About

Kenneth Iczkowski, M.D., is a professor at UC Davis Health within the Department of Pathology and Laboratory Medicine. His clinical specialties include pathology, with a focus on surgical pathology, cardiovascular pathology, transplant pathology, and urologic pathology. He serves as the head of urological anatomic pathology and practices in cardiovascular and general pathology, providing diagnostic consultation for prostate, bladder, testicular, kidney, and penile cancers. Dr. Iczkowski's research centers on pathologic-radiologic correlation using artificial intelligence to match microscopic prostate findings with radiological imaging. He is involved in collaborations related to therapeutic targeting of bladder cancer and has a background in basic science research on prostate cancer, including cell adhesion molecules and pathways involved in growth and metastasis. His academic background includes a B.A. in Biology from Columbia College of Columbia University, an M.D. from St. Louis University, and specialized training through residencies and fellowships at Dartmouth-Hitchcock Medical Center, University of Kansas, and Mayo Clinic Rochester. He has received numerous awards for his educational contributions and research, including the Medical College of Wisconsin's Teacher of the Year and the Paul Albert Grawitz medal from the International Society of Urological Pathology.

Research topics

• Biology
• Genetics
• Cancer research
• Nuclear medicine
• Radiology
• Cell biology
• Internal medicine
• Medicine

Selected publications

• How does AI perform compared to human expert panels in medical Delphi studies? A pilot study through the lens of pathology

  Journal of Pathology Informatics · 2026-04-01

  articleOpen access

  Background: Since their inception, Delphi studies have been a key part of medical literature. They consist of an expert panel tasked with coming to consensus on answers to various questions where obtaining objective results is difficult or impossible, with ranked responses based on a Likert scale. The ability of artificial intelligence (AI), particularly large language models (LLMs), to perform this role traditionally assigned to a panel of experts has been scarcely explored in medicine. This study accordingly aimed to explore the feasibility of an "AI-run" Delphi study applied to the practice of pathology. Methods: A prior human-based Delphi study (PMID: 36603288) employed to forecast the future role of AI in pathology was repeated, but this time with LLMs (Llama 3, ChatGPT-4, and ChatGPT-3.5 based on availability at the time of the study). This was done at various temperature settings (0, 0.7, and 1.0), a measurement of how much an LLM prioritizes determinism versus creativity. Low temperature caused the models to be more deterministic and focused, whereas high temperature increased creativity. "Delphi-GPT" was created to automate prompts that entailed 5 trials for 180 questions, leading to data that were compared to the original human expert panel. Findings: All LLM and temperature combinations were able to reach consensus for a greater percentage of the 180 questions posed than human experts. Newer ChatGPT-4 and Llama 3 models performed better than ChatGPT-3.5. Whereas AI models and human experts did not always agree, the amount of agreement increased when the temperature setting was increased across all LLMs. Interpretation: LLMs are shown here to successfully be able to simulate a Delphi study in medicine. The data show that generative AI models were consistently able to reach greater degrees of consensus than human experts in their responses to 180 prompts related to the future practice of pathology. This serves as a proof-of-concept that one day, pending further robust methodological validation, AI could even serve as a surrogate for de novo Delphi studies that ordinarily would have relied on feedback from a panel of experts. The reliability of consensus/concordance achieved will depend upon the combination of LLM and temperature setting selected.

  PublisherDOI

• Genitourinary Pathology Society and International Society of Urological Pathology White Paper on Defining Indolent Prostate Cancer

