AI-guided optimization and eco-friendly assessment of in lab stability-indicating HPLC method for empagliflozin and sitagliptin in combined pharmaceutical dosage form
Abstract
An accurate analytical method is necessary for accurate quantification and stability assessment of novel antidiabetic fixed dose combination of empagliflozin and sitagliptin phosphate monohydrate. Artificial intelligence has shown promise in predicting chromatographic conditions, but experimental validation has significance for pharmaceutical applications. The primary objective of this study was to develop and compare a stability-indicating HPLC method for the simultaneous quantification of empagliflozin and sitagliptin phosphate monohydrate using AI tools (ChatGPT 4.5 (OpenAI), Perplexity.ai Sonar (Perplexity AI)) predicted and laboratory-optimized methods. In the laboratory, C18 column was used and the mobile phase was 0.1% triethylamine in water (pH 3.5 adjusted with 0.01% orthophosphoric acid), methanol and acetonitrile (40,35,25, v/v/v) as optimized chromatographic condition. The retention time of sitagliptin phosphate monohydrate and empagliflozin was found to be 2.95 min and 7.17 min, respectively. The method was validated as per ICH guidelines and was subjected to forced degradation studies like acidic, alkaline, oxidative, thermal and photolytic degradation. The Red Green Blue model was used to estimate greenness. The AI-predicted method exhibited similar chromatographic conditions; however, experimental implementation produced less favourable peak shape and lower resolution than those achieved with the optimized laboratory-developed method. The developed in-lab HPLC method was validated and the greenness assessment showed the greenness of the developed method. The proposed method showed lower LOD/LOQ than the published method. To the best of our knowledge, this is the first study to predict chromatographic conditions through AI tools for the simultaneous determination of empagliflozin and sitagliptin phosphate monohydrate and to directly compare their experimental performance with a laboratory-optimized method. The proposed method is simple, sensitive, robust and suitable for routine quality control analysis.
Related
- Green Micelle-Enhanced Spectrofluorimetric Determination of Gefitinib in Pharmaceutical Tablets Using Cremophor RH 40: Response-Surface Optimization and Time-Dependent Density Functional Theory (TD-DFT) Interpretation
- Fabrication and Optimization of 3D-Printed TiO2/g-C3N4 Scaffolds for Degradation of Organic Dye and Pharmaceutical Pollutant
- Comparative Feasibility Assessment of a Simplified Mossel EE–MacConkey Workflow for Detection of Enterobacteriaceae in Selected Non-Sterile Pharmaceutical Products
- Unlocking delafloxacin fluorescence with 9-fluorenyl methyl chloroformate: A sensitive spectrofluorimetric method for pharmaceutical and plasma analysis with greenness and whiteness assessment
- Stability-Indicating RP-HPLC Determination of Acetylsalicylic Acid, Dextromethorphan Hydrobromide, Phenylephrine Bitartrate, and Chlorpheniramine Maleate in an Effervescent Tablet Formulation
BibTeX
@article{akabari2026aiguided,
title = {{AI-guided optimization and eco-friendly assessment of in lab stability-indicating HPLC method for empagliflozin and sitagliptin in combined pharmaceutical dosage form}},
author = {Ashok H Akabari and Ashim Kumar Sen and Yadav Ragini and Ketan V. Shah and Mohammad Khishe and Ghanshyam G. Tejani},
journal = {Results in Chemistry},
year = {2026},
month = jun,
doi = {10.1016/j.rechem.2026.103581},
url = {https://doi.org/10.1016/j.rechem.2026.103581},
note = {CC BY},
}
RIS
TY - JOUR
TI - AI-guided optimization and eco-friendly assessment of in lab stability-indicating HPLC method for empagliflozin and sitagliptin in combined pharmaceutical dosage form
AU - Ashok H Akabari
AU - Ashim Kumar Sen
AU - Yadav Ragini
AU - Ketan V. Shah
AU - Mohammad Khishe
AU - Ghanshyam G. Tejani
PY - 2026
DA - 2026/06/25
JO - Results in Chemistry
DO - 10.1016/j.rechem.2026.103581
UR - https://doi.org/10.1016/j.rechem.2026.103581
AB - An accurate analytical method is necessary for accurate quantification and stability assessment of novel antidiabetic fixed dose combination of empagliflozin and sitagliptin phosphate monohydrate. Artificial intelligence has shown promise in predicting chromatographic conditions, but experimental validation has significance for pharmaceutical applications. The primary objective of this study was to develop and compare a stability-indicating HPLC method for the simultaneous quantification of empagliflozin and sitagliptin phosphate monohydrate using AI tools (ChatGPT 4.5 (OpenAI), Perplexity.ai Sonar (Perplexity AI)) predicted and laboratory-optimized methods. In the laboratory, C18 column was used and the mobile phase was 0.1% triethylamine in water (pH 3.5 adjusted with 0.01% orthophosphoric acid), methanol and acetonitrile (40,35,25, v/v/v) as optimized chromatographic condition. The retention time of sitagliptin phosphate monohydrate and empagliflozin was found to be 2.95 min and 7.17 min, respectively. The method was validated as per ICH guidelines and was subjected to forced degradation studies like acidic, alkaline, oxidative, thermal and photolytic degradation. The Red Green Blue model was used to estimate greenness. The AI-predicted method exhibited similar chromatographic conditions; however, experimental implementation produced less favourable peak shape and lower resolution than those achieved with the optimized laboratory-developed method. The developed in-lab HPLC method was validated and the greenness assessment showed the greenness of the developed method. The proposed method showed lower LOD/LOQ than the published method. To the best of our knowledge, this is the first study to predict chromatographic conditions through AI tools for the simultaneous determination of empagliflozin and sitagliptin phosphate monohydrate and to directly compare their experimental performance with a laboratory-optimized method. The proposed method is simple, sensitive, robust and suitable for routine quality control analysis.
SN - 2211-7156
ER -