Improving the clinical meaning of surrogate endpoints: An empirical assessment of clinical progression in phase III oncology trials.

Disease progression in clinical trials is commonly defined by radiologic measures. However, clinical progression may be more meaningful to patients, may occur even when radiologic criteria for progression are not met, and often requires a change in therapy in clinical practice. The objective of this study was to determine the utilization of clinical progression criteria within progression-based trial endpoints among phase III trials testing systemic therapies for metastatic solid tumors. The primary manuscripts and protocols of phase III trials were reviewed for whether clinical events, such as refractory pain, tumor bleeding, or neurologic compromise, could constitute a progression event. Univariable logistic regression computed odds ratios (OR) and 95% CI for associations between trial-level covariates and clinical progression. A total of 216 trials enrolling 148,190 patients were included, with publication dates from 2006 through 2020. A major change in clinical status was included in the progression criteria of 13% of trials (n = 27), most commonly as a secondary endpoint (n = 22). Only 59% of trials (n = 16) reported distinct clinical progression outcomes that constituted the composite surrogate endpoint. Compared with other disease sites, genitourinary trials were more likely to include clinical progression definitions (16/33 [48%] vs. 11/183 [6%]; OR, 14.72; 95% CI, 5.99 to 37.84; p < .0001). While major tumor-related clinical events were seldom considered as disease progression events, increased attention to clinical progression may improve the meaningfulness and clinical applicability of surrogate endpoints for patients with metastatic solid tumors.

Use of win time for ordered composite endpoints in clinical trials.

Consider the choice of outcome for overall treatment benefit in a clinical trial which measures the first time to each of several clinical events. We describe several new variants of the win ratio that incorporate the time spent in each clinical state over the common follow-up, where clinical state means the worst clinical event that has occurred by that time. One version allows restriction so that death during follow-up is most important, while time spent in other clinical states is still accounted for. Three other variants are described; one is based on the average pairwise win time, one creates a continuous outcome for each participant based on expected win times against a reference distribution and another that uses the estimated distributions of clinical state to compare the treatment arms. Finally, a combination testing approach is described to give robust power for detecting treatment benefit across a broad range of alternatives. These new methods are designed to be closer to the overall treatment benefit/harm from a patient's perspective, compared to the ordinary win ratio. The new methods are compared to the composite event approach and the ordinary win ratio. Simulations show that when overall treatment benefit on death is substantial, the variants based on either the participants' expected win times (EWTs) against a reference distribution or estimated clinical state distributions have substantially higher power than either the pairwise comparison or composite event methods. The methods are illustrated by re-analysis of the trial heart failure: a controlled trial investigating outcomes of exercise training.

An information-theoretic approach for the assessment of a continuous outcome as a surrogate for a binary true endpoint based on causal inference: Application to vaccine evaluation.

Within the causal association paradigm, a method is proposed to assess the validity of a continuous outcome as a surrogate for a binary true endpoint. The methodology is based on a previously introduced information-theoretic definition of surrogacy and has two main steps. In the first step, a new model is proposed to describe the joint distribution of the potential outcomes associated with the putative surrogate and the true endpoint of interest. The identifiability issues inherent to this type of models are handled via sensitivity analysis. In the second step, a metric of surrogacy new to this setting, the so-called individual causal association is presented. The methodology is studied in detail using theoretical considerations, some simulations, and data from a randomized clinical trial evaluating an inactivated quadrivalent influenza vaccine. A user-friendly R package Surrogate is provided to carry out the evaluation exercise.

Familywise error for multiple time-to-event endpoints in a group sequential design.

We investigate the familywise error rate (FWER) for time-to-event endpoints evaluated using a group sequential design with a hierarchical testing procedure for secondary endpoints. We show that, in this setup, the correlation between the log-rank test statistics at interim and at end of study is not congruent with the canonical correlation derived for normal-distributed endpoints. We show, both theoretically and by simulation, that the correlation also depends on the level of censoring, the hazard rates of the endpoints, and the hazard ratio. To optimize operating characteristics in this complex scenario, we propose a simulation-based method to assess the FWER which, better than the alpha-spending approach, can inform the choice of critical values for testing secondary endpoints.

