Integrated In-Silico Toxicological Risk Assessment of Vonoprazan Degradation Products using Complementary QSAR Approaches

Bhairavi Vijayanand Saraf¹*, Preeti Rajeev Mehta¹, Rahul Subhash Somani²

¹Department of Chemistry, Faculty of School of Basic and Applied Sciences,

 Sangam University, Bhilwara (Rajasthan) - 311001 India.

²Alkem Laboratories Limited, Mumbai 400013

*Corresponding Author E-mail: saraf.bhairavi@gmail.com

ABSTRACT:

Vonoprazan is a potassium-competitive acid blocker widely used for the management of acid-related gastrointestinal disorders. Although the chemical structures of vonoprazan have been elucidated, a systematic toxicological evaluation has not been reported in public literature. In the present study, an in-silico toxicological risk assessment was conducted for two major degradation products of vonoprazan namely N-dealkylated vonoprazan (DP1) and vonoprazan aldehyde (DP2) using QSAR methodologies recommended under ICH M7(R2) for mutagenicity assessment, along with additional evaluation of other toxicity endpoints for broader toxicological understanding. A complementary quantitative structure-activity relationship (QSAR) strategy was employed, as recommended under ICH M7(R2), integrating statistical-based predictions (ADMETlab), expert rule-based structural alert analysis (ToxTree), and additional consensus modeling using the Online Chemical Modeling Environment (OCHEM). Mutagenicity, carcinogenicity, and selected organ toxicity endpoints were evaluated for the intact degradation products as well as relevant structural fragments. DP1 was consistently predicted to be non-mutagenic and non-carcinogenic across all applied models and was therefore classified as an ICH M7 Class 5 (non-mutagenic) impurity, permitting control under standard ICH Q3B(R2) limits. In contrast, DP2 triggered structural alerts for potential genotoxicity due to the presence of an aldehyde functionality and was classified as an ICH M7 Class 3 impurity. However, in-silico metabolic assessment indicated rapid oxidation of DP2 to the corresponding carboxylic acid metabolite, which was devoid of genotoxic and carcinogenic alerts, supporting a mitigated toxicological risk. This study establishes a robust, toxicological framework aligned with ICH M7(R2) principles for mutagenicity assessment for the safety evaluation and control of vonoprazan degradation products and provides a scientifically justified basis for impurity specification setting and regulatory risk management in pharmaceutical development.

KEYWORDS: Vonoprazan, Impurity Profiling, In-silico toxicology, ICH M7(R2), QSAR.

INTRODUCTION:

Vonoprazan fumarate is a novel potassium-competitive acid blocker, its quality and safety of vonoprazan formulations is of critical importance throughout pharmaceutical development and product shelf life.¹ While impurities may be introduced during synthesis from starting materials and reagents, or form as degradation products under stress conditions such as hydrolysis and oxidation, vonoprazan has been extensively investigated using chromatographic and spectroscopic techniques to ensure rigorous impurity profiling and stability evaluation in line with modern regulatory standards.^2-23 Systematic forced degradation studies are therefore routinely employed to elucidate degradation pathways, establish the intrinsic stability of drug molecules, and support the development of stability-indicating analytical methods in compliance with regulatory expectations.^24-33 Several studies have reported the development and validation of stability-indicating HPLC and UPLC methods for vonoprazan, enabling the separation, detection, and quantification of process-related impurities and degradation products.^2-4,10,11 Furthermore, forced degradation investigations coupled with LC-MS/MS-based structural characterization have demonstrated the susceptibility of vonoprazan to hydrolytic and oxidative degradation, leading to the formation of chemically distinct degradation products.^3,5,38 In our previous research, a stability-indicating HPLC method was developed and the structures of two major degradation products identified during force degradation- DP1 (formed via hydrolytic N-dealkylation, known vonoprazan impurity 5) and DP2 (formed via oxidative stress, known vonoprazan sulfonyl aldehyde impurity) were successfully characterized using LC-MS/MS.³⁸ These analytical studies provide critical insight into impurity identity and formation pathways, integration with systematic toxicological risk assessment allows a more complete evaluation of the biological safety of the identified degradation products. Compliance with regulatory guidelines is mandatory such as ICH M7(R2) addresses the identification, classification, and control of DNA-reactive (mutagenic) impurities, mandating an appropriate safety evaluation.^39-40 In-silico toxicology, utilizing quantitative structure-activity relationship (QSAR) models, is now a regulatory-accepted approach for the initial hazard assessment of pharmaceutical impurities.⁴¹ This study therefore serves as a direct extension of our previous analytical work, aiming to perform a robust toxicological risk assessment of the identified degradation products (DP1 and DP2) using complementary statistical and rule-based QSAR methodologies, and to evaluate their metabolic fate in order to propose a scientifically justified control strategy in compliance with ICH Q3B(R2) and ICH M7(R2) guidelines.

