Radiotherapy, especially at young age, increases the risk for de novo brain tumors in patients treated for pituitary tumors

Pia Burman, André P. Van Beek, Beverly M.K. Biller, Cecilia Camacho-Hubner, Anders F. Mattsson

Research output: Contribution to journalMeeting AbstractAcademic


Background: Excess mortality due to de novo malignant brain tumors was recently found in a national study of patients with hypopituitarism following treatment of pituitary tumors. Here, we examined a larger multi-national cohort to corroborate and extend this observation.

Objective: To investigate the incidence of malignant brain tumors and benign meningiomas in a large cohort of patients treated for pituitary tumors.

Patients and Methods: Between 1994-2012, 8917 hypopituitary patients with a pituitary tumor treated with growth hormone (GH) replacement were followed in KIMS (non-interventional Pfizer International Metabolic Database). For 4936 patients, pituitary tumor treatment was surgery/medical therapy, and in 3227 radiotherapy (RT) was used alone or with surgery. For 754 patients, tumor treatment data were missing. Reference incidence rates of malignant brain tumors (ICD10: C70-C72) stratified for age, gender and country (Cancer Incidence in 5 regions, Vol IX, IARC 2007) were used to calculate standardized incidence ratios (SIR). Risk indicators for 2nd central nervous system neoplasia (new malignant brain tumors and benign meningiomas) were analyzed by multiple Poisson regression.Relative risks (RRs) were based on internal comparisons.

Results: During 53,881 patient-years at risk, 21 malignant brain tumors (SIR 3.92 95% CI: 2.43-6.00), and 27 benign meningiomas were reported. Fourteen malignant tumors occurred in patients treated with RT, SIR 6.50 (95% CI 3.55-10.91) vs 7 in those without RT. The RR of RT vs no RT for malignant brain tumors was 2.43 (95% CI 1.02-6.29), and for meningiomas was 5.43 (95% CI 2.06 -14.35).On average, the risk for malignant brain tumors increased by 5.9% per year of decreasing age at first radiotherapy (95% CI: 1.2% - 10.3%; p=0.01) and for meningiomas by 5.7% (95% CI: 1.6% - 9.6%; p=0.003). Median times from RT exposure to 2nd tumor diagnosis were 20.3, range 3.1-46.1 yrs, and 22.2, range 8.1-39.3 yrs, for malignant tumors and meningiomas, respectively. Meningiomas were more common in females (18 vs 9, RR=2.26, 95% CI: 1.01-5.05), but there was no difference between genders for malignant brain tumors. The underlying pituitary tumor etiology, number of surgeries, and IGF-I SDS on GH treatment did not influence the risk of 2nd tumors.

Conclusion: Radiotherapy is associated with increased risk of developing malignant brain tumors and meningiomas, in particular when given at younger ages. In balancing risks and benefits of RT in the treatment of pituitary tumors, our findings emphasize that special consideration should be given to the age of the patient.
Original languageEnglish
Article numberOR18
JournalEndocrine reviews
Issue number2
Publication statusPublished - Apr-2016
Event98th Annual Meeting of the Endocrine Society - Boston, United States
Duration: 1-Apr-20164-Apr-2016


  • endogenous compound
  • growth hormone
  • somatomedin C
  • brain cancer
  • brain tumor
  • cancer incidence
  • cancer susceptibility
  • controlled study
  • data base
  • exposure
  • female
  • gender
  • human
  • hypophysis tumor
  • ICD-10
  • major clinical study
  • male
  • meningioma
  • radiotherapy
  • remission
  • risk factor
  • standardized incidence ratio
  • statistical model
  • surgery

Cite this