Design of randomized clinical trials with a binary endpoint: Conditional versus unconditional analyses of a two-by-two table.

When designing a randomized clinical trial to compare two treatments, the sample size required to have desired power with a specified type 1 error depends on the hypothesis testing procedure. With a binary endpoint (e.g., response), the trial results can be displayed in a 2 × 2 table. If one does the analysis conditional on the number of positive responses, then using Fisher's exact test has an actual type 1 error less than or equal to the specified nominal type 1 error. Alternatively, one can use one of many unconditional "exact" tests that also preserve the type 1 error and are less conservative than Fisher's exact test. In particular, the unconditional test of Boschloo is always at least as powerful as Fisher's exact test, leading to smaller required sample sizes for clinical trials. However, many statisticians have argued over the years that the conditional analysis with Fisher's exact test is the only appropriate procedure. Since having smaller clinical trials is an extremely important consideration, we review the general arguments given for the conditional analysis of a 2 × 2 table in the context of a randomized clinical trial. We find the arguments not relevant in this context, or, if relevant, not completely convincing, suggesting the sample-size advantage of the unconditional tests should lead to their recommended use. We also briefly suggest that since designers of clinical trials practically always have target null and alternative response rates, there is the possibility of using this information to improve the power of the unconditional tests.

The impact of allocation bias on test decisions in clinical trials with multiple endpoints using multiple testing strategies.

BACKGROUND: Considering multiple endpoints in clinical trials provide a more comprehensive understanding of treatment effects and may lead to increased power or reduced sample size, which may be beneficial in rare diseases. Besides the small sample sizes, allocation bias is an issue that affects the validity of these trials. We investigate the impact of allocation bias on testing decisions in clinical trials with multiple endpoints and offer a tool for selecting an appropriate randomization procedure (RP). METHODS: We derive a model for quantifying the effect of allocation bias depending on the RP in the case of two-arm parallel group trials with continuous multiple endpoints. We focus on two approaches to analyze multiple endpoints, either the Sidák procedure to show efficacy in at least one endpoint and the all-or-none procedure to show efficacy in all endpoints. RESULTS: To evaluate the impact of allocation bias on the test decision we propose a biasing policy for multiple endpoints. The impact of allocation on the test decision is measured by the family-wise error rate of the Sidák procedure and the type I error rate of the all-or-none procedure. Using the biasing policy we derive formulas to calculate these error rates. In simulations we show that, for the Sidák procedure as well as for the all-or-none procedure, allocation bias leads to inflation of the mean family-wise error and mean type I error, respectively. The strength of this inflation is affected by the choice of the RP. CONCLUSION: Allocation bias should be considered during the design phase of a trial to increase validity. The developed methodology is useful for selecting an appropriate RP for a clinical trial with multiple endpoints to minimize allocation bias effects.

A note on familywise error rate for a primary and secondary endpoint.

Hung et al. (2007) considered the problem of controlling the type I error rate for a primary and secondary endpoint in a clinical trial using a gatekeeping approach in which the secondary endpoint is tested only if the primary endpoint crosses its monitoring boundary. They considered a two-look trial and showed by simulation that the naive method of testing the secondary endpoint at full level α at the time the primary endpoint reaches statistical significance does not control the familywise error rate at level α. Tamhane et al. (2010) derived analytic expressions for familywise error rate and power and confirmed the inflated error rate of the naive approach. Nonetheless, many people mistakenly believe that the closure principle can be used to prove that the naive procedure controls the familywise error rate. The purpose of this note is to explain in greater detail why there is a problem with the naive approach and show that the degree of alpha inflation can be as high as that of unadjusted monitoring of a single endpoint.

