About

Irene M. Ong is an Assistant Professor at the University of Wisconsin in the Departments of Obstetrics and Gynecology with a joint appointment in Biostatistics and Medical Informatics. She also serves as the Associate Director of the University of Wisconsin Carbone Cancer Center's Cancer Informatics Shared Resource. Her research interests encompass data mining, artificial intelligence, machine learning, probabilistic methods, dynamical models, inductive logic programming, and statistical relational learning, with a focus on applications to biological and medical data. Irene is particularly interested in the integration and analysis of diverse molecular data types including clinical, genomics, transcriptomics, proteomics, immunome, metagenomics, and metabolomics data, especially as these relate to precision medicine. Irene earned dual bachelor's degrees in Computer Sciences and Biological Sciences from Cornell University and completed her PhD in Computer Sciences under the mentorship of David Page at the University of Wisconsin-Madison.

Research topics

• Biology
• Genetics
• Medicine
• Pathology
• Immunology
• Cancer research
• Internal medicine
• Environmental health
• Endocrinology
• Virology
• Cell biology
• Oncology

Selected publications

• A Genome‐Wide Association Study of Colorectal Cancer Mortality Outcomes Among Individuals of African and Admixture Ancestry

  Molecular Carcinogenesis · 2026-01-29

  articleOpen access

  ABSTRACT African Americans have the highest colorectal cancer (CRC) mortality rates in the United States. We performed the first genome‐wide association study (GWAS) of overall and CRC‐specific mortality among African Americans with incident CRC to identify genetic contributors to CRC outcomes. Participants enrolled in the Southern Community Cohort Study in 2002–2009; incident CRC and mortality were identified via state cancer registries and the National Death Index. SNPs were genotyped across the genome via Illumina platforms and imputed using the Michigan Imputation Server with Minimac4. Associations with mortality were estimated as hazard ratios (HRs) with 95% confidence intervals (CIs) using Cox proportional hazards models, adjusted for age, sex, stage, and five principal components for ancestry. In total, 500 Black‐identifying participants were analyzed, including 316 deaths and 184 CRC‐specific deaths. Two novel loci in linkage disequilibrium ( r 2 = 1) within LTBP1 were associated with higher CRC‐specific mortality risk: rs34071846 and rs12712337 (per allele HR: 2.74, CI: 1.91–3.92, p = 3.78 × 10 −8 ). An additional variant mapped to a gene for a noncoding RNA was associated with CRC‐specific mortality: rs10103953 (per allele HR: 0.52, CI: 0.42–0.66, p = 2.03 × 10 −8 ). One loci mapping to MCTP2 was associated with lower overall mortality risk: rs7171579 (per allele HR: 0.59, CI: 0.50–0.71, p = 2.13 × 10 −8 ). In conclusion, evidence from the present study supports LTBP1 and MCTP2 as important to CRC mortality.

  PublisherOA PDFDOI

• Abstract A023: Differential Patterns of Immune Infiltration in the Tumor Immune Microenvironment Associate with Therapeutic Response in Primary Prostate Cancer Following Chemohormonal Therapy

