Biotech is a lot like the tagline from a certain fashion television show: "One day you're in, and the next, you're out." In the past two business days, Exelixis (Nasdaq: EXEL) has experienced this sort of boom and bust, watching one candidate fizzle while another took flight.

On Friday, the Rule Breakers pick got word that partner GlaxoSmithKline (NYSE: GSK) was rejecting its option to take over development of Exelixis' diabetic nephropathy drug candidate XL784. GSK's decision was expected, because XL784, designed to treat a kidney disease, failed to meet its efficacy endpoints in phase 2 testing. Exelixis had previously stated that it would halt further development of the drug.

Glaxo has options remaining on any two of a pool of seven drugs in Exelixis' pipeline, including several more interesting candidates. Thus far, Glaxo's selected one of Exelixis' candidates (XL880), and now rejected two others (XL647 and XL784). Exelixis continues to develop XL647 on its own, with plans to bring it into phase 3 testing in the first half of this year.

Even though XL784 didn't work out, another Exelixis drug candidate, XL139, was picked up by partner Bristol-Myers Squibb (NYSE: BMY) yesterday. XL139 inhibits cancer cells in a way that's different from any other approved drugs today. Finding novel anti-cancer targets is important, because it's harder to get the FDA's approval for me-too drugs than for new approaches to fighting a disease.

Exelixis' collaboration agreement with Bristol-Myers is akin to an earlier-stage version of its deal with GSK. Exelixis discovers the drugs, and shepherds them through preclinical trials and into phase 1 testing. Bristol-Myers then gets to cherry-pick and share development and U.S. profits on up to three of those drug candidates. XL139 is the first of Bristol-Myers' picks.

It's never a dull month for Exelixis and its stable of drugs in development. As long as it can keep generating new data for its compounds, and discover new compounds to replace failures like XL784, Exelixis shareholders should be fine in the long run.

More Foolishness on drug developers: