Background: Most of the world's population is not covered by cancer surveillance systems or vital registration, and worldwide/UN-regional cancer incidence is estimated using a variety of methods. Quantifying the cancer burden in children (<15 years) is more challenging than in adults; childhood cancer is rare and often presents with non-specific symptoms that mimic those of more prevalent infectious and nutritional conditions.

Methods: A Baseline Model (BM) was constructed comprising a set of quality assured sex- and age-specific cancer rates derived from the US Surveillance, Epidemiology and End Results (SEER) program, for diagnostic groups of the International Classification of Childhood Cancers (ICCC-3) 3^rd edition, and information on a known risk factor for endemic Burkitt lymphoma and Kaposi's sarcoma. These rates were applied to global country-level population data for 2015 to estimate the global and regional incidence of childhood cancer. Results were compared to GLOBOCAN 2018, extrapolations from the International Incidence of Childhood Cancer (IICC-3) and estimates from the Global Childhood Cancer (GCC) model (based on IICC-3 data combined with information on health care systems and other parameters).

Results: The BM estimated 360,114 total childhood cancers occurring worldwide in 2015; 54% in Asia and 28% in Africa. BM estimated standardised rates ranged from ∼178 cases per million in Europe and North America, through to ∼218 cases per million in West and Middle Africa. Totals from GLOBOCAN and extrapolations from the IICC-3 study were lower (44.6% and 34.7% respectively), but the estimate from the GCC model was 10.2% higher. In all models, agreement was good in countries with very high human development index (HDI), but more variable in countries with medium and low HDIs; the discrepancies correlating with registration coverage across these settings.

Conclusion: Disagreements between the BM estimates and other sources occur in areas where health systems are insufficiently equipped to provide adequate access to diagnosis, treatment, and supportive care. Incorporating aetiological evidence into the BM enabled the estimation of the additional burden of Burkitt lymphoma and Kaposi sarcoma; similar adjustments could be applied to other cancers, as and when information becomes available.