Survival of Patients Diagnosed With Cancer During the COVID-19 Pandemic
Menée à partir de données des registres américains des cancers portant sur 1 008 012 patients atteints d'un cancer diagnostiqué pendant la pandémie de la COVID-19 (2020-2021), cette étude analyse la survie spécifique à 1 an
Importance : The effects of COVID-19 pandemic–related disruptions on cancer diagnosis in the US have been widely observed, but their impact on short-term survival has not been assessed.
Objective : To examine 1-year cause-specific survival (CSS) rates among patients diagnosed with cancer during 2020 and 2021 using high-quality cancer registry data.
Design, Setting, and Participants : This population-based cohort study of cancer survival between 2015 and 2021 used the Surveillance, Epidemiology, and End Results 21 Registries (SEER-21) database. The population included individuals with an invasive cancer diagnosis and complete follow-up reported to registries included in SEER-21 between January 1, 2015, and December 31, 2021. Data were analyzed from May 13 to May 27, 2025.
Exposure : COVID-19 pandemic.
Main Outcomes and Measures : The primary outcome was 1-year CSS rates by stage at diagnosis for patients diagnosed with cancer in 2020 and 2021 compared with trends in 1-year CSS rates among patients diagnosed between 2015 and 2019. Additional site-specific analyses were performed on common cancer sites identified as having low survival (5-year relative survival <33%) or high incidence and high survival (incidence >20.0 per 100 000 and 5-year relative survival ≥66%).
Results : A total of 1 008 012 individuals were diagnosed with cancer during the first 2 years of the COVID-19 pandemic, including 473 781 in 2020 (50.1% female; 51.0% diagnosed at ≥65 years; 11.5% Black, 15.8% Hispanic, 64.1% White, and 7.4% other race) and 534 231 in 2021 (50.3% female; 51.9% diagnosed at ≥65 years; 51.9% diagnosed at ≥65 years; 11.7% Black, 16.2% Hispanic, 62.9% White, and 7.9% other race). Compared with prepandemic trends, significant absolute reductions in 1-year CSS rates occurred for early-stage diagnoses in 2020 (−0.44 [95% CI, −0.54 to −0.34] percentage points) and 2021 (−0.27 [95% CI, −0.37 to −0.16] percentage points) and late-stage diagnoses in 2020 (−1.34 [95% CI, −1.75 to −0.93] percentage points) and 2021 (−1.20 [95% CI, −1.69 to −0.71] percentage points). Survival reductions resulted in an estimated 17 390 more cancer-related deaths (13.1%) within 1 year of diagnosis than expected during the first 2 years of the COVID-19 pandemic. Absolute survival reductions greater than 1.00 percentage point occurred in both years for late-stage diagnoses among individuals of other non-Hispanic race and ethnicity (ie, American Indian and Alaska Native, Asian or Pacific Islander, or unknown race) and individuals aged 65 years or older. Significant site-specific survival reductions also existed in both 2020 and 2021, respectively, for early-stage diagnoses of esophageal cancer (−3.89 and −3.67 percentage points) and colorectal cancer (−1.08 and −0.78 percentage points) and late-stage diagnoses of prostate cancer (−0.64 and −0.77 percentage points).
Conclusions and Relevance : This cohort study found that individuals diagnosed with cancer in 2020 and 2021 experienced worse short-term survival than those diagnosed between 2015 and 2019, suggesting substantial harms associated with cancer care disruptions during the first 2 years of the COVID-19 pandemic.
JAMA Oncology , article en libre accès, 2026