This week in Pathology

Nov 21, 2026

Latest episode · 11 min
Editor’s pick

Human Epidermal Growth Factor Receptor 2 Testing and Clinical Decision-Making in Gastroesophageal Adenocarcinoma: Guideline Update From the College of American Pathologists in Collaboration With the American Society for Clinical Pathology.

A GRADE-based guideline update keeps immunohistochemistry with reflex in situ hybridization as primary HER2 testing in advanced gastroesophageal adenocarcinoma, and permits retesting at progression, including liquid biopsy.

Archives of Pathology & Laboratory Medicine · 2026 · PubMed

This week’s highlights

  1. 01Human Epidermal Growth Factor Receptor 2 Testing and Clinical Decision-Making in Gastroesophageal Adenocarcinoma: Guideline Update From the College of American Pathologists in Collaboration With the American Society for Clinical Pathology.

    A GRADE-based guideline update keeps immunohistochemistry with reflex in situ hybridization as primary HER2 testing in advanced gastroesophageal adenocarcinoma, and permits retesting at progression, including liquid biopsy.

    Bartley AN et al. · Archives of Pathology & Laboratory Medicine · 2026

    PMID 43042926

  2. 02Real-world comparison of routine and companion diagnostic mismatch repair immunohistochemistry in a large Chinese pan-tumour cohort.

    Across 1443 tumours, routine mismatch repair staining agreed less with the companion diagnostic in endometrial and miscellaneous tumours, and two-protein testing would miss deficient cases.

    Yan S et al. · Histopathology · 2026

    PMID 43059199

  3. 03Multicenter characterization of enterocolitis associated with BCMA-directed CAR T cell therapy.

    Enterocolitis after ciltacabtagene autoleucel affected about 3 percent of recipients, showed cytotoxic T cell duodenal injury despite often normal endoscopy, and 17 of 41 patients died with active disease.

    Mikkilineni L et al. · Nature Medicine · 2026

    PMID 43058222

  4. 04Length of Ischemic Time Is Critical for Accurate Determination of Homologous Recombination Capacity by Immunofluorescence Staining in Ovarian Cancer Formalin-Fixed Paraffin-Embedded Samples.

    Prolonged ischemic time lowered RAD51 signal and produced false homologous recombination deficiency calls in ovarian carcinoma, whereas fixation, ethanol storage, necrosis and hypoxia did not significantly affect scores.

    Karagöz E et al. · Archives of Pathology & Laboratory Medicine · 2026

    PMID 43042964

  5. 05Tumor Cell-Free DNA Assay Analytical Performance in 101 Clinical Laboratories Participating in a College of American Pathologists Proficiency Testing Program.

    Proficiency testing across 101 laboratories showed high cell-free DNA sensitivity and specificity, with sequencing-based assays detecting low-fraction EGFR mutations more often than non-sequencing methods.

    Lockwood CM et al. · Archives of Pathology & Laboratory Medicine · 2026

    PMID 43042949

  6. 06Assessment of Distributive DNA Methylation Profiling for Central Nervous System Tumor Classification Between Laboratories in the United States and Brazil.

    Methylation data generated in Brazil and classified by a United States clinical classifier matched histology in about 93 percent of 151 brain tumours, supporting distributive testing feasibility.

    Furtado LV et al. · Archives of Pathology & Laboratory Medicine · 2026

    PMID 43042952

  7. 07A Unified Two-Tiered Grading Scheme for Low Risk Salivary Gland Cancer: A Proposal to Simplify Pathologic Grading Based on Mitotic Activity and Tumor Necrosis.

    In 621 low-risk salivary carcinomas, high grade defined by mitoses or necrosis independently predicted worse disease-specific and recurrence-free survival in a retrospective cohort.

    Katabi N et al. · Modern Pathology · 2026

    PMID 43051593

  8. 08Diagnostic utility of p53 immunohistochemistry in Barrett's neoplasia: practice patterns and diagnostic challenges in a referral laboratory.

    Mutant p53 staining was strongly associated with definite Barrett's dysplasia but did not separate grades, and consensus review upgraded half of reviewed indefinite cases to dysplasia.

    Uthamalingam P et al. · Virchows Archiv · 2026

    PMID 43033250

  9. 09CD40 expression is highly sensitive and specific for differentiating between classic Hodgkin lymphoma and ALK-negative anaplastic large cell lymphoma.

    CD40 immunohistochemistry showed about 99 percent sensitivity and 100 percent specificity for classic Hodgkin lymphoma versus anaplastic large cell lymphoma, including PAX5-negative and null-phenotype cases.

