To evaluate the effectiveness of fast frozen rapid automated immunohistochemistry (FFRA-IHC) in improving tumor classification during intraoperative assessments, highlighting its potential impact on surgical decision-making.
Key Findings:
FFRA-IHC allowed specific classification in 68% of previously unclassifiable tumors, including various tumor types.
All 5 ambiguous resection margins were resolved with FFRA-IHC.
FFRA-IHC potentially altered intraoperative management in approximately 12% of cases.
Turnaround time for FFRA-IHC averaged 21 minutes, with total diagnosis time around 40 minutes.
Interpretation:
FFRA-IHC demonstrates potential to enhance intraoperative decision-making by providing rapid and specific tumor classifications, though its clinical impact needs further validation in broader settings.
Limitations:
Small sample size and single-center design limit generalizability.
Assessment relied on a semiquantitative scoring system not externally validated, affecting reliability.
No evaluation of patient outcomes or long-term clinical impact, limiting conclusions.
Conclusion:
Larger studies with expanded antibody panels are needed to assess the broader applicability and clinical outcomes associated with FFRA-IHC in intraoperative pathology, particularly in translating management changes into measurable surgical and oncologic outcomes.
Across six experiments—including a blinded, real-world ER evaluation—an OpenAI large language model outperformed physician baselines on multiple clinical reasoning tasks, though not on key safety endpoints such as cannot-miss diagnoses
