Hematological neoplasms are a group of diseases in which the hematopoietic system, the one in charge of the differentiation and proliferation of erythrocytes, thrombocytes, granulocytes and lymphocytes, is involved. The altered proliferation and/or differentiation of the different lineage precursors drives to a wide group of diseases with diverse etiology, presentation and outcomes, and depending on the lineage we find the first stratification of these neoplasms: myeloid neoplasms, lymphoid neoplasms (LNs) and histiocytic/dendritic neoplasms.

LNs, which comprise a heterogeneous group of more than 60 histological subtypes, are ranked as the 6^th to 7^th most common cancer worldwide (1). Broadly, they are grouped into four main categories: Hodgkin lymphoma, precursor lymphoid neoplasms, mature B-cell neoplasms, and mature T- and NK-cell neoplasms. Changes in our understanding of LNs have resulted in the continuous update of classification schemes over the past 60 years (2). The World Health Organization (WHO) with its first classification of tumors of hematopoietic and lymphoid tissues published in 2001 (3), and subsequently updated in 2008 (4), 2016 (5) and 2022 (6), is currently regarded as the gold standard for the diagnosis and research of LNs. Furthermore, the correspondence established between the WHO classification and the last editions of the International Classification of Diseases for Oncology (ICD-O-3) codes (7), has provided international standards for the registration and epidemiological surveillance of these diseases.

During the last decades, population-based data of LNs has become more available, with multiple studies from hematology-specialized registries (8, 9) or large European and North American datasets (10–14). Specifically in Spain, however, few regional studies have reported incidence and survival data of LNs (15, 16), while there is a lack of subtype-specific data at a nationwide level. Therefore, the aim of this study was to describe incidence and survival patterns of LNs subtypes in Spain over the period 2002-2013, and to estimate the number of LNs expected in 2023, using data from the Spanish Network of Cancer Registries (REDECAN).

This paper presents population-based incidence and survival of LNs in Spain during the period 2002-2013. Our study, which includes 39,156 cases diagnosed and registered by the 13 Spanish PBCRs that make up REDECAN, further complements existing European data by providing incidence patterns, projections for 2023 and survival analysis of LNs by subtypes.

Incidence rates for the most frequent subtypes are in line with those reported in large European datasets (10), as well as by hematology-specialized registries from the United Kingdom (Haematological Malignancy Research Network, HMRN) (8). By contrast, we reported lower rates in comparison to France (Côte d’Or) (9), to the United States 2016 projected incidence (14), and to Australia’s data for 1982-2006 (22), presumably due to differences in the study periods and classification. Regarding specific entities, we reported lower rates of marginal zone lymphoma and higher rates of follicular lymphoma when comparing to the HMRN (8), probably due to differences in classification and different completeness of cancer registries. Similarly, we report higher rates of chronic lymphocytic leukemia/small lymphocytic lymphoma and mycosis fungoides/Sézary syndrome, mainly attributable to a higher registration of indolent subtypes by REDECAN rather than a higher actual incidence in Spain. Our data also show higher rates of mantle cell lymphoma which is related to environmental and genetic risk factors like farm life, atopy, and allergy (23).

Incidence of LNs showed a male predominance, with the main exception of follicular lymphoma (sex ratio 0.94), also described in the literature (8, 9). With respect to age at diagnosis, most LNs are diagnosed at advanced ages, with a median age at diagnosis around 67 years. In line with previous studies, cases of precursor lymphoid neoplasms, Hodgkin lymphomas, and Burkitt lymphoma/leukemia were diagnosed at younger ages. In addition, we evidenced that most NOS cases are diagnosed at advanced ages, with an incidence that increases markedly after 70 years. This suggests a decrease in the quality of diagnoses in older people, since they are less likely to receive aggressive diagnostic tests due to comorbidities and/or frailty (24).