Design of paediatric trials with benefit-risk endpoints using a composite score of adverse events of interest (AEI) and win-statistics.

A fundamental problem in the regulatory evaluation of a therapy is assessing whether the benefit outweighs the associated risks. This work proposes designing a trial that assesses a composite endpoint consisting of benefit and risk, hence, making the core of the design of the study, to assess benefit and risk. The proposed benefit risk measure consists of efficacy measure(s) and a risk measure that is based on a composite score obtained from pre-defined adverse events of interest (AEI). This composite score incorporates full aspects of adverse events of interest (i.e. the incidence, severity, and duration of the events). We call this newly proposed score the AEI composite score. After specifying the priorities between the components of the composite endpoint, a win-statistic (i.e. win ratio, win odds, or net benefit) is used to assess the difference between treatments in this composite endpoint. The power and sample size requirements of such a trial design are explored via simulation. Finally, using Dupixent published adult study results, we show how we can design a paediatric trial where the primary outcome is a composite of prioritized outcomes consisting of efficacy endpoints and the AEI composite score endpoint. The resulting trial design can potentially substantially reduce sample size compared to a trial designed to assess the co-primary efficacy endpoints, therefore it may address the challenge of slow enrollment and patient availability for paediatric studies.

Development of Novel, Value-Based, Digital Endpoints for Clinical Trials: A Structured Approach Toward Fit-for-Purpose Validation.

Novel digital endpoints gathered via wearables, small devices, or algorithms hold great promise for clinical trials. However, implementation has been slow because of a lack of guidelines regarding the validation process of these new measurements. In this paper, we propose a pragmatic approach toward selection and fit-for-purpose validation of digital endpoints. Measurements should be value-based, meaning the measurements should directly measure or be associated with meaningful outcomes for patients. Devices should be assessed regarding technological validity. Most importantly, a rigorous clinical validation process should appraise the tolerability, difference between patients and controls, repeatability, detection of clinical events, and correlation with traditional endpoints. When technically and clinically fit-for-purpose, case building in interventional clinical trials starts to generate evidence regarding the response to new or existing health-care interventions. This process may lead to the digital endpoint replacing traditional endpoints, such as clinical rating scales or questionnaires in clinical trials. We recommend initiating more data-sharing collaborations to prevent unnecessary duplication of research and integration of value-based measurements in clinical care to enhance acceptance by health-care professionals. Finally, we invite researchers and regulators to adopt this approach to ensure a timely implementation of digital measurements and value-based thinking in clinical trial design and health care. SIGNIFICANCE STATEMENT: Novel digital endpoints are often cited as promising for the clinical trial of the future. However, clear validation guidelines are lacking in the literature. This paper contains pragmatic criteria for the selection, technical validation, and clinical validation of novel digital endpoints and provides recommendations for future work and collaboration.

The development of a core outcome set for clinical effectiveness studies of bordered foam dressings in the treatment of complex wounds.

AIM: The aim of this project was to develop a core outcome set (COS) for clinical effectiveness studies of bordered foam dressings in the treatment of complex wounds. METHODS: The research project followed the Core Outcome Measures in Effectiveness Trials (COMET) initiative and consisted of two phases. The first phase prepared the background and process, while the second phase had three steps: outcome list generation via systematic review and qualitative study, Delphi consensus study, and consensus meeting. The study has been registered in the Core Outcome Measures in Effectiveness Trials database. RESULTS: The systematic review resulted in 82 outcomes and 20 additional outcomes were obtained during the interviews. After refinement, 111 panellists from 23 countries rated a list of 51 outcomes. In the following consensus meeting, six outcomes were prioritized to be included in the core outcome set. After the consensus meeting, a patient-reported outcome was added to the core outcome set. CONCLUSION: The COS for evaluating the effectiveness of bordered foam dressings in treating complex wounds includes 7 outcomes: "ability to stay in place", "leakage", "pain", "dressing related periwound skin changes", "change in wound size over time", and "overall satisfaction". These identified outcomes are correlated with contemporary bioengineering testing and evaluation methods for dressing performance, which underpins the need for a close multidisciplinary collaboration to advance the field of wound dressings. The outcome 'overall satisfaction' reflects the impact of complex wounds and their treatment on a patient's daily life. The use of these outcomes is recommended to improve data synthesis and promote evidence-based practice. Future developments in COS development involve creating measurement instruments and relevant endpoints for these outcomes.

