Genomics is becoming increasingly integrated into multiple areas of human health, from prenatal diagnosis and early cancer detection to the prevention of genetic disease and precision treatment. At the 25th Anniversary International Symposium on the Human Genome, dedicated sessions on reproductive genomics and cancer genomics examined both recent scientific advances and the challenges involved in translating genomics discoveries into clinical practice.
Dennis Lo, Academician of the Chinese Academy of Sciences and President of The Chinese University of Hong Kong, discussed the clinical potential of cell-free DNA (cfDNA). In 1997, Lo was the first to demonstrate the presence of fetal-derived cfDNA in the peripheral blood of pregnant women, a discovery that laid the foundation for non-invasive prenatal testing (NIPT) for Down syndrome. The technology has since been adopted in more than 100 countries and has benefited more than 100 million pregnant women worldwide. His pioneering achievements in this field have earned him the Future Science Prize in Life Sciences (2016), the Breakthrough Prize in Life Sciences (2021), and the Lasker–DeBakey Clinical Medical Research Award.
Lo subsequently extended the concept of cfDNA analysis to the early detection of nasopharyngeal carcinoma. Using plasma Epstein–Barr virus DNA, his team conducted a cohort study involving more than 20,000 individuals at elevated risk. Repeated testing enabled risk stratification and demonstrated that persistently positive results were associated with substantially increased cancer risk, providing clinical evidence for more personalized screening strategies. Importantly, cfDNA carries not only genetic sequence information; features such as fragment length and DNA fragment-end structure also have diagnostic value.
His presentation also highlighted fragmentomics as an emerging direction in genomic diagnostics. Cell-free DNA contains information beyond its nucleotide sequence: fragment length, DNA end structures, and fragmentation patterns can themselves provide diagnostically relevant signals. Lo’s team developed the FRAGMA analytical approach to infer methylation information from DNA cleavage patterns, as well as a “four-end sequencing” strategy designed to preserve native DNA-end information and combine it with artificial intelligence to extract multidimensional fragment features.
The field is moving beyond simply detecting target DNA toward a more comprehensive interpretation of fragment-generation mechanisms, tissue of origin, and disease associations. Fragmentomics is therefore expanding the information content of genomic diagnostics and opening a new path toward the next generation of non-invasive liquid biopsy technologies.
Peter Campbell, Fellow of the Royal Society and former Head of the Cancer, Ageing and Somatic Mutation Programme and Senior Group Leader at the Wellcome Sanger Institute, has played a leading role in cancer genomics. He led the Pan-Cancer Analysis of Whole Genomes (PCAWG) project, which systematically characterized the whole-genome landscapes of more than 2,600 cancers spanning 38 cancer types. He discussed somatic mutations in normal tissues. Whereas sequencing cancer genomes can help reconstruct tumor evolution, sequencing normal cells can reveal cellular lineages and patterns of clonal evolution. Shared mutations between cells can indicate common ancestry, enabling researchers to reconstruct phylogenetic relationships and estimate how mutations accumulate over time.
The findings presented by Campbell demonstrated that somatic mutations accumulate approximately linearly with age across multiple tissues, while tissue architecture is also strongly influenced by selection. Clones carrying driver mutations may expand with age in tissues including the esophagus, lung, endometrium, and blood without necessarily progressing to cancer. Research into chronic liver disease further suggests that disease-associated environments can reshape clonal structures, illustrating the complex relationship among mutation, selection, aging, and disease.
Sijia Lu, Co-founder of Yikon Genomics and a co-developer of MALBAC technology, reviewed the application of single-cell whole-genome amplification to preimplantation genetic testing. His team used MALBAC-based technology in 2014 in a strategy that resulted in the birth of the world’s first “MALBAC baby,” designed to prevent hereditary multiple exostoses. The team has subsequently developed approaches targeting more than 1,200 single-gene disorders and has served more than 400,000 patient families.
Lu also proposed a “seed, soil, and sunlight” framework for reproductive health, integrating single-cell genomics and transcriptomics to support precision embryo selection and the assessment of endometrial function.
专题讨论三 | Panel Discussion 3
The subsequent “Genomic Medicine” panel discussion was moderated by Yanyi Huang, a leading scientist at Changping Laboratory and Professor at Peking University. Dennis Lo, Peter Campbell, Yunlong Cao, and Sijia Lu engaged in an in-depth discussion shifting toward a central translational question: once a genomic innovation leaves the laboratory, what is required for it to become part of clinical practice?
