Citations

Abstract Large language models (LLMs) and vision‐language models represent a fundamentally different category of artificial intelligence (AI) compared to prior image analysis approaches in digital pathology, which have largely been based on convolutional neural network architectures. This review from the American Society of Cytopathology Clinical Practice Committee examines the current evidence for LLM and vision‐language model applications in cytopathology, including structured reporting, diagnostic assistance, quality control, education, and workflow integration. The distinction between applications with preliminary evidence and those that remain hypothetical is described. A detailed assessment of the challenges that must be addressed before clinical deployment, including hallucination risk, limited explainability, bias, data privacy, validation gaps, and infrastructure barriers is discussed. A review of the regulatory landscape in the United States and European Union as it applies to AI‐enabled software as a medical device is provided. Recommendations addressing cytopathology‐specific benchmarks, multi‐institutional validation, transparent governance, and incremental deployment beginning with low‐risk applications are suggested. In the current environment, LLMs have the potential to augment cytopathology practice, but responsible adoption requires rigorous validation and sustained collaboration among cytopathologists, AI researchers, and regulatory bodies.

Background Systemic corticosteroids (SCS) remain the cornerstone of eosinophilic granulomatosis with polyangiitis (EGPA) and hypereosinophilic syndrome (HES) therapy but are associated with substantial toxicity. Mepolizumab, an anti-interleukin-5 biologic, has demonstrated steroid-sparing effects in addition to EGPA and HES disease control, yet evidence of reductions in SCS-related complications is limited. Methods This retrospective cohort study used US claims data from the Komodo Research Database to quantify SCS-related complications, healthcare resource utilization, and costs in patients with EGPA or HES treated with mepolizumab vs. chronic SCS. Patients were stratified as mepolizumab (300 mg) or chronic SCS users (≥6 months continuous SCS use with >7.5 mg/day for EGPA and >10 mg/day for HES) based on treatment received during a 6-month landmark period post-index (first claim of mepolizumab or SCS). An inverse probability of treatment weighting approach was applied to the cohorts to minimize confounding. Outcomes were assessed up to 12 months post-landmark. Results 578 patients with EGPA (305 mepolizumab users vs. 273 chronic SCS users) and 272 patients with HES (161 vs. 111) were included. In patients with EGPA, rates of any SCS-related complications were 39% lower with mepolizumab compared with chronic SCS (rate ratio [95% confidence interval]: 0.61 [0.49, 0.76], p < 0.001). Mepolizumab users had 57% lower rate of all-cause hospitalizations (0.43 [0.27, 0.69], p < 0.001), and lower outpatient and emergency department (ED) visits vs. chronic SCS. Patients with HES had significantly lower rates of any SCS-related complications with mepolizumab vs. chronic SCS (0.59 [0.42, 0.82], p = 0.002). Mepolizumab users had 66% lower rates of all-cause hospitalizations (0.34 [0.17, 0.65], p = 0.001), and lower outpatient visits relative to chronic SCS users. In EGPA and HES SCS-complication-related hospitalizations and ED visits were lower with mepolizumab compared with chronic SCS. SCS-complication-related medical costs per patient per year were $16,194 ( p = 0.008) and $3,976 ( p = 0.676) lower, in the mepolizumab vs. chronic SCS cohort of patients with EGPA and HES, respectively. Conclusion This first of its kind study suggests that real-world mepolizumab use is associated with reduced SCS-related complications in EGPA and HES, reinforcing its clinical and economic steroid-sparing benefits in reducing healthcare burden among EGPA and HES populations.

Background: Polyetheretherketone (PEEK) is a commonly used material for cranioplasties due to its advantages over other materials and autologous bone grafts. PEEK has an increased malleability, which increases postoperative aesthetics and adapts well to the cranial contours; it is also associated with a lower rate of infection when compared with autologous cranioplasties. Despite these benefits, there is limited information on the practical aspects of modifying PEEK cranioplasties after their initial placement. Case Series: We present 2 cases of patients who had previously undergone PEEK cranioplasty being treated with deep-brain stimulation (DBS), which was placed through their PEEK implant. The first was an 18-year-old male who had previously undergone cranial vault closure with bone grafts following a craniotomy. Later, when a deep-brain stimulator was planned, it was discovered that the bone grafts had significant resorption and a lack of structural integrity. Therefore, bone grafts were replaced with a custom PEEK implant cranioplasty before the DBS electrodes were placed via drilling through the PEEK implant to access the brain at a later surgery. The second patient was a 39-year-old male who developed epilepsy after a stroke, requiring craniotomy. DBS was unable to be performed initially due to significant resorption of the patient’s native bone. Therefore, a PEEK cranioplasty was performed in preparation for a staged, subsequent placement of DBS electrodes through the implant. Both patients recovered uneventfully following surgery, with noticeable improvement in functional (reduction in seizure frequency) and aesthetic outcomes. Systematic Review: A literature review was conducted using PubMed with keywords related to PEEK cranioplasties. After screening titles, abstracts, and full-text papers, it was found that 3 previously documented cases have involved drilling through PEEK implants: one by Mallela et al, involving SEEG electrode placement; one by Marrone et al, concerning a burr hole for evacuating a chronic subdural hematoma; and one by Ledesma et al, regarding the placement of subdural grid electrodes. Conclusions: This case series highlights how PEEK cranioplasties can be drilled through for therapeutic procedures, such as DBS electrode implantation, without compromising their structure. These cases illustrate the versatility, safety, and adaptability of PEEK implants for additional necessary interventions.