Core outcomes for pressure ulcer prevention trials: results of an international consensus study: Classification: Outcomes and qualitative research.

BACKGROUND: There is substantial heterogeneity between trial outcomes in pressure ulcer prevention research. The development of core outcome sets is one strategy to improve comparability between trial results and thus increase the quality of evidence. OBJECTIVES: To identify core outcomes for pressure ulcer prevention trials. METHODS: A workshop was held with service users to discuss their views and understanding of the outcomes identified by a scoping review and to identify any missing outcomes. In a next step, a Delphi survey comprising three rounds was conducted to evaluate a compiled list of outcomes by their importance. Afterwards the preselection from the Delphi survey was discussed in a virtual consensus meeting with the aim of agreeing on a final set of core outcomes. Individuals who had completed all three rounds of the Delphi survey were eligible to participate in this meeting. Participants included practitioners, service users, researchers and industry representatives. The OUTPUTs project is registered in the COMET database and is part of the Cochrane Skin Core Outcome Set Initiative. RESULTS: The workshop did not reveal any missing outcomes, but highlighted the need for further efforts to make lay people understand what an outcome is in a study setting. The Delphi survey took place between December 2020 and June 2021. After the three rounds, 18 out of 37 presented outcomes were rated to be critically important. In the following consensus meeting, six outcomes were prioritized to be included in the core outcome set for pressure ulcer prevention trials: (i) pressure ulcer occurrence; (ii) pressure ulcer precursor signs and symptoms; (iii) mobility; (iv) acceptability and comfort of intervention; (v) adherence/compliance; and (vi) adverse events/safety. CONCLUSIONS: Based on a comprehensive list of outcomes in pressure ulcer prevention research, there was clear agreement on the six identified core outcomes in three international Delphi rounds and in the consensus meeting. Although outcome measurement instruments need to be identified next, the six identified core outcomes should already be considered in future trials, as service users, practitioners, researchers and industry representatives have agreed that they are critically important. What is already known about this topic? There are numerous trials on pressure ulcer prevention, but evidence on the effectiveness of preventive measures is limited due to heterogeneity between trial outcomes. The development of a core outcome set is one strategy to improve comparability between trial results. What does this study add? A service user workshop, a three-round Delphi survey and an online consensus meeting with practitioners, service users, researchers and industry representatives were conducted to identify core outcomes for pressure ulcer prevention trials. Six core outcomes were defined: (i) pressure ulcer occurrence, (ii) pressure ulcer precursor signs and symptoms, (iii) mobility, (iv) acceptability and comfort of intervention, (v) adherence/compliance and (vi) adverse events/safety. What are the clinical implications of this work? Better evidence of interventions for pressure ulcer prevention will help health professionals and service users to decide which interventions are most appropriate and effective. Better evidence may contribute to better pressure ulcer prevention.

Sample size estimation using a latent variable model for mixed outcome co-primary, multiple primary and composite endpoints.

