NetraMark Holdings Inc. announced the presentation of new data describing how its proprietary NetraAI clinical trial solution identified novel biomarkers and protein-protein interaction (PPI) pathways associated with specific forms of non-small cell lung cancer (NSCLC) and colorectal cancer (CRC) using small data sets and a self-learning algorithm that obviates the need for large training data sets. These insights, or NetraPerspectives, have the potential to advance the personalized medicine landscape through patient enrichment strategies while also enabling novel diagnostic and therapeutic avenues, enhancing patient care and outcomes in these complex indications. Dr. Joseph Geraci, PhD, Founder and Chief Scientific Officer of NetraMark, presented the data yesterday in two posters at the American Association of Cancer Research (AACR) Annual Meeting 2024, which is taking place April 5-10, 2024 in San Diego, California.

Both posters were presented in the "Late-Breaking Research: Bioinformatics, Computational Biology, Systems Biology, and Convergent Science 2" session, which took place yesterday. The poster titled "NetraAI-driven discovery of novel biomarkers in MSI-high colon cancer for precision immunotherapy" (Abstract #LB395) describes the use of Attractor AI algorithms to identify causal clusters of variables (hypotheses) that explain specific sub-populations of patients with microsatellite instability-high (MSI-H) CRC. MSI-H tumors are characterized by an extensive mutational load, which fosters the production of neoantigens and amplifies immune visibility, making them prime candidates for immunotherapy.

However, these same factors contribute to heterogeneity that further complicates the efficacy of targeted therapies. NetraAI was applied to a data set consisting of tens of thousands of RNA expression variables from 390 samples from CRC patients. These profiles included 44 MSI-H and 21 MSI-low (MSI-L) samples and the dataset used consisted of a total of 22,283 variables.

Key findings from the analysis include: In one NetraPerspective, there was a MSI-H subpopulation identified, consisting of 29 MSI-H and 2 MSI-L samples. This subpopulation is characterized by expression of CATSPERB (p=1.2 x 10-7), MLPH (p=4.9 x 10-5), FUT8 (p=8.6 x 10-5), DUSP4 (p=1.1 x 10-3), and PLLP (p=0.01). Constructing PPI networks based on the identified variables suggests a complex interplay among them, particularly in the context of spermatogenesis.

Mismatch repair (MMR) is essential for ensuring genetic integrity during sperm production. The findings of the NetraAI analysis suggest that defects in MMR play a causal role in the genetic instability seen in MSI-H CRC. The specificity of CATSPERB to an MSI-H colon cancer subgroup posits it as a potential biomarker for identifying patients who might benefit from tailored therapeutic approaches, contributing to the personalized medicine landscape.

CATSPERB protein is primarily associated with calcium channels in sperm, but its over-expression in a subset of MSI-H CRC patients suggests that the protein may modulate calcium signaling in tumor cells, which is known to play a role in a variety of cellular processes that drive cancer cell proliferation, survival, and metastasis. The poster titled "The power of NetraAI: Precision medicine in oncology through sub-insight learning from small data sets" (Abstract #LB396) describes the use of Attractor AI algorithms to identify variables defining specific subpopulations of patients with NSCLC. A small data set consisting of 104 gene expression samples of adenocarcinoma (ADC) and squamous cell carcinoma (SSC) was compiled from two NSCLC datasets.

Key findings from the analysis include: One NetraPerspective showed multiple explainable subpopulations of NSCLC, primarily stratified as ADC or SCC subpopulations. Interestingly, there were multiple subpopulations of each subtype suggesting that different combinations of variables drive specific ADC and SCC subtypes. Examination of each subpopulation using NetraAI's unique zoom capabilities of NetraAI, identified the specific patients and characterizing variables.

In one NetraPerspective, NetraAI distinguished between ADC and SCC subtypes through unique genetic signatures, with 9 out of 10 variables correlating with known NSCLC markers. This validates the methods and technology used by NetraAI. PIGX emerged as a novel target due to its previously unexplored role in cancer biology.

Further investigation into PPI networks revealed a significant connection between PIGX and BACE1, a protein implicated in NSCLC brain metastasis. This opens new avenues for understanding molecular mechanisms underlying cancer progression and metastasis. PIGX is also related to PIGN which is associated with genomic instability and regulates spindle assembly checkpoint proteins in leukemia transformation and progression.