- Autoimmune inflammatory symptoms are a common side effect of immune checkpoint inhibitors for cancer therapy, but as these agents are relatively new, the details are not yet fully understood.
- This study analyzed data from a German registry that collects information from rheumatologists about patients experiencing just such events.
- It found that these events were typically more serious than reported in previous studies, and most patients required therapies beyond corticosteroids to bring symptoms under control.
Autoimmune reactions to immune checkpoint inhibitors (ICIs) in cancer patients referred to rheumatologists were generally more severe than previously reported, German researchers found, with corticosteroids usually insufficient for adequate control.
Analysis of 60 cases from a registry of rheumatologist-treated immune-related adverse events (IRAEs) from ICI treatment revealed that almost one-quarter required hospitalization and 38% had to stop ICIs permanently, according to Didzis Gailis, MD, of Ludwig-Maximilians-Universität in Munich, and colleagues.
Nearly 60% did not resolve with steroids alone and 32 of the 60 went on to receive some type of disease-modifying anti-rheumatic drug (DMARD), including biologic agents, the group reported in ACR Open Rheumatology.
These prevalences are greater than in other studies, in which cases weren't necessarily limited to those treated by rheumatologists. Gailis and colleagues cited one study, as an example, in which 15% of patients permanently discontinued ICI therapy. They also pointed to two other studies that found only 10%-15% of IRAE cases treated with DMARDs, only methotrexate or other old-line synthetic products.
Their findings, Gailis's group wrote, highlight "the urgent need for strategies to sustain cancer immunotherapy despite rheumatic toxicity," given the high cancer response rates with ICIs that have revolutionized oncology in recent years.
Study Details
Data for the study came from the ERIN registry, which has collected information on patients with ICI-IRAEs treated by rheumatologists at six German centers since 2024. Patients were included in the analysis as long as symptoms weren't "clearly attributable" to other causes besides ICI treatment but were adequate to support an IRAE diagnosis.
IRAEs arise in this context because ICIs work essentially by unleashing the immune system to attack tumor cells that have blocked this activity. As a result, the newly activated immune cells may sometimes turn their attention to other tissues in the body. Many organ systems can be affected; events can be life-threatening when the heart is the target. Joint pain and inflammation, though, is probably the most common IRAE associated with ICI therapy.
Of the 60 patients included, 34 were men, and the median age was 66. Eleven different cancer types were represented, with melanoma accounting for 40%, non-small cell lung cancer 15%, and urothelial cancer 12%. The most common ICIs were pembrolizumab (Keytruda), nivolumab (Opdivo), and the combination of nivolumab and ipilimumab (Yervoy), but six others were used in a few cases.
Onset of IRAEs after starting ICIs varied considerably, with median lags of 2.1 to 13.7 months; the overall median was 7.2 months. The most common event type was polyarthritis resembling rheumatoid arthritis (RA), seen in 25 cases, followed by polymyalgia rheumatica-like symptoms (11) and symptoms similar to peripheral spondyloarthritis (SpA; seven). Others included vasculitis, newly aggressive osteoarthritis, noninflammatory joint pain, arthritis in a single joint, and a condition resembling axial SpA.
This last event was seen in four patients and Gailis's team found it especially interesting in that their age was substantially younger -- median 46 years, with only one older than 50 -- and the time to onset was relatively short at median 4.8 months. The researchers noted that diagnosing it is complicated insofar as the underlying cancer, past trauma, and even ordinary degenerative processes can produce the same features seen on imaging. "Delineating this entity will require larger cohorts and standardized axial imaging," they wrote.
Although truly severe IRAEs were rare -- just two were rated at grade 4 -- 21 cases were considered moderate (grade 3). Twenty-one cases were treated with conventional synthetic DMARDs and 11 with biologic drugs. Use of DMARDs was especially common with the RA-like and peripheral SpA-like cases. Other IRAE types were too uncommon to support conclusions.
Data were also insufficient to form conclusions about oncologic outcomes: a minority of patients saw cancer progression, a similar proportion reached some level of remission, and the rest had stable disease. The heterogeneity of cancer and IRAE types precluded meaningful further analysis.
Rheumatologic outcomes, however, were clearly disappointing. Only 20 patients had full symptom resolution, although partial remission was achieved by 31. Also, more than one-third of patients receiving DMARDs had to change drugs because of inadequate response or their own adverse effects. Median follow-up was 10 months with an interquartile range of 4.1-31.1, so it's possible that some patients will improve further with longer treatment.
Gailis and colleagues also looked for factors linked with the need for DMARDs. Two were statistically significant: pre-existing rheumatic disease was actually negatively associated (OR 0.09, P=0.013), while male sex quintupled the need for escalation (OR 5.20, P=0.004). Other factors with substantial numerical associations, but falling short of statistical significance, included high baseline C-reactive protein levels (OR 1.75) and ICI combination therapy (OR 4.23).
The study came with a number of limitations, the most notable of which was that these data came from patients referred to rheumatologists, and therefore represented those that were too complex to be managed by other specialists. Thus, the results may not be generalizable to the full span of ICI-IRAE cases. As well, "remission" and severity grades in this context have no universally agreed definition, especially since clinical presentations are so variable. And the number of cases was too small to permit extensive subgroup analysis.
Disclosures
The study had no commercial funding.
Gailis reported having no relevant financial interests. Co-authors reported relationships with multiple pharmaceutical companies and other commercial entities.
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.