Mixed outcome endpoints that combine multiple continuous and discrete components are often employed as primary outcome measures in clinical trials. These may be in the form of co-primary endpoints, which conclude effectiveness overall if an effect occurs in all of the components, or multiple primary endpoints, which require an effect in at least one of the components. Alternatively, they may be combined to form composite endpoints, which reduce the outcomes to a one-dimensional endpoint. There are many advantages to joint modeling the individual outcomes, however in order to do this in practice we require techniques for sample size estimation. In this article we show how the latent variable model can be used to estimate the joint endpoints and propose hypotheses, power calculations and sample size estimation methods for each. We illustrate the techniques using a numerical example based on a four-dimensional endpoint and find that the sample size required for the co-primary endpoint is larger than that required for the individual endpoint with the smallest effect size. Conversely, the sample size required in the multiple primary case is similar to that needed for the outcome with the largest effect size. We show that the empirical power is achieved for each endpoint and that the FWER can be sufficiently controlled using a Bonferroni correction if the correlations between endpoints are less than 0.5. Otherwise, less conservative adjustments may be needed. We further illustrate empirically the efficiency gains that may be achieved in the composite endpoint setting.

Calculating power for the Finkelstein and Schoenfeld test statistic for a composite endpoint with two components.

The Finkelstein and Schoenfeld (FS) test is a popular generalized pairwise comparison approach to analyze prioritized composite endpoints (eg, components are assessed in order of clinical importance). Power and sample size estimation for the FS test, however, are generally done via simulation studies. This simulation approach can be extremely computationally burdensome, compounded by increasing number of composite endpoints and with increasing sample size. Here we propose an analytical solution to calculate power and sample size for commonly encountered two-component hierarchical composite endpoints. The power formulas are derived assuming underlying distributions in each of the component outcomes on the population level, which provide a computationally efficient and practical alternative to the standard simulation approach. Monte Carlo simulation results demonstrate that performance of the proposed power formulas are consistent with that of the simulation approach, and have generally desirable objective properties including robustness to mis-specified distributional assumptions. We demonstrate the application of the proposed formulas by calculating power and sample size for the Transthyretin Amyloidosis Cardiomyopathy Clinical Trial.

Weighted composite time to event endpoints with recurrent events: comparison of three analytical approaches.

BACKGROUND: In clinical trials the study interest often lies in the comparison of a treatment to a control regarding a time to event endpoint. A composite endpoint allows to consider several time to event endpoints at once. Usually, only the time to the first occurring event for a patient is thereby analyzed. However, an individual may experience more than one non-fatal event. Including all observed events in the analysis can increase the power and provides a more complete picture of the disease. Thus, analytical methods for recurrent events are required. A challenge is that the different event types belonging to the composite often are of different clinical relevance. In this case, weighting the event types according to their clinical relevance is an option. Different weight-based methods for composite time to event endpoints were proposed. So far, there exists no systematic comparison of these methods. METHODS: Within this work we provide a systematic comparison of three methods proposed for weighted composite endpoints in a recurrent event setting combining non-fatal and fatal events of different clinical relevance. We consider an extension of an approach proposed by Wei and Lachin, an approach by Rauch et al., and an approach by Bakal et al.. Comparison is done based on a simulation study and based on a clinical study example. RESULTS: For all three approaches closed formula test statistics are available. The Wei-Lachin approach and the approach by Rauch et al. show similar results in mean squared error. For the approach by Wei and Lachin confidence intervals are provided. The approach by Bakal et al. is not related to a quantifiable estimand. The relevance weights of the different approaches work on different level, i.e. either on cause-specific hazard ratios or on event count. CONCLUSION: The provided comparison and simulations can help to guide applied researchers to choose an adequate method for the analysis of composite endpoints combining (recurrent) events of different clinical relevance. The approach by Wei and Lachin and Rauch et al. can be recommended in scenarios where the composite effect is time-independent. The approach by Bakal et al. should be applied carefully.

Correlates of immune exacerbations in leprosy.