  The American Journal of Surgical Pathology · 2025-05-30 · 6 citations

  articleOpen accessCorresponding

  A significant subset of well-differentiated prostatic acinar neoplasms with invasive histologic features will not spread outside of the prostate, become symptomatic, or shorten a patient's life even if the tumor is left untreated. Overdiagnosis and overtreatment of these indolent prostate cancers (PCa) remain a significant health care problem despite the improved risk assessment and uptake in acceptance of conservative management. While detection of indolent PCa on an entirely resected prostate is possible, recognition of indolent PCa on a needle biopsy (NBX) cannot be reliably made as Grade Group 1 (GG1) PCa diagnosis on NBX is not always identical to one from radical prostatectomy due to a variety of reasons. Further, some of the initially diagnosed GG1 PCas on NBX and carefully monitored on active surveillance (AS) are later reclassified with higher grades. At the same time, other GG1 PCas never progressed on long-term follow-up while receiving no therapy. The overarching goal of this white paper by the 2 leading uropathology organizations, Genitourinary Pathology Society (GUPS) and International Society of Urological Pathology (ISUP), is to help identify a path toward a more meaningful multidisciplinary solution addressing the pervasive problem of overdiagnosis of indolent PCa and its downstream negative effects. Herein, GUPS and ISUP jointly release statements that address why recognition of indolent PCa cannot be reliably made in NBX and why various contemporary multidisciplinary approaches are needed to help improve the detection of indolent PCa in NBX.

  PublisherOA PDFDOI

• Alteration in expression and subcellular localization of the androgen receptor- regulated FAM111A protease is associated with emergence of castration resistant prostate cancer

  Neoplasia · 2025-05-29

  articleOpen access

  • The AR represses FAM111A protease transcription in multiple castration sensitive and resistant prostate cancer cells. • Metastatic lesions exhibit decreased FAM111A expression when compared to primary PCa tumors. • FAM111A subcellular localization is restricted to nucleoli in castration sensitive cells but becomes progressively redistributed to the nuclear and cytoplasmic compartments with acquisition of castration resistance. • Depletion of FAM111A sensitizes castration sensitive and resistant cells to PARP inhibitors. • Expression of AR-regulated transcripts decreases on FAM111A depletion indicative of an AR-FAM111A regulatory loop. The androgen receptor (AR) is a pivotal regulator of growth and survival of prostate cancer (PCa) and the majority of lethal castration-resistant prostate cancers (CRPC) remain reliant on AR signaling. PCa exhibits variability in progression and responses to treatment suggesting genetic heterogeneity. Two independent studies identified PCa predisposing single nucleotide polymorphisms (SNPs) within the FAM111A protease gene, but the mechanistic basis of this association remained elusive. Our in vitro and in vivo studies uncovered that AR represses FAM111A in castration sensitive and resistant cells via an AR binding site within the FAM111A gene. FAM111A levels are significantly lower in matched castration-resistant than in castration-sensitive cells and xenografts, and lower in metastatic lesions than in primary tumors. We discovered that FAM111A is AR-repressed in castration sensitive PCa xenograft and multiple PCa cells. Additionally, FAM111A subcellular localization changes dramatically with acquisition of castration resistance, where in castration sensitive cells FAM111A is predominantly in the nucleoli, but with castration resistance it becomes more dispersed in the nucleus and in the cytoplasm. FAM111A depletion in castration sensitive and resistant cells enhances the efficacy of PARP1 inhibitors olaparib and niraparib, consistent with its role in DNA repair. Moreover, FAM111A depletion reduces AR target gene prostate specific antigen ( PSA) and transmembrane serine protease 2 ( TMPRSS2) transcription, indicating that FAM111A modulates AR-dependent gene expression forming a FAM111A-AR co-regulatory loop in PCa. Our studies argue that AR-dependent FAM111A regulation modulates PCa gene expression, acquisition of castration resistance, and sensitivity to agents that target DNA damage repair.

