Precision public health perspective: Is genomic population screening a viable option for the early detection of risks for oncological diseases? Reflections from Serbia

This paper aims to discuss the term precision public health and its application in the early detection of oncological diseases and genomic risks for these diseases. The use of artificial intelligence and genomic profiling are briefly elaborated on, including reflections on the current status of innovative approaches in the Republic of Serbia. An environment for big genetic and health data sets storage that is protected and interoperable needs to be created, in keeping with the findability, accessibility, interoperability, and reusability (FAIR) principles. Wider use of genomics in the early detection of risks for oncological diseases must meet the well-established criteria for population screening, which is currently not the case. The risk factors that can trigger the penetrance of genes, thus leading to oncological disease, are also well known, and are as follows: unhealthy environments including air pollution, smoking and exposure to second-hand smoke, stress, unbalanced diet, and other harmful factors. Therefore, health governance bodies and health policy decision-makers should focus on building a healthy environment as much as on working on the technical and technological health information infrastructure necessary for further development of genomics and precision public health.

MALIGNANT DISEASES IN SERBIA

Malignant diseases are a significant public health problem in the Republic of Serbia. The age-standardized incidence rate (ASIR) is 250.4 per 100,000 population [1], while the mortality rate for all cancer types, in 2022, was 128.4 per 100,000 population, which puts Serbia in fourth place in Europe [2]. Several risk factors contribute to these unfavorable statistics, such as a high prevalence of smoking among adults (36.6%) [3], high levels of air pollution during the entire year (an annual average PM2.5 concentration of 19.4 μg/m³, as compared to 14 μg/m³ in the European Region) [4], a national diet rich in fat and processed foods and depleted in fiber [5], but also an aging population (in 2022, the share of persons aged 65 and over was 22.0%) [6], in combination with a low total fertility rate (1.63 in 2022) [7].

The threat of a high incidence of malignant diseases and mortality caused by these diseases was observed as early as 20 years ago, and national policymakers announced programs for the prevention and early detection of cancers as the strategy which was supposed to improve the length and quality of life. Consequently, population screening programs were launched for three types of cancers in 2012: breast cancer (mammography followed by biopsy), cervical cancer (Papa test followed by biopsy), and colorectal test (fecal occult blood test - FOBT, followed by colonoscopy and biopsy) [8],[9],[10]. Although the National Cancer Screening Office was established to coordinate a proactive approach in inviting the population believed to be healthy (ostensibly healthy) to undergo screening, the response rates remained modest, far below the desired goals. Despite all efforts and the implementation of public health campaigns to raise awareness of the importance of early detection of cancers, according to Serbia: Health System Review 2019, the response rates were 11.4%, 15.8%, and 5.0% of the target population, for breast, cervical and colorectal cancer, respectively, which was far below the reported response rates for other European countries [11],[12],[13].

As public health professionals with a comprehensive societal perspective, we must consider whether the innovative approaches available in the 21^st century can meaningfully be used in the early detection of oncological diseases and the assessment of risks within ostensibly healthy populations. Furthermore, we should examine whether these approaches can reduce mortality rates and enhance survival and health-related quality of life.

INNOVATIVE APPROACHES

We are living in the era of the fourth industrial revolution, and the availability of innovative technologies, artificial intelligence (AI), and genome profiling are making a revolutionary difference in all aspects of health care. Therefore, there is a valid and justified question: can these innovative approaches improve the early detection and treatment of oncological diseases? While we leave the aspect of oncological diseases related to innovative diagnostic methods and therapies to the consideration of clinicians, pathologists, and molecular biology specialists, let us consider the first part, i.e., the early detection of oncological diseases. There are ongoing efforts and projects related to training AI models to be used in radiology for the valid detection of suspicious changes/lesions, particularly for reading mammography reports [14],[15],[16],[17]. Several studies have proven that using AI in the first reading of radiology reports can be efficient and cost-effective. It can overcome an important barrier, such as the lack of highly qualified radiologists, known as one of the biggest bottlenecks in the healthcare systems worldwide. This is especially the case in developing countries, due to brain drain and because of the extensive training that doctors have to undergo before becoming specialists [18]. However, although AI has shown promise in mammography-based screening in retrospective evaluations, few prospective studies exist, which is why the scope of its application is still limited [19]. It is expected that in the future, AI might contribute to the early detection of clinically relevant cancers thus reducing the screen-reading workload without increasing false positives [20].

