Whole Exome Sequencing Uncovered the Genetic Architecture of Growth Hormone Deficiency Patients.

Purpose Congenital growth hormone deficiency (GHD) is a rare and etiologically heterogeneous disease. We aim to screen disease-causing mutations of GHD in a relatively sizable cohort and discover underlying mechanisms via a candidate gene-based mutational burden analysis. Methods We retrospectively analyzed 109 short stature patients associated with hormone deficiency. All patients were classified into two groups: Group I (n=45) with definitive GHD and Group II (n=64) with possible GHD. We analyzed correlation consistency between clinical criteria and molecular findings by whole exome sequencing (WES) in two groups. The patients without a molecular diagnosis (n=90) were compared with 942 in-house controls for the mutational burden of rare mutations in 259 genes biologically related with the GH axis. Results In 19 patients with molecular diagnosis, we found 5 possible GHD patients received known molecular diagnosis associated with GHD (NF1 [c.2329T>A, c.7131C>G], GHRHR [c.731G>A], STAT5B [c.1102delC], HRAS [c.187_207dup]). By mutational burden analysis of predicted deleterious variants in 90 patients without molecular diagnosis, we found that POLR3A (p = 0.005), SUFU (p = 0.006), LHX3 (p = 0.021) and CREB3L4 (p = 0.040) represented top genes enriched in GHD patients. Conclusion Our study revealed the discrepancies between the laboratory testing and molecular diagnosis of GHD. These differences should be considered when for an accurate diagnosis of GHD. We also identified four candidate genes that might be associated with GHD.