Subacute Cutaneous Lupus Erythematosus-Like Eruption Induced by Durvalumab: A Case Report and Literature Review

Introduction

Immune checkpoint inhibitors (ICIs), which target cytotoxic T-lymphocyte-associated protein-4 (CTLA-4), programmed cell death 1(PD-1) and programmed cell death-ligand 1 (PD-L1), have been established as a dominant paradigm in management of advanced malignancies. Blocking specific molecular targets inhibits cancer cell growth, progression, and spread, often resulting in long-term survival. However, by increasing the activity of the immune system, these blockades can lead to inflammatory effects, which are often termed immune-related adverse events.¹ These events have a significant impact on patient’s quality of life and may impair ICIs treatment efficacy due to ICIs dose-limiting effects. Immune-related cutaneous adverse events (irCAEs) are the most common and earliest to occur in ICIs patients; thus, understanding their clinicopathologic features and effective management strategies is critical to a successful oncodermatologic practice.² Recently, cutaneous connective tissue disease-like eruptions have been observed in patients receiving ICIs, with subacute cutaneous lupus erythematosus (SCLE) being the most common manifestation.³

The newly developed anti-PD-L1 agents also have emerged in drug-related cutaneous adverse events. Regarding the National Comprehensive Cancer Network (NCCN) guidelines, durvalumab is a PD-L1 blocker which has been approved by United State FDA in 2020 as a first-line treatment for adult patients with extensive-stage small cell lung carcinoma (ES-SCLC) in combination with platinum-etoposide as a result of the Phase III CASPIAN trial.²

Case Report

A 74-year-old male with no known underlying disease developed chronic productive cough for 9 months. He had history of heavy smoking (50 pack-years) for several decades. The initial chest radiography revealed a lung mass. His computerized tomography (CT) demonstrated a large lobulated lung mass at the right lower lobe with surrounding reticulonodular opacities. A transbronchial lung biopsy was performed and revealed SCLC on histopathology. He also had leptomeningeal and ipsilateral pleural metastasis. As a result, the diagnosis of ES-SCLC was made. The oncologist began treatment with standard-of-care chemotherapies including carboplatin and etoposide. One month later, he received the second cycle of standard chemotherapies plus durvalumab (Imfinzi^®). The therapeutic agents were administered intravenously once a month with no dose adjustments. Before the fourth cycle of treatment (after receiving 3 infusion cycles of durvalumab), he noticed asymptomatic erythematous skin lesions on the extensor surface of his forearms and back. He frequently engaged in outdoor activities with insufficient photoprotection. Three weeks after the fourth cycle, the skin eruptions worsened, becoming darker and more widespread on his face, neck, trunk, and extensor surface of upper extremities. Dermatologic consultation was made. Physical examination showed multiple dusky red to brownish papules and patches covered by scales and crusts with some erosions predominately on sun-exposed areas (face, upper chest, back, dorsum of both forearms) and spared the mucosa (Figure 1A–D). Nikolsky’s sign was positive. A punch biopsy obtained from the lesion on the back revealed a superficial perivascular infiltration and epidermal atrophy with marked interface change (Figure 2A). Remarkably, thin epidermis was necrosed and separated with few dysmaturation of atypical basal keratinocytes (Figure 2B). ANA was detected by using indirect immunofluorescence antibody (IFA) technique showing a titer of 1:320 with a fine speckled pattern. Anti-Ro/SSA was positive (2+) while anti-La/SSB and anti-histone were negative. The erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) levels were elevated. Durvalumab-induced SCLE was diagnosed in our patient based on history, clinicopathological and serological findings. There was no evidence of systemic lupus erythematosus.

Figure 1 Multiple dusky red to brownish papules and patches covered by scales and crusts with some erosions predominately on face (A), upper chest (B), back (C), and dorsum of both forearms (D).

Figure 2 Superficial perivascular infiltration of lymphocytes and melanophages with marked interface change, H&E 100X (A) Epidermal atrophy and few atypical basal keratinocytes, H&E 400X (B). Abbreviation: H&E, hematoxylin and eosin.

