The Hematopoietic Cell Transplantation-Specific Comorbidity Index (HCT-CI) assigns a high-risk score to patients who develop secondary hematologic malignancies following solid tumors, indicating an increased risk of non-relapse mortality (NRM). This study aimed to evaluate the impact of prior solid tumors on outcomes after hematopoietic stem cell transplantation (HSCT).

From a cohort of 2,382 patients who underwent HSCT for acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), or myelodysplastic syndrome (MDS) between January 2014 and July 2024, we included 43 (1.8%) with a history of prior solid tumors and 82 matched controls for analysis by 1:2 propensity score matching.

The solid tumor cohort predominantly comprised breast cancer (48.8%). With a median follow-up of 31.0 months, only one patient exhibited post-transplant relapse or metastasis of the solid tumor. Compared to the control group, patients with solid tumors exhibited higher ECOG scores (≥ 2: 23.1% vs. 9.5%, P = 0.049), lower platelet counts (35.5 vs. 72×10⁹/L, P = 0.010), a higher incidence of complex karyotypes (16.3% vs. 3.7%, P = 0.031). No significant differences were noted in 3-year overall survival (OS) (64.3% vs. 71.9%, P = 0.468), leukemia-free survival (LFS) (57.6% vs. 70.8%, P = 0.218), graft-versus-host disease/relapse-free survival (GRFS) (43.3% vs. 53.0%, P = 0.359) and NRM (23.9% vs. 11.7%, P = 0.246). In an exploratory landmark analysis, the solid tumor cohort appeared to have significantly lower OS (P = 0.030), LFS (P = 0.009), and GRFS (P = 0.038) from 2 years after transplantation. Multivariable analysis identified age greater than 55 years, baseline platelet counts less than 50×10⁹/L as significant predictors of inferior OS and LFS in solid tumor patients.

Patients with hematologic diseases secondary to solid tumors showed no significant increase in overall transplantation risk. However, their adverse clinical characteristics and reduced long-term survival rates beyond 2 years post-transplantation, underscore the need to refine HCT-CI scoring and improve management strategies.

While immune-related adverse events (irAEs) are associated with better prognosis in advanced esophageal squamous cell carcinoma (ESCC), the prognostic impact of single-organ irAE (uni-irAE), multiple-organ irAEs (multi-irAEs), and organ-specific irAEs remains poorly understood. This study aimed to address this gap by evaluating the effects of various irAEs on survival and characterizing the co-occurrence patterns of multi-irAEs in ESCC patients.

We retrospectively analyzed 213 ESCC patients treated with immune checkpoint inhibitor (ICI), dividing them into non-irAE, uni-irAE, and multi-irAEs groups to compare their efficacy and prognosis. Baseline characteristics and efficacy outcomes were compared by Chi-square test. Prognostic analysis was performed using Kaplan-Meier survival analysis with the log-rank test and Cox proportional hazard models. The Mann-Whitney U test was used to compare the time to onset of irAEs. Additionally, logistic regression analysis was conducted to identify risk factors associated with the development of multi-irAEs.

Patients who developed irAEs exhibited a significantly higher disease control rate (DCR) compared to patients without irAEs (94.9% vs. 82.1%, p = 0.007). This was observed in both the uni-irAE group (93.4% vs 82.1%, p = 0.036) and as a trend in the multi-irAEs group (100% vs. 82.1%, p = 0.078) when compared to the non-irAE group. Multivariate analysis revealed that the development of uni-irAE was an independent protective factor for both progression-free survival (PFS; hazard ratio [HR] 0.57, 95% confidence interval [CI] 0.39-0.83, p = 0.003) and overall survival (OS; HR 0.64, 95% CI 0.44-0.95, p = 0.028). Similarly, multi-irAEs were identified as an independent protective factor for OS (HR 0.41, 95% CI 0.20-0.86, p = 0.019). Analysis of co-occurrence patterns showed that endocrine irAEs were frequently leading to multi-irAEs. Furthermore, a multivariate Cox regression confirmed that endocrine irAEs and mild (grade 2 or lower) irAEs were independently associated with favorable survival outcomes.

