WHO’s childhood leukaemia map reveals a gulf in survival—and in what countries can measure

For a family confronting childhood cancer, the decisive question is not simply whether an effective treatment exists. It is whether a child can receive the right diagnosis, begin appropriate treatment promptly, complete it safely and obtain supportive care along the way. A newly accessible global dataset shows how sharply that chain differs from place to place.

On September 22, the World Health Organization released the complete dataset behind a report it had published on May 8. It provides the first estimates made with a common methodology for all 194 WHO member states, using five-year survival from lymphoid leukaemia as a window onto childhood-cancer care.

WHO reports estimated survival of roughly 80–90% in high-income settings, compared with about 40% where health systems are weaker. Estimated levels are generally higher in Europe and the Western Pacific, while Africa and parts of the Eastern Mediterranean face persistent gaps. Adolescents aged 15–19 have a wider survival divide than children aged 0–14, especially in low- and middle-income countries.

Those patterns matter because childhood cancers are often treatable. Yet where a child lives can determine whether the necessary system—clinicians, laboratories, medicines, blood products, hospitals and follow-up—works as a connected whole.

What the number actually measures

“Five-year net survival” is not a head count of children confirmed alive five years after diagnosis. The International Agency for Research on Cancer defines net survival as an estimate that removes the effect of deaths expected from other causes. In simplified terms, it asks what survival would look like if the cancer were the only possible cause of death.

That adjustment helps researchers compare populations with different background mortality. It also makes the figure more abstract than an individual prognosis. A national estimate cannot tell a family what will happen to one child, and it should not be read as a precise ranking of hospitals or countries.

The distinction is particularly important because the underlying observations are uneven. WHO says nearly 80% of the world’s population is not covered by a population-based cancer registry, and in some regions many childhood cancers may never be diagnosed or treated. A standardized estimate can fill out a global comparison, but it does not turn missing records into directly observed outcomes.

The measure is also deliberately narrow. Lymphoid leukaemia can reveal whether a health system reliably delivers complicated, sustained care, but it cannot represent every brain tumour, lymphoma, bone cancer or other childhood malignancy. Different cancers require different combinations of surgery, radiotherapy and medicines.

A second dataset supplies a different view

An IARC study known as SURVCAN-3 offers a useful comparison. Researchers followed 16,821 cancer cases among children younger than 15 recorded by 47 population-based registries in 23 predominantly low- and middle-income countries. The cases covered diagnoses during 2008–2017 and multiple cancers.

For leukaemia, reported three-year survival ranged from just above 30% in Kenya to almost 90% in Puerto Rico. These are registry-based observations for the participating populations, not estimates covering every country.

The two projects therefore answer related but different questions. SURVCAN-3 follows recorded cases in selected registries and supplies observed benchmarks. WHO offers a standardized five-year lymphoid-leukaemia baseline spanning every member state. Their time periods, ages, survival endpoints and geographic coverage differ, so their percentages cannot be treated as interchangeable measurements.

Read together, however, they reinforce the same broad finding: survival varies profoundly with the systems surrounding a child.

A map of questions, not answers

A low estimate cannot identify the broken link. WHO’s childhood-cancer guidance lists missed, incorrect or delayed diagnosis; inaccessible treatment; treatment abandonment; toxicity; and relapse among the causes of avoidable deaths in lower-resource settings.

That turns each survival estimate into a practical checklist. Are children being diagnosed accurately? Can families reach treatment and afford to remain there? Are essential medicines and blood products consistently available? Can hospitals manage infections and other complications? Do registries follow patients long enough to show whether care is improving?

The new dataset cannot answer those questions by itself. Its real value is that every country now has a comparable signal showing where investigators should start—and where better records may be as necessary as better treatment.