Citations

Abstract Deep learning has advanced digital pathology and multi-omics data analysis. However, there is still a critical need to integrate these modalities to better understand complex diseases. Deep learning models, such as Prov-Gigapath, a state-of-the-art pathology foundation model, have been developed for analyzing whole-slide images (WSI). Our previously published deep-learning framework, OmicsFootPrint (OFP), transforms high-dimensional multi-omics data into intuitive genomically-arranged circular images and utilizes transfer learning to overcome the challenges of high-dimensional, low-sample-size data. In this study, we aimed to combine WSI and multi-omics data to develop a framework capable of distinguishing between invasive ductal carcinoma (IDC) and invasive lobular carcinoma (ILC), which is clinically significant as it may help explain differential responses to therapies, such as tamoxifen versus letrozole. We enhanced the performance of our OFP model by dividing each multi-omics circular image (1024x1024x3 pixels) into 16 equal-sized patches (256x256x3 pixels). Each patch was trained using a separate EfficientNetV2 architecture with pre-trained ImageNet weights. The probability outcomes from these 16 models were subsequently combined through a meta-learner model using the AutoGluon framework. Concurrently, WSI images corresponding to the same patients were independently trained, maintaining consistent train/validation/test splits. The Prov-Gigapath model, which was trained using a dataset from the Providence Health System, was employed to extract patch-level embeddings from the WSI data. These embeddings were fine-tuned using a LongNet-based masked autoencoder to predict outcomes. Finally, predictions from the OFP model and the WSI-based Prov-Gigapath model were integrated through a meta-learner, preserving the original train/validation/test split. Model performances (AUC) were evaluated using the held-out test set. The gold standard used for the analysis was histological identification of IDC versus ILC, against which the performance of omics data alone, WSI data alone, and the multi-modal integration approach were compared. We evaluated our integrated model as a proof-of-concept using The Cancer Genome Atlas (TCGA) breast cancer cohort, comprising 92 ILC and 391 IDC samples. Multi-omics data, including gene expression, microRNA, copy number variation, and RPPA, were analyzed using the OFP framework, while WSI data were processed using the Prov-Gigapath method. The results demonstrated that the multi-modal integration approach significantly outperformed individual modalities, achieving an AUC of 0.98 compared to 0.87 for OFP (original), 0.95 for OFP (16-patches), and 0.92 for WSI. Integrating WSI and multi-omics data improves predictive accuracy, significantly improving the classification of breast cancer subtypes over histology alone. Citation Format: Naresh Prodduturi, Xiaojia Tang, Hamid R. Tizhoosh, Kevin J. Thompson, Richard Weinshilboum, Karthik V. Giridhar, Judy C. Boughey, Eric W. Klee, Liewei Wang, Matthew P. Goetz, Vera Suman, Krishna R. Kalari. A multi-modal learning framework to integrate digital pathology image and multi-omics data in breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2429.

Pancreatic ductal adenocarcinoma (PDAC) is one of the most aggressive cancers, primarily due to its complex tumor microenvironment (TME), which drives both disease progression and therapy resistance. Understanding the molecular mechanisms governing TME dynamics is essential for developing new treatment strategies for this devastating disease. In this study, we uncover an oncogenic role for Galectin-1 (Gal1), a glycan-binding protein abundantly expressed by activated pancreatic stellate cells (PSCs), a key component of the PDAC TME that orchestrates tumor progression. Our findings reveal that Gal1 expression is elevated in the nucleus of human PSCs in both tissue samples and cultured cell lines. Using chromatin immunoprecipitation followed by sequencing analysis (ChIP-seq), we identify Gal1 occupancy at the promoters of several cancer-associated genes, including KRAS , a pivotal oncogene involved in PDAC pathogenesis. We demonstrate that Gal1 binds to the KRAS promoter, sustaining KRAS expression in PSCs, which, in turn, maintains PSC activation and promotes the secretion of protumorigenic cytokines. Mechanistically, Gal1 is required to preserve histone H3 lysine 4 monomethylation levels and to recruit the histone methyltransferase MLL1 to target promoters. Collectively, our findings define a nuclear function of Gal1 in modulating the transcriptional landscape of cancer-associated genes in PSCs within the PDAC TME, mediated through an epigenetic mechanism. These insights enhance our understanding of PDAC pathology and open potential avenues for therapeutic interventions targeting intracellular Gal1.

