Psoriasis Vulgaris Exacerbation during Treatment with a PD-1 Checkpoint Inhibitor: Case Report and Literature Review

The use of immune checkpoint inhibitors (ICI) is exponentially increasing as it has become the standard of care for several cancer types. Currently, 2 types of ICI are used in the clinic: first, anti-cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) and second, anti-programmed cell death 1 (PD-1) or anti-programmed death ligand 1 (PD-L1) inhibitors. Inherent to the mechanism of action, immune-related adverse events (irAEs) are seen. Every organ is at risk, but skin toxicity is among the most frequent adverse events. Nonspecific maculopapular rash and pruritus represent the most common manifestations [1]. Other entities are less frequent and not so well documented. We present a case of psoriasis vulgaris exacerbation in a patient treated with nivolumab (anti-PD-1). The purpose of this paper is to point to the possibility of psoriasis vulgaris exacerbation as a potential irAE and to discuss the management of this adverse event.

A 65-year-old woman presented with multiple itchy erythematosquamous plaques on both lower and upper limbs ongoing for 1 week. In addition, psoriasiform scales on the scalp and retroauricular were observed. Full-body inspection revealed no other skin lesions. The patient was in good general condition. She had no other complaints and felt generally well. She denied systemic complaints such as weight loss, night sweats, fever, dyspnea, cough, or gastrointestinal symptoms. The patient had been treated with nivolumab 3 mg/m² (anti-PD-1) every 2 weeks for a stage IV melanoma. At the time of presentation, the treatment had been administered 11 times. Two years ago, she was diagnosed with a stage IV melanoma, positive for the BRAF V600 mutation. She was diagnosed with lymph node, subcutaneous, and brain metastases. Nivolumab was initiated as a third-line treatment, after a BRAF enzyme inhibitor and ipilimumab (CTLA-4 inhibitor). Ipilimumab was stopped after 2 cycles because of grade 3 diarrhea. Since the treatment with nivolumab, a disease stabilization was observed. The patient had a known history of scalp psoriasis, type II diabetes, and hypertension. Her regular medication included lorametazepam, gliclazide, metoprolol, pantoprazole, and momethasone nasal spray.

Based on the clear clinical image, the patient was diagnosed with a psoriasis vulgaris exacerbation. The clinical image is shown in Figure 1. No skin biopsy was obtained. Local treatment with corticosteroids was initiated. Additionally, on patient request, the interval of nivolumab was extended from 2 weeks to 3 weeks, because she noticed a flare-up of the skin lesions after every nivolumab administration. Dosing at 3-week intervals, in combination with the local corticoid treatment, led to successful control of the psoriatic lesions. No systemic corticoids were administered. The patient had a stable disease for 14 months after the start of nivolumab. Then, she developed progressive brain metastasis with an intracranial hemorrhage and died.

A bibliographic search was conducted on PubMed using the key words: “psoriasis” and “nivolumab,” “pembrolizumab,” “atezoluzimab,” “anti PD-1,” or “anti PDL-1.” Thirty-four cases with psoriasis linked to anti-PD-1 or anti-PDL-1 treatment were retained. An overview is given in Table 1 and Table 2. Twelve individual cases were described. Two authors published a collection of 17 and 5 cases, respectively, in 1 publication [2, 3]. Two additional publications also reported on psoriasis exacerbation but were not included in the tables because detailed information is missing [4, 5].

We describe an exacerbation of a mild pre-existing psoriasis under anti-PD-1 therapy. An extension of the disease with new localizations was observed. We hypothesize that nivolumab was the trigger to the psoriasis exacerbation in our patient. However, we should mention that a psoriasis exacerbation can be triggered by multiple factors, such as stress and skin injury. In our case, the sequence of events and a clearly observed flare-up of the lesions after each anti-PD-1 infusion suggest a causal link between the administration of nivolumab and the psoriasis exacerbation. Indeed, several cases of psoriasis exacerbation following PD-1 or PDL-1 inhibition have been reported. In Table 1 and Table 2, we describe 34 other cases. The majority of cases show a psoriasis flare in patients with pre-existing disease, but a new-onset disease has been reported in 5 cases. Apart from these case reports, 2 publications on adverse events of anti-PD-1 mention psoriasis exacerbation. Danlos et al. [4] report a psoriasis flare in 4 out of 13 patients with psoriasis and Menzies et al. [5] report a flare in 3 out of 6 patients . The number of cases reported suggests that psoriasis exacerbation after ICI may be more common. However, reliable data on the prevalence and incidence of psoriasis in patients treated with anti-PD-1/PDL-1 antibodies are still lacking. In order to detect side effects at an early stage, it is recommended that these side effects of immunotherapy are searched for during each clinical examination. Registration in a systematic way is needed to obtain reliable epidemiologic data. This might also have a predictive value as some retrospective data suggested a better outcome for patients with cutaneous irAEs [6, 7]. At this moment it is unclear whether this is true for psoriasis induced by ICI.

A possible rationale for the pathogenesis can be given but remains speculative. Psoriasis is known as a T-cell- and dendritic-cell-mediated disease. IL-17 and IL-22 produced by T helper (Th) 1, 17, and 22 cells play an important role [8]. Th cells are downregulated by the PD-1 pathway. By inhibiting this pathway, an upregulation of Th-17 cells is observed [9]. Th-17 upregulation might be the explanation why this patient had psoriasis exacerbation parallel to the anti-PD-1 infusion.

The recognition of psoriasis induced by ICI is important for adequate management. General management for skin toxicity of ICI describes the use of topical emollients, antihistamines, and corticoids [10]. For psoriasis, a more specific approach is needed. Prior recommendations for drug-induced psoriasis were to stop the causal drugs and start classic treatment options for psoriasis vulgaris. These treatment options are currently topical corticosteroids, vitamin D analogues, ultraviolet-based phototherapy, and systemic treatments, such as methotrexate, acitretin, and fumaric acid esters, and biologics [11]. Because of the underlying malignant condition, cyclosporine and many of the novel biologic agents are preferably avoided in patients with PD-1/PDL-1-induced psoriasis unless other severe irAEs are present, e.g., ICI-induced colitis not responding to systemic corticotherapy [3]. Our patient was successfully treated with topical steroids and prolongation of the dosing interval of nivolumab. Indeed, some cases describe good results with topical treatment with corticoids and vitamin D analogues [12-15]. However, in several cases local treatment was insufficient with need for oral prednisolone [2, 16, 17], acitretin (vitamin A derivate), or phototherapy [2, 16, 18, 19]. One patient with concomitant psoriasis arthritis was also treated with oral methotrexate [20]. In most cases, psoriatic lesions were controlled without the need to permanently discontinue the anti-PD-1/PDL-1 therapy; often, only a short break is reported. If psoriasis exacerbation or de novo psoriasis is suspected, we suggest a rapid referral to a dermatologist to initiate a proper local and if needed systemic treatment. With proper treatment, a discontinuation of the ICI can be avoided in most cases.

Psoriasis exacerbation is an established side effect of PD-1/PDL-1 checkpoint inhibitors with 35 reported cases. Early recognition and management are challenging with no clear guidelines available. A close collaboration between oncologist and dermatologist is mandatory to manage this specific irAE.

Written informed consent was obtained from the patient.

The authors have no conflicts of interest to disclose.

There were no funding sources for this work.