    Spivey T et al. · Modern Pathology · 2026

    PMID 43051594

  10. 10Residual Pure Intralymphatic Breast Carcinoma After Neoadjuvant Therapy: Clinicopathologic Analysis and Comparative Outcomes Across Published Studies.

    Residual purely intralymphatic breast carcinoma after neoadjuvant therapy occurred in 13 of 2000 patients and is staged ypT0, so current frameworks may misclassify it as complete response.

    Bahmad HF et al. · Archives of Pathology & Laboratory Medicine · 2026

    PMID 43042959

Read the weekly briefing

This AudioScholar briefing is generated by artificial intelligence from PubMed abstracts for healthcare professionals and is not medical advice.

This week: an updated College of American Pathologists guideline for HER2 testing in gastroesophageal cancer, a delayed and sometimes fatal gut toxicity after BCMA CAR T therapy, and CD40 as a marker separating Hodgkin lymphoma from anaplastic large cell lymphoma.

Welcome to This Week in Pathology. We're covering 10 papers this week: 3 in depth, 4 in brief, 3 quick signals.

Archives of Pathology and Laboratory Medicine published a guideline update from the College of American Pathologists with the American Society for Clinical Pathology, led by Bartley, on HER2 testing in advanced gastroesophageal adenocarcinoma [1]. An expert panel followed National Academy of Medicine standards for trustworthy guidelines. The panel defined key questions, ran a systematic review of literature published since the 2016 joint guideline, and graded recommendations with the GRADE approach. The panel issued three strong recommendations. Six statements from the 2016 guideline went out for public comment: four were reaffirmed as good practice statements and two were archived. At initial diagnosis of advanced disease, the panel recommends testing the highest-quality tumor specimen. Immunohistochemistry, followed by in situ hybridization when indicated, remains the primary assessment of HER2 status. Genomic testing may be used at the same time or afterward for clinical decisions. For patients being considered for further therapy after progression, the panel allows testing on relapsed or recurrent tumor, and it states that tissue is preferable but liquid biopsy is acceptable. The main limitation is that a guideline is only as firm as the literature behind it, and the panel describes this field as still evolving. The update adds formal backing for retesting at progression and a defined place for genomic and liquid biopsy testing alongside immunohistochemistry.

Staying with predictive immunohistochemistry, Histopathology published a retrospective multicentre study by Yan and colleagues comparing routine mismatch repair immunohistochemistry with the Food and Drug Administration approved companion diagnostic, the VENTANA MMR RxDx Panel [2]. The cohort was 1443 resection specimens from five hospitals in China, covering colorectal, endometrial, gastric and other solid tumours. Agreement between the historical clinical-record result and the companion diagnostic was highest in colorectal and gastric carcinoma. Endometrial carcinoma had the lowest positive agreement, at about 96 percent, and the mixed other-tumour group had the lowest negative agreement, at about 93 percent. A third panel reviewed 34 discordant cases and supported the VENTANA classification in 31 of them. Subclonal loss appeared in about one in seven deficient cases, mostly involving MSH6 and MLH1. Reduced two-protein testing performed worse: the best pairing, PMS2 with MSH6, was estimated to miss 33 deficient cases compared with the four-protein classification. The design is retrospective, and three discordant cases stayed unresolved without orthogonal molecular testing. Like the HER2 guideline, the study bears on how reliably a stain selects patients for targeted or immune therapy, and it adds real-world support for full four-protein assessment and companion-diagnostic interpretation rules.

Nature Medicine published a multicenter retrospective characterization, by Mikkilineni and colleagues from the United States Multiple Myeloma Immunotherapy Consortium, of immune effector cell-associated enterocolitis after ciltacabtagene autoleucel [3]. Of 54 suspected cases from eight institutions, 41 were adjudicated as this enterocolitis, and 31 of those were confirmed by anti-camelid variable heavy-chain immunohistochemistry. Cumulative incidence was about 3 percent of 1,308 recipients, and median symptom onset came 75 days after infusion. The pathology carries much of the diagnostic weight. Upper endoscopy was normal in about two fifths of patients examined, yet duodenal biopsies showed villous blunting, apoptotic bodies and a dense cytotoxic T cell infiltrate, most often CD4 positive. A dominant T cell clone was found in 13 of 17 cases tested. The course was severe: almost three quarters of patients needed parenteral nutrition, and 17 of 41 patients died with active enterocolitis. Combined biologic and T cell-directed therapy had the highest resolution rate, but in very small subgroups. The design is retrospective, so treatment comparisons are tentative. The study adds a defined histologic picture and a tissue confirmation tool for a delayed, potentially fatal toxicity.

Now, four in brief.

