(One quick note: Merck’s Gardasil and similar vaccines, which are frequently referred to as “cancer vaccines,” aren’t, really. These vaccines do what vaccines have always done, which is boost the body’s defenses against infection — in this case, by the human papilloma virus, which can cause cervical cancer. These vaccines don’t attack tumors directly.)

Cancer vaccines really don’t get much respect these days, after a series of high-profile failures and disappointments. The idea is simple enough: Use proteins unique to tumors to “train” the immune system so that it recognizes and attacks tumors, which normally skate past the body’s defenses like groupies waving backstage passes at a bouncer. But no cancer vaccine has ever worked unequivocally in human trials, as Dendreon’s experience with Provenge shows. At least Dendreon is still chugging along, though: Several other companies have gone down the tubes when their vaccines didn’t work, including all-but-forgotten names like CancerVax and Therion Biologics.

Enter Bellicum, a Houston biotech whose Latin name roughly translates as “call to war,” and whose strategy puts an interesting genetic twist on the basic notion of a cancer vaccine. (See our previous coverage in the sixth item here.) Like Dendreon, Bellicum’s basic strategy relies on dendritic cells, which are workhorses of the immune system that activate other defensive cells and teach them how to recognize invaders. (Tumors usually evade this system because, to the body’s watchdogs, they pretty much look like any other normal cell.)

To make Provenge, Dendreon extracts dendritic cells from a patient, marinates them with a protein “antigen” (that is, a molecule that stimulates an immune response) specific to prostate tumors, then re-infuses them into the patient. Theoretically, these newly primed dendritic cells will next begin recruiting an immune-cell army ready to have it out with any tumor cells they might encounter. Whether Provenge actually works or not, however, won’t really be known until an ongoing 500 patient trial produces some data, perhaps as early as next year.

Scientists at Houston-based Bellicum, however, think that process still has a number of shortcomings. Activated dendritic cells typically shut themselves off after about 24 hours and then die off another day later, limiting their ability to properly stimulate an immune response. That led a Baylor research team led by David Spencer — who is now also Bellicum’s chief scientific officer — to look for ways to either prolong that activation period or to “toggle” it so that the cells spent their entire activated lifetime in the lymph nodes, which are hotbeds for triggering immune reactions.

Those researchers focused on a dendritic-cell receptor protein — imagine sort of a docking port on the cell surface — called CD40, which is a sort of control switch for the dendritic-cell immune response. Normally, CD40 is only “switches on” when a dendritic cell docks with a particular type of T cell. The Baylor researchers, however, figured out how to genetically engineer a version of CD40 that could be switched on remotely by administering a particular drug molecule. If you’re interested, that team explained its work in a 2005 Nature Medicine paper that I’ve uploaded here (PDF).

The upshot of all this is that — theoretically, at least — Bellicum should be able to expose a patient’s dendritic cells to tumor antigens, genetically modify their CD40 receptors, return them to the patient and then switch them on once they’ve returned to the lymph nodes, thereby maximizing the dendritic-cell immune-stimulating effect. The process would look something like this:

To summarize, this slide shows dendritic cells (DCs) returned to the patient after modification and exposure to tumor antigens, where they migrate to the lymph nodes and activate killer T cells (formally known as cytotoxic T lymphocytes, or CTLs), which specialize in taking out infected cells. Here, the killer cells swarm tumors and destroy them. In theory, of course.

Bellicum says the system has worked well in animals and in laboratory tests using human dendritic cells, although it hasn’t yet tried the vaccine in humans. The company, however, hopes to move into early-stage tests in prostate-cancer patients by early next year.

Of course, no one knows if the strategy will work any better than other cancer vaccines, and there are certainly grounds for skepticism. There are an awful lot of moving parts in Bellicum’s vaccine-production process, which of course increases the chance that something unexpected will go wrong. What’s more, the immune system has a tendency to outfox even the most sophisticated attempts at manipulating it. Still, when in comes to opening up new avenues in the struggle against cancer, any attempt to refine what has so far been a disappointing strategy should probably be welcomed.

That dream suffered a major setback earlier today, when the FDA threw up a fresh roadblock to the approval of Dendreon‘s Provenge, the first such cancer vaccine to go before the agency for regulatory approval. In March, an FDA advisory panel recommended approval of Provenge as a treatment for prostate cancer, despite some reservations about its efficacy in clinical trials. Instead, the FDA issued what is known as an “approvable” letter — a misnomer that actually means the agency won’t approve a drug’s application without additional data. Providing that sometimes requires additional lengthy and expensive clinical trials.

