measure the API concentrations with similar accuracy (validation samples within the United States Pharmacopeia (USP) limits), compared to univariate CCSA models and multivariate models for an established Raman spectrometer. Specifically, the multivariate CCSA models applied to the suspensions of 6-MP and ACM demonstrate improved accuracy of 63% and 31%, respectively, compared to the univariate CCSA models [1]. On the other hand, the PLS models for the solutions WS and LID showed a reduced accuracy compared to the univariate models [1].
紧凑型复合探头被确定为缓解小型化单元操作和制药制造平台空间限制的优先事项。因此，在这项原理验证研究中，评估了一种紧凑型复合传感器阵列 (CCSA)，该阵列在 380×30mm 的外壳（长度×直径，探头头部直径为 7mm）中结合了四个不同波长（280、340、600 和 860nm）的紫外和近红外特征，通过使用偏最小二乘 (PLS) 回归模型的多元分析来监测溶液和悬浮液的原位浓度。使用了四种模型活性药物成分 (API)：异丙醇合华法林钠 (WS)、盐酸利多卡因一水合物 (LID)、6-巯基嘌呤一水合物 (6-MP) 和对乙酰氨基酚 (ACM) 在其水溶液和混悬剂配方中用于评估。结果表明，与单变量 CCSA 模型和用于建立的拉曼光谱仪的多变量模型相比，PLS 模型可用于 CCSA 原型以类似的准确度（验证样品在美国药典 (USP) 范围内）测量 API 浓度。具体而言，与单变量 CCSA 模型相比，应用于 6-MP 和 ACM 混悬液的多变量 CCSA 模型分别提高了 63%和 31%的准确度[1]。另一方面，WS 和 LID 溶液的 PLS 模型与单变量模型相比表现出较低的准确性[1]。
