Copy Number Variants Contributing to Combined Pituitary Hormone Deficiency

Round 1

Reviewer 1 Report

This is a straightforward copy number variation (CNV) analysis conducted for 32 patients with multiple pituitary hormone deficiencies who were negative for diagnosis by whole exome sequencing.  The authors did not find any large (300 kb) CNV throughout the genome.  Their secondary filtering analysis examined CNV in 353 genes expressed in the pituitary gland and known hypopituitarism genes.  They identified CNV in 5 genes in 9 patients.  Thus, 28% of the patients had CNV that could contribute to disease.  4 of the genes are known hypopituitarism genes: HESX1, LHX4, OTX2 and SIX3. To date, no CNV have been reported in these genes.  Thus, it is speculative to conclude that overdosage of these genes causes disease.  The authors report an interstitial deletion in ASH1L, a gene for which loss of function of one allele is associated with autism and intellectual disability.  The clinical description of the affected patients should include information about this as well as craniofacial abnormalities, vision defects, optic nerve hypoplasia etc. as other cases of mutations in HESX1, OTX2 and SIX3 often carry such features.  While no functional studies were carried out, it is notable that the yield of CNV was high and suggests future studies of hypopituitarism patients should include CNV detection.

1. Please provide additional clinical description.

2. Figure 2

• This figure was truncated in the pdf such that I could not evaluate the results from patients 9, 11 or 12.
• It is difficult to see the genes and discern where one gene ends and another begins, i.e. HESX1 and DNAH12. This is important for determining whether a gene is duplicated in its entirety or contains an interstitial disruption.

3. Conclusions should be made with the caveat that identification of additional CNV in CPHD patients or functional studies are necessary to determine whether the CNV reported here are causal.

Author Response

This is a straightforward copy number variation (CNV) analysis conducted for 32 patients with multiple pituitary hormone deficiencies who were negative for diagnosis by whole exome sequencing.  The authors did not find any large (300 kb) CNV throughout the genome.  Their secondary filtering analysis examined CNV in 353 genes expressed in the pituitary gland and known hypopituitarism genes.  They identified CNV in 5 genes in 9 patients.  Thus, 28% of the patients had CNV that could contribute to disease.  4 of the genes are known hypopituitarism genes: HESX1, LHX4, OTX2 and SIX3. To date, no CNV have been reported in these genes.  Thus, it is speculative to conclude that overdosage of these genes causes disease.  The authors report an interstitial deletion in ASH1L, a gene for which loss of function of one allele is associated with autism and intellectual disability.  The clinical description of the affected patients should include information about this as well as craniofacial abnormalities, vision defects, optic nerve hypoplasia etc. as other cases of mutations in HESX1, OTX2 and SIX3 often carry such features.  While no functional studies were carried out, it is notable that the yield of CNV was high and suggests future studies of hypopituitarism patients should include CNV detection.

1. Please provide additional clinical description.

We updated clinical description of patients. We agree with reviewer that accompanying clinical features like craniofacial abnormalities, vision defects, optic nerve hypoplasia often is seen in patients presenting pathogenic loss of function mutations of HESX1, OTX2 and SIX3. We did not note such abnormalities in our patients.

2. Figure 2

This figure was truncated in the pdf such that I could not evaluate the results from patients 9, 11 or 12.

The figure was loaded separately to the system due to the requirements of quality checks (high-resolution). It was possibly automatically cropped during the file generation. In the present version, provided by the editor Figure was fitted to the page. 

It is difficult to see the genes and discern where one gene ends and another begins, i.e. HESX1and DNAH12. This is important for determining whether a gene is duplicated in its entirety or contains an interstitial disruption.

We modify the coloring of the genes (gene names with corresponding genomic span) so it is now clear where gene starts and ends. We also added a flanking dashed line determining a position of abnormality for a rough estimation. Unfortunately, Affymetrix software (ChAS) is not offering any options to adopt and customize microarray plot to human genome. For precise breakpoints determination, we provided raw genomic coordinates in Table 1 (HGVS column).      

3. Conclusions should be made with the caveat that identification of additional CNV in CPHD patients or functional studies are necessary to determine whether the CNV reported here are causal.

We also agree with the reviewer on this matter. Therefore, a clear statement was added at the end of the conclusions (Page 8, Lines 220-223).

On behalf of authors, I would like to thank the Reviewer for time spent on constructive criticism, that significantly improved our paper

Reviewer 2 Report

This is a well conducted study that includes a catalogue of information that will be of interest to those studying the genetics of CPHD and provide new targets for future research. Without functional studies, the implications of the copy number variation found is not clear, as is the mechanism underlying altered pituitary functions. However, as a source of inspiration for further study of previously uncharacterised genes the paper is a valuable resource.

Author Response

On behalf of the authors, I would like to thank the Reviewer for reading our paper and a positive opinion