How many infants worldwide, not just in the United States
also called global infant leukemia burden · worldwide infant leukaemia cases per year
About 132 million babies are born each year worldwide, and if leukemia hits infants everywhere at the US rate, roughly 5,400 babies a year develop acute leukemia. No cancer registry anywhere actually counts that number, so it is an estimate built from two solid figures and one shaky assumption.
The same thing, for a scientist
There is no published global incidence figure for infant (<1 year) acute leukemia. The defensible construction is US infant incidence (~41 per million infants per year) applied to the UN's ~132 million annual live births, giving ~5,400 infant acute leukemia cases worldwide per year, of which perhaps 2,100-2,400 are KMT2A-rearranged infant ALL and ~1,200 KMT2A-rearranged infant AML. The assumption that a US rate transfers globally is the weak link: registry coverage outside high-income countries is partial, infant leukemia is under-ascertained where diagnostic cytogenetics is scarce, and competing infant mortality truncates the at-risk population in exactly the regions with the most births.
The denominator, stated first2 deeper
Denominator: live births worldwide in one year (~132 million), as a proxy for infant person-years at risk 1. Every number below is cases per year among children in their first year of life, worldwide. It is not per million population, and it is not per million children aged 0-14. Converting between those three denominators is where this literature goes wrong.
| Quantity | Value | Source |
|---|---|---|
| Global live births per year | ~132 million | UN World Population Prospects 2024 1 |
| US infant acute leukemia incidence | ~41 per million infants/yr | Brown 2013, see epi-infant-leukemia-incidence 2 |
| Implied global infant acute leukemia cases | ~5,400 per year | arithmetic, this node |
| Implied global infant ALL (~56% of infant acute leukemia) | ~3,000 per year | arithmetic 2 |
| Implied global infant KMT2A-r ALL (70-80% of infant ALL) | ~2,100-2,400 per year | arithmetic 2 |
| Implied global infant KMT2A-r AML (~50% of ~2,400 infant AML) | ~1,200 per year | arithmetic 2 |
The multiplication is trivial. The honesty is in the caveats.
Why the US rate probably does not transfer
Three separate biases push in different directions, and nobody has quantified their net effect.
- Ascertainment. The IARC international incidence programme accepted only 153 of 532 invited registries as meeting quality standards for 2001-10, and coverage is thinnest in sub-Saharan Africa and South Asia, which is where most of the 132 million births happen 3. An infant who dies of undiagnosed leukemia in a setting without paediatric haematology never enters a registry, so the true count is under-recorded but the estimate above is not.
- Competing mortality. Regions with the highest birth counts also have the highest infant mortality from other causes, so person-years lived in the first year of life are fewer than the birth count implies. This biases the estimate upward.
- Real biological variation. Registered childhood cancer incidence differs severalfold between IARC regions, and some of that is genuine 3. Whether infant KMT2A-rearranged leukemia specifically varies by region is an open question that the transplacental-exposure hypothesis predicts should be answerable.
The defensible statement is: the worldwide infant acute leukemia population is of the order of a few thousand children a year, and the worldwide infant KMT2A-rearranged population is of the order of three thousand a year. Any tighter number is false precision.
What this does to the platform argument
Three thousand infants a year worldwide is still a tiny commercial market, but it is roughly twenty times the ~160 US infant cases that the US-only framing produces. That matters for two decisions the lab will actually face.
- Trial feasibility. An international consortium can accrue in a year what a US-only design accrues in twenty. Interfant-06 enrolled 651 infants across many countries over a decade; no single country could have done it 4. See
trialdesign-basket-and-umbrella-for-rare-fusions. - Regulatory economics. Orphan thresholds are jurisdiction-specific (US: fewer than 200,000 prevalent patients; EU: not more than 5 in 10,000), so a global population of a few thousand qualifies comfortably everywhere while still being far too small to fund a conventional programme. See
reg-orphan-drug-designation.
| metric | derived annual case count |
|---|---|
| value | 5400 |
| unit | infant acute leukemia cases per year |
| population | infants under 1 year |
| region | worldwide |
| year | 2024 |
| source | derived: UN WPP 2024 births x US infant incidence (Brown 2013) |
| denominator | ~132,000,000 live births per year worldwide |
| assumption flagged | US incidence applied unchanged to global births; not validated against any non-US infant-specific registry series |
derivation — the numbers behind the chart
[
{
"step": "global live births per year",
"value": 132000000
},
{
"step": "US infant acute leukemia incidence per million infants",
"value": 41
},
{
"step": "implied global infant acute leukemia cases per year",
"value": 5400
},
{
"step": "implied global infant ALL cases per year",
"value": 3000
},
{
"step": "implied global infant KMT2A-r ALL cases per year (70-80%)",
"value": [
2100,
2400
]
},
{
"step": "implied global infant KMT2A-r AML cases per year (~50% of ~2400)",
"value": 1200
}
]- A global case count built by multiplying a US rate by global births can be quoted back at the lab as if it were measured. Every downstream use of this number must carry the derivation, or a reviewer who checks will find no registry behind it.
- If the true global burden is concentrated in countries that cannot access a menin inhibitor at Western prices, then 'we can pump out therapies' describes access for a few hundred children, not a few thousand.
- Does any national registry publish infant (<1 year) leukemia incidence separately from the 0-4 age band, and if so do the rates agree with the US figure?
- Is infant KMT2A-rearranged leukemia incidence genuinely different in Latin America, where the transplacental topoisomerase-II-poison hypothesis has been most actively studied?
- How much of the apparent deficit relative to the US rate is under-ascertainment versus competing infant mortality, and can the two be separated using infant mortality rate as a covariate?
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.dataset
There are approximately 132 million live births worldwide per year according to the UN's 2024 revision of World Population Prospects.
World Population Prospects 2024: Summary of Results · United Nations Department of Economic and Social Affairs, Population Division · United Nations 2024What this evidence cannot show
A model-based estimate for the current period, not a count; birth registration is incomplete in several high-birth countries, which is the same coverage problem that affects the cancer numerator.
Evidence that argues against this (1)
- A global live-birth projection is a demographic model with its own uncertainty intervals, and applying a single incidence rate to it assumes infant leukaemia incidence is geographically uniform. The registry evidence in this same entry says it is not measured uniformly, so the product of the two carries both uncertainties. The Lancet Oncology 2017
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.review
US infant acute leukemia incidence is about 41 per million infants per year (~160 US cases: ~90 ALL, ~70 AML), with KMT2A rearrangement in 70-80% of infant ALL and about 50% of infant AML.
Treatment of infant leukemias: challenge and promise · Brown P · Hematology Am Soc Hematol Educ Program 2013What this evidence cannot show
A 2013 educational review, US-specific, and the partner frequencies come from referral cohorts rather than population registries. Using it as the global rate is the explicit assumption this node is built on and flags.
Evidence that argues against this (1)
- Applying any single national rate globally is methodologically weak; IARC's own regional rates for childhood leukaemia vary severalfold, so a single multiplier cannot be right everywhere. source 2017
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Only 153 of 532 invited cancer registries, from 62 countries, met IARC quality standards for the 2001-10 decade, and registered childhood cancer incidence varies considerably between and within world regions.
International incidence of childhood cancer, 2001-10: a population-based registry study · Steliarova-Foucher E, Colombet M, Ries LAG, et al. · Lancet Oncology 2017What this evidence cannot show
Covers 2001-10, so it is now historical. Reports rates for ages 0-14 and 0-19 with age-specific rates for 0-4, not for infants under one year specifically, which is exactly the granularity this beat needs and does not have.
Evidence that argues against this (1)
- This is the number that should bound every global estimate in the beat: only 153 of 532 invited registries, from 62 countries, met IARC quality standards. Global infant leukaemia burden is therefore extrapolated from a minority of registries concentrated in high-income settings, and infant leukaemia is exactly the presentation most likely to be missed where diagnostic access is limited - biasing global counts downward. The Lancet Oncology 2017
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Interfant-06 accrued 651 infants with newly diagnosed acute lymphoblastic leukemia through an international consortium, showing that infant-leukemia trials are feasible only at multinational scale.
Outcome of Infants Younger Than 1 Year With Acute Lymphoblastic Leukemia Treated With the Interfant-06 Protocol: Results From an International Phase III Randomized Study · Pieters R, De Lorenzo P, Ancliffe P, et al. · Journal of Clinical Oncology 2019What this evidence cannot show
Enrolment spanned roughly a decade and is weighted towards high-income registries, so the enrolled population is not a global sample.
Evidence that argues against this (1)
- Interfant-06 accruing 651 infants demonstrates consortium reach, not incidence. Trial accrual is a function of participating centres and eligibility, so it provides a floor on cases in participating countries over the accrual period and cannot be converted to a global rate. The Lancet Oncology 2017
Counter-evidence (NCT05327894 Interfant-21, read 2026-09-18): the consortium's own current KMT2A-restricted protocol plans 160 infants over ~4.8 accrual years across 24 countries, i.e. ~33/yr, not the ~48/yr this node assumes. Every accrual-year figure in the node rises by ~45%.
- ← related toChildhood leukaemia worldwide, ages 0-14
Counter-evidence (Ward 2019, PMID 30824204): registries recorded 224,000 of an estimated 397,000 global childhood cancers in 2015, so 43% went undiagnosed (3% in western Europe and North America, 57% in western Africa). Registry-rate-times-population gives diagnosed cases, not incident cases, and the shortfall sits in the highest-birth regions.
- supports →How many children this affects
scales the US infant rate to global births
worldwide annual case estimate for this disease
- supports →Interfant-06 — 651 infants, and the definitive answer that myeloid-style consolidation does not help
multinational accrual is the only feasible route at this incidence
national decomposition of the global estimate