TRPS I is an infrequently diagnosed cause of short stature that is associated with subtle dysmorphia. It displays clinical heterogeneity, even within a family. In the case reported here, although the diagnosis was apparent in retrospect, it was not until the findings of 3 affected siblings were combined that the diagnosis was made.
Genetics and pathophysiology of TRPS I
Consistent with the TRPS I phenotype, cartilage and hair follicles are among the limited tissues that express the gene. It is postulated that TRPS1 (the protein product of
TRPS1) deficiency impairs endochondral cartilage differentiation and epithelial/mesenchymal cell interactions in developing hair follicles [
15]. Mechanistically, it is hypothesized that TRPS1 deficiency induces mitotic arrest in prometaphase cells as a result of abnormal chromatin condensation caused in part by excessive histone deacetylation [
16]. Normally, TRPS1 homodimers complex with GATA binding protein sequences in DNA to repress transcription of target genes, an effect that is suppressed when TRPS1 is complexed with either the dynein light-chain LC8 or the ring finger protein RNF4.
Mutation analysis of
TRPS1 has revealed that missense mutations restricted to the GATA DNA-binding domains (aa 886–952) caused the more severe musculoskeletal anomalies associated with TRPS III. In 2 cases where missense mutations occurred outside the GATA zinc finger and in all reported nonsense single base pair mutations, the milder TRPS I phenotype arose [
17] as in our case. The previously unreported mutation in this family is predicted to stop translation between the 7
th and 8
th zinc finger domains of TRPS1. In theory, translation of the nuclear translocation signal persists, but the mutation prevents extension of TRPS1 to the 2 Ikaros family zinc fingers at the protein’s carboxy terminus. These are necessary for repression of GATA-driven transcriptional signals [
18]. Although the mutation also interrupts the RNF4 binding site between amino acids 985 – 1184, this change should be inconsequential in the absence of the Ikaros domains.
Ludecke et al. reported the average height of 75 patients with TRPS to be 1.41 SD below the average height of the respective population (SD = 1.15). Their data included children and adults with TRPS I and III [
10]. Four previous reports of growth hormone treatment for short stature in patients with TRPS I have yielded conflicting results (Table
2). Naselli et al. first reported no growth promotion by GH in monozygotic twins with TRPS I [
11]. Although the report indicated both children had GH deficiency, the actual GH stimulation tests results were not included. Prior to the start of growth hormone, both girls were prepubertal and had bone age delays of 2–3 years. Their treatment was discontinued after 12 months owing to its ineffectiveness.
Table 2
Summary of the reports on GH axis evaluation and treatment in patients with TRPS I
| Patient 1 (F) | Not reported | Deficient, based on GH stim test and IGF1 level, no data | 11.1 y | 8–9 y | Prepubertal | 0.23 | None |
Patient 2 (F) | Not reported | Deficient, based on GH stim test and IGF1 level, no data | 11.1 y | 8–9 y | Prepubertal | 0.23 | None |
| Patient 3 (M) | Not reported | Partial deficiency, Peak after arginine 6.8, after insulin 12.7, low nocturnal GH 2.7 | 12 y | 9 y 6 m | G2, PH2, T4 ml bilaterally | 0.26 | + 0.7 SDS over 5 years |
Patient 4 (M) | c2722C > T (p.R908X) | Partial deficiency. Peak after Clonidine 10.2, peak after insulin5.4, low nocturnal GH 2.38 | 9 y 9 m | 7 y8 m | Prepubertal | 0.26 | + 1.9 SDS over 7 years |
Sarafoglou et al. 2010 [ 14] | Patient 5 (M) | Not reported | No deficiency. Low IGF1 and normal IGFBP3 | 7 y | ~3 y delay | Prepubertal | 0.3–0.43 | +1.81 SDS over 3 years |
Patient 6 (F) | Not reported | No deficiency. Low IGF1 and normal IGFBP3 | 6.95 y | ~ 6 m delay | Prepubertal | 0.34–0.54 | +1.95 SDS over 2 years |
| Patient 7 (F) | c2520dupT (p.Arg841LysfsX3) | Deficient (peak after insulin 3.17 and after L-dopa 5) | 4 y | 2.6 y | Prepubertal | 0.2 | + 0.4 SD over 10 years |
Patient 8 (M) | c1630C > T (p.Arg544X) | Not deficient (peak after glucagon 9.86, after L-dopa 9.7) | 14 y | 16 y | Pubertal | Not reported | None (1 cm over 6 months) |
Sohn et al. also reported 2 Korean patients with TRPS I who were unresponsive to GH treatment [
12]. One was a 4 year old girl who was GH deficient based on stimulation testing. Despite a 10 year course of GH (0.2 mg/kg/week), her height SDS response to treatment was not impressive as she only gained 0.4 SDS change in her height over that time. The other was a 14 year old boy with a bone age of 16 years and a peak circulating GH concentration of 9.9 ng/mL after receiving glucagon. He did not show any growth velocity response after a 6 month GH trial, a circumstance that is expected with his advanced bone age of 16 years.
