NGRL- Wessex

_
_

Meetings

New and Developing Technologies 04

New and Developing Technologies 06

New and Developing Technologies 08

New and Developing Technologies 10

BCR-ABL Meeting 08

Agenda and talks

Summary Feedback

RNA splicing meeting 08

Home

Meetings>

BCR-ABL meeting

BCR-ABL Meeting '08 - Summary of key points from meeting

1. The two pilot UK NEQAS trials have revealed considerable variation in the performance of RQ-PCR for BCR-ABL. It is currently unclear how much of this attributable to UK labs and in the future it would be useful to separate out UK results in order to (hopefully) monitor improvements over time. UK NEQAS will perform further rounds and plans to introduce a scoring system to enable poor performers to be identified.

2. As part of a European LeukemiaNet programme and other parallel initiatives to promote better comparability of BCR-ABL testing, NGRL (Wessex) will act as a central reference laboratory for the UK and Ireland. Please can any lab that is interested in establishing the International Scale for BCR-ABL contact me (ncpc@soton.ac.uk) to arrange the initial exchange of samples.

3. The survey revealed 20 labs in the UK and Ireland that were performing RQ-PCR for BCR-ABL, with a further 4 labs planning to set it up. As expected there was wide variation in sample throughput, hardware and performance criteria. Key findings were:

Of the 18 labs who returned results, all employed ABL as a control gene except one who uses BCR (but plans to switch to ABL).
Several labs use less than the recommended 10mls of peripheral blood for analysis (although the number of ABL transcripts in the same volume of cDNA is ultimately the best judge of quality).
Most labs use EAC primer sets (but I do not think it is desirable to try and compel everyone to use identical conditions).
5 labs use lymphoprep to remove red cells. Largely unpublished data has shown that this gives different results to red cell lysis, particularly for bone marrow specimens. Labs using lymphoprep should switch to red cell lysis.
There was a general consensus that 3 RQ replicates for BCR-ABL should be performed and at least two replicates for the control gene.
There was a wide range in the number of control gene transcripts that were considered acceptable. It is very difficult to define a clear cut off since this may vary depending on clinical context however the sensitivity for all negative specimens should be given and I suggest that labs should be aiming for a median of at least 104 and preferably 3x104 ABL transcripts (in the same volume of cDNA used for each BCR-ABL reaction).
The use and nature of quality controls was very variable

4. There is a clear need to develop best practice guidelines for BCR-ABL testing that needs to go hand in hand with the development of quality control materials. Best practice guidelines should formally involve the BSH and CMGS and should be undertaken with representative from all interested laboratories. Any consideration of generic comments to be added to reports should be undertaken in consultation with the NCRN CML working group.

5. It was felt that it could be useful to develop QC rounds in addition to those run by NEQAS as part of an improvement and educational programme modelled on the European Biomed and Europe against Cancer groups. It would make sense to link this to the development of best practice guidelines.

6. The Adelaide laboratory routinely uses high and low BCR-ABL standards to determine the validity of any run by applying Westgard QC rules (see Adelaide QC.ppt; Westgard paper also attached for anyone who is really keen). NGRL(W) agreed to grow up a large batch of high and low standards for distribution to labs in order to pilot this process in the UK. Hopefully these reagents will be available in April-May this year.

7. The possibility of NGRL(W) sending out plasmids dilutions was also discussed. There is some concern here about contravening the EU IVD Directive but possibly this can be considered as part of a performance evaluation process.

8. Rare BCR-ABL fusions are seen in about 1-2% of cases and cannot be detected by standard primer sets. It seems that most of these cases are not monitored by molecular methods and many are not even picked up in the first place by appropriate analysis of a pretreatment sample. It would be useful to establish one or more designated centres that could monitor patients with rare transcripts. If you are keen to do this please let me know. If nobody else is keen then we can probably do this in Salisbury.

9. Approximately 7 labs perform RQ-PCR for p190 BCR-ABL and it would make sense to include p190 within the remit of this group.

10. Most people consider that RQ-PCR follow up is most appropriate once patients have achieved CCR as assessed by bone marrow metaphase analysis. Assessments prior to CCR by RQ-PCR or FISH on total leucocytes often produce misleading results. This study will hopefully be written up by Mike Griffiths in due course.

11. The European LeukaemiaNet is developing an RQ-PCR reporting programme which should be available soon. Further details from David Grimwade.

12. Nine labs currently perform of have just started BCR-ABL mutation testing with at least four more planning to set this up. All labs cover the entire region where mutations have been reported to occur, either upfront or after eliminating the presence of T315I. Development of controls for mutation testing was felt to be unnecessary, at least at this stage.

13. There was a general consensus that the meeting was useful and should be repeated once additional comparative data has been collected (hopefully before summer 2008). Future meetings will be held in London, probably at the Hammersmith Hospital, and funding will be sought from the pharmaceutical industry (this meeting was kindly supported by BMS). Details to follow in due course.

Last Updated: 29 July, 2008 by G.Watkins