JClinPath_BMJ

HER2 testing in the antibody–drug conjugate era 🔬

Liu and colleagues summarise the 2024 Chinese guidelines for HER2 testing in breast cancer, with a focus on reliable assessment of low HER2 expression.

Key themes include careful specimen handling, gradient controls, review of results near scoring thresholds, and reassessment of recurrent or metastatic lesions. The guidelines recommend reporting the IHC score and ISH result, with HER2-low and HER2-ultralow treated as therapeutic classifications rather than labels in pathology reports.

Watch the video highlights and read the open-access paper:
doi.org/10.1136/jcp-20…

#JClinPath #BreastPathology #HER2 #Pathology

Are our pathology guidelines keeping pace with reality?

📖 A thought-provoking proposal to make diagnostic criteria practical, accessible and globally applicable in this excellent paper by the authors - Philippe Benjamin Stephenson, Olaleke Folaranmi, José Jiménez Heffernan, and Sanjay Mukhopadhyay, published in the BMJ Journal of Clinical Pathology.

Three powerful messages:
🔹 Make criteria feasible across resource settings
🔹 Focus on what truly impacts diagnosis and treatment
🔹 Guideline panels should include community pathologists and represent the global community of pathologists

The best guidelines should not just define what is ideal—they should help every pathologist make the best possible diagnosis with the resources available.

Link to the full paper here bit.ly/4cvgo6K

Can AI predict BRAF status from H&E slides? 🔬

Lv and colleagues developed a deep learning pipeline using 272 cutaneous melanoma cases and tested it on 76 external TCGA cases, achieving an AUC of 0.915.

This is a proof of concept. External sensitivity was 86%, and negative predictive value was 68% in a cohort with 75% BRAF-mutant tumours. A negative prediction cannot rule out a mutation or justify deferring molecular testing. Prospective validation and calibration are needed before clinical use; biopsies and metastases were not studied.

Watch the video highlights and read the paper:
doi.org/10.1136/jcp-20…

#JClinPath #Melanoma #BRAF #DigitalPathology #AI

🔬 Beyond Root Cause Analysis: Rethinking Patient Safety in Histopathology

Diagnostic errors rarely arise from a single failure. They often reflect complex interactions between people, processes and systems.

This insightful review by Omar E Rakha and Emad A Rakha, proposes a practical, systems-based approach to incident investigation—shifting the focus from assigning blame to redesigning systems, preventing recurrence and driving continuous improvement.

Because safer pathology begins not just with identifying what went wrong, but with understanding why and making lasting changes.
Link to the full article here bit.ly/3VF3AVU

What does the pathologist of the future look like?

Not just an interpreter. A diagnostic architect!

Seven challenges are reshaping modern pathology:
🧬 PD-L1 → Lead biomarker stewardship
💰 NGS → Engage with reimbursement & sustainability
🤖 AI triage → Automate the routine, elevate expert interpretation
🔗 Integrated diagnostics → Break down diagnostic silos
💊 Germline testing → Turn actionable genomics into clinical practice
🧪 Clinical trials → Build biomarkers into treatment decisions
🛡️ AI assurance → Keep pathologists at the center of diagnostic quality

As diagnostics become increasingly multimodal, computational and interconnected, the pathologist’s role expands from interpreting individual tests to integrating, governing and shaping the diagnostic ecosystem.

Read the full review here: bit.ly/4AGHNNI

#JournalOfClinicalPathology #Pathology #DigitalPathology #AIinPathology #MolecularPathology #PrecisionMedicine #IntegratedDiagnostics #NextGenerationSequencing #Biomarkers #DiagnosticPathology #FutureOfPathology