Chemosensitivity Testing: Lessons Learned

Like all physicians and scientists engaged in the study of cancer biology and cancer treatment, I had accepted that cancer was a disease of abnormal cell growth. I remember reading the lead article in the New England Journal of Medicine (NEJM) that described the clonogenic assay (Salmon, S. E., Hamburger, A. W., Soehnlen, B. S., et al. 1978. Quantitation of differential sensitivity of human tumor stem cells to anticancer drugs. N Engl J Med 298:1321–1327).

I sat in a laboratory at Georgetown University reading about a lab test that could accurately predict the outcome of cancer patients, without first having to give patients toxic drugs. It seemed so logical, so elegant, so inherently attractive. Sitting there as a medical student, far removed from my formal cancer training, I thought to myself, this is a direction that I would like to pursue.

But I was only a first year student and there were miles to go before I would treat cancer patients. Nonetheless, selecting drugs based on a laboratory assay was something I definitely wanted to do. At the time I had no idea just how difficult that could prove to be.

After medical school I found myself in California. There I met an investigator from the National Cancer Institute who had recently joined the faculty at the University of California, Irvine. He too had read the NEJM paper. Being several years ahead of me in training he had applied the clonogenic technique at his laboratory at the National Cancer Institute. Upon his arrival in California, he had continued his work with the clonogenic assay.

All was going along swimmingly until the NEJM published their report documenting the results of five years experience with the clonogenic assay.  It wasn’t a good report card. In fact the clonogenic assay got an “F.”

Despite the enthusiastic reception that the assay had previously enjoyed, the hundreds of investigators around the world who had adopted it and the indefatigable defense of its merits by leading scientists, it seemed that something was very wrong with the clonogenic assay and I desperately needed to know what that was.

It so happens that in parallel to clonogenic assays, my colleague was working on a simpler, faster way to measure drug effects. Using the appearance of cells under the microscope and their staining characteristics, one could skip the weeks of growth in tissue culture and jump right to the finish line. The simple question to be answered was: Did the drugs and combinations kill cancer cells in the test tube? And if they did kill cancer cells in the test tube, would those drugs work in the patient? The answer was, “YES!”

Despite the clonogenic assay’s supporters, it turned out that killing cancer cells outright in the test tube was a much, much better way to predict patient’s outcomes. It would be years before I understood the depth of this seemingly simple observation and the historical implications it would have for cancer therapy.

In Chapter 7 of my soon-to-be-released book, Outliving Cancer I examine the impact of programmed cell death on human biology.