Anomolies in proliferation assays

David J. Adams adams041 at
Thu Mar 5 10:16:37 EST 1998


I am writing to seek advice on some problems I am having with a
cytotoxicity assay used in my laboratory to assess anticancer drug
sensitivity in patients with leukemia.  Tumor cells are cultured in
96-well Costar microtiter plates for 72 h in the presence or absence of
drugs, then percent growth inhibition (vs untreated control) is measured
by MTS metabolic assay (Promega).  A problem arises when low levels of
drug produce more MTS signal than untreated controls, which leads to as
much as 160% of control values. This "hump" in the dose response curve
dramatically effects the IC50 endpoint.  The problem appears randomly,
and is not related to any particular drug.  I have ruled out an edge
effect on the plate (controls are in row H), plating and pipeting
errors, and position in the incubator.  I cannot rule out a random
problem with out plate reader (BioTek EL340), but I think it unlikely. 
Recently, I have noticed, in completely separate experiments, that
untreated breast cancer cells appear to be dying in certain wells of
Costar 12-well plates.  This makes me wonder whether the plasticware is
at the root of both problems.  Have any of you had similar experiences? 
If so, I would appreciate your thoughts.  These issues are extremely
costly to my research, and any experience or solutions are rarely
published.  Thanks for your help!  Please respond directly to my e-mail
address:  adams041 at

David J. Adams, Ph.D.
Associate Research Professor of Medicine
Duke Comprehensive Cancer Center
Box 3843 MSRB
Research Drive
Durham, NC 27710

More information about the Cellbiol mailing list