This Week in Pathology — Aug 30, 2026

Welcome to This Week in Pathology. This week we're covering 10 notable papers spanning molecular biomarker interpretation and its reproducibility, classification updates in soft tissue, skin and gastrointestinal pathology, and the recurring theme of how morphology and genomics inform each other. Let's dive in.

We start with liquid biopsy, where the European Society of Pathology has published an expert opinion paper in Virchows Archiv staking out pathology's claim to blood-based genomic testing [1]. Fusco and colleagues are explicit that tissue remains the cornerstone of diagnosis, but circulating tumour DNA offers a repeatable, minimally invasive source of tumour genomic information when tissue is unavailable, insufficient, outdated, or unlikely to capture heterogeneity. They walk through two implementation models. In advanced non-small cell lung cancer, plasma testing allows rapid detection of actionable alterations and identification of acquired resistance at progression, but tissue remains essential for histological classification, PD-L1 assessment, recognising histologic transformation, and evaluating the tumour microenvironment. In metastatic breast cancer, liquid biopsy has become central to dynamic molecular management of hormone-dependent disease. The practical message for laboratories is about infrastructure rather than novelty: assay selection, awareness of clonal haematopoiesis as a confounder, attention to tumour fraction and variant allele fraction, and quality assurance. The Society flags marked inequity in access across Europe and argues that molecular biologists and bioinformaticians now belong inside the pathology laboratory, not adjacent to it.

The second theme is one every practising pathologist will recognise — the reproducibility of biomarker scoring when therapy hinges on faint staining. Also in Virchows Archiv, Hursitoglu and colleagues tackled the HER2-null, ultralow and low categories in 125 treatment-naive invasive breast carcinoma excisions, each read in two separate topographical blocks, plus 78 matched core biopsies and 58 lymph node metastases, scored independently by two blinded pathologists using the 4B5 antibody [4]. Agreement between pathologists was substantial across specimen types, but agreement between two blocks of the same tumour was only moderate — spatial heterogeneity, not just observer subjectivity, is driving the discrepancies. Concordance between the final excision status and the core biopsy was substantial, but concordance with nodal metastases was only moderate, largely because expression falls off in nodes: of 22 ultralow primaries, 15 reverted to null in the corresponding metastasis. HER2 tier showed no independent relationship with overall survival. Their pragmatic recommendation is to examine a second tumour block and obtain a second pathologist's confirmation for ultralow cases. Sitting alongside this, Modern Pathology reports the largest cohort to date of HER2-positive salivary duct carcinoma treated with trastuzumab plus docetaxel — 98 patients studied by Utsumi and colleagues [8]. Using the 2023 ASCO/CAP definition, intratumoral heterogeneity was found in about one in nine cases, and only ever in resected primary tumours; in every heterogeneous case the matched metastases were homogeneous and usually HER2-positive. Immunohistochemistry and dual-colour in situ hybridisation agreed closely, with one sarcomatoid exception. The objective response rate was about 81 percent, with median progression-free survival close to ten months and median overall survival just under four years. Critically, neither heterogeneity nor the degree of amplification predicted response — a contrast with breast and gastric cancer. The clinical implication is that finding a heterogeneous HER2 pattern in a salivary duct carcinoma primary should not be used to withhold HER2-targeted therapy.

