NATURE npj Precision Oncology: Real-world performance analysis of Genomate® in lung cancer
Read More
Lung cancer remains the leading cause of cancer-related deaths globally, accounting for approximately 1.8 million deaths each year. While molecular testing has become more common, treatment decisions are still often based on single biomarkers, which fail to reflect the complexity of most tumor genomes, many of which contain four or more driver mutations. This oversimplified approach yields underutilized genomic data and inconsistent outcomes. Molecular tumor boards can be helpful, but research indicates that their recommendations are highly discordant and challenging to implement on a large scale in everyday clinical practice.
To address these challenges, Genomate Health developed the Digital Drug Assignment (DDA) system (commercially known as Genomate®), a computational reasoning engine that analyzes a tumor’s entire molecular profile and ranks available therapies based on predicted efficacy. Unlike opaque machine learning algorithms, DDA is a knowledge-based system that simulates expert reasoning, integrating published scientific evidence to deliver clear, personalized therapy rankings. It enables oncologists to understand why a therapy is recommended and how it relates to a patient’s molecular profile.
In a new study published in npj Precision Oncology, researchers assessed the real-world clinical impact of Genomate® by analyzing treatment outcomes from 111 lung cancer patients. The central question: Do treatments ranked highly by Genomate® lead to better outcomes?
This retrospective study analyzed the systemic treatment histories of patients enrolled in a precision oncology program in a hospital specialized in pulmonary diseases between 2018 and 2022.
All patients underwent comprehensive tumor molecular profiling, and their results were processed through Genomate’s DDA system. The DDA algorithm generated ranked therapy recommendations, which were shared with physicians as a decision support tool. It’s essential to note that DDA was used solely for decision support; the clinical specialists ultimately made the treatment decisions. This approach enabled researchers to retrospectively assess whether high DDA treatment scores were associated with improved clinical outcomes.
In total, 155 systemic lines of therapy (LOTs) were included in the analysis. Each treatment was classified into one of the following categories:
The primary clinical outcomes assessed included:
This real-world analysis revealed that lung cancer patients who received treatments aligned with Genomate Health’s Digital Drug Assignment (DDA) recommendations experienced significantly better clinical outcomes.
Patients who received any MTA had significantly improved survival outcomes compared to those treated with only standard chemotherapies:
Patients who received any MTA had significantly improved survival outcomes compared to those treated with only standard chemotherapy.
Not all targeted therapies performed equally. The study further stratified MTA treatments based on their DDA score, a computational score representing how well the therapy matches the tumor’s complete molecular profile.
Patients who received high-scoring MTAs (DDA score ≥ 1000) had markedly better clinical outcomes compared to those who received low-scoring MTAs or chemotherapy:
Patients receiving top-ranked therapies by DDA lived longer, responded better, and were more likely to achieve long-term control.
Survival analyses by treatment DDA scores.
Progression-free survival (PFS) probability of high-score MTA treatments was plotted against all other treatments (SC and low-score MTAs) (left) and overall survival (OS) of patients who received at least one high-score MTA line of treatment (LOT) during their treatment course versus OS of patients who did not receive any high-score MTA LOT during their treatment course (right). High-score MTA treatments are indicated with blue, the corresponding comparator sets in red in both Kaplan-Meier plots.
This study provides robust real-world evidence that Genomate® is clinically predictive: therapies with higher DDA scores consistently led to better patient outcomes across multiple endpoints, including progression-free survival, overall survival, and response rates.
By simulating expert reasoning, Genomate® enables oncologists to make data-driven, individualized treatment decisions based on the entire molecular profile of the tumor, not just single biomarkers. This shift from one-size-fits-all to n-of-1 treatment selection represents a significant advance toward consistent, scalable, and equitable precision oncology.
Importantly, this real-world evidence demonstrates the feasibility of implementing digital precision oncology in routine care settings. Genomate®’s explainable, non–machine learning approach offers clarity, consistency, and confidence in treatment decisions, qualities that are especially valuable when managing complex cases and limited therapeutic windows.
The study also showed that 50% of the patients were matched to high-scoring, FDA-approved targeted therapies for lung cancer, demonstrating that Genomate®’s insights are not just theoretical; they are clinically actionable today, without the need for experimental compounds or inaccessible treatments.
This study demonstrates that Genomate® meaningfully improves real-world outcomes for lung cancer patients. By helping oncologists prioritize therapies based on the full molecular complexity of each tumor, Genomate®-guided decisions led to longer survival, higher response rates, and greater disease control.
These findings mark a critical advancement in precision oncology, showing that computational reasoning can bring structure, clarity, and clinical benefit to an increasingly complex field. Unlike black-box algorithms, Genomate® provides transparent and explainable recommendations that are both scientifically grounded and immediately actionable.
As precision oncology continues to expand, Genomate® provides a scalable, reproducible, and clinically validated pathway to personalized cancer care, enabling the right treatment, at the right time, for every patient.
.png)