Abstract Background Patients’ self-report of their symptoms can provide important data for the evaluation of treatment benefit and tolerability of oncology drugs. Contemporary treatment approaches, including immunotherapy and molecular targeted therapies, have unique toxicities based on their novel mechanisms of action. This scoping review aimed to summarize evidence from existing reviews and clinical practice guidelines to examine the type and prevalence of toxicities including symptomatic adverse events (sympAEs) for adult cancer patients to inform clinical care and therapeutic trials. Methods A systematic search of PubMed, Web of Science, and Embase was performed using predefined eligibility criteria. Thirty-one literature reviews and 3 clinical practice guidelines met inclusion criteria and were selected for review and data abstraction. Results Findings from this scoping review demonstrated several leading sympAEs that were reported across immunotherapy and targeted therapy drugs, including fatigue, diarrhea, and rash. In addition to these more prevalent sympAEs, there were some less frequently reported class-specific sympAEs, which had potential for significant harm or disability to the patient if not properly identified and treated. Many studies reported toxicities as AEs or syndromes solely using data reported by clinicians without additional self-report from patients. Conclusion We identified several core sympAEs experienced by patients participating in oncology trials using immunotherapy and targeted therapy agents, which has implications for future trial design and drug labeling. Future cancer trials should assess patient-reported sympAEs based on the identified drug mechanism to inform the tolerability of these newer agents and enhance patient safety during trial participation and clinical care.

Abstract Pancreatic neuroendocrine tumors (PanNETs) have an age-adjusted incidence of 1.5 per 100,000 people, with a notable rise in the incidence of small (≤2 cm) non-functional PanNETs (NF-PanNETs) in recent decades. While surgery is traditionally the preferred treatment for localized NF-PanNETs, active surveillance is now an accepted management strategy for tumors smaller than 2 cm due to their relatively benign behavior. However, this approach has not yet been fully integrated into routine clinical practice. There is considerable histopathological heterogeneity observed in NF-PanNETs, which results in significant variability in clinical presentation, behavior and treatment outcomes. Hence, tumor size alone does not provide sufficient certainty regarding a benign clinical course for decision-making. Although studies advocate for incorporating WHO grade into clinical prognostic assessments, this marker also has limitations. Several established tissue-based markers, such as ATRX and DAXX alterations, alternative lengthening of telomeres, and copy number variations, can be used for PanNET subtyping and correlate with metastatic risk. Combining these markers with traditional histopathological parameters may yield a more comprehensive and accurate prognostic assessment. This review discusses the advantages and limitations of current prognostication methods for small NF-PanNETs and highlights recently established prognostic markers, along with the requirements for their implementation into routine clinical practice. It also proposes practical solutions to address the challenges associated with the immediate integration of these biomarkers into routine care.

Diffuse large B-cell lymphoma (DLBCL) is the most prevalent and aggressive subtype of non-Hodgkin lymphoma comprising 30% to 40% of cases. While DLBCL frequently arises in the lymph nodes, up to 40% of cases originate in extranodal tissues. Primary splenic DLBCL is extremely rare comprising only ~1% of DLBCL cases and can be complicated by rare entities including gastrosplenic fistula (GSF). In contrast to DLBCL, polycythemia vera (PV) is a myeloproliferative malignancy. Polycythemia vera can transform into other hematologic malignancies including post-polycythemia myelofibrosis, but associations with DLBCL are uncommon. We present the first case of PV with concomitant primary splenic DLBCL complicated by GSF. While the majority of splenic lesions are benign, they have a broad differential diagnosis including malignant etiologies. Cystic or solid morphology, the number of lesions, and vascularity on imaging heavily guide further management. Due to concern for imminent massive upper gastrointestinal bleeding in the setting of GSF, our patient was diagnosed and managed with prompt splenectomy and gastrectomy. Percutaneous splenic biopsy is also a safe and effective diagnostic modality, but was deferred in our case given increased bleeding risk with PV. In conclusion, primary splenic lymphoma should be in the differential for a splenic mass regardless of whether the patient has a prior hematologic malignancy, and management should be prompt especially if complicated by a GSF.

