Publications by Year: 2026

The transition from trainee to independent investigator is one of the most challenging and formative phases of a scientific career. It requires not only scientific expertise but also the skills to lead, mentor, manage, and communicate effectively. The Arthur and Sandra Irving Cancer Immunology Symposium serves as a platform for established investigators to mentor trainees and early-career faculty as they navigate this transition to independence. Through sharing personal experiences and lessons from their own careers, senior leaders provide guidance on the scientific, professional, and personal challenges that shape a successful career in cancer immunology-emphasizing how curiosity, persistence, and a translational mindset can make a lasting real-world impact. This commentary highlights key themes, including leadership, communication, recruitment, and fundraising. Altogether, these insightful thoughts provide a framework for the next generation of cancer immunologists as they establish their independent careers as future leaders in the field.

OBJECTIVE: Focal hand dystonia (FHD) severely impairs task-specific motor control, yet the optimal surgical target for stereotactic intervention remains uncertain. This study aimed to identify the precise thalamic lesion site associated with symptomatic improvement and to clarify its network connectivity.

METHODS: We retrospectively analyzed 164 patients with FHD (mean age = 42.0 years, 26.2% women) who underwent stereotactic thalamotomy of the ventral lateral thalamus. Voxel-wise probabilistic lesion mapping was applied to relate lesion locations to clinical outcomes. Structural connectivity analyses assessed fiber tracts linked to the optimal lesion site. Model performance was evaluated by 10-fold cross-validation, validation in an out-of-sample cohort, and testing in reoperation cases.

RESULTS: We identified that lesioning the border zone between the ventralis oralis posterior (Vop) and ventralis intermedius (Vim) nuclei was associated with improvement of FHD. The predictive model achieved high accuracy in cross-validation (area under the curve [AUC] = 0.836) and performed robustly in independent validation. Connectivity analyses showed that the Vop-Vim border zone was linked to cerebellothalamic and pallidothalamic afferents as well as thalamocortical projections to the supplementary motor area and premotor cortex. In contrast, lesions extending into the ventralis oralis anterior nucleus were associated with an increased risk of motor complications.

INTERPRETATION: Precise targeting of the Vop-Vim border maximizes clinical benefit while minimizing adverse effects in FHD thalamotomy. These findings establish the first evidence-based thalamic target for FHD, offering practical guidance for stereotactic interventions and advancing understanding of dystonia pathophysiology. ANN NEUROL 2026.

BACKGROUND: People experiencing homelessness are disproportionately affected by opioid use disorder (OUD), yet treatment engagement remains low. Understanding opioid-related health beliefs in this population may inform more tailored interventions.

OBJECTIVE: To characterize opioid-related health beliefs among homeless-experienced individuals with OUD and examine their alignment with opioid use severity.

METHODS: We conducted a cross-sectional survey of individuals initiating opioid treatment at a homeless healthcare program in Boston, Massachusetts. The survey assessed key Health Belief Model domains related to opioid use and incorporated the World Health Organization's Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST) to quantify opioid use severity. We used nonparametric trend tests to examine associations between ASSIST quartiles (Q1 ≤ 23.5, Q2 23.6-33.0, Q3 33.1-38.0, and Q4 > 38.0) and belief domains.

RESULTS: Among 148 participants, the mean age was 41.7 years; 51.4% identified as White, 16.2% as Black, and 38.5% as Hispanic. Higher ASSIST quartiles, indicating higher opioid use severity, were significantly associated with greater perceived threat, including being very worried about opioid use (25% in Q1 vs. 72.7% in Q4; p = 0.001), perceiving very high overdose risk (4.2% vs. 40.9%; p = 0.006), and viewing overdose as very dangerous (87.5% vs. 100%; p = 0.036). Higher ASSIST quartiles were also significantly associated with greater perceived benefits, including that abstaining from opioids (78.3% vs. 100%; p = 0.004) and taking MOUD (69.6% vs. 95.2%; p = 0.016) would improve quality of life.

CONCLUSIONS: Individuals with greater opioid use severity endorsed higher perceived opioid-related threats and treatment-related benefits. Interventions that highlight these beliefs may help enhance treatment engagement in this high-risk population.

BACKGROUND: The use of artificial intelligence (AI) in medicine has surged. Given the sensitive nature of palliative care, it is crucial to apply AI tools in a patient-centered and ethical manner.

AIM: We sought to understand the ethical implications of implementing AI-based tools in palliative care from various stakeholders.

DESIGN: Using a framework of ethics of care, we conducted thematic analysis to identify emerging themes.

SETTING/PARTICIPANTS: We recruited and interviewed 22 participants: six patients, four care-partners, five clinicians, three bioethicists, and four clinician-bioethicists, recruited through a single hospital clinic for patients and care-partners and purposeful snowball sampling for clinicians and bioethicists based on their knowledge and interest in AI-based tools. The mean age of participants was 48 years old; 64% were male. Clinician participants had practiced for a mean of 11 years, and ethicists for 15 years. We aimed to recruit clinicians and bioethicists from multiple states and specialties.

