The Association for Molecular Pathology (AMP) annual meeting is currently being held at the Seattle Convention Center. Several studies are being presented, highlighting the role of molecular diagnostics in pediatric and women’s health, and artificial intelligence (AI) pros, cons, and trends in MDx.

What can I learn about molecular testing for pediatric diseases? The following studies will be presented:

• Abstract G017: Whole-genome sequencing in an adolescent cohort: Results of rapid and ultra-rapid testing. On November 14 at 9:15 a.m., Jennifer Schleit, Ph.D. of Rady Children’s Institute for Genomic Medicine will present her team’s poster, revealing the importance of genome testing at birth.
• Abstract G027: Phenotype-driven reanalysis of clinical genome sequencing data shows comparable yield in originally negative and positive reports. Jing Zhang, M.D., Ph.D. of Rady Children’s Institute for Genomic Medicine will discuss reanalyzing previous test results when a child’s symptoms change on November 14 at 9:15 a.m.
• Abstract ST015: Clinical evaluation of whole-genome enzymatic methylation sequencing for integrated classification and copy number profiling of pediatric central nervous system tumors. Jianling Ji, M.D. of Children’s Hospital Los Angeles and Keck School of Medicine at USC will address if a single test can detect childhood brain tumors rather than multiple assays on November 14 at 1:45 p.m.
• Abstract H011: An ensemble machine learning classifier for B-ALL molecular subtyping via targeted RNA-seq: A cost-effective approach for resource-limited settings. On November 14 at 9:15 a.m., Joaquín García-Solorio, Ph.D. of Instituto Nacional de Medicina Genómica will discuss AI and advanced childhood leukemia testing accessibility.
• Abstract G022: Prevalence of cancer predisposition variants in medulloblastoma and Abstract ID045: Neonatal healthcare-associated Mpox cluster linked to reverse vertical transmission and community spread in Sindh, Pakistan will be presented today at 9:15 a.m. and 12:55 p.m., respectively.

What can I learn about the role of molecular diagnostics for women’s health? The following studies will be presented:

• Mycoplasma genitalium and macrolide resistance prevalence across a large women's health cohort: Implementation of a reflex testing algorithm that supports CDC-recommended resistance-guided treatment. On November 14 at 9:15 a.m., John Fulmer, M.S., MB (ASCP) of Unified Women’s Healthcare will present his team’s study, highlighting the importance of molecular testing for personalized STI treatment and antibiotic decision-making.
• Methylomic stratification of recurrence risk in primary, low-grade, copy number-low endometrioid adenocarcinoma of the uterus. Today at 9:15 a.m., Matthew Zunitch of the University of California, San Fransico, will discuss the importance of molecular profiling for endometrial cancer risk assessment and tumor biology.

What can I learn about the reliability of AI in the MDx space? The following studies will be presented:

• Abstract H071: An epigenomic foundation model for hierarchical WHO/ICC classification of hematolymphoid tumors. Pedro Orsini, M.D., of ALMA Genomics will reveal findings showing how an AI model distinguished blood cancer subtypes. The poster will be presented on November 14 from 9:15–10:15 a.m.
• Abstract I069: Deep learning-based prediction of genomic alterations from histopathology using large-scale pre-trained representations in gastric adenocarcinoma. On November 14 at 9:15 a.m., Kiran Nijjer of Stanford University will discuss how AI can aid with pathology slide analysis, saving labs time and money.

These AI studies will be presented today:

• Abstract I040: Turnkey precision oncology an open, validated read-to-report platform for guideline-concordant, AI-assisted molecular pathology reporting at 9:15 a.m.
• Abstract I068: Measuring the accuracy, hallucination, bias and guideline adherence of AI-assisted molecular pathology reporting at 9:15 a.m.
• Abstract I070: Real-time human-in-the-loop histopathology AI genomic test triage reduces rapid EGFR testing by 33% in lung adenocarcinoma at 9:15 a.m.
• Abstract I056: Zero-shot whole slide image analysis distinguishes hepatocellular carcinoma from cholangiocarcinoma using pathology language foundation models at 9:15 a.m.