Hematological neoplasms (HNs) are divided into leukemias and lymphomas. They account for one-third of all childhood cancers and are among the most common cancers in children (1). Leukemias are a group of diseases involving an uncontrolled proliferation of hematopoietic stem cells in the bone marrow caused by several risk factors such as genetic (e.g., chromosomal translocations, rare germline mutations, or epigenetics) and environmental factors (e.g., infections and exposure to chemicals or ionizing radiation) (2, 3). Meanwhile, lymphomas are a diverse group of diseases that arise from the clonal proliferation of lymphocytes (4). Due to the heterogeneity of these malignancies and the improvement of diagnostic methods based on genetic and pathological examinations (5), the International Classification of Childhood Cancer (ICCC), which classifies cancer histology codes in children, has been updated to the latest edition, the third edition (ICCC-3) of 2017 (1).

The most recent international collaborative study on childhood cancer (0–14 years), published in 2017 and covering the period 2001–2010, found an age-standardized rate (ASR) using the Segi world standard population (ASR_SEGI) of 61.6 cases per million child-years, of which leukemias accounted for 46.4 and lymphomas for 15.2 (6). Slightly lower rates were estimated in a European study, covering the period 1988–1997, which compared childhood cancer incidence between geographical areas in Europe (7). Both studies reported higher incidence rates, particularly in southern Europe, compared with the rest of the world (6, 7).

The latest European study assessing incidence trends showed a statistically significant increase in the annual percentage change (APC) of 0.7% per year from 1991 to 2010 for childhood leukemias, whereas no increase was observed for lymphomas (8). In addition, some specific countries or regional population-based cancer registries have shown different results on childhood HN incidence trends (9, 10). Previously published data on childhood HNs in Spain covered the period 1983–2002 and reported an ASR_SEGI of 64.4 cases per million children, of which 49.9 cases were leukemias and 18.5 were lymphomas. The incidence trends for the same period were also discussed and showed an increase in the early years (1983–1991), followed by a stabilization in the second period (1992–2002) (11). Later studies covering a longer period up to 2013 for the incidence and up to 2007 for the incidence trends reported similar results for all HNs and leukemias (1, 12, 13); however, lymphomas showed stability throughout the period (13). The lack of updated results on the incidence and incidence trends of HNs in Spain requires an analysis of more recent years.

Previous studies have reported that developed countries tend to have higher incidence rates of HNs and southern European countries have higher incidence rates of lymphoma (6), in addition to a variation in trends (9, 10). Although stable incidence rates have previously been described in Spain (12, 13), we hypothesize that similar results to those reported in Europe will currently be observed, with higher incidence rates and changes in trends for both HN groups. Therefore, this study aims to provide an overview of the incidence and incidence trends of childhood HNs in Spain. This will be performed by adding more recent years and coverage areas to the previous studies, taking into account the most recent classification of childhood HNs, by age group, sex, and cancer type. Furthermore, we aim to compare the results with other population-based cancer registries in southern Europe.

Information on the epidemiology of childhood HNs is scarce, as most studies focus on adults due to their higher incidence compared to children (6). Therefore, this study, using data from 15 PBCRs, allows us to provide a broad descriptive analysis of the incidence and incidence trends of all childhood HNs over 36 years, from 1983 to 2018 in Spain. In addition, the long time period of this study and the increased population coverage allow us to update the epidemiological results of these malignancies since previous publications in Spain covered a smaller population area and a shorter time period until 2007 (12, 13).

New histology codes created after the 2005 ICCC-3, due to multiple new diagnostic methods in progress using molecular techniques, were accurately grouped according to the updated version, the 2017 ICCC-3 (5). The latest classification is not widely used in other studies due to its recent publication. At the national level in Spain, previous publications on HNs in children have used the ICCC-3 of 2005 (12, 13, 24). Therefore, the analysis of incidence and incidence trends not only provides insight over a long period of time but also provides more recent results in terms of histological aggregation. At the international level, only one study was found that used the ICCC-3 of 2017 (25), since the most recent studies used the ICCC-3 of 2005 (26–28). Therefore, we suggest the use of this more recent classification for future studies in European PBCRs. Furthermore, ASR exhibited variability between the different standard populations used. In particular, the use of the Segi 1960 standard population yielded the highest incidence rates due to its emphasis on younger age groups (20). Conversely, the ASR_E and the ASR_WHO demonstrated comparable rates, suggesting that the new WHO standard reflected the study population structure more accurately. In light of these findings, we recommend adopting the new WHO standard to be used for international comparisons.

