Intelligencia Portfolio OptimizerTM

AI Clinical Development Software

AI opens a new window of innovation for drug development. Using Portfolio Optimizer, you can optimize your drug pipeline, select promising drug candidates early, assess the competitive landscape, and make highly informed business development decisions.

Explore an asset’s potential to clear all technical and regulatory hurdles and receive FDA approval.

Do your internal probability of technical and regulatory success (PTRS) estimates utilize the best possible data available?

Are you concerned that your PTRS assessments may be too subjective?

Are your benchmarks inconsistent and based on limited information? 

Does the process of expert solicitation take significant resources and time? 

Do you need additional insights to inform asset and portfolio strategies? 

Do you need to understand the competitive landscape for your asset better? 

Do you want to understand how each trial design parameter affects PTRS assessment?

Solutions and Insights to Drive Success

Assess current pipeline with real-time AI-driven PTRS for faster, more confident decisions.

Gain transparency and context with AI Explainability with PTRS drivers to understand the impact they have on the assessment.

See phase transition and PTRS for industry-led, interventional FDA-track trials, including that of competitors.
Explore how a specifc drug or combinations of drugs advance through clinical development phases
 
View clinical outcomes by phase, arm and cohort and access primary and secondary endpoints to understand clinical design by specific indication.

Together, Let’s Drive Your Strategic Portfolio Management Forward

Experience What Portfolio Optimizer Can Do for You

 

Challenge

Expensive and time-consuming expert elicition and KOL process.

Solution

Leverage Portfolio Optimizer to complement and enhance internal process with consistent and comprehensive data.

Benefits

Save valuable resources while minimizing manual effort and human bias.

Challenge

Limited data and high risk on early stage licensing. High-costs and competition on later stage assets.

Solution

Use Portfolio Optimizer to compare PTRS of various candidates, assess them against objective benchmarks, and compare the competitive landscape to support your valuation efforts.

Benefits

Identify promising assets earlier and more confidently by incorporating data-driven probability of success assessment to support your business case and negotiations.

Challenge

Lack of consistent and comprehensive benchmarking data make for difficult decisions on which drug candidates to move forward.

Solution

Use Portfolio Optimizer to inform decisions on therapeutic area expansion, indication and phase benchmarking to rank drug development opportunities.

Benefits

Leverage objective assessments to generate a list of prioritized assets and foster internal alignment and understand where your asset fits in the competitive landscape.

The predictive performance of our algorithms in forecasting FDA approvals for Phase 2 oncology programs boasts a prospectively validated accuracy of 83%*, and 90% when measured retrospectively.

*Based on Intelligencia AI predictions as of May 2022 for oncology
programs with definite announcements in 2023.

Our Industry-Leading Accuracy Is Validated With Four Distinct Methods

Retrospective Accuracy

Area under the curve (AUC) is a metric which indicates whether the model assigns higher probability to the approved programs higher than the failed ones. The higher the AUC the better.

Sensitivity/ Specificity

The True Positive (sensitivity) and True Negative (specificity) rates highlight the ability of the model to correctly identify approved and rejected programs, accordingly. The higher the rates the better.

Brier Score Loss

Brier Score Loss (BSL) is a metric that quantifies the accuracy of probability predictions. BSL penalizes high probabilities of failed programs and low probabilities of approved programs. The lower the BSL the better.

Prospective Accuracy

The Calibration Curve measures prospectively how closely the predicted probabilities match the actual and observed outcomes, helping to understand the reliability of the model.