PO.PS01.02 · 人群科学

美国早发性结直肠癌和胰腺癌:至2040年的发病率和死亡率预测

Early‑onset colorectal and pancreatic cancer in the United States: Incidence and mortality projections through 2040

编号 3575 展板 25 时间 4/20 02:00–05:00 区域 Section 34 主讲 Woo Joo Lee, MD
分会场 Cancer Surveillance: Emerging Cancer Trends and Population Differences
该海报暂无可下载的资料 AACR 官方页面

作者与单位 Authors & Affiliations

Woo Joo Lee1, Sumbal Aziz1, Seon Hye Won2, Muhammad Sohaib Asghar1, Robin Park3, Thomas Shimshak1

1Internal Medicine, AdventHealth Sebring, Sebring, FL,2Department of Family Medicine, Dongguk University Ilsan Hospital, Goyang-si, Korea, Republic of,3Department of Head and Neck-Endocrine Oncology, Moffitt Cancer Center, Tampa, FL

摘要 Abstract

中文摘要
背景:包括美国在内的许多高收入国家,早发性结直肠癌(CRC)和胰腺癌(PC)的发病率一直在上升,但针对年轻成人的中期预测仍然有限。卫生保健系统需要按年龄划分的预测,以便为50岁以下成人规划筛查、预防和生存者服务。 方法:我们使用了1999-2022年美国基于人群的CRC和PC发病率数据以及1999-2023年的死亡率数据,汇总为13个五岁年龄组(20-24岁至80-84岁)。采用泊松似然拟合年龄-时期-队列(APC)模型,使用自然三次样条进行描述性分析,并使用贝叶斯APC(BAPC)框架结合集成嵌套拉普拉斯近似(INLA)和二阶随机游走先验进行至2040年的预测。对于每种癌症和终点(发病率或死亡率),获得按年龄划分的后验均值和95%可信区间(CrI),然后在20-49岁范围内求和,利用美国人口预测得出预测病例数和每100,000人的发病率。 结果:2020年,20-49岁成人CRC的建模发病率为每100,000人12.9例(95% CrI,12.3-13.5),相当于16,854例(16,092-17,616),预计到2030年将增至每100,000人21.6例(10.9-32.3),到2040年增至52.2例(0-162),约为2020年均值估计的四倍,但不确定性极大。该年龄组的PC发病率在2020年为每100,000人2.08例(1.89-2.27),相当于2,707例(2,460-2,953),预计到2030年升至2.83例(2.23-3.43),到2040年升至4.98例(2.02-7.94),约增加2.4倍。相比之下,20-49岁成人的CRC死亡率预计增幅较为温和,从2020年的每100,000人2.81例死亡(2.64-2.97)升至2030年的3.25例(2.45-4.06)和2040年的3.41例(0.51-6.31)。PC死亡率预测高度不确定:均值率从2020年的每100,000人2.29例(2.11-2.47)降至2030年的0.76例(0.33-1.19)和2040年的0.38例(0-1.08),但CrI包含接近零及接近2020年基线的值。 结论:我们的APC建模表明,如果近期趋势持续,美国20-49岁成人的早发性CRC发病率到2040年可能会大幅上升,PC发病率也会有更温和但仍具意义的上升。预测的死亡率趋势在CRC方面不太明显,在PC方面高度不确定,这强调这些估计应被解读为基于情景的预测,而非精确的预报。这些发现支持在年轻成人中加强预防和早期检测策略,同时凸显随着更多数据积累而更新预测的必要性。
查看英文原文 English abstract
Background Incidence of early‑onset colorectal cancer (CRC) and pancreatic cancer (PC) has been rising in many high‑income countries, including the United States, but medium‑term projections for young adults remain limited. Health‑care systems need age‑specific forecasts to plan screening, prevention, and survivorship services for adults younger than 50 years. Methods We used population‑based U.S. incidence data for CRC and PC from 1999-2022 and mortality data from 1999-2023, aggregated into 13 five‑year age groups (20-24 to 80-84 years). Age‑period‑cohort (APC) models were fitted with Poisson likelihoods using natural cubic splines for descriptive analyses and a Bayesian APC (BAPC) framework with integrated nested Laplace approximation (INLA) and second‑order random‑walk priors for projections to 2040. For each cancer and endpoint (incidence or mortality), age‑specific posterior means and 95% credible intervals (CrIs) were obtained and then summed over ages 20-49 years to derive projected case counts and rates per 100,000 persons using U.S. population projections. Results In 2020, the modeled incidence of CRC among adults aged 20-49 years was 12.9 cases per 100,000 persons (95% CrI, 12.3-13.5), corresponding to 16,854 cases (16,092-17,616), and is projected to increase to 21.6 (10.9-32.3) per 100,000 in 2030 and 52.2 (0-162) per 100,000 in 2040, or approximately fourfold higher than the 2020 mean estimate with very wide uncertainty. PC incidence in this age group was 2.08 (1.89-2.27) per 100,000 in 2020, corresponding to 2,707 cases (2,460-2,953), and is projected to rise to 2.83 (2.23-3.43) in 2030 and 4.98 (2.02-7.94) in 2040, roughly a 2.4‑fold increase. In contrast, CRC mortality among adults aged 20-49 years is projected to increase more modestly, from 2.81 (2.64-2.97) deaths per 100,000 in 2020 to 3.25 (2.45-4.06) in 2030 and 3.41 (0.51-6.31) in 2040. PC mortality projections are highly uncertain: the mean rate decreases from 2.29 (2.11-2.47) per 100,000 in 2020 to 0.76 (0.33-1.19) in 2030 and 0.38 (0-1.08) in 2040, but CrIs include near‑zero and values close to the 2020 baseline. Conclusions Our APC modeling suggests that, if recent trends persist, early‑onset CRC incidence in U.S. adults aged 20-49 years might increase substantially through 2040, with more moderate but still meaningful increases in PC incidence. Projected mortality trends are less pronounced for CRC and highly uncertain for PC, underscoring that these estimates should be interpreted as scenario‑based projections rather than precise forecasts. These findings support intensified prevention and early‑detection strategies in younger adults while highlighting the need to update projections as additional data accrue.
利益披露 Disclosure
W. Lee, None.. S. Aziz, None.. S. Won, None.. M. Asghar, None.. R. Park, None.. T. Shimshak, None.

← 返回 AACR 2026 检索