Cold-Induced Urticaria Exacerbated During Treatment With Pembrolizumab

Abstract

An 80-year-old woman was assessed for cutaneous edema at the site of her intravenous (IV) access during infusion of normal saline (Figure 1). She had no associated symptoms and was otherwise well. The same reaction occurred in the contralateral arm with dextrose 5%. She was treated with IV diphenhydramine and the reaction subsided. The patient was on a research protocol including cyclophosphamide, pembrolizumab, and antisurvivin tumor vaccine to treat refractory diffuse large B-cell lymphoma. Her previous medical conditions included chronic hepatitis B on tenofovir and hypothyroidism on levothyroxine. She described a more than 20-year history of cold intolerance, reporting local hives with exposure to subzero degree centigrade temperatures, as well as throat tightness and epigastric pain after consuming frozen foods. She managed her symptoms by cold avoidance alone. Her symptom severity increased after starting pembrolizumab. Cold-induced urticaria (CIndU) was suspected on the basis of history and hypothesized as the cause of her cutaneous reaction to room temperature saline. To support the diagnosis of CIndU, skin testing was performed. Latex and chlorhexidine skin prick test (SPT) results were negative, ruling out the possibility of contact urticaria. Results for the ice cube test and SPT with room temperature saline were both positive (60 mm and 40 mm wheals, respectively), with histamine control of 4 mm; SPT result to saline warmed to 30°C was negative. After pretreatment with 20 mg of cetirizine, repeat SPT result to room temperature saline was negative. She then tolerated challenge to an IV infusion of 250 mL of 30°C saline. In the context of her history and diagnostic testing, CIndU triggered by room temperature IV fluid was diagnosed. The patient was started on daily oral antihistamine therapy (20 mg of bilastine). Subsequent IV infusions were successfully administered by heating the IV lines at the point of insertion to 30°C. She was able to receive additional chemotherapy doses in this manner but was moved to palliative care before any additional testing could be performed. Interestingly, although there is often no identified underlying disease trigger for CIndU, the comorbidities of our patient (hypothyroidism, chronic hepatitis B, malignancy) are all established potential etiologies. However, her CIndU predated these health conditions and did not worsen until pembrolizumab therapy. The cold protein diseases (cryoglobulins, cryofibrinogen, or cold agglutinins) should also be considered in the differential of cold-exacerbated diseases, especially cryoglobulinemic vasculitis, which can be linked to chronic hepatitis and malignancy. However, cold proteins have not been clearly linked to the pathophysiology of CIndU, and this patient was not experiencing symptoms compatible with vasculitis or worsening anemia, making cold protein disease unlikely.1Maltseva N. Borzova E. Fomina D. Bizjak M. Terhorst-Molawi D. Košnik M. et al.Cold urticaria – what we know and what we do not know.Allergy. 2020; 76: 1077-1094Crossref PubMed Scopus (30) Google Scholar Why was the patient’s CIndU acutely exacerbated after many years of stable disease? We hypothesized that pembrolizumab might have lowered the patient’s CIndU temperature threshold. Indeed, pembrolizumab is an IgG4 antibody that inhibits the programmed cell death protein 1 (PDI) inhibitor on lymphocytes. PDI is an “immune checkpoint” on activated effector T and B cells that binds to its cognate receptor PD-L1, inhibiting activated lymphocytes. Many cancer cells produce PD-L1 as an immune evasion strategy. Pembrolizumab removes this inhibition, enabling lymphocytes to better target and attack cancer cells. However, pembrolizumab also has many off-target effects, mainly immune-mediated adverse events. Cutaneous adverse events (CAEs) are the most common type of adverse event occurring on pembrolizumab, affecting approximately 25% of patients. Most CAEs are nonspecific rash or pruritis, but urticaria can occur in 0.7% of patients.2Wongvibulsin S. Pahalyants V. Kalinich M. Murphy W. Yu K. Wang F. et al.Epidemiology and risk factors for the development of cutaneous toxicities in patients treated with immune-checkpoint inhibitors: a United States population-level analysis.J Am Acad Dermatol. 2022; 86: 563-572Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar A mechanism for exacerbating preexisting CIndU is not established. Extrapolating from other immune-mediated CAEs on pembrolizumab, such as bullous pemphigoid and vitiligo, PD-1 blockade may have caused increased autoantibody formation (IgE or IgG autoantibodies to FcεRI or IgE), thus decreasing the patient’s CIndU threshold. Chronic urticaria index or autologous skin testing, when positive, would have provided evidence of autoantibody formation to support this hypothesis, but was not possible for this patient. This case highlights the need for ongoing research on both the pathophysiology of CIndU and the expanding field of immune-mediated side effects of checkpoint inhibitors.