Objective: to assess the potential risk of cancer and its risk factors in paediatric transplant recipients.

Design: analytical retrospective longitudinal observational study conducted in the framework of the US Transplant Cancer Match (TCM) Study.

Setting: the data were retrieved from the Scientific Registry of Transplant Recipients (SRTR) database, which has records of all solid organ transplantations performed in patients aged less than 18 years in 16 states of the United States between 1987 and 2011.

Study population: out of 18 150 transplantations in patients aged less than 18 years included in the TCM Study (45% of all solid organ transplantations in the United States), 17 958 were selected, corresponding to 16 732 individuals, and 1% were excluded because the recipients did not belong to any of the included race categories.

Risk factor assessment: the unit of analysis was transplantation of a solid organ. The authors assessed demographic and clinical patient characteristics, the indication for transplantation, patient EBV serology status before transplantation and the race of transplant recipients.

Outcome measurement: the primary endpoint was the development of cancer classified according to the third edition of the Classification of Diseases for Oncology. At-risk time spanned from the transplant date or the start of cancer registry coverage to the first of the following events: death, graft failure, retransplantation, loss to followup or end of cancer registry coverage. Followup extended beyond 18 years of age. The authors summarised the results by means of standardised incidence ratios (SIRs) calculated by the direct method (dividing observed cancer counts among recipients by expected counts based on the general population rates) for total cancer and for specific types of cancer. Since non-Hodgkin lymphoma (NHL) was the most frequent type of cancer, the authors also calculated specific SIRs for NHL by age, transplanted organ, time to development of cancer and EBV status. The authors performed multivariate Cox regression to examine associations with potential risk factors.

Main results: of the 18 150 selected transplant recipients, 17 958 were included in the analysis (1% excluded). Of the total, 54.3% were male, 54.5% were aged less than 9 years, and 53.8% were Caucasian, and the median duration of followup was 4 years (interquartile range [IQR]: 1-7 years). The EBV status was known in 48% of the patients, of who 46% were seronegative. A total of 392 cancers were diagnosed, with a SIR of 212.95% (95% confidence interval [95 CI]: 188 to 238). The most frequent type was NHL (SIR: 18.5%; 95 CI: 13 to 26), with an increased incidence relative to the general population independently of age, years since transplantation and EBV status. The risk factors for NHL identified in the multivariate analysis were first year after transplant, EBV seronegative status, transplant type (heart, lung and intestine), and duration of immunosuppression, although there were no significant differences based on age, sex or race.

Conclusion: paediatric transplant recipients are at higher risk of developing cancer compared to the general population. The most frequent cancer type was non-Hodgkin lymphoma, with a higher risk in the first year post transplant, in recipients susceptible to primary infection by EBV and in intestine transplant recipients.

Conflicts of interest: one of the authors is employed by Grail (a company dedicated to the early detection of cancer), the rest declared no conflicts of interest.

Funding source: none noted.

Justification: in recent years, there has been an increase in the number of transplantations in Spain and worldwide. In Spain, 4821 organ solid transplantations were performed in 2016, and 136 were performed in recipients aged less than 15 years in 2015, which amounts to an annual rate of 3 transplantations per million individuals, above the European average.¹ Previous studies have described an increased risk of tumour development in paediatric transplant recipients.² Since the life expectancy of these patients has increased and thus there is more time for malignancy to occur, studying the conditions under which tumours develop is evidently relevant in order to establish preventive measures, where possible, or even to plan the followup of patients.

Validity or scientific rigour: the population was well defined, as were the predictor variables. The method for data collection was appropriate, the diagnoses were confirmed, and only 1% of the population was excluded. The time at risk of cancer was well defined. The authors used standardised incidence rates based on general population registries and assessed the independent effect of different risk factors on tumour development in transplant recipients with a Cox regression model. The median duration of followup was 4 years and the median time elapsed between transplantation and tumour development was 2.5 years, so the duration of followup seems sufficiently long, although it is not possible to know the long-term incidence of tumours in this group of patients or the incidence of tumour types for which the expected number of cases during the followup was less than 1.

Clinical relevance: during the followup, the incidence of cancer in transplant recipients was 19 times greater than the one in the general population (SIR: 19.1 [95 CI: 17.3 to 21.1]), with a predominance of NHL, whose incidence was 212 times that in the general population, and especially high in patients aged less than 5 years, intestine transplant recipients, in the first year post transplantation and in EBV-seronegative recipients, with the incidence of NHL in the latter group being 2.7 times greater than the incidence in EBV-seropositive recipients (hazard ratio [HR]: 2.71; 95 CI: 1.82 to 4.05). Excluding NHL, the incidence of tumours was 6 times greater compared to the general population (SIR: 5.89; 95 CI: 4.86 to 7.08). The effect size in tumour development seems considerable, as does the increased incidence in EBV-seronegative recipients. Although this effect would not in principle alter the decision to perform transplantation, it may be key in shaping the followup plan for these patients. The cancer incidence in transplant recipients was similar to those published in previous studies,^2,3 as was the increased incidence of tumours in solid transplant recipients that were EBV-seronegative; in the future, the latter patients may be eligible for vaccination which, while not currently available, may become a beneficial intervention.^4-6

Applicability to clinical practice: the paediatric sample in this study is the largest analysed to date. It describes the increased risk of developing certain tumours compared to the general population, and suggests that EBV infection is a key risk factor, especially for the development of NHL, so in the future vaccination against this virus may play an important role in preventing infection at a later date, and perhaps become an effective preventive measure.

Conflicts of interest: the authors of the commentary have no conflicts of interest to declare.

Martín Masot R, Ortega Páez E. Pediatric solid organ transplantation is a risk factor for future cancers. Evid Ped. 2017;13:53.

1. Newsletter Transplant. International figures on donation and transplantation 2015. EDQM. 21. 2016; 21: 38.
2. Simard JF, Baecklund E, Kinch A, Brattström C, Ingvar A, Molin D, et al. Pediatric organ transplantation and risk of premalignant and malignant tumors in Sweden. Am J Transplant. 2011;11:146-51.
3. Webster AC, Craig JC, Simpson JM, Jones MP, Chapman JR. Identifying high risk groups and quantifying absolute risk of cancer after kidney transplantation: a cohort study of 15,183 recipients. Am J Transplant. 2007;7:2140-51.
4. Dharnidharka VR, Lamb KE, Gregg JA, Meier-Kriesche HU. Associations between EBV serostatus and organ transplant type in PTLD risk: analysis of the SRTR National Registry Data in the United States. Am J Transplant. 2012;12:976-83.
5. Cohen JI. Epstein–Barr virus vaccines. Clin Transl Immunology. 2015;4:e32.
6. Schober T, Framke H, Kreipe Schulz TF, Großhennig A, Hussein K, et al. Characteristics of early and late PTLD development in pediatric solid organ transplant recipients. Transplantation. 2013;95: 240-6.