Drug Development Life Cycle Research Articles

Artificial intelligence integration in the drug lifecycle and in regulatory science: policy implications, challenges and opportunities.

Artificial intelligence tools promise transformative impacts in drug development. Regulatory agencies face challenges in integrating AI while ensuring reliability and safety in clinical trial approvals, drug marketing authorizations, and post-market surveillance. Incorporating these technologies into the existing regulatory framework and agency practices poses notable challenges, particularly in evaluating the data and models employed for these purposes. Rapid adaptation of regulations and internal processes is essential for agencies to keep pace with innovation, though achieving this requires collective stakeholder collaboration. This article thus delves into the need for adaptations of regulations throughout the drug development lifecycle, as well as the utilization of AI within internal processes of medicine agencies.

VariantSurvival: a tool to identify genotype-treatment response.

Motivation: For a number of neurological diseases, such as Alzheimer's disease, amyotrophic lateral sclerosis, and many others, certain genes are known to be involved in the disease mechanism. A common question is whether a structural variant in any such gene may be related to drug response in clinical trials and how this relationship can contribute to the lifecycle of drug development. Results: To this end, we introduce VariantSurvival, a tool that identifies changes in survival relative to structural variants within target genes. VariantSurvival matches annotated structural variants with genes that are clinically relevant to neurological diseases. A Cox regression model determines the change in survival between the placebo and clinical trial groups with respect to the number of structural variants in the drug target genes. We demonstrate the functionality of our approach with the exemplary case of the SETX gene. VariantSurvival has a user-friendly and lightweight graphical user interface built on the shiny web application package.

Planning Benefit-Risk Assessments Using Visualizations.

A Benefit Risk Assessment Plan (BRAP) describes the assessments planned to determine whether the benefits of an investigational drug outweigh the risks. The plan can have two sections, one with timelines for aligning resources with decision milestones and the other for pre-specifying assessments for decision milestones. Regulatory guidance recommends a proactive planning process over an ad-hoc process. However, very little has been published about proactive plans themselves. This article works through a hypothetical example visualizing a series of assessments across the drug development lifecycle. Based on a regulatory framework, the planning process starts with assessing the medical condition and current treatment options. These early assessments bring out major considerations in assessing the investigational drug.

A novel Hyperledger blockchain-enabled decentralized application for drug discovery chain management

The aim of the drug discovery process is to create a brand-new drug that is both safe and useful for treating disease in patients. The trustworthy collaboration of the organisations involved in the drug discovery process and the integrity of their contributions are essential for this process. Current drug discovery chains use centralised systems, which are susceptible to lockdown by cyberattacks. Blockchain, with its many characteristics such as accountability, immutability, integrity, privacy, and security, has the potential to be extremely useful in drug discovery chain management. The purpose of this work is to create a drug discovery framework utilizing the strengths of blockchain technology in combination with machine learning (ML). Such a system would provide a secure, efficient, and faster drug development life cycle. This study presents a novel Hyperledger fabric-based drug discovery application that empowers the permissioned organisations to upload, update, view, and verify contributions. ML is used to preprocess data and visualise features. In the proposed work, a unique identifier is assigned to each contribution asset using the secure hash algorithm (SHA-256). The proposed design also enables the regulatory authority to issue certificates proving the ownership of contributions to the contributing organisations. The blockchain ledger has been used to store the meta-data and InChIKey of drug contributions, and actual contributions are in off-chain storage. In this work, we have successfully built an end-to-end decentralised drug discovery application with a front-end interface and demonstrated chaincode algorithms. The end-to-end application is not available in any previous work. The Caliper tool has been used to investigate throughput, latency, and resource statistics. The performance and comparative analysis show that the proposed design is scalable and promising for drug discovery chain management.

Walking the talk in digital transformation of regulatory review.

Cloud-based regulatory platforms have the potential to substantially transform how regulatory submissions are developed, transmitted, and reviewed across the full life cycle of drug development. The benefits of cloud-based submission and review include accelerating critical therapies to patients in need globally and efficiency gains for both drug developers and regulators. The key challenge is turning the theoretical promise of cloud-based regulatory platforms into reality to further the application of technology in the regulatory processes. In this publication we outline regulatory policy journeys needed to effect the changes in the external environment that would allow for use of a cloud-based technology, discuss the prerequisites to successfully navigate the policy journeys, and elaborate on future possibilities when adoption of cloud-based regulatory technologies is achieved.

