Toward Smarter Lumping and Smarter Splitting: Rethinking Strategies for Sepsis and Acute Respiratory Distress Syndrome Clinical Trial Design

Both quality improvement and clinical research efforts over the past few decades have focused on consensus definition of sepsis and acute respiratory distress syndrome (ARDS). Although clinical definitions based on readily available clinical data have advanced recognition and timely use of broad supportive treatments, they likely hinder the identification of more targeted therapies that manipulate select biological mechanisms underlying critical illness. Sepsis and ARDS are by definition heterogeneous, and patients vary in both their underlying biology and their severity of illness. We have long been able to identify subtypes of sepsis and ARDS that confer different prognoses. The key is that we are now on the verge of identifying subtypes that may confer different response to therapy. In this perspective, inspired by a 2015 American Thoracic Society International Conference Symposium entitled "Lumpers and Splitters: Phenotyping in Critical Illness," we highlight promising approaches to uncovering patient subtypes that may predict treatment responsiveness and not just differences in prognosis. We then discuss how this information can be leveraged to improve the success and translatability of clinical trials by using predictive enrichment and other design strategies. Last, we discuss the challenges and limitations to identifying biomarkers and endotypes and incorporating them into routine clinical practice.

Applying Precision Medicine to Trial Design Using Physiology. Extracorporeal CO2 Removal for Acute Respiratory Distress Syndrome

In clinical trials of therapies for acute respiratory distress syndrome (ARDS), the average treatment effect in the study population may be attenuated because individual patient responses vary widely. This inflates sample size requirements and increases the cost and difficulty of conducting successful clinical trials. One solution is to enrich the study population with patients most likely to benefit, based on predicted patient response to treatment (predictive enrichment). In this perspective, we apply the precision medicine paradigm to the emerging use of extracorporeal CO₂ removal (ECCO₂R) for ultraprotective ventilation in ARDS. ECCO₂R enables reductions in tidal volume and driving pressure, key determinants of ventilator-induced lung injury. Using basic physiological concepts, we demonstrate that dead space and static compliance determine the effect of ECCO₂R on driving pressure and mechanical power. This framework might enable prediction of individual treatment responses to ECCO₂R. Enriching clinical trials by selectively enrolling patients with a significant predicted treatment response can increase treatment effect size and statistical power more efficiently than conventional enrichment strategies that restrict enrollment according to the baseline risk of death. To support this claim, we simulated the predicted effect of ECCO₂R on driving pressure and mortality in a preexisting cohort of patients with ARDS. Our computations suggest that restricting enrollment to patients in whom ECCO₂R allows driving pressure to be decreased by 5 cm H₂O or more can reduce sample size requirement by more than 50% without increasing the total number of patients to be screened. We discuss potential implications for trial design based on this framework.

Antidepressant drug development: Focus on triple monoamine reuptake inhibition

Many patients with major depressive disorder (MDD) only partially respond, and some have no clinically meaningful response, to current widely used antidepressant drugs. Due to the purported role of dopamine in the pathophysiology of depression, triple-reuptake inhibitors (TRIs) that simultaneously inhibit serotonin (5-HT), norepinephrine (NE) and dopamine reuptake could be a useful addition to the armamentarium of treatments for MDD. A TRI should more effectively activate mesolimbic dopamine-related reward-networks, restore positive mood and reduce potent 5-HT reuptake blockade associated "hypodopaminergic" adverse effects of decreased libido, weight gain and "blunting" of emotions. On the other hand, dopaminergic effects raise concern over abuse liability and TRIs may have many of the cardiovascular effects associated with NET inhibition. Several clinical development programs for potential TRI antidepressants have failed to demonstrate significantly greater efficacy than placebo or standard of care. Successful late-stage clinical development of a TRI is more likely if experimental research studies in the target population of depressed patients have demonstrated target engagement that differentially and dose-dependently improves assessments of reward-network dysfunction relative to existing antidepressants. TRI treatment could be individualized on the basis of predictive markers such as the burden of decreased positive mood symptoms and/or neuroimaging evidence of reward network dysfunction. This review focuses on how the next generation of monoamine-based treatments could be efficiently developed to address unmet medical need in MDD.

Biomarkers in acute respiratory distress syndrome

PURPOSE OF REVIEW

This article provides an overview of protein biomarkers for acute respiratory distress syndrome (ARDS) and their potential use in future clinical trials.

RECENT FINDINGS

The protein biomarkers studied as indices of biological processes involved in the pathogenesis of ARDS may have diagnostic and/or prognostic value. Recently, they also proved useful for identifying ARDS phenotypes and assessing heterogeneity of treatment effect in retrospective analyses of completed clinical trials.

SUMMARY

This article summarizes the current research on ARDS biomarkers and provides insights into how they should be integrated as prognostic and predictive enrichment tools in future clinical trials.

An explainable machine learning-based phenomapping strategy for adaptive predictive enrichment in randomized controlled trials

Randomized controlled trials (RCT) represent the cornerstone of evidence-based medicine but are resource-intensive. We propose and evaluate a machine learning (ML) strategy of adaptive predictive enrichment through computational trial phenomaps to optimize RCT enrollment. In simulated group sequential analyses of two large cardiovascular outcomes RCTs of (1) a therapeutic drug (pioglitazone versus placebo; Insulin Resistance Intervention after Stroke (IRIS) trial), and (2) a disease management strategy (intensive versus standard systolic blood pressure reduction in the Systolic Blood Pressure Intervention Trial (SPRINT)), we constructed dynamic phenotypic representations to infer response profiles during interim analyses and examined their association with study outcomes. Across three interim timepoints, our strategy learned dynamic phenotypic signatures predictive of individualized cardiovascular benefit. By conditioning a prospective candidate's probability of enrollment on their predicted benefit, we estimate that our approach would have enabled a reduction in the final trial size across ten simulations (IRIS: -14.8% ± 3.1%, pone-sample t-test=0.001; SPRINT: -17.6% ± 3.6%, pone-sample t-test<0.001), while preserving the original average treatment effect (IRIS: hazard ratio of 0.73 ± 0.01 for pioglitazone vs placebo, vs 0.76 in the original trial; SPRINT: hazard ratio of 0.72 ± 0.01 for intensive vs standard systolic blood pressure, vs 0.75 in the original trial; all with pone-sample t-test<0.01). This adaptive framework has the potential to maximize RCT enrollment efficiency.