The occurrence of both uni-irAE and multi-irAEs was associated with favorable prognosis in ESCC patients treated with ICIs. Furthermore, patients who developed endocrine irAEs or mild irAEs also demonstrated improved efficacy, suggesting their potential as clinical response markers for a positive response to therapy. This finding emphasizes the necessity of vigilant monitoring and early intervention for irAEs in patients undergoing ICIs.

Pancreatic ductal adenocarcinoma (PDAC) is a lethal malignancy characterized by significant heterogeneity. We conducted a multi-omics integrated clustering analysis to categorize PDAC molecular subtypes.

Multi-omics data from The Cancer Genome Atlas-Pancreatic Adenocarcinoma (TCGA-PAAD) were integrated using ten clustering algorithms. Comparisons across PDAC subtypes were performed regarding prognosis, gene mutations, pathways, tumor microenvironment (TME), and chemotherapy sensitivity. A prognostic model was constructed utilizing Cox and Lasso regression based on subtype-related genes.

Samples from the TCGA-PAAD cohort were classified into two subtypes. The CS1 subtype was identified as a high-risk, immunosilent subtype, while the CS2 subtype was characterized as a low-risk, immunoactive subtype. Compared to CS2 subtype, CS1 subtype exhibited shorter survival, higher frequency of genetic mutations, more aggressive tumor-promoting nature, lower TME immune score, and increased sensitivity to chemotherapy. The prognostic model related to PDAC subtypes displayed robust predictive efficiency; IL20RB gene emerged having superior predictive capability.

We successfully identified two distinct PDAC subtypes. The developed prognostic model exhibited strong predictive efficacy; and the upregulation of IL20RB was identified as a promising therapeutic target for PDAC.

Alectinib was approved by the US, Europe and China in 2024 as the first adjuvant targeted therapy for ALK+ NSCLC, lowering risk of disease recurrence or death by 76%. Alectinib addresses a critical gap in postoperative adjuvant therapy for ALK+ NSCLC. From Chinese healthcare-system perspective, this study evaluates the impact of introducing alectinib in adjuvant therapy for stage IB (tumor ≥ 4 cm) to IIIA (UICC/AJCC 7th edition) ALK+ NSCLC on prevention of recurrence and the associated direct medical costs, compared to platinum-based chemotherapy.

A Markov model was developed to estimate the number of locoregional and metastatic recurrences over a 10-year period by defining four health states: disease-free survival, locoregional recurrence, metastatic recurrence, and death. In the control group, all patients received platinum-based chemotherapy, while in the intervention group, 75% received alectinib and 25% received platinum-based chemotherapy. Clinical data were collected from open-label, randomized phase 3 trials ALINA and ALEX. Cost parameters were derived from local charges, expert consultation, and published literature.

Compared to control group, the intervention group would reduce recurrences by 11,300 cases over 10 years, including 3,684 locoregional and 7,616 metastatic cases. This corresponds to a 45.82% lower recurrence rate. Estimated recurrence-related cost savings amounted to 6.910 billion RMB, with 1.445 billion RMB saved from locoregional recurrences and 5.465 billion RMB from metastatic recurrences. This represents a 41.49% reduction in costs compared to control group. These findings were robust across various scenario analyses.

Using alectinib in postoperative adjuvant therapy significantly reduces both the recurrence rate and recurrence-related treatment costs for stage IB (tumor ≥ 4 cm) to IIIA ALK+ NSCLC patients, compared to platinum-based chemotherapy. From perspective of Chinese healthcare system, this approach shows substantial potential for preventing recurrence and achieving cost savings.

Accurately identifying disease diagnoses from electronic health records (EHRs) is crucial for clinical/biomedical research; however, this is challenging when diagnoses are complex and require data from several sources, e.g., multiple myeloma (MM) and its precursor condition, MGUS. Leveraging the national Veterans Health Administration EHRs, we developed and validated a large language model (LLM)-based pipeline that utilizes only clinical notes from randomly selected patients identified via ICD codes for MGUS/MM. Among the evaluated LLMs and alternative approaches, Llama-3-8B-based pipeline with prompt engineering achieved the best performance. This pipeline not only saved the preprocessing steps and shortened the overall processing time but also outperformed rule-based or machine learning-based methods for identifying MGUS and achieved comparable performance for MM, solely relying on clinical notes. Our work demonstrates that the developed LLM-based pipeline can efficiently and effectively identify MGUS/MM diagnoses to replace manual chart abstraction and rule- or machine learning-based natural language processing methods.

