Recommendations on Managing Cutaneous Side Effects of ICIs

WASHINGTON, DC — Cutaneous immune-related adverse events (cirAEs) in oncology patients receiving immune checkpoint inhibitors (ICIs) should be treated in as targeted a fashion as possible and with judicious usage and dosing of prednisone when deemed necessary, Blair Allais, MD, said during a session on supportive oncodermatology at the ElderDerm conference on dermatology in the older patient hosted by the George Washington University School of Medicine and Health Sciences.

“It’s important when you see these patients to be as specific as possible” based on morphology and histopathology, and to treat the rashes in a similar way as in the non-ICI setting,” said Allais, a dermato-oncologist at the Inova Schar Cancer Institute in Fairfax, Virginia.

cirAEs are the most frequently reported and most visible adverse effects of checkpoint inhibition — a treatment that has emerged as a standard therapy for many malignancies since the first ICI was approved in 2011 for metastatic melanoma.

And contrary to what the phenomenon of immunosenescence might suggest, older patients are no less prone to cirAEs than younger patients. “You’d think you’d have fewer rashes and side effects as you age, but that’s not true,” said Allais, who completed a fellowship in cutaneous oncology after her dermatology residency.

A 2021 multicenter international cohort study of over 900 patients aged ≥ 80 years treated with single-agent ICIs for cancer did not find any significant differences in the development of immune-related adverse events among those younger than 85, those 85-89 years of age, and those 90 and older. Neither did the ELDERS study in the United Kingdom; this prospective observational study found similar rates of high-grade and low-grade immune toxicity in its two cohorts of patients ≥ 70 and < 70 years of age.

Allais, who coauthored a 2023 review in Immunological Reviews of cirAEs from ICIs, reviewed recent developments and provided the following advice:

New diagnostic criteria: “Really exciting” news for more precise diagnosis and optimal therapy of cirAEs, Allais said, is a position paper published this year in the Journal for ImmunoTherapy of Cancer that offers consensus-based diagnostic criteria for the 10 most common types of dermatologic immune-related adverse events and an overall diagnostic framework. “Luckily, through the work of a Delphi consensus group, we can now have [more diagnostic specificity],” which is important for both clinical care and research, she said.

Most cirAEs have typically been reported nonspecifically as “rash,” but diagnosing a rash subtype is “critical in tailoring appropriate therapy that it is both effective and the least detrimental to the oncology treatment plan for patients with cancer,” the group’s coauthors wrote.

The 10 core diagnoses include psoriasis, eczematous dermatitis, vitiligo, Grover disease, eruptive atypical squamous proliferation, and bullous pemphigoid. Outside of the core diagnoses are other nonspecific presentations that require evaluation to arrive at a diagnosis, if possible, or to reveal data that can allow for targeted therapy and severity grading, the group explains in its paper.

“To prednisone or not to prednisone”: The development of cirAEs is associated with reduced mortality and improved cancer outcomes, making the use of immunosuppressants such as corticosteroids a therapeutic dilemma. “Patients who get these rashes usually do better with respect to their cancer, so the concern has been, if we affect how they respond to their immunotherapy, we may minimize that improvement in mortality,” said Allais, also assistant professor at the University of Virginia, Charlotteville, Virginia, and clinical assistant professor of dermatology at Geroge Washington University, Washington, DC.

A widely discussed study published in 2015 reported on 254 patients with melanoma who developed an immune-related adverse event during treatment with ipilimumab — approximately one third of whom required systemic corticosteroids — and concluded that systemic corticosteroids did not affect overall survival or time to (cancer) treatment failure. This study from Memorial Sloan-Kettering Cancer Center, New York City “was the first large study looking at this question,” she said, and the subsequent message for several years in conferences and the literature was that steroids do not affect the efficacy of checkpoint inhibitors.

“But the study was not without limitations,” Allais said, “because the patients who got prednisone were mainly those with higher-grade toxicities,” while those not treated with corticosteroids had either no toxicities or low-grade toxicities. “If higher-grade toxicities were associated with better (antitumor) response, the steroids may have just [blunted] that benefit,” she said.

The current totality of data available in the literature suggests that corticosteroids may indeed have an impact on the efficacy of ICI therapy, she noted. “Subsequent studies have come out in the community that have shown that we should probably think twice about giving prednisone to some patients, particularly within the first 50 days of ICI treatment, and that we should be mindful of the dose,” Allais said.

