RELIABLE & HIGH-COVERGE REPORTS
FROM OUR PROPRIETARY GENOMIC MARKERS AND MODELS
A RIGHT DECISION
FROM CA-GENOME RX's DIGITAL BIOPSY
At the heart of Ca-Genome Rx’s innovative approach lies the Ca-Genome Rx report, a revolutionary diagnostic tool conceptualized as a “digital biopsy.” This state-of-the-art platform integrates propriety genomic markers with cutting-edge machine learning to deliver precise predictions about patient responsiveness to targeted therapies, immunotherapies, and chemotherapy. By incorporating these propriety genomic biomarkers and models, Ca-Genome Rx extends the diagnostic potential of conventional genomic assays, enabling oncologists to tailor treatments with unprecedented specificity.
Overcoming Immunotherapy Challenges: Advancing Biomarkers for Broader Impact
Immunotherapy has heralded a paradigm shift in cancer treatment, offering unprecedented hope to many patients. However, significant hurdles persist, as its benefits are realized in only a small subset of individuals. Specifically, immune checkpoint blockade (ICB)—a cornerstone of modern immunotherapy—demonstrates efficacy in merely 12.5% of cancer patients. Compounding this limitation, 10-25% of patients experience severe adverse effects, while up to 29% endure paradoxical tumor progression, potentially accelerating their disease trajectory. The economic burden is equally substantial, with treatment costs ranging from $100,000 to $200,000 annually per patient, underscoring the urgent need for more effective and cost-efficient therapeutic strategies.
While existing biomarkers, such as PD-L1 expression and tumor mutation burden (TMB), provide some predictive value in identifying ICB responders, their limitations are increasingly evident. Many malignancies—including triple-negative breast cancer, ovarian cancer, uterine cancer, and esophageal cancer—exhibit robust responses to ICB despite having inherently low TMB. This phenomenon highlights the existence of cryptic tumor antigens and underscores the critical need for next-generation biomarkers capable of uncovering these hidden vulnerabilities. By enabling precise patient stratification and tailored therapeutic interventions, advanced biomarkers can significantly expand the transformative potential of immunotherapy, bringing its benefits to a broader population.
Transformative Genomic Biomarkers for
Precision Cancer Immunotherapy
Represents a transformative biomarker by quantifying cryptic intragenic rearrangements that alter exon structures within tumor genomes, a previously overlooked area in cancer genetics. IRB burden has emerged as the most influential predictor of T-cell inflammation in TMB-low, IGR-dominant cancers, such as breast, ovarian, uterine, and esophageal cancers, as well as in tumors treated with platinum-based therapies. Clinical trial datasets, including the MEDI4736 trial for esophageal adenocarcinoma, NeoPembrOv trial for high-grade serous ovarian cancer, two clinical trials for metastatic triple-negative breast cancer, and the IMVigor210 trial for urothelial carcinoma, have demonstrated IGR burden’s strong correlation with immune infiltration and its ability to predict immune checkpoint blockade benefits. These findings highlight IGRs as a dark source of neoantigens and a groundbreaking predictive biomarker for improving immunotherapy precision in TMB-low and IGR dominant cancer entities as well as platinum-treated tumors.
Represents a groundbreaking metric for evaluating tumor antigenicity in PD-L1-negative patients. TAAs, normal proteins overexpressed in tumors and immune-privileged organs, have long been explored as vaccine targets. Our recent findings highlight three prerequisites for mounting an effective TAA-reactive immune response: 1) a high TAA burden within the tumor, 2) a non-exhausted tumor immune context, and 3) costimulatory signals from immune-stimulating molecular patterns, such as double-strand DNA breaks. Our studies have shown that TAA burden correlates with immune checkpoint blockade efficacy in urothelial carcinoma and head and neck cancers with low T cell exhaustion and negative PD-L1 expression on immune cells. This underscores TAB’s potential as a predictive biomarker for identifying immunotherapy responders in PD-L1 negative patients.
Our proprietary machine learning and artificial intelligence solutions tailored for the hallmarks of genomic data
With the advent of ultra-low-cost deep sequencing, both transcriptome and genome sequencing are on the brink of transforming omics-based precision oncology into a clinical routine within the next decade. Despite these advancements, the development of genome-wide models capable of guiding patient-specific therapeutic interventions remains limited. Current machine learning methods, primarily adapted from fields like image and language processing, struggle to address the unique challenges of cancer genomic data. These include: 1. Ultra-high dimensionality: An enormous feature set but limited sample sizes. 2. Imprecision: Due to sequencing errors and experimental variability. 3. Selective oncogenic pathway influence: Affecting drug sensitivity in only a minority of subjects. 4. High multicollinearity: Among genomic features due to co-expression and co-occurrence of genetic events. These issues are compounded by the opaque nature of machine learning algorithms, which hinders their adoption in clinical settings where transparency is crucial. Therefore, there is a critical need for innovative, resilient, and transparent modeling methods specifically designed for genomic data.
Introducing iGenSig and iGenSig-Rx:
Revolutionizing Omics-based Precision Oncology
The iGenSig-Rx technology represents a paradigm shift in modeling cancer therapeutic responses using multi-omics data. Unlike conventional black-box machine learning tools, iGenSig-Rx employs a “white-box” approach, offering transparency in its predictive modeling. By leveraging high-dimensional redundant genomic features, iGenSig-Rx reinforces predictive robustness akin to strengthening pillars with steel rods, ensuring resilience against sequencing errors and dataset heterogeneity. Validated across clinical trials for HER2-targeted therapies and other treatment regimens, iGenSig-Rx has demonstrated consistent predictive power for therapeutic responses. It has also revealed clinically actionable insights into the pathways driving therapeutic sensitivity and resistance, enabling informed clinical decisions. Through iGenSig-Rx, Ca-Genome Rx bridges the gap between complex genomic datasets and actionable treatments, offering a robust and interpretable solution to guide treatment strategies across diverse cancer types. Furthermore, iGenSig-AI, our advanced machine learning framework, enhances the ability to conduct in silico drug screening, accelerating the identification of optimal therapeutic interventions for individualized care