本文将选择性分类与 HINT 结合用于临床试验结果预测，通过对低置信样本弃权（不执行）来量化不确定性，显著提升各期试验的 PR-AUC 等指标。
摘要：
在机器学习的各个领域中，不确定性量化的重要性正日益得到认可。准确评估模型预测的不确定性，有助于为研究人员和从业者提供更深入的理解与信心。这一点在医学诊断和药物发现领域尤为关键，因为可靠的预测会直接影响研究质量和患者健康。在本文中，我们提出将不确定性量化纳入临床试验结果预测。我们的主要目标是增强模型辨别细微差异的能力，从而显著提升其整体性能。为实现这一目标，我们采用了选择性分类方法，并将其与处于临床试验预测建模前沿的 Hierarchical Interaction Network（HINT，层次交互网络）无缝集成。选择性分类涵盖一系列不确定性量化方法，使模型在面对被标记为模糊或低置信度的样本时能够弃权（不执行）决策，从而提高其选择分类的实例的预测准确率。一系列综合实验表明，将选择性分类纳入临床试验预测可显著增强模型性能，PR-AUC、F1、ROC-AUC 和总体准确率等关键指标均有明显提升。具体而言，在 I 期、II 期和 III 期试验结果预测中，所提方法相对于基础模型（HINT）在 PR-AUC 上分别实现了 32.37%、21.43% 和 13.27% 的相对提升。在预测 III 期时，我们的方法达到了 0.9022 的 PR-AUC 分数。这些发现说明了该策略在临床试验预测领域中的稳健性和潜在效用，并有可能在该领域设立新的基准。
英文摘要：
The importance of uncertainty quantification is increasingly recognized in the diverse field of machine learning. Accurately assessing model prediction uncertainty can help provide deeper understanding and confidence for researchers and practitioners. This is especially critical in medical diagnosis and drug discovery areas, where reliable predictions directly impact research quality and patient health.
In this paper, we proposed incorporating uncertainty quantification into clinical trial outcome predictions. Our main goal is to enhance the model's ability to discern nuanced differences, thereby significantly improving its overall performance.
We have adopted a selective classification approach to fulfill our objective, integrating it seamlessly with the Hierarchical Interaction Network (HINT), which is at the forefront of clinical trial prediction modeling. Selective classification, encompassing a spectrum of methods for uncertainty quantification, empowers the model to withhold decision-making in the face of samples marked by ambiguity or low confidence, thereby amplifying the accuracy of predictions for the instances it chooses to classify. A series of comprehensive experiments demonstrate that incorporating selective classification into clinical trial predictions markedly enhances the model's performance, as evidenced by significant upticks in pivotal metrics such as PR-AUC, F1, ROC-AUC, and overall accuracy.
Specifically, the proposed method achieved 32.37\%, 21.43\%, and 13.27\% relative improvement on PR-AUC over the base model (HINT) in phase I, II, and III trial outcome prediction, respectively. When predicting phase III, our method reaches 0.9022 PR-AUC scores.
These findings illustrate the robustness and prospective utility of this strategy within the area of clinical trial predictions, potentially setting a new benchmark in the field.
