Clear skin is a trial result, not a promise

Modern biologics have changed what a person with severe psoriasis can hope for. Fifteen years ago, when I marketed topical dermatology treatments, a good outcome meant fewer plaques and less itching. Today many patients in trials reach PASI 90, a ninety per cent reduction in the severity score, and a substantial minority reach PASI 100, skin that is completely clear. It is natural for companies to put that hope at the centre of their marketing. It is also where marketing in this class most often goes wrong.

I want to argue for a discipline that sounds obvious and is commercially uncomfortable. An efficacy claim should describe what happened in a trial, to whom, compared with what, and it should say who did not benefit. A claim that does less is not only a regulatory risk. It is a small breach of trust with a doctor who will repeat it to a patient.

The chart we pulped

In April 2025 our agency delivered a refreshed campaign for our IL-17A antibody. Its centrepiece was a detail aid comparing our week-12 PASI 100 rate with a competitor's, drawn from two separate trials. The bars looked decisive. But the trials had enrolled different populations, with different baseline severity and different proportions of patients previously treated with biologics, and they had used different designs. A chart setting their results side by side implied a head-to-head result that nobody had produced.

Our medical reviewer flagged it, and I withdrew it. We pulped 38,000 printed detail aids, removed the digital versions and wrote off RMB 6.8 million. Our head of sales argued that competitors used similar charts and that representatives needed a reason for a dermatologist to switch. The first point may have been true; it did not make the chart accurate. The second point was right, and the answer was to give representatives something accurate to say.

The replacement told the story of our own pivotal trial against placebo. Seventy-six per cent of patients reached PASI 90 at week 12 and 41 per cent reached PASI 100. About one patient in nine did not reach PASI 75. The common adverse events, including candida infections, appeared on the same page as the efficacy results rather than on the back. Dermatologists told our medical colleagues that the new materials were more useful, and our share among new patients in hospitals we already served held through the seven-week delay.

Cross-trial comparisons tempt every marketer in a crowded class because they are easy to make and hard for a busy reader to dismantle. Network meta-analyses, properly conducted and published, have their place in clinical discussion and in submissions to payers. A bar chart in a detail aid is not that. If a company believes its medicine is superior, the honest route is a head-to-head trial; if it cannot run one, it should describe its own results well.

Saying whom the medicine does not help

The most valuable sentence in our materials is the one that says some patients do not respond. It reads like a concession. In practice it helps dermatologists choose: whom to start on our antibody, what to look for at week 12, and when to switch. A doctor who has been told honestly about non-response does not feel misled when a patient fails, and is more likely to believe the next thing a representative says.

Non-response also has a shape that doctors can use. In our trial, patients who had not improved meaningfully by week 12 rarely caught up later, and those who had already failed another biologic responded less often than those who had not. Saying so costs a few prescriptions at the margin. It also spares patients months on a medicine unlikely to help them, and spares us their disappointment.

Honest claims matter for safety as well. Blocking IL-17A is associated with mucocutaneous candida infections and calls for caution in patients with inflammatory bowel disease. A campaign built around clear skin tends to push such facts into the small print. When they appear beside the efficacy figures, doctors screen and counsel better, and patients are less likely to abandon treatment in fright at the first infection.

I would extend the same discipline to the outcomes patients care about. Skin clearance is what trials measure; itching, sleep and the confidence to wear short sleeves are what patients describe. Trials capture some of this through quality-of-life measures, and those deserve a place in our materials. What they do not justify is the slide from a trial result to a personal promise: the patient story that implies every viewer will look like the person in the photograph.

Online, where patients read alone

The hardest territory is online. Chinese law confines advertising of prescription medicines to professional medical publications, so a company may not run a brand campaign aimed at patients. But patients search anyway: on online pharmacy product pages, in patient communities and in videos made by people describing their own treatment. Our 2025 campaign included short videos in which a lifestyle influencer described her skin before and after treatment. We cancelled them, because no framing could turn a branded before-and-after story into anything other than patient-facing promotion of a prescription medicine.

What a company can do online is provide disease education without brand names, written for patients and reviewed by medical colleagues: what moderate-to-severe psoriasis is, what treatments exist, what screening a biologic requires, and when to ask a doctor about stopping. When I worked at an online pharmacy I saw how often patients read a product page alone, late at night, with nobody to explain the difference between a trial average and their own prospects. Content that will be read in those conditions should be more careful, not less.

There is a commercial argument here too, and I do not want to hide behind ethics alone. Our antibody is reimbursed, and hospitals decide whether to stock it through committees that read evidence files. A company with a record of accurate claims gets a better hearing in those rooms, and its representatives are more welcome in departments that have learnt to distrust promotion. Honest claims are slower to persuade and much slower to disappoint. In a chronic disease, where a patient may take a medicine for years, that is the better bargain.