Matthew R. Cooperberg
Has grant
Research Interests
Make your next move.
Prepare your application
Review drafts before you send them.
Explore related searches
Recent Grants
Grant: Close
Improving Prostate Cancer Outcome Prediction Through Noninvasive exRNA Assessment
Open Date: 2016-08-10
Close Date: 2021-07-31
Grant: Close
Prospective validation of a multi-marker prostate cancer prediction model
Open Date: 2012-08-08
Close Date: 2017-05-31
Articles (15)
Background Multiparametric MRI is useful for early detection of clinically significant prostate cancer (csPCa), but its standard Apparent Diffusion Coefficient (ADC) has limited utility as a quantitative metric for automated, patient-level detection of csPCa. Restriction Spectrum Imaging (RSI), an advanced diffusion technique, yields a quantitative biomarker (RSIrs) that improves csPCa detection. RSIrs is typically calculated from a dedicated multi- b -value acquisition. RSIrs estimated from conventional MRI has not been studied. Purpose To evaluate the accuracy and validity of RSI metrics estimated post-hoc from conventional diffusion-weighted imaging (DWI) to serve as a viable surrogate for a dedicated RSI acquisition. Materials and Methods We conducted a retrospective, multi-center study of patients with both a dedicated RSI acquisition and conventional DWI. We compared three different RSI restriction score (RSIrs) calculation methods: from the dedicated acquisition (RSIrs dedicated ), from conventional DWI alone (RSIrs post-hoc ), and from a combination of conventional DWI with only the high b -values from the RSI acquisition (RSIrs combo ). We compared these methods for quantitative agreement and csPCa detection performance (Area under the Receiver Operating Characteristic [AUC, 95% Confidence Interval]) of maximum RSIrs (RSIrs max ) in the prostate compared to that of minimum ADC (ADC). Results Data from n=1095 patients (16 centers) were analyzed. Post-hoc RSIrs max differed systematically from RSIrs dedicated by a median of +156 (RSIrs post-hoc ) and −59 (RSIrs combo ), respectively. AUCs for csPCa detection were 0.51 [0.47,0.54], 0.60 [0.57,0.64], 0.70 [0.67,0.74], and 0.77 [0.74,0.80] for ADC, RSIrs post-hoc , RSIrs combo , and RSIrs dedicated , respectively. Conclusion Even when estimated using conventional DWI, RSIrs is a superior quantitative biomarker to ADC for automated, patient-level detection of csPCa. A dedicated RSI acquisition gives the best performance. A compromise would be to acquire high b- values (1500 s/mm 2 , 2500 s/mm 2 ) to complement low b- values (<1000 s/mm 2 ) from conventional DWI.
Year:
2025
Objectives To evaluate the longitudinal transcriptomic changes that occurred over time in patients from the ENACT trial (ClinicalTrials.gov identifier: NCT02799745). ENACT evaluated patients with low‐ or intermediate‐risk prostate cancer, comparing the efficacy and safety of enzalutamide plus active surveillance (AS) to AS alone. Patients and Methods Gene‐expression profiling was conducted using the Decipher Genomic Resource for Intelligent Discovery (GRID) platform (Veracyte, Inc.) on 131 patient samples (enzalutamide plus AS: n = 57; AS alone: n = 74) collected at screening, Year 1, and Year 2. Pre‐defined GRID signatures were analysed for associations with biological and clinical features, including immune activity, treatment sensitivity, metastatic risk, and molecular subtypes. Statistical analyses utilised Cox proportional hazards models and logistic/linear regression, while longitudinal changes were evaluated using linear mixed‐effects models. Results At Year 1, patients treated with enzalutamide plus AS showed significant transcriptomic changes, such as downregulation of androgen‐receptor signalling and immune‐suppressor signatures, alongside upregulation of immune‐activated and basal‐like markers. Many of these transcriptomic alterations, including immune‐related changes, were transient and reverted to baseline levels at Year 2 after enzalutamide treatment cessation. Conclusion This exploratory biomarker analysis highlights transcriptomic changes in patients undergoing AS and those treated with enzalutamide, providing insights into molecular disease progression on surveillance and the effects of transient exposure to enzalutamide. The observed immune modulation during enzalutamide treatment suggests opportunities to explore immunotherapy strategies.
