Exelixis (EXEL 2.02%) fell nearly 40% today on word that the company was continuing its phase 3 trial testing Cometriq in prostate cancer patients after the Independent Data Monitoring Committee took an interim peek at the data.

On the surface, the news doesn't seem so bad: The trial will continue to completion when Cometriq could still show an improvement of overall survival compared to patients taking prednisone.

But investors are right to be a little jittery. Here are three reasons why.

Big potential
Exelixis was a $1.2 billion company before today, but Exelixis sold a measly $4.3 million worth of Cometriq in the fourth quarter. The drug is approved for a type of thyroid cancer that isn't particularly common.

Prostate cancer is a big market however. Johnson & Johnson (JNJ -0.69%) sold $495 million worth of Zytiga in the fourth quarter. Medivation (MDVN) and its partner Astellas racked up $126 million for Xtandi in the U.S. alone.

Investors have priced in some of the potential sales in prostate cancer and, to a lesser extent, liver, kidney, and lung cancer, which are further behind in the clinic. If the interim look had caused the trial to be stopped for positive efficacy, Exelixis would have been able to capture some of the prostate cancer market sooner, and the Food and Drug Administration and EU regulators would surely give the drug an expanded approval.

Instead, we'll have to wait for the final analysis later this year.

No news isn't good news
Some interim looks at clinical trial data have futility analyses built into them that check to see if there's a low likelihood of the drug showing a statistical significant improvement in the final read of the data. If it appears unlikely that the drug will reach the threshold, the trial is stopped early because there's little reason to continue.

Unfortunately there was no futility analysis in the interim peek, so there's no way to know if Cometriq is close to showing an improvement in overall survival or a long way off. All we can be certain of is that Cometriq isn't performing substantially worse than prednisone because, if there was some safety issue, the Independent Data Monitoring Committee would have stopped the trial. Of course that's not saying much.

Competitors could do it
Investors were hoping the trial would be stopped early for positive efficacy because trials testing Johnson & Johnson's Zytiga and Medivation and Astellas' Xtandi were both stopped early at their interim looks.

But these aren't apples to apples comparisons.

The Zytiga trial in late-stage prostate cancer patients was stopped after 552 events. For Xtandi, the interim look was at 520 events. Exelixis, on the other hand, took its interim peek after just 387 events. The more events there are, the easier it is to show statistical significance with the same relative magnitude of difference between the drug and the comparator.

The patient populations in the trials are also different. Patients in Exelixis' trial could have failed Sanofi's Taxotere, Zytiga, and/or Xtandi, but enrolment in Johnson & Johnson's and Medivation/Astellas' trials were limited to patients that had failed Taxotere since their drugs weren't approved yet.

The potentially sicker patients, especially if they failed multiple treatments, could make it harder for Exelixis to estimate the overall survival of the control group. If the control group is living longer than expected, it's harder to show a statistically significant increase in survival.

Buying opportunity?
There's certainly more risk in owning Exelixis with only one shot to show Cometriq improves overall survival.

And it's possible the difference in survival won't be as pronounced if Cometriq does pass the trial at the final look than it would have been if it had passed the interim look. As a last line of defense, any statistically significant improvement in survival should be enough to get the drug approved for prostate cancer, but could signal that it isn't as active and make it harder to compete with Zytiga and Xtandi earlier in the disease progression.

But at its market cap of $760 million, I think investors are probably overreacting, producing a potential buying opportunity for those that can tolerate the higher risk-reward ratio. Even if Cometriq is only used as a last line of defense -- ballpark $400 million in sales per year -- Exelixis is worth more than twice what it's worth now. Of course it's got to pass the trial first.