METHODS AND MATERIALS:

Data Set and Compound Selection:

The chemical structures of Vonoprazan and its degradation products were derived from our previously published LC-MS characterization data. The chemical structures and Simplified Molecular Input Line Entry System (SMILES) strings for vonoprazan and its degradation products were verified using the PubChem database for all the m/z ratios identified in our previous study⁴².

Figure 1: Acid/Alkali treated vonoprazan Degradation³⁸

Figure 2: Peroxide treated vonoprazan Degradation [DP2] Pathway ³⁸

In-silico Prediction Models:

In accordance with ICH M7(R2) guidelines, which recommend the use of complementary QSAR methodologies, a combined modelling approach was employed. The QSAR models applied in this study follow the OECD principles for QSAR validation, including a defined endpoint, an unambiguous algorithm, a defined domain of applicability, appropriate measures of goodness-of-fit, robustness and predictivity, and mechanistic interpretation where possible. This approach improves prediction reliability and minimizes model-specific bias. Details of the tools used are provided below:

1. Statistical-Based Model:

(ADMETlab 3.0):^43-44 The ADMETlab 3.0 platform was used to predict probabilities for AMES (test procedure developed by Bruce Ames) mutagenicity, carcinogenicity, and hepatotoxicity/drug induced liver injury (DILI). This web-based platform employs machine learning models trained on large, curated datasets of experimentally validated compounds across multiple ADMET-related endpoints. The "Evaluation" and "Screening" modules were utilized to generate probability scores ranging from 0 to 1, where a score >0.5 typically indicates a positive prediction.

2. Rule-Based Model:

ToxTree v3.1:⁴⁵ ToxTree was used to identify structural alerts (toxicophores) using defined knowledge bases and to classify the compounds using defined decision trees. The following decision trees were applied: Cramer Rules: To assign a toxicity class (Class I, II, or III) based on structural complexity, ISS Mutagenicity (AMES) Rule Base: To identify structural alerts associated with bacterial mutagenicity and DNA-reactive alerts and ISS Carcinogenicity (Genotox and Non-Genotox) Rule Base: To screen for alerts related to carcinogenic potential.

OCHEM (Online Chemical Modeling Environment): ⁴⁶ The OCHEM platform was employed for a dual-layer analysis. First, the "ToxAlerts" module was used to perform a high-throughput screening of structural alerts, allowing for the isolation of new toxicophores present in the degradants but absent in the parent drug. Second, the "Run Predictions" module was utilized to generate AMES mutagenicity predictions based on multiple underlying QSAR models, providing an independent QSAR-based prediction to complement the ADMETlab results.

3. Classification of impurities: Classification of impurities was determined according to the ICH M7(R2) categories, reported below: Class 1/2: Known mutagen, Class 3: Alerting structure, unrelated to the parent, Class 4: Alerting structure, same as the parent and Class 5: No mutagenic/carcinogenic alerts.

RESULTS:

ICH M7(R2) classification decisions were based primarily on the intact degradation products DP1 (m/z 204) and DP2 (m/z 330). The lower-molecular-weight fragments derived from LC–MS/MS fragmentation patterns of the degradation products observed in the previous analytical study were not considered as independent impurities but were evaluated to support mechanistic interpretation and to identify structural alerts associated with specific substructures of the parent degradation products. The details on SMILES for each m/z are reported below in Table 1 and Table 2.

Table 1: Chemical Structures and SMILES Strings of vonoprazan and DP1

Compounds	m/z	Smiles
Parent Drug under hydrolytic forced degradation	346	CNCC1=C[NH+] (C(=C1) C2=CC=CC=C2F) S(=O) (=O) C3=CN=CC=C3
DP1 (degradant)	204	CNCC1=CNC(=C1) C2=CC=CC=C2F
1^st Fragment of DP1	189	CCC1=CNC(=C1) C2=CC=CC=C2F
2^nd Fragment of DP1	110	CNCC1=CNC=C1
3^rd Fragment of DP1	96	C1=CC=C(C=C1) F

Table 2: Chemical Structures and SMILES Strings of vonoprazan and DP2

Compounds	m/z	Smiles
Parent Drug under oxidative forced degradation	346	CNCC1=C[NH+] (C(=C1) C2=C(C=CC=C2) F) S(=O) (=O) C3=CN=CC=C3
DP2 (degradant)	330	C1=CC=C(C(=C1) C2=CC(=CN2S(=O) (=O) C3=CN=CC=C3) C=O) F
1^st Fragment of DP2	189	C1=CC=C(C(=C1) C2=CC(=CN2) C=O) F
2^nd Fragment of DP2	161	C1=CC=C(C(=C1) C2=CC=CN2) F
3^rd Fragment of DP2	95	C1=CNC=C1C=O

While the parent compound (m/z 346) remains chemically consistent, separate (Q)SAR evaluations were performed for each degradation pathway to enable a rigorous baseline comparison of DP1 and DP2 within their respective stress-specific contexts. A wide range of toxicity endpoints were generated using computational tools, however, only endpoints of primary regulatory relevance were selected for detailed discussion and have been summarized statistically in Table 3 and Table 4, through Rule-Based Structural Alert Analysis in Table 5, Table 6, Table 7 and Table 8. The complete data on toxicity predictions for the degradants is available in the supplementary data.