University of Cambridge（剑桥大学）；Yonsei University（延世大学）；University of Virginia（弗吉尼亚大学）；University of Oxford（牛津大学）；UCLA（加利福尼亚大学洛杉矶分校）
Clinical trials in the medical domain are constrained by budgets. The number of patients that can be recruited is therefore limited. When a patient population is heterogeneous, this creates difficulties in learning subgroup specific responses to a particular drug and especially for a variety of dosages. In addition, patient recruitment can be difficult by the fact that clinical trials do not aim to provide a benefit to any given patient in the trial. In this paper, we propose C3T-Budget, a contextual constrained clinical trial algorithm for dose-finding under both budget and safety constraints…
Significant advancements have been made in recent years to optimize patient recruitment for clinical trials, however, improved methods for patient recruitment prediction are needed to support trial site selection and to estimate appropriate enrollment timelines in the trial design stage. In this paper, using data from thousands of historical clinical trials, we explore machine learning methods to predict the number of patients enrolled per month at a clinical trial site over the course of a trial's enrollment duration. We show that these methods can reduce the error that is observed with curre…
Interventional cancer clinical trials are generally too restrictive, and some patients are often excluded on the basis of comorbidity, past or concomitant treatments, or the fact that they are over a certain age. The efficacy and safety of new treatments for patients with these characteristics are, therefore, not defined. In this work, we built a model to automatically predict whether short clinical statements were considered inclusion or exclusion criteria. We used protocols from cancer clinical trials that were available in public registries from the last 18 years to train word-embeddings, a…
A major impediment to successful drug development is the complexity, cost, and scale of clinical trials. The detailed internal structure of clinical trial data can make conventional optimization difficult to achieve. Recent advances in machine learning, specifically graph-structured data analysis, have the potential to enable significant progress in improving the clinical trial design. TrialGraph seeks to apply these methodologies to produce a proof-of-concept framework for developing models which can aid drug development and benefit patients. In this work, we first introduce a curated clinica…
University of Wisconsin School of Medicine and Public Health（威斯康星大学医学院与公共卫生学院）
In the past decade, there has been much discussion about the issue of biased reporting in clinical research. Despite this attention, there have been limited tools developed for the systematic assessment of qualitative statements made in clinical research, with most studies assessing qualitative statements relying on the use of manual expert raters, which limits their size. Also, previous attempts to develop larger scale tools, such as those using natural language processing, were limited by both their accuracy and the number of categories used for the classification of their findings. With the…
University of Cambridge（剑桥大学）；Newcastle University（纽卡斯尔大学）
High-dimensional biomarkers such as genomics are increasingly being measured in randomized clinical trials. Consequently, there is a growing interest in developing methods that improve the power to detect biomarker-treatment interactions. We adapt recently proposed two-stage interaction detecting procedures in the setting of randomized clinical trials. We also propose a new stage 1 multivariate screening strategy using ridge regression to account for correlations among biomarkers. For this multivariate screening, we prove the asymptotic between-stage independence, required for family-wise erro…
University of Virginia（弗吉尼亚大学）；University of Pennsylvania（宾夕法尼亚大学）；University of Cambridge（剑桥大学）；University of California, Los Angeles（加利福尼亚大学洛杉矶分校）
Phase I dose-finding trials are increasingly challenging as the relationship between efficacy and toxicity of new compounds (or combination of them) becomes more complex. Despite this, most commonly used methods in practice focus on identifying a Maximum Tolerated Dose (MTD) by learning only from toxicity events. We present a novel adaptive clinical trial methodology, called Safe Efficacy Exploration Dose Allocation (SEEDA), that aims at maximizing the cumulative efficacies while satisfying the toxicity safety constraint with high probability. We evaluate performance objectives that have opera…
This paper proposes a new method for analysis of adverse event data in clinical trials. The method is illustrated by application to data on 4 phase III clinical trials, two on breast cancer and two on diabetes mellitus.
Clinical trials are considered as the golden standard for medical device validation. However, many sacrifices have to be made during the design and conduction of the trials due to cost considerations and partial information, which may compromise the significance of the trial results. In this paper, we proposed a model-based pre-clinical trial framework using statistical model checking. Physiological models represent disease mechanism, which enable automated adjudication of simulation results. Sampling of the patient parameters and hypothesis testing are performed by statistical model checker. …
Existing trials had not taken enough consideration of their population representativeness, which can lower the effectiveness when the treatment is applied in real-world clinical practice. We analyzed the eligibility criteria of Bevacizumab colorectal cancer treatment trials, assessed their a priori generalizability, and examined how it affects patient outcomes when applied in real-world clinical settings. To do so, we extracted patient-level data from a large collection of electronic health records (EHRs) from the OneFlorida consortium. We built a zero-inflated negative binomial model using a …
The classification procedure of streaming data usually requires various ad hoc methods or particular heuristic models. We explore a novel non-parametric and systematic approach to analysis of heterogeneous sequential data. We demonstrate an application of this method to classification of the delays in responding to the prompts, from subjects with bipolar disorder collected during a clinical trial, using both synthetic and real examples. We show how this method can provide a natural and systematic way to extract characteristic features from sequential data.
