← Back
Understanding leukemia

How to understand the statistics

After a diagnosis, many people go straight online to look for numbers: “What percentage survive?” Those numbers can bring hope, and they can also frighten. It matters to read them correctly. This topic calmly explains what leukemia statistics measure and what they do not, why the numbers differ from one source to another, and how to relate them to your own life.

Last updated: ~7 min read
A calm crowd of hundreds of small translucent human silhouettes on a dark navy background; most of the silhouettes glow with a soft warm light, with a faint DNA helix and blood cells drifting between them, and no numbers or text.
Figure 1.Concept: statistics combine many people’s stories into one number, but each light is a separate person.

01Short answer

Statistics are data collected from large groups of people, and they cannot predict exactly what will happen to any one person. The most widely used measure is 5-year survival — the percentage of people alive 5 years after diagnosis; it does not mean “living for 5 years”, because many people live much longer. Relative survival compares people with leukemia to the general population of the same age and sex. In the US (SEER, 2016–2022), 5-year relative survival for all leukemias combined is 68.6%, but the differences between types are large: 90.2% for CLL, 73.2% for ALL, 71.1% for CML, 33.4% for AML and 86.7% for childhood leukemia. These figures refer to people diagnosed several years ago and do not fully reflect the newest treatments. Results have improved considerably over the decades: in the US, 5-year survival was about 33% for people diagnosed in 1975 and about 73% for those diagnosed in 2018. Only a hematologist who knows all the details can estimate your personal outlook.

02Simple explanation

Statistics summarise the results of many people. They answer the question “roughly how many out of 100”, but they cannot tell which of those 100 people is you.

Five-year survival is the percentage of people alive 5 years after diagnosis. It does not mean people die after 5 years — the observation simply covers a 5-year period. Many people stay in long-term remission and live for decades.

Leukemia statistics usually differ a lot by type. A general figure such as “leukemia survival” may have little to do with your disease. Information about your own type, subtype and risk group is more useful.

Numbers always look backwards: to measure a 5-year result, you have to wait at least 5 years. Meanwhile new drugs and treatments appear, so results for people treated today are often better.

It is natural to fear numbers or to get stuck on them. But you are not an average. Discuss the numbers that apply to your situation with your hematologist, and decide for yourself how much you want to know.

03Scientific explanation

Types of survival (NCI): cancer-specific survival — the percentage of people who have not died from their cancer within a given period; relative survival — without using cause-of-death information, it compares the survival of people with cancer to the general population of the same age and sex; overall survival — the percentage of people who have not died from any cause; disease-free survival — the percentage of people with no signs of cancer after treatment (also called recurrence-free or progression-free survival). Clinical trials often use event-free survival (EFS) and median overall survival — the time at which half of the patients in a group are still alive.

US figures (SEER): about 67,790 new cases of leukemia and 23,910 deaths are expected in 2026; the median age at diagnosis is 68, and 7.4% of cases occur in people under 20. The rate of new cases is 14.7 per 100,000 people per year and the death rate is 5.7; about 1.6% of people will be diagnosed with leukemia during their lifetime. In 2023, an estimated 563,581 people in the US were living with leukemia (including those in long-term remission). Age-adjusted death rates fell by an average of 1.8% a year over 2015–2024.

Why do the numbers differ? Registry data (such as SEER) cover everyone diagnosed, including older people and those with other illnesses, while clinical trials study selected patients. For example, SEER reports 5-year relative survival of 71.1% for CML, while the authors of the WHO classification note that with TKI therapy and careful monitoring, 10-year overall survival in CML is 80–90%. In AML the median age is about 70, and the overall figure reflects the harder outcomes in older patients; in children, according to the ACS, 5-year survival for AML is about 65–70%. Conditional survival describes the chance of surviving a further period for people who are already alive a certain time after diagnosis, for example 2 years (NCI).