  Cancer Research · 2026-01-20

  article

  Abstract Background: There is a critical need to develop novel therapeutic strategies and diagnostic tools to precisely deliver treatments to improve survival for men with prostate cancer (PCa). To support this development, improved strategies are needed to better understand heterogenous tumor microenvironments and tumor biology that associate with variable treatment responses. We hypothesized that the tumor immune microenvironment (TIME) plays a critical role in treatment resistance. In this study we aimed to evaluate TIME signatures of treatment response and resistance utilizing a novel, integrated technological tool to identify response patterns and enable precision sampling for comparative cellular and molecular analysis. Methods: 30 patients with newly-diagnosed, locally advanced, high-risk, primary PCa underwent 18F-DCFPyL PSMA PET with multiparametric MRI (mpMRI) imaging on a dedicated PET/MRI scanner followed by 3 cycles of chemohormonal therapy (NCT03358563). Repeat PSMA PET/MRI was performed prior to prostatectomy and scans were interpreted by an experienced radiologist and nuclear medicine physician as complete response (CR), partial response (PR), no response (NR) or normal tissue. MRI scans were used to print a 3D mold of the prostate to allow PET and MRI directed mapping and microdissection of regions of interest from the resected prostate with slice-by-slice and lesion-to-lesion correlation. Cellular infiltrates were analyzed by flow cytometry in 3 to 5 tissue specimens per patient. Results: The frequency of CD8+ T cells in the total CD45+ infiltrate was highest in normal and CR areas and was significantly reduced in PR vs CR (p<0.01). CXCR3+CD8+ and CD103+CD8+ Tcell frequencies were also reduced in PR vs CR foci (p<0.01, p<0.05, respectively). Meanwhile, the frequency of CXCR3+CD8+ T and CXCR3+++CD8+ T cells was highest and significantly elevated in CR vs normal tissue suggesting enrichment of activated, homing, Tc1 CD8T cells. An increase in total CD8+ and CD103+CD8+ T cells was associated with longer progression-free survival. DESeq2 analysis of bulk mRNA sequencing showed enrichment of CD8a and ITGAE (CD103) gene expression differential in CR vs PR lesions (p=0.028, p=0.00029, respectively). CD8a expression was reduced in panCK- AOI of resistant (PR) foci by GeoMx WTA spatial transcriptomic analysis in two model patients with multi-focal tumors. Flow analysis of EpCAM+ cells had a significant increase in B7H3 expression in PR vs CR lesions (p<0.05). We are currently integrating analysis of myeloid cells and expand spatial transcriptomic analysis of matched multi-focal tumors to further dissect patterns of therapeutic response in our study cohort. Conclusions: In conclusion, PSMA /PET and mpMRI based precision sampling of tumor tissue associated with differential therapeutic response patterns captured differences in the TIME infiltrates and these observations may provide hypothesis to test biological mechanisms to expedite discovery of targetable mechanisms to improve tumor stratification and targeting in high-risk prostate cancer. Citation Format: Erika Heninger, Jamie M. Sperger, Kristin Weinstein, Brian P. Johnson, Peter G. Geiger, Shane A. Wells, Steve Y. Cho, Wei Huang, Philippos Tsourkas, Sean McIlwain, Irene M. Ong, David Quigley, David F. Jarrard, Sheena C. Kerr, David J. Beebe, Joshua M. Lang. Differential Patterns of Immune Infiltration in the Tumor Immune Microenvironment Associate with Therapeutic Response in Primary Prostate Cancer Following Chemohormonal Therapy [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Innovations in Prostate Cancer Research and Treatment; 2026 Jan 20-22; Philadelphia PA. Philadelphia (PA): AACR; Cancer Res 2026;86(2_Suppl):Abstract nr A023.

  PublisherDOI

• A Head and Neck Cancer Patient-Specific Microphysiological System for Predicting Response to Chemoradiation

  bioRxiv (Cold Spring Harbor Laboratory) · 2026-04-30

  articleOpen access

  Head and neck cancer (HNC) is the 6th most common malignancy worldwide. 60% of patients present with advanced disease and approximately 50% of patients recur following primary treatment. Chemoradiation remains a standard of care for most patients. However, clinicians lack functional tools to predict which patients will respond to chemoradiation prior to treatment and current models, including organoids and animal model systems, fail to capture either full complexity or patient-to-patient heterogeneity of the individual HNC tumor and microenvironment (TME). Here, we have developed, characterized, and tested a patient-specific microphysiological system (MPS) that reconstructs the HNC TME in a vascularized 3D environment. This MPS was constructed from malignant cells, fibroblasts, and immune cells from a patient's surgically resected tumor, seeded within a 3D hydrogel with molded endothelial lumens. Single-cell RNA sequencing confirmed that the MPS preserved 12 transcriptionally distinct cell populations found in matched native tissue. The platform recapitulated tumor hypoxia, with a 12-fold increase in hypoxic marker expression that altered radiation response, consistent with clinical HNC biology. Compartment-resolved imaging revealed distinct treatment dynamics in tumor, stromal, and vascular regions, and individual patients exhibited divergent responses to chemoradiation in spheroid morphology, cell viability, and migration. We found the slope of spheroid area change with treatment tracked with tumor recurrence, suggesting this metric could serve as a functional predictor of therapeutic response.