Two papers in Archives of Pathology and Laboratory Medicine address assay quality. Karagöz and colleagues tested how sample handling affects a functional homologous recombination assay, which quantifies RAD51 by immunofluorescence, in retrospective and prospective cohorts of formalin-fixed high-grade serous ovarian carcinoma [4]. Prolonged ischemic time substantially reduced nuclear RAD51 signal, lowering scores and producing false calls of homologous recombination deficiency, the status that predicts PARP inhibitor sensitivity. Formalin fixation, ethanol storage, necrosis and hypoxia did not significantly change scores, and variation within a tumour was low. The finding is firm for the variable tested, and the authors advise caution with archival material that lacks handling records.

Also in Archives, Lockwood and colleagues analysed College of American Pathologists proficiency testing for tumour cell-free DNA assays, with 5088 responses from 101 laboratories on 16 variants across five oncogenes [5]. Sensitivity by gene ranged from about 87 to 100 percent, and specificity stayed near 99 percent or higher. For EGFR mutations at the lowest variant allele fractions, between 0.1 and 0.3 percent, next-generation sequencing reached about 96 percent sensitivity against about 78 percent for other methods. The data come from 2018 and 2019 and use engineered proficiency specimens, so they describe analytical rather than clinical performance.

Furtado and colleagues, also in Archives, tested distributive DNA methylation profiling for central nervous system tumours, with data generated in Brazil and classified by a United States clinical-grade classifier [6]. Across 151 tumours, confident classifications matching histology were reached in 141, about 93 percent, with ancillary testing supporting the call in six more. Preanalytic metrics were comparable to the home laboratory, and concordance did not correlate with preanalytic or methylation metrics. This is a feasibility study at one partner site, with concordance judged against histopathology and ancillary testing.

Modern Pathology published a proposal from Katabi and colleagues for a unified two-tiered grade in salivary gland carcinomas generally considered low risk, with high grade defined by a mitotic count of at least 5 per 2 square millimetres, tumour necrosis, or both [7]. In 621 retrospective cases, about a fifth of tumours were high grade, ranging from 43 percent of acinic cell carcinomas to none of the non-apocrine intraductal carcinomas. On multivariable analysis, high grade was associated with worse disease-specific, disease-free, locoregional and distant metastasis-free survival, independent of stage and other pathologic factors. The evidence is retrospective, and the scheme remains a proposal.

Three quick signals.

In Virchows Archiv, a referral-laboratory audit by Uthamalingam and colleagues of 468 patients with Barrett's oesophagus found a mutant p53 staining pattern strongly associated with definite dysplasia, though it did not reliably separate low from high grade [8]. Consensus review upgraded half of the indefinite-for-dysplasia cases reviewed to definite dysplasia. This is a retrospective audit from a single service.

Also in Modern Pathology, Spivey and colleagues report CD40 immunohistochemistry with about 99 percent sensitivity and 100 percent specificity for classic Hodgkin lymphoma against anaplastic large cell lymphoma, including PAX5-negative Hodgkin cases and ALK-negative null-phenotype cases [9]. Background staining in histiocytes needs morphologic correlation, and the authors call for comparative studies.

In Archives, Bahmad and colleagues found residual pure intralymphatic breast carcinoma in 13 of 2000 patients after neoadjuvant therapy [10]. All of these tumours were staged ypT0, so current frameworks can record them as complete response. The authors propose a new American Joint Committee on Cancer modifier for residual lymphovascular invasion. The numbers are very small.

Stepping back across this week's papers, most of the evidence concerns how reliably a test performs rather than what happens to patients after the result. Proficiency data, concordance comparisons and pre-analytic experiments establish analytical validity well. Links to outcomes come mainly from retrospective cohorts, and the guideline rests on literature its own panel calls evolving. Prospective studies tying assay choice, specimen handling and grading schemes to treatment response or survival remain scarce across these topics.

If you only have time for one paper this week, make it the College of American Pathologists HER2 guideline update for gastroesophageal adenocarcinoma [1]. It sets the reference approach for HER2 assessment at diagnosis and formally opens testing at progression, including on liquid biopsy.

That's your roundup for This Week in Pathology. The full transcript and references are available on the episode page. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And think of one colleague, in any specialty, who never has time to keep up with the literature. Tell them about AudioScholar: a ten-minute weekly for every specialty, to listen to in any podcast app, or to read at audioscholar dot C C.

This is an automated summary generated by artificial intelligence, which can make mistakes. Use it to choose which papers to read in full.

Get Pathology every week

By email, Spotify, Apple or any podcast app.

Listen on Apple PodcastsListen on Spotify

Spot something worth flagging?