Machine learning-based prediction of drug approvals using molecular, physicochemical, clinical trial, and patent-related features

ABSTRACT

Background

Drug development productivity has been declining lately due to elevated costs and reduced discovery rates. Therefore, pharmaceutical companies have been seeking alternative ways to determine and evaluate drug candidates.

Research design and methods

In this work, we proposed a new computational approach to directly predict the regulatory approval of drug candidates, and implemented it as a method called ‘DrugApp.’ To accomplish this task, we employed multiple types of features including molecular and physicochemical properties of drug candidates, together with clinical trial and patent-related features, which are then processed by random forest classifiers to train our disease group-specific approval prediction models.

Results

Our evaluations indicated DrugApp has a high and robust prediction performance. Within a use-case study, we showed our method can predict phase IV trial drugs that are later withdrawn from the market due to severe side effects. Finally, we used DrugApp models to forecast the approval of drug candidates that are currently in phases I/II/III of clinical trials.

Conclusions

We hope that our study will aid the research community in terms of evaluating and improving the process of drug development. The datasets, source code, results, and pre-trained models of DrugApp are freely available at https://github.com/HUBioDataLab/DrugApp.

KEYWORDS:

• Approval of drugs
• clinical trials
• drug patents
• machine learning
• molecular structures
• physicochemical properties
• predictive modeling

Acknowledgement

The authors thank Dr. Rengul Cetin-Atalay, Dr. Erden Banoglu, Dr. Yesim Aydin Son and Dr. Aybar Can Acar for their valuable comments and fruitful discussions throughout the study.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/17460441.2023.2153830

Author contributions

T. Doğan conceptualized the study and designed the methodology. F. Ciray prepared the datasets, did the coding, and performed data analyses. F. Ciray and T. Doğan evaluated and discussed findings. F. Ciray and T. Doğan wrote the manuscript. T. Doğan supervised the overall study. Both authors approved the manuscript.

Additional information

Funding

Fulya ÇIRAY was supported by TUBITAK BIDEB-2211-A PhD Fellowship Program.