  PublisherDOI

• 790 Current Reporting Trends, Practices, and Resource Utilization in Penile Cancer: A Survey among Genitourinary Pathologists

  Laboratory Investigation · 2025-03-01

  article
  PublisherDOI

• 818 Digital Whole-Slide Inter-Observer Concordance of Contemporary Prostatic Biopsy Grading

  Laboratory Investigation · 2025-03-01

  article1st authorCorresponding
  PublisherDOI

• Clinical Application of Large Language Models in Generating Pathologic Images

  JCO Clinical Cancer Informatics · 2025-07-01

  article

  PURPOSE: This study investigates the potential of DALL·E 3, an artificial intelligence (AI) model, to generate synthetic pathologic images of prostate cancer (PCa) at varying Gleason grades. The aim is to enhance medical education and research resources, particularly by providing diverse case studies and valuable teaching tools. METHODS: This study uses DALL·E 3 to generate 30 synthetic images of PCa across various Gleason grades, guided by standard Gleason pattern descriptions. Nine uropathologists evaluated these images for realism and accuracy compared with actual hematoxylin and eosin (H&E)-stained slides using a scoring system. RESULTS: < .05), with Gleason 5 images achieving the highest scores and accurately depicting critical pathologic characteristics. Limitations included a lack of fine nuclear detail, essential for identifying malignancy, which may affect the images' diagnostic utility. CONCLUSION: DALL·E 3 shows promise in generating customized pathologic images that can aid in education and resource expansion within pathology. However, ethical concerns, such as the potential misuse of AI-generated images for data falsification, highlight the need for responsible oversight. Collaboration between technology firms and pathologists is essential for the ethical integration of AI in pathology practices.

  PublisherDOI

• 922 Determinants of Oncologic Outcomes in High Grade Organ-Confined Prostate Cancer After Radical Prostatectomy

  Laboratory Investigation · 2025-03-01

  article
  PublisherDOI

• Corrigendum to “Magnetic Resonance Imaging (MRI)–Adapted Prostate Cancer Risk Tool Incorporating Cribriform and Intraductal Carcinoma” (Modern Pathology 2025 Dec;38(12):100852)

  Modern Pathology · 2025-10-07

  article
  PublisherDOI

• Magnetic Resonance Imaging (MRI)–Adapted Prostate Cancer Risk Tool Incorporating Cribriform and Intraductal Carcinoma

  Modern Pathology · 2025-07-25 · 1 citations

  article
  PublisherDOI

• Immunoexpression of MED12 in lobular carcinoma in situ confined within fibroadenoma of breast

  Human Pathology Reports · 2025-11-01

  articleOpen accessCorresponding

  Fibroadenomas are common lesions in young women. Approximately 60–80% of fibroadenomas harbor a somatic mutation in exon 2 of the mediator complex MED12 subunit. MED12 regulates physiologic processes important for cell development. Dysregulation of MED12 has been described as linked to estrogen and TGF-β signaling. MED12 subunit mutations in fibroadenomas localize to the stromal cells and not the epithelial cells. Frequent MED12 subunit mutations have been detected in breast cancer as well as estrogen-dependent benign tumors, such as fibroadenoma, uterine leiomyoma, and phyllodes tumor. An association of lobular carcinoma in situ with fibroadenoma has been reported. The LCIS in this case was entirely confined to the fibroadenoma. This is the first reported case of immunohistochemical overexpression of MED12 in LCIS arising within a fibroadenoma.

  PublisherDOI

Frequent coauthors

• David J. Grignon

  Indiana University School of Medicine

  10 shared

• Lars Egevad

  10 shared

• Glen Kristiansen

  University Hospital Bonn

  9 shared

• Kátia Ramos Moreira Leite

  Universidade de São Paulo

  9 shared

• Martin Eklund

  9 shared

• James G. Kench

  Royal Prince Alfred Hospital

  8 shared

• Hemamali Samaratunga

  University of Queensland

  8 shared

• Nihal Ahmad

  William S. Middleton Memorial Veterans Hospital

  7 shared

Labs

• Department of Pathology and Laboratory MedicinePI

Awards & honors

• Faculty Teacher of the Year plaque. Medical College of Wisco…
• Outstanding Medical Student Teacher, 2016-2017, 2014-2015, M…
• Paul Albert Grawitz medal awarded by International Society o…
• Distinction in Research Award, and Herbert B. Taylor Award i…
• Good Catch Award, Pathology at UC-Davis (diagnostic error pr…