Is there guidance or an industry standard for pull windows (+/- number of days)?

[John O’Neill]

#22528

Posted for a Subscriber
Question:
Is there guidance or an industry standard for pull windows (+/- number of days a pull can be done from the target pull date for each timepoint)? Are most LIMS systems capable of different pull window strategies – if client A wants +/- 3 days from target pull date at the 6M timepoint, but client B wants +/- 7 days from target pull date at the 6M timepoint?

[John O’Neill]

#22529

Excellent question and a timeless bugaboo for Stabilitarians. I’m not aware of a specific guidance on this aspect of pull points, and common practices have been governed by what companies have executed without regulatory citations and those instances where regulatory input influenced a company to follow a very specific (and almost always) tighter policy. For example, one company which had practiced a +/- 7 day window had an FDA inspection and following a citation (not based on Guidance, but what the inspectors called an expectation of good practice), revised their policy to No Early Pulls and a Late Pull limited to “Next Business Day”. Other companies have gone through inspections where early or late pull policies were justified with data showing that no more than insignificant changes would occur because of the pull shifts. Many of those justifications were accepted.
All that to say, that Pull ranges should be kept to the tightest practical window and there should be solid scientific justification for the policy you apply. In general, early pulls can raise a red flag for inspectors and very late pulls can raise issues of control in your practice. Working with your QA and Regulatory colleagues will help you arrive at an appropriate comfort level with the pathway you choose.
I would suggest contacting some of the larger LIMS providers about study-specific pull windows in their software. Since some companies apply a sliding scale to various types of studies, or even early vs. late stage of study, I wouldn’t be surprised if there could be study-specific windows, though I have never personally seen that in practice.
So, Stabilitarian audience, What response would you give to the writer?
Best regards,
John O’Neill

[Walter Routh]

#22532

Short answer–No, there is no industry standard. I came from a long history of strictly no early pulls. That requirement, however, came from a very old FDA audit observation and the program was terribly out of control at the time with very inadequate procedures. After several PSDG meetings even from 10 years ago, I was pushing to loosen that requirement in our program to allow early and late pulls on a sliding scale based on the age. The last pull should always be exactly on the date, however (or very very close). Back over 20 years ago, the issue from an audit perspective was the appearance of pulling a sample earlier in order to “help” it meet specification, so your procedures need to give strong scientific/statistical justification for your window, if allowing wider pull windows. That said, I was never able to implement a wider pull window. We wanted to primarily to enable more testing efficiency (it increases the abiltiy to pool testing together across multiple studies (if they’ve started just a few days/weeks apart). But, our systems and resources were just not able to monitor, track and plan future pulls adequately. We were more afraid of missing pulls among the chaos.

[Author]

Posts