Statistical-Based Toxicity Assessment (ADMETlab 3.0):

Vonoprazan (Parent): The parent drug showed a negative prediction for AMES mutagenicity (Probability: 0.11) and carcinogenicity (Probability: 0.33), establishing a baseline safety profile. It exhibited high probabilities for hERG inhibition (0.70) and DILI (0.94), which are known class effects. DP1 (m/z 204): The hydrolytic product demonstrated a favorable safety profile. The AMES mutagenicity probability was low (0.35), and the carcinogenicity probability (0.28) was lower than the parent. DP2 (m/z 330): The oxidative product presented a mixed profile. While the AMES mutagenicity prediction remained Negative (Probability: 0.41), the carcinogenicity probability increased to 0.56 (Positive), and the DILI probability was notably high (0.96). The genotoxicity was predicted as high.

Table 3: Statistical-Based Prediction of Toxicological Risk Assessment for DP1

Compounds	m/z	Key AdMETlab genotoxicity endpoints	Other key AdMETlab toxicity/ ADME
Parent Drug under hydrolytic forced degradation	346	AMES: Negative. Genotoxicity: Low. Alerts: Non-genotoxicity and non-carcinogenicity alerts only.	High Risk: DILI, Respiratory, Neurotoxicity, hERG (0.70–0.75). ADME: Low HIA, High PAMPA. Ecology: Non-biodegradable, aquatic toxicity alerts.
DP1 (degradant)	204	AMES: Negative. Genotoxicity: Low. Alerts: None detected.	Toxicity: Moderate/Low DILI. ADME: Drug-like profile
1^st Fragment of DP1	189	AMES: Negative. Genotoxicity: Low. Alerts: Non-genotoxicity and non-carcinogenicity alerts only.	Toxicity: Moderate (DILI, Receptors). ADME: Suitable (Lipophilic fragment)
2^nd Fragment of DP1	110	AMES: Negative. Genotoxicity: Low. Alerts: None detected.	Toxicity: Low to Moderate. ADME: Typical (Small polar fragment).
3^rd Fragment of DP1	96	AMES: Borderline. Genotoxicity: Low. Alerts: Non-genotoxicity and non-carcinogenicity alerts only.	High Risk: BBB, PPB. Toxicity: Moderate (DILI, Organs). Ecology: Non-biodegradable, Aquatic tox.
Keys: ADME: Absorption, Distribution, Metabolism, and Excretion, BBB: Blood-Brain Barrier, PPB: Plasma Protein Binding, DILI: Drug‑Induced Liver Injury, hERG: human Ether‑à‑go‑go‑Related Gene, HIA: Human Intestinal Absorption, PAMPA: Parallel Artificial Membrane Permeability Assay.

Table 4: Statistical-Based Prediction Of Toxicological Risk Assessment for DP2

Compounds	m/z	Key AdMETlab genotoxicity endpoints	Other key AdMETlab toxicity/ ADME
Parent Drug under oxidative forced degradation	346	AMES: Negative. Genotoxicity: High. Alerts: None (Strong QSAR signal only).	High Toxicity: DILI, Respiratory, Neurotoxicity. hERG: High Other: FAF-Drugs4 substructures, Aquatic toxicity.
DP2 (degradant)	330	AMES: Borderline. Genotoxicity: High	High Toxicity: DILI, Respiratory, Neurotoxicity. CYP: Strong 3A4/2C8. Other: Skin sensitivity, Aquatic toxicity, FAF-Drugs4.
1^st Fragment of DP2	189	AMES: Moderate. Genotoxicity: Moderate. Alerts: Non-genotoxic and non-carcinogenicity.	Toxicity: Moderate-High DILI. ADME: Fragment compatible. Ecology: Aquatic toxicity, non-biodegradable.
2^nd Fragment of DP2	161	AMES: Negative. Genotoxicity: Low. Alerts: None.	Toxicity: Moderate flags. ADME: Typical (Heteroaromatic fragment).
3^rd Fragment of DP2	95	AMES: Low. Genotoxicity: Low. Alerts: None.	Toxicity: Low to Moderate. ADME: Typical (Volatile fragment).
Keys: DILI: Drug‑Induced Liver Injury, FAF-Drugs4 substructures: Free ADME-Tox Filtering Tool 4, hERG: human Ether‑à‑go‑go‑Related Gene. Note- The classification of DP2 as genotoxic is primarily driven by rule-based structural alerts (ISS model) rather than statistical AMES probability alone.

Rule-Based Structural Alert Analysis (ToxTree)

Cramer Classification:

Both DP1 and DP2 were classified as Class III (High Toxicity) due to the presence of fused heterocyclic rings (pyridine/pyrrole).

