New data reveals insights into cancer and cardiovascular safety of JAK inhibitors

Now, the work presented at the 2025 annual EULAR congress in Barcelona adds two important pieces to the puzzle. First, a large-scale real-world study reporting no significantly higher risk of cancer in RA patients treated with JAKi compared to bDMARDs, and second an abstract looking at whether the glucagon-like peptide-1 receptor agonists (GLP-1RA) which are causing waves in many fields including diabetes and obesity might offer cardiovascular protection in RA. 

Romain Aymon and colleagues set out to assess the cancer incidence in RA patients treated with JAKi compared to biologic disease-modifying anti-rheumatic drugs (bDMARD), using data from 13 registers. Cancers were linked to treatments within 5 years of cessation or until follow-up loss, death, or study end, whichever came first. Overall they identified 53,169 treatment initiations in 33,127 patients; 219 non-melanoma skin cancers (NMSC) were reported, plus 638 non-NMSC cancers. The crude incidence of non-NMSC cancer was lower for TNFi than both JAKi and bDMARDs with other mechanism of action (OMA), with rates of 2.2, 2.9, and 3.1 per 1000 patient-years. But further statistical analysis found no significant difference in the incidence of either cancer type between JAKi and TNFi, nor between JAKi and OMA. 

A sub-analysis was performed on patients aged 50 and older who had at least one cardiovascular risk factor – mimicking the inclusion criteria from ORAL Surveillance. This high-risk cohort made up 39.4% of treatment courses, and showed a higher cancer incidence in each treatment group: 3.2, 4.2, and 4.1 per 1000 patient-years for TNFi, JAKi, and OMA, respectively. But similarly to the overall population, there was no significant difference in the incidence of either cancer type between JAKi and either bDMARD group. 

The authors note that further analyses are planned, including additional registries to enhance statistical power and that they will evaluate the cancer incidence across different exposure periods – but in the meantime the findings look reassuring. 

Of note, the incidence of keratinocyte cancer – largely comprised of basal cell and squamous cell carcinoma – is increasing in the general population. This could be due to UV light, but immune alteration has also been linked to increased risk – including in people with RA, but largely with data from the pre-biologic era. New work has set out to assess the risk of keratinocyte cancer in modern RA patients initiating treatment with b/tsDMARDs, including the risk of a second keratinocyte cancer. 21,121 people were identified from the Swedish Rheumatology Quality Register. Overall, 94 keratinocyte cancers were identified in people starting a JAKi, 407 with OMA, and 628 with a TNFi. When compared to TNFi, the hazard ratio for a keratinocyte cancer was 1.72 with JAKi and 0.81 with OMA, Among RA patients with a history of keratinocyte cancers, the hazard ratio for a second diagnosis was 2.76 for JAKi versus TNFi, and 1.54 for OMA versus TNFi. 

Presenting the work, Viking Huss said "these findings suggest that people with RA treated with a JAKi have a higher incidence of a first keratinocyte cancer, primarily driven by basal cell carcinoma. Additionally – although based on a limited number of events – we saw a higher incidence of a second cancer of this type with JAKi compared to TNFi." 

These findings support the notion that skin examination would be beneficial in this population, and skin cancer risks with JAKi treatment should be monitored. 

It is well-known that as well as increased cancer risks, people with RA are prone to cardiovascular complications, and in ORAL Surveillance JAKi were associated with adverse cardiovascular events. Conversely, GLP-1RA have demonstrated possible cardioprotective effects in other patient populations, with some agents now specifically approved for reducing cardiovascular risk. 

Asmaa Beltagy presented findings from a retrospective analysis of two cohorts of RA patients over the age of 40. All 2,449 patients were on JAKi, but one group subsequently initiated GLP-1RA, while the other group did not. The work assessed primary outcomes occurring within 5 years, including the incidence of acute coronary syndromes, cerebral infarction, acute peripheral arterial thrombosis, deep venous thrombosis, and overall arterial events. 

The results show that patients taking a GLP-1RA had a significantly lower risk of acute coronary syndromes and deep venous thrombosis, with a trend towards a lower risk of acute cerebral infarction and peripheral arterial disease events, although the difference was not statistically significant. The overall incidence of cardiovascular events was significantly lower in GLP-1RA group. A survival analysis showed comparable probabilities between the two groups for all studied outcomes, with no significant difference in median survival. 

These findings underscore the potential of GLP-1RA to reduce certain CV risks in people with RA being treated with JAKi. However, not all effects were significant, and there remains a need for further research to validate the results.