Although the near-term impact on Dendreon’s stock was severe — its shares plunged by 60 percent on the news — chances are good that Provenge or at least one of many other cancer vaccines now in clinical trials will eventually work. To explain why, I’ll set the stage in this post with a walk through the troubled history of cancer vaccines in general and Provenge in particular. In subsequent posts, I’ll take a look at other vaccines currently in late-stage trials and the efforts of venture companies in this area.

First, though, a quick note on terminology: This class of therapeutic cancer vaccines shouldn’t be confused with more traditional infectious-disease vaccines like Merck‘s Gardasil. The Merck product protects against the human papilloma virus linked to cervical cancer, but isn’t really a “cancer vaccine” except in the loosest possible sense of the term.

One of the things that makes tumors dangerous is the way they evade the immune-system surveillance that guards against infection by bacteria and viruses. The goal of most cancer vaccines is to somehow stimulate the body’s defenders to identify tumor cells as enemy “others,” effectively calling in the immune system’s beat cops and strike teams to swarm and overwhelm them. (For a detailed description of how this should work in practice, click here.)

The problem has long been figuring out exactly how to trigger an immune response strong enough to attack the tumor but specific enough to leave other healthy parts of the body alone. There are two basic challenges: Identifying proteins specific to tumors that can be used to inoculate cancer patients, and finding the right way to expose those “antigens” to the guardians of the immune system.

Researchers have tried a variety of approaches, most of which have failed. In recent years, for instance, Antigenics has seen a cancer vaccine directed against “heat shock” proteins extracted from patients’ own tumors fail twice in large trials — once against kidney cancer, and again in skin cancer. A vaccine against pancreatic cancer made by Therion Biologics didn’t improve patient survival, effectively imploding the closely held company. Similarly, a startup called CancerVax pulled the plug on a melanoma vaccine in mid-2005 and later effectively vanished in a reverse merger with the European biotech Micromet.

Provenge, known generically as sipuleucel-T, took a different yet still relatively straightforward approach. Prostate-cancer patients underwent pheresis to filter out certain immune-system cells known as “antigen-presenting cells.” In the lab, researchers then cultured the cells together with a tumor protein called prostatic acid phosphatase in order to “prime” them for attack once they were reintroduced to the patient.

But Provenge’s history has been anything but smooth. The company’s first large-scale trial of the vaccine failed to demonstrate that the vaccine could hold prostate cancer in check, although Dendreon tried to argue that it appeared to work better in men with less aggressive prostate cancer — an after-the-fact analysis that amounted to moving the goalposts during the game. (For a technical summary of clinical-trial statistics and why Dendreon’s strategy didn’t hold water, click here.) The company rejiggered an existing trial to study Provenge’s effect in this particular subgroup of men, only to discover that a followup analysis of the original patients showed that men who received Provenge appeared to live longer than their counterparts. That analysis, however, involved just 127 men, a group so small that no one could be sure if the survival data was merely a statistical fluke.

Although the second trial also suggested that Provenge improved survival, it was also quite small and failed tests of statistical significance. What’s more, neither trial showed that the vaccine could actually slow the progress of the disease — a paradox that concerned experts assembled for an FDA advisory panel in March, although it nevertheless ended up recommending approval of Provenge by a 13-4 vote.

Dendreon’s shares shot up almost fivefold following that recommendation, although naysayers continued to warn that the vaccine hadn’t proven itself. Two no-voting members of the advisory panel wrote letters to FDA — which quickly leaked to the newsletter Cancer Letter — urging rejection of Provenge. In the end, the FDA apparently agreed, although it isn’t yet clear what sort of additional data the agency is looking for. Under one scenario, Dendreon may have to complete a large clinical trial that’s already underway, but which may not report data until 2010.

UPDATE: I’ve revised the description of Provenge’s clinical-trial history for accuracy.

Next: The outlook for other late-stage cancer vaccines.

The gist was that an experimental test for a protein called EPCA-2 appears to find prostate tumors missed by current diagnostics, while also avoiding “false positives” that can lead to unnecessary biopsies. On its face, that certainly sounds like great news, particularly since the most widely used test today — which looks for a protein called prostate-specific antigen (PSA) — is indeed wildly inaccurate, although many doctors still consider it better than nothing.

So far, so good. There’s just one problem: The trial described in Urology was too small and not designed properly for anyone to draw the conclusion that EPCA-2 works any better than PSA — or any of what are likely dozens of other protein “biomarkers” that various research teams are studying as possible diagnostics for prostate cancer.

To understand why, it helps to recall two important things about clinical trials. The first is that more subjects you enroll in a trial, the more weight the results will carry. (This is only true up to a point, but it’s a good approximation for our purposes.) If you’re only looking at a diagnostic’s performance in a limited number of patients, the odds are much higher that whatever results you see may be the result of sheer chance or other bias.

