Bispecific Dual‐Immune Checkpoint Inhibitor Associated Cutaneous Toxicity: A Report of Lorigerlimab Adverse Skin Reaction in Two Cancer Patients

ABSTRACT

Lorigerlimab is a dual bispecific antibody (BsAb) targeting cytotoxic T‐lymphocyte‐associated protein 4 and programmed cell death protein 1 that is used for treatment of advanced solid cancers such as metastatic castration‐resistant prostate carcinoma. Reported cutaneous immune‐related adverse events (irAEs) of lorigerlimab, such as pruritus and rash, are mostly manageable. However, to the best of our knowledge, the clinical histopathologic features of irAEs of the skin to dual immune checkpoint blockade with BsAbs have not been characterized. We report herein on lorigerlimab‐associated lichenoid and vacuolar interface dermatitis in two patients with metastatic castration‐resistant prostate adenocarcinoma. Case 1 had a violaceous, papular eruption and a lichenoid, perivascular histopathologic reaction pattern that responded well to temporary withholding of lorigerlimab and to corticosteroids. Case 2 presented with a more severe rash characterized by mixed clinical features of psoriasiform, eczematous, and morbilliform eruption and a histopathologic pattern of vacuolar interface changes with superficial perivascular lymphocytic infiltrate, scattered dyskeratotic keratinocytes, and focal acantholysis. The eruption was refractory to treatment, despite the patient withholding lorigerlimab, leading to complete discontinuation of lorigerlimab. This report highlights the importance of histopathologic evaluation in guiding treatment decisions for lorigerlimab‐associated irAEs and underscores the need for further research into its dermatologic irAEs.

1. Introduction

Bispecific antibodies (BsAbs), which target two different antigens or epitopes, have become a part of treatment for hematologic malignancies and some advanced solid tumors. By recognizing two distinct antigens, they can recruit immune effector cells to attack tumor‐specific targets. Based on their mechanisms of action, BsAbs can be classified into different groups, including bispecific T‐cell engagers, immune checkpoint inhibitors (ICIs), and signaling pathway blockers [1]. BsAbs are reported to have relatively manageable side effects, if mild and include cytokine release syndrome and neurotoxicity; however, infrequently severe, life‐threatening forms of these reactions that lead to organ dysfunction may occur and require intensive clinical monitoring and therapies [2, 3].

Tumor‐infiltrating lymphocytes frequently co‐express cytotoxic T‐lymphocyte‐associated protein 4 (CTLA‐4) and programmed cell death protein 1 (PD‐1), which is uncommon in peripheral T lymphocytes [4]. Lorigerlimab is a dual immune checkpoint–inhibiting BsAb that works via targeting CTLA‐4 and PD‐1. It has been effective in treating castration‐resistant prostate carcinoma when used in combination with chemotherapy, with fatigue, pruritus, skin rash, pyrexia, and hypothyroidism reported as common immune‐related adverse events (irAEs) [5]. The clinical manifestations of lorigerlimab associated adverse reactions have not been described in detail and are mostly reported as a maculopapular rash [5]. Adverse skin reactions to other PD‐1 × CTLA‐4 BsAbs (e.g., cadonilimab) may exhibit clinical features of erythematous macules, papules, and patches, and rarely as erythroderma with blistering or toxic epidermal necrolysis‐like features [6, 7]. However, to the best of our knowledge, the pathologic features of cutaneous irAEs of dual immune checkpoint–inhibiting BsAbs have not been fully characterized. Herein we report on two patients with cutaneous irAEs associated with lorigerlimab use.

