Jump to comment:

• Caution with genotype-phenotype conclusions in 22q11 Deletion Syndrome
  Anne S. Bassett
  Published on: 27 April 2016

• Published on: 27 April 2016

  Caution with genotype-phenotype conclusions in 22q11 Deletion Syndrome

  • Anne S. Bassett, Psychiatrist
  • Other Contributors:
    • Eva W.C. Chow, and Rosanna Weksberg

  Dear Editor,

  The conclusions of Rauch et al. [1] with respect to positive genotype-phenotype correlations in 22q11 Deletion Syndrome (22qDS) must be viewed with caution. They report on extensive fluorescence in situ hybridization (FISH) studies of 350 patients with features of 22qDS ascertained from 3 sources. Based on a case series of 3 subjects found to have distal (atypical) deletions that would not have been detecte...

  Show More

  Dear Editor,

  The conclusions of Rauch et al. [1] with respect to positive genotype-phenotype correlations in 22q11 Deletion Syndrome (22qDS) must be viewed with caution. They report on extensive fluorescence in situ hybridization (FISH) studies of 350 patients with features of 22qDS ascertained from 3 sources. Based on a case series of 3 subjects found to have distal (atypical) deletions that would not have been detected using commonly used clinical probes (TUPLE1 or N25), they draw provocative conclusions about genotype-phenotype associations. The methods used and data presented do not justify the generalizations made, however.

  We have calculated 95% confidence intervals (CIs) (see Table 1.) for the Rauch et al. data presented since percentages alone may be misleading. CIs allow for an assessment of statistical significance while accounting for sample size and rarity of events.[2] All of these CIs are non-significant, that is contain expected values, such as ~18% for positive clinical FISH in conotruncal congenital heart defect (ctCHD).[3] In addition, the 95% CI for finding three distal deletions in 350 subjects (0.86%) is 0.18-2.5%, which overlaps the 0-5% frequency reported in the literature, as would the results if only the sample of genetics referrals with features suggestive of 22qDS were used as the denominator: 3/77 (3.9%, 95% CI 0.8-8.3%).

  To support their assertion of a genotype-phenotype correlation, the authors compared the frequency of atypical distal deletions in a sub-sample (n=63) from the 77 genetics referrals to that in 3 other samples. However, using a selected partial sample prevents a proper interpretation of results, and limits applicability to other populations. If the 22qDS phenotype were the issue, the 77-subject genetics referral sample should be used, since these subjects had multiple features consistent with 22qDS, as compared to those with a single feature such as ctCHD. On the other hand, if the issue were whether "typical" 22qDS facial features were related to atypical distal deletions then the relevant comparison should be between the 14 subjects with "typical" 22qDS facies and the 63 without these facies from the same ascertainment source. As illustrated in Table 2, the frequency of distal deletions is significantly different (p<_0.05 no="no" correction="correction" for="for" multiple="multiple" comparisons="comparisons" in="in" only="only" _3="_3" non-overlapping="non-overlapping" all="all" of="of" which="which" involve="involve" deletions="deletions" distal="distal" to="to" the="the" _3mb="_3mb" region.="region." conclusion="conclusion" one="one" can="can" draw="draw" from="from" these="these" results="results" is="is" that="that" a="a" _22q11.2="_22q11.2" deletion="deletion" more="more" likely="likely" be="be" found="found" subjects="subjects" with="with" than="than" feature="feature" _22qds.="_22qds." _="_" p="p"/>

  The assessment of atypicality of phenotypic expression is challenging in the phenotypically diverse condition 22qDS. Terms such as "atypical" should be used with caution as phenotype will vary with age, completeness of assessment, assessors, and ascertainment.[4] Mild learning difficulties may not be noticeable until later childhood, and absence of psychosis could not be truly determined until well into adulthood (in contrast to Table 3. of Rauch et al. where infants are indicated to have no psychosis). Hyperactivity is a common feature in children with 22qDS.[5] The sister from the affected sibpair with two different length 22q11.2 deletions may thus be considered to have typical neurobehavioural features of 22qDS.

  Larger sample sizes and detailed consideration of phenotypic and statistical methodology will be needed before one can conclude that there is “a significant correlation between deletion site and phenotypic expression” in 22qDS.

