Researchers at the Johns Hopkins Kimmel Cancer Center have developed a promising diagnostic assay utilizing a combination of machine learning and liquid biopsy technologies to enhance the detection of ovarian cancer. Presented at the American Association for Cancer Research (AACR) Annual Meeting 2024 in San Diego, this new assay utilizes cell-free DNA (cfDNA) fragment patterns and the levels of proteins CA125 and HE4 to distinguish patients with ovarian cancer from those with benign ovarian masses or no cancer.
Ovarian cancer is notorious for its late detection due to its asymptomatic early stages. Being the fifth leading cause of cancer deaths among women in the US, it presents a significant challenge to women's health. The disease's covert progression and the lack of efficient screening tools have contributed to its high mortality rate. According to Jamie Medina, Ph.D., a postdoctoral fellow involved in the study, "The lack of efficient screening tools, combined with the asymptomatic development of ovarian cancer, contributes to late diagnoses when effective treatment options are limited."
Working to address this challenge, the researchers turned to liquid biopsy technologies, specifically DELFI (DNA Evaluation of Fragments for Early Interception), which uses a newer approach known as fragmentomics. This method analyzes the size and distribution of cfDNA fragments across the genome, capturing the complex genomic landscape of cancer cells which contrasts with that of healthy cells. "By carefully analyzing these fragments across the entire human genome, we can detect subtle patterns indicating the presence of cancer," Medina elaborated.
The integration of fragmentomics with plasma levels of CA125 and HE4, through a sophisticated machine-learning algorithm, marks a significant advancement in the noninvasive detection of ovarian cancer. Victor Velculescu, M.D., Ph.D., FAACR, senior author of the study, highlighted how imperative it is for improved biomarkers in ovarian cancer detection, stating, "Ovarian cancer is an incredibly deadly disease with no great biomarkers for screening and early intervention. Our goal was to overcome this challenge by combining genome-wide cell-free DNA fragmentation with protein biomarkers to develop a new high-performance approach for early detection of ovarian cancer."
The research team analyzed plasma samples from a cohort including 134 women diagnosed with ovarian cancer, 204 cancer-free individuals, and 203 women with benign adnexal masses. The results revealed that the screening model, boasting a specificity of over 99%, identified a significant proportion of ovarian cancer cases across all stages, with an area under the curve (AUC) of 0.97, indicating a high level of accuracy. This performance notably surpasses that of current biomarkers, including CA125 levels alone.
While the diagnostic model demonstrated the ability to effectively differentiate between ovarian cancer and benign masses with an AUC of 0.87, the researchers plan to further validate their findings in larger cohorts. Velculescu expressed optimism about the potential of this approach, stating, "This study contributes to a large body of work from our group demonstrating the power of genome-wide cell-free DNA fragmentation and machine learning to detect cancers with high performance."
This innovative assay represents a significant step forward in the early detection of ovarian cancer, offering hope for more effective treatment and improved survival rates for patients.
Ovarian cancer is notorious for its late detection due to its asymptomatic early stages. Being the fifth leading cause of cancer deaths among women in the US, it presents a significant challenge to women's health. The disease's covert progression and the lack of efficient screening tools have contributed to its high mortality rate. According to Jamie Medina, Ph.D., a postdoctoral fellow involved in the study, "The lack of efficient screening tools, combined with the asymptomatic development of ovarian cancer, contributes to late diagnoses when effective treatment options are limited."
Working to address this challenge, the researchers turned to liquid biopsy technologies, specifically DELFI (DNA Evaluation of Fragments for Early Interception), which uses a newer approach known as fragmentomics. This method analyzes the size and distribution of cfDNA fragments across the genome, capturing the complex genomic landscape of cancer cells which contrasts with that of healthy cells. "By carefully analyzing these fragments across the entire human genome, we can detect subtle patterns indicating the presence of cancer," Medina elaborated.
The integration of fragmentomics with plasma levels of CA125 and HE4, through a sophisticated machine-learning algorithm, marks a significant advancement in the noninvasive detection of ovarian cancer. Victor Velculescu, M.D., Ph.D., FAACR, senior author of the study, highlighted how imperative it is for improved biomarkers in ovarian cancer detection, stating, "Ovarian cancer is an incredibly deadly disease with no great biomarkers for screening and early intervention. Our goal was to overcome this challenge by combining genome-wide cell-free DNA fragmentation with protein biomarkers to develop a new high-performance approach for early detection of ovarian cancer."
The research team analyzed plasma samples from a cohort including 134 women diagnosed with ovarian cancer, 204 cancer-free individuals, and 203 women with benign adnexal masses. The results revealed that the screening model, boasting a specificity of over 99%, identified a significant proportion of ovarian cancer cases across all stages, with an area under the curve (AUC) of 0.97, indicating a high level of accuracy. This performance notably surpasses that of current biomarkers, including CA125 levels alone.
While the diagnostic model demonstrated the ability to effectively differentiate between ovarian cancer and benign masses with an AUC of 0.87, the researchers plan to further validate their findings in larger cohorts. Velculescu expressed optimism about the potential of this approach, stating, "This study contributes to a large body of work from our group demonstrating the power of genome-wide cell-free DNA fragmentation and machine learning to detect cancers with high performance."
This innovative assay represents a significant step forward in the early detection of ovarian cancer, offering hope for more effective treatment and improved survival rates for patients.