Detection of interactions between treatment effects and patient descriptors in clinical trials is critical for optimizing the drug development process. The increasing volume of data accumulated in clinical trials provides a unique opportunity to discover new biomarkers and further the goal of personalized medicine, but it also requires innovative robust biomarker detection methods capable of detecting non-linear, and sometimes weak, signals. We propose a set of novel univariate statistical tests, based on the theory of random walks, which are able to capture non-linear and non-monotonic covari…
Bayesian adaptive designs have gained popularity in all phases of clinical trials with numerous new developments in the past few decades. During the COVID-19 pandemic, the need to establish evidence for the effectiveness of vaccines, therapeutic treatments and policies that could resolve or control the crisis emphasized the advantages offered by efficient and flexible clinical trial designs. In many COVID-19 clinical trials, due to the high level of uncertainty, Bayesian adaptive designs were considered advantageous. Designing Bayesian adaptive trials, however, requires extensive simulation st…
Effective decision making from randomised controlled clinical trials relies on robust interpretation of the numerical results. However, the language we use to describe clinical trials can cause confusion both in trial design and in comparing results across trials. ACceptability Curve Estimation using Probability Above Threshold (ACCEPT) aids comparison between trials (even where of different designs) by harmonising reporting of results, acknowledging different interpretations of the results may be valid in different situations, and moving the focus from comparison to a pre-specified value to i…
TWIN-GPT: Digital Twins for Clinical Trials via Large Language Model
School of Public Health, Zhejiang University（浙江大学公共卫生学院）；Rensselaer Polytechnic Institute（伦斯勒理工学院）；University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
A sufficient condition for the uniqueness of multinomial sequential unbiased estimators is provided generalizing a classical result for binomial samples. Unbiased estimators are applied to infer the parameters of multidimensional or multinomial Random Walks which are observed until they reach a boundary. An application to clinical trials is presented.
In many clinical trials, individuals in different subgroups have experience differential treatment effects. This leads to individualized differences in treatment benefit. In this article, we introduce the general concept of predictive directions, which are risk scores motivated by potential outcomes considerations. These techniques borrow heavily from sufficient dimension reduction (SDR) and causal inference methodology. Under some conditions, one can use existing methods from the SDR literature to estimate the directions assuming an idealized complete data structure, which subsequently yields…
The COVID-19 pandemic continues to affect the conduct of clinical trials globally. Complications may arise from pandemic-related operational challenges such as site closures, travel limitations and interruptions to the supply chain for the investigational product, or from health-related challenges such as COVID-19 infections. Some of these complications lead to unforeseen intercurrent events in the sense that they affect either the interpretation or the existence of the measurements associated with the clinical question of interest. In this article, we demonstrate how the ICH E9(R1) Addendum o…
Individualized treatment rules (ITRs) tailor treatments according to individual patient characteristics. They can significantly improve patient care and are thus becoming increasingly popular. The data collected during randomized clinical trials are often used to estimate the optimal ITRs. However, these trials are generally expensive to run, and, moreover, they are not designed to efficiently estimate ITRs. In this paper, we propose a cost-effective estimation method from an active learning perspective. In particular, our method recruits only the "most informative" patients (in terms of learn…
The analysis and planning methods for competing risks model have been described in the literatures in recent decades, and non-inferiority clinical trials are helpful in current pharmaceutical practice. Analytical methods for non-inferiority clinical trials in the presence of competing risks were investigated by Parpia et al., who indicated that the proportional sub-distribution hazard model is appropriate in the context of biological studies. However, the analytical methods of competing risks model differ from those appropriate for analyzing non-inferiority clinical trials with a single outcom…
Phase III randomized clinical trials play a monumentally critical role in the evaluation of new medical products. Because of the intrinsic nature of uncertainty embedded in our capability in assessing the efficacy of a medical product, interpretation of trial results relies on statistical principles to control the error of false positives below desirable level. The well-established statistical hypothesis testing procedure suffers from two major limitations, namely, the lack of flexibility in the thresholds to claim success and the lack of capability of controlling the total number of false pos…
A new approach to adaptive design of clinical trials is proposed in a general multiparameter exponential family setting, based on generalized likelihood ratio statistics and optimal sequential testing theory. These designs are easy to implement, maintain the prescribed Type I error probability, and are asymptotically efficient. Practical issues involved in clinical trials allowing mid-course adaptation and the large literature on this subject are discussed, and comparisons between the proposed and existing designs are presented in extensive simulation studies of their finite-sample performance…
One of Pranab K. Sen's major research areas is sequential nonparametrics and semiparametrics and their applications to clinical trials, to which he has made many important contributions. Herein we review a number of these contributions and related developments. We also describe some recent work on nonparametric and semiparametric inference and the associated computational methods in time-sequential clinical trials with survival endpoints.
