How to read a peptide study: 10 questions

In short: a study is not proof in itself. It becomes proof only when we understand the model, sample size, control, dose, outcome and limitations. The title and abstract are the least reliable parts of the text.

This guide is intended for reading the peptide literature. It does not replace expert judgement and does not provide instructions for use in humans.

1. Who is the author and who funds the study?

Check:

• whether the authors are affiliated with an institution you can verify;
• whether funding is stated;
• whether conflicts of interest are disclosed;
• whether the study is part of a larger programme or a single experiment.

Funding in itself does not mean the study is wrong. But it does mean that the results should be read together with any interests involved.

2. What is the model?

First find out whether it is:

• a cell culture (in vitro);
• an animal model (in vivo);
• a human study (observational or interventional);
• a literature review or meta-analysis.

Each model answers a different question. A result in cells does not prove an effect in humans. A result in animals is an important preclinical signal, not clinical proof.

3. What is the sample size?

A small sample increases uncertainty. With small groups, the observed effect is more likely to be due to chance.

Ask:

• how many people or animals were in each group;
• whether a sample size calculation was performed;
• whether exclusions of participants are explained;
• whether the results are stable across repeated experiments.

4. Was the control adequate?

A control group is essential for separating the effect of a substance from other factors. Check whether it was:

• placebo or vehicle;
• comparison with an active substance;
• the same route of use;
• the same environment and handling.

Without an adequate control, the result is hard to attribute to the substance under study.

5. Was there randomisation and blinding?

Randomisation reduces bias in allocation to groups. Blinding prevents participants or researchers from knowing who received which substance.

In preclinical studies other measures are common, such as random allocation of animals and blinded assessment of outcomes. If this information is absent, caution is needed.

6. What were the dose, route and form?

The result applies to a specific formulation and a specific route of use. Check:

• dose and frequency;
• route (e.g. local, systemic, in the model);
• salt form and purity;
• duration of exposure;
• whether the same form was used as in other studies.

Two studies with the same peptide name are not necessarily comparable if they differ in formulation.

7. Which outcome was measured?

The outcome must be clearly defined and measurable. In peptides, common ones are:

• biochemical markers;
• histological changes;
• behavioural tests;
• healing time;
• molecular signalling.

Ask whether the outcome is a surrogate marker or a clinically important outcome. A change in a marker does not yet mean that a condition important to the patient has improved.

8. What is the effect size and uncertainty?

Statistical significance indicates how likely it is that the result is due to chance. It does not indicate how large the effect is.

Look for:

• effect size;
• confidence interval;
• standard deviation;
• number of measurements;
• whether the effect was replicated in an independent group.

A small effect with a narrow confidence interval may be more reliable than a large effect with a very wide interval.

9. Have the results been independently replicated?

One study is a beginning, not a conclusion. Check whether the results have been replicated by:

• another research group;
• another formulation;
• another model;
• a larger study.

In peptides, much of the literature is concentrated in a limited number of groups. This does not mean the data are wrong, but independent replication is an important part of trust.

10. What do the authors themselves acknowledge as a limitation?

A good study has a limitations section. Check whether the authors mention:

• a small sample;
• a lack of long-term follow-up;
• possible biases;
• differences between the model and humans;
• the need for further research.

If the text promises more than the methods allow, that is a warning sign.

The evidence ladder in practice

The ladder is not absolute. A poorly conducted randomised trial may be less reliable than a well-conducted preclinical programme.

Red flags

• the title promises a treatment, but the study is in cells;
• there is no control group;
• the sample is very small while the conclusions are very broad;
• the dose or form is not stated;
• the results have not been replicated;
• the journal is not clearly indexed;
• the text refers to “studies” without links;
• the authors do not state limitations.

The link to substance documentation

A study describes what happened in a particular experiment. Batch documentation describes what is in a particular vial. These are two different questions.

To understand analytical data, read HPLC and mass spectrometry and How to read a certificate of analysis. For a comparison of two frequently mentioned peptides, the article BPC-157 versus TB-500 is useful.

Conclusion

Reading studies is a skill, not a search for a single sentence that confirms an opinion. Best practice is slow reading: model, sample, control, dose, outcome, uncertainty and limitations.

If a study does not allow a clear answer, that may be an important answer in itself.