Of Mice and Cancer 109
Maximum Prophet points out a series of articles in Slate about the role of mice and rats in the fight against cancer. The first article discusses the problem of using the same type of animal for many tests; the reactions may be consistent, but they can also be different from the reactions a human has to the same treatment. "The inbred, factory-farmed rodents in use today—raised by the millions in germ-free barrier rooms, overfed and understimulated and in some cases pumped through with antibiotics—may be placing unseen constraints on what we know and learn." The second article focuses on one particular type of mouse, bred specifically for consistency and for its suitability to labwork, which has come to dominate biological testing. The final piece examines what researchers are trying to learn from the naked mole rat, a species that doesn't seem to get cancer on its own, and is resistant to attempts to induce cancer. "Buffenstein and her students tried one of these shortcuts. They placed some mole rats in a gamma chamber and blasted their pale, pink bodies with ionizing rays. The animals were unimpressed."
Re:Yeah, we knew that already. (Score:5, Interesting)
I had to read the article a few times. It doesn't make sense. Then I found it. He didn't see a result he wanted, and blamed the mice. Now he is wrapping himself up in a process of deflecting things he doesn't want as a flaw in the mice.
I mean, reread that article and think about that. He logical fallacy pops up a few times.
Re:So is there an alternative? (Score:5, Interesting)
This is part of the motivation for developing computational models of cancer. Code up the biological assumptions, calibrate to mouse data, validate to the mouse. If it works, then the biology and calibration protocols are probably fine. Re-calibrate to humans (with changes to geometry, tissue properties, cell parameters, etc.), run the models on clinical data (pathology, imaging, proteomics, etc.), and see how it does.
Now, actually doing this is the subject of tricky ongoing work by many many teams of people (see the work in the NCI Physical Sciences Oncology Network [cancer.gov]), but it's being driven by just the types of problems stated in this thread.
We've been testing various aspects of this on breast cancer and lymphoma, and the results are encouraging, ranging from explaining "tissue artifacts" in pathology (due to fast timescale biophysics) to predicting correlations between mammography and pathology (due in part to necrotic core biomechanics + oxygen diffusion limitations), to predicting DCIS excision volumes. (See stuff here [mathcancer.org] and a few movies [mathcancer.org].)