Artera, a pioneer in multimodal artificial intelligence (MMAI) for prognostic and predictive cancer testing, is set to present two oral abstracts at the 2024 American Society for Radiation Oncology (ASTRO) Annual Meeting. The company’s MMAI platform utilizes advanced AI algorithms on digitized images of hematoxylin and eosin (H&E)-stained pathology slides, integrating these visual data with patients’ clinical information to deliver personalized insights.
The upcoming presentations will showcase validation data demonstrating the MMAI biomarker’s potential to enhance treatment decision-making for patients diagnosed with oligometastatic castration-sensitive prostate cancer (omCSPC). This early stage of metastatic prostate cancer is critical, as timely intervention can prevent further progression. Metastasis-directed therapy (MDT), a common treatment approach for omCSPC, targets metastatic lesions through radiation or surgical means.
“We’re excited to unveil our latest findings at ASTRO 2024, which highlight how Artera’s biomarker can cater to the specific needs of this unique patient demographic,” said Andre Esteva, CEO of Artera. “Patients with omCSPC stand at a crucial juncture in their cancer journey, and we are committed to using our digital pathology AI to empower them and their healthcare providers with tailored insights for more individualized treatment plans.”
Key Highlights from the Abstracts:
- The biomarker has been validated as prognostic for overall survival (HR=4.38, p=0.022).
- It demonstrates predictive capabilities for MDT in omCSPC, revealing that patients with high MMAI scores (HR=0.32, 95% CI=0.12-0.90; p=0.03) benefited from MDT, unlike those with low MMAI scores (HR=1.59, 95% CI=0.63-4.04; p=0.33). This indicates an improvement in metastasis-free survival for those receiving MDT (p-interaction=0.02).
- The biomarker aligns with established genomic mutations related to cancer aggressiveness. DNA sequencing indicates that patients with high MMAI scores tended to have more WNT pathway (APC/CTNNB1) mutations and significantly more BRCA2/ATM mutations compared to those with lower scores (p=0.13 and p=0.008, respectively). Conversely, patients with low MMAI scores exhibited a higher prevalence of SPOP mutations, which correlate with better prognosis (p=0.03). This suggests the biomarker effectively captures patterns linked to known biological pathways.
Dr. Phuoc Tran, Professor and Vice Chair of Research at the University of Maryland Greenebaum Comprehensive Cancer Center, remarked, “The findings from this research affirm the MMAI biomarker’s potential to tailor treatments for omCSPC patients. The ability of the AI algorithm to identify mutations and transcriptional pathways associated with metastasis should enhance clinician and patient confidence in its application.”
The ASTRO conference is scheduled for September 29 to October 2, 2024, at the Walter E. Washington Convention Center in Washington, DC, where the two oral abstracts will be presented on October 2. Attendees can visit Artera’s booth (#2606) to learn more about the company’s offerings, including the commercially available ArteraAI Prostate Test. Below are details on each abstract:
- Validation of a Digital Pathology-Based Multimodal Artificial Intelligence Model in Oligometastatic Castration-Sensitive Prostate Cancer (Abstract 334)Multiple randomized trials have shown improvements in progression-free survival for omCSPC patients undergoing MDT. However, variability in clinical outcomes highlights the need for prognostic and predictive biomarkers. Trained using localized prostate cancer data, the MMAI biomarker demonstrates prognostic value for overall survival in omCSPC patients and predictive benefits from MDT, indicating a need for further validation to personalize treatment strategies effectively.
- Association of a Digital Pathology Multimodal Artificial Intelligence Algorithm with Pro-Metastatic Genomic Pathways in Oligometastatic Prostate Cancer (Abstract 333)Artera’s MMAI algorithm, previously validated in localized prostate cancer, has now been assessed for its prognostic capability in omCSPC. The study reveals a correlation between MMAI scores and genomic mutations associated with metastatic pathways, suggesting that digital pathology-based MMAI algorithms can identify phenotypic features representative of underlying genomic and transcriptomic pathways.