The takeaways from these studies — all published in the past few years — are to use prednisone early and liberally for life-threatening toxicity, to use it at the lowest dose and for the shortest course when there is not an appropriate alternative, to avoid it for diagnoses that are not treated with prednisone outside the ICI setting, and to “have a plan” for a steroid-sparing agent to use after prednisone, she said.

Allais recommends heightened consideration during the first 50 days of ICI treatment based on a multicenter retrospective study that found a significant association between use of high-dose glucocorticoids (≥ 60 mg prednisone equivalent once a day) within 8 weeks of anti-programmed cell death protein 1 (PD-1) monotherapy initiation and poorer progression-free and overall survival. The study covered a cohort of 947 patients with advanced melanoma treated with anti-PD-1 monotherapy between 2009 and 2019, 54% of whom developed immune-related adverse events.

This study and other recent studies addressing the association between steroids and survival outcomes in patients with immune-related adverse events during ICI therapy are described in Allais’ 2023 review of cirAEs from ICIs.

Approach to morbilliform eruptions: This rash is “super common” in patients on ICIs, occurring generally within 2-3 weeks of starting treatment. “It tends to be self-limited and can recur with future infusions,” Allais said.

Systemic steroids should be reserved for severe or refractory eruptions. “Usually, I treat the patients with topical steroids, and I manage their expectations (that the rash may recur with subsequent infusions), but I closely follow them up” within 2-3 weeks, she said. It’s important to rule out a severe cutaneous adverse drug eruption, of course, and to start high-dose systemic steroids immediately if necessary. “Antibiotics are a big culprit” and often can be discontinued, she noted.

Soak and smear: “I’m obsessed” with this technique of a 20-minute soak in plain water followed by application of steroid ointment, said Allais, referring to a small study published in 2005 that reported a complete response after 2 weeks in 60% of patients with psoriasis, atopic dermatitis, and other inflammatory skin conditions (none had cancer), who had failed prior systemic therapy. All patients had at least a 75% response.

The method offers a way to “avoid the systemic immunosuppression we’d get with prednisone,” she said. One just needs to make sure the older patient can get in and out of their tub safely.

ICI-induced bullous pemphigoid (BP): BP occurs more frequently in the ICI setting compared with the general population, with a median time to development of 8.5 months after ICI initiation. It is associated in this setting with improved tumor response, but “many oncologists stop anticancer treatment because of this diagnosis,” she said.

In the supportive oncodermatology space, however, ICI-induced BP exemplifies the value of tailored treatment regimens, she said. A small multi-institutional retrospective cohort study published in 2023 identified 35 cases of ICI-BP among 5636 ICI-treated patients and found that 8 out of 11 patients who received biologic therapy (rituximab, omalizumab, or dupilumab) had a complete response to ICI-BP without flares following subsequent ICI cycles. And while statistical significance was not reached, the study showed that no cancer-related outcomes were worsened.

“If you see someone with ICI-induced BP and they have a lot of involvement, you could start them on steroids and get that steroid-sparing agent initiated for approval…And if IgE is elevated, you might reach for omalizumab,” said Allais, noting that her favored treatment overall is dupilumab.

Risk factors for the development of ICI-induced BP include age > 70, skin cancer, and having an initial response to ICI on first imaging, the latter of which “I find fascinating…because imaging occurs within the first 12 weeks of treatment, but we don’t see BP popping up until 8.5 months into treatment,” she noted. “So maybe there’s a baseline risk factor that could predispose them.”

Caution with antibiotics: “I try to avoid antibiotics in the ICI setting,” Allais said, in deference to the “ever-important microbiome.” Studies have demonstrated that the microbiomes of responders to ICI treatment are different from those of nonresponders, she said.

And a “fascinating” study of patients with melanoma undergoing ICI therapy showed not only a higher abundance of Ruminococcaceae bacteria in responders vs nonresponders but a significant impact of dietary fiber. High dietary fiber was associated with significantly improved overall survival in the patients on ICI, with the most pronounced benefit in patients with good fiber intake and no probiotic use. “Even wilder, their T cells changed,” she said. “They had a high expression of genes related to T-cell activation…so more tumor-infiltrating lymphocytes.”

A retrospective study of 568 patients with stages III and IV melanoma treated with ICI showed that those exposed to antibiotics prior to ICI had significantly worse overall survival than those not exposed to antibiotics, she noted. “Think before you give them,” Allais said. “And try to tell your older patients to eat beans and greens.”

Allais reported having no relevant disclosures.