Differences between countries: according to the WHO, more than 80% of children with cancer survive in high-income countries, compared with less than 30% in many low- and middle-income countries. This depends less on the biology of the disease than on the timing of diagnosis, access to treatment and supportive care. Country estimates of new cases and deaths are published by the Global Cancer Observatory (GLOBOCAN) of the International Agency for Research on Cancer (IARC), which covers 185 countries. The survival figures on this site are US data and may differ from results in other countries.

04What happens in the body?

Yesterday’s numbers, today’s treatment

To calculate 5-year survival, people have to be followed for at least 5 years after diagnosis. So the figures published today refer to people diagnosed in 2016–2022. In the US, 5-year survival in leukemia has risen from about 33% in 1975 to about 73% for people diagnosed in 2018 — and progress continues.

A graceful arc of light on a dark navy background rising gently from left to right; blood cells along it grow brighter and healthier, with a soft golden light on the right; no chart axes, numbers or text.
Figure 2.Over the decades, survival in leukemia has improved considerably, and progress continues.

Reading percentages correctly

“70% survival” means that about 70 out of 100 similar people are alive after a given time. Studies also use phrases such as “the risk falls by 50%”: this is a relative reduction, and without knowing the starting risk its meaning is incomplete. In such cases it helps to ask your doctor, “How many out of 100 people is that?”

You are not an average

Every person has their own age, general health, leukemia genetics and response to treatment. Even within the same type of leukemia, the outlook can vary widely. Statistics are like a map: they show the direction but not every turn of your path. The doctor who knows you is best placed to estimate your prognosis.

Among pale translucent human silhouettes on a dark navy background, one figure stands out with a clearer, warmer light; a soft glow fills its chest and bones, with calm blood cells around it.
Figure 3.You are not an average: each person’s disease and path are their own.

Living alongside the numbers

Some people want to know every number; others prefer not to. Both choices are normal. Use reliable sources (such as NCI, SEER and ACS), do not rely on single stories from forums, and if the numbers worry you, share this with your team or a psychologist. Numbers should not distract you from your treatment and your plans.

05How does a number come about?

Four steps from diagnosis to published statistics — and their limits.

Diagnoses are recorded — Cancer registries collect thousands of cases.1Diagnoses are recordedCancer registries collect thousands of cases.
Years of follow-up — People are followed for 5 years or more.2Years of follow-upPeople are followed for 5 years or more.
Calculation — Survival is compared, allowing for age and sex.3CalculationSurvival is compared, allowing for age and sex.
Publication and interpretation — The number appears years later; your doctor relates it to you.4Publication and interpretationThe number appears years later; your doctor relates it to you.
About a groupA number does not predict an individual’s fate.
Looks backwardsThe newest treatments are not yet reflected.
Shows progressResults have improved greatly over the decades.

06Key figures by type of leukemia in the US (SEER)

TypeAcute / chronicMedian age at diagnosis5-year relative survival (2016–2022)
AMLAcute7033.4%
ALLAcute1873.2%
CMLChronic6771.1%
CLLChronic7190.2%
All leukemiasAll6868.6%
Childhood leukemia (ages 0–19)All—86.7%

* Source: NCI SEER, Cancer Stat Facts. These are averages for the US population; results vary greatly by subtype, genetics, age and response to treatment. Discuss with your hematologist how far these figures apply to you.

07Key terms

Incidence
The number of new cases of a disease diagnosed each year.
Prevalence
The total number of living people who have or have had a disease at a given point in time.
Five-year survival
The percentage of people alive 5 years after diagnosis.
Relative survival
Survival of patients compared with the general population of the same age and sex.
Overall survival
The percentage of people who have not died from any cause.
Event-free survival (EFS)
Time without relapse or other defined events.
Median survival
The time at which half of the people in a group are still alive.
Conditional survival
The chance of further survival for people who have already survived a certain time.
Relative risk reduction
A reduction in risk expressed as a percentage; incomplete without the starting risk.
Registry
A database that systematically collects cancer cases, such as SEER.