The second point is that the only reliable way to know that a diagnostic like EPCA-2 works is to test it in a prospective, or “forward looking,” trial. In this case, a prospective trial would most likely involve testing a large group of healthy men regularly with both the PSA and EPCA-2 tests and watching to see which ones developed prostate cancer and which test better predicted it. (Again, I’ve simplified things somewhat for clarity.) By contrast, a retrospective, or “backward looking,” test involves taking patients whose cancer status is already known and seeing if the experimental test correctly identifies them. It’s not a bad way to get a rough-and-ready sense of whether an experimental diagnostic is worth pursuing further, but it’s nowhere near as statistically reliable as a prospective trial.

One more piece of background: To date, most studies of biomarkers that are supposed to predict cancer or heart disease have involved small groups of patients in retrospective studies. The track record for such studies is not great: One recent large and rigorous effort to “validate” more than 80 biomarkers thought to predict heart attacks and related problems found that only one of them even came close to working as advertised.

You can probably tell where I’m headed. The Urology study, of course, was retrospective, and its key findings involved tests in relatively small groups of men — 30 in one case, 40 in another, 18 in a third — whose cancer status was already known. (The person performing the diagnostic was “blinded” to the patient status, which eliminates one possible source of bias, but doesn’t affect the larger statistical problem.) Which puts the latest EPCA-2 results pretty much squarely in the realm of promising but unproven.

In fact, the Urology report itself notes that the test values reported “are not necessarily reflective of a screening population,” and Robert Getzenberg, the Johns Hopkins University researcher who led the study, confirmed to me via e-mail that “it is clear that further validation is required.” (Getzenberg went on to tell me that he thought the EPCA-2 test would produce “equal if not better” results in a broader population. He holds a patent on the test and has consulted for and received a grant from Onconome, a Seattle biotech developing EPCA-2 as a commercial product.)

To its credit, the USA Today story put those caveats squarely before the reader. Here’s the lead paragraph from reporter Liz Szabo:

Researchers at the Johns Hopkins University School of Medicine are trying to develop a more reliable way to find prostate cancer. While experts say the new test is promising, they say it’s too soon to know whether it really works better than older screening methods.

Contrast that with the following from the LAT’s Susan Brink:

A new prostate test that relies on measuring levels of a blood protein called EPCA-2 accurately found cancer 94% of the time, a significant improvement over the current PSA test, according to a study released Wednesday.

Or, worse, the WaPo’s David Brown, whose lead reports the researchers’ belief in the test as if it were fact:

An experimental blood test for prostate cancer may help eliminate tens of thousands of unnecessary biopsies at the same time that it detects many tumors that are now missed by the test commonly used, its developers said yesterday.

Somewhat more incredibly, Brown managed not to mention either of the study’s key limitations anywhere in the text of his story, although he did note uncritically that the EPCA-2 test could be commercially available by 2008. The LAT story hints at those limitations by noting that EPCA-2 must still undergo further testing, but only barely.

I don’t mean to pick on the WaPo and the LAT or their reporters, who are generally much better than this. Mostly, I think, it points up some of the difficulty today’s medical reporters have interpreting biomarker-test findings, which can look a whole lot more clear-cut than they actually are. Absent an understanding of the statistics and trial-design issues here, the simple fact that a new test appears to outperform an older one can look a lot like important news, even when it’s really not.

I also don’t have any particular beef with this way the Johns Hopkins folks designed and conducted this study. Large prospective trials are great, but they’re also expensive and time-consuming, so it makes perfect sense to take a rough cut at the question you’re hoping to answer with a faster but more limited trial. It’s just that you can’t read too much into the results when you do that.

I can’t help but wonder, though, why this story got picked up by so many major newspapers, as Urology isn’t exactly in the top tier of medical journals that reporters keep an eye on. (Check out this recent table of contents and you’ll understand why.) Although I don’t know for certain, I suspect that the Johns Hopkins PR office gave the story a big push, and that the university’s institutional authority lent it credibility so far as reporters were concerned. Check out the university’s press release on the subject and marvel at the similarities between its opening paragraph –

EPCA-2 testing curtails unnecessary biopsies and can differentiate disease that has spread outside the prostate from cancer within the prostate, Hopkins team says.

– and the lead of Brown’s WaPo story.

Onconome, of course, might have pushed for press coverage as well — and of course it put out its own press release on the findings — but as a tiny biotech no one’s ever heard of, it would have faced an uphill battle getting attention at a major newspaper.

Just to be clear, I’d be more than happy to see EPCA-2 eventually proven as a new and more reliable cancer diagnostic. (Just about anyone with a prostate — or who cares about someone who has one — probably feels the same.) It just makes no sense to let our hopes run way ahead of the science, or for the media to do likewise.