Insulin-like growth factor-I (IGF-I) and IGF-binding protein-3 (IGFBP-3) concentrations compared to stimulated growth hormone (GH) in the evaluation of children treated for malignancy

Objective: The aim of this investigation was to evaluate the utility of IGF-I and IGFBP-3 determinations in screening for GH deficiency (GHD) in children previously submitted to treatment for childhood malignancy, Patients and methods: We compared the GH responses to two pharmacological tests (arginine and levo-dopa) with the IGF-I and IGFBP-3 levels in 48 patients (29 boys) who had undergone bone marrow transplantation (BMT) (36 patients) or treatment for a solid cranial tumor (12 patients). Results: 22 patients (45.8%) showed GHD (i.e. GH peak <8 ng/ml in both tests), and only three (13.6%) of the GHD patients had concomitant low IGF-I levels (i.e. -2 SD below the normal mean) and only one (4.5%) an abnormal IGFBP-3 value (i.e. -2 SD below the normal mean). Among the 26 children with normal GH secretion, 21 (80.8%) also showed normal IGF-I and IGFBP-3 levels, three (11.5%) had a concomitant low IGF-I value and two (7.7%) a concomitant low IGPBP-3 value. A significant correlation was found between GH secretion and age at diagnosis (r=0.26, P<0.05), and between IGF-I and IGFBP-3 (r=0.52, P<0.0001), but not between GH and IGF-I or IGFBP-3. Comparing the growth pattern of these patients from diagnosis to the first year after therapy or BMT, we found that while individual height changes did not correlate with the GH peak, a significant correlation was found between height SDS decrease and IGF-I (r=0.31, P<0.05) or IGFBP-3 SDS (r=0.37, P<0.01). Conclusion: Our results indicate that the cut-off of -2 SD for IGF-I and IGFBP-3 was insensitive in screening for GHD. A normal value did not exclude a subnormal GH response to provocative tests and therefore although IGF-I and IGFBP-3 levels may be indicators of the growth pattern, they cannot be used alone as a tool for identifying GHD children after treatment for childhood malignancy.