The incidence rate of early-onset colorectal cancer (EoCRC, age < 45) has increased every year, but this population is younger than the recommended age established by national guidelines for cancer screening. In this paper, we applied 10 different machine learning models to predict EoCRC, and compared their performance with advanced large language models (LLM), using patient conditions, lab results, and observations within 6 months of patient journey prior to the CRC diagnoses. We retrospectively identified 1,953 CRC patients from multiple health systems across the United States. The results demonstrated that the fine-tuned LLM achieved an average of 73% sensitivity and 91% specificity.

Breast cancer treatment involves surgery, radiation, chemotherapy, and endocrine therapy, with recurrence risk depending on treatment execution. We propose a weighted Cox mixtures model that integrates treatment plans and clinical data to estimate recurrence risk. Data from Mayo Clinic (US) and the National Institute of Oncology (Morocco) inform the model. We enhance expectation maximization within the Cox mixtures model using three weighting strategies: Inverse Probability of Treatment Weighting, Adaptive Weights with focal loss, and Prioritizing Subgroups. In the Mayo Clinic cohort, Adaptive Weights improve predictive accuracy (C-index: 0.67-0.88), outperforming the standard Cox model. In the Moroccan cohort, Adaptive Weights also enhance C-index values (0.60-0.71), though with larger confidence intervals. Our findings demonstrate that weighting strategies refine recurrence risk prediction, particularly in imbalanced cohorts. Expanding datasets, especially in underrepresented populations, is crucial for improving model reliability and clinical applicability.

This study developed and evaluated methods to adjust misclassification errors in electronic health record (EHR)-derived covariates using group-wise and individualized weights based on observed sensitivity and specificity to reduce bias in predictive modeling. Logistic regression, XGBoost, and neural networks predicted follow-up adherence in lung cancer screening. The Lung-RADS category, extracted via natural language processing (NLP), was adjusted using group-wise weights and individualized weights from kernel and multinomial regression. Models with adjusted covariates were compared to naïve (unadjusted) and oracle (true value) models. Performance assessed by the area under the receiver operating characteristic (AUROC) curve across 10%, 20%, and 30% validation sets, showed that adjusted models outperformed naïve models, improving AUROC by 0.3%-10.4%. Compared to oracle models, adjusted models reduced the AUROC gap to 2.0%-7.5%. Individualized weights provided more precise corrections than group-wise weights. This scalable framework mitigates misclassification bias in EHR-derived covariates, enhancing predictive accuracy without resource-intensive manual review.

Clinical variability in prostate radiation therapy (RT) planning is well documented, but little is known about how radiation oncologists experience and adapt to the factors that drive it. This study explores variability as a human-centered design challenge, with the goalofinformingclinicaldecision support (CDS) design through real-timeinsight into planning decisions. We conducted observation sessions with the think aloud method followed by semi-structured interviews with five radiation oncologists while they contoured prostate cases. Using the Systems Engineering Initiative for Patient Safety (SEIPS) framework, we thematically analyzed the contributors to variability across tasks, technology, and organizational conditions. Results suggest that variability arises not only from anatomical or guidelineambiguity, butalso fromindividual interpretations of inputs, variation in contouring decisions, andadaptive strategies such as reliance on prior experience and estimation under uncertainty. Findings support the design of context-sensitive CDS tools that reflect real-world clinical reasoning while preserving clinical flexibility.

Accurate extraction of thyroid nodule features from radiology and pathology reports is clinically essential for guiding patient management decisions, such as surgical intervention or active surveillance. However, manual data extraction from electronic health records is labor-intensive and prone to inter-rater variability. To address this challenge, we evaluated open-source large language models (LLMs) for automating the extraction and matching of these critical nodule features. Using a retrospective dataset of 451 ultrasound and pathology report pairs, we developed an annotation schema capturing nodule characteristics. Two LLMs-Llama-3.3 70B and QwQ-32B-were benchmarked against manual annotations. Both models demonstrated near-perfect extraction accuracy for clinically relevant features such as location, size, and biopsy results. Notably, QwQ-32B achieved an F1 score of 0.987 on the complex multi-step reasoning task of matching nodules across reports. Our findings suggest integrating LLMs into clinical annotation workflows can significantly reduce clinician workload and inter-rater variability while maintaining high accuracy.