Is Stelara Safe Long Term? 5 Years of Crohn's Evidence

Is Stelara safe long term? Across 2,575 ustekinumab-treated patients and 4,826 patient-years in the final pooled inflammatory bowel disease safety analysis, serious infections occurred at 3.71 per 100 patient-years compared with 5.52 on placebo, and cancer rates were close to what you would expect in a comparable population not taking the drug at all (1). Many of us get vague reassurance instead of numbers, or a frightening list with no sense of scale. Here are the actual rates beside placebo, and where the evidence stops. For how Stelara works and is dosed, see our Stelara patient guide.

Key Takeaways

• The longest controlled long term Crohn's dataset runs through 5 years, not 10 or 20, and risk did not climb with added exposure (1)
• Serious infections ran at 3.71 per 100 patient-years on ustekinumab versus 5.52 on placebo, and all infections at 88.87 versus 108.64 (1)
• Cancers excluding skin ran at 0.39 per 100 patient-years versus 0.32 on placebo, and no lymphomas were reported (1)(2)
• Antibodies to ustekinumab appeared in only 5.8% of patients, 31 of 532, across 5 years (2)
• Stelara carries no boxed warning, unlike JAK inhibitors and anti-TNF biologics, but its label still flags possible malignancy risk (5)
• Among 39 pregnancies in the trial program there were 26 live births, all normal newborns, with no congenital anomalies (4)

What "Long Term" Actually Means for Stelara

For Stelara in Crohn's disease, the honest ceiling is 5 years of controlled data, from the final pooled safety analysis of the ustekinumab trial program, covering 5 years in Crohn's and 4 years in ulcerative colitis (1). There is no published 10-year or 20-year randomised Crohn's dataset, and implying otherwise fills a gap with confidence.

Most year-by-year numbers here come from the IM-UNITI long-term extension, which followed maintenance patients to Week 252 (2).

How Far the Crohn's Evidence Really Goes

Beyond 5 years, the real-world picture comes mostly from psoriasis. The PSOLAR registry followed 12,093 patients across 40,388 patient-years and found an unadjusted serious infection rate of 0.93 per 100 patient-years for ustekinumab, against 2.91 for infliximab and 1.91 for other biologics, with no biologic linked to increased malignancy, major adverse cardiovascular events or mortality (3).

That is reassuring, and it is also a different disease, dose and population, so treat PSOLAR as supporting context rather than Crohn's proof. The evidence cannot tell you what year 12 looks like. It can tell you whether risk climbed across 5 years in Crohn's, and it did not.

Infection Risk Over Five Years

In the pooled analysis, serious infections ran at 3.71 per 100 patient-years on ustekinumab against 5.52 on placebo (1,389 patients, 943 patient-years), all infections at 88.87 versus 108.64, and serious adverse events at 18.85 versus 29.39 (1). Every comparison favoured the drug.

It is tempting to read that as protection. It is not. A rate lower than placebo means the trials detected no increased infection signal, not that Stelara defends you against pneumonia. Untreated Crohn's carries its own infection risk, from abscesses and fistulas to the effects of steroids, which is part of why the placebo arms looked worse.

IM-UNITI found the same pattern across 5 years. Per 100 patient-years, ustekinumab versus placebo: all adverse events 327.6 versus 440.3, serious adverse events 17.5 versus 19.3, infections 93.8 versus 99.8, and serious infections 3.4 versus 3.9 (2).

Serious infections reported in the extension included anal abscess, pneumonia, cellulitis, diverticulitis, gastroenteritis, abdominal abscess, perirectal abscess, pyelonephritis, sepsis and cholecystitis (2). Several are complications of Crohn's itself rather than exotic opportunistic infections, which is worth knowing when deciding what to watch for.

Tuberculosis Screening Before You Start

Tuberculosis was rare: three cases across the entire pooled IBD program, one on ustekinumab and two on placebo (1). The label still requires evaluating for TB before starting, and treating latent TB first (5). Latent TB is not a permanent barrier, it is a sequencing question for your IBD team.

Cancer Risk: What the Numbers Show

Cancer is usually the fear underneath this question. In the pooled analysis, malignancies excluding non-melanoma skin cancer occurred at 0.39 per 100 patient-years on ustekinumab versus 0.32 on placebo, and non-melanoma skin cancers at 0.38 versus 0.32 (1).

More informative is the standardised incidence ratio, comparing observed cancers against what SEER general-population data would predict for a similar group. It came out at approximately 1.0 (1): about the number expected in a population not taking the drug.