  PublisherOA PDFDOI

• Priming versus propagating: distinct immune effects of alpha- versus beta-particle emitting radiopharmaceuticals when combined with immune checkpoint inhibition in mice

  Nature Communications · 2026-01-26 · 1 citations

  articleOpen access

  Radiopharmaceutical therapy (RPT) synergises with immune checkpoint inhibitors (ICI), but comparison of immunomodulation by different radioisotopes is lacking. Here, we evaluate mechanisms of response and timing of ICI administration relative to α- (225Ac) and β-emitting (90Y, 177Lu) radioisotope therapy, coupled with alkylphosphocholine NM600, when combined with dual (anti-PD-L1 and anti-CTLA4) ICI, using syngeneic poorly immunogenic (B78 and Myc-CaP) and immunogenic (MC38) murine models. Regardless of the isotope, RPT delivering 2 Gy mean tumor dose promotes tumor regression and improves survival in B78 or MC38 tumor-bearing mice when combined with early ICI administration. Greatest anti-tumor responses are seen in MC38 to 90Y-NM600 + ICI and in B78 and Myc-CaP to 225Ac-NM600 + ICI. Flow cytometry and single-cell RNA and T cell receptor sequencing reveal that, combined with ICI, β-emitting radioisotopes expand existing adaptive immunity, whereas α-emitting radiopharmaceuticals initiate immune priming. Thus, appropriate application of α- or β-emitting RPT in combination with ICI achieves distinct antitumor immune responses. Preclinical studies indicate a synergistic effect of radiotherapy treatment (RT) and Immune checkpoint inhibitors (ICI) on tumor growth and metastasis. However, little is known about the immunomodulatory performance of different radioisotopes on the tumor microenvironment. Here, the authors employ alpha- versus beta-particle emitting radiopharmaceuticals in combination with dual ICI therapy and dissect mechanisms of in vivo immunomodulation and timing of ICI administration relative to RT, by comparing responses in immunogenic and non-immunogenic preclinical mouse models.

  PublisherDOI

• Low-dose radiation by radiopharmaceutical therapy enhances GD2 <i>TRAC</i> -CAR T cell efficacy in localized neuroblastoma

  Science Advances · 2025-06-06 · 5 citations

  articleOpen access

  Chimeric antigen receptor (CAR) T cells have limited efficacy against solid tumors including neuroblastoma. Here, we evaluated whether low-dose radiation delivered by radiopharmaceutical therapy (RPT), known to potentiate immune checkpoint inhibitors, can synergize with CRISPR-edited GD2 TRAC- CAR T cells to improve outcomes in neuroblastoma. We found that in the localized model of neuroblastoma, low-dose radiation delivered by 177 Lu-NM600, an alkylphosphocholine mimetic RPT agent, followed 9 days later by GD2 TRAC- CAR T cells led to complete tumor regression. Irradiation of neuroblastoma before GD2 TRAC- CAR T cells enhanced the release by CAR T cells of perforin, granzyme B, tumor necrosis factor–α, and interleukin-7 while abrogating transforming growth factor–β1. Low-dose RPT up-regulated the death receptor Fas on neuroblastoma, potentially enabling CAR-independent killing. This suggests that low-dose RPT can enhance suboptimal CAR T cell efficacy against solid tumors. However, optimization of radiation dose and timing may be needed for each patient and RPT agent to account for varied tumor radiosensitivity and dosimetry.