RESULTS: From the interviews we extracted five main themes: (1) Primacy of the doctor-patient relationship over AI performance; (2) Humans have intuition and nuance that AI lacks; (3) Agreement about the importance of oversight of AI tools; (4) New AI technologies should include a process for patient education; and (5) AI increases efficiency, scalability, and a more unified approach to serious illness.

CONCLUSIONS: When building and implementing AI-based tools, we recommend: establishing oversight committees; reflecting on the unique contributions of humans to care; proactively educating patients and contextualizing the tools; and ensuring data use is restricted to clinical care.

Astrocytes are glial cells in the brain essential for maintaining neural homeostasis, modulating synaptic activity through gliotransmission, and supporting metabolic processes. As part of Alzheimer's disease (AD) progression, astrocytes undergo significant morphological and functional changes, transitioning to reactive states that can contribute to both neuroprotection and neurodegeneration. This review aims to summarize current knowledge on the roles of astrocytes in AD, focusing on their contributions to amyloid-β (Aβ) and tau pathologies, neuroinflammation, disrupted calcium signaling, and age-related changes. We synthesized findings from published studies investigating astrocytic sodium channels (Nav1.6), key molecular pathways such as apolipoprotein E (ApoE), oxidative stress, and excitatory amino acid transporter 2 (EAAT2), as well as emerging astrocytic biomarkers including GFAP, YKL-40, and MAO-B. Optogenetic studies and other experimental approaches with high spatiotemporal resolution were also considered to understand astrocyte involvement in circuit impairments and sleep deficits in AD. Astrocytes in AD exhibit altered calcium signaling, impaired gliotransmission, and dysregulated sodium channel activity. Reactive astrocytes influence Aβ and tau pathology, contribute to neuroinflammation, and show altered biomarker expression. Molecular dysfunctions, including changes in ApoE, EAAT2, and oxidative stress pathways, exacerbate disease progression. Emerging therapeutic strategies targeting astrocytic pathways, such as siRNA therapy and gene editing, show promise for mitigating these pathological changes. Understanding the complex roles of astrocytes in AD highlights their dual protective and detrimental functions and identifies novel avenues for therapeutic intervention. Targeting astrocytic dysfunction may offer strategies to slow disease progression and improve cognitive outcomes.

BACKGROUND & AIMS: Metabolic dysfunction-associated steatotic liver disease (MASLD) spans from simple steatosis to metabolic dysfunction-associated steatohepatitis (MASH) and can progress to cirrhosis or hepatocellular carcinoma. Despite its prevalence, effective therapies are lacking. Recent genome-wide association studies identified a common missense variant (rs2642438) in the Mitochondrial Amidoxime Reducing Component 1 (MTARC1) gene that protects against liver cirrhosis without increasing cardiovascular disease risk. Biochemical and disease risk signatures associated with carriers of this missense variant also aligned with those of a known loss-of-function MTARC1 variant, suggesting mARC1 inhibition as a potential MASLD treatment.

METHODS: To validate mARC1 loss-of-function as protective against MASLD, we generated Mtarc1 knockout (KO) mice and placed them on a choline-deficient, L-amino acid-defined, high-fat diet (CDAHFD). Effects of Mtarc1 KO on obesity and type 2 diabetes were explored using a high-fat diet. Hepatocytes from Mtarc1 KO mice were isolated to explore the molecular mechanisms by which Mtarc1 KO impacts lipid metabolism.

RESULTS: Mtarc1 KO mice exhibited no vital growth or development defects. With a high-fat diet-induced obesity model, obese Mtarc1 KO mice exhibited reduced liver mass and lower cholesterol levels, with no effect on glucose homeostasis. In a CDAHFD-induced MASLD model, mARC1 deficiency significantly reduced liver steatosis, profibrosis, and inflammation. Untargeted metabolomics profiling further showed hepatic enrichment of phospholipids in Mtarc1 KO mice. Primary hepatocytes isolated from Mtarc1 KO mice exhibited reduced lipid droplet accumulation, decreased fatty acid uptake, and increased lipid secretion.

CONCLUSIONS: These findings support mARC1 inhibition as a promising therapeutic strategy for MASLD/MASH.

OBJECTIVE: To explore the population level trends in new incidence of uveitis diagnoses as they relate to seasonal variation across the United States.

DESIGN: Retrospective cross-sectional study.

SUBJECTS: Subjects with uveitis in the TriNetX Research USA "No Date Shift" network between 2016 and 2024.

