SURVMARK-2 aims to establish the most up-to-date international cancer survival benchmarks for cancers of the lung, oesophagus, stomach, ovary, colon, rectum, liver, and pancreas across six comparable high-income countries, as well as to provide a greater understanding of the impact of stage at diagnosis and stage-specific survival. This work will be carried out in three Elements (A, B, and C).For each jurisdiction and for each cancer, SURVMARK-2 will:
A. Provide up-to-date measures of cancer survival, incidence, and mortality.
B. Provide an overview of jurisdiction adherence to international coding frameworks, including staging classifications, to enable robust comparisons and to recommend how these could be harmonized.
C. Quantify the impact of local registration practices on short-term cancer survival, to improve the sensitivity of international comparisons.
Data from all participating jurisdictions will be collected following a standardized protocol. Comprehensive quality checks and conversions wherever necessary will be carried out in order to ensure data quality and comparability. All analyses will be performed centrally at IARC.
The overarching aim of the Element A work on epidemiological benchmarking is to unpick the reasons for survival differences between countries in order to eliminate survival disparities in the near future. This will be done by developing a comprehensive and quality-assured set of country-specific indicators for benchmarking across countries.
Additionally, Element A will provide report cards for the most recent cancer survival estimates using multiple indicators and presentational approaches, including synergetic analyses of incidence and mortality by country, period, sex, age, cancer site and stage. The focus of this element is to highlight and elucidate survival variations between-jurisdiction.
Element A will incorporate standard survival measures as well as those that provide insight into the lives of cancer patients beyond their initial diagnosis (conditional survival or loss in life expectancy) and reasons for observed improvements in survival (better therapy or palliative care versus specific biases or artifacts) based on changes in proportion cured and the median survival time of those who die.The Element A work will be carried out by the IARC team.
During the course of the work performed in Element B, an assessment of adherence to international standards of coding and classification will be carried out. The impact of differentials in such practices on explaining survival differences will be determined, and subsequently, international guidelines to ensure robust data for international survival benchmarking purposes will be developed.
The focus of Element B is on optimizing cross-country comparability while quantifying the impact how survival differences are modified by inaccurate, partial or missing cancer-specific information. A full understanding in differences in coding, classification and local registration practice (Elements B and C) will be obtained through surveys, discussion with key informants and in-depth sample studies in each jurisdiction.
In addition, a special Staging Advisory Board will be assembled to provide initial and final recommendations for collection of summary stage information that is feasible across jurisdictions.The Element B work will be carried out by the London team.
During the course of Element C, an assessment of local registration practices including quantification of the impact of specific factors in accentuating/attenuating short-term survival differences will be performed. Several methods adjusting for these differences in the analysis phase will be explored.
The focus of Element C lies in optimizing cross-country comparability in registry practices while quantifying how survival differences are modified by differing registry practices. Structured questionnaires and discussions with key informants will be carried out as base for jurisdiction-specific reports, followed by simulation-based methods to assess the impact of various cancer registration errors by cancer types and country. Adjustment factors, modeled against various determinants e.g. site, age, will be tested.The Element B work will be carried out by the Leicester/Oslo team.