Apart from AI, the other innovative approach we would like to consider is the use of genomics. Thanks to the advanced techniques for genome profiling, one of which is known as Next Generation Sequencing (NGS), genomic analyses of diseased tissues are nowadays an almost inevitable part of tumor profiling, which leads to prescribing the most effective personalized treatments that target biomolecules responsible for cancer development [21]. However, the availability of NGS and innovative therapies depends on the financial resources of the healthcare system, and there are different models for securing funding for these approaches [22],[23].

The aforementioned innovative approaches are not unknown in Serbia. At the national level, thirteen organizations equipped to perform NGS have been mapped. Their sequencing systems are from different manufacturers (Illumina, BGI, Oxford Nanopore Technologies, Thermo Fisher Scientific, PacBio). All of these organizations are publicly owned [24]. The availability of innovative therapies is also increasing. According to recent announcements, the National Health Insurance Fund of the Republic of Serbia is financing 17 innovative therapies for oncological diseases, and the overall expenditure for them was 18.5 billion dinars in 2024, which is almost 157 million euros [25].

PRECISION MEDICINE AND PRECISION PUBLIC HEALTH

In the past 20 years, due to the availability of NGS, which enables significantly faster and cheaper analyses of the human genome, the concept of precision medicine (generally considered analogous to personalized medicine) evolved to include genetic profiling in clinical decision-making. Although there is no universally accepted definition of personalized medicine, the Horizon 2020 Advisory Group defines it as “a medical model using characterization of individuals’ phenotypes and genotypes (e.g. molecular profiling, medical imaging, lifestyle data) for tailoring the right therapeutic strategy for the right person at the right time, and/or to determine the predisposition to disease and/or to deliver timely and targeted prevention“. The “one size fits all” approach to cancer treatment was gradually replaced by tailoring treatments to each individual, which has placed new challenges before clinicians and healthcare systems. However, successful precision medicine requires a population perspective, paving the way to establishing the precision public health (PPH) concept [21]. Precision public health can be defined as “a means of improving population health through the use of new technologies, particularly genomics and digital, which would guide public health practice by generating more individually tailored interventions and policies” [26]. In other words, PPH is about delivering “the right intervention, at the right time, to the right population” [27], or using the best available data to more effectively and efficiently target those in need [28]. The two biggest drivers in PPH are genomic research and health information technologies that collect health-related data in a machine-readable form, owing to the interoperability of health information systems. However, specialists in health systems, decision-makers, and public health professionals need to understand innovative, digital, and “omics” technologies (the latter is the suffix that applies to genomics, epigenomics, proteomics, metabolomics, and metagenomics). They are not the goal in themselves, but tools for health promotion and disease prevention, if applied reasonably and responsibly, i.e., if followed by appropriate risk management strategies such as counseling and surveillance. Still, many methodological considerations and concerns remain to be addressed in order to justify their use on a large scale, which is particularly relevant in their use for population genomic screening to identify risk factors for malignant diseases [26],[29].

GENOMICS IN THE WORLD AND SERBIA

Numerous projects and ongoing initiatives related to genomic profiling in patients and the general population are nowadays being applied worldwide, mostly in developed but also in developing countries [21],[30]. For example, Genomics England launched an initiative called “100,000 Genomes Project” aiming to sequence 100,000 genomes from around 85,000 patients registered with their National Health Service (NHS), who are suffering from either cancer or a rare disease, while the European Union announced ”1+ Million Genomes” (1+MG), an initiative aiming to secure access to genomics and the corresponding clinical data across Europe [31],[32]. The “Genome of Europe” project aims to build a European network of national genomic reference cohorts to collect data on the genetic composition of both healthy and diseased individuals (sequenced genomes) [33].

In Serbia, various efforts have been implemented to use the power of genetic data for improvements in the well-being of citizens, but they are mostly fragmented and represent unrelated institutional activities [34]. Since November 2023, as a more focused national effort, Serbia has also become a part of the “Genome of Europe” initiative/consortium, together with 29 European Union (EU) countries and only three countries outside the EU (besides Serbia, also Turkey and Norway) [33],[35]. These findings are expected to lead to groundbreaking insights into genetic structure, enabling “personalized healthcare treatments with the potential to improve disease prevention” [32].

Capacities for genomic analyses have been mapped in Serbia, and it seems that the country has reasonable potential and willingness to keep up with the global trends in this domain [24],[35]. However, certain challenges remain, particularly the lack of qualified research and medical staff for analyzing and interpreting sequencing results. To address this issue, in 2025, the Ministry of Education launched the first Bioinformatics Master’s Program to train the next generation of students in bioinformatics, genomics, molecular biology, clinical data analysis, pharmacogenetics, the application of artificial intelligence in biology and medicine, as well as in big data analysis [36]. At the national level, a study on voluntary genomic profiling among 1,000 individuals 18-40 years old, from the general (healthy) population, is being conceptualized [37]. This needs to be followed by the project of establishing a Genetic and Biomedical Data Registry, i.e., an information system for storing and exchanging genetic data under the highest technical and security standards, for research and development purposes in biotechnology, healthcare, and genetics [37]. These activities represent a great incentive and opportunity, but also a great responsibility to deliver the most cost-effective interventions to persons with identified genomic risks.

