When examining drug sensitivities across a panel of cell lines, I am not sure that a fixed-time measure is appropriate (e.g. live cell count after 48 hours with increasing doses of the drug). In particular, drugs that affect DNA replication (e.g. Topoisomerase inhibitors) will obviously affect slow-cycling cells less as they will have a smaller population going through S phase in 48 hours. Thus, slow cycling cells tend to appear more resistant to the drug. But are they really?
I thought of reporting the survival at a time when untreated cells have increased e.g. 4-fold (an average of 2 cell cycles), which would happen at different times for different lines.
Can anyone offer opinions about the validity of either approach or offer a third alternative?
A fixed-time measure would not be appropriate. You need the drug present for at least one full cell cycle in an exponentially growing (log-phase) population, and maybe 1.5 is better. Likewise for measurement time; two full cycles for each individual cell line. Better yet, leave the drug on for two full cycles and measure colony-forming ability (reproductive survival). Good luck.
Thanks for the prompt reply. I did almost identical experiments in the 1960's, and I mixed cultures of two cell lines and completely eliminated one of them by timed administration cycles of FU. The assay was colony formation -- the ultimate test of reproductive survival.
Typically drug sensitivity is expressed as drug level that will kill 90% of the cells using colony formation as an end point. You can also measure drug potency as the amount of drug that inhibits growth by 50%. In the former case which is more robust
you treat cells either continuously or sometimes for a shorter time period and then wash and trypsinize cells and replate at several hundred per dish and count colonies after typically 10 days or so depending upon doubling time. Cell doubling time is not a factor with either approach.
The US NCI has tested (and continues to actively test) not far from one million compounds.
They analyzed the drug-induced effects on cancer cell population growth after 48h.
Sharing at 100% the remark of Paul, we always use 72h of cell-drug incubation.
It is obvious that data generated after 24 or 48 or 72 or 96 hours will totally differ from each other for a same compound on a same cancer cell line and with a same concentration.
However, unfortunately, cancers in patients are very heterogeneous and their cell doubling times amazingly changes from one subclone to another one in the same cancer.
We have to keep in mind that some "poor cancer cells prinsoners of plastic flasks" are far to be representative of the "biological reality" in a cancer developing in a patient.
Thus, I think that it is better to "stay" on 48 or 72h to be able to compare "our" data to those generated by our colleagues.
I am attaching here articles relating to the NCI 60-cell-line approach and also some articles on the complex biology in "true" cancers.
Best regards
Robert
- Badges
- Science topic
- Similar topics
- Biological Science
- Anatomy
- Tissue