  Anne S. Bassett, MD, FRCPC^1,2
  Rosanna Weksberg, PhD, MD, FRCPC^{3, 4}
  Eva W.C. Chow, MD, MPH, FRCPC^1,2

  ¹Clinical Genetics Research Program, Centre for Addiction and Mental Health, Toronto, Ontario, Canada
  ²Department of Psychiatry, University of Toronto, Toronto, Ontario, Canada
  ³Division of Clinical and Metabolic Genetics, Hospital for Sick Children, Toronto, Ontario, Canada
  ⁴Department of Molecular & Medical Genetics, University of Toronto, Toronto, Ontario, Canada

  Table 1. Frequencies from Rauch et al. with percentages and 95% CI

  22q11.2 deletions	Number	Sample	%	95 % CI*		Number	Sample	%	95 % CI*
  				Lower Bound	Upper Bound				Lower Bound	Upper Bound
  	Phenotype: Conotruncal congenital heart defect (ctCHD)					Phenotype: Genetics referrals with various features suggestive of 22qDS
  Positive clinical FISH	37	200	18.5	13.4	24.6	11	77	14.3	7.4	24.1
  3 Mb (common)	33	200	16.5	11.6	22.4	11	77	14.3	7.4	24.1
  1.5 Mb (proximal nested)	3	200	1.5	0.3	4.3	0	77	0	0	3.8
  Atypical with proximal extension	1	200	0.5	0.01	2.8	0	77	0	0	3.8
  Atypical distal nested	0	200	0	0	1.5	1	77	1.3	0.03	7
  Atypical distal to 3 Mb region	0	200	0	0	1.5	2	77	2.6	0.3	9.1
  Negative clinical FISH	73	73	100	96	100
  Atypical distal nested	0	73	0	0	4
  Atypical distal to 3 Mb region	0	73	0	0	4
  	Multiple congenital anomalies and/or mental retardation without features of 22qDS (MCA/MR)					Healthy controls
  Positive clinical FISH	0	100	0	0	3
  3 Mb (common)	0	100	0	0	3
  1.5 Mb (proximal nested)	0	100	0	0	3
  Atypical distal nested	0	250	0	0	1.2	0	285	0	0	1
  Atypical distal to 3 Mb region	0	300	0	0	1	0	285	0	0	1

  ^*Where the frequency is 0, the 95% CI is one-sided

  
  

  Table 2. Comparisons of frequency of distal deletions in different samples from Rauch et al. using Fisher’s exact testing*

  	Distal nested deletion		Deletions distal to the 3Mb region		Any distal deletion
  Group vs Genetics Referral Sample	Parameters	p-value	Parameters	p-value	Parameters	p-value
  Retrospective ctCHD	0/73 vs 1/77	1	0/73 vs 2/77	0.5	0/73 vs 3/77	0.25
  Prospective ctCHD	0/200 vs 1/77	0.28	0/200 vs 2/77	0.08	0/200 vs 3/77	0.02
  All ctCHD	0/273 vs 1/77	0.22	0/273 vs 2/77	0.048	0/273 vs 3/77	0.01
  MCA/MR sample with data on distally nested deletion	0/250 vs 1/77	0.24	not applicable		not applicable
  MCA/MR sample with data on deletions distal to the 3Mb region	not applicable		0/300 vs 2/77	0.041	not applicable
  Healthy controls	0/285 vs 1/77	0.21	0/285 vs 2/77	0.045	0/285 vs 3/77	0.009
  Within the Genetic Referral Sample
  Typical facies vs. non-typical facies	0/14 vs 1/63	1	0/14 vs 2/63	1	0/14 vs 3/63	1

  *Statistically significant results are bolded

  References

  1. Rauch A, Zink S, Zweier C, Thiel CT, Koch A, Rauch R, Lascorz J, Huffmeier U, Weyand M, Singer H, Hofbeck M. Systematic assessment of atypical deletions reveals genotype-phenotype correlation in 22q11.2. J Med Genet 2005.

  2. Richardson WS, Wilson MC, Williams JWJ, Moyer VA, Naylor CD. Users' guides to the medical literature: XXIV. How to use an article on the clinical manifestations of disease. J Am Med Assoc 2000;284:869-875.

  3. Goldmuntz E, Clark BJ, Mitchell LE, Jawad AF, Cuneo BF, Reed L, McDonald-McGinn D, Chien P, Feuer J, Zackai EH, Emanuel BS, Driscoll DA. Frequency of 22q11 deletions in patients with conotruncal defects. J Am Coll Cardiol 1998;32:492-498.

  4. Cohen E, Chow EWC, Weksberg R, Bassett AS. Phenotype of adults with the 22q11 Deletion Syndrome: A review. Am J Med Genet 1999;86:359-365.

  5. Feinstein C, Eliez S, Blasey C, Reiss AL. Psychiatric disorders and behavioral problems in children with velocardiofacial syndrome: Usefulness as phenotypic indicators of schizophrenia risk. Biol Psychiatry 2002;51:312-318.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top