The role of Real-World Data and evidence in oncology medicines approved in EU in 2018–2019

Use of Real-World Data (RWD) has gained the interest of different stakeholders in cancer care. The aim of this study was to identify and describe the use of RWD/RWE during the pre-authorization phase of products authorized by the EMA in 2018 and 2019 (n = 111), with the focus on oncology medicines (n = 24). Information was extracted from the European Public Assessment Report (EPAR) summaries and recorded for 5 stages (11 categories) of the drug development lifecycle (discovery, early development, clinical development, registration/market launch, lifecycle management). Specific chapters of full EPAR were reviewed to substantiate the findings on RWD/RWE use in clinical trial design, efficacy, safety, and effectiveness evaluation. RWD/RWE is present in all stages of the oncology drug development; 100.0 % in discovery, 37.5 % early development, 58.3 % in clinical development, 62.5 % in registration decision and 100.0 % in post-authorization lifecycle management. Examples showed that trial design supported by RWD/RWE included use of open label/single arm studies; efficacy was about using either comparison of results to historical controls, supplying survey data obtained outside the clinical trial or utilizing expert panel advice; safety about including literature findings in evidence; and effectiveness on comparison of trial results of the given product to historical data or existing standard of care. The findings of this study provide specific insights into how RWD/RWE is used in development of cancer therapeutics, how it contributes to regulatory decision making and can guide further policy developments in this field.

MCANet: shared-weight-based MultiheadCrossAttention network for drug-target interaction prediction.

Accurate and effective drug-target interaction (DTI) prediction can greatly shorten the drug development lifecycle and reduce the cost of drug development. In the deep-learning-based paradigm for predicting DTI, robust drug and protein feature representations and their interaction features play a key role in improving the accuracy of DTI prediction. Additionally, the class imbalance problem and the overfitting problem in the drug-target dataset can also affect the prediction accuracy, and reducing the consumption of computational resources and speeding up the training process are also critical considerations. In this paper, we propose shared-weight-based MultiheadCrossAttention, a precise and concise attention mechanism that can establish the association between target and drug, making our models more accurate and faster. Then, we use the cross-attention mechanism to construct two models: MCANet and MCANet-B. In MCANet, the cross-attention mechanism is used to extract the interaction features between drugs and proteins for improving the feature representation ability of drugs and proteins, and the PolyLoss loss function is applied to alleviate the overfitting problem and the class imbalance problem in the drug-target dataset. In MCANet-B, the robustness of the model is improved by combining multiple MCANet models and prediction accuracy further increases. We train and evaluate our proposed methods on six public drug-target datasets and achieve state-of-the-art results. In comparison with other baselines, MCANet saves considerable computational resources while maintaining accuracy in the leading position; however, MCANet-B greatly improves prediction accuracy by combining multiple models while maintaining a balance between computational resource consumption and prediction accuracy.

Patient engagement in pharmaceutical development: Where are we? – Report from a symposium

In recent years, the term “patient engagement” has found its way into healthcare and specifically into the field of drug development. To better understand the actual status of “patient engagement” in drug development, a symposium was organized by the Drug Research Academy of the University of Copenhagen (Denmark) on November 16, 2022. The symposium brought together experts from regulatory authorities, industry, academia and patients to share their views and experience of and with patient engagement in drug product development. The symposium led to intensive discussions among the speakers and the audience, confirming that viewpoints and experiences of the different stakeholder provide important input into the promoting patient engagement along the entire drug development life cycle.

LC-HRMS and NMR studies for characterization of forced degradation impurities of ponatinib, a tyrosine kinase inhibitor, insights into in-silico degradation and toxicity profiles

The degradation profile of ponatinib was established during the present study by exposing it to various stress conditions. In-silico degradation pattern of ponatinib was outlined by using Zeneth software. Five degradation impurities were formed during the stress testing of ponatinib. High performance liquid chromatographic method was developed to separate these degradation impurities which includes ammonium acetate of pH 4.75 (A) and methanol (B) as mobile phase in gradient elution mode and Waters Reliant C18 (4.6 × 250 mm, 5 µm) column as stationary phase. Optimised flow rate, injection volume and detection wavelength of the HPLC method were 1.0 mL/min, 10 µL and 254 nm, respectively. Chemical structures of degradation impurities were proposed by high resolution mass spectrometry further, major degradation products were isolated, enriched and investigated thoroughly with the aid of nuclear magnetic resonance spectroscopy studies. The degradation impurities were identified as 4-aminophthalaldehyde (DP 1), 4-((4-methylpiperazin-1-yl)methyl)− 3-(trifluoromethyl) benzenamine (DP 2), 3-(2-(imidazo[1,2-b]pyridazin-3-yl)acetyl)− 4-methylbenzoic acid (DP 3), 3-(2-(imidazo[1,2-b]pyridazin-3-yl)ethynyl)− 4-methylbenzoic acid (DP 4) and N-oxide impurity (DP 5) which are new and were not reported in the literature till date. Additionally, toxicity and mutagenicity profiles of ponatinib and its degradation impurities were predicted in-silico by using DEREK and SARAH software. This whole study gives meaningful insights about chemical stability of ponatinib which is useful in its drug development lifecycle.

