Leukemia in Canada: Understanding the Numbers

The current projected figures

The Canadian Cancer Society estimates 6,500 new leukemia diagnoses and 3,200 deaths from leukemia in Canada in 2026. These are projected annual figures, not a count of everyone living with the disease. The source page was medically reviewed in April 2026. ^[1]

Incidence, mortality and prevalence

Incidence counts new diagnoses during a defined period. Mortality counts deaths. Prevalence concerns people alive who have previously received a diagnosis, and depends on the time window and definition used. These measures cannot be substituted for one another. ^[1][2]

An annual estimate cannot establish how many leukemia survivors live in Canada. For that question, use a specifically identified prevalence dataset with its reference date and inclusion criteria.

Read leukemia as a group of diseases

Leukemia includes different diseases whose course and patient populations differ. Acute and chronic forms should not be compressed into a single clinical story. The National Cancer Institute's overview identifies leukemia as a broad group and explains that classification depends partly on the affected cell type and rate of progression. ^[3]

For communication, this means a national headline needs context. It cannot tell an individual their prognosis, which treatment they need or whether a symptom represents leukemia.

Use statistics carefully in research and planning

Our recommendation is to carry four items with every quoted figure: the measure, geography, year and source. When comparing provinces or time periods, also check whether figures are counts or rates and whether methods are comparable.

Population burden can motivate research, but it does not demonstrate that a particular AI product improves survival or reduces diagnostic delays. Such statements need separate product-specific evidence.

CellSight uses this distinction in communicating about blood-cell image research: explain the public-health context clearly, then evaluate technology on its own evidence.