Portfolio Optimizer

Portfolio Optimizer's Causal Disease Modeling capability, developed by PhaseV, is designed to help R&D teams unlock the power of biomarker-driven clinical insights. By establishing causal relationships between biomarkers and clinical outcomes, the platform enables life sciences organizations to quantify biomarker impact, identify key biological pathways, and simulate "what if" scenarios — all in support of more informed drug development, combination therapy design, and indication selection decisions.

This solution is built for clinical researchers, translational scientists, and portfolio strategists who need to predict the probability of trial success and accelerate discovery with data-efficient, explainable modeling — even when working with limited sample sizes.

Core Capabilities

• Establishes causal links between biomarkers and clinical outcomes
• Quantifies the impact of changes in biomarkers on clinical benefit
• Identifies potential pathways and mediators influencing clinical outcomes
• Enables exploration of "what if" scenarios, including combination therapies and alternative patient populations
• Supports informed discovery and preclinical R&D decision-making
• Guides optimal indication selection and sequencing

Causal Modeling Workflow

1. Develop a Causal Graph: Model disease progression using a prognostic framework to represent biological relationships structurally.
2. Build the Control Arm: Leverage a Causal Directed Acyclic Graph (DAG) and integrate Real-World Data (RWD) to create a robust representation of the untreated patient population.
3. Incorporate Treatment Dynamics: Utilize RWD where available to account for treatment effects, and integrate insights from preclinical or early clinical data to enrich the model.
4. Quantitatively Assess Impact: Analyze changes to key nodes within the graph by propagating their effects throughout the entire causal structure to measure downstream clinical impact.

Use Cases

• Identify new treatments and combination therapy opportunities
• Generate and estimate new patient populations for trial planning
• Make informed indication selection and treatment sequencing decisions
• Conduct benchmarking and evaluation of therapeutic strategies
• Accurately recapitulate biology even under small sample size constraints

Key Advantages

• Data Efficient: Delivers meaningful insights even when data availability is limited, reducing dependency on large datasets.
• Explainable: Provides transparent, interpretable models that support scientific and regulatory communication.
• Flexible Interactions: Allows users to explore a wide range of scenarios and interventions within the causal framework.

PhaseV's Causal Disease Modeling is part of the broader Portfolio Optimizer solution suite, accessible to teams at any stage of clinical development. The platform is based in Cambridge, MA, and supports organizations seeking to transform their clinical trial strategies through rigorous, data-driven causal analysis.