Automation and Active Learning for the Multi‐Objective Optimization of Antibody Formulations
Source abstract
Abstract
ABSTRACT Over the last forty years, monoclonal antibodies have become increasingly important therapeutic agents, with most manufactured as preformulated solutions. However, bioformulation of complex proteins is a difficult engineering challenge; formulations must be tailored to individual therapies, necessitating time‐ and material‐intensive campaigns to select combinations of excipients to simultaneously optimize various design criteria. These additives complicate formulation design with unintuitive and non‐linear relationships, creating a vast multidimensional design space that is intrinsically difficult to optimize using traditional techniques. To address this challenge, we investigated a high‐throughput discovery pipeline using machine learning to model and predict the effects of GRAS excipients on formulation behavior of a model antibody. This was supported by automation‐assisted “on‐demand” formulation to produce dozens of uniquely formulated antibody solutions for downstream evaluation and biophysical characterization. This pipeline was then integrated into an iterative closed‐loop cycle of automated Design‐Build‐Test‐Learn (DBTL), where new rounds of experiments are designed by the model. The process yielded both improved formulations and accurate predictive models of formulation behavior across multiple target objectives (melting temperature, diffusivity, and high‐concentration viscosity). This validates the utility of this technique to both map the underlying property‐function landscape and effectively guide formulation development while balancing multiple competing design requirements.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Monoclonal and Polyclonal Antibodies Research · Protein purification and stability · Biosimilars and Bioanalytical Methods
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Biomedical Engineering Rutgers The State University of New Jersey Piscataway New Jersey USA
Funding
National Institutes of Health · National Science Foundation
References (77)
Show source-provided references
1. Development of Therapeutic Antibodies for the Treatment of Diseases
2. 10.4161/mabs.1990910.4161/mabs.19909 ↗
3. 10.1080/19420862.2024.244353810.1080/19420862.2024.2443538 ↗
4. 10.1093/abt/tbac02110.1093/abt/tbac021 ↗
5. Therapeutic Antibodies in Medicine10.3390/molecules28186438 ↗
6. 10.3390/antib802003610.3390/antib8020036 ↗
7. 10.1016/j.xphs.2021.03.01710.1016/j.xphs.2021.03.017 ↗
8. 10.1002/jps.2072710.1002/jps.20727 ↗
9. 10.3109/1547691X.2013.82156410.3109/1547691X.2013.821564 ↗
10. 10.1016/j.ddtec.2020.08.00510.1016/j.ddtec.2020.08.005 ↗
11. 10.1021/acs.molpharmaceut.5b0064310.1021/acs.molpharmaceut.5b00643 ↗
12. 10.1021/acs.jpcb.8b0952710.1021/acs.jpcb.8b09527 ↗
13. 10.1080/19420862.2015.112860610.1080/19420862.2015.1128606 ↗
14. 10.1016/j.xphs.2023.03.02610.1016/j.xphs.2023.03.026 ↗
15. I. AbbVie Skyrizi (risankizumab‐rzaa) injection for subcutaneous or intravenous use: Prescribing Information (AbbVie Inc. 2026) accessed 2026/05/01 https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=7148c8eb‐b39e‐e20a‐6494‐a6df82392858.
16. I. Regeneron Pharmaceuticals Dupixent (dupilumab) injection for subcutaneous use: Prescribing Information (Regeneron Pharmaceuticals Inc. 2025) accessed 2026/05/01 https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=595f437d‐2729‐40bb‐9c62‐c8ece1f82780.
17. 10.1039/D4ME00201F10.1039/D4ME00201F ↗
18. 10.1002/jps.2219810.1002/jps.22198 ↗
19. 10.1016/j.addr.2011.07.00610.1016/j.addr.2011.07.006 ↗
20. 10.1016/j.xphs.2022.12.01110.1016/j.xphs.2022.12.011 ↗
21. Machine Learning Applied to Determine the Molecular Descriptors Responsible for the Viscosity Behavior of Concentrated Therapeutic Antibodies10.1021/acs.molpharmaceut.0c01073 ↗
22. 10.1080/19420862.2022.202620810.1080/19420862.2022.2026208 ↗
23. 10.1073/pnas.161640811410.1073/pnas.1616408114 ↗
24. 10.3390/biom1003036010.3390/biom10030360 ↗
25. 10.1080/19420862.2023.221118510.1080/19420862.2023.2211185 ↗
26. 10.1016/j.addr.2023.11510810.1016/j.addr.2023.115108 ↗
27. 10.1021/acs.chemrev.1c0002210.1021/acs.chemrev.1c00022 ↗
28. 10.1109/JPROC.2015.249421810.1109/JPROC.2015.2494218 ↗