Identification of drug-side effect association via restricted Boltzmann machines with penalized term.

In the entire life cycle of drug development, the side effect is one of the major failure factors. Severe side effects of drugs that go undetected until the post-marketing stage leads to around two million patient morbidities every year in the United States. Therefore, there is an urgent need for a method to predict side effects of approved drugs and new drugs. Following this need, we present a new predictor for finding side effects of drugs. Firstly, multiple similarity matrices are constructed based on the association profile feature and drug chemical structure information. Secondly, these similarity matrices are integrated by Centered Kernel Alignment-based Multiple Kernel Learning algorithm. Then, Weighted K nearest known neighbors is utilized to complement the adjacency matrix. Next, we construct Restricted Boltzmann machines (RBM) in drug space and side effect space, respectively, and apply a penalized maximum likelihood approach to train model. At last, the average decision rule was adopted to integrate predictions from RBMs. Comparison results and case studies demonstrate, with four benchmark datasets, that our method can give a more accurate and reliable prediction result.

Identification of a monoclonal antibody clipping variant by cross-validation using capillary electrophoresis – sodium dodecyl sulfate, capillary zone electrophoresis – mass spectrometry and capillary isoelectric focusing – mass spectrometry

Critical quality attributes (CQAs) of recombinant monoclonal antibody therapeutics are constantly monitored throughout the life cycle of drug development and manufacturing. In the past few decades, numerous analytical techniques have been developed for the characterization of CQAs. In this regard, non-reduced and reduced capillary electrophoresis – sodium dodecyl sulfate (CE-SDS) methods have been widely adopted by the biopharmaceutical industry for the evaluation of size-related heterogeneities in biologics. In this work we demonstrate that, with recent development of capillary electrophoresis – mass spectrometry (CE-MS) technologies, a clipping variant of bevacizumab may be identified directly by both capillary zone electrophoresis – mass spectrometry (CZE-MS) and capillary isoelectric focusing – mass spectrometry (cIEF-MS) approaches, providing a powerful addition to the traditional CE-SDS analysis workflow. In this novel workflow, linear regression between the mobility and molecular weight first results in an approximate size range of this variant. The intact masses of all species in the bevacizumab are then obtained, after deconvolution of all features identified in the CZE-MS analysis. Subsequent CZE-MS analysis of the subunits of bevacizumab leads to the confirmation of a clipped heavy chain. Furthermore, cIEF-MS of the intact bevacizumab confirms the existence of this clipping variant. The cross-validation between CE-SDS, CZE-MS, and cIEF-MS, creates a comprehensive roadmap for monoclonal antibody size variants profiling. These CE-based analytical techniques are complementary to each other, leading to orthogonal verification for size heterogeneity characterization.

Applications of Model Informed Drug Development (MIDD) in Drug Development Lifecycle and Regulatory Review.

Applications of Model Informed Drug Development (MIDD) in Drug Development Lifecycle and Regulatory Review.

SA38 Cross-Functional Perceptions on Value and Utility of Real-World Evidence (RWE) within an Emerging Biopharmaceutical Company (EBP)

RWE is becoming increasingly accepted by healthcare stakeholders (e.g., regulators, payers, providers, patients, patient advocacy groups) to support decision-making across the product drug development lifecycle. However, it is unclear how the value and utility of RWE is perceived cross-functionally by EBPs. The objective of this survey was to assess baseline RWE knowledge of varied personnel representing medical and clinical affairs at an EBP and identify key learning opportunities to leverage RWE and optimize execution in the product development lifecycle.

Landscape Analysis of the Application of Artificial Intelligence and Machine Learning in Regulatory Submissions for Drug Development From 2016 to 2021.