Because the skin lesions progressed to grade 3 (covering more than 30% of body surface area [BSA] with moderate pruritus) according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0,³ all chemotherapies and durvalumab were discontinued. The patient received 1 mg/kg/day of oral prednisolone for the first 2 weeks, followed by hydroxychloroquine (HCQ) 200 mg daily and topical 1.25% hydrocortisone ointment twice daily. Due to significant clinical improvement, topical and systemic corticosteroid was discontinued in the next 2 weeks. Overall, the skin lesions resolved with post-inflammatory hyperpigmentation within one month of drug withdrawal (Figure 3A–D). Furthermore, the patient was advised to undergo strict photo-protection. The standard chemotherapies were resumed without durvalumab and the same hydroxychloroquine dose was maintained for 5 months. During the fifth and sixth cycles of chemotherapies, no new lesions appeared. A CT chest follow-up after 6 months of chemotherapies revealed a reduction in the size of right lung mass with a slightly increased small right pleural effusion and nodular pleural thickening. Unfortunately, the patient failed to follow-up with the oncologist after receiving the sixth cycle of chemotherapies. He passed away 3 months later due to disease progression with no cutaneous eruption recurrence.

Figure 3 One month after Durvalumab was discontinued, the lesions resolved with post inflammatory hyperpigmentation on face (A), upper chest (B), back (C), and dorsum of both forearms (D).

Discussion

Checkpoint inhibitors are monoclonal antibodies that block CTLA-4, PD-1, or PD-L1 leading to anti-tumor immune response. They can cause a wide range of side effects, most of which are immune-related adverse events that manifest in the gastrointestinal tract, liver, and skin, though any organ can be affected. The diverse groups of cutaneous reactions have been implicated in checkpoint inhibitions. Maculopapular rash, pruritus, psoriasiform and lichenoid eruptions are the most prevalent subtypes.³ Immune-related adverse events occur in 70% of patients treated with PD-1 or PD-1 inhibitor, typically in cutaneous system. Bullous pemphigoid-like, vitiligo-like, and psoriasiform eruptions are frequently reported.³

A large retrospective cohort study found that only 0.025% of patients developed skin lesions resembling de novo cutaneous connective tissue disease. Median time to onset of skin lesions was 8 months (0.5–26). SCLE is the most common, accounting for 72.7% of all cases, followed by SLE with photosensitivity rash, eosinophilic fasciitis, and dermatomyositis (all 9.1%).⁴ To date, atezolizumab is the only PD-L1 inhibitor associated with SCLE. Table 1 demonstrates a literature review on immune checkpoint inhibitors-associated CLE.

Table 1 Review of Reports on Checkpoint Inhibitors-Associated CLE

Durvalumab has been approved as a first-line treatment for adult patients with ES-SCLC in combination with standard-of-care chemotherapies, etoposide plus either carboplatin or cisplatin (platinum-etoposide). The regimen demonstrated a statistically significant improvement in overall survival. The most common adverse events of durvalumab (more than 20%) in patients with ES-SCLC are neutropenia, anemia, nausea, and alopecia.² The increase in frequency of dermatologic toxicities has been noticed suggesting the possible association between the drug and cutaneous adverse events. The cutaneous lesions possibly associated with durvalumab consist of psoriasiform eruptions,⁵ cutaneous sarcoidosis,⁶ and dermatomyositis.⁷ To the best of our knowledge, no case of durvalumab-related CLE has ever been documented in the literature.