METHODS: Subjects with uveitis as identified by International Classification of Disease -10 encounter diagnoses codes were collected. They were further stratified based on anatomic location (anterior, intermediate, panuveitis, posterior, scleritis, unspecified/systemic) or etiology (infectious, noninfectious, idiopathic). Uveitis diagnosis was calculated as incident cases by season from a baseline ophthalmology patient population and further analyzed using analysis of variation (ANOVA) to isolate significant trends.

MAIN OUTCOME MEASURES: Incidence and difference between group means of new uveitis diagnosis stratified by season, anatomic, and etiologic classification.

RESULTS: ANOVA post-hoc comparisons between seasons showed no association between all permutations of seasonal variations. When stratified by etiology, infectious uveitis had significantly lower incidence in the spring months compared to summer (Mean -3.75, 95% CI -7.49 to -0.02, p = 0.049).

CONCLUSIONS: This study presents a novel observation that infectious uveitis may vary based on season. Prospective studies are needed to verify these observations.

BACKGROUND: Survival for patients with high-risk neuroblastoma remains poor despite current-era multimodality treatment that includes postconsolidation GD2-directed immunotherapy. Given the promising responses in patients who receive dinutuximab-based chemoimmunotherapy in the relapsed setting, the Children's Oncology Group ANBL19P1 study evaluated the feasibility of administering irinotecan, temozolomide, dinutuximab, and sargramostim after frontline consolidation with tandem autologous stem cell transplantation (ASCT).

METHODS: Patients with high-risk neuroblastoma who received induction therapy followed by tandem ASCT and had no evidence of progressive disease (PD) were eligible. Treatment included five 28-day cycles of temozolomide and irinotecan (days 1-5), dinutuximab (days 2-5), and sargramostim (days 6-12). Isotretinoin (days 8-21) was given during cycles 1-6. Therapy was deemed feasible if the 95% confidence interval placed on the percentage of patients that completed five cycles of chemoimmunotherapy without PD within 30 weeks contained 75% in the absence of excessive toxicity. Event-free survival and overall survival were determined from the time of enrollment.

RESULTS: Forty eligible patients enrolled, and 35 (87.5%; 95% confidence interval, 73.9%-94.5%) completed five cycles without PD within 30 weeks, meeting the feasibility threshold. No unacceptable toxicities occurred on protocol therapy, including no toxic deaths. Five patients discontinued therapy early because of physician determination (n = 2), patient/parent refusal of further therapy (n = 2), and PD (n = 1). The 2-year event-free and overall survival rates were 82.5% ± 6.1% and 92.5% ± 4.2%, respectively.

CONCLUSIONS: The administration of chemoimmunotherapy in the postconsolidation setting after tandem ASCT is feasible and tolerable. Future studies will be needed to define the population most likely to benefit from this augmented postconsolidation therapy.

BACKGROUND: Traumatic brain injuries (TBI) account for 30–40% of global injury mortality. In India, high out-of-pocket expenditure (OOPE; 58.7% of total health costs) exacerbates catastrophic health expenditures (CHE). This study estimates OOPE, along with the risk of CHE and distress financing, in patients with isolated TBI.

METHODS: We conducted a prospective cohort study across two centers, enrolling patients with isolated TBI over one month. Direct (medical and non-medical) and indirect (wage loss) costs were recorded using medical diaries. 3 months post discharge rehabilitation expenses were recorded.

RESULTS: Eighteen participants (mean age: 43 years, eleven males) were included; four had health insurance, and seven were primary earners. The mean OOPE was INR 88,485 comprising INR 58,319 in direct and INR 30,166 in indirect costs. CHE was observed in fifteen participants and distress financing in eleven participants.

CONCLUSION: Despite subsidized care, the substantial OOPE for isolated TBI underscores the financial burden, and the imperative for equitable, sustainable healthcare policies.

SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12962-026-00716-1.

Since 2022, artificial intelligence (AI) methods have progressed far beyond their established capabilities of data classification and prediction. Large language models (LLMs) can perform logical reasoning, enabling them to plan and orchestrate complex workflows. By using this planning ability and equipped with the ability to act upon their environment, LLMs can function as agents. Agents are (semi-)autonomous systems capable of sensing, learning and acting upon their environments. As such, they can interact with external knowledge or external software and can execute sequences of tasks with minimal or no human input. In cancer research and oncology, evidence for the capability of AI agents is rapidly emerging. From autonomously optimizing drug design and development to proposing therapeutic strategies for clinical cases, AI agents can handle complex, multistep problems that were not addressable by previous generations of AI systems. Despite rapid developments, many translational and clinical cancer researchers still lack clarity regarding the precise capabilities, limitations, and ethical or regulatory frameworks associated with AI agents. Here we provide a primer on AI agents for cancer researchers and oncologists. We illustrate how this technology is set apart from and goes beyond traditional AI systems. We discuss existing and emerging applications in cancer research and address real-world challenges from the perspective of academic, clinical and industrial research.