Landscape Analysis of the Application of Artificial Intelligence and Machine Learning in Regulatory Submissions for Drug Development From 2016 to 2021.

Review: Role of Model-Informed Drug Development Approaches in the Lifecycle of Drug Development and Regulatory Decision-Making.

Model-informed drug development (MIDD) is a powerful approach to support drug development and regulatory review. There is a rich history of MIDD applications at the U.S. Food and Drug Administration (FDA). MIDD applications span across the life cycle of the development of new drugs, generics, and biologic products. In new drug development, MIDD approaches are often applied to inform clinical trial design including dose selection/optimization, aid in the evaluation of critical regulatory review questions such as evidence of effectiveness, and development of policy. In the biopharmaceutics space, we see a trend for increasing role of computational modeling to inform formulation development and help strategize future in vivo studies or lifecycle plans in the post approval setting. As more information and knowledge becomes available pre-approval, quantitative mathematical models are becoming indispensable in supporting generic drug development and approval including complex generic drug products and are expected to help reduce overall time and cost. While the application of MIDD to inform the development of cell and gene therapy products is at an early stage, the potential for future application of MIDD include understanding and quantitative evaluation of information related to biological activity/pharmacodynamics, cell expansion/persistence, transgene expression, immune response, safety, and efficacy. With exciting innovations on the horizon, broader adoption of MIDD is poised to revolutionize drug development for greater patient and societal benefit.

On the brink of transformation

The research on drug development life cycle and bringing sole new drug to the market is a million dollar question for pharmaceutical organization. Any clinical trial consumes average of 10 to 15 years and USD 1.5-2.0 billion with uncertainty of medications for its effectiveness for human use. Hardly, one out of 10 compounds entering into the clinical trial that reaches to the market rendering a major loss to pharmaceutical or biotech company in case of trial failure. Conversely, with changing time and an increase in the number of medicines approved by regulatory authorities, the regulatory teams are increasing networks for monitoring and assembling adverse event reports from varied sources. This in turn, has increases annual exponential rise in data volumes and the companies are facing a huge challenge in processing it. To meet such challenges, organizations must sharpen their ability to introduce new wearables for clinical trials and provide advanced cognitive solution to handle large and complex datasets. This has summoned concepts like Artificial Intelligence to expedite medical science and clinical trial and pharmacovigilance attain success.

How can natural language processing help model informed drug development?: a review.

ObjectiveTo summarize applications of natural language processing (NLP) in model informed drug development (MIDD) and identify potential areas of improvement.Materials and MethodsPublications found on PubMed and Google Scholar, websites and GitHub repositories for NLP libraries and models. Publications describing applications of NLP in MIDD were reviewed. The applications were stratified into 3 stages: drug discovery, clinical trials, and pharmacovigilance. Key NLP functionalities used for these applications were assessed. Programming libraries and open-source resources for the implementation of NLP functionalities in MIDD were identified.ResultsNLP has been utilized to aid various processes in drug development lifecycle such as gene-disease mapping, biomarker discovery, patient-trial matching, adverse drug events detection, etc. These applications commonly use NLP functionalities of named entity recognition, word embeddings, entity resolution, assertion status detection, relation extraction, and topic modeling. The current state-of-the-art for implementing these functionalities in MIDD applications are transformer models that utilize transfer learning for enhanced performance. Various libraries in python, R, and Java like huggingface, sparkNLP, and KoRpus as well as open-source platforms such as DisGeNet, DeepEnroll, and Transmol have enabled convenient implementation of NLP models to MIDD applications.DiscussionChallenges such as reproducibility, explainability, fairness, limited data, limited language-support, and security need to be overcome to ensure wider adoption of NLP in MIDD landscape. There are opportunities to improve the performance of existing models and expand the use of NLP in newer areas of MIDD.ConclusionsThis review provides an overview of the potential and pitfalls of current NLP approaches in MIDD.