Year:
2025
Supplementary Figure 6 shows clinical data analysis of neoadjuvant androgen axis inhibition in primary prostate cancer as well as expression analysis of AR, MHCI, and IFNG activity in mHSPC clinical samples.
Year:
2025
Year:
2025
Supplementary Figure 6 shows clinical data analysis of neoadjuvant androgen axis inhibition in primary prostate cancer as well as expression analysis of AR, MHCI, and IFNG activity in mHSPC clinical samples.
Year:
2025
Year:
2025
Tumors escape immune detection and elimination through a variety of mechanisms. Here, we used prostate cancer as a model to examine how androgen-dependent tumors undergo immune evasion through downregulation of the major histocompatibility complex class I (MHCI). We report that response to immunotherapy in late-stage prostate cancer is associated with elevated MHC expression. To uncover the mechanism, we performed a genome-wide CRISPR interference (CRISPRi) screen and identified androgen receptor ( AR ) as a repressor of the MHCI pathway. Syngeneic mouse models of aggressive prostate cancer deficient in Ar also demonstrated increased tumor immunogenicity and promoted T cell–mediated tumor control. Notably, the increase in MHCI expression upon AR blockade is transient and correlates with resistance to AR inhibition. Mechanistic studies identified androgen response elements upstream of MHCI transcription start sites which increased MHCI expression when deleted. Together, this body of work highlights another mechanism by which hormones can promote immune escape. Significance: Immunotherapy options for immune cold tumors, like prostate cancer, are limited. We show that AR downregulates MHCI expression/antigen presentation and that AR inhibition improves T-cell responses and tumor control. This suggests that treatments combining AR inhibitors and checkpoint blockade may improve tumor immune surveillance and antitumor immunity in patients.
Year:
2025
Background: The prognosis of diabetic men with advanced prostate cancer is poorly understood and understudied. Hence, we studied associations between diabetes and progression to metastases, prostate cancer–specific mortality (PCSM) and all-cause mortality (ACM) in men with nonmetastatic castration-resistant prostate cancer (nmCRPC). Methods: Data from men diagnosed with nmCRPC between 2000 and 2017 at 8 Veterans Affairs Health Care Centers were analyzed using Cox regression to determine HRs and 95% confidence intervals (CI) for associations between diabetes and outcomes. Men with diabetes were classified according to: (i) ICD-9/10 codes only, (ii) two HbA1c values > 6.4% (missing ICD-9/10 codes), and (iii) all diabetic men [(i) and (ii) combined]. Results: Of 976 men (median age: 76 years), 304 (31%) had diabetes at nmCRPC diagnosis, of whom 51% had ICD-9/10 codes. During a median follow-up of 6.5 years, 613 men were diagnosed with metastases, and 482 PCSM and 741 ACM events occurred. In multivariable-adjusted models, ICD-9/10 code-identified diabetes was inversely associated with PCSM (HR, 0.67; 95% CI, 0.48–0.92) while diabetes identified by high HbA1c values (no ICD-9/10 codes) was associated with an increase in ACM (HR, 1.41; 95% CI, 1.16–1.72). Duration of diabetes, prior to CRPC diagnosis was inversely associated with PCSM among men identified by ICD-9/10 codes and/or HbA1c values (HR, 0.93; 95% CI, 0.88–0.98). Conclusions: In men with late-stage prostate cancer, ICD-9/10 ‘code-identified’ diabetes is associated with better overall survival than ‘undiagnosed’ diabetes identified by high HbA1c values only. Impact: Our data suggest that better diabetes detection and management may improve survival in late-stage prostate cancer.
