Another wonder drug fails

First it was Vioxx and Celebrex, which really helped arthritis but unfortunately, due to their mode of action, also increased the risk of heart disease. Now, a major new drug used to treat type II diabetes, Avandia (rosiglitazone), may be in trouble. Preliminary studies (which must be confirmed with a larger sample size) indicate an over 40% greater risk of heart attacks for patients on Avandia than in control groups receiving other standards of care.

Avandia works by activating a receptor in the cell, PPAR-gamma, which then modifies the transcription of various genes having to do with lipid metabolism, and other things. It is useful in diabetes because it sensitizes cells to insulin, supposedly helping to reduce hyperglycemia and other side effects of insulin resistance. But, it appears that there is either a non-specific effect leading to increased heart attacks, or PPAR-gamma has some other role yet to be discovered. We’ll see–the results publicized in the news are just preliminary, but it proves the adage that my old clinical medicine instructor drilled into us: “The only drugs without side effects are new drugs.”

Magic Bullet #72873982938

I do love breathless science headlines in the news, especially since the overwhelming majority of them are blown way out of proportion. So, this morning, when I read, “S. Korean Scientists Say Cancer-Killing Virus Developed”, I admit I was skeptical. The article makes it sound like a new panacea has been discovered. I can’t say that isn’t the case, but the research has a long way to go. Nevertheless, what they did is pretty cool.

A hot line of research in cancer therapy over the past several years has been to try to selectively target cancer cells, sparing normal bystanders. Standard chemotherapy is terribly bad at doing that: chemo typically works by preferentially killing cells that are rapidly growing, like cancer (and hair, and the lining of your gut and mouth, and your immune system, unfortunately). Oncologists hope to kill as much of the cancer as possible without killing you in the process. But, needless to say, the side effects of that strategy are less than desirable.

So, the current trend in cancer research is to find cures that don’t touch normal cells, or touch them less frequently or harm them less severely at least. There have been sporadic successes on this front, based on the increasing knowledge we have about how cancer cells are different from normal cells. The most famous of these new treatments is Gleevec, a drug that works by targeting specifically an altered component of a specific type of cancer’s cells. It works beautifully, although resistance has started to crop up. I’ve written on why that is at length. Other so-called molecular targets for new drugs are starting to be used successfully, and I think there is great promise for curing very specific types of cancer, or at least making them into chronic diseases instead of deadly ones, using them.

The Korean group in this news article, however, take a different approach, one that I think has a brilliant future provided safety concerns are adequately addressed. They are using genetically-engineered viruses that specifically target cancer cells, leaving normal cells untouched. I didn’t know how they did this until I started digging around, and it’s very clever. Cancer cells are different from normal cells by some combination of growing faster or dying less frequently, and they get that way by disrupting normal cellular pathways that keep growth in check and/or kill cells that have problems. The specific type of virus that is being used in these new therapies is only able to replicate itself (and in the process, kill its host cells) by disrupting those same cellular pathways in very similar ways. So, if you make a defective virus that can’t disrupt cellular pathways on its own, it will, at least theoretically, only be able to survive in cells that already have the pathways messed up. Like cancer.

The problem that researchers have run into with this method is that the virus has a hard time spreading from cell to cell, because of the sticky, thick “matrix” that surrounds and supports cells in our body. It kills the cancer cells it infects, but can’t get to the ones next door to kill them, too. So, only a small percentage of cells die, and the tumor survives and keeps growing. What the Korean group realized was, if they made another modification to the virus, causing it to make something to chop up the matrix (“relaxin,” a hormone, in this case), they might be able to get past this sticky problem. It appears to work, at least in the mouse models they published and upon which the news story was based. Up to 90% of the cancer cells are now killed by this modified virus. They can’t tell much about normal tissue yet, because the virus only infects human cells. Clinical trials have to be done, but there’s no real reason to expect that the newly modified virus will be any different from the old one, which definitely doesn’t hurt normal human cells.

That said, I do wish the media would not make these news stories into what they’re not. This is very preliminary research, which hasn’t yet been tested clinically. Just because human tumors injected into immune-compromised mice are much more effectively killed doesn’t mean that real cancer patients will benefit. The modified virus might just not work for an unforeseen reason–happens all the time. Also, I have one big worry about this new technique, one that I didn’t see addressed in the actual scientific manuscript. Another common characteristic of cancer is an ability to chew up that same “matrix”, so as to spread itself more effectively and “metastasize” to other areas of the body. I’m a bit concerned about trying to kill cancer cells more effectively by introducing a virus into them that encodes a greater ability to degrade extracellular matrix, even if that virus is supposed to kill them. What if, as almost always happens, cancer evolves in such a way as to evade the virus? Well, it could still make use of the “relaxin” gene, and spread even better than before.

We’ll have to see how the clinical trials go, but despite how cool the science is, I can’t help but assign this new discovery to my list of “magic bullets” that end up being duds…