This Week in Pathology — May 14, 2026

Welcome to This Week in Pathology. This week we're covering 10 notable papers spanning advances in cancer pathology, including a new CAR T-cell target for glioblastoma and a critical safety signal in gene therapy. We'll also explore the challenges in neuropathology diagnostics for conditions like CTE and autism, and finally, we'll look at novel insights into immune-mediated diseases and the new tools making these discoveries possible. Let's dive in.

We begin this week in oncology, with a trio of papers highlighting therapeutic innovation and a crucial note of caution.

First, in glioblastoma, a historically difficult-to-treat tumor, a new target for CAR T-cell therapy has been identified. A study in *Science Translational Medicine* used multiomic analysis of patient-derived primary and recurrent glioblastoma cell lines to pinpoint the urokinase plasminogen activator receptor, or uPAR [9]. The authors found that uPAR was highly expressed in recurrent glioblastoma and appeared to be a marker of brain tumor-initiating cells. When they generated uPAR-specific CAR T-cells, they observed potent antitumor activity in patient-derived xenograft models. Interestingly, the therapeutic effect wasn't just from direct tumor cell killing. uPAR was also expressed on glioblastoma-associated macrophages, allowing the CAR T-cells to target both the cancer itself and key cells within the tumor microenvironment.

Sticking with the theme of the tumor microenvironment, a paper in *Nature* provides new insights into triple-negative breast cancer, or TNBC [3]. Researchers performed single-cell and spatial transcriptomic analysis on tumors from over 100 patients before they received neoadjuvant chemotherapy. They identified eight distinct cellular communities, which they termed 'ecotypes', based on the co-occurrence of cancer cells and various immune and stromal cell types. While past studies have often focused on T-cells, this work highlights the importance of macrophage subtypes and specific cancer-cell gene programs related to interferon signaling and cell cycle activity. These factors were strongly associated with a good response to chemotherapy, offering a more nuanced view of why some patients with this aggressive cancer respond well to treatment while others do not.

While these papers highlight progress, a case report in *The New England Journal of Medicine* serves as a critical reminder of the potential risks of novel therapies [8]. The report details the case of a 5-year-old boy with severe mucopolysaccharidosis type I who had received intracisternal AAV9 gene therapy four years prior. The child developed a neuroepithelial tumor. Molecular analysis of the resected tumor showed clonal integration of the adeno-associated virus vector directly into the PLAG1 gene, a known oncogene, resulting in a chimeric AAV-PLAG1 transcript. This case provides direct evidence of AAV vector-mediated oncogenesis in a human, a risk previously observed in neonatal mouse models, and will be a crucial data point for the safety evaluation of gene therapies moving forward.

Next, we turn to neuropathology, where two papers tackle the complexities of diagnosis and etiology, framed by a commentary on the challenge of overlapping diseases.

First, a study in *Nature Medicine* raises serious questions about the clinical diagnosis of chronic traumatic encephalopathy, or CTE [1]. Currently, CTE can only be definitively diagnosed postmortem. To bridge this gap, consensus criteria were developed for a clinical diagnosis called traumatic encephalopathy syndrome, or TES. This study evaluated the diagnostic accuracy of these TES criteria in a brain bank cohort of over 1,000 individuals. Of the 25 cases that met clinical criteria for TES, only 6 actually had CTE pathology on autopsy. This translates to a positive predictive value of just 24 percent. The authors found that a diagnosis was driven more by a history of repetitive head impacts than by the specific core or supportive clinical features of TES, which were just as common in cases without CTE. This poor performance raises substantial concern that athletes may be incorrectly diagnosed with a progressive neurodegenerative disease.

These findings are a stark illustration of a broader issue discussed in a *Science* commentary: the challenge of 'copathology' in dementia [2]. The piece notes that clinicians and researchers are increasingly aware that multiple disease processes often overlap in a single patient, leading to nuanced and often non-specific clinical presentations. The CTE study shows this in action, where symptoms thought to be characteristic of TES were not specific for underlying CTE pathology. This growing recognition is pushing the field toward more sophisticated testing and treatment strategies that account for this complexity.

On the basic science front, a paper in *Nature* identifies a new genetic factor in autism spectrum disorder, or ASD [4]. By analyzing whole-genome sequencing data from thousands of individuals, researchers implicated an X-linked long non-coding RNA called *PTCHD1-AS* as an ASD-susceptibility gene. Microdeletions affecting this gene were associated with more than double the risk for ASD in males. To explore its function, they created knockout mouse models. Male mice lacking *Ptchd1-as* showed core ASD-like features, such as repetitive behaviors and impaired social interaction, but notably, they did not have the cognitive comorbidities or ADHD-like behaviors often seen with other ASD-linked genes. The study traced the dysfunction to the dorsal striatum, where loss of the non-coding RNA altered myelination and synaptic plasticity.