Application of Machine Learning in Drug Discovery and Development Lifecycle

Machine learning and Artificial Intelligence have significantly advanced in recent years owing to their potential to considerably increase the quality of life while reducing human workload. The paper demonstrates how AI and ML are used in the drug development process to shorten and enhance the overall timeline. It contains pertinent information on a variety of Machine Learning approaches and algorithms that are used across the whole drug development process to speed up research, save expenses, and reduce risks related to clinical trials. A range of QSAR analysis, hit finding, and de novo drug design applications are used in the pharmaceutical industry to enhance decision-making. As technologies like high-throughput screening and computation analysis of databases used for lead and target identification and development create and integrate vast volumes of data, machine learning and deep learning have grown in importance. It has also been emphasized how these cognitive models and tools may be used in lead creation, optimization, and thorough virtual screening. In this paper, problem statements and the corresponding state-of-the-art models have been considered for target validation, prognostic biomarkers, and digital pathology. Machine Learning models play a vital role in the various operations related to clinical trials embracing protocol optimization, participant management, data analysis and storage, clinical trial data verification, and surveillance. Post-development drug monitoring and unique industrially prevalent ML applications of pharmacovigilance have also been discussed. As a result, the goal of this study is to investigate the machine learning and deep learning algorithms utilised across the drug development lifecycle as well as the supporting techniques that have the potential to be useful.

Use of Real-World Data and Evidence in Drug Development of Medicinal Products Centrally Authorized in Europe in 2018-2019.

Real‐world data/real‐world evidence (RWD/RWE) are considered to have a great potential to complement, in some cases, replace the evidence generated through randomized controlled trials. By tradition, use of RWD/RWE in the postauthorization phase is well‐known, whereas published evidence of use in the pre‐authorization phase of medicines development is lacking. The primary aim of this study was to identify and quantify the role of potential use of RWD/RWE (RWE signatures) during the pre‐authorization phase, as presented in the initial marketing authorization applications of new medicines centrally evaluated with a positive opinion in 2018–2019 (n = 111) by the European Medicines Agency (EMA). Data for the study was retrieved from the evaluation overviews of the European Public Assessment Reports (EPARs), which reflect the scientific conclusions of the assessment process and are accessible through the EMA website. RWE signatures were extracted into an RWE Data Matrix, including 11 categories divided over 5 stages of the drug development lifecycle. Nearly all EPARs included RWE signatures for the discovery (98.2%) and life‐cycle management (100.0%). Half of them included RWE signatures for the full development phase (48.6%) and for supporting regulatory decisions at the registration (46.8%), whereas over a third (35.1%) included RWE signatures for the early development. RWE signatures were more often seen for orphan and conditionally approved medicines. Oncology, hematology, and anti‐infectives stood out as therapeutic areas with most RWE signatures in their full development phase. The findings bring unprecedented insights about the vast use of RWD/RWE in drug development supporting the regulatory decision making.

Evaluation of Drug Lags in Development Initiation, New Drug Application and Approval Between Japan and the USA and the Impact of Local Versus Multi-regional Clinical Trials.

Drug time lags occur between the date that new drugs are first approved, often in the USA, and approval is granted in other countries. Multi-regional clinical trials (MRCTs) are a key strategy for simultaneous global development and regulatory submission of new drugs. However, no studies have evaluated the impact of MRCT versus local development on key time points in the drug development lifecycle between the USA and Japan. It is important for pharmaceutical companies planning drug development in Japan to understand when they can start development, when they can catch up in case of development initiation delay, length of time the development period might take, and amount of time that market exclusivity is lost, if Japan does not participate in the MRCT. The aim of this study was to investigate differences in drug lag in development initiation, New Drug Application (NDA) submission and drug approval, as well as differences in the development and review periods, by local trials and MRCTs between Japan and the USA. We also assessed the advantages and disadvantages of MRCTs for these lags in Japan. We analyzed drug approvals in Japan between 2016 and 2020 and divided them into local and MRCT groups. Lags in development initiation, NDA submission, and approval of new drugs were calculated by subtracting each date in Japan from the corresponding date in the USA. Our study period was divided into three periods based on the International Conference on Harmonization (ICH) E17 guideline, published in 2017, and the guideline for the Phase I trials in the Japanese population prior to MRCTs, published in 2014. In addition, subgroup analyses by therapeutic area, regulatory background, modality, capital style, and sales ranking (2020) were conducted. We analyzed 174 approvals in Japan and the USA. The differences in the drug lags for development initiation, NDA submission, and approval between the local and MRCT groups were 4.9, 3.5, and 3.2 years, respectively. All three lag times were shorter for the MRCT group than the local group. A development initiation lag in the local group has expanded since publication of the guidelines. For the people of Japan, important drug lags were identified in development initiation, NDA submission, and drug approval dates between local trials and MRCTs that include Japan. It is difficult to recover fully from the delay caused by local development, and it is important to understand the further expansion of drug lags, in cases where Japan is not involved in the MRCT.