DP1 Assessment:

No structural alerts were identified for mutagenicity or carcinogenicity in the ISS rule bases. This aligns with the ADMETlab results.

DP2 Assessment:

The ISS rule base triggered a specific Structural Alert for Genotoxic Carcinogenicity, flagging the aldehyde moiety. This alert explains the elevated carcinogenicity score observed in the statistical model.

Table 5: Rule-Based Structural Alert Prediction of Toxicological Risk Assessment for DP1

Compounds	m/z	ToxAlerts (structure‑based)	ToxTree ISS (genotox /AMES)	Cramer class (ISS)
Parent Drug under hydrolytic forced degradation	346	Alerts: Heteroaromatic/Sulfonyl DNA-reactive: None.	Genotoxicity: Negative (non-genotoxic alerts only). AMES: No alerts. TA100: Unlikely.	Cramer Class III.
DP1 (degradant)	204	Pattern: Vonoprazan-like DNA-reactive: None.	Genotoxicity: No structural alerts. AMES: No structural alerts. S. typhimurium: Negative.	Cramer Class III.
1^st Fragment of DP1	189	Alerts: Generic aromatic DNA-reactive: None.	Genotoxicity: Negative. AMES: No alerts. TA100: Unlikely.	Cramer Class III.
2^nd Fragment of DP1	110	Alerts: Few (Small size) DNA-reactive: None.	Genotoxicity: No alerts. AMES: No alerts. TA100: Not indicated.	Cramer Class III.
3^rd Fragment of DP1	96	Alerts: Aromatic/Fluoro DNA-reactive: None.	Genotoxicity: Negative (non-genotoxic alerts only). AMES: No alerts. TA100: Unlikely.	Cramer Class III.
Keys: DNA: Deoxyribonucleic Acid, ISS: Istituto Superiore di Sanità, TA100: Salmonella mutagenicity test

Table 6: Rule-Based Structural Alert Toxicological Risk Assessment for DP2

Compounds	m/z	ToxAlerts (structure‑based)	ToxTree ISS (genotox /AMES)	Cramer class (ISS)
Parent Drug under oxidative forced degradation	346	Alerts: Dense (Sulfonyl-heteroaryl). Genotoxicity: Potential.	Genotoxicity: Positive (SA11gen). AMES: Positive. TA100: Unlikely.	Cramer Class III.
DP2 (degradant)	330	Alerts: Multiple (Sulfonyl-heteroaryl). Pattern: Matches DP2-346.	Genotoxicity: Positive. AMES: Positive (SA11gen). TA100: Unlikely.	Cramer Class III.
1^st Fragment of DP2	189	Alerts: Inherited (DP2 scaffold). Count: Reduced.	Genotoxicity: Negative (non-genotoxic alerts only). AMES: No alerts. TA100: Unlikely.	Cramer Class III.
2^nd Fragment of DP2	161	Alerts: Limited. DNA-reactive: None.	Genotoxicity: Negative. AMES: Negative. TA100: Unlikely.	Cramer Class III.
3^rd Fragment of DP2	95	Alerts: Few (Minimal structure) DNA-reactive: None.	Genotoxicity: Negative. AMES: Negative. TA100: Negative.	Cramer Class III.
Keys: ISS: Istituto Superiore di Sanità, SA11gen: Structural Alert Aldehydes Genotoxic Carcinogenicity, TA100: Salmonella mutagenicity test. Note: Although the statistical AMES probability was below the binary cut-off, the presence of a DNA-reactive aldehyde structural alert in rule-based models warranted conservative genotoxic classification in accordance with ICH M7(R2).

Rule-Based Structural Alert Analysis (OCHEM):

OCHEM-based predictions and structural alert screening were consistent with ADMETlab and ToxTree outputs, supporting the classification of DP1 as non-genotoxic and DP2 as structurally alerting.

Table 7: Summary of OCHEM-Based QSAR Predictions for DP1

Compound	m/z	OCHEM AMES Prediction	Structural Alerts (ToxAlerts)	Non-Genotoxic Carcinogenicity	Interpretation
Parent Drug under hydrolytic forced degradation	346	Negative	Non-Significant	No relevant alerts	Non-genotoxic
DP1 (degradant)	204	Negative	None	No alerts	Non-genotoxic
1^st Fragment of DP1	189	Negative	None	No alerts	Non-genotoxic
2^nd Fragment of DP1	110	Negative	None	No alerts	Non-genotoxic
3^rd Fragment of DP1	96	Borderline / Negative	None	No alerts	Non-genotoxic

Table 8: Summary of OCHEM-Based QSAR Predictions for DP2

Compound	m/z	OCHEM AMES Prediction	Structural Alerts (ToxAlerts)	Non-Genotoxic Carcinogenicity	Interpretation
Parent Drug under oxidative forced degradation	346	Negative / Weak	Minor alerts	Possible alerts (non-DNA reactive)	Low concern
DP2 (degradant)	330	Positive / Alert	Aldehyde structural alert	Present (structural)	Genotoxic concern
1^st Fragment of DP2	189	Negative	Reduced alerts	No alerts	Non-genotoxic
2^nd Fragment of DP2	161	Negative	None	No alerts	Non-genotoxic
3^rd Fragment of DP2	95	Negative	None	No alerts	Non-genotoxic

