Next generation sequencing (NGS) appeared to be the main focus of many presentations at ASHG 2013 in Boston. In 2009 I was predicting a quick decrease in costs of NGS and a rapid invasion of the technique into the clinics. As we currently see the prices have fallen down to nearly affordable, however, they are still widely variable from $3000 to $10,000 all inclusive test for WGS. The analytical methods are also maturing rapidly with a selection of tools available from open source to highly priced commercial software solutions.
The movement of NGS into the clinic requires standardization of the wet bench assays as well as the following sequence and variant analysis. While the clinical labs are good at optimizing and validation of the tests, the important part of the process is proficiency testing (PT) mandated by CLIA. At the moment there are no PT standards for NGS. In 2012 ACMG produced a set of guidelines for NGS. Nazneen Aziz presented a talk at ASHG on PT aspect of these recommendations. It is quite obvious that there will have to be a change from the current 6 PT samples a year due to financial aspects, yet at the moment no decision has been made as to how NGS process will vary from the standard testing.
It is obvious though that PT will have to include wet part of the assay, bioinformatic analysis and eventually a clinical interpretation. It seems that a wet bench process is probably going to be the easiest to introduce PT. On the other hand bioinformatics part is going to be much more challenging. Despite relatively high concordance between the different technologies there are still some differences. It will be critical to be able to correctly identify a set of unique SNVs with no false negatives or false positives. It is clear that a selection of the analysis tools may become very important. The pilot PT according to N. Aziz will contain 200 concordant loci and will consist of SNVs, indels and wild type sequences, which will have to be correctly called by the participating labs. It is expected that the pilot will be available in 2014 across the USA. The labs in the USA will also be able to participate in the European PT program.
The final piece of the puzzle for QC of the NGS services offered by the testing laboratories will be the appropriate report, including not just results but also all limitations of the test. All reports will have to include the types of variants identified, properly annotated according to HGVS nomenclature, with clinical classification according to ACMG guidelines. This part is very important for a patient. However, this part may need some attention from the geneticists as D. MacArthur pointed out in his ASHG presentation that a deleterious or disruptive variant does not have to be pathogenic. Yet the databases frequently used by us do not necessarily properly discriminate clinical outcomes of the variants.
In summary it seems that NGS is slowly coming of age in the clinic. There still are and will be issues for some time but at least once the QC system is introduced we will be more confident in the data we are getting back.
Leave a comment