TempoGap: A Molecular "Recovery Clock" for Predicting Outcomes in Critical Illness

This technology is a computational framework that uses gene expression and protein data from blood immune cells to estimate a patient's biological "immune recovery time" and compare it against actual time since injury or illness onset. By quantifying the gap between biological and chronological recovery, it enables earlier, more precise identification of patients at risk for poor outcomes such as prolonged organ failure or death across trauma, sepsis, and other acute critical illnesses.

Description

The invention models immune recovery as a continuous, individualized biological process rather than relying on a single snapshot measurement. A predictive model is trained on longitudinal transcriptomic and/or proteomic profiles from circulating immune cell populations to infer an "immune recovery time" for a given patient. This predicted value is then compared to the patient's actual chronological time since injury or disease onset, producing a quantitative deviation metric that reflects whether a patient's immune system is recovering ahead of, in line with, or behind their clinical timeline. Alongside this predictive scoring, the framework identifies interpretable molecular signature sets that underlie the model's predictions including a cellular stress-associated signature linked to poor outcomes and a recovery-associated signature linked to favorable immune regulation. Because the stress-associated signature was found to be broadly conserved across multiple disease contexts, the same underlying framework can be applied across different acute conditions without requiring a disease-specific model rebuild, while still providing both a prognostic score and a biologically interpretable explanation for that score addressing a key limitation of "black-box" predictive models.

Applications

- Clinical decision-support tools for ICU and trauma care, flagging patients at elevated risk of organ failure or delayed recovery
- Companion diagnostics for sepsis and critical illness management platforms
- Patient stratification tools for clinical trial enrollment and endpoint selection in critical care and immunology trials
- Treatment-response monitoring assays for tracking a patient's trajectory during hospitalization
- Precision medicine platforms extending molecular risk-scoring to other acute or systemic immune-mediated conditions

Advantages

- Models immune recovery as a dynamic, individualized process rather than a static, one-time measurement
- Provides a single quantitative metric capturing the divergence between predicted biological recovery and actual clinical time course
- Applicable across multiple disease contexts (e.g., trauma, sepsis, viral infection) without requiring disease-specific redesign
- Delivers both a predictive score and biologically interpretable molecular signatures, avoiding the limitations of black-box models
- Compatible with minimally invasive blood-based molecular measurements, supporting practical clinical translation

Invention Readiness

The technology has progressed beyond initial concept to a validated computational framework, with existing software and both transcriptomic and proteomic data supporting model performance. The approach has been evaluated on data from multiple independent patient cohorts, demonstrating consistent predictive performance across disease contexts. Further studies are needed to validate the framework in prospective clinical trials and to establish standardized, disease-specific chronological benchmarks to optimize predictive performance in additional clinical settings.

IP Status

Patent Pending