Drugs fail their clinical trials for all kinds of reasons. Sometimes, it's related to efficacy; sometimes, it's safety related. Most of the time, there isn't a way to spot those problems before the clinical trials read out. If you want to invest in biotech companies, you just have to roll the dice and hope for the best.

But there are a few areas where complications can arise, lowering the likelihood of success.

Placebo effect
In a double-blind clinical trial, neither the doctor nor the patient know whether the patient is receiving the study drug or the comparator, either an already approved drug or a placebo. Since patients don't know which group they're in, they may assume they have gotten the study drug and think they feel better, even if the placebo isn't doing anything.

Clinical trials where patients report how they feel -- pain, rheumatoid arthritis, or depression, for example -- are a lot more susceptible to the placebo effect than drugs with more concrete endpoints, such as, overall survival, or something that can be determined from a blood test.

The placebo effect should increase the observed effect of the drug being tested, as well -- patients who don't respond to the drug will still report that they feel better -- so the placebo effect isn't a problem, in theory. Where drugs run into problems is when they assume one level for the placebo, and then see a much higher level.

That happened to BioSante's female sexual dysfunction drug LibiGel, which seemed to work wonderfully, increasing the number of satisfying sexual events by 83%. Unfortunately, the placebo group increased the number of satisfying sexual events by a whopping 65%, so the LibiGel effect wasn't statistically significant.

Ditto for Lexicon Pharmaceuticals (NASDAQ:LXRX). Its rheumatoid arthritis drug LX2931 seemed to produce a pretty solid 60% response rate. By comparison, Pfizer's (NYSE:PFE) Xeljanz produced 59.8% and 65.7% response rates at two doses in its own trial, but Xeljanz passed its clinical trial, and is on the market because the placebo group managed just a 26.7% response rate. In Lexicon's trial, 49% of the patients taking placebo responded.

Difficult to measure endpoints
Sometimes, drugs do exactly what they're supposed to do, but fail because measuring success is difficult.

Lupus is an excellent example. There was a graveyard of drugs that failed clinical trials in lupus over the last few decades, but I'd be willing to bet that at least some of them probably worked; however, because lupus is a complex disease, the drug developers had a hard time measuring success.

Human Genome Sciences and GlaxoSmithKline (NYSE:GSK) finally broke the drought with Benylsta by developing their own measures of success that involved a composite endpoint.

Lack of data
It's a lot easier to have confidence that a phase 3 trial will work when the trial looks similar to the phase 2 trial. Unfortunately, some companies will take one aspect of a phase 2 trial and test it in a larger phase 3 trial.

The better way to do it is to run a smaller phase 2 trial testing the earlier finding specifically, and then, if that trial is positive, move into a larger phase 3 trial. Many biotechs skip that step, however, because jumping straight to phase 3 is quicker and less expensive.

It's also more risky.

A prime example is Aeterna Zentaris' and Keryx Biopharmaceuticals' (NASDAQ:KERX) cancer drug perifosine, which went down in flames after the positive effect that looked good in a subset of patients in the phase 2 trial didn't carry over to the larger phase 3 trial.

Avoid them?
That's certainly one strategy. And I wouldn't fault you for following it.

If you do choose to invest in companies with risky clinical trial situations, make sure you're being compensated for it. There are certainly opportunities where the reward clearly outweighs the risk. Human Genome Sciences rose 6,300% from its lows when few people believed Benylsta would pass its clinical trials.

Fool contributor Brian Orelli has no position in any stocks mentioned. The Motley Fool has no position in any of the stocks mentioned. Try any of our Foolish newsletter services free for 30 days. We Fools may not all hold the same opinions, but we all believe that considering a diverse range of insights makes us better investors. The Motley Fool has a disclosure policy.