These are my notes from week 4 of Harvard’s Genetics 228: Genetics in Medicine from Bench to Bedside course, with lectures delivered by Dr. Brian Skotko and Dr. Faycal Guedj on February 20, 2015.

Brian Skotko - Medical care of individuals with Down syndrome (DS)

Medical care is conducted largely according to guidelines set out in [Bull 2011]. About 75% of children with DS have hearing loss, and 60% have vision issues. In both cases, most of the conditions are treatable. About 15% have thyroid problems, which are also treatable. Hypothyroidism can be treated with levothyroxine, hyperthyroidism can also be treated with drugs or surgery. Compensated hypothyroidism must be monitored closely as it can sometimes indicate autoimmunity. About 5% have Celiac disease, which can be managed with a gluten-free diet. Most children with Down syndrome have constipation, to the point that it is assumed until proven otherwise. The constipation often manifests with behavioral issues, but can be treated simply with laxatives. 75% of DS children develop obstructive sleep apnea by age 18, often requiring tonsil or adenoid surgery. It is recommended that every child with Down syndrome undergo a sleep study by age 4. If tonsil and adenoid removal doesn’t fix the sleep apnea, a CPAP is sometimes required. ADHD, OCD, depression and anxiety are also common. About 6% are diagnosed with classic autism, and 18% with autism spectrum disorder. When DS children diagnosed with autism (or all ASD?) receive a genetic workup, about 30-40% of the time a causal variant can be identified, according to current ACMG guidelines [Schaefer 2013]. DS children have slower metabolisms, and the majority are overweight. Spine issues are fairly common - atlantoaxial instability or occipitoaxial instability - and can put the child at very high risk for pinching a nerve, especially when participating in active sports. Seizures are sometimes observed, usually in the form of “infantile spasms”. One of the most infamous and dangerous, though rare (<1%) comorbid conditions of DS is leukemia. This comes in the form of either acute lymphocytic leukemia (ALL) or acute myeloid leukemia (AML), both striking before the age of 6. Chemotherapy is the first line of treatment. A higher proportion, about 10-20% of DS children, have transient myeloproliferative disorder (TMD) early on. This is a sort of preleukemia, but most of the time, for reasons we don’t understand, it goes away and does not lead to cancer. Adults with DS have an increased risk of Alzheimer disease and dementia.

Several types of solid tumors, including breast cancer, occur at dramatically decreased rates in individuals with Down syndrome. The reasons for this are currently under investigation. It is also thought that people with DS have increased resilience against Aβ toxicity, because many of them develop Aβ plaques by age 25 or 30 yet do not present with any symptoms of Alzheimer disease until age 50.

There are 58 DS specialty clinics across the United States, though in total probably only about 5% of the population can access them. Dr. Skotko represents the Down Syndrome Clinic at Mass General.


One hears an oft-cited figure that 90% of expecting mothers receiving a prenatal diagnosis of Down syndrome choose to terminate. That figure is from an international study and may or may not reflect rates in the U.S. The most recent U.S. study was done 10 years ago and reported a figure of 74% based on three states - California, Maine and Hawaii. We really have no good up-to-date statistics on the proportion of mothers choosing termination today, nor do we have any ability to quantify how it varies between different places in the U.S. What we do know is that the number of children with Down syndrome born each year has remained constant even while the population has grown and average maternal age has increased, to the point that one would have expected the number of Down syndrome births to increase two- or three-fold.

Faycal Guedj - Down syndrome treatment: from myth to reality

Down syndrome neurological phenotypes

By gestation week 15-18, brain weight is reduced by about 10%. By adolescence or adulthood the gap grows to 15-17%. This decrease is distributed unevenly, with particularly large reductions in the cerebellum and hippocampus. The reduction in brain weight and volume is due to a lower total number of cells. Though Down syndrome is phenotypically heterogeneous, 100% of patients exhibit cognitive deficits. These are usually quantified by testing IQ, which declines (relative to non-DS controls, I assume) progressively beginning at 6 months of age, and can be severe (20) to mild (55) (by adulthood?).

Genetics and mouse models

By looking at a large number of individuals with different partial trisomies of chromosome 21, some of whom had a DS phenotype and some of whom did not, there was identified a region from CBR1 to ERG which segregated almost (though not quite) perfectly with the DS phenotype. See for instance [Delabar 1993].

In modeling DS in the mouse, you don’t want the mother to have DS, as maternal influences in the gestation period could be a confounder. Therefore you want DS to be transmitted from the father, which means the males need to be fertile. Another challenge is deciding whether endpoints for preclinical tests in the mouse should be embryonic, neonatal or adult, and whether the treatment should be given only during gestation or after birth as well. The goal is to find a drug that could be administered to pregnant women and their fetuses, which sets a very high bar for safety.

The syntenic regions of human chr21 are scattered across Mmu10, 16 and 17. Many different mouse models are used in the study of DS. Some are monogenic models used to explore the contribution of a single candidate gene to the DS phenotype, while others try to capture as many of the human chr21 gene ortholgos as possible. We’ll focus on Ts65Dn, which has a duplication of a region of Mmu16 including 144 genes.

DYRK1A is considered one of the strongest candidate genes for making a major contribution to the DS phenotype. Patients with a heterozygous truncation of DYRK1A have microcephaly reminiscent to that seen in DS [Moller 2008]. As a kinase, it has many substrates, and could plausibly affect many different pathways. Mice heterozygous for a disrupted Dyrk1a gene have reduced brain weight, and mice overexpressing Dyrk1a have increased brain weight [Guedj 2009, Guedj 2012]. Dyrk1a overexpressers appear to have deficits in short-term memory, as they are not as interested in novel objects as wild-type mice. Overexpression also affects synapse and spine density.

Candidate therapies

It has reported that the protein product of DYRK1A is the target of epigallocatechin gallate or EGCG, a small molecule found in green tea [Bain 2003]. This motivated Dr. Guedj to investigate the effects of EGCG in mouse models of Dyrk1a overexpression and, later on, in the Ts65Dn model as well. This led to a placebo-controlled randomized clinical trial of 10 mg/kg/day EGCG in DS patients (n = 15 placebo, 15 treated), with two sites, in Paris and Barcelona [De la Torre 2014]. There is another clinical trial now ongoing with 90 patients. Results are expected to be reported this year or next. Dr. Guedj is also investigating the potential of apigenin, a flavonoid antioxidant, as a potential therapeutic.

For journal club, two students presented on Green tea polyphenols rescue of brain defects induced by overexpression of DYRK1A [Guedj 2009].