Guide · Markets

Phase 3 Trial Results Explained for Biotech Investors

A Phase 3 trial is the large final study a drug needs before approval. Investors judge it by its primary endpoint, the hazard ratio and the p-value.

In two days in November 2026, two cancer drugs moved their companies' shares in opposite directions by huge amounts. On Monday, November 16, Summit Therapeutics fell 16.47% after the US Food and Drug Administration declined to approve its lung cancer drug. On Tuesday, November 17, SELLAS Life Sciences rose 156% after its Phase 3 leukaemia trial succeeded. To understand why, you need to know what a trial measures and which numbers regulators care about.

Drugs are tested in phases. Phase 1 trials give a new drug to a small group to check safety and dose. Phase 2 trials test whether it works in a few dozen to a few hundred patients. Phase 3 trials are the large, usually randomised studies that compare the drug with a placebo or the existing standard treatment, and they are what a company files with the FDA in a New Drug Application or, for biological products such as vaccines and antibodies, a Biologics License Application. The FDA then has a target date to decide, after which it either approves the drug or sends a complete response letter explaining why it cannot approve it yet.

Every trial names a primary endpoint before it starts, and that is the test it passes or fails. In cancer the strongest endpoint is overall survival: how long patients live. A weaker but faster one is progression-free survival: how long patients live before the disease gets worse. Summit's trial improved progression-free survival but its gain in overall survival was not statistically significant, and the FDA said effectiveness was not established. SELLAS's REGAL trial was designed around overall survival from the start, and median survival was 24.0 months on its vaccine against 14.7 months on the best available treatment.

Two numbers summarise the result. The hazard ratio compares the risk of the event, here death, in the treated group with the control group at any moment during the study: 1.0 means no difference and below 1.0 favours the drug. SELLAS reported 0.59, a 41% lower risk of death. The p-value estimates how likely a difference at least that large would be if the drug did nothing; the usual threshold is 0.05, and SELLAS reported 0.035. A p-value just under the line, from a trial with relatively few events, means the result is real by the standard rules but may be examined closely by regulators.

For investors the reaction depends on how much of the company's value rests on the drug. A large pharmaceutical group with dozens of products barely moves on one trial; a company with one late-stage asset can double or halve, and the move happens at the open, before any stop-loss can act. Success often brings a second effect: SELLAS announced a $150 million share sale the same evening and the stock fell 10.57% after hours, because a company that needs cash to reach approval usually raises it when its share price is highest.

Before a readout, check the endpoint, the size of the trial and what the market has priced in, and remember that the options market often prices a far bigger move than for an ordinary stock. After it, read the full release: the primary endpoint, the hazard ratio and its confidence interval, the p-value, safety, and the next regulatory step and date. A headline that says a trial was positive without naming its primary endpoint deserves suspicion.

By Trade2Spike Team, in-house editorial team. About us · Methodology