2. Case Reports

2.1. Case 1

A 64‐year‐old man with metastatic castration‐resistant prostate cancer with bone involvement and prior treatment failure was started on docetaxel and lorigerlimab. One week after the first cycle, he experienced a pruritic eruption characterized by pink to violaceous papules on the anterior and posterior trunk (Figure 1). A skin punch from a representative papule on the back demonstrated a band‐like, perivascular lymphohistiocytic inflammatory infiltrate with scattered pigmented macrophages and rare dyskeratotic cells (Figure 2). Eosinophils were not identified. Periodic acid‐Schiff (PAS) stain and blood cultures were negative for microorganisms. Overall, the histopathologic features were consistent with a lichenoid irAE associated with lorigerlimab therapy.

FIGURE 1.

Clinical presentation of multiple violaceous papules on the back in Case 1. The marked area in the photograph corresponds to the biopsy site.

FIGURE 2.

Lichenoid dermatitis associated with lorigerlimab therapy in Case 1. (A, B) Histopathologic examination using hematoxylin and eosin (H&E) stains revealed band‐like and perivascular lymphohistiocytic infiltrates at a magnification of ×40 (A) and ×100 (B). (C, D) H&E stains of the epidermis showed acanthosis, mild spongiosis, vacuolar changes at the dermal–epidermal junction, and scattered dyskeratotic cells at a magnification of ×200 (C) and ×400 (D).

The patient was first managed with topical corticosteroids and oral antihistamines with no improvement. The rash progressed to the trunk and extremities forming coalescing plaques. Additionally, he developed asymptomatic, subtle reticulated white patches on the buccal mucosa consistent with lichenoid mucositis. Lorigerlimab was held for one cycle. A short prednisone course and acitretin were added to the treatment regimen with gradual rash improvement. Upon oral steroid taper, while continuing acitretin, the eruption flared on the feet; however, it responded to higher potency topical corticosteroids. The patient tolerated lorigerlimab rechallenge with no further significant skin exacerbations. Acitretin was discontinued after 3 months.

2.2. Case 2

A 62‐year‐old man with metastatic castration‐resistant prostatic adenocarcinoma was started on docetaxel and lorigerlimab. A mild, itchy, transient rash developed 2 weeks after the first cycle. This rash recurred 3–4 days after the second cycle, appearing on the back and chest. It was described as Grade 1 eczematous dermatitis involving less than 10% of the body surface area that was managed with topical corticosteroids. At this time, the urticarial phase of bullous pemphigoid (BP) was also considered, but serum‐specific antibodies were negative for BP autoantibodies. After two doses, lorigerlimab was held due to immunotherapy‐associated myocarditis and myasthenia gravis. Nonetheless, his cutaneous eruption continued to worsen and generalize into pruritic pink plaques with sparing of areas of skin folds on the trunk, exhibiting a mixed clinical pattern of eczematous, morbilliform, and psoriasiform lesions without mucosal involvement (Figure 3). A skin punch demonstrated vacuolar interface alteration of the dermal‐epidermal junction, mild superficial perivascular lymphohistiocytic infiltrate, scattered dyskeratotic cells, and focal intraepithelial acantholysis (Figure 4). PAS stain and blood cultures were negative for microorganisms. The overall findings were interpreted as a vacuolar interface and perivascular irAE.

FIGURE 3.

Clinical presentation of erythematous, eczematous papules and plaques on the trunk in Case 2. The marked area in the photograph corresponds to the biopsy site.

FIGURE 4.

Interface dermatitis with focal acantholysis associated with lorigerlimab therapy in Case 2. (A, B) Histopathologic examination using H&E stains revealed cell‐poor vacuolar interface changes along the dermal–epidermal junction (DEJ) and dermal perivascular lymphohistiocytic infiltrates at a magnification of ×40 (A) and ×100 (B). (C, D) H&E stains of the epidermis showed few lymphocytes along the DEJ and scattered dyskeratotic cells at a magnification of ×200 (C) and a focal area of suprabasal acantholysis at a magnification of ×400 (D).