Background: Young female patients undergoing hip arthroscopy have an increased prevalence of baseline capsular laxity of the hip joint. This laxity, along with superimposed postoperative iatrogenic capsular deficiency secondary to an unrepaired capsule, could potentially lead to worse outcomes after arthroscopic treatment of femoroacetabular impingement (FAI) in this population. Purpose: To compare outcomes and revision rates for young female patients undergoing hip arthroscopy for FAI and labral tear with capsular closure (CC group) versus capsular nonclosure (CNC group). Study Design: Cohort study; Level of evidence, 3. Methods: A retrospective review of prospectively collected data of patients who underwent index hip arthroscopy using interportal capsulotomies by a single surgeon between January 2014 and February 2020 was performed. Female patients aged 12 to 21 years who underwent hip arthroscopy with cam or pincer osteoplasty and labral repair with a 2-year minimum follow-up were included. The Beighton score was assessed. The Hip Outcome Score–Activities of Daily Living (HOS-ADL) and Hip Outcome Score–Sport-Specific Subscale (HOS-SSS) were obtained preoperatively and at 3 months, 1 year, and 2 years postoperatively. Data were analyzed using the Mann-Whitney U test and Fisher exact test. Results: A total of 23 hips (20 patients) were included in the CC group and 17 hips (16 patients) were included in the CNC group. The groups were not different regarding characteristics and preoperative patient-reported outcome scores. At all follow-up intervals postoperatively, the CC group scored significantly higher HOS-ADL and HOS-SSS. Fifteen of 17 (88.2%) patients in the CC group returned to sports versus 8 of 14 (57.1%) in the CNC group ( P = .03). Four (17.4%) hips in the CC group had revision surgery compared with 9 (52.9%) hips in the CNC group (odds ratio, 5.1; 95% CI, 1.2-22.5; P = .02). Conclusion: Young female patients treated with CC while undergoing arthroscopic FAI had improved outcomes, fewer revisions, and a higher return-to-sports rate than those treated without CC.

Background: Latinas with mammographic dense breasts are at increased risk of breast cancer. This randomized controlled trial tests the effectiveness of three breast density (BD) educational approaches on adherence to subsequent mammography screening among Latinas receiving care at a Federally Qualified Health Center (FQHC). Measure(s): Adherence was ascertained using electronic health record and survey responses. Kaplan–Meier estimates of the time to subsequent mammogram were used to obtain adherence rates at 1- and 2-years post baseline mammogram. Cox modeling assessed whether adherence differed by patient characteristics or group assignment. Results: This analysis was limited to Latinas enrolled between October 27, 2016, and December 21, 2018 ( n = 946; 66.1% <50 years of age, 53.5% with dense breasts). Adherence rates at 1 year was 24.8% increasing to 51.7% by year 2. Latinas randomized to the Promotora + brochure + letter arm (hazard ratio [HR]: 1.09 [95% confidence interval [95% CI] 0.87 to 1.36]) or brochure + letter (HR: 1.03 [95% CI 0.82 to 1.29]) arm were not more likely to be adherent to subsequent mammography compared to the letter only arm ( p = 0.76) . Adjusting for age and study group, having more prior mammograms, being “extremely likely” to get an annual mammogram, and having more confidence to get an annual mammogram at baseline were significant drivers of subsequent adherence. Conclusion: Informational interventions targeting BD education alone are unlikely to significantly improve adherence to subsequent mammography among Latinas receiving care in FQHCs.

Abstract The aging process is universal, and it is characterized by a progressive deterioration and decrease in physiological function leading to decline on the organismal level. Nevertheless, a number of genetic and nongenetic interventions have been described, which successfully extend healthspan and lifespan in different species. Furthermore, a number of clinical trials have been evaluating the feasibility of different interventions to promote human health. The goal of the annual Biological Sciences Section of the Gerontological Society of America meeting was to share current knowledge of different topics in aging research and provide a vision of the future of aging research. The meeting gathered international experts in diverse areas of aging research including basic biology, demography, and clinical and translational studies. Specific topics included metabolism, inflammaging, epigenetic clocks, frailty, senescence, neuroscience, stem cells, reproductive aging, inter-organelle crosstalk, comparative transcriptomics of longevity, circadian clock, metabolomics, and biodemography.