Metabolic Safety Assessment:

The predictive metabolite (vonoprazan-carboxylic acid) was generated using an in-silico metabolism prediction tool and was not experimentally observed in the LC-MS/MS analysis. QSAR evaluation of DP2 structural fragments lacking the aldehyde functionality demonstrated consistent non-genotoxic predictions, with no associated mutagenicity or carcinogenicity alerts. This observation indicates that the aldehyde moiety is the primary contributor to the observed genotoxic concern, and that its removal or transformation is associated with a significant reduction in toxicological risk. These findings provide mechanistic support for the proposed metabolic detoxification pathway of DP2.

DISCUSSION:

The integration of analytical data from our previous study with the current in-silico toxicological assessment provides a comprehensive safety profile for the degradation products of vonoprazan. The safety profile of the hydrolytic degradation pathway was established through a comprehensive genotoxicity evaluation of the parent drug and the specific degradation product DP1 (N-dealkylated vonoprazan, m/z 204), along with its characteristic mass fragments (m/z 189, 110, and 96). QSAR modeling consistently classified the intact DP1 (m/z 204) and all related fragments as non-genotoxic. Although the fragment at m/z 96 initially presented a borderline AMES alert, this finding was effectively overruled by predictions across complementary models. Structurally, the formation of DP1 involves the hydrolytic cleavage of the sulfonyl group, which exposes the electron-rich pyrrole ring (secondary amine). While unsubstituted pyrrole systems can theoretically undergo metabolic activation to form reactive epoxides or radical intermediates capable of DNA alkylation, our analysis confirms that the specific electronic environment of the fluorinated phenyl-pyrrole scaffold in DP1 does not trigger such mutagenic activity. Consequently, DP1 was categorized as an ICH M7 Class 5 impurity, indicating the absence of mutagenic or carcinogenic concern. As such, DP1 does not require control at the stringent Threshold of Toxicological Concern (TTC) applicable to DNA-reactive impurities and may instead be regulated as a standard organic impurity under ICH Q3B(R2), subject to dose-dependent qualification thresholds as defined under ICH Q3B(R2) (e.g. NMT 0.15% or 1.0 mg/day, depending on the maximum daily dose).

In contrast to the hydrolytic pathway, assessment of oxidative degradation products initially revealed significant toxicological concern. The parent drug subjected to oxidative stress and the specific degradation product DP2 (vonoprazan aldehyde, m/z 330) demonstrated positive structural alerts for mutagenicity in the rule-based ISS model and were therefore conservatively classified as genotoxic. This positive prediction was primarily driven by the presence of a reactive aldehyde moiety, which triggered rule-based ISS alerts, supported by statistical predictions from ADMETlab and independent QSAR predictions generated using the OCHEM platform. Mechanistically, this electrophilic carbonyl group poses a direct genotoxic threat due to its ability to react with the exocyclic amino groups of DNA bases (particularly guanine and adenine). This nucleophilic interaction can result in the formation of covalent Schiff base adducts and DNA-protein crosslinks, potentially interfering with DNA replication and inducing mutations. Under a conservative regulatory approach, such a DNA-reactive impurity would be controlled below the TTC of 0.0025µg/kg/day (approximately 0.15µg/day for a 60kg adult), in accordance with ICH M7(R2) guidance.

However, a deeper structural and metabolic analysis supports a refined risk interpretation. First, the characteristic mass fragments of DP2 (m/z 189, 161, and 95) were consistently predicted to be non-genotoxic, indicating that the genotoxic concern is specific to the intact aldehyde functionality rather than the core molecular scaffold. Second, ICH M7(R2) guidelines explicitly allow positive structural alerts to be contextualized using expert knowledge when rapid detoxification can be demonstrated. Physiologically, aldehydes are transient metabolic intermediates that are efficiently oxidized by the aldehyde dehydrogenase (ALDH) enzyme family. Metabolic simulation in the present study identified vonoprazan-carboxylic acid (m/z 346) as the primary detoxification product. This rapid metabolic conversion neutralizes the electrophilic aldehyde center, supporting the interpretation that DP2 behaves as a short-lived intermediate rather than a persistent mutagen in vivo.

It is critical to distinguish this detoxification metabolite from the parent drug vonoprazan, which also possesses a nominal mass of m/z 346. Although the parent compound and the metabolite are isobaric, their chemical structures and toxicological profiles are fundamentally distinct. The parent drug retains the sulfonyl moiety essential for potassium-competitive acid blocker activity, whereas the metabolite represents an oxidized carboxylic acid derivative of the pyridine side chain. Computational analysis confirmed that this metabolite is devoid of the genotoxic alerts associated with the aldehyde precursor. This metabolic read-across provides strong evidence that the genotoxic risk identified for DP2 is mitigated by rapid metabolic clearance.