THE POTENTIAL OF GENOMICS IN POPULATION SCREENING ASSOCIATED WITH MALIGNANT DISEASES

According to the American College of Medical Genetics and Genomics (ACMG) and the United States Centers for Disease Control and Prevention (CDC) Tier 1 conditions, the most common and recommended application of population genetic screening is for the identification of genetic mutations that predispose an individual to one of the following conditions: hereditary breast and ovarian cancer syndrome (HBOC), which presents a risk for breast, ovarian, tubal, peritoneal and other cancers due to mutations in breast cancer gene 1 (BRCA1) or breast cancer gene 2 (BRCA2), and Lynch syndrome (LS), which represents an increased risk for colorectal, endometrial, ovarian and other cancers [30],[38].

Mutations of the BRCA1 and BRCA2 genes are known for having relatively high penetrance: the cumulative risk for breast cancer among mutation carriers at the age of 70 was 57% for BRCA1 (95% confidence interval (CI) 47%-66%) and 49% for BRCA2 (95% CI 40%- 57%), and even lower for ovarian cancer (40% and 18% for BRCA1 and BRCA2 respectively, 95% CI 35%-46% and 95% CI 13%-23%), as found in a meta-analysis conducted by Chen and Parmigiani [39]. On the other hand, the prevalence of BRCA mutations among patients with breast cancers is in the range of 1.8% in Spain to 36.9% in the United States, according to the systematic review of 70 studies worldwide conducted by Armstrong et al. [40]. In Nordic countries, the heritability of female breast cancer was found to be 31%, while for colorectal cancer it was found to be 15% [41]. In Serbia, testing for germline mutations in cancer susceptibility genes for hereditary breast and ovarian cancer and Lynch syndrome is performed at the national level, at the Institute for Oncology and Radiology of Serbia. The most frequent mutations in breast and ovarian cancer were found to be in BRCA1 (7.6%), BRCA2 (4.8%), the partner and localizer of BRCA2 gene (PALB2) (4.1%), and the checkpoint kinase 2 gene (CHEK2) (3.8%) [42]. Identifying pathogenic and likely pathogenic variants in genes that carry a high predisposition for hereditary cancers allows for the stratification of families by risk and has become a widely accepted global practice.

It is important to note that if an individual is a carrier of the aforementioned genetic mutations, this does not necessarily mean that this person is unhealthy or will be unhealthy. This is, obviously, a matter of probability which can be modified. It is well known that disease appears as an interaction between genetic and environmental risk factors. These risk factors include exposure to stress, unfavorable environmental conditions, such as air pollution, and unhealthy lifestyles (e.g., smoking, lack of physical activity, and unhealthy nutrition), but how exactly the disease develops and what and when triggers the penetrance of gene mutations, or the appearance of new ones and the expression of disease, remains unclear. More information is needed from longitudinal studies on lifetime exposure to risk factors, in addition to genetics, to build knowledge on the natural course of disease. Digital health information might help in individual data collection, while AI is useful in the analysis of big data and identification of patterns and associations between individual lifetime experiences, exposures, and health outcomes. However, when oncological disease eventually happens, each tumor has its own genetics, which can be proven by NGS and genomic analyses, and innovative therapies are available to address the genetic mutations as the root cause of malignant disease, which is the foundation of precision medicine.

SOCIAL, LEGAL, AND ETHICAL CONSIDERATIONS

Bearing all the previously described facts in mind, we cannot but ask ourselves: what are the benefits or the added value of knowing our genetic status when we are healthy? How does it affect our chances of getting a particular oncological disease, our survival, or our health-related quality of life? Currently, the answer is lacking, as is the knowledge related to the natural course of disease [43]. Would the information on genetic status increase an individual’s compliance with regular checkups and improve healthy lifestyle choices? These and many other questions need to be carefully considered, and the answers to these questions are by no means uniform [44]. Although there is no doubt that the use of genomic data holds significant potential for precision medicine, special considerations are necessary in order to use this information ethically, responsibly, safely, and beneficially, without potential harm or misuse [45].