Leprosy is still a considerable health threat in pockets of several low and middle income countries worldwide where intense transmission is witnessed, and often results in irreversible disabilities and deformities due to delayed- or misdiagnosis. Early detection of leprosy represents a substantial hurdle in present-day leprosy health care. The dearth of timely diagnosis has, however, particularly severe consequences in the case of inflammatory episodes, designated leprosy reactions, which represent the major cause of leprosy-associated irreversible neuropathy. There is currently no accurate, routine diagnostic test to reliably detect leprosy reactions, or to predict which patients will develop these immunological exacerbations. Identification of host biomarkers for leprosy reactions, particularly if correlating with early onset prior to development of clinical symptoms, will allow timely interventions that contribute to decreased morbidity. Development of a point-of-care (POC) test based on such correlates would be a definite game changer in leprosy health care. In this review, proteomic-, transcriptomic and metabolomic research strategies aiming at identification of host biomarker-based correlates of leprosy reactions are discussed, next to external factors associated with occurrence of these episodes. The vast diversity in research strategies combined with the variability in patient- and control cohorts argues for harmonisation of biomarker discovery studies with geographically overarching study sites. This will improve identification of specific correlates associated with risk of these damaging inflammatory episodes in leprosy and subsequent application to rapid field tests.

Correlating efficacy and immunogenicity in malaria vaccine trials.

The availability of an effective and appropriately implemented malaria vaccine would form a crucial cornerstone of public health efforts to fight this disease. Despite many decades of research, however, no malaria vaccine has yet shown satisfactory protective efficacy or been rolled-out. Validated immunological substitute endpoints have the potential to accelerate clinical vaccine development by reducing the required complexity, size, duration and cost of clinical trials. Besides facilitating clinical development of existing vaccine candidates, understanding immunological mechanisms of protection may drive the development of fundamentally new vaccination approaches. In this review we focus on correlates of protection in malaria vaccine development: Does immunogenicity predict malaria vaccine efficacy and why is this question particularly difficult? Have immunological correlates accelerated malaria vaccine development in the past and will they facilitate it in the future? Does Controlled Human Malaria Infection represent a valid model for identifying such immunological correlates, or a correlate of protection against naturally-acquired malaria in itself?

Systems approaches to correlates of protection and progression to TB disease.

Tuberculosis (TB) is the leading cause of death due to a single infectious disease and an effective vaccine would substantially accelerate global efforts to control TB. An immune correlate of protection (CoP) from TB disease could aid vaccine optimization and licensure. This paper summarises opportunities for identifying CoP and highlights results from correlates of risk studies. Although we don't have CoP, there are ongoing efficacy trials with both disease and infection endpoints which provide opportunities for such an analysis. Transcriptomics has successfully identified robust CoR, with transcripts found in the Type I IFN pathway. Correlates of lower risk include BCG antigen specific IFN-γ and natural killer cells. Collating evidence from multiple studies using a range of systems approaches supports a role for IFN-γ in protection from TB disease. In addition, the cells that express the IFN-γ receptor are also important in protective immunity. Protection is a culmination not only of the amount of IFN-γ produced by T cells and NK cells but by the ability of IFN-γ receptor expressing monocytes to respond to IFN-γ. To better understand IFN-γ as a correlate we need to understand host-factors such as age, sex, co-infection, nutritional status and stress which may alter or impair the ability of cells to respond to IFN-γ. These studies highlight recent advances in our understanding of the immune mechanisms of TB disease risk and show the importance of whole systems approaches to correlates of risk analysis. CoP may be useful tools for specific vaccine products in specific populations, but a well-designed CoR analysis can identify novel immune mechanisms and provide insights critical for the development of new and better TB vaccines.

Correlates of immune and clinical activity of novel cancer vaccines.