According to our case, the dusky red lesions and skin denudation can be explained by severe apoptotic injury of the epidermis, which correlates with epidermal necrosis and separation in histologic features. Interface dermatitis with foci of vacuolar alteration of basal keratinocytes alternating with areas of lichenoid dermatitis, prominent epidermal atrophy, follicular plugging, and mild or absent basement membrane thickening are all classic features of SCLE skin biopsy.⁹ The median time interval between drug exposure and the appearance of drug-induced SCLE (DI-SCLE) was reported to be 6 weeks (3 days to 11 years). The period of ICIs use before the onset of SCLE appears longer, averaging 4 months (2 weeks to 20 months).^8,9 In our case, anti-cancer drugs consisting cytotoxic chemotherapies and durvalumab were the new medications initiated before the skin eruption in preceding 3 months and 2 months, respectively. However, DI-SCLE due to chemotherapeutic agents is uncommon, with few published cases associated with taxanes, tamoxifen, capecitabine, 5-fluorouracil, pemetrexed, and carboplatin.¹⁰ There is no report on etoposide-associated SCLE in the literature. Moreover, anti-Ro/SSA can be positive in up to 80% in patients with ICIs-associated SCLE, whereas anti-La/SSB positivity occurs in only 25%.⁹ The percentage of anti-Ro/SSA and ANA positivity among DI-SCLE associated with other drugs and ICIs-related SCLE were roughly the same, at 80% and 82%, respectively. However, positive anti-La/SSB is more common in classic DI-SCLE, accounting for approximately 48%.⁸ Anti-histone antibodies are almost always positive in DI-SLE and patients usually lack cutaneous manifestations. Approximately one-third of the classic DI-SCLE triggers (ie, thiazide diuretics, calcium channel blockers, and terbinafine) has also been reported to be associated with anti-histone antibodies;⁸ however, our patient had a negative result and no systemic involvement. Future studies are required to determine the frequency and pattern of LE-associated autoantibodies in ICI-related SCLE.

We conclude that the clinicopathological features, drug history and its compatible temporal relationship with rash development and positive lupus-associated serology (ie, ANA and anti-Ro/SSA) makes durvalumab associated SCLE-like eruption the most likely diagnosed. In this patient, the skin lesion resolved within one month of discontinuing immunotherapy, without recurrence after resuming the original chemotherapies.

The pathomechanism of ICIs-associated SCLE remains unclear, but many potential theories have been proposed. Firstly, the multi-hit hypothesis presumes that the PD-1 or PD-L1 blockade may unmask an immune response to a previously tolerated medication, resulting in SCLE phenotype.^11,12 Another mechanism is that SCLE may be stimulated by ultraviolet (UV)-B radiation resulting in translocation of Ro (SSA) antigen and increase of cell membrane antigen expression via epitope spreading.^13–15 Our patient frequently engaged in outdoor activities with inadequate photoprotection which may have aggravated the symptoms.¹⁶ Lastly, focusing on mechanism of action, anti-PD-1 or PD-L1 may modulate humoral immunity which enhances pre-existing autoantibodies and unmask latent auto-immunity.^17–19 Therefore, anti-PD-L1 might be involved with immune-related adverse events, not only by increasing the activity of the immune system against antigens in both cancers and healthy tissue, but also by increasing levels of pre-existing autoantibodies. These mechanisms may explain the positive autoantibodies in our case.

No prospective trials have defined the best treatment approach for immune-related cutaneous adverse events. Systemic corticosteroids may be indicated for widespread or severe disease while moderate- to high-potency topical corticosteroids are first-line treatments for mild to moderate eruptions.^3,20–22 In our case, the rash dramatically improved with 1 mg/kg/day of oral prednisolone. The use of antimalarial drugs has typically been recommended as the first-line systemic treatment for CLE, with response rates ranging from 31% to 91%.²³ There is no strong evidence on using antimalarials in treatment for ICIs-associated SCLE. However, HCQ administered at 200–400 mg/day has been employed with good outcome and without any adverse events.^9,24–28 In a previous study, approximately 40% showed serologic improvement at the 8-month follow-up in drug-induced CLE with positive anti-Ro/SSA.²⁹

In conclusion, ICIs are increasingly used in cancer therapy, the clinical-pathologic correlation is important to diagnose specific dermatologic adverse events. Prospective trials are required to define the best treatment approaches and recommendations for minimizing immunotherapy interruption and avoiding immunosuppressive medications whenever possible.