Early Cancer Detection with Liquid Biopsy

1. Background

Cancer remains a critical global health challenge, with increasing incidence and mortality rates, particularly in developing countries such as Vietnam. Despite advancements in diagnostic imaging and traditional biopsy techniques, early cancer detection continues to be hindered by high costs, invasiveness, and limited accessibility. Liquid biopsy, which analyzes cell-free DNA (cfDNA) in blood samples, has emerged as a promising non-invasive diagnostic tool. This study aims to develop an early detection model for six prevalent cancers in Vietnam—lung, liver, breast, colorectal, gastric, and kidney cancer—by integrating liquid biopsy and clinical epidemiological factors.

2. Research Approach

This research was conducted in collaboration with the Ludwig Center for Cancer Genetics and Therapeutics at Johns Hopkins University, following a case-control study design. The study was implemented across multiple hospitals in Ho Chi Minh City in cooperation with the Vietnam Osteoporosis Study (VOS). The primary objectives of this research included:

• Identifying clinical and epidemiological factors associated with cancer in the Vietnamese population.
• Establishing genetic mutation profiles for six common cancer types.
• Developing an early cancer diagnostic model using cfDNA and clinical epidemiological factors.
• Validating the effectiveness of the model in distinguishing early-stage cancer patients from healthy individuals.

3. Data Collection and Management

3.1 Patient Recruitment

Cancer patients were recruited based on confirmed pathological diagnoses from four hospitals:

• Ho Chi Minh City Oncological Hospital
• Ho Chi Minh City University of Medicine and Pharmacy Hospital
• Pham Ngoc Thach Hospital
• Binh Dan Hospital

Control groups were matched by age and sex and sourced from the Vietnam Osteoporosis Study (VOS).

3.2 Sample Collection

Blood samples were collected and processed to extract cell-free DNA (cfDNA) for genomic analysis.

3.3 Genetic and Clinical Data Analysis

• Whole-genome sequencing (WGS) and bioinformatics techniques were utilized to detect genetic mutations, methylation patterns, and fragmentation profiles indicative of cancer.
• Machine learning algorithms, including logistic regression, support vector machines (SVM), and random forest models, were tested to construct the most effective predictive model.

4. Key Findings

• Breast Cancer Trends (1):

  • The incidence of breast cancer in Ho Chi Minh City has increased by approximately 70% over the past two decades.
  • Vietnamese women tend to develop breast cancer at younger ages compared to white populations, suggesting that screening programs should target younger women.

• Colorectal Cancer (CRC) Trends (2):

  • The incidence of CRC has risen significantly, especially among men, older populations, and individuals born after 1975.
  • The increasing prevalence of CRC is likely linked to the adoption of a Westernized lifestyle.

• Thyroid Cancer Trends (3):

  • There has been a marked increase in thyroid cancer incidence in both genders, particularly in the papillary subtype.
  • The average age at diagnosis has gradually declined over time.

• Dietary Influence on Cancer Risk (4):

  • Greater consumption of vegetables, fruits, soybean products, coffee, and eggs is associated with a reduced risk of breast cancer.
  • These findings provide a basis for culturally tailored dietary recommendations to lower breast cancer risk, particularly in low- and middle-income countries (LMICs).

• Genomic Insights (5):

  • Alu elements, particularly AluS subfamily elements, show altered representation in the cfDNA of cancer patients.
  • Further research is required to determine whether Alu representation can enhance the sensitivity of WGS-based diagnostics.

• Advancements in Liquid Biopsy (6):

  • The study introduced MethylSaferSeqS, a novel approach that simultaneously evaluates both genetic and epigenetic alterations in the same DNA molecules.
  • This technique has demonstrated its capability to detect mutations, copy number variations, and methylation patterns in cfDNA from both cancer patients and healthy individuals.

• Diagnostic Model Performance:

  • The developed diagnostic model based on fragmentomics and clinical epidemiological factors exhibited high sensitivity and specificity in distinguishing cancer patients from healthy individuals, particularly in early-stage cancers.
  • Liquid biopsy proved to be a highly effective, minimally invasive tool for early cancer detection.

5. Future Directions

• Expanding the study to include a larger and more diverse population to enhance the generalizability of the findings.
• Integrating additional biomarkers such as proteins and RNA to improve diagnostic accuracy.
• Conducting clinical trials to facilitate the implementation of liquid biopsy-based diagnostic models in routine cancer screening programs.

6. Conclusion

This research significantly contributes to Vietnam’s National Cancer Control Strategy by providing a scientific basis for improving early cancer detection. The implementation of innovative, accessible, and minimally invasive diagnostic methods has the potential to reduce cancer-related mortality and improve patient outcomes.

Saigonmec promotes open research and international collaboration in precision medicine by encouraging data sharing and collaborative research. We believe open access to our well-curated datasets can accelerate research and enable breakthroughs. If you have research ideas that could benefit from our datasets, please contact us at contact@saigonmec.org

Article date: Feb 14, 2025. Author: Dr. Lan T. Ho Pham