By Monica Heger
This story was originally published Nov. 20.
HONG KONG — The Chinese University of Hong Kong plans to launch a noninvasive, sequencing-based Down syndrome test for high-risk pregnancies in Hong Kong in the next one to two months. Longer term, it is aiming to develop prenatal targeted sequencing-based tests to diagnose beta-thalassemia and other genetic diseases.
The technique, which was developed in Dennis Lo's lab at the university and published last year in the British Medical Journal, can detect trisomy 21 by sequencing plasma from pregnant women in the late first trimester or early second trimester (IS 1/18/2011). Lo, who licensed a patent for the technique to Sequenom for its MaterniT21 test, maintained the rights to market the test in Hong Kong.
Rossa Chiu, a professor of chemical pathology and first author of the BMJ paper,told Clinical Sequencing News that the team will launch the test on the Illumina HiSeq 2000 and will seek accreditation for it from Hong Kong.
In preparation for the launch, the lab recently acquired another HiSeq instrument and is taking steps to validate the protocol on the new system. Using a new machine with new reagents can affect the algorithm used to determine the number of copies of chromosome 21 in the fetal DNA, added Allen Chan, a chemical pathologist at the Chinese University of Hong Kong who helped develop the bioinformatics for analyzing the fetal genome.
Certification from Hong Kong requires the personnel running the test to be accredited and trained, and the person signing the report has to be a trained pathologist, said Chiu. Because the lab is co-located within the Prince of Wales hospital in Hong Kong, where Chiu and Chan both have co-appointments in the clinical diagnostic lab, the accreditation process should go smoothly, she said.
She anticipated that the test would cost around $1,000, but would vary depending on demand and the level of multiplexing. The more full the sequencing machine and the more samples that can be pooled, the lower the cost, she said.
Additionally, the team has submitted a proposal to have the test covered by Hong Kong's public health system, but Chiu anticipated that the test would have to first demonstrate its cost effectiveness before Hong Kong would cover it.
The test will initially focus on high-risk pregnancies. It must be ordered by a physician, and the team will deliver results back to the physician, not the patient. The lab is aiming for a turnaround time of about two weeks.
Around six samples will be pooled per sequencing lane, but that is also a moving target, as the level of multiplexing that can be done depends on the number of reads. To accurately make a diagnosis of trisomy 21, around 2 million aligned reads are needed per sample. When the team published the initial BMJ paper, they were using Illumina's Genome Analyzer, so could only pool two samples to achieve the required read depth. But with the HiSeq's higher throughput, and newer chemistry that yields even greater throughput, more samples can be pooled.
The team demonstrated earlier this year that it can also identify trisomies 13 and 18 (CSN 7/6/2011), but Chiu stressed that detection of those still aneuploidies needs to be validated.
One consideration for these aneuploidies is that detection is impacted by the GC content of the chromosomes. Chromosome 13 has very low GC content, while chromosome 18 has high GC content. Since sequencing platforms tend to have a GC bias, normalization algorithms to correct for the bias are designed based on the specific platform and reagents. As a result, the algorithms will need to be re-validated on the new HiSeq, since the different machine and reagents will affect the test's sensitivity, said Chan.
Fetal Genome Sequencing
Aside from trisomy 21 testing, the Lo lab is focusing on refining its method of whole-genome fetal sequencing so that it can detect other genetic diseases. Chiu envisions eventually using a targeted sequencing approach to test for an array of diseases in the fetus.
In a proof-of-principle paper published last year in Science Translational Medicine, the team demonstrated that it could sequence the entire fetal genome from maternal plasma at a 65-fold coverage (IS 12/14/2010) and could accurately detect the presence of a mutation causing beta-thalassemia, a type of genetic anemia common in Southeast Asian countries.
The next step is to validate the protocol in additional samples — no easy feat, said Chan. Detecting trisomies in fetal DNA requires just counting chromosomes copies, but a test to diagnose a disease like beta-thalassemia must be able to not only detect point mutations, but determine whether those point mutations are located in the fetal genome or the mother's genome, and if they are in the fetal genome, whether they were passed on from the mother or the father.
To do this, an important step is to first detect paternal mutations in the maternal plasma, the presence of which would indicate the mutation has been passed on from the father to the fetus. The next step is to determine if a maternal mutation has been passed on to the fetus, which can be done by determining the concentration of the mutation. A mutation that has been passed on, will be present at a slightly higher level.
Initially, the team tried to do the analysis without separating considering the maternal and paternal portions, but the presence of the paternal DNA "makes the analysis confusing," he said, because "certain parts are identical to the maternal chromosome and the presence of the paternal portions may even out the dosage difference between the maternal portions that have been passed on and those that have not been passed on."
The algorithm calculates all possible scenarios to come up with the right answer.
The team is validating this method in additional samples, and also trying to determine how much sequencing needs to be done to accurately call point mutations in the fetal genome. The more sequencing depth, the more accurate the test, but also the more costly, said Chiu.
Once the protocol is validated, Chiu said that a targeted sequencing method could be used to test for the disease in the fetus, which would also help reduce costs. Down the road, such a method could be applied to a panel of diseases, she said.
The lab is currently focusing on population-specific diseases, and especially those that have a high impact in Southeast Asia, such as beta thalassemia. But the same principal could also be applied to cystic fibrosis or other genetic diseases, she said.
While the lab's main focus is in fetal genomics, Chiu said that it will also pursue research in cancer. The same technique used to detect fetal DNA in maternal plasma can be used to detect circulating tumor DNA in a patient's blood, which could be used to monitor metastasis, relapase, and drug response. Chiu said that arm of research is still in the early stages, though.
Have topics you'd like to see covered by Clinical Sequencing News? Contact the editor at mheger [at] genomeweb [.] com.