Cancer vaccines are solely meant to amplify the pool of type 1 cytokine oriented CD4+ and CD8+ T cells that recognize tumor antigen and ultimately foster control and destruction of a growing tumor. They are not designed to deal with all aspects of immune ignorance, exclusion, suppression and escape that are generally in place in patients with cancer and may prevent the T cells to enter the tumor or to exert their effector function. This simple fact prompted for a reappraisal of the many recent trials in which therapeutic cancer vaccines have been examined as monotherapy. In this review, I focus on trials examining therapeutic cancer vaccines at different stages of existing disease. The analysis of vaccine-induced immune responses and clinical activity of therapeutic cancer vaccines revealed four levels of evidence for vaccine efficacy. The lowest levels, reflect the many trials in which the strength of the tumor-reactive T cell response of vaccinated patients is associated with better clinical outcome or change in tumor marker. The highest levels indicate occasional regressions of tumors and metastases after vaccination or reflect a stronger clinical impact of vaccine in a randomized trial. A whole series of trials in which vaccine-induced tumor immunity correlates with the clinical impact of cancer vaccines in premalignant diseases, settings of low tumor burden or tumor regressions in patients with cancer, form an attest to the fact that cancer vaccines work. While the current number of true clinical responders in each cancer trial is too low for firm conclusions on immune correlates of clinical reactivity in cancer, extrapolation of the results from vaccinated patients with pre-cancers suggest a requirement of broad type 1 T cell reactivity.

Correlates of vaccine-induced protective immunity against Ebola virus disease.

Ebola virus disease is a deadly infection which occurs in sporadic outbreaks. Several vaccine candidates have been developed. The most advanced candidate is the recombinant VSVΔG-ZEBOV-GP vaccine, in which the Vesicular Stomatitis Virus (VSV) envelope glycoprotein is replaced by the Zaire strain Ebola virus (ZEBOV) glycoprotein (GP). This vaccine demonstrated 100% protection in a ring vaccination trial performed in Guinea in 2015, was granted "Breakthrough Therapy Designation" by the FDA and PRIority Medicines (PRIME), and is currently (June 2018) used to support outbreak control in Democratic Republic of Congo. rVSVΔG-ZEBOV-GP elicits a strong and durable antibody response in most vaccinees. This sustained Ebola GP-specific antibody response correlates with an early activation of innate immunity, especially of monocytes and of type-I interferon induced genes. Despite significant progress in the characterization of vaccine-induced immunity, human correlates of protection against Ebolavirus infection have not yet been fully established. A systems biology approach, integrating clinical, immunological, transcriptomic and metabolomic data from pre-clinical and clinical vaccine studies, together with data from disease survivors, will be instrumental to identify Ebola vaccine correlates of protection. The information generated for the rVSVΔG-ZEBOV-GP vaccine may also help identify the correlates of protection of the other Ebola vaccine candidates.

Win ratio approach for analyzing composite time-to-event endpoint with opposite treatment effects in its components.

There is an increasing interest in the use of win ratio with composite time-to-event due to its flexibility in combining component endpoints. Exploring this flexibility further, one interesting question is in assessing the impact when there is a difference in treatment effect in the component endpoints. For example, the active treatment may prolong the time to occurrence of the negative event such as death or ventilation; meanwhile, the treatment effect may also shorten the time to achieving positive events, such as recovery or improvement. Notably, this portrays a situation where the treatment effect on time to recovery is in a different direction of benefit compared to the time to ventilation or death. Under such circumstances, if a single endpoint is used, the benefit gained for other individual outcomes is not counted and is diminished. As consequence, the study may need a larger sample size to detect a significant effect of treatment. Such a scenario can be handled by win ratio in a novel way by ranking component events, which is different from the usual composite endpoint approach such as time-to-first event. To evaluate how the different directions of treatment effect on component endpoints will impact the win ratio analysis, we use a Clayton copula-based bivariate survival simulation to investigate the correlation of component time-to-event. Through simulation, we found that compared to the marginal model using single endpoints, the win ratio analysis on composite endpoint performs better, especially when the correlation between two events is weak. Then, we applied the methodology to an infectious disease progression simulated study motivated by COVID-19. The application demonstrates that the win ratio approach offers advantages in empirical power compared to the traditional Cox proportional hazard approach when there is a difference in treatment effect in the marginal events.