Cancer Prevalence Projections in Japan and Decomposition Analysis of Changes in Cancer Burden, 2020-2050: A Statistical Modeling Study

Abstract

Background: We provide comprehensive sex-stratified projections of cancer prevalence for 22 cancer sites in Japan from 2020 to 2050.

Methods: Using a scenario-based approach, we projected cancer prevalence by combining projected incidence cases and survival probabilities. Age-specific incidences were forecasted using age-period-cohort models, while survival rates were estimated using a period-analysis approach and multiple parametric survival models. To understand changes in cancer prevalence, decomposition analysis was conducted, assessing the contributions of incidence, survival, and population demographics.

Results: By 2050, cancer prevalence in Japan is projected to reach 3,665,900 (3,210,200 to 4,201,400) thousand cases, representing a 13.1% increase from 2020. This rise is primarily due to a significant increase in female survivors (+27.6%) compared with a modest increase in males (+0.8%), resulting in females overtaking males in prevalence counts from 2040 onward. In 2050, the projected most prevalent cancer sites in Japan include colorectal, female breast, prostate, lung, and stomach cancers, accounting for 66.4% of all survivors. Among males, the highest absolute increases in prevalence are projected for prostate, lung, and malignant lymphoma cancers, while among females, the highest absolute increases are expected for female breast, colorectal, and corpus uteri cancers.

Conclusions: These findings emphasize the evolving cancer prevalence, influenced by aging populations, changes in cancer incidence rates, and improved survival. Effective prevention, detection, and treatment strategies are crucial to address the growing cancer burden.

Impact: This study contributes to comprehensive cancer control strategies and ensures sufficient support for cancer survivors in Japan.

Figure 1.

All-site combined 5-year cancer projections by sex, 2020–2050. A, Scenario-specific projections with sensitivity range of prevalence count; (*) the main projection scenario with the dynamic population demographics, incidence rates, and survival rates; the red dashed lines signify the projected peak of prevalence counts (horizontal line) and the corresponding year of the peak (vertical line). B, Age distribution of cancer survivor. C, Decomposed absolute changes over 2020–2050 in prevalence count; Absolute changes in a specific year X is calculated as Prevalence in year X – Prevalence in reference year. Prevalence in year 2020 is used as reference year.

Figure 2.

Site-specific 5-year prevalence projections in 2020 and 2050, by sex. The plot shows the projections for both-sex (A), male (B), and female (C). † represents prevalence count in thousand cases; ‡ represents percent in all cancers (e.g., sum up to 100%); The y-axis title is ordered based on sex-specific prevalence count in year 2020, with the highest at the top and the lowest at the bottom.

Figure 3.

Age distribution of 5-year cancer survivors in 2020 and 2050 by cancer site and sex. The plot presents both-sex (A), male (B), and female (C). The y-axis title is ordered based on sex-specific prevalence count in year 2020, with the highest at the top and the lowest at the bottom.

Figure 4.

Site-specific decomposed absolute and relative changes of 5-year prevalence count over 2020–2050, by sex. The plot shows the absolute changes for both-sex (A), male (B), and female (C), and relative changes for both-sex (D), male (E), and female (F). Absolute changes over 2020–2050 is calculated as Prevalence 2050 – Prevalence 2020; Relative changes over 2020–2050 is calculated as (Prevalence 2050 – Prevalence 2020) / Prevalence 2020; The y-axis title is ordered based on sex-specific total absolute/relative changes in prevalence count over 2020–2050, with the highest at the top and the lowest at the bottom.