Incidence and Outcomes of Rare Paediatric Non‐hodgkin Lymphomas

Non-Hodgkin lymphomas (NHL) represent approximately 15% of all cancers in patients younger than 21 years. The majority of paediatric NHLs (pNHL) are comprised of lymphoblastic, anaplastic, large B cell and Burkitt lymphomas. The remaining subtypes of pNHL are very heterogeneous and their prognostic information is often extrapolated from older patients. By encompassing a large number of patients with prolonged follow-up, population-based data sets can provide important insight into the incidence and outcomes of rare pNHL that may be unique to this age group. We used data from the National Cancer Institute's Surveillance, Epidemiology, and Endo Results (SEER-18, corresponding to 27·8% of the US population) to determine incidence and outcome of de novo follicular lymphoma (FL; ICD-O-3 9690/3, 9691/3, 9695/3 and 9698/3), mycosis fungoides/primary cutaneous T-cell lymphoma (MF; ICD-O-3 9700/3 and 9709/3), marginal zone lymphoma, non-otherwise specified (MZL-NOS, ICD-O-3 9699/3), angioimmunoblastic T-cell/peripheral (mature) T-cell lymphoma (PTCL, ICD-O-3 9702/3 and 9705/3), primary cutaneous CD30-positive T-cell lymphoproliferative disorder (PCLPD, ICD-O-3 9718/3), extranodal Natural Killer/T-cell lymphoma (ENKL, ICD-O-3 9719/3), and hepatosplenic (γ/δ)T-cell lymphoma (HSTCL; ICD-O-3 9716/3), diagnosed between 2001 and 2012 in patients younger than 21 years. Individuals of all races were included. We excluded cases reported from death certificate or autopsy only. Age-adjusted incidence rates per 1 000 000 persons and 95% confidence interval (95% CI) were calculated using the Rate Session in SEER*Stat Version 8.1.5 (http://seer.cancer.gov/seerstat/). All incidence rates were adjusted for the 2000 US standard population. Overall survival (OS) rates at 5 years post-diagnosis were estimated using the Kaplan–Meier method. Greenwood variance estimates were used to construct corresponding 95% CIs. For survival analysis, all deaths were considered events, irrespective of cause and subjects were censored at the last time they were reported alive. Due to the nature of the SEER database, information regarding systemic treatment, including use of haematopoietic stem cell transplant (SCT) was not available. Two-sided P values <0·05 were considered to indicate statistical significance. A total of 552 cases of rare pNHL met entry criteria for the study, corresponding to 17·5% (n = 3158) of all pNHL cases reported between 2001 and 2012. The incidences varied from 0·05 to 0·54 cases/million population*year for the different subtypes of pNHL (Table 1). Previous reports using SEER or Southern and Eastern European (SEE) registries did not specifically detail incidence by morphological subtypes of pNHL (Georgakis et al, 2016; Karalexi et al, 2017), but they support significant variation worldwide. For instance, certain subtypes, such as ENKL or PCLPD, are peculiarly more frequent in Asian countries or in individuals of Pacific Island origin (Swerdlow et al, 2008). The incidence of cutaneous forms may be underestimated because those cases are often misdiagnosed with other skin conditions. Additional epidemiological studies are needed to confirm our findings. As in the most widespread forms of pNHL, the majority of rare pNHL histologies affected predominantly males (Table 1) (Scheurer et al, 2016). The proportion of affected females was similar to males only among patients with cutaneous forms of pNHL (MF and PCLPD). The median age at presentation among all pNHL subtypes varied from 12 to 17 years (Table 1). The mechanisms for the observed gender disparities and increased incidence among adolescents are probably multifactorial and beyond the scope of this article. An important contribution of the current study was to demonstrate contemporary data on outcomes of rare forms of pNHL (Fig 1). We confirmed that patients with B cell NHL subtypes (MZL and FL) have an excellent survival. For MZL patients, the OS remained very high at 10 years (5-year and 10-year OS 96·2 ± 2·2%). That was not the case for patients with FL (5-year OS 96·2 ± 2·2% and 10-year OS 89 ± 5·3%). This is in contrast to previous reports describing OS of 100%, but over shorter follow-up periods (O'Suoji et al, 2016). Our data suggest that, as in adults, paediatric FL has an indolent course and patients may experience recurrences or transformation to more aggressive lymphomas with potential impact in long term survival (Swerdlow et al, 2008). Paediatric patients with rare systemic forms of T cell NHL (PTCL, ENKL and HSTCL) had more guarded prognoses (Fig 1). In this category, PTCL patients had the best outcome (5-year OS 59·0 ± 5·1% and 10-year OS 56·3 ± 5·5%), which was similar to a European cohort of paediatric PTCL patients (Mellgren et al, 2016). In children, the outcome is suggested to be dependent on disease staging (Hutchison et al, 2008; Windsor et al, 2008). In contrast, the majority of adult patients present with advanced disease and have much inferior outcomes (Mahadevan et al, 2013). Both ENKL and HSTCL are exceedingly rare in the United States. The overall prognosis of ENKL for all stages was poor, with a 5-year OS of 47 ± 11·2%. Although information regarding long term OS for paediatric patients with ENKL is lacking in the literature, response to therapy in children may be similar to adults (Qi et al, 2016). Patients with HSTCL had dismal outcomes. In this series, the 5-year OS was only 9·6 ± 8·8%, confirming previous reports (Mellgren et al, 2016). This study was met by some limitations. Unfortunately, the SEER registry provides no information regarding therapeutic interventions, other than surgical resection or radiation therapy. Additionally, there is no central review of pathology, staging, response assessment or clinical factors, such as serum level of lactate dehydrogenase or others. In summary, this large and contemporary registry-based analysis confirms that rare subtypes of paediatric B cell NHL and cutaneous T cell lymphomas have excellent outcomes. The issue remains for patients with rare subtypes of systemic T cell pNHL. Those patients still face dismal outcomes that are comparable to their adult counterparts. Therefore, these patients constitute an unmet medical need and require new therapies that aim to improve their outcomes. M. Cairo receives funding support from the Pediatric Cancer Research Foundation (PCRF) and St. Baldrick Foundation. C. Sorge designed the study, collected and analysed the data, wrote the first draft of the manuscript and approved its final version. L.J. Costa, J.W. Taub and M. Cairo designed the study, significantly edited the manuscript and approved its final version. A.C. Xavier designed the study, collected and analysed the data, significantly edited the manuscript and approved its final version. The authors have not conflict of interest to disclose.