One absence is worth naming. No lymphomas were reported in the pooled analysis, and none across the IM-UNITI extension (1)(2). For patients who have read about lymphoma risk with thiopurines or anti-TNF therapy, that is a meaningful difference.

None of this rounds down to "no cancer risk." The label states plainly that ustekinumab is an immunosuppressant and may increase malignancy risk, and post-approval reports describe rapid onset of multiple skin cancers in susceptible patients (5).

Skin Cancer Deserves Its Own Mention

Non-melanoma skin cancer ran at 0.38 per 100 patient-years versus 0.32 on placebo (1), counted separately because these behave differently. With the label's note about rapid onset of multiple skin cancers in susceptible people (5), annual skin checks and sun protection are worth raising with your doctor, especially if you have had skin cancer, burn easily, or have had heavy sun or phototherapy exposure.

The FDA Label in Plain Language

Stelara carries no boxed warning (5). That is the FDA's most serious warning format, and its absence is a genuine difference from JAK inhibitors such as upadacitinib and from the anti-TNF biologics, which carry boxed warnings for serious infections and malignancy.

What it does carry is eight Warnings and Precautions, worth reading plainly rather than skimming in fear:

• Infections. May lower your ability to fight infection; do not start during an active infection.
• Theoretical vulnerability to particular infections. People with rare inherited IL-12/IL-23 deficiencies are unusually susceptible to mycobacterial and salmonella infections, so the label flags a theoretical concern.
• Pre-treatment tuberculosis evaluation. Screen before the first dose; treat latent TB first.
• Malignancies. An immunosuppressant that may increase malignancy risk.
• Serious hypersensitivity reactions. Anaphylaxis and angioedema reported.
• Posterior reversible encephalopathy syndrome (PRES). A rare neurological event.
• Immunisations. Live vaccines need planning around treatment.
• Noninfectious pneumonia. Rare lung reactions reported.

Two deserve detail, because rare is not the same as ignorable. PRES has presented with headaches, seizures, confusion and visual disturbances, with recovery after stopping the drug (5); no cases appeared in the pooled IBD program (1). Reported lung reactions include interstitial, eosinophilic and cryptogenic organising pneumonia, and the label says to discontinue if confirmed (5). New breathlessness or a persistent dry cough warrants a call.

On vaccines, the label advises avoiding BCG during treatment and in the year before starting and the year after stopping (5). Non-live vaccines, including flu shots, are generally not restricted the same way. Our guide to vaccines for Crohn's patients covers getting immunisations up to date before your first dose.

Why No Boxed Warning Is Not the Same as No Risk

A boxed warning reflects a regulatory judgement about a specific hazard. Its absence means the FDA has not found a risk severe enough to warrant one. It does not mean nothing on the Warnings list can happen to you.

Does It Keep Working? Antibodies and Durability

Patients ask about durability alongside safety, because a drug that stops working is one you will have to replace. Immunogenicity, the body making antibodies against the drug, stayed low: antibodies were detected in 5.8% of patients, 31 of 532, across 5 years, and 14 of those 31 at a single visit only (2).

Antibody rates were similar with or without a concomitant immunosuppressant: 7.7% of 181 patients on one versus 4.8% of 351 not on one (2). That differs from anti-TNF therapy, where combination therapy with a thiopurine or methotrexate is often used specifically to suppress antibodies. With Stelara, much of that rationale falls away.

Remission numbers need careful reading. At Week 252, remission was 34.4% on the every-8-week dose and 28.7% on every-12-week by intent-to-treat; among patients who actually entered the extension, 54.9% and 45.2% (2). They differ because intent-to-treat counts everyone randomised, including those who left for any reason, as not in remission, while the higher figure covers only patients still in the trial, already selected for doing well. The first is conservative; the second is closer to what to expect if Stelara works for you.

Among those still in the trial, remission declined roughly 5% per year from year 1 to year 5, and antibodies did not clearly predict losing it: 46.7% in remission with antibodies versus 48.2% without (2).

Pregnancy, Family Planning and Vaccines

Across the ustekinumab IBD trials there were 39 pregnancies, 28 in Crohn's and 11 in ulcerative colitis, producing 26 live births, all described as normal newborns with no congenital anomalies (4). The spontaneous abortion rate was 20.5%, 8 of 39, which the authors call generally comparable to the roughly 17% background rate in the US population (4).

