The impact of pembrolizumab on patients with pre-existing autoimmune diseases.

Abstract

2660 Background: Immune checkpoint inhibitors (ICI) have changed the therapeutic landscape across a range of solid and hematologic malignancies. With the mechanistic activation of the adaptive immune response and known unique development of immune-related adverse effects (irAE), there is a theoretical risk of causing detrimental effects on patients with pre-existing autoimmune diseases (AID). However, this patient population was excluded from many clinical trials, and there is limited and conflicting retrospective data speculating upon the impacts of using ICI therapy in patients with AID. This study aims to report the impact of pembrolizumab in patients with AID. Methods: Patients who received pembrolizumab for treatment of any type of cancer between 1/1/2017-8/1/2021 were retrospectively reviewed, and patients with AID (rheumatologic or non-rheumatologic) were identified. Data including age, gender, race, ECOG, and primary cancer diagnosis were collected. AID was characterized by type, activity (symptomatic vs. asymptomatic at ICI initiation), and immunosuppressive treatment (IST) use. Outcomes included flare of AID, irAE, and ICI discontinuation. Results: Out of 810 patients, 12 with pre-existing AID were found and being treated for non-small cell lung cancer (NSCLC) (58% N=7), renal cell carcinoma (RCC) (16.6% N=2), colorectal cancer (CRC) (8% N=1), head and neck cancer (8% N=1), and cervical cancer (8% N=1). Around 67% of these patients were females with a median age of 65 years (50-77). Median ECOG was 1 (0-2). 9 patients (75%) had rheumatologic disease (5 rheumatoid arthritis (RA), 1 systemic lupus erythematosus (SLE), 2 psoriatic arthritis, and 1 systemic sclerosis (SS)). 3 patients (25%) had non-rheumatologic AID (1 multiple sclerosis (MS), 1 autoimmune hepatitis, and 1 Evan’s syndrome). All patients were asymptomatic prior to initiation of therapy. At the time of ICI initiation, 11 patients (92%) were on therapy, including steroids, methotrexate, and other IST. Due to concern for severe flare, pembrolizumab was discontinued in 3 (25%) MS, RA, and SS patients treated for RCC, CRC, and NSCLC, respectively. 3 patients (25%) developed irAE unrelated to their AID (pneumonitis, dermatitis, arthritis, gastritis) and received steroids. 4 patients (33.3%) had documented flares while on pembrolizumab requiring escalation of their ongoing therapy, and 1 NSCLC patient had a severe exacerbation of SS requiring cyclophosphamide and rituximab infusions. Conclusions: This analysis adds to the limited available literature that patients with an AID and on IST appear to have minimal AID-associated flares requiring intervention and tolerate pembrolizumab with a similar rate of irAE compared to patients without AID. Additional larger studies are needed to assess the efficacy of treatment in this population.