29. Machine Learning in Chemoinformatics and Medicinal Chemistry10.1146/annurev-biodatasci-122120-124216 ↗
30. 10.1016/j.ailsci.2022.10005010.1016/j.ailsci.2022.100050 ↗
31. 10.1016/j.bmcl.2018.06.04610.1016/j.bmcl.2018.06.046 ↗
32. The Artificial Intelligence‐Powered New Era in Pharmaceutical Research and Development: A Review10.1208/s12249-024-02901-y ↗
33. Celecoxib Nanoformulations With Enhanced Solubility, Dissolution Rate, and Oral Bioavailability: Experimental Approaches Over In Vitro/In Vivo Evaluation10.3390/pharmaceutics15020363 ↗
34. Machine Learning Directed Drug Formulation Development10.1016/j.addr.2021.05.016 ↗
35. 10.1038/s41467‐025‐61113‐510.1038/s41467‐025‐61113‐5 ↗
36. 10.1002/bit.2866910.1002/bit.28669 ↗
37. 10.1016/j.ejpb.2022.12.00410.1016/j.ejpb.2022.12.004 ↗
38. 10.1016/j.crmeth.2022.10037410.1016/j.crmeth.2022.100374 ↗
39. 10.1038/nrd.2017.23210.1038/nrd.2017.232 ↗
40. 10.1016/j.ipha.2025.01.00110.1016/j.ipha.2025.01.001 ↗
41. 10.1021/acs.chemrev.4c0005510.1021/acs.chemrev.4c00055 ↗
42. Accelerating Materials Discovery Using Artificial Intelligence, High Performance Computing and Robotics10.1038/s41524-022-00765-z ↗
43. Machine‐Assisted Discovery of Chondroitinase ABC Complexes Toward Sustained Neural Regeneration10.1002/adhm.202102101 ↗
44. 10.1002/adma.20220180910.1002/adma.202201809 ↗
45. Data‐Driven Design of Novel Polymer Excipients for Pharmaceutical Amorphous Solid Dispersions10.1021/acs.bioconjchem.4c00294 ↗
46. 10.1021/acs.molpharmaceut.1c0046910.1021/acs.molpharmaceut.1c00469 ↗
47. 10.1021/acs.molpharmaceut.5c0059110.1021/acs.molpharmaceut.5c00591 ↗
48. 10.1016/j.ab.2009.12.00310.1016/j.ab.2009.12.003 ↗
49. Additive Effects of the New Viscosity‐Reducing and Stabilizing Excipients for Monoclonal Antibody Formulation10.1016/j.ijpharm.2025.125451 ↗
50. Nano Differential Scanning Fluorimetry for Comparability Studies of Therapeutic Proteins10.1016/j.ab.2020.113581 ↗
51. 10.1002/jps.2195510.1002/jps.21955 ↗
52. 10.3390/pharmaceutics1302017210.3390/pharmaceutics13020172 ↗
53. 10.1021/acs.jpcb.3c0392810.1021/acs.jpcb.3c03928 ↗
54. 10.1016/j.xcrp.2022.10104110.1016/j.xcrp.2022.101041 ↗
55. Navigating the Design Space of Inorganic Materials Synthesis Using Statistical Methods and Machine Learning10.1039/D0DT02028A ↗
56. On Latin Hypercube Sampling
57. 10.1016/j.jmp.2018.03.00110.1016/j.jmp.2018.03.001 ↗
58. M.Balandat B.Karrer D.Jiang et al. “BoTorch: A Framework for Efficient Monte‐Carlo Bayesian Optimization ” inAdvances in Neural Information Processing Systems(Curran Associates Inc. 2020) 21524–21538.
59. 10.1109/TIT.2011.218203310.1109/TIT.2011.2182033 ↗
60. Ginsbourger D. · 2010 · 13110.1007/978-3-642-10701-6_6 ↗
61. M. T.Ribeiro S.Singh andC.Guestrin “Why Should I Trust You?″: Explaining the Predictions of Any Classifier ” inProceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining(Association for Computing Machinery 2016) 1135–1144 https://doi.org/10.1145/2939672.2939778.10.1145/2939672.2939778 ↗
62. S. M.LundbergandS.‐I.Lee “A Unified Approach to Interpreting Model Predictions ” inAdvances in Neural Information Processing Systems(Curran Associates Inc. 2017) 4765–4774.
63. 10.1038/s42256‐019‐0138‐910.1038/s42256‐019‐0138‐9 ↗
64. 10.1016/j.xphs.2019.10.06510.1016/j.xphs.2019.10.065 ↗
65. 10.1093/abt/tbad02210.1093/abt/tbad022 ↗
66. Opalescence of an IgG1 Monoclonal Antibody Formulation Mediated by Ionic Strength and Excipients
67. 10.1214/07‐AOAS14810.1214/07‐AOAS148 ↗
68. 10.1007/s10994‐021‐05946‐310.1007/s10994‐021‐05946‐3 ↗
69. Machine Learning Models of Antibody–Excipient Preferential Interactions for Use in Computational Formulation Design10.1021/acs.molpharmaceut.0c00629 ↗
70. Bayesian Optimization and Machine Learning for Vaccine Formulation Development
71. 10.1016/j.xphs.2017.06.01710.1016/j.xphs.2017.06.017 ↗
72. Molecular Modeling and Machine Learning for Predicting High‐Concentration Antibody Viscosity10.1016/j.addr.2026.115839 ↗
73. 10.1038/s41551‐023‐01074‐610.1038/s41551‐023‐01074‐6 ↗
74. Analysis of Protein Stability and Ligand Interactions by Thermal Shift Assay10.1002/0471140864.ps2809s79 ↗
75. 10.1016/j.xphs.2018.04.01710.1016/j.xphs.2018.04.017 ↗
76. Dick J. · 201010.1017/CBO9780511761188 ↗
77. On the Distribution of Points in a Cube and the Approximate Evaluation of Integrals10.1016/0041-5553(67)90144-9 ↗