# Reducing prostate cancer mortality in African American men by early disease detection using DNA fragmentation profiles and methylated DNA patterns of plasma derived cell free DNA

> **NIH NIH R21** · BOSTON CHILDREN'S HOSPITAL · 2024 · $475,759

## Abstract

Project Summary/Abstract:
African American (AA) individuals with prostate cancer (PCa) face significantly worse clinical outcome than
their European American (EA) counterparts. There is a strong correlation between the stage of disease
(localized versus metastatic) at diagnosis and long-term survival. While patients diagnosed with localized
disease have a 99% 5-year survival, patients presenting with metastatic disease have significantly worse
prospects, an about 32% 5-year survival. A minimally invasive, plasma based diagnostic method could
significantly improve chances of detecting PCa at an early stage and thus reduce PCa related mortality.
There are two, next generation sequencing (NGS)-based methods that likely have the required sensitivity and
specificity of early detection of PCa from liquid, plasma biopsies. The first is based on the altered fragmentation
profile of cancer derived cell free DNA (cfDNA). In this, whole genome sequencing (WGS) is applied to plasma
derived cfDNA and the ratio of short to normal fragment size indicates the presence of cancer. The second
method is based on a specific pattern of methylated DNA loci across the genome and combines cell-free
methylated DNA immunoprecipitation with high throughput sequencing.
We found that the altered cell free DNA fragmentation profile is a highly sensitive, robust indicator of the presence
of metastatic prostate cancer. In this proposal we will investigate whether this method can identify cases when
AA men present at diagnosis or at later time-point after post-primary treatment with metastatic PCa. We will also
investigate whether a clinically useful sensitivity and specificity is retained as we analyze samples at earlier
stages of disease.
We will benchmark and optimize an experimental and computational protocol to detect prostate cancer at various
stages based on the fragmentation profile of plasma derived cell free DNA. This will be applied to AA patients
that presented with metastatic disease at diagnosis or later developed metastatic disease, to AA PCa patients
with localized or locally advanced disease, and to AA PCa cases where sequential plasma was collected at
various times, ranging from right before surgery to ten years before disease progression.
We will determine the sensitivity and specificity of a fragmentomics profile-based method for early detection of
PCa of various disease stages.
Similarly, we will benchmark and optimize an experimental and computational protocol to detect prostate cancer
at various stages based on cell free methylated DNA profiles. The next generation sequencing based methylation
profiling of plasma derived cfDNA samples will be applied to AA PCa of various disease stages and we will
determine the sensitivity and specificity of DNA methylation profile-based method for early detection of PCa of
various disease stages.
This will establish a non-invasive method for the early detection of PCa of AA men.

## Key facts

- **NIH application ID:** 10951041
- **Project number:** 1R21CA292755-01
- **Recipient organization:** BOSTON CHILDREN'S HOSPITAL
- **Principal Investigator:** Zoltan Szallasi
- **Activity code:** R21 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2024
- **Award amount:** $475,759
- **Award type:** 1
- **Project period:** 2024-09-06 → 2026-08-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10951041

## Citation

> US National Institutes of Health, RePORTER application 10951041, Reducing prostate cancer mortality in African American men by early disease detection using DNA fragmentation profiles and methylated DNA patterns of plasma derived cell free DNA (1R21CA292755-01). Retrieved via AI Analytics 2026-05-26 from https://api.ai-analytics.org/grant/nih/10951041. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*