Based on this comprehensive assessment, a scientifically justified control strategy is proposed. DP1 is confirmed as a non-genotoxic impurity and should be regulated as a standard organic impurity with a specification limit of NMT 0.15% in accordance with ICH Q3B(R2). In contrast, DP2 is formally classified as an ICH M7 Class 3 impurity due to the presence of an alerting structure unrelated to the parent molecule. While TTC-based approaches are generally applied for DNA-reactive impurities, the metabolic safety data presented herein supports the potential derivation of a compound-specific permissible daily exposure (PDE). However, in the absence of definitive in vitro AMES data, the most scientifically and regulatorily conservative approach is to maintain control at the DNA-reactive TTC or to perform a confirmatory AMES test to enable possible reclassification of DP2 as Class 5. Furthermore, given the susceptibility of vonoprazan to oxidative degradation leading to DP2 formation, formulation strategies should avoid excipients containing peroxide impurities (e.g., povidone or polyethylene glycol) to minimize aldehyde generation.

CONCLUSION:

This study presents a comprehensive in-silico toxicological risk assessment of vonoprazan degradation products using complementary QSAR models, in compliance with ICH M7(R2) and Q3B(R2) guidelines.

The hydrolytic product, DP1, consistently demonstrated an absence of genotoxic alerts and is classified as an ICH M7 Class 5 (non-mutagenic) impurity, permitting control under standard ICH Q3B(R2) limits. Conversely, the oxidative product, DP2, was classified as an ICH M7 Class 3 impurity due to a reactive aldehyde structural alert. However, metabolic simulation demonstrated the rapid oxidation of DP2 to a non-genotoxic carboxylic acid, supporting a mitigated risk assessment via metabolic read-across. While confirmatory AMES testing is recommended for definitive down-classification, these findings establish a scientifically justified, regulatory-compliant framework for the impurity profiling and control of vonoprazan formulations.

ACKNOWLEDGMENTS:

Authors are thankful to Venture Centre, Pune for their assistance with the LC-MS analysis. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors and is purely academic in nature.

REFERENCES

1. U.S. National Library of Medicine. DailyMed: Vonoprazan fumarate [Internet]. Bethesda (MD): National Institutes of Health; [cited 2025 May 10]. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0cc52ac5-77ec-4d66-a770-762a1a960914

2. Liu L, Cao N, Ma X, Xiong K, Sun L, Zou Q. Identification, characterization, and high-performance liquid chromatography quantification of process-related impurities in vonoprazan fumarate. J Sep Sci. 2016; 39(7): 1232–41. doi:10.1002/jssc.201501154

3. Luo Z, Liu A, Liu Y, Wang G, Chen X, Wang H, et al. Development of a stability-indicating HPLC method for simultaneous determination of ten related substances in vonoprazan fumarate drug substance. J Pharm Biomed Anal. 2018; 149: 133–42. doi: 10.1016/j.jpba.2017.11.011

4. Kanaan BM, Algohary AM, Ibrahim AM. Optimized reversed-phase liquid chromatography methodology for the determination of vonoprazan fumarate impurities: towards Six Sigma quality standards and sustainability assessment. Microchem J. 2024; 206: 111535. doi: 10.1016/j.microc.2024.111535

5. Abuothman M, Deeb AA, Hailat M, Abuyaman O, Aldoqum HM. A novel fast analytical method for the determination of N-nitroso vonoprazan in vonoprazan tablets and raw materials using LC-ESI-MS/MS. Int J Environ Anal Chem. 2024; 104(9): 1–100. doi:10.1080/03067319.2024.2407916

6. Qiao Y, Huang J, Xu Y, Zhao J, Wang Q. Determination of vonoprazan pyroglutamate and vonoprazan fumarate by HPLC. China Pharmacist. 2018; 12: 535–8.

7. Abdelazim AH, Abdel-Fattah A, Osman AO, Abdel-Kareem RF, Ramzy S. Spectrophotometric quantitative analysis of aspirin and vonoprazan fumarate in recently approved fixed-dose combination tablets using ratio spectra manipulating tools. J AOAC Int. 2023; 106(2): 490–5. doi:10.1093/jaoacint/qsac128

8. Saleh AM, El-Kosasy AM, Fares NV. UV spectrophotometric method development and validation of vonoprazan fumarate in bulk and pharmaceutical dosage form; green profile evaluation via eco-scale and GAPI tools. Egypt J Chem. 2023; 66(8): 141–8. doi:10.21608/ejchem.2022.161704.6948

9. Alzaghal NM, El-Mossalamy ES, El-Sayed GO. Method development and validation for estimation of vonoprazan by RP-HPLC method in bulk and tablets dosage form. Egypt J Chem. 2024;67(2):145–59. doi:10.21608/ejchem.2023.193129.7593