08Key points

  • Statistics describe a group; they do not predict exactly what will happen to one person.
  • Five-year survival does not mean “living for 5 years”.
  • The numbers differ greatly between types of leukemia.
  • Numbers refer to the past; today’s treatments are often better.
  • In the US, 5-year leukemia survival has risen from ~33% (1975) to ~73%.
  • Registry and clinical trial figures reflect different groups.
  • Results also differ between countries.
  • Interpret the numbers together with your hematologist.

09Frequently asked questions

Does “70% five-year survival” mean I will live for 5 years?

No. It means that about 70% of similar people are alive 5 years after diagnosis. Many of them go on to live for many more years; 5 years is simply the standard observation period.

Why do I see different numbers online?

Numbers may come from different countries, years and age groups, and from different kinds of survival (overall, relative, event-free). Clinical trial results concern selected patients, while registry data cover everyone diagnosed. That is why it is important to check the source and what exactly was measured.

Are there figures for Azerbaijan?

The International Agency for Research on Cancer (IARC) publishes country estimates of new cases and deaths in its Global Cancer Observatory. Reliable country-level survival figures are not available everywhere. The survival figures on this site are mainly US data.

If I am older, do the numbers look worse for me?

On average, older people have harder outcomes in some leukemias, because they may tolerate intensive treatment less well and more often have other illnesses. But “biological age” and overall health matter more than age in years, and many newer treatments are suitable for older patients too.

What does “cure” mean, and when do doctors say someone is cured?

According to the NCI, cure means there are no traces of cancer after treatment and it will never come back. Some doctors may say that people who stay in complete remission for 5 years or more are cured, but follow-up continues, because a relapse can occasionally happen later.

The numbers frighten me. What should I do?

You can take a break from searching for numbers, stick to reliable sources and discuss your questions with your doctor. Sharing your feelings with loved ones or a psychologist also helps. Remember: a number does not write your personal story.

10Questions for your doctor

Take these questions to your appointment. The ones you save are collected on the “My questions” page and stay on this device only.

  1. How well do the statistics I read online apply to my situation?

  2. Which years of treatment do these figures come from?

  3. How do my age and genetics change these probabilities?

  4. What matters more for me than the statistics?

My questions

12Sources

  1. National Cancer Institute (NCI). Understanding Cancer Prognosis
  2. National Cancer Institute (NCI), SEER. Cancer Stat Facts: Leukemia
  3. National Cancer Institute (NCI), SEER. Cancer Stat Facts: Acute Myeloid Leukemia (AML)
  4. National Cancer Institute (NCI), SEER. Cancer Stat Facts: Acute Lymphocytic Leukemia (ALL)
  5. National Cancer Institute (NCI), SEER. Cancer Stat Facts: Chronic Myeloid Leukemia (CML)
  6. National Cancer Institute (NCI), SEER. Cancer Stat Facts: Chronic Lymphocytic Leukemia (CLL)
  7. National Cancer Institute (NCI), SEER. Cancer Stat Facts: Childhood Leukemia (Ages 0–19)
  8. National Cancer Institute (NCI), Surveillance Research Program. Cancer Survival Statistics Overview
  9. National Cancer Institute (NCI). NCI Dictionary of Cancer Terms: relative survival rate
  10. National Cancer Institute (NCI). NCI Dictionary of Cancer Terms: median overall survival
  11. American Cancer Society. Prognostic Factors and Survival Rates for Childhood Leukemia
  12. World Health Organization. Childhood cancer — Fact sheet
  13. International Agency for Research on Cancer (IARC/WHO). Global Cancer Observatory (GLOBOCAN)
  14. Leukemia (Nature) — WHO 5th edition. The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: Myeloid and Histiocytic/Dendritic Neoplasms
Medical disclaimerThis information is for educational purposes only and does not replace medical advice.
Last updated: ← Back to the list

Open the interactive site →