Cancer survival fell and deaths rose during first two years of COVID-19

Patients diagnosed with cancer in the U.S. during the first two years of the COVID-19 pandemic had significantly lower one-year cancer survival rates, leading to an estimated 17,390 more deaths than expected compared with prepandemic trends, according to a new study published Feb. 5, 2026, in JAMA Oncology.

Disruptions to cancer care during the COVID-19 pandemic raised early concerns about missed screenings, delayed diagnoses and interrupted treatment, but much less has been known about how those disruptions affected patient survival. With national cancer registry data now available for patients diagnosed through 2022, researchers are examining whether the pandemic translated into worse short-term outcomes for people recently diagnosed with cancer, according to the study.

Researchers from the University of Kentucky and the Medical University of South Carolina looked at one-year cancer-specific survival among U.S. patients diagnosed in 2020 and 2021 to determine whether pandemic-related disruptions were associated with higher cancer-related mortality compared with prepandemic trends.

This population-based cohort study used data from the SEER-21 registries, which cover about 42% of the U.S. population. Cases found only by death certificate or autopsy, in situ cancers and cases with missing stage or survival time were excluded. Survival was looked at overall and by sex, age, race and ethnicity, urbanicity and stage at diagnosis, with age stratified at 65 years or older.

Cancer-specific survival trends from 2015 to 2019 were modeled to estimate expected survival in 2020 and 2021 and compared with observed outcomes.

Researchers found that during the first two years of the COVID-19 pandemic, more than 1 million people in the U.S. were diagnosed with cancer, and one-year cancer-specific survival declined compared with prepandemic trends. Overall, one-year survival was 84.84% in 2020 and 85.69% in 2021, resulting to an estimated 17,390 more cancer-related deaths than expected within the first year after diagnosis.

Survival fell for both early-stage and late-stage cancers, with larger declines seen among patients diagnosed at later stages. Early-stage cancers had one-year survival rates of about 96% in both years, but outcomes were still significantly lower than expected based on prior trends. Late-stage cancers showed sharper reductions, with one-year survival just under 75% in both 2020 and 2021.

Declines were found across most population subgroups, with the greatest losses affecting adults ages 65 years and older and those of other non-Hispanic race. Site-specific reviews showed a reduction in survival for low-survival cancers such as brain, lung and pancreatic cancer, as well as for high-incidence cancers including colorectal and prostate cancer. These two cancers together accounted for a large share of excess deaths during the pandemic period.

The study’s findings of reduced 1-year cancer survival during the first two years of the COVID-19 pandemic match earlier observations from former NCI Director Norman Sharpless, M.D., who told Managed Healthcare Executive in 2021 that U.S. cancer diagnoses dropped by nearly 50% in the months after the pandemic began.

Sharpless, now managing director of Jupiter Bioventures, a biotechnology company creating new cancer-focused startups, said the drop was caused by delayed care and missed screenings rather than a true decrease in cancer cases. He warned that postponed diagnoses would likely lead to more advanced-stage cancers and higher deaths. He also noted that underserved populations faced barriers to telehealth and routine care, which may have worsened delays.

Based on the new JAMA study’s approach and findings, its main strength is its use of high-quality, population-based cancer registry data, which allows for a national view of short-term cancer survival during the pandemic. Focusing on cancer-specific survival helped limit the influence of COVID-related deaths.

However, the analysis is limited by only 1 year of follow-up, possible differences in disease severity within stage groups, and potential delays or misclassification in cause-of-death reporting. The authors also note the lack of data on prior COVID-19 infection. They suggested further research to clarify these effects and stressed rebuilding screening capacity, improving health system resilience and reducing delays in diagnosis and treatment to prevent similar outcomes in the future.