Overall, we reported that two-thirds of all childhood HNs were leukemias and the remaining third were lymphomas. Similar proportions have been reported in southern and eastern Europe (29). An ASR_E of 64 cases per million child-years of HNs was observed during the study period, with a statistically significant increase in incidence during the first 3 years, followed by stable rates for the remaining years. To the best of our knowledge, no publication in the literature provides an overview of the incidence and incidence trends of HNs as a whole. Most of the authors report the incidence and incidence trends of HNs divided into two large groups: leukemias and lymphomas (6, 7, 30). Differences between sexes and age groups have been observed, with boys and children aged 1–4 years tending to have higher incidence rates. Furthermore, the sex differences in HNs are greater than in other cancers, but the endogenous causes are largely unknown despite all the research that has been performed (31, 32).

Lymphoid leukemias were the main subgroup of all leukemias, followed by a smaller proportion of AML. An ASR_SEGI of 47.3 cases of childhood leukemia per million child-years was observed, with similar results to those previously reported in Spain and other European regions (12, 24, 30). These results contrast with the lower values reported in developing countries, such as sub-Saharan Africa and South Asia (6). The peak in leukemia incidence in the age group 1–4 years, specifically observed in LL, is a pattern that has been reported in previous studies (12, 24, 33). At present, this peak was first identified in the early 20th century, and it was suggested that the risk factors triggering this increase were unknown (34). However, subsequent studies have suggested that benzene, agricultural exposures, smoking and alcohol consumption, cigarettes, and illicit drugs during pregnancy are predisposing factors for childhood LL (35). Meanwhile, the reported incidence peak at the age of <1 year, corresponding to AML, is mainly attributed to hereditary conditions such as Down syndrome and Fanconi anemia (36). From an international perspective, the leukemia incidence patterns observed in this study by age group, sex, and subtype were similar to those in France and Portugal (1, 30, 33). In addition, Surveillance, Epidemiology, and End Results (SEER) leukemia incidence rates were similar to those observed in this study, whereas Cancer Research UK incidence rates were higher (37, 38).

Similar proportions were observed for HL, NHL, and BL, with ASR_E values of 6.0, 5.4, and 6.0 cases per million child-years, respectively, out of a total of 20 cases per million child-years for all lymphomas. The most recent study published in Spain reported lower incidence values for lymphomas as a whole, but this can be attributed to the exclusion of MLN and unspecified and other specified lymphomas subtypes from the incidence analysis (13). Internationally, southern Europe, and therefore the region included in this study, has one of the highest lymphoma incidence rates in the world, with the exception of other Mediterranean regions such as North African countries (6, 7). It has been suggested that the elevated incidence rates observed in the Mediterranean region may be due to environmental factors, as different ethnic groups living in the region, including Caucasians, Arabs, and Jews, exhibit higher incidence rates of these malignancies (39). Aside from this region, Cancer Research UK lymphoma incidence rates were similar to those observed in this study, while SEER results were higher, specifically for HL (37, 38).

Sex and age differences were also observed in our results, which have been widely reported by other authors (6, 12, 13). Although the main risk factors for these sex differences are currently unknown, some authors suggest that boys have an innate susceptibility associated with immune surveillance that makes them more vulnerable to proto-oncogenic mutations (6, 31, 40). The association between lymphoma incidence and age is due to the fact that these malignancies are more common in adolescents, with the age group of 10–14 years having the highest age-specific incidence rate of all age groups (13, 41).

Our findings showed that the overall incidence trends of HNs were similar to those previously reported for leukemias in Spain, with a statistically significant increase in the early years of the period, followed by a stabilization since 1989 (12). We suggest that this steep increase in leukemia incidence during 1985–1988 is probably an artifact due to the lack of cases registered by the provinces/islands during the first years of the study. In our study, the incidence rates were found to be stable in most recent years added. Meanwhile, other studies have shown different patterns of leukemia incidence trends, with some reporting a steady statistically significant increase without a stabilization (8, 10, 26, 27). However, others reported a stabilization over the whole period of the study (9, 42, 43), especially for LL and AML in Germany for the most recent period available in the study (28).

This heterogeneity in incidence trends was also observed for lymphomas, as our results revealed a statistically significant steady increase of 1% per year, in contrast to previous findings in southern Europe and Spain, where no increase was detected (8, 13). However, a similar increase was observed in Denmark and Australia for the periods 1977–2014 and 1983–2015, respectively (9, 26). We hypothesized that the increase in lymphoma incidence could be due to an increase in environmental risk factors in the Mediterranean region, as these have been suggested to be the main risk factors for these malignancies in this area (39). However, other countries, such as Switzerland, Italy, Japan, and the England, reported a stabilization since the beginning of the period analyzed (10, 42, 43). Many factors may contribute to the differences in incidence trend results, despite all the standardized definitions of quality criteria by IARC. One factor that may contribute to these differences is the use of different versions of the ICCC in the histological grouping, as this may complicate the interpretation of incidence trends (44).

Despite all the studies discussed on incidence trends, the number of collaborative studies is scarce. One of the reasons contributing to this may be the implementation of the General Data Protection Regulation in Europe since 2018, which, among other things, may affect researchers and make data sharing more difficult than before (45).