Dennis Lo revisited the winding path from his initial discovery of fetal cfDNA in maternal blood to the ultimate clinical adoption of non-invasive prenatal testing (NIPT). His 1997 breakthrough originated with the detection of Y-chromosome sequences in women carrying male fetuses. Subsequently, his team realized that with sufficiently sensitive analytical tools, fetal DNA did not need to be physically isolated; rather, fetal genetic status could be inferred directly through subtle shifts in chromosomal dosage. The advent of digital PCR and next-generation sequencing (NGS) provided the necessary technological engine. What began as a seemingly simple observation ultimately revolutionized global prenatal screening through decades of rigorous experimentation, refinement, and technological innovation.
Peter Campbell then turned the discussion to cancer genomics, emphasizing that somatic mutation is a fundamental biological process common to all human cells, upon which cancer cells undergo selection akin to Darwinian evolution. Although certain overarching principles govern oncogenesis, every tumor traverses a unique evolutionary trajectory. Consequently, relying on a single driver mutation to guide therapy serves merely as a starting point; the primary challenge lies in deciphering generalizable evolutionary patterns that can be harnessed for precision treatment and prevention. Addressing the debate over whether cancer is predominantly a matter of “bad luck,” Campbell acknowledged the influence of stochastic processes but noted that marked geographical variations in cancer incidence correlate far more strongly with environmental exposures than with genetic background alone. This underscores that a substantial proportion of cancer risk remains preventable, with genomic methodologies—such as mutational signature analysis—offering powerful tools to pinpoint these environmental drivers.
Yunlong Cao, a leading scientist at Changping Laboratory and Professor at Peking University, initially focused his doctoral research on high-sensitivity DNA methylation detection and its clinical utility in oncology. Representing Changping Laboratory's early cancer detection team, Cao evaluated the health economics of multi-cancer early detection (MCED). Citing data from GRAIL's early clinical trials as a case study, he noted that at an estimated cost of US1 million in direct costs. However, because such trials identified only about three treatable, early-stage cancers, the effective cost per actionable case detected reached approximately US$330,000.
To address this challenge, Cao’s team is exploring multifaceted strategies, including reducing sequencing and reagent costs, enhancing sensitivity for early-stage malignancies, optimizing computational algorithms, and refining high-risk population stratification. Together, these approaches could substantially improve the health-economic viability of MCED.
Cao emphasized that evaluating a screening technology should not rely solely on sensitivity and specificity; positive predictive value (PPV) is equally critical. When the underlying prevalence of a disease within a target population is exceptionally low, even a test with high specificity will yield a significant proportion of false positives. From a health-economics standpoint, therefore, combining precise high-risk cohort targeting with drastic cost reductions is essential if genomic screening is to achieve viable, population-scale implementation.
Sijia Lu shared his insights on translating reproductive genetic technologies from bench discovery into large-scale clinical application. Originally trained in physics, Lu transitioned into medical research within the exploratory, interdisciplinary environment of Xiaoliang Sunney Xie’s laboratory. He emphasized that delivering genuine clinical value requires continuous, proactive dialogue with clinicians, domain experts, and regulatory agencies. “Don’t view regulators as stumbling blocks in your development; treat them as partners.”
From diagnosing a single index case to establishing standardized protocols, scaling clinical services, and tailoring assays to diverse global populations, technology translation extends far beyond the laboratory bench. Lu further noted that robust regulatory frameworks establish consistent operational standards, prevent technology misuse, and bolster public trust—ultimately creating the essential conditions for validating, disseminating, and sustaining scientific innovations.
The journey from laboratory discovery to clinical implementation in genomic medicine represents a multifaceted ecosystem bridging discovery, validation, health economics, regulatory oversight, and clinical interpretation. At each junction, a technological innovation can chart a distinct translational trajectory.
Twenty-five years ago, the Human Genome Project set out primarily to answer a singular, fundamental question: What is the sequence? Today, genomic technologies are actively delivering tangible benefits to patients and the public - spanning early cancer detection, oncological diagnosis and targeted therapy, and reproductive carrier screening. Together, these advances are shaping a clinical landscape that was virtually unimaginable at the inception of the Human Genome Project.
Watch the full recording of Panel Discussion 3 to hear leading scientists share their experiences, insights, and vision for the rapid evolution of genomic medicine.