AI-Enhanced Blood Test Shows Promise in Early Detection of Ovarian Cancer

Johns Hopkins Kimmel Cancer Center Leads Study on Novel Screening Method

A groundbreaking study conducted by researchers at the Johns Hopkins Kimmel Cancer Center, in collaboration with several institutions across the United States and Europe, has demonstrated the potential of a novel blood test to detect ovarian cancer at its early stages. Using artificial intelligence (AI) to analyse cancer-related genetic changes and protein biomarkers, the test marks a significant advancement in cancer screening.

Published Findings and Principal Researchers

The study, published on 30th September in the journal Cancer Discovery, utilised AI-powered analyses of DNA fragments coupled with two known protein biomarkers, cancer antigen 125 (CA-125) and human epididymis protein 4 (HE4). While CA-125 and HE4 alone struggled to reliably detect ovarian cancer, their combination with AI analysis significantly improved screening accuracy by distinguishing between cancerous and benign ovarian growths.

Dr. Victor E. Velculescu, the senior author of the study, and professor of oncology at the Johns Hopkins Kimmel Cancer Center, highlighted the potential of this approach. "The combination of artificial intelligence, cell-free DNA fragmentomes and a pair of protein biomarkers in a simple blood test improved detection of ovarian cancer even in patients with early-stage disease," he explained. This method, according to Velculescu, could become an affordable and accessible screening tool.

Ovarian Cancer Context

Ovarian cancer is the fifth deadliest cancer among women in the United States, with a five-year survival rate approximating 50%, as per the Centers for Disease Control and Prevention (CDC). Early detection is critical, yet most women are diagnosed at a later stage when survival rates plummet.

Jamie Medina, Ph.D., a postdoctoral fellow at the Johns Hopkins Kimmel Cancer Center and co-first author, elaborated on the significance of early diagnosis. "Early detection of ovarian cancer may save lives, but most women are diagnosed late in the course of the disease when survival rates are much lower," Medina stated. The paucity of specific early symptoms and effective biomarkers has historically hampered early detection efforts.

Methodology and Results

The DELFI (DNA Evaluation of Fragments for Early Interception) test, which the researchers employed, uses fragmentomics—analysing the fragmented DNA in blood derived from tumour cells. Cancer cells, unlike healthy cells, leave disorganised DNA fragments when they die and break apart. This disorganisation was exploited in the DELFI-Pro test to enhance detection capabilities.

Utilising blood samples from 94 women with ovarian cancer, 203 women with benign ovarian tumours, and 182 women without known ovarian growths, the study had a robust sample size. The DELFI-Pro test successfully identified 72%, 69%, 87%, and 100% of ovarian cancer cases in stages I to IV, respectively. In comparison, CA-125 alone detected 34%, 62%, 63%, and 100% at the same specificity.

Further confirming these findings, the test was applied to a smaller American sample group consisting of 40 ovarian cancer patients, 50 with benign growths, and 22 without any known ovarian lesions. Even in this smaller cohort, the test detected 73% of all cancers and 81% of high-grade serous ovarian carcinoma, the most aggressive form of the disease, while producing almost no false positives.

Potential Impact and Future Directions

The study's promising results suggest that the AI-enhanced DELFI-Pro test could significantly improve the accuracy of ovarian cancer screenings and may even reduce the need for exploratory surgeries in women with benign growths detected through ultrasound.

Akshaya Annapragada, an M.D./Ph.D. student at the Johns Hopkins University School of Medicine and co-first author, underscored the test's ability to distinguish between cancerous and benign lesions, a distinction not possible with current ultrasound exams.

While the researchers are optimistic, Velculescu and his team plan to validate the test further through larger samples from randomised clinical trials. "This study provides further evidence demonstrating the benefit of genome-wide, cell-free DNA fragmentation and artificial intelligence to detect cancers with high accuracy. Our results show that this combined approach has higher performance for screening than existing biomarkers," Velculescu concluded.

References

  • Jamie E. Medina et al., Early detection of ovarian cancer using cell-free DNA fragmentomes and protein biomarkers, Cancer Discovery (2024), DOI: 10.1158/2159-8290.CD-24-0393

Source: Noah Wire Services