Our results are strengthened by the use of population-based incidence data from REDECAN, in which more than 95% of cases were verified microscopically. In addition, the large area covered in Spain and the long time period allowed us to perform the main analysis with high statistical power. Furthermore, few studies grouped the histology codes according to the latest ICCC-3 of 2017 and provided the ASR using the new WHO world standard population (2000–2025). However, the sample size was limited when performing all sub-analyses involving less common pathologies, such as calculating the incidence of specific sex, age group, and HN subgroup. Another limitation of the study was the lack of period homogeneity between cancer registries, but this was addressed by ensuring that most of the provinces/islands contributed to each year of the trend. In addition, this study does not represent the entire Spanish child population, as incidence rates may differ between the covered and non-covered areas, as a previous study has revealed differences in incidence rates between urban and rural areas in Spain (46). Finally, changes in histological grouping and diagnostic criteria throughout the period could have also affected the results obtained and the comparisons with previous studies.

This study presents an updated population-based analysis of the incidence and incidence trends of childhood HNs in Spain covering a long period from 1983 to 2018 and a large area. Our results showed that leukemias are the most common HNs in children, and their incidence has remained stable since 1988, while lymphomas are less common but their incidence is increasing every year. The observed incidence rates of lymphomas are similar to those reported in other southern European countries, while the similarities in leukemia incidence rates were limited to the southwestern European countries. Furthermore, we recommend the use of the new ICCC-3 from 2017 and the use of the new WHO world standard population (2000–2025) in future studies conducted by European PBCRs. Collaborative cancer registry projects, such as the REDECAN, provide the opportunity to assess epidemiological indicators of less common cancers, such as pediatric HNs. Consequently, all these reported epidemiological findings could help health authorities and clinicians to have updated results of these malignancies for the more recent years in Spain. Furthermore, the results suggest the need for more studies focusing on the risk factors of childhood lymphomas, as their incidence is increasing every year in Spain.

REDECAN Working Group: Girona Cancer Registry – Coordination of hematology neoplasms in REDECAN: Rafael Marcos Gragera, Montse Puigdemont, Anna Vidal Vila, Arantza Sanvisens, and Jan Trallero; Tarragona Cancer Registry – Data coordination in REDECAN: Marià Carulla, Alberto Ameijide, Clàudia Pla, and Jaume Galceran; Álava Cancer Registry: Arantza López de Munain, Patricia Sancho, and María Luisa Iruretagoyena; Albacete Cancer Registry: Cristina Ramírez; Asturias Cancer Registry: Susana Merino Perera, Virginia Menéndez García, and Marta Rodríguez Camblor; Bizkaia Cancer Registry: Visitación de Castro, Marta De La Cruz, and Joseba Bidaurrazaga; Castellón Cancer Registry: Consol Sabater Gregori, Isabel Sáez Lloret, Ana Vizcaíno Batllés, and Xavier Peñalver Herrero; Ciudad Real Cancer Registry: Cristina Díaz del Campo; Cuenca Cancer Registry: Ana Isabel Marcos and Rosario Jimenez Chillarón; Gipuzkoa Cancer Registry: Amaia Aizpurua; Canary Islands Cancer Registry: María Dolores Rojas Martín, María Araceli Alemán Herrera, Emilio De Miguel, María Magdalena Ramos Marrero, María Carmen Gábas, Beatriz Hidalgo Martín, and Olga Leonor Velázquez Castilla; Granada Cancer Registry: María José Sánchez, Daysi Yoe-Ling Chang-Chan, and Miguel Rodríguez Barranco; La Rioja Cancer Registry: María Isabel Palacios and Enrique Ramalle; Mallorca Cancer Registry: Paula Franch, Patricia Ruiz Armengol, and Carmen Sánchez Contador; Murcia Cancer Registry: María Dolores Chirlaque, Antonia Sánchez Gil, and Ricardo José Vaamonde; Navarra Cancer Registry: Marcela Guevara and Eva Ardanaz; Castilla y León Cancer Registry: Pilar Gutiérrez, Rufino Álamo, and Enrique Cabrera; Registro Español de Tumores Infantiles: Adela Cañete, Elena Pardo, and Rafael Peris Bonet; Registro de Tumores Infantiles y Adolescentes de la Comunitat Valenciana: Ana Vizcaino, Fernando Almela Vich, and Noura Jeghalef El Karoni.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by CEIM Girona Institut d’Investigació Biomèdica de Girona Dr. Josep Trueta (IDIBGI). Written informed consent from the participants’ legal guardian/next of kin was not required to participate in this study in accordance with the national legislation and institutional requirements.

JT, AS, and RM-G contributed to the study conception and design. Data collection was performed by FA, NJ, IS, CD-D-C, AM-N, AAA, PS, MO, MS, JP, PF, MC, MG, CR, MC, MA, PG, and RM-G. Data analysis was performed by JT, AS, and AA. The first draft of the manuscript was written by JT. All authors contributed to the article and approved the submitted version.