ESPE Yearbook of Paediatric Endocrinology

To read the full abstract: J Clin Endocrinol Metab. 2018 Aug 1; 103(8):2785

GH deficiency (GHD) is common among children treated for cancer, especially among childhood cancer survivors (CCS) with tumors/surgery in the hypothalamic-pituitary (HP) region, CCS exposed to HP radiation or CCS exposed to cranial (CIR), craniospinal (CSI), or total body irradiation (TBI). However, most of existing data on GHD diagnostic work-up derive from people who were not CCSs, despite suspected differences related to the specific nature of the radiation-induced HP axis damage. Therefore, an Endocrine Society taskforce was charged to develop guidelines on the management of growth disorders in CCSs. This is a preliminary report produced by the taskforce, aiming to evaluate the existing data on GHD screening in CCS by the use of IGF-1 or IGFBP-3 measurements, compared with GH provocative tests and diagnosing GHD by using different GH dynamic tests.

Fifteen studies were analyzed [8 studies examined IGF-1 and 7 studies analyzed IGFBP-3]; 477 patients were included. Overall, both IGF-1 and IGFBP-3 showed poor diagnostic accuracy. The studies were remarkably heterogeneous; when calculations were possible, the sensitivity and specificity of IGF-1 varied from 47% to 66% and 77% to 100%, respectively. IGFBP-3 had a lower sensitivity (20%). IGF-1 and IGFBP-3 were strongly correlated and the simultaneous use of both tests did not improve the diagnostic accuracy. Provocative tests remained the most accurate tools to identify GHD in CCS, despite remarkable variability in the testing protocols and the absence of standardized testing strategies. The insulin tolerance test (ITT) seems to be the most accepted reference test, when used alone or in combination with arginine. The ability of GHRH stimulation, with or without arginine, to diagnose GHD was equivocal across different studies; in one study GHRH with arginine stimulation had 66% sensitivity and 88% specificity compared to ITT. Insufficient data were available to assess the accuracy of serial GH testing (nocturnal or over 24 hours).

Evaluating the GH axis in CCSs allows an opportunity to treat these patients with available and effective replacement therapy. The controversy regarding the benefit-to-risk ratios of such therapies highlights the importance of appropriate patient selection and accuracy of GHD testing. According to this systematic review, dynamic tests are still the cornerstone of GHD diagnosis. Additional research is needed to establish the best provocative test for CCS. In the meantime, reliance on the ITT (as the gold standard) seems to be appropriate, even if this test is not feasible at many institutions. The use of GHRH with arginine stimulation test should be limited in CCS. The pituitary gland is located outside the blood-brain barrier and intravenous GHRH infusion may act as a maximal direct stimulus on the somatotropic cells. In CCS patients, this test can elicit a high and falsely normal GH response, because it does not investigate the true spontaneous function of the HP axis damaged by primary tumour, surgery and/or therapeutic irradiation. CCSs probably need a lower testing threshold because of their established cranial pathologies and/or HP radiation. In the presence of low height velocity associated or not to concurrent HP defects, a blunted GH response to any of the provocative tests should be considered sufficient to establish GHD diagnosis, due to the high pre-test diagnostic suspicion.

On the other hand, a direct bone damage with skeletal growth impairment independently of GHD, due to treatment with agents such as TBI, imatinib, and cis-retinoic acid, may contribute to poor linear growth, short adult stature and/or altered body proportions.