We want to be explicit about the limits, because this is where patients are most often given false confidence. Thirty-nine pregnancies is a small sample: it can show that no obvious signal appeared, but cannot rule out an uncommon risk. The FDA label takes the same position, stating that available data have not identified a Stelara-associated risk of birth defects, miscarriage or other adverse maternal or fetal outcomes (5).

Two practical points. The BCG restriction applies to the infant too, since biologics cross the placenta later in pregnancy, so your baby's vaccination schedule is part of the conversation. And third-trimester dosing belongs with your own IBD team rather than any article. Our pregnancy management guide covers the wider picture.

What the Five-Year Picture Adds Up To

Taken together, this is about as reassuring as controlled trial evidence gets for an immunosuppressant: no boxed warning, cancer rates close to general-population expectations, serious infection rates no higher than placebo, no lymphomas, low antibody formation, and no sign any of it climbed year over year.

What it is not is a guarantee, or a 20-year record. If you are weighing whether to stay on Stelara or move to a newer IL-23 inhibitor, safety is one input alongside how well your disease is controlled; our Skyrizi versus Stelara comparison covers that directly. The most useful next step is to bring these numbers to your appointment and ask how they apply to your history, medications and screening needs. Research describes populations; your gastroenterologist can translate it into your decision.

Frequently Asked Questions

How many years of Stelara safety data actually exist for Crohn's disease?

The longest controlled dataset covers 5 years in Crohn's disease and 4 years in ulcerative colitis (1), with the IM-UNITI extension following patients to Week 252 (2). There is no published 10-year or 20-year randomised Crohn's dataset, so claims about two decades of safety are extrapolation, not evidence.

Does Stelara increase cancer risk?

Malignancies excluding non-melanoma skin cancer ran at 0.39 per 100 patient-years versus 0.32 on placebo, with a standardised incidence ratio of roughly 1.0 against general-population expectations and no lymphomas reported (1). The label still calls ustekinumab an immunosuppressant that may increase malignancy risk, with skin cancer the most flagged concern (5).

Why were infection rates lower on Stelara than on placebo?

Because placebo did not mean no disease. Those patients had less controlled Crohn's, which itself drives abscesses, fistulas and steroid use. The figures of 3.71 serious infections per 100 patient-years versus 5.52 on placebo (1) show no increased infection signal. They do not mean the drug protects you from infection.

Will Stelara stop working over time?

Some loss of response happens, but antibodies are rarely the reason. Only 5.8% of patients developed antibodies across 5 years (2). Among those still in the trial, remission declined about 5% per year from year 1 to year 5, and antibodies did not clearly predict losing remission, 46.7% versus 48.2% (2).

Is it safe to stay on Stelara during pregnancy?

Across 39 pregnancies there were 26 live births, all described as normal newborns with no congenital anomalies, and a spontaneous abortion rate of 20.5% against a background rate of about 17% (4). The FDA label has not identified a Stelara-associated risk of birth defects or miscarriage (5), though on a small sample, so dosing decisions belong with your IBD team.

Which vaccines do I need to avoid on Stelara?

The label advises avoiding BCG during treatment, and in the year before starting and the year after stopping, with the restriction also applying to infants exposed in utero (5). Other live vaccines need discussion with your team. Non-live vaccines, including influenza, are handled differently and are easiest to arrange before your first dose.

What should I ask my doctor about long-term Stelara safety?

Useful questions include: has my latent TB screening been done and does it need repeating; do I need annual skin checks given my history; are my vaccinations up to date; what symptoms should prompt me to call rather than wait; and is Stelara still the best option for my disease.

References

1. Safety of Ustekinumab in Inflammatory Bowel Disease: Pooled Safety Analysis Through 5 Years in Crohn's Disease and 4 Years in Ulcerative Colitis. 2024. Read study
2. Five-Year Efficacy and Safety of Ustekinumab Treatment in Crohn's Disease: The IM-UNITI Trial. 2022. Read study
3. Safety Surveillance for Ustekinumab and Other Psoriasis Treatments From the Psoriasis Longitudinal Assessment and Registry (PSOLAR). 2015. View on PubMed
4. Ustekinumab Exposure in Pregnant Women From Inflammatory Bowel Disease Clinical Trials: Pregnancy Outcomes Through Up To 5 Years in Crohn's Disease and 2 Years in Ulcerative Colitis. 2022. Read study
5. Janssen Biotech. STELARA (ustekinumab) injection, solution - FDA Prescribing Information. DailyMed, 2025. Read label