Potentially, the information that a person is a carrier of a gene mutation might have some advantages, such as increased motivation for regular monitoring of their health status and refraining from being exposed to behavioral risk factors that are under individual control. At the same time, this information could cause fear and anxiety. Furthermore, risk factors originating from the environment are public responsibility, and governments and societies have to ensure that appropriate and recommended policies related to building a healthy environment are launched and applied. The best negative example of this in the Republic of Serbia is the tobacco control strategy that was signed and ratified almost 20 years ago, followed by a set of national legislation that envisaged creating smoking-free public spaces, among other initiatives. However, nowadays, indoor smoking (and therefore passive exposure) in Serbia is still not banned in many restaurants and cafes, which means that people are greatly exposed to second-hand smoke. This is a direct risk factor for all oncological diseases, primarily lung cancer, whose ASIR was 40.4 per 100,000 population in 2022, which put Serbia in fourth place globally, preceded by Hungary, China, and New Caledonia [46].

Being aware of the fact that one is a carrier of a gene mutation with a relatively high penetrance potential to develop oncological disease may have a considerable negative impact on one’s psychological well-being, especially if not communicated properly. The effect it has on family members of pathogenic mutation carriers is also an aspect to be considered. Good practice proposes counseling before and after genomic testing and adequate disease risk management [47]. Without these resources in place, it would be irresponsible to leave individuals to deal with genomic testing themselves, in any setting. In more developed environments, there are some ideas and projects related to promoting genomic testing in primary healthcare services [48],[49]. To be successful and do no harm, these projects have to be followed by complex infrastructure that enables preventive interventions and high-quality communication between providers and individuals/ patients, assuring that advantages are well understood and strategies for risk management are in place after receiving test results.

Criteria for any kind of population screening were established in 1968, and genomic testing for detecting genes associated with malignant diseases in healthy populations is no exception [50],[51]. Criteria are related to the availability of an appropriate screening test, which has to be sufficiently sensitive and specific, available at a low cost, easy to apply, acceptable, non-invasive, safe, comfortable, and cost-effective. The majority of these criteria could be met by NGS as a screening tool, with cost-effectiveness still being debated, even in simulated cohorts with modeled expected outcomes, such as incidence and quality-adjusted life years (QALY) after survival [52],[53],[54].

The other criterion for population screening is the availability of effective interventions that make a difference in survival and QALY. Basically, in the case of “omics” markers of risks for oncological disease, these interventions are just traditional public health promotion and prevention interventions, which have proven to be “best-buy” strategies. These are tobacco control measures, efforts related to the decrease of air pollution, and promotion of healthy lifestyles, but also programs for early detection of clinical diseases by traditional screening instruments with sufficient levels of sensitivity (ability to correctly identify cases of disease) and specificity (ability to correctly identify individuals who are disease free, i.e. true negative cases). This brings us back to the beginning, i.e., sending the message that without “old fashioned” individual and population measures, optimized and in place, we cannot expect robust evidence that early detection of genomic risk factors can improve survival and QALY in cases of known actionable genetic conditions such HBOC, and Lynch syndrome. Even when gene panel screenings for these conditions are performed as a personal choice, typically by those who can afford or are highly motivated to pursue genomic testing, the same individual and public health promotion and prevention measures should be available and accessible. Public health measures affect everyone and should already be in place (such as creating tobacco-free environments, for example), while individual health promotion measures (healthy lifestyles, for example) depend on the income and socio-economic status of an individual. Therefore, the overall response to the results of genetic testing might increase health inequalities [55].

POLICY IMPLICATIONS

Public health measures, health promotion, and prevention related to well-known risk factors for oncological diseases did not lose their importance, and innovations such as genomics and artificial intelligence did not decrease their significance either [56]. Health governance bodies and health policy decision-makers should focus on building a healthy environment as much as on working on the technical and technological health information infrastructure necessary for further development of genomics and precision public health [57],[58]. An environment for big genetic and health data sets storage that is protected and interoperable according to FAIR (findability, accessibility, interoperability, and reusability) principles needs to be created [59]. Leveraging biomedical research and development in Serbia is associated with creating a national register of genomics and health data, as a great national step forward towards the health promotion of our citizens, through precision medicine and precision public health.

Funding statement: This study was supported by the MSD Global Oncology Policy Grant; the Ministry of Science, Technological Development and Innovation of the Republic of Serbia Agreements No. 451- 03-66/2024-03/200110 and No. 451-03-136/2025- 03/200043, and the Horizon Europe STEPUPIORS Project (HORIZON-WIDERA-2021-ACCESS-03, European Commission, Agreement No. 101079217). The funders had no role in the study design, data collection and analysis, the decision to publish, or the preparation of the manuscript.