Houston is home to the Texas Medical Center, one of the largest concentrations of hospitals, medical schools and research institutions anywhere. People who work in a place like that learn a fact the rest of us rarely hear: the normal fate of a promising drug candidate is to fail. The surprise is not that most programs end in disappointment. It is how early, and how often, it happens.

That base rate is the missing context behind nearly every health headline about a new compound, and it applies to peptides as much as to any other class. Before weighing a claim about what a peptide does, it helps to know how far up the ladder the evidence has climbed.

The number that frames everything

The most cited benchmark comes from a 2021 report by the Biotechnology Innovation Organization, Informa Pharma Intelligence and QLS Advisors, which analyzed ten years of development data from 2011 to 2020. It found that a drug candidate entering Phase 1 human trials had a 7.9 percent chance of eventually winning approval from the US Food and Drug Administration. That works out to fewer than one in twelve.

The figure was lower than in the group’s previous report, which put the same measure at 9.6 percent for 2006 to 2015. The time it takes is also long. The report put the average journey from the start of Phase 1 to approval at about ten and a half years, and every year of that is spent on programs that had already survived the laboratory.

Behind those averages sits a wide spread. Likelihood of approval from Phase 1 ranged from roughly 24 percent in hematology to under 4 percent in urology, and about 5 percent in oncology. Programs aimed at rare diseases did better, at around 17 percent. Where a candidate starts matters, so one headline percentage can only be a rough guide.

The ladder, rung by rung

Development is a sequence of filters, and each one asks a different question.

The first rung is preclinical work: experiments on cells in dishes and on animals. It asks whether a compound does something measurable and whether early safety signals look acceptable. It is cheap, fast and generates the most exciting early results, because a dish can be controlled in ways a person cannot.

Phase 1 is the first test in people, usually small, and focused on safety and how the body handles the compound. Phase 2 enlarges the group and looks for a signal that the compound does what the theory predicted. According to the same dataset, a candidate that has just entered Phase 2 has about a 15 percent chance of approval.

Phase 3 is the large confirmatory stage, with hundreds to thousands of participants and a comparison against placebo or an existing treatment. Candidates entering it have about a 52 percent chance of reaching approval, which means roughly half still fail at the last major hurdle. After that comes regulatory review, where around nine in ten applications that get that far succeed.

The averages for time spent are worth noting too. The report’s figures put Phase 1 at about 2.3 years, Phase 2 at about 3.6 years, Phase 3 at about 3.3 years and regulatory review at about 1.3 years. The slowest and most expensive stages are the ones that demand the most rigorous evidence, which is no accident.

Why failure is the normal outcome

Analyses of failed programs tend to point to the same causes. The most common is a lack of efficacy: a mechanism that looked convincing in cells or animals does not produce a meaningful benefit in people. The second is safety, where an unexpected harm appears in a larger or longer study. Smaller shares fail because the compound behaves poorly in the body, or because the business case collapses.

The deeper reason is that biology resists shortcuts. A cell in a dish has no immune system, no blood supply and no liver to break down what it receives. An animal differs from a person in anatomy, metabolism and disease. Every rung of the ladder exists because the one below it turned out to be a poor predictor of the one above.

That is why a result from an early rung, however striking, is best read as a reason to run the next experiment and not as a conclusion.

Peptides are on the same ladder

Peptides have a long and successful record in medicine. Insulin, first used in patients in 1922, is a peptide hormone and the classic example. More recent peptide-based medicines, including semaglutide and tirzepatide, are approved prescription drugs, and they earned that status through large trials that followed participants for many months.

That success shows what the ladder demands. It does not show that every peptide behaves the same way. Whether a given peptide should be grouped with small molecules or with biologics depends on its size and how it is made, and the reported approval rates differ between those groups: roughly 9 percent for biologics and under 6 percent for conventional small molecules in the same report. For a specific compound, the base rate is a rough guide, not a prediction.

Where many peptide claims actually live

Here the ladder gets interesting, because many products that mention peptides never had to climb it. In the United States, cosmetics reach shelves without FDA premarket approval, and dietary supplements are marketed under a 1994 law that does not require approval before sale. Products in those categories can rest on early-rung evidence and small, short studies, as long as their claims stay within the rules.

Collagen is a good illustration. Most trials of collagen peptides in skin are short, use small groups and rely on instrument readings of hydration and elasticity, while longer-term outcomes for bone and joints are studied less often and remain less settled. A short summary of why most collagen trials stop at skin-measurement endpoints on one Texas information site makes the same point. The evidence is real and also early, and it is valuable to know which rung it sits on.

Copper peptides, common in skincare, follow a similar pattern. As far as the published record shows, much of the foundational work was done in cell cultures and animals, and the human studies are small and mostly topical. This copper peptide overview is a reasonable place to see how laboratory findings and human findings are separated on the page.

Reading any claim by its rung

A simple routine turns the ladder into a practical tool. When you meet a claim about a peptide, a supplement or any new compound, ask where its strongest evidence comes from.

• Was it cells, animals or people?
• If people, how many, and for how long?
• Was there a control group, and was the comparison blinded?
• What was measured: a clinical outcome, or a stand-in such as a lab value or an instrument reading?
• Has anyone outside the original group repeated it?
• Is the compound approved for the use being described, or only for a different one?

The answers place the claim on the ladder. A claim built on cells and a handful of volunteers is a hypothesis with some support. A claim built on large, controlled, repeated trials and regulatory review is something else.

What a hub like Houston teaches

A visit to any large research center shows how much effort goes into the middle rungs. Teams design trials, recruit participants, monitor safety and wait years for answers that often disappoint. It is slow because getting it wrong is costly, to patients first and to everyone else afterward.

Readers in Texas who want a plain-language guide to separating established medical uses from preliminary findings can start with Texas Peptides, an information site that describes its aim as offering evidence-based insight for Texans.

The next time a headline announces that a compound shows promise, find the rung. Ask what the study could not have shown at that stage, and what the next rung would need to show before the claim should change anyone’s mind. If the answer is a lot, the claim may still be worth following. It just has a long way to climb.