Year:
2025
UNSTRUCTURED Background Manual abstraction of unstructured clinical data is often necessary for granular clinical outcomes research, but is time consuming and can be of variable quality. Large language models (LLMs) show promise in medical data extraction, yet integrating them into research workflows remains challenging and poorly described. We developed and integrated an LLM-based system for automated data extraction from unstructured electronic health record (EHR) text reports within an established clinical outcomes database. Methods We implemented a generative AI pipeline (UODBLLM) utilizing a flexible language model interface that supports various LLM implementations, including HIPAA-compliant cloud services and local open-source models. We used XML-structured prompts and integrated using an open database connectivity interface to generate structured data from clinical documentation in the EHR. We evaluated UODBLLM's performance on completion rate, processing time, and extraction capabilities across multiple clinical data elements, including quantitative measurements, categorical assessments, and anatomical descriptions. System reliability was tested across multiple batches to assess scalability and consistency. Results Piloted against MRI reports, UODBLLM processed 1,800 clinical documents with a 100% completion rate and an average processing time of 8.90 seconds per report. Token utilization averaged 2,692 tokens per report, with an input-to-output ratio of approximately 6.5:1, resulting in a processing cost of $0.009 per report. UODBLLM had consistent performance across 18 batches of 100 reports each and completed all processing in 4.45 hours. From each report, UODBLLM extracted 16 structured clinical elements, including prostate volume, PSA values, PI-RADS scores, clinical staging, and anatomical assessments. All extracted data was automatically validated against predefined schemas and stored in standardized JSON format. Conclusion We demonstrated successful integration of an LLM-based extraction system within an existing clinical outcomes database, achieving rapid, comprehensive data extraction at minimal cost. UODBLLM provides a scalable, efficient solution for automating clinical data extraction while maintaining PHI security. This approach could significantly accelerate research timelines and expand feasible clinical studies, particularly for large-scale database projects.
Year:
2024
Background Local conditions where people live continue to influence prostate cancer outcomes. By examining local characteristics associated with trends in Black-White differences in prostate cancer–specific mortality over time, we aim to identify factors driving county-level prostate cancer–specific mortality disparities over a 15-year period. Methods We linked county-level data (Area Health Resource File) with clinicodemographic data of men with prostate cancer (Surveillance, Epidemiology, and End Results registry) from 2005 to 2020. Generalized linear mixed models evaluated associations between race and county-level age-standardized prostate cancer–specific mortality, adjusting for age; year of death; rurality; county-level education; income; uninsured rates; and densities of urologists, radiologists, primary care practitioners, and hospital beds. Results In 1085 counties, 185 390 patients were identified of which 15.8% were non-Hispanic Black. Racial disparities in prostate cancer–specific mortality narrowed from 2005 to 2020 (25.4 per 100 000 to 19.2 per 100 000 overall, 57.9 per 100 000 to 38 per 100 000 for non-Hispanic Black patients, and 23.4 per 100 000 to 18.3 per 100 000 for non-Hispanic White patients). For non-Hispanic Black and non-Hispanic White patients, county prostate cancer–specific mortality changes varied greatly (-65% to +77% and -61% to +112%, respectively). From 2016 to 2020, non-Hispanic Black patients harbored greater prostate cancer–specific mortality risk (relative risk = 2.09, 95% confidence interval [CI] = 2.01 to 2.18); higher radiation oncologist density was associated with lower mortality risk (relative risk = 0.93, 95% CI = 0.89 to 0.98), while other practitioner densities were not. Conclusion Although overall rates improved, specific counties experienced worsening race-based disparities over time. Identifying locations of highest (and lowest) mortality disparities remains critical to development of location-specific solutions to racial disparities in prostate cancer outcomes.
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2023
Collaborators (22)
Thomas O'Loughlin
Icahn School of Medicine at Mount Sinai
Anobel Y Odisho
University of California, San Francisco
William Aronson
Chief of Urologic Oncology
VA Greater Los Angeles Healthcare System
Davide Ruggero
Professor
University of California
David A. Quigley
Assistant Professor
University of California, San Francisco
Wilbert Zwart
Senior Groupleader / professor
Eindhoven University of Technology
Luke Gilbert
Assistant Professor
University of California, San Francisco
Alan Ashworth
University of California, San Francisco
Zheng Xia
Associate Professor
University of Texas at Austin
Adam Sowalsky
Senior Investigator
National Cancer Institute
Michelle Badura
Lecturer
Santa Clara University
Tyler Seibert
Assistant Professor
University of California, San Diego
Duygu Kuzuoglu Ozturk
Assistant Prof
Sabancı University
Raunak Shrestha
Assistant Professor
Uniformed Services University of the Health Sciences
Peter Carroll
University of California, San Francisco
Zachary Klaassen
-
Lawrence Fong
Fred Hutch Cancer Center
Arian Lundberg
Assistant Professor at KTH Royal Institute of Technology
KTH Royal Institute of Technology
Felix Y. Feng
University of California
Samuel Washington
University of California, San Francisco
Ilona Csizmadi
-
Jessica Hawley
Assistant Professor
University of Washington
Make your next move.
Prepare your application
Review drafts before you send them.