Radiation therapy to the pancreas during childhood confers an increased risk of diabetes during adulthood, according to findings from a retrospective cohort study involving more than 2,500 survivors of childhood cancer.
Specifically, a dose-response relationship was seen between radiation to the tail of the pancreas – where the islets of Langerhans are concentrated – and subsequent diabetes development, reported Florent de Vathaire, Ph.D., of the Center for Epidemiology and Public Health of INSERM at Gustave Roussy Institute in Villejuif, France, and colleagues.
Courtesy Dr. Florent de Vathaire
Dr. Florent de Vathaire
The findings are published online in the Aug. 23 issue of The Lancet Oncology.
Of 2,520 survivors of childhood cancer who returned a survey for the study, 1,632 received radiotherapy during childhood, and 65 had verifiable diabetes, which the authors said was likely type II in most cases. The cumulative incidence of diabetes at age 45 years was significantly greater in those who received radiation therapy (6.6% vs. 2.3%), and those who received radiotherapy to the tail of the pancreas at a dose of 10 Gy or more during childhood were significantly more likely to develop diabetes than were those who did not receive radiotherapy (cumulative incidence of 16.3%).
This relationship persisted even after adjustment for body-mass index (relative risk of 12.6 at a mean radiation dose of 24.2 Gy), the investigators said.
The risk, which was similar in men and women, increased strongly in a dose-dependent fashion up to 20-29 Gy, reaching a plateau at higher doses, they noted (Lancet Oncology 2012 Aug. 23 [doi:10/1016/S1470-2045(12)70323-6]).
No association was found between radiotherapy to other parts of the pancreas and subsequent diabetes development.
Of note, diabetes was diagnosed in 3% of 739 patients who were younger than age 2 years at diagnosis, and the increase per Gy was higher in these patients, compared with older patients. Also, diabetes incidence strongly varied based on the type of childhood cancer, with a cumulative incidence of diabetes at age 45 of 14.7% in survivors of nephroblastoma, compared with 3.1% in survivors of other cancers.
Although these variations by cancer type were explained by differences in age at radiotherapy and by radiation dose to the pancreas tail, patients with nephroblastoma comprised the majority of patients who received high radiation doses to the tail of the pancreas, which raises uncertainty about the attribution of cause, the investigators noted.
No evidence of a significant role for chemotherapy in the calculation of diabetes risk or for chemotherapy overall acting as a modifier of the dose-response for radiation was found in this study, they said.
For the study, the investigators surveyed survivors of childhood cancer – including solid cancer or lymphoma, but excluding leukemia – who were treated in eight centers in France and the United Kingdom before 1986 and followed for a mean of 30 years. Radiation doses to the tail, body, and head of the pancreas were estimated using mathematical modeling, details from the patients’ records, and information about equipment, treatment techniques, and guideline used at the time of treatment.
Though limited by factors such as the high proportion of survivors not included in the analysis and inherent difficulties with estimating radiation dose, the findings of the specificity of the radiation dose to the tail of the pancreas is "plausibly explained by the fact that the islet of Langerhans concentration is higher in the tail than in the body and head of the pancreas," they noted.
"Our investigation emphasizes the importance of long-term follow-up of childhood cancer survivors; almost no diabetes mellitus was seen in our cohort, or those of others, before 20 years of follow-up," they said, noting that the findings also underscore the need for contouring the pancreas when planning radiation therapy to achieve the lowest possible radiation dose to the organ.
This study was funded by Ligue National Contre le Cancer, Institut de Recherche en Santé Publique, Programme Hospitalier de Recherche Clinique, Institut National du Cancer, Agence Française de Sécurité Sanitaire et des Produits de Santé and Fondation Pfizer pour la santé de l’enfant et de l’adolescent. The authors reported having no conflicts of interest.