Randomization is a common technique used in clinical trials to eliminate potential bias and confounders in a patient population. Equal allocation to treatment groups is the standard due to its optimal efficiency in many cases. However, in certain scenarios, unequal allocation can improve efficiency. In superiority trials with more than two groups, the optimal randomization is not always a balanced randomization. In non-inferiority trials, additive margin with equal variance is the only instance with balanced randomization. Optimal randomization for non-inferiority trials can be far from 1:1 an…
Most clinical trials involve the comparison of a new treatment to a control arm (e.g., the standard of care) and the estimation of a treatment effect. External data, including historical clinical trial data and real-world observational data, are commonly available for the control arm. Borrowing information from external data holds the promise of improving the estimation of relevant parameters and increasing the power of detecting a treatment effect if it exists. In this paper, we propose to use Bayesian additive regression trees (BART) for incorporating external data into the analysis of clini…
[Invited Discussion] Randomization Tests to Address Disruptions in Clinical Trials: A Report from the NISS Ingram Olkin Forum Series on Unplanned Clinical Trial Disruptions
Confusion often arises when attempting to articulate target estimand(s) of a clinical trial in plain language. We aim to rectify this confusion by using a type of causal graph called the Single-World Intervention Graph (SWIG) to provide a visual representation of the estimand that can be effectively communicated to interdisciplinary stakeholders. These graphs not only display estimands, but also illustrate the assumptions under which a causal estimand is identifiable by presenting the graphical relationships between the treatment, intercurrent events, and clinical outcomes. To demonstrate its …
Adaptive designs for clinical trials permit alterations to a study in response to accumulating data in order to make trials more flexible, ethical and efficient. These benefits are achieved while preserving the integrity and validity of the trial, through the pre-specification and proper adjustment for the possible alterations during the course of the trial. Despite much research in the statistical literature highlighting the potential advantages of adaptive designs over traditional fixed designs, the uptake of such methods in clinical research has been slow. One major reason for this is that …
Conditional (European Medicines Agency) or accelerated (U.S. Food and Drug Administration) approval of drugs allow earlier access to promising new treatments that address unmet medical needs. Certain post-marketing requirements must typically be met in order to obtain full approval, such as conducting a new post-market clinical trial. We study the applicability of the recently developed harmonic mean Chi-squared test to this conditional or accelerated approval framework. The proposed approach can be used both to support the design of the post-market trial and the analysis of the combined evide…
Recently, the strategy for dose optimization in oncology has shifted to conduct Phase 2 randomized controlled trials with multiple doses. Optimal biologic dose selection from Phase 1 trial data to determine candidate doses for Phase 2 trials has been gaining attention. This study proposes a novel adaptive cohort size determination method for optimal biologic dose-finding to accelerate trials. The cohort size expansion is determined adaptively depending on the toxicity and efficacy data of the ongoing trial. In a simulation, the proposed method shortened the trial duration and maintained accura…
Historical data from previous clinical trials, observational studies and health records may be utilized in analysis of clinical trials data to strengthen inference. Under the Bayesian framework incorporation of information obtained from any source other than the current data is facilitated through construction of an informative prior. The existing methodology for defining an informative prior based on historical data relies on measuring similarity to the current data at the study level and does not take advantage of individual patient data (IPD). This paper proposes a family of priors that uti…
Suppose we are interested in the effect of a treatment in a clinical trial. The efficiency of inference may be limited due to small sample size. However, external control data are often available from historical studies. Motivated by an application to Helicobacter pylori infection, we show how to borrow strength from such data to improve efficiency of inference in the clinical trial. Under an exchangeability assumption about the potential outcome mean, we show that the semiparametric efficiency bound for estimating the average treatment effect can be reduced by incorporating both the clinical …
The recent 21st Century Cures Act propagates innovations to accelerate the discovery, development, and delivery of 21st century cures. It includes the broader application of Bayesian statistics and the use of evidence from clinical expertise. An example of the latter is the use of trial-external (or historical) data, which promises more efficient or ethical trial designs. We propose a Bayesian meta-analytic approach to leveraging historical data for time-to-event endpoints, which are common in oncology and cardiovascular diseases. The approach is based on a robust hierarchical model for piecew…
In a group sequential clinical trial, accumulated data are analysed at numerous time-points in order to allow early decisions about a hypothesis of interest. These designs have historically been recommended for their ethical, administrative and economic benefits. In this work, we discuss a collection of new Stata commands for computing the stopping boundaries and required group size of various classical group sequential designs, assuming a normally distributed outcome variable. Following this, we demonstrate how the performance of several designs can be compared graphically.