Due to worsening rash, he was started on oral acitretin, along with IV methylprednisolone, which was transitioned to oral corticosteroids after 2 days. Upon attempts to taper systemic steroids, his eruption continued to flare with interval development of scaly plaques on the soles, despite being on acitretin and topical corticosteroids. The decision was made to initiate biologic therapy with risankizumab, an interleukin‐23 inhibitor used for the treatment of psoriasis, as clinical concern for a psoriasiform eruption was favored over a lichenoid eruption. He demonstrated ongoing clinical improvement. Systemic steroids were tapered to baseline dose for the management of neurotoxicity and acitretin was discontinued after 4 months.

3. Discussion

ICIs, including monoclonal anti‐PD‐1 and anti‐CTLA‐4 antibodies, are known to cause a variety of inflammatory cutaneous irAEs. These include interface and lichenoid dermatitis, dermal hypersensitivity reactions, psoriasiform dermatitis, bullous pemphigoid, erythema nodosum, toxic epidermal necrolysis/Stevens–Johnson syndrome, Grover disease, scleroderma, and sarcoid‐like reactions [8, 9].

Lichenoid irAEs of monospecific ICIs represent a well‐recognized histopathologic reaction pattern and are commonly encountered in biopsies [10]. Lichenoid irAEs may exhibit histopathologic features identical to those of lichen planus (LP) or may exhibit features of LP with additional findings of spongiosis, perivascular lymphohistiocytic infiltrates in the reticular dermis, and eosinophils. A feature of lichenoid irAEs that aids in the distinction from lichen planus is the histopathologic variation that may be seen in multiple biopsy samples taken from different anatomic sites at the same time for a patient receiving ICIs [10]. Other possible morphologies include hyperplastic lichenoid irAE that mimics superficially invasive squamous cell carcinoma and lichenoid irAE with lymphocytic atypia that mimics mycosis fungoides [11, 12]. Furthermore, there are cases with acantholytic lichenoid irAE. This lichenoid irAE pattern may be suprabasal and thus histopathologically mimics pemphigus vulgaris or be focal resembling Grover disease [13]. Rarely, there may be an immunobullous acantholytic lichenoid irAE (e.g., paraneoplastic pemphigus‐like reaction) from ICI therapy may occur [14]. Awareness of these morphologic entities is critical in evaluating lichenoid irAEs in patients receiving ICI therapy.

Treatment with dual immune checkpoint–inhibiting BsAbs has also been associated with skin rash in 1.9% of patients receiving therapy, although detailed histopathologic features associated immune checkpoint blockade with dual BsAbs are lacking [15]. One case report described a mucosal lichenoid reaction caused by cadonilimab, a dual immune checkpoint–inhibiting BsAb targeting PD‐1 and CTLA‐4 used to treat hepatocellular carcinoma [16]. Biopsy examination in both of our cases demonstrated both vacuolar interface dermatitis that exhibited either low (cell‐poor) or high (cell rich) density of inflammation along with additional features. In Case 1, which was clinically more lichenoid than Case 2, pathology revealed cell‐rich, band‐like lymphohistiocytic infiltrate and mild spongiosis, and perivascular infiltrate. In Case 2, the initial clinical presentation was a mixed pattern of lichenoid, eczematous, and psoriasiform features. Histopathology demonstrated a cell‐poor, vacuolar interface and perivascular lymphohistiocytic dermatitis. The mixed histologic patterns observed in both cases are common findings in patients with drug reactions. However, neither case had evident eosinophils, a finding typically associated with drug‐induced skin eruption. The persistence of the rash in Case 2 despite discontinuation of lorigerlimab may be attributed to sustained immune activation triggered by treatment with this dual immune checkpoint–inhibiting BsAb, a phenomenon that has been observed with monospecific ICI therapy [17, 18].

In summary, with the use of BsAbs on the rise, further studies aimed at characterizing their associated cutaneous adverse reactions are necessary to improve management of irAEs in patients treated with these novel antibodies.

Ethics Statement

This case report was conducted in accordance with the ethical principles of MD Anderson.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.