Our third theme is classification — where criteria are being revised, validated, or defended. In the American Journal of Surgical Pathology, Lengyel and colleagues assembled the largest series of vulvar smooth muscle tumours to date, 35 cases with follow-up in 30 patients over a median of just over five years, and compared three competing schemes: the 1979 and 1996 site-specific vulvar criteria and the uterine criteria [6]. Most tumours presented as a painless mass in premenopausal patients, and the three systems agreed in 80 percent of cases. The disagreements are instructive: several tumours that the uterine criteria called smooth muscle tumours of uncertain malignant potential were labelled leiomyosarcoma by the vulvar criteria, while two tumours with tumour cell necrosis and brisk mitoses but small size and circumscribed borders were called leiomyosarcoma by uterine criteria and benign leiomyoma by both site-specific systems. Overall the data support using the uterine criteria at this site, with the caveat that the uncertain-malignant-potential category still needs further study. Staying with the same journal, Zhang and colleagues addressed a problem that generates real diagnostic anxiety — the diverted rectum [7]. In a blinded, rectum-exclusive comparison of 63 diverted rectal resections, 17 without inflammatory bowel disease and 46 with, graded across five histologic domains, de novo diversion colitis was uncommon in the non-inflammatory bowel disease group, occurring in about one in six specimens, with most of those cases showing only diversion change. By contrast, 43 of 46 inflammatory bowel disease cases showed a chronic colitis pattern and scored higher across all five parameters, and prediversion distal margin involvement was frequent. Ulcerative colitis specimens were more severely affected than Crohn's disease. The authors argue that in most cases persistent inflammatory bowel disease, not severe diversion colitis, is the better interpretation, and they propose a three-tier reporting vocabulary after clinicopathological correlation: diversion change, diversion colitis, and inflammatory bowel disease with superimposed diversion changes. Two reviews round out this theme. Histopathology carries an update on cutaneous lymphoma from Kempf and Mitteldorf, tracking the fifth-edition WHO classifications and the International Consensus Classification, including the promotion of several formerly provisional entities and the renaming of primary cutaneous CD8-positive acral T-cell lymphoma to a lymphoproliferative disorder on the strength of its excellent prognosis — a reminder that nomenclature drives treatment intensity and patient counselling [2]. And in Virchows Archiv, Holm and colleagues review duodenal and ampullary tumours, emphasising that assigning origin depends on locating the tumour centre and therefore on meticulous grossing, that lineage-specific antibody panels help subtype ampullary carcinomas and guide chemotherapy choice, and that routine molecular profiling for predictive biomarkers is now warranted [5].

Finally, three papers illustrate morphology and immunohistochemistry as gatekeepers to molecular information. Human Pathology offers a review from Anderson-Calleja and colleagues on the morphologic correlates of genetically defined acute myeloid leukaemia subtypes, making the case that reproducible bone marrow and blood findings can raise early genetic suspicion, guide triage of ancillary testing, and anticipate biologic behaviour before cytogenetics returns [3]. In Modern Pathology, Kasajima and colleagues characterised 199 resected pancreatic neuroendocrine tumours and found that about one in ten narrowed the intrapancreatic bile duct [9]. These lesions were exclusively non-functioning, more common in women, had higher Ki-67 indices, and — when involving the lower, periampullary bile duct — shorter progression-free survival. Nearly all expressed PDX1, most lower duct lesions expressed CDX2, and gastrin and somatostatin expression was common while insulin and glucagon were absent, a profile resembling duodenal neuroendocrine tumours and periampullary glands rather than conventional pancreatic tumours, suggesting a non-islet origin. Also in Modern Pathology, Shi and colleagues examined albumin messenger RNA in situ hybridisation across 63 cholangiocarcinomas and 157 tumours of other origins [10]. Nearly all cholangiocarcinomas carrying FGFR fusions or IDH1 and IDH2 mutations — the alterations with approved targeted therapies — showed strong albumin positivity, whereas tumours with BAP1, ARID1A or PBRM1 alterations were less often positive and stained more weakly. That raises the possibility of using a widely available in situ hybridisation assay to help flag which liver tumours are most likely to harbour an actionable target.

If you only have time for one paper this week, make it the HER2 reproducibility study in Virchows Archiv [4]. Ultralow HER2 now determines eligibility for antibody-drug conjugate therapy, and this paper gives you two concrete, immediately actionable steps — a second block and a second reader — to reduce the misclassification that decides who gets treated.

Here are the key takeaways from this week in Pathology. First, minimal HER2 expression is unstable across tumour blocks and drops off in nodal metastases, so score a second block and get a second opinion on ultralow cases. Second, in salivary duct carcinoma, HER2 heterogeneity in the primary does not predict failure of trastuzumab-based therapy and should not be used to deny treatment. Third, the European Society of Pathology positions circulating tumour DNA as an extension of the pathologist's role, with tissue still required for classification, PD-L1, transformation and microenvironment assessment. Fourth, for vulvar smooth muscle tumours the uterine criteria outperform the older site-specific systems. Fifth, in a diverted rectum with known inflammatory bowel disease, chronic active colitis usually means persistent disease rather than diversion colitis — and report it using explicit, context-aware terminology. And finally, albumin in situ hybridisation may help identify cholangiocarcinomas most likely to carry FGFR or IDH alterations where molecular testing is not readily available.

That's your roundup for This Week in Pathology. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. 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 free 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. Always review the original source materials.