Citations

Abstract Motivation Compositional data are frequently encountered in many disciplines, such as in next-generation sequencing experiments widely used in biomedical studies. Regression analysis with compositional data as either responses or predictors has been well studied. However, when both responses and predictors are compositional, the inventory of analysis tools is surprisingly limited, especially in the high-dimensional setting. Among the few existing methods, most of them rely on a log-ratio transformation to move compositional data from the simplex to real numbers. Yet, a serious weakness of these methods is their failure to handle the substantial fraction of zeroes observed in data collected from next-generation sequencing experiments. Results To investigate associations between two high-dimensional multi-omics compositions, we propose a composition-on-composition (COC) regression analysis method which does not require log-ratio transformations and hence can handle zeroes in the data. To account for high dimensionality, we estimate regression coefficients using a penalized estimation equation approach. Finally, inference procedures for COC regression are also proposed. Superior performance of COC is demonstrated through both comprehensive numerical simulations and case studies. Availability and implementation Source R codes to implement COC method is available at https://github.com/nrios4/COC.

Abstract Anaplastic large cell lymphoma (ALCL) cases were discussed at the 2023 Society of Hematopathology/European Association for Haematopathology workshop held on November 9-11, 2023, in Houston, Texas. This session focused on the clinical and pathologic spectrum of 3 types of ALCL: anaplastic lymphoma kinase (ALK)-positive, ALK-negative, and breast implant associated (BIA). Cases of primary cutaneous ALCL were excluded because they were included in another session. Although the diagnosis of ALCL is often straightforward, the 42 submitted cases encompassed unusual clinical presentations, morphologic variants, and atypical immunophenotypes, exemplifying potential difficulties and challenges in establishing the diagnosis of ALCL. Distinguishing ALCL from CD30-positive peripheral T-cell lymphoma, not otherwise specified, or classic Hodgkin lymphoma was discussed. In patients with a previous history of mycosis fungoides and other T-cell lymphoma/leukemia, the differential diagnosis of a CD30-positive T-cell lymphoma mimicking ALCL (mycosis fungoides transformation vs de novo ALCL) was also discussed. In patients with suspected BIA-ALCL, it is critical to properly handle the periprosthetic fluid when the disease first presents and the capsule at the time of initial capsulectomy to make a correct diagnosis and pathologic staging because a missed diagnosis may lead to disease progression. Comprehensive immunohistochemical analysis; fluorescence in situ hybridization for DUSP22, TP63, or JAK2 rearrangement; assessment of clonality of the T-cell receptor and immunoglobulin genes; and sequencing for mutations were performed as part of the workup of the submitted cases, particularly on ALK-negative ALCL cases, emphasizing the importance of ancillary testing in establishing the diagnosis.

ABSTRACT Purpose Optical coherence tomography (OCT) is a widely used imaging modality in ophthalmology. Accurate semantic segmentation of these images is critical for both clinical and research applications, yet existing convolutional neural network (CNN)-based methods face challenges in generalizability and robustness. This study introduces MAYOCTransformer, the first transformer-based deep learning model for comprehensive semantic segmentation of OCT images, and evaluates its performance against CNN-based models. Methods A large dataset of 3,500 OCT images was manually segmented using an iterative deep learning-assisted workflow. The MAYOCTransformer model, based on the Mask2Former architecture, was trained and compared against CNN-based segmentation models, including U-Net, U-Net++, FPN, and DeepLabV3+. Comprehensive segmentation tasks included 10 retinal layer segmentation, choroid stroma and vessel segmentation, and the identification of 9 types of discrete pathological findings including intraretinal fluid (IRF), subretinal fluid (SRF), pigment epithelial detachment (PED), subretinal hyper-reflective material (SHRM), intraretinal hyper-reflective foci, and reticular pseudodrusen. Model performance was evaluated using the Dice similarity coefficient (DSC) on a hold-out test set with five-fold cross-validation. Additional validation was performed using external datasets, open-source segmentation models, and a randomized blinded expert evaluation. Results MAYOCTransformer outperformed CNN-based models in most segmentation tasks. Choroid segmentation performance was comparable between MAYOCTransformer and CNN models. External validation demonstrated the model’s generalizability, achieving higher DSC scores than publicly available segmentation models. A blinded expert evaluation showed that MAYOCTransformer’s segmentation was non-inferior to manual annotations. Conclusion MAYOCTransformer provides improved segmentation performance over CNN-based models. Its ability to generalize to external datasets suggests potential applicability in clinical and research settings.

PURPOSE Stereotactic radiation (SRS/SRT) as opposed to whole-brain radiation (WBRT) represents the standard of care for patients with a limited number of brain metastases given the relatively favorable toxicity profile associated with stereotactic treatment. However, in patients with small cell lung cancer (SCLC), WBRT remains standard because of a lack of prospective data supporting SRS/SRT and concerns related to intracranial progression and neurologic death when WBRT is omitted. We conducted a single-arm, multicenter, phase II trial of SRS/SRT in patients with SCLC and 1-10 brain metastases to assess neurologic death rates relative to historical controls managed with WBRT (ClinicalTrials.gov identifier: NCT03391362 ). METHODS Patients were eligible if they had SCLC or an extrathoracic small cell primary and 1-10 brain metastases. Previous brain-directed radiation including prophylactic cranial irradiation was not permitted. Neurologic death was defined as marked, progressive, radiographic brain progression accompanied by corresponding neurologic symptomatology without systemic disease progression or systemic symptoms of a life-threatening nature. Close imaging-based surveillance of the brain post-SRS/SRT was used. RESULTS Between February 2018 and April 2023, 100 patients were enrolled. The median number of brain metastases was 2 (IQR, 1-4; range, 1-10). The median overall survival was 10.2 months; only 22% of patients required salvage WBRT. In total, 20 neurologic deaths were observed, relative to 64 non-neurologic deaths. The neurologic death rate at 1 year was 11.0% (95% CI, 5.8 to 18.1); the historical rate in patients managed with WBRT was 17.5%. CONCLUSION Our prospective, multi-institutional study demonstrated low rates of neurologic death when SRS/SRT as opposed to WBRT is used in patients with SCLC and 1-10 brain metastases who are surveilled closely post-treatment, supporting the utility of stereotactic approaches in this population.

