Chronic Pleuritis and Recurrent Pleural Effusion After Atezolizumab for Small Cell Lung Cancer

Background

As immune checkpoint inhibitors are being increasingly used and investigated against various malignancies, we are learning more about their toxicity profiles, particularly immune-related adverse events. Pneumonitis causes lung parenchymal changes and is the most recognized pulmonary immune-related adverse event. Pulmonary complications from PD-1 inhibitors have been extensively described but adverse effects from PD-L1 inhibitors such as atezolizumab have not. Specifically, little is known about how PD-L1 inhibitors might affect the pleura and the pleural space. Here, we describe the first report ed case of biopsy-proven pleuritis manifesting as recurrent pleural effusion in a patient treated with atezolizumab.

Case Report

A 66-year-old woman with a history of extensive-stage small cell lung cancer presented to our clinic for evaluation of a new left-sided pleural effusion. She reported a 2-month history of progressive shortness of breath during which she was no longer able to exercise or perform household chores. She endorsed a dry cough when laying on her left side. She denied fevers, chills, night sweats, or weight loss. She had a prior 50 pack-year smoking history and quit 2 years earlier.

She had previously received systemic therapy with carboplatin, etoposide, and atezolizumab. After completing 4 cycles of this regimen, she had been on atezolizumab maintenance therapy but developed pneumonitis after 2 additional months of treatment; immunotherapy was therefore subsequently held. By the time of presentation to our clinic, she had been off of chemo-therapy for 8 months and off of immunotherapy for 6 months. Radiation to the primary left upper lobe tumor and a cerebellar metastasis was completed 1 year prior to presentation.

Due to concern for recurrence of malignancy, she underwent a thoracentesis, with 750 mL of serous fluid removed. Pleural fluid analysis revealed a lymphocyte-predominant exudative effusion with eosinophilia (13%). Pleural fluid cytology revealed no malignant cells. Since the concern for malignant recurrence remained high, she underwent left-sided medical thoracoscopy. Inspection of the parietal pleura did not reveal any abnormalities, including no evidence of radiation changes (Figure 1). Random parietal pleural biopsies were taken, which revealed chronic pleuritis with reactive mesothelial hyperplasia and no tumor cells (Figure 2).

Given her completely normal-appearing pleura, radiation pleuritis was ruled out. It was felt that chemotherapy too long ago to be a present cause of her pleuritis. There was very low suspicion for pulmonary embolism given the normal right ventricle and tricuspid valve function on echocardiogram and resolution of symptoms with thoracentesis. As such, after extensive workup to rule out malignant recurrence and given the known ability of immunotherapy to have ongoing effects even after discontinuation, the eosinophilic pleural effusion was felt to be due to pleuritis from atezolizumab. The effusion eventually recurred and was drained again approximately 3 months later; in total, her effusion has been drained 5 times over 12 months, and cytology has remained negative for malignancy.

Discussion

Pulmonary toxicity related to immune checkpoint inhibitors has an overall incidence of 2–3% during cancer treatment [1,2]. Pneumonitis causes lung parenchymal changes and is the most recognized pulmonary immune-related adverse event. This is more commonly encountered with PD-1 or PD-L1 inhibitors than with cytotoxic T-lymphocyte antigen (CTLA)-4 inhibitors [3].

Less is known regarding how ICIs affect the pleura and pleural space. In the current literature, pleural effusions associated with ICIs are rarely encountered and have been reported with nivolumab use in case reports [3-6]. Socinski et al listed 1 case of pleurisy in a group of patients receiving combined atezolizumab, bevacizumab, paclitaxel, and carboplatin, but whether this referred simply to pain on inspiration and how this was diagnosed was not described [7]. To the best of our knowledge, this is the first case report of biopsy-proven pleuritis after atezolizumab.

Atezolizumab is a PD-L1 antibody that has been approved for use in metastatic non-small cell lung cancer, small cell lung cancer, and urothelial cancers [7]. In a meta-analysis comparing the safety profiles of ICIs, atezolizumab ranked among the safest in terms of incidence of overall adverse events compared to nivolumab, pembrolizumab, ipilimumab, and tremelimumab [8]. Pneumonitis associated with atezolizumab occurred in only 1% of cases [9]. Atezolizumab can cause a sarcoid-like reaction, but even large studies have not described pleural effusion as an adverse event [7,10,11].

Pleural effusion associated with immune checkpoint inhibitors is a diagnosis of exclusion. When patients with active or recent cancer develop a pleural effusion, initial management involves evaluation for malignant pleural involvement. In effusions with negative cytology after thoracentesis and high suspicion of malignancy, thoracoscopy with pleural biopsies should be performed next [12]. Nevertheless, in a study of thoracoscopy in patients with active malignancy and pleural effusion, the majority of patients did not have malignant pleural involvement but were deemed to have nonspecific pleuritis [13]. The effusion was then further subclassified based on clinical context and determined to be due to radiation, chemotherapy, or other etiologies. Radiation-induced pleuritis was accompanied by discolored, thickened pleura and often with a line of demarcation between normal and abnormal pleura that coincided with the radiation field. Chemotherapy-induced pleuritis usually occurred with agents known to cause effusion and relatively soon after therapy – at a mean time of 38 days after last treatment. In our case, given completely normal-appearing pleura, radiation-induced pleuritis was felt to be ruled out. Additionally, given the extended period of time since chemotherapy cessation (8 months), chemotherapy-induced pleuritis was also unlikely. As such, the remaining explanation for the effusion was atezolizumab. Of note, the patient had eosinophilia on her pleural fluid studies. In one study of drug-related effusions, 25% had eosinophilic effusions, supporting our belief that her effusion could be attributable to previous medical therapy [13]. Pulmonary embolism is a known cause of eosinophilic pleural effusions but suspicion was low in this case with a normal echocardiogram (normal right ventricular size and function with trace tricuspid regurgitation) and resolution of symptoms with each thoracentesis. Moreover, the repeated recurrence may be consistent with a prolonged immunotherapy effect. Chronic immune checkpoint inhibitor pneumonitis is a recently described entity (ICI pneumonitis that persists despite ICI cessation) [14,15]. In our case, we may additionally be observing the first reported case of chronic immune checkpoint inhibitor pleuritis.

Conclusions

Immune checkpoint inhibitors will be used more frequently as clinicians incorporate these therapies into cancer treatment regimens. With increased use, adverse events will be more frequently encountered and need to be better understood. We present the first case of chronic pleuritis with recurrent pleural effusion after atezolizumab to increase awareness for clinicians to recognize this as a potential etiology of pleural effusion. We propose that this is the first reported case of chronic immune checkpoint inhibitor pleuritis.