Stagi et al. [
13] described 2 GH-responsive patients with TRPS I who had normal GH responses to provocative testing [
13]. The authors however considered the patients to both have partial growth hormone deficiency based on low nocturnal mean GH concentrations. One had a height SDS of −2.2 and a bone age of 9.5 years at age 12 years. After 5 years of GH treatment, his adult height was −1.5 SDS, well above his mid-parental target height of −3.5 SDS. The other patient had a height SDS of −2.0 and bone age of 7 years 8 months at age 9 years 9 months. His growth velocity doubled during the first year of treatment and his final height was −0.1 SDS. The rate of bone age progression during GH treatment was not reported for either patient and unfortunately, neither experienced improvements of their pre-treatment osteopenia.
Sarafoglou et al. [
14] described accelerated growth and improved bone mineral density after GH treatment in 2 siblings with TRPS I [
14]. A 6.3 year old male had a height SDS of −2.7, weight SDS −3.6, pre-treatment growth velocity of 2.5 cm/year, and bone age of 3.5 years. In the face of normal GH provocative testing, his serum IGF-1 concentration of 81 ng/mL was considered to indicate GH resistance, notwithstanding an IGFBP3 concentration of 1.7 mcg/mL. Growth hormone (0.3 – 0.43 mg/kg/week) for 3.2 years resulted in an increased height SDS to −1.37, weight SDS −1.06, mean growth velocity of 6.7 cm/year and bone age progression to 8 years. This patient’s sister had a height SDS of −2.76, weight SDS of −3.35 at 6.3 years, pre-treatment growth velocity of 3.9 cm/year, and bone age of 5.75 years. She too was designated as GH resistant on the basis of normal provocative GH testing, serum IGF1 concentration of 81 ng/mL, and serum IGFBP3 concentration of 3.5 mg/L. Over 3 years of GH treatment, her height SDS rose to −1.1 and her bone age increased to 7.8 years. The IGF1 concentrations of both children rose by 2 to 5 fold over their baseline levels while they received exogenous GH doses of 0.3 to 0.5 mg/kg/week. Improved BMD was reported with GH treatment, but only in the female patient.
Combined with the family reported here, the literature allows a series of questions to be addressed.
Is TRPS I associated with GH resistance?
Although Sarafoglou et al. claimed their patients both had GH resistance, the baseline serum IGF1 SDS scores were −1.5 in both patients and these were accompanied by normal IGFBP3 concentrations. With weight SDS scores significantly below the height SDS scores, one should consider the possibility that malnutrition in the baseline state may have contributed to selective depressions of IGF-1. Nutritional counseling was not addressed in the report. Finally, the marked increase of circulating IGF1 after exogenous GH administration is inconsistent with a GH resistant state. The normal pre-treatment serum IGF1 and IGFBP3 concentrations in the children reported here and their responses to exogenous GH administration are also evidence against a growth hormone resistant state.
Is short stature in TRPS I GH responsive?
Increased growth velocity and final height after GH treatment are well described in short children with normal function of their GH-IGF1 axis. The patient reported here increased his growth velocity by 63 percent over the 2 years of treatment. This compares to a mean change in growth velocity of 66–168 percent among the responding patients in the previously published reports. The previous reports may indicate that the early initiation of GH treatment is associated with better height outcomes. Similar observations were reported after GH treatment in children with GH-sufficient, idiopathic short stature [
22]. Sohn et al. however reported suboptimal growth in a girl with TRPS I, despite starting a 10 year course of GH supplementation at 4 years of age [
12]. Factors contributing to her poor response may include the low GH dose (0.2 mg/kg/week), her midparental height, and the adherence to her recommended treatment. A dose dependent effect may be reflected in better outcomes reported by Sarafoglou et al. [
14] in 2 children receiving GH doses of 0.3 to 0.5 mg/kg/week.
The mechanism by which growth hormone therapy could accelerate linear growth in children with TRPS I is unknown and could be complex. In a cell culture model that may mimic TRPS1 mutations however, IGF-1 expression by a chondrogenic cell line derived from a murine teratoma (ATDC5) was reduced by blockade of TRPS1 expression with microRNA [
23]. It is therefore possible that high systemic IGF-1 concentrations, resulting from growth hormone therapy, compensate for low local IGF-1 concentrations in the growth plates of individuals with TRPS1 mutations.
These cases also incidentally indicate the need to consider whether routine evaluation of bone mineral density is indicated in patients with TRPS I and whether it may be increased by GH treatment.