Cancer Surveillance Research (CSR) is an observational science of cancer data ascertained in population-based cohorts, notably, cancer registries. CSR is used to: track cancer incidence, mortality, burden, prevalence, and survivorship; quantify cancer differences; inform cancer control programs; test hypotheses; uncover etiologic clues; gauge effectiveness of screening and therapy; and characterize cancer’s natural history and its evolution over time. To date, most CSR studies have relied on a limited set of non-parametric statistical tools applied to relatively small datasets. Currently, attempts to scale such studies are surprisingly labor intensive and require specialized expertise. Advances in biostatistics and data science have the potential to usher in a ‘golden age’ where more data and better data are universally accessible, contemporary methods from biostatistics and data science are rapidly and freely deployable, and current knowledge more readily mined to inform both cancer control and cancer research.
The aim of this research topic is to show how to get there – the golden age – from the here and now. There are three pieces to the equation for advancement: data, statistics, and hypotheses.
1. How can information technology and data science facilitate access to and integration of public use cancer data in a format suitable for novel studies of cancer incidence, mortality, and survivorship? How can associated demographic and risk factors data be integrated with cancer surveillance data into the workflows?
2. Can advances in non-parametric statistics and mathematical models, especially, age-period-cohort models, provide a unified framework for descriptive and hypothesis-based analysis of cancer trends and patterns?
3. How can information technology and data science help researchers interrogate, synthesize, and statistically analyze cancer data to inform etiological hypotheses and cancer control programs?
As well, we aim to highlight why this effort is important:
4. What are the big questions in CSR? Which current trends and patterns are most concerning? How is cancer evolving within and between countries and populations stratified by factors including human development indices, race/ethnicity/ancestry, sex, age, location, and risk factor profiles?
1. Cancer epidemiology, data science, and integrative studies in CSR
a. From APIs to analytical files
b. Record linkage: tools and resources
c. Web-based analysis tools
d. AI prompt engineering and web library applied to CSR
2. Non-parametric statistics and mathematical models in CSR
a. Spatial and temporal analysis of cancer outcomes and risk factors
b. Scalable alternatives to the Join Point method
c. Comparative methods and hypothesis testing
d. Advances in age-period-cohort analysis
e. Forecasting models
3. Knowledge Mining in CSR
a. Generative languages: examples, promises and pitfalls
b. Advances in meta-analysis
4. Novel methods in surveys of current and future cancer trends
a. Slowing or growing? Cancer incidence, mortality, and survival within and between populations worldwide
b. The many faces of breast neoplasms: trends by subtype and behavior
c. Racial and ethnic cancer disparities
d. Epidemiology of early-onset cancers
Original contributions, reviews, reviews, meta-analyses, and tutorials are welcome.
Please note: manuscripts consisting solely of bioinformatics, computational analysis, or predictions which are not accompanied by illustrative real-world examples are not suitable for publication in this special issue. Ideally, all data should be freely available.
Cancer Surveillance Research (CSR) is an observational science of cancer data ascertained in population-based cohorts, notably, cancer registries. CSR is used to: track cancer incidence, mortality, burden, prevalence, and survivorship; quantify cancer differences; inform cancer control programs; test hypotheses; uncover etiologic clues; gauge effectiveness of screening and therapy; and characterize cancer’s natural history and its evolution over time. To date, most CSR studies have relied on a limited set of non-parametric statistical tools applied to relatively small datasets. Currently, attempts to scale such studies are surprisingly labor intensive and require specialized expertise. Advances in biostatistics and data science have the potential to usher in a ‘golden age’ where more data and better data are universally accessible, contemporary methods from biostatistics and data science are rapidly and freely deployable, and current knowledge more readily mined to inform both cancer control and cancer research.
The aim of this research topic is to show how to get there – the golden age – from the here and now. There are three pieces to the equation for advancement: data, statistics, and hypotheses.
1. How can information technology and data science facilitate access to and integration of public use cancer data in a format suitable for novel studies of cancer incidence, mortality, and survivorship? How can associated demographic and risk factors data be integrated with cancer surveillance data into the workflows?
2. Can advances in non-parametric statistics and mathematical models, especially, age-period-cohort models, provide a unified framework for descriptive and hypothesis-based analysis of cancer trends and patterns?
3. How can information technology and data science help researchers interrogate, synthesize, and statistically analyze cancer data to inform etiological hypotheses and cancer control programs?
As well, we aim to highlight why this effort is important:
4. What are the big questions in CSR? Which current trends and patterns are most concerning? How is cancer evolving within and between countries and populations stratified by factors including human development indices, race/ethnicity/ancestry, sex, age, location, and risk factor profiles?
1. Cancer epidemiology, data science, and integrative studies in CSR
a. From APIs to analytical files
b. Record linkage: tools and resources
c. Web-based analysis tools
d. AI prompt engineering and web library applied to CSR
2. Non-parametric statistics and mathematical models in CSR
a. Spatial and temporal analysis of cancer outcomes and risk factors
b. Scalable alternatives to the Join Point method
c. Comparative methods and hypothesis testing
d. Advances in age-period-cohort analysis
e. Forecasting models
3. Knowledge Mining in CSR
a. Generative languages: examples, promises and pitfalls
b. Advances in meta-analysis
4. Novel methods in surveys of current and future cancer trends
a. Slowing or growing? Cancer incidence, mortality, and survival within and between populations worldwide
b. The many faces of breast neoplasms: trends by subtype and behavior
c. Racial and ethnic cancer disparities
d. Epidemiology of early-onset cancers
Original contributions, reviews, reviews, meta-analyses, and tutorials are welcome.
Please note: manuscripts consisting solely of bioinformatics, computational analysis, or predictions which are not accompanied by illustrative real-world examples are not suitable for publication in this special issue. Ideally, all data should be freely available.