10. Salva C, Galla R. Development and validation of a stability-indicating RP-UPLC method for the simultaneous estimation of clarithromycin, amoxicillin, and vonoprazan in a physical mixture. J Appl Pharm Sci. 2024; 14(5): 193–202. doi:10.7324/JAPS.2024.165836

11. Moneim MM, Hamdy M. Chromatographic assay of recently approved co-formulation of vonoprazan fumarate with low-dose aspirin: AGREE, Complex MoGAPI, and RGB 12-model assessments. BMC Chem. 2024; 18(1): 230. doi:10.1186/s13065-024-01344-7

12. Ali NAM, El-Gindy AE, Wahba MEK, Mostafaa AE. Ecofriendly chromatographic method for the separation and quantification of vonoprazan fumarate, a novel potassium-competitive acid blocker with amoxicillin and clarithromycin effective in the treatment of Helicobacter pylori. Acta Chromatogr. 2025; 37: 1–13. doi:10.1556/1326.2025.01306

13. Yoneyama T, Teshima K, Jinno F, Kondo T, Asahi S. A validated simultaneous quantification method for vonoprazan (TAK-438F) and its four metabolites in human plasma by liquid chromatography–tandem mass spectrometry. J Chromatogr B. 2016; 1015–16:42–9. doi: 10.1016/j.jchromb.2016.01.051

14. Saraya RE, Hassan YF, Eltukhi WE, Salman BI. Ultra-sensitive fluorimetric method for estimation of vonoprazan in human plasma and content uniformity test. J Fluoresc. 2022; 32(5): 1725–32. doi:10.1007/s10895-022-02979-2

15. Chen X, Chen T, Huang Y, Wang M, Wang Y, Wu P, et al. LC–MS/MS method for rapid detection of vonoprazan fumarate in human plasma: development, validation and application to a bioequivalence study. Biomed Chromatogr. 2024;38(6): e5860. doi:10.1002/bmc.5860

16. El-Hamd MA, El-Maghrabey M, Magdy G, Soltan OM, Abdelrahman KS, Obaydo RH, et al. Factorial design-aided derivatization-free fluorimetric ultrasensitive assay of vonoprazan: application in uniformity of dosage units and plasma samples analysis. Microchem J. 2024; 205: 111320. doi: 10.1016/j.microc.2024.111320

17. Gupta A, Yadav J, Rawat S, Gandhi M. Method development and hydrolytic degradation study of doxofylline by RP-HPLC and LC-MS/MS. Asian J Pharm Anal. 2011; 1(1): 14–8.

18. Kasad PA, Muralikrishna KS. Method development and acid degradation study of rivaroxaban by RP-HPLC in bulk. Asian J Pharm Anal. 2013; 3(2): 62–5.

19. Gupta A, Rawat S, Gandhi M, Yadav J. Method development and acid degradation study of doxofylline by RP-HPLC and LC-MS/MS. Asian J Pharm Anal. 2011; 1(1): 10–3.

20. Gupta A, Rawat S, Pandey A. Method development and photolytic degradation study of doxofylline by RP-HPLC and LC-MS/MS. Asian J Pharm Anal. 2011; 1(2): 29-33.

21. International Council for Harmonisation (ICH). Q1A(R2): Stability testing of new drug substances and products. Geneva: ICH; 2003.

22. International Council for Harmonisation (ICH). Q1B: Photostability testing of new active substances and medicinal products [Internet]. Geneva: ICH; 1998 [cited 2025 May 10]. Available from: https://www.ich.org

23. International Council for Harmonisation (ICH). Q2(R1): Validation of analytical procedures: text and methodology [Internet]. Geneva: ICH; 2005 [cited 2025 May 10]. Available from: https://www.ich.org

24. Arous B, Al-Mardini MA, Ghazal H, Al-Lahham F. Stability-indicating method for the determination of rivaroxaban and its degradation products using LC-MS and TLC. Res J Pharm Technol. 2018; 11(1): 212–20. doi:10.5958/0974-360X.2018.00040.9

25. Manasa M, Aanandhi VM. Stability-indicating method development and validation of semaglutide by RP-HPLC in pharmaceutical substance and product. Res J Pharm Technol. 2021; 14(3): 1385–9. doi:10.5958/0974-360X.2021. 00247.X

26. Maruthi R, Chandan RS, Tengli A. LC-MS/MS and NMR characterization of impurities in epalrestat. Res J Pharm Technol. 2021; 14(1): 11–3. doi:10.5958/0974-360X.2021.00003.2

27. Patel RA, Patel MP, Shah HR, Shah N. Development and validation of stability-indicating high-performance liquid chromatographic method for indapamide. Asian J Pharm Technol. 2015;5(3):158–64. doi:10.5958/2231-5713.2015.00023.9

