Investor Biotech Primer

P-value -- A measure of how statistically significant a trial's results are. In other words, it's the percentage chance that a result is not true. The standard "line in the sand" p-value is 0.05, meaning 5%. Results with a p-value of greater than 0.05 are considered statistically insignificant. The results have a greater than 5% chance of being false. Let's take a look at a real-world example.

Last year, Celgene (CELG Quote) compared a combination of its thalidomide with melphalan and prednisone vs. just melphalan and prednisone in patients newly diagnosed with multiple myeloma. The company reported that the combination that included thalidomide led to improvement in overall survival. The p-value was 0.001 -- meaning the results had only a 1/10 of one percent chance of being false.

Contrast those results with Amgen's (AMGN Quote) trial for Aranesp for cancer-caused anemia. Amgen released the results in its fourth-quarter earnings report. In the trial, 18% of the patients taking Aranesp needed red blood cell transfusions vs. 24% in the placebo group. Sounds promising, right? But the p-value was 0.15, indicating there was only an 85% certainty that the results were accurate. There were some other problems in the study, but this example shows that you can have positive results, but without a low p-value, they're meaningless.

Hazard Ratio -- The relative risk of a patient in the experimental group compared to the control arm. If a drug increased survival by 100%, the hazard ratio would be 0.01. If there was no increase in survival, the hazard ratio would be 1. So clearly, a lower number is better. Returning to our Aranesp example -- after 16 weeks the hazard ratio was 0.89, indicating only a 12% increase in survival.

These are just a few of the basic terms you should know when investing in biotech stocks. I'll cover others as time goes by. In the meantime, feel free to email me with your questions or write Adam Feuerstein, who fields questions each week in Feuerstein's Biotech Mailbag.