  PublisherDOI

• Uncovering Mechanisms of Lung Epithelial Injury in Acute Lung Transplant Rejection Using Organoids From Bronchoalveolar Lavage Fluid

  American Journal of Respiratory and Critical Care Medicine · 2025-05-01

  article

  Abstract RATIONALE: Lung transplantation is the last resort for people with end-stage lung disease, but high rates of rejection limit quality of life and survival post-transplant. Half of patients experience acute rejection in the first year after transplant, most commonly acute cellular rejection (ACR). ACR also increases the risk of chronic lung allograft dysfunction (CLAD), the leading cause of death after one year. ACR is thought to result from T cell and possibly NK cell activation, but how immune attack injures the lung is largely unknown due to a lack of experimental models. To uncover the features of epithelial injury in ACR, we established lung organoids from bronchoalveolar lavage (BAL) fluid from lung transplant patients. METHODS: We obtained fresh BAL fluid from clinical bronchoscopies at the University of Wisconsin-Madison and Mass General Brigham, under approved IRB protocols. We recently published methods for deriving airway and alveolar organoids from BAL. Since ACR may be related to CLAD, an airway-centered disease, we focused on airway organoids for our study. We characterized organoid phenotypes, cellular composition, and single-cell transcriptomics from patients with and without ACR. RESULTS: We generated airway organoids from 24 BAL samples from lung transplant recipients. Thirteen samples were from patients who had current or past rejection of grade A2, B1, or higher, as determined by transbronchial biopsy at the time of BAL, while 11 patients never had rejection. We excluded patients with indeterminate grade A1 rejection or active infection. We found that significantly more organoids form from the BAL of patients who experienced rejection, while patients without rejection resemble healthy non-transplanted controls. The organoid numbers did not correlate with the total epithelial or immune cell counts in BAL fluid, cell viability, patient age, or microbiology. To determine potential mechanisms for the discrepancy, we performed single-cell RNA-sequencing on airway organoids from 3 patients who never had rejection and 6 patients with current (n = 5) or past (n = 1) rejection. Preliminary analysis indicates that significant gene expression differences persist in the cultured airway epithelial cells. Immune, inflammatory, cell death, and stress response pathways are upregulated in ACR, while epithelial progenitor cell renewal, differentiation, and metabolism appear downregulated. CONCLUSIONS: We provide proof of principle that BAL-derived lung organoids can be used to model epithelial injury in acute lung transplant rejection. While further analysis and validation are underway, the results may ultimately advance a mechanistic understanding of ACR and identify more targeted therapies.

  PublisherDOI

• Loss of GATA2 promotes invasion and predicts cancer recurrence and survival in uterine serous carcinoma

  JCI Insight · 2025-04-01

  articleOpen access

  BACKGROUNDA priori knowledge of recurrence risk in patients with nonmetastatic (International Federation of Gynecology and Obstetrics [FIGO] stage I) uterine serous carcinoma (USC) would enable a risk-stratified approach to the use of adjuvant chemotherapy. This would greatly reduce treatment-related morbidity and be predicted to improve survival.METHODSGATA2 expression was scored by IHC across a retrospective multiinstitutional cohort of 195 primary USCs. Associations between GATA2 levels and clinicopathologic metrics were evaluated using Student's t test, Fisher's exact test, Kaplan-Meier method, and Cox proportional hazard ratio. Invasion in patient-derived USC cells was assessed by Student's t test. RNA-Seq, anti-GATA2 ChIP-Seq, and confirmatory Western blotting enabled identification of GATA2 targets.RESULTSPatients with FIGO stage I GATA2hi USCs had 100% recurrence-free and 100% cancer-related survival, which was significantly better than patients with GATA2lo USCs. In patients for whom adjuvant chemotherapy was omitted, patients with GATA2hi USC had 100% recurrence-free 5-year survival compared with 60% recurrence-free survival in patients with GATA2lo USC. Depletion of GATA2 in patient-derived USC cells increased invasion in vitro.CONCLUSIONRoutine GATA2 IHC identifies 33% of patients with FIGO stage I USC who have a greatly reduced risk of posthysterectomy USC recurrence. Our results suggest that a GATA2-guided personalized medicine approach could be rapidly implemented in most hospital settings, would reduce treatment-related morbidity, and would likely improve outcomes in patients with USC.FUNDINGNIH grants R01 DK068634, P30 CA014520, S10 OD023526, K08 DK127244, T32 HL007899, the UW-Madison Department of Pathology and Laboratory Medicine, the UW-Madison Centennial Scholars Program, the Diane Lindstrom Foundation, the American Cancer Society, the V Foundation, The Hartwell Foundation, and the UMN Department of Obstetrics, Gynecology, and Women's Health.

  PublisherOA PDFDOI

• Abstract 2443 Fluctuations in central metabolite availability regulate global and site-specific chromatin mechanisms

  Journal of Biological Chemistry · 2025-05-01

  articleOpen access

  Central metabolites function as essential cosubstrates for chromatin modifying enzymes, directly linking cellular metabolism to chromatin regulation. Accordingly, dynamic changes in the global availability of these metabolites have been observed to influence downstream phenotypes through chromatin-dependent mechanisms. We have previously shown how dietary or in vitro restriction of the essential amino acid methionine leads to decreased availability of its downstream product S-adenosylmethionine (SAM), the universal methyl-donor of the cell.

  PublisherOA PDFDOI

• Depletion of individual dietary amino acids induce distinct metabolic and chromatin states

  Journal of Biological Chemistry · 2025-12-17 · 2 citations

  articleOpen access

  Reducing dietary levels of protein or specific essential amino acids (EAAs) promote favorable metabolic reprogramming, including improved glucose tolerance, increased insulin sensitivity, and reduced fat mass. However, the extent to which shared or EAA-specific mechanisms facilitate diet-associated phenotypes remains unclear. Here, we compared the physiological and molecular responses to dietary levels of methionine, leucine, and isoleucine by feeding C57BL/6J mice diets in which each of these specific AAs is depleted. Dietary depletion of Met, Leu, or Ile (Met-D, Leu-D, or Ile-D) elicited distinct, AA-specific physiological and hepatic molecular (transcriptome, metabolome, histone proteome) responses that were not phenocopied by mTORC1 inhibition via rapamycin treatment. Ile-D yielded the most distinct and dramatic responses, highlighted by expression of select chromatin modifying and metabolic enzymes that led to a prominent epigenetic state of histone H2A/H4 hypoacetylation and maintained hepatic acetyl-CoA levels despite downregulated β-oxidation. Multi-omics factor analysis of 14,139 data points objectively affirmed Ile-D phenotypes are distinct from Met-D or Leu-D and identified metabolic and chromatin features as primary discriminators. We further demonstrated the metabolic and epigenetic responses to Ile-D can be recapitulated in vitro, suggesting that these responses are cell intrinsic. Together, these results demonstrate that dietary depletion of EAAs induce unique phenotypes and highlight distinct molecular mechanisms by which individual EAAs may control metabolic health.

  PublisherDOI

• Supplementary Materials, Figure S1-S4, Table S1-S9 from Functional and Clinical Characterization of Variants of Uncertain Significance Identifies a Hotspot for Inactivating Missense Variants in RAD51C

  2025-11-24

  articleOpen access

  &lt;p&gt;Supplementary Materials, References, Tables and Figures&lt;/p&gt;

  PublisherDOI

Recent grants

• Project 2: Relating the functional capacities of the GI and nasal microbiomes to immune development in children from traditional and modern exposures

  NIH · $36.4M · 2013–2028

Frequent coauthors

• Sean McIlwain

  University of Wisconsin–Madison

  123 shared

• Randall J. Kimple

  63 shared

• Michael A. Newton

  University of Wisconsin–Madison

  49 shared

• Kwangok P. Nickel

  43 shared

• Kristina A. Matkowskyj

  Mayo Clinic in Arizona

  38 shared

• Paul M. Sondel

  University of Wisconsin–Madison

  34 shared

• Robert Landick

  University of Wisconsin–Madison

  33 shared

• Mark R. Albertini

  University of Wisconsin Carbone Cancer Center

  30 shared

Labs

• Irene Ong LabPI

  Data mining, artificial intelligence, machine learning, probabilistic methods, dynamical models, inductive logic programming, and statistical relational learning with applications to biological and medical data.

Education

• Ph.D., Biostatistics

  University of Wisconsin-Madison

  2000

• M.S., Biostatistics

  University of Wisconsin-Madison

  1996

• B.S., Mathematics

  University of Wisconsin-Madison

  1994