The question of how individual patient data from cohort studies or historical clinical trials can be leveraged for designing more powerful, or smaller yet equally powerful, clinical trials becomes increasingly important in the era of digitalisation. Today, the traditional statistical analyses approaches may seem questionable to practitioners in light of ubiquitous historical covariate information. Several methodological developments aim at incorporating historical information in the design and analysis of future clinical trials, most importantly Bayesian information borrowing, propensity score…
Multi-arm designs provide an effective means of evaluating several treatments within the same clinical trial. Given the large number of treatments now available for testing in many disease areas, it has been argued that their utilisation should increase. However, for any given clinical trial there are numerous possible multi-arm designs that could be used, and choosing between them can be a difficult task. This task is complicated further by a lack of available easy-to-use software for designing multi-arm trials. To aid the wider implementation of multi-arm clinical trial designs, we have deve…
Clinical trials disruption has always represented a non negligible part of the ending of interventional studies. While the SARS-CoV-2 (COVID-19) pandemic has led to an impressive and unprecedented initiation of clinical research, it has also led to considerable disruption of clinical trials in other disease areas, with around 80% of non-COVID-19 trials stopped or interrupted during the pandemic. In many cases the disrupted trials will not have the planned statistical power necessary to yield interpretable results. This paper describes methods to compensate for the information loss arising from…
Defining the Inclusion/Exclusion (I/E) criteria of a trial is one of the most important steps during a trial design. Increasingly complex I/E criteria potentially create information imbalance and transparency issues between the people who design and run the trials and those who consume the information produced by the trials. In order to better understand and quantify the impact of a category of I/E criteria on observed treatment effects, a concept, named the Selection Induced Contrast Estimate (SICE) effect, is introduced and formulated in this paper. The SICE effect can exist in controlled cl…
Rheumatoid arthritis clinical trials are strategically designed to collect the disease activity score of each patient over multiple clinical visits, meanwhile a patient may drop out before their intended completion due to various reasons. The dropout terminates the longitudinal data collection on the patients activity score. In the presence of informative dropout, that is, the dropout depends on latent variables from the longitudinal process, simply applying a model to analyze the longitudinal outcomes may lead to biased results because the assumption of random dropout is violated. In this pap…
Various adaptive randomization procedures (adaptive designs) have been proposed to clinical trials. This paper discusses several broad families of procedures, such as the play-the-winner rule and Markov chain model, randomized play-the-winner rule and urn models, drop-the-loser rule, doubly biased coin adaptive design. Asymptotic theories are presented with several pivotal proofs. The effect of delayed responses, the power and variability comparison of these designs are also discussed.
Comparative effectiveness evidence from randomized trials may not be directly generalizable to a target population of substantive interest when, as in most cases, trial participants are not randomly sampled from the target population. Motivated by the need to generalize evidence from two trials conducted in the AIDS Clinical Trials Group (ACTG), we consider weighting, regression and doubly robust estimators to estimate the causal effects of HIV interventions in a specified population of people living with HIV in the USA. We focus on a non-nested trial design and discuss strategies for both poi…
Generalizability and transportability methods have been proposed to address the external validity bias of randomized clinical trials that results from differences in the distribution of treatment effect modifiers between trial and target populations. However, such studies present many challenges. We review and summarize state-of-the-art methodological considerations. We additionally provide investigators with a step-by-step guide to address these challenges, illustrated through a published case study. When conducted with rigor, such studies may play an integral role in regulatory decisions by …