28. Chandana OSS, Ravichandra Babu R. Stability-indicating HPLC method development and validation for thalidomide and its impurity determination. Asian J Pharm Anal. 2016; 6(2): 115–8. doi:10.5958/2231-5675.2016. 00017.X

29. Patil S, Amurutkar S, Upasani C. Development and validation of stability-indicating RP-HPLC method for empagliflozin. Asian J Pharm Anal. 2016; 6(4): 201–6. doi:10.5958/2231-5675.2016.00030.2

30. Patel Y, Noolvi MN, Raj H, Patel M. Development and validation of a stability-indicating RP-HPLC method for estimation of donepezil HCl from bulk drug. Asian J Pharm Clin Res. 2015; 8(1): 14–8. doi:10.5958/2231-5691.2015.00014.3

31. Housheh S. Development of a rapid, simple and stability-indicating method for determination of azithromycin using RP-HPLC. Asian J Pharm Clin Res. 2017; 10(1): 9–13. doi:10.5958/2231-5691.2017.00009.0

32. Shinde N, Bangar B, Deshmukh S, Sulake S, Sherekar D. Pharmaceutical forced degradation studies with regulatory consideration. Asian J Res Pharm Sci. 2013; 3(4): 178–88.

33. Roge AB, Tarte PS, Kumare MM, Shendarkar GR, Vadvalkar SM. Forced degradation study: an important tool in drug development. Asian J Pharm Res. 2013; 3(4): 198–201.

34. Pharmaffiliates. Vonoprazan impurities [Internet]. [cited 2025 May 30]. Available from: https://www.pharmaffiliates.com/en/parentapi/vonoprazan-impurities

35. SynZeal Research Pvt Ltd. Vonoprazan reference standards and impurities [Internet]. [cited 2025 May 30]. Available from: https://www.synzeal.com/en/vonoprazan

36. Brambilla G, Martelli A. Genotoxic and carcinogenic risks from sulfoxides and sulfones. Mutat Res Rev Mutat Res. 2009; 681(2–3): 135–47. doi: 10.1016/j.mrrev.2006.09.003

37. European Medicines Agency (EMA). Impurities: Guideline for residual solvents and degradation products [Internet]. London: EMA; 2011 [cited 2025 May 10]. Available from: https://www.ema.europa.eu

38. Saraf BV, Mehta PR, Somani RS. Development and Validation of a Stability-Indicating HPLC Method for Vonoprazan with LC-MS Characterization of Degradation Products. Research Journal of Pharmacy and Technology. 2026 (Accepted, In Press).

39. International Council for Harmonisation (ICH). M7(R2): Assessment and control of DNA-reactive (mutagenic) impurities in pharmaceuticals to limit potential carcinogenic risk [Internet]. Geneva: ICH; 2017 [cited 2025 May 10]. Available from: https://www.ich.org.

40. International Council for Harmonisation (ICH). Q3B(R2): Impurities in new drug products [Internet]. Geneva: ICH; 2006 [cited 2025 May 10]. Available from: https://www.ich.org.

41. Dobo KL, Greene N, Fred C, Glowienke S, Harvey JS, Hasselgren C, et al. In silico methods combined with expert knowledge rule out mutagenic potential of pharmaceutical impurities: an industry survey. Regul Toxicol Pharmacol. 2012; 62(3): 449–63. doi: 10.1016/j.yrtph.2012.01.007.

42. National Center for Biotechnology Information. PubChem [Internet]. Bethesda (MD): National Library of Medicine (US), National Center for Biotechnology Information; 2004 – [cited 2025 Dec 13]. Available from: https://pubchem.ncbi.nlm.nih.gov/ .

43. ADMETlab 3.0. Evaluation Module [Internet]. Xiangya School of Pharmaceutical Sciences, Central South University; 2024 [cited 2025 Dec 13]. Available from: https://admetlab3.scbdd.com/server/evaluation.

44. ADMETlab 3.0. Screening Module [Internet]. Xiangya School of Pharmaceutical Sciences, Central South University; 2024 [cited 2025 Dec 13]. Available from: https://admetlab3.scbdd.com/server/screening.

45. Ideaconsult Ltd. ToxTree: Toxic Hazard Estimation by Decision Tree Approach (v3.1.0) [Computer Software]. SourceForge; 2018 [cited 2025 Dec 13]. Available from: http://toxtree.sourceforge.net.

46. Sushko I, Novotarskyi S, Körner R, Pandey AK, Rupp M, Teetz W, et al. Online chemical modeling environment (OCHEM): web platform for data mining, modeling and molecular properties prediction. J Comput Aided Mol Des. 2011; 25(6): 533–54. [Available from: https://ochem.eu/login/show.do]

Received on 02.02.2026      Revised on 11.03.2026

Accepted on 10.04.2026 Published on 16.04.2026

Available online from April 18, 2026

Asian Journal of Pharmaceutical Analysis. 2026; 16(2):81-88.

DOI: 10.52711/2231-5675.2026.00011

©Asian Pharma Press All Right Reserved

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Creative Commons License.