Earlier this year, I cautioned investors to tread lightly with potential Alzheimer's disease drugs. I sure hope they listened. Medivation (Nasdaq: MDVN) is down 67% after results from a phase 3 trial, CONNECTION, showed that its drug candidate, Dimebon, doesn't work.

Medivation and partner Pfizer (NYSE: PFE) might as well change the drug's name to Divebomb, because there doesn't seem to be any positive news from the clinical trial. The drug failed to meet its primary and secondary endpoints, and patients taking Dimebon even fared worse than those taking placebo for some measures. The placebo group did a little better than expected, but that didn't affect the outcome; even if the placebo group had ended up within its expected range, the trial would have failed.

Dimebon is still being tested in a pair of clinical trials, in combination with Eisai's and Pfizer's Aricept and Forest Labs' (NYSE: FRX) Namenda, but I wouldn't hold out much hope for success.

For Pfizer, the results aren't nearly as disastrous. Assuming Dimebon remains a flop, Pfizer will be out the $225 million it put up to gain the rights, and 60% of the costs associated with the trial. But the company has eight other drugs to treat Alzheimer's disease in the works. Will all of them work? Highly unlikely, but I'd rather have nine shots on goal than one. Investors won't have to wait too long for the next results, either; the first phase 3 trials for bapineuzumab, partnered with Johnson & Johnson (NYSE: JNJ) and Elan (NYSE: ELN), are expected to read out later this year.

Today's flop, combined with earlier ones from drugmakers like Myriad Genetics (Nasdaq: MYGN), just reinforces the high-risk nature of developing drugs for complicated diseases. Take the risk if you want -- there's plenty of money to be made if the drug succeeds -- but make sure you keep that risk to a reasonable portion of your portfolio.