I have Outsourced my Method and now it has Stopped Working?

In the first edition of the help desk we discussed the retention time stability of assays, but another area where challenges are often faced is during the transfer of a methods. This generally involves several organisations all trying to apportion the blame on one of the other parties, but could it be simply that the method is being run on two different sets of instrumentation? The helpdesk team have all come across this phenomenon before, and are aware that the initial response would be that the different instrumentation should not really affect the performance of the chromatography and certainly should not affect the selectivity between two components. This would appear to be a logical explanation of what should happen but the reality can be very different. The following example exemplifies what is happening.

The helpdesk were initially asked to help in the development of a separation of some very polar compounds for a purity study.

In this example a fully porous column was being used. When running with standards the original method worked well. However, with real samples the column began to overload resulting in the primary peak beginning to merge into the one of the impurities which required quantification. Thus the role of the help desk was to improve the resolution between the active pharmaceutical compound and the impurity to allow real sample to be analysed.

The method development went well and a method was developed; the chromatography of the final method is shown in Figure 1. As previously stated the separation of standards was good for the original method but for the real samples, where the major component was overloaded, the resolution of the impurities from the main component was compromised which affected the quantification of the impurities.

The method was validated and was used with real samples with no issues. Eventually the customer looked to outsource the method and this is where it started to get a little messy. The contractor received the method and came back to the client stating that the method “Was not working”, as “There was a shift in the retention times of some of the components”. Initially it was assumed that the contractor had purchased a column that was not working, and so purchased a replacement which also exhibited the same issue.
So the method went back to the original customer to check that the assay was working, and no issues were observed. As can happen in this situation, an investigation then ensued into the method parameters to determine the cause of the variation, and it was identified that the isocratic portion of the method resulted in some movement of the compounds along the column which was affected by the delay volume of the pump. The original customer had used a very low dwell volume pumping system to develop the original separation, which the contractor did not have access to. Consequently the contractor was using pumps which had a larger delay volume, which in turn affected the length of time to deliver the gradient to the pump. Figure 2 demonstrates the effect with the two systems clearly having different retention times for some of the peaks, in particular looking at the earlier eluting compounds the effect is quite dramatic.

Unlock and downlaod the full article for more...