Editor's note: This column originally appeared on RealMoney on April 14; we're offering it today as a bonus to TheStreet.com readers. If you'd like to read Adam's commentary in real time, please click here for information about a free trial to RealMoney.

To succeed over the long haul, biotech investors need to be savvy in the ways of the market like everyone else, but they also require fairly specialized medical and scientific knowledge. Drug development can be amazingly complex, and if most investors are like me, their science education ended in college, if not high school.

Some readers have asked for "Biotech 101" columns to help them make better investing decisions. This column is for those people who don't know the difference between an IND (which is an investigational new drug) and a phase II study, but really want to know.

Drug Discovery

Drug development starts small. In the research lab, scientists test thousands of chemical compounds and molecules against disease targets to find just a few promising drug candidates. This is the image that most people have when they think of drug research -- a bunch of scientists in white lab coats playing around with cells that are isolated in petri dishes and test tubes.

More recently, the unraveling of the human genome and the accompanying technology championed by companies such as Human Genome Sciences ( HGSI) and Millennium Pharmaceuticals ( MLNM) have promised to speed up this time-consuming and tedious process, but will never eliminate it entirely.

Preclinical Testing

Compounds that graduate from drug discovery move into preclinical, or animal, testing. In a lengthy series of tests, researchers need to learn how the drug candidate behaves when it enters the animal. Does the drug have the desired effect on the disease target? What are the toxicities? What doses can be safely used? How is the drug metabolized, or broken down, by the animal? What formulation (pill or injection) can be used? Are there any side-effects from long-term exposure, such as cancer or birth defects?

Once this small mountain of animal data is collected, and assuming it's positive enough to move on, researchers file an IND application with the Food and Drug Administration. Regulators then decide if the experimental drug is safe enough to move into human testing.

Phase I Studies

Phase I studies also start small, typically enrolling less than 100 healthy volunteers who are given the drug candidate in escalating doses to determine its safety profile. The goal here is to also determine where the drug goes in the body and how it is metabolized. Sometimes you'll see companies talk about early signs of efficacy from phase I studies, but be very careful: Because these studies are small and generally use healthy people, talk of drug efficacy here is more about wishful thinking than reality.

Phase II Studies

Assuming that a drug's safety passed muster in the phase I studies, a series of phase II studies enrolling upwards of several hundred patients will be ordered. The big difference here is that phase II studies are where the drug candidate is finally given to actual, sick patients. Different doses, formulations and dosing schedules might also be tried out. All this work is directed toward looking for signs that the drug does what it's designed to do. Well-designed and successful phase II studies build a foundation for a company to eventually design and run pivotal, phase III studies.

Phase II is where the real work in determining drug efficacy begins; it's also where most investors should start paying attention -- with extreme vigilance. There's an old saw among experienced biotech investors that says, drugs never look better than they do in phase II studies. Biotech companies certainly play this to their advantage, trumpeting positive phase II results to the public and to investors in an effort to drive stock prices.

Be mindful when sifting through what biotech companies disclose about "positive" phase II studies. The biotech graveyard is littered with the rotting corpses of drug candidates that failed miserably in phase III studies, despite showing great promise in earlier phase II studies. There are no sure things in drug development.

Generally speaking, phase II studies are not "controlled," which means the studies do not enroll a group of patients who are given a placebo instead of the drug candidate. In some phase II studies, patients are treated with an experimental drug given in combination with approved drugs. Without a control group, however, it's difficult to determine which drug is helping patients. (This has been a problem that's dogged Genta ( GNTA) and its antisense cancer drug Genasense, for example.)

Often, you'll see companies compare uncontrolled phase II study results to results from similar patients culled from previous studies. These historical comparisons are helpful in judging a drug candidate's efficacy, but only to a point. Continuous, incremental advances in medicine means that patients treated today generally fare better than patients treated yesterday -- which can skew comparisons made from older studies.

Without a control group, doctors and patients involved in phase II studies know they're being given a drug that might help them get better. This can lead to biased results, even if the bias is unintentional.

Finally, ensure that positive results are clinically relevant. For example, an experimental cholesterol-lowering drug may do just that in a phase II study; but if it underperforms rival drugs already approved, the data -- albeit positive --could be commercially meaningless.

Certainly there are positive phase II studies that accurately presage the approval of a new, effective drug. It just doesn't happen all the time, or even as often as most investors would like. So be smart when interpreting phase II results and realize that companies won't always offer the fullest, clearest picture of their experimental drugs.

Phase III Studies

Phase III is the ultimate showdown: large, well-designed (hopefully) clinical studies enrolling hundreds, sometimes thousands, of patients, that are the final arbiter of a drug's efficacy and safety. The data collected from phase III studies, if positive, makes up the bulk of a company's marketing application to the FDA.

Because of their prominence and importance, this is probably the area of drug development in which investors are most knowledgeable. Unlike phase II studies, phase III, or pivotal, studies almost always contain a control group of patients in which to compare and judge the efficacy and safety of the drug candidate. Good phase III studies are also "randomized" and "double blinded," which means patients are randomly assigned to either the drug arm or the control arm; thus, neither patients nor their doctors know which arm of the study they've been assigned. This eliminates potential bias.

In order to succeed, a phase III study must show that the experimental drug being tested demonstrates a statistically significant increase in efficacy compared with the control, which can be a placebo or another existing, approved drug. The measure of statistical significance is expressed as a so-called p value, which is simply the odds that efficacy was a result of random chance and not the drug being tested.

In order for the efficacy endpoint of a phase III study to reach statistical significance, the "p value" must come in at .05 or below. This means there is a 5% chance or less that results occurred randomly.

The lower the "p value" the stronger the study results. The much-heralded colon cancer study involving Genentech's ( DNA) Avastin last spring came in with a "p value" of .00003 for its primary endpoint, meaning there was a microscopically small .003% chance the results were random.

While it's harder for companies to fudge results from phase III studies, it does happen. Beware of companies trumpeting a drug's safety in phase III studies before discussing efficacy endpoints. Safety data from a phase III study is important, for sure, but the primary endpoint of these pivotal studies is almost always based on the drug's efficacy.

Also, be aware that some companies sometimes pull a bait and switch when it comes to efficacy data -- choosing to highlight a less-important, secondary or tertiary efficacy endpoint if the primary endpoint fails to hit statistical significance (which technically speaking, means the study failed). While this doesn't always doom a drug, it certainly raises a major red flag that could delay or prevent a drug from being approved.

One last note: I want to recommend a helpful book, From Alchemy to IPO: The Business of Biotech by Cynthia Robbins-Roth.

This is one of the first books I read to get myself ready to cover biotech three-and-half years ago for TheStreet.com. It's a fast read, providing a richly detailed overview of the biotech industry (including the birth of early pioneers Genentech and Amgen ( AMGN)), and easily understood explanations of the drug development process. As the title implies, Robbins-Roth also has the investor in mind. Beginner biotech investors, as well as those more experienced, will learn something from this book. Some of Robbins-Roth's insights were included in this story.
Adam Feuerstein writes regularly for RealMoney.com. In keeping with TSC's editorial policy, he doesn't own or short individual stocks, although he owns stock in TheStreet.com. He also doesn't invest in hedge funds or other private investment partnerships. He invites you to send your feedback to adam.feuerstein@thestreet.com.