The global incidence trend of LN was stable during the study period in accordance with reported results of other Western countries (14, 25), although significant variations were found in specific subtypes. However, the continuous update of the WHO classification of LNs across the period of study hampers the interpretation of these trends. Whilst studies have been shown that computer converted ICD-O-3 historical codes for lymphoma subtypes are generally reliable, and that such agreement is improved when those codes are grouped (26), some changes are hard to overcome. For instance, in 2008 the International Workshop on chronic lymphocytic leukemia changed the definition of the disease, now requiring an absolute B-cell count of 5,000 cells/μL rather than the previous absolute lymphocyte count of 5,000 cells/μL, which caused that many former Rai stage 0 cases now should be considered as a pre-malignant condition (monoclonal B-cell lymphocytosis) (27). This may explain the negative trend in the incidence of chronic lymphocytic leukemia/small lymphocytic lymphoma evidenced in our study. On the other hand, we reported a marked decrease in the incidence of NOS cases across the period of study. This might be attributed to more specific clinical diagnoses and also to improved coding in Spanish PBCRs. These improvements in coding are probably due to the training courses organized by REDECAN and given to registry professionals, in particular hematological neoplasm coding courses held during the study period.

As expected, 5y-NS was widely different across LNs subtypes, and our results are broadly consistent with those reported by specialized registries in the UK (8), France (9), and in European (HAEMACARE, EUROCARE-5) (11–13) and North American datasets (28). However, the survival estimates that we present are higher than those reported by studies using the HAEMACARE database (11). These differences can be due to the HAEMACARE’s study period (2000–2002), prior to the appearance of new treatment lines and more accurate diagnostic techniques that significantly improved survival, and to the use of the Hakulinen method for estimating survival instead of the Pohar-Perme method (29). By contrast, our results for most subtypes were lower than those presented by the HMRN (8).

Overall, survival rates differed considerably across age-groups, a well-established prognostic factor for most cancers (30). The poor survival of elderly patients with LN is mainly attributed to the presence of comorbidities or to their frail status, which hampers firstly, seeking for a specific diagnosis, and subsequently, the application of several treatment protocols, such as several chemotherapy regimens or stem-cell transplantation (31–33). On the other hand, we evidenced differences in the survival of mature B, T and NK-cell neoplasms between sexes that would be related to the impact of chromosomal and hormonal control of immunity (34), among other factors, and its consequent cancer evolution and response to treatment (35, 36). Finally, NS for all LNs improved significantly during the period of study, which was mostly attributed to a better prognosis of mature B-cell neoplasms, particularly in plasma cell neoplasms and mantle cell lymphoma. This could be related to changes in treatment patterns, such the introduction of rituximab during 2004-2005 for mantle cell lymphoma, and the application of novel agents (thalidomide, bortezomib, lenalidomide) together with autologous stem cell transplantation in 2009 for multiple myeloma (37, 38). The favorable effect on survival of these therapeutic changes were also reported by European PBCRs (39–41).

According to our projections, we expect 17,926 new cases of LNs in 2023. However, these data take into account neither the changes in the WHO classification (new updates were launched in 2016 and 2022) (5, 6), nor the impact of coronavirus disease 19 (COVID-19). Regarding the latter, according to multiple studies, the COVID-19 pandemic may have caused significant delays in diagnosis and initiation of treatment in many cancer sites. This may lead to a subsequent increase in the number of cases diagnosed in more advanced stages, with implications on survival, quality of life, and economic costs (42–44). Finally, the projections presented here could be underestimated due to the continuous improvement in the capture of hematological neoplasms by PBCRs and the development in recent years of more effective diagnostic tools. However, the methods and strategy used in data analysis provide a reasonably accurate estimate of the current incidence of LNs in Spain, based on the greatest population coverage currently provided by cancer registries in this country.

Some limitations must be considered when interpreting our data. First, changes in the classifications and diagnostic criteria of the subtypes studied hamper the interpretation of our results, as well as comparisons with previous studies. In addition, lack of homogeneity among REDECAN registries, since there is no centralized pathology and clinical review, can lead to problems of quality and completeness. For instance, this may have implications in the percentage of NOS cases in all provinces or affect the estimates of subtypes such as chronic lymphocytic leukemia or Waldenström’s macroglobulinemia, which have been shown to be underreported by PBCR. Nonetheless, during the last years, REDECAN has implemented training programs to improve the coding and registering of hematological neoplasms in non-specialized Spanish registries, following the example of the French Network of Cancer Registries (FRANCIM) (45). In addition, some LNs subtypes, such as mature T-cell and NK-cell neoplasms, are exceptionally rare; therefore, our estimates may be less robust for those entities. Finally, lack on information on disease stage at diagnosis and therapeutic patterns makes interpretation of survival rates difficult. High-resolution studies collecting such data would provide further evidence on the effectiveness of new treatments, as well as to explain geographic variations across Europe (12).

This study presents the most comprehensive population-based analysis of LNs incidence and survival in Spain. It features some useful points for public health authorities and clinicians, such as the evolution of the incidence of several subtypes, and the expected cases in 2023 based on these data. Survival for all cases as a whole improved along the period of study, although rates are still poor for several subtypes. This study also provides information on the quality of the registries, suggesting, with a negative trend of NOS cases, improvements in the registration and/or a more specific diagnosis, although there is still room for improvement.

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

REDECAN Working Group: Girona Cancer Registry – Coordination of hematology neoplasms in REDECAN: RM-G, Montse Puigdemont, Anna Vidal Vila, AS, Andrés Castillo-Bonilla; Tarragona Cancer Registry – Data coordination in REDECAN: MC, AA, CP, and JG; Álava Cancer Registry: AL-d-M, Patricia Sancho, María Luisa Iruretagoyena; Albacete Cancer Registry: Cristina Ramírez; Asturias Cancer Registry: SM, Virginia Menéndez García, Marta Rodríguez Camblor; Bizkaia Cancer Registry: Visitación de Castro, Marta De La Cruz, Joseba Bidaurrazaga; Castellón Cancer Registry: Consol Sabater Gregori, Isabel Sáez Lloret, Ana Vizcaíno Batllés, Xavier Peñalver Herrero; Ciudad Real Cancer Registry: CD-d-C; Cuenca Cancer Registry: AM, Rosario Jimenez Chillarón; Gipuzkoa Cancer Registry: AA; Gran Canaria Cancer Registry: MDR, Emilio De Miguel, María Carmen Gabas; Granada Cancer Registry: MJS, Daysi Yoe-Ling Chang-Chan; Miguel Rodríguez Barranco; La Rioja Cancer Registry: María Isabel Palacios, Enrique Ramalle; Mallorca Cancer Registry: Paula Franch, Patricia Ruiz Armengol, Carmen Sánchez Contador; Murcia Cancer Registry: MDC, Antonia Sánchez Gil, Ricardo José Vaamonde; Navarra Cancer Registry: MG, Eva Ardanaz; Tenerife Cancer Registry: MAA, Leonor Olga Velázquez, María Magdalena Ramos Marrero; Castilla y León Cancer Registry: Pilar Gutiérrez, Rufino Álamo, Lorena Estévez; Registro Español de Tumores Infantiles: Adela Cañete, Elena Pardo, Rafael Peris Bonet; Registro de Tumores Infantiles y Adolescentes de la Comunitat Valenciana: Consol Sabater Gregori, Ana Vizcaino Batllés, Fernando Almela Vich.

RM-G, MS, and CP contributed to the study conception and design. Data collection was performed by CP, AS, MC, MDR, MAA, IS-L, CD-d-C, AIM-N, LS-d-A, AA-A, AL-d-M, MJS, JP, PF, MDC, MG, JG, SM, and RM-G. Data analysis was performed by AA and AS. The first draft of the manuscript was written by CP. All authors contributed to the article and approved the submitted version.

This work was partially funded by the Josep Carreras Leukaemia Research Institute (Grant No.: FIJC1100).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.