Abstract RAS pathway (RASp) mutations induce proliferative features, and promote transformation in chronic myelomonocytic leukemia (CMML). However, the unique clonal landscape and hierarchy of distinct RASp mutations remain unexplored. To characterize the landscape, architecture, and implications of unique RASp mutations in CMML, we evaluated a cohort of 814 patients with CMML. We identified 461 RASp mutations among 342 patients (42%). N/KRAS and CBL mutations were the most common, frequently involved the P-loop or RING domains, respectively, and frequently appeared as dominant events (63% and 65%, respectively). BRAF, NF1, and PTPN11 mutations spanned throughout the gene structure, and frequently appeared as subclonal events (75%, 64%, and 59%, respectively). CBL mutations frequently occurred in codominance with SRSF2 and multihit TET2, and were enriched for KIT mutations. PTPN11 mutations more frequently co-occurred with SETBP1 and DNMT3A mutations, and were infrequently codominant with TET2 or ASXL1. RASp mutations predicted for shorter overall survival (hazard ratio [HR], 1.55; 95% confidence interval [CI], 1.15-2.07; P = .0075) and leukemia-free survival (LFS; HR, 1.67; 95% CI, 1.26-2.20; P = .0011), influenced outcomes of myelodysplastic and TET2 mutant CMML, and cooperated with IDH2 and RUNX1 mutations to induce shorter LFS. These data set the bases for refined genomic classifications of CMML, and underscore the need to develop RAS-directed therapies for patients with CMML.

Abstract Despite the growing interest in network-guided neuromodulation for epilepsy, uncertainty about the safety and long-term efficacy of thalamocortical stimulation persists. Our evaluation focused on the use of a four-lead open-loop implantable pulse generator for thalamocortical network neuromodulation. We retrospectively reviewed seven patients with diverse seizure networks—poorly localized, regional, or multifocal—undergoing thalamocortical neuromodulation. Employing a four-lead system, electrodes targeted both thalamic and cortical seizure network nodes. We assessed seizure severity, life satisfaction and sleep quality on a 10-point scale, and seizure frequency was assessed via telephone interviews and chart review. Outcomes were assessed by the Wilcoxon sign-rank test at the 0.05 significance level. Seven patients with a median age at implant of 22 years (range 14–42 years) had a median disabling seizure reduction of 93% (range 50–100%, P = 0.0156), with 100% responder rate (≥50% reduction in seizure frequency) after a median of 17 months post-implantation (range 13–60). The median improvement in seizure severity was 3.5 out of 10 points (P = 0.0312), life satisfaction 4.5 points (P = 0.0312) and quality of sleep 3 points (P = 0.062). No perioperative complications occurred. Transient seizure exacerbations (n = 2) and stimulation-related sensory/motor side-effects (n = 2) quickly resolved with parameter adjustments. One patient required surgical revision due to delayed infection. Six patients had permanent electrode placement refined by intracranial EEG trial stimulation; five patients had >90% reduction in seizure frequency during trial stimulation. Thalamocortical network neuromodulation using a four-lead open-loop system is safe and associated with significant improvements in seizure control and patient quality of life. Trial stimulation during intracranial EEG shows promise for enhancing seizure network engagement and parameter optimization but requires further study. Prospective controlled trials are needed to further characterize and validate the efficacy and side-effect profile of thalamocortical network neuromodulation for epilepsy.

Abstract Purpose: Intratumoral injection of Clostridium novyi-nontoxic (C. novyi-NT), lacking alpha-toxin, germinates and subsequently replicates in the tumor hypoxic regions, causing cell lysis and inflammation. This phase Ib study investigated the safety and synergistic effects of pembrolizumab and C. novyi-NT in advanced solid tumors. Patients and Methods: Patients with injectable solid tumors received one intratumoral C. novyi-NT injection (day 8) across four dose cohorts (3 × 104 to 100 × 104 spores) with pembrolizumab (200 mg intravenously every 3 weeks, day 0–24 months). Sixteen patients were enrolled with a median age of 62.5 years, and the most common tumor types were breast cancer and melanoma. Results: The combination therapy demonstrated a manageable safety profile, with one grade 3 dose-limiting toxicity of abscess formation and the remainder of treatment-related adverse events in ≥10% of patients being grade 1 or 2 injection site reaction (25%), pyrexia (19%), pruritus (13%), leukopenia (13%), and anemia (13%). The maximum tolerated dose was determined to be pembrolizumab intravenously at 200 mg every 3 weeks and a single dose of 100 × 104C.novyi-NT spores intratumorally on day 8. The confirmed overall objective response rate of 25% was observed among four patients with nonkeratinizing undifferentiated nasopharyngeal squamous carcinoma, human papilloma virus–positive squamous cell carcinoma of the base of the tongue, vulvar melanoma, and chordoma. Among these patients, we obtained three partial responses and one complete response. The median duration of the response was 10.93 months, and stable disease was observed in 69% of patients. Conclusions: This study highlights the potential of the C. novyi-NT and pembrolizumab combination as a promising therapeutic strategy for treatment-refractory solid tumors.