Finally, we'll cover new insights into immune-mediated diseases and the powerful tools that are enabling these discoveries.

A review in *The New England Journal of Medicine* provides a comprehensive update on inflammatory myopathies [7]. It outlines the classification of these autoimmune diseases into five major subtypes: inclusion-body myositis, immune-mediated necrotizing myopathies, antisynthetase syndrome, overlapping myositis, and dermatomyositis. The review emphasizes that these are not a monolithic entity; each has distinct clinical features, prognoses, and pathomechanisms. A key point for practicing pathologists and clinicians is the central role of myositis-specific autoantibodies, which are critical for accurate diagnosis, subtyping, and prognosis, and are paving the way for more targeted therapies.

Illustrating this move toward targeted therapy, a study in *Science Translational Medicine* uncovers a key immune evasion mechanism in pulmonary fibrosis [10]. The authors investigated why the immune system fails to clear senescent fibroblasts, a process thought to drive fibrosis. Using single-cell RNA sequencing, they found that senescent fibroblasts in fibrotic lungs express high levels of HLA-E. This molecule acts as a ligand for NKG2A, an inhibitory checkpoint receptor on natural killer, or NK, cells. This interaction effectively puts the brakes on the NK cells, creating an immune-privileged niche for the senescent cells. The exciting therapeutic implication is that blocking this axis could restore immune function. Indeed, in a mouse model, blocking NKG2A led to clearance of senescent fibroblasts and resolution of fibrosis. Furthermore, the clinical-grade NKG2A inhibitor, monalizumab, was able to reactivate patient-derived NK cells to kill human senescent fibroblasts in vitro, suggesting a promising strategy for diseases like idiopathic pulmonary fibrosis.

These biological discoveries are increasingly powered by novel high-throughput methods. Two papers in *Cell* introduce such tools. First, a method called VIS-seq, for variant in situ sequencing, provides a new way to assess the functional impact of genetic variants [5]. Instead of a single functional readout, VIS-seq is an image-based method that captures high-dimensional morphological profiles of cells. Applied to thousands of variants in the *LMNA* and *PTEN* genes, it could measure complex changes in protein abundance, localization, and cell architecture simultaneously. This multi-dimensional view allowed it to distinguish, for example, between autism-linked and cancer-linked variants in *PTEN*, which often elude simpler assays.

Second, a computational framework called D-SPIN was developed to make sense of large-scale single-cell RNA-sequencing data from perturbation experiments [6]. D-SPIN can build regulatory network models that explain how stimuli, like drugs or gene knockdowns, reconfigure cellular interactions and change cell fates. The authors demonstrated its power by modeling how combinations of immunomodulatory drugs work together and even simulating immune cell population shifts at unobserved drug dosages, providing a powerful tool for understanding and predicting cellular responses.

If you only have time for one paper this week, make it the study in *Nature Medicine* on the diagnostic accuracy of traumatic encephalopathy syndrome criteria [1]. Its finding that the clinical criteria for CTE have a positive predictive value of only 24 percent is a crucial, practice-changing piece of information for any clinician counseling patients with a history of head trauma and cognitive or behavioral symptoms.

Here are the key takeaways from this week in Pathology.

First: Be highly skeptical of a clinical diagnosis of traumatic encephalopathy syndrome, or TES. New data show a positive predictive value of only 24% for underlying CTE pathology, with the diagnosis driven more by head impact exposure history than by specific clinical symptoms.

Second: In recurrent glioblastoma, the uPAR receptor is an emerging dual-action CAR T-cell target, as it's expressed on both the tumor cells and tumor-associated macrophages.

Third: A key immune evasion mechanism in pulmonary fibrosis has been identified, where senescent fibroblasts use the HLA-E/NKG2A checkpoint to suppress NK cells. Blocking this axis with drugs like monalizumab is a promising therapeutic strategy.

Fourth: While AAV gene therapies are advancing, a case report from *The New England Journal of Medicine* confirmed the risk of insertional oncogenesis in a human patient, after an AAV vector integrated into the *PLAG1* oncogene, causing a tumor.

Fifth: As reviewed in *The New England Journal of Medicine*, accurate subtyping of inflammatory myopathies using myositis-specific autoantibodies is essential, as the five major subtypes have different prognoses and require different management strategies.

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.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.