Pericardial Tamponade Caused by a Hypersensitivity Response to Tuberculosis Reactivation after Anti-PD-1 Treatment in a Patient with Advanced Pulmonary Adenocarcinoma.

We read with interest the editorial comments1Reungwetwattana T. Adjei A.A. Anti-PD-1 antibody treatment and the development of acute pulmonary tuberculosis.J Thorac Oncol. 2016; 11: 2048-2050Abstract Full Text Full Text PDF PubMed Scopus (37) Google Scholar on a case report by Fujita K. et al.2Fujita K. Terashima T. Mio T. Anti-PD1 antibody treatment and the development of acute pulmonary tuberculosis.J Thorac Oncol. 2016; 11: 2238-2240Abstract Full Text Full Text PDF PubMed Scopus (101) Google Scholar describing a patient with NSCLC in whom acute pulmonary tuberculosis (TB) developed while he was receiving immune checkpoint inhibitor therapy. The expression of programmed death protein 1 (PD-1) and its ligand (PD-L1) can be dysregulated by tumors to serve as a mechanism against host immune surveillance.3Pardoll D.M. The blockade of immune checkpoints in cancer immunotherapy.Nat Rev Cancer. 2012; 12: 252-264Crossref PubMed Scopus (8958) Google Scholar On the other hand, increased expression of interferon-γ and upregulation of the PD-1/PD-L1 pathway are also observed in patients with TB infection.4Jurado J.O. Alvarez I.B. Pasquinelli V. et al.Programmed death (PD)-1:PD-ligand 1/PD-ligand 2 pathway inhibits T cell effector functions during human tuberculosis.J Immunol. 2008; 181: 116-125Crossref PubMed Scopus (209) Google Scholar The role of PD-1/PD-L1 in the complex interplay between cancer and TB infection when both diseases coexist is beyond our current knowledge. Two hypothesis have been proposed by the editors for the anti–PD-1 treatment–related acute TB infection in cancers.1Reungwetwattana T. Adjei A.A. Anti-PD-1 antibody treatment and the development of acute pulmonary tuberculosis.J Thorac Oncol. 2016; 11: 2048-2050Abstract Full Text Full Text PDF PubMed Scopus (37) Google Scholar The first one is a hypersensitive response similar to immune reconstitution inflammatory syndrome. The other hypothesis attributes TB infection to immune checkpoint–related lymphopenia. We would like to share the case of our patient with lung cancer in whom tuberculous pericarditis developed through a hypersensitivity response after 3 months of nivolumab treatment, in which the presentation supports the first hypothesis. Stage IV adenocarcinoma of the lung with EGFR exon 19 deletion was diagnosed in a 59-year-old male former smoker in September 2011. The patient was heavily treated in the following 4 years, during which a partial response to gefitinib as first-line targeted therapy occurred along with stable or progressive disease in response to several lines of chemotherapy. Rebiopsy of metastatic lung tumors did not identify the T790M mutation. The patient started receiving nivolumab, 3 mg/kg every 2 weeks, in March 2016, with a partial response after two cycles of treatment. He was found to have rapid accumulation of pericardial effusion and increased pericardium thickness (Fig. 1) accompanied by signs of pericardial tamponade after three cycles of nivolumab. Throughout the course of nivolumab, there was no significant lymphopenia. Pathologic examination of the pericardium biopsy specimen showed granulomatous inflammation and acid-fast bacilli. The culture of pericardial effusion yielded Mycobacterium tuberculosis. The patient has been receiving anti-TB therapy since 2016 June and has skipped one cycle of nivolumab. Prednisolone, 1 mg/kg, was also given for 1 month. The pericardial and pleural effusion regressed after 1 month of treatment. There was no relapse of effusion after resumption of nivolumab treatment. On review of the computed tomography scans, we found that TB reactivation over the pericardium should have been present at the beginning of nivolumab treatment, but the finding of pericardial thickening was interpreted as cancer invasion at that time (see Fig. 1A). We retrospectively assessed PD-L1 expression in his archived tumor samples, pericardium biopsy specimen, and cell block of pericardial effusion. There was no PD-L1 expression on cancer cells in the initial specimen, but more than 1% of the tumor cells expressed PD-L1 when the patient was heavily treated (Fig. 2). On the other hand, the immune cells (CD68 positive) in the tuberculous pericardium showed strong expression of PD-L1 before anti-TB treatment, but after 2 weeks of anti-TB medication, the monocytes in the pericardial effusion became PD-L1–negative (Fig. 3). He completed 6 months of anti-TB treatment successfully. The duration of response to nivolumab was 6 months. After nivolumab failure and successful TB treatment, a rebiopsy specimen of the new metastatic lung tumor revealed that more than 50% of the cancer cells were strongly positive for PD-L1 (see Fig. 2F).Figure 2Cancer specimen cohort of programmed death ligand 1 (PD-L1) expression. The initial specimen (A) from a mediastinal lymph node biopsy (hematoxylin and eosin) exhibited metastatic adenocarcinoma that was negative for PD-L1 (B) by immunohistochemistry. A rebiopsy specimen from a metastatic lung tumor (C) after the patient being heavily treated showed acquired expression of PD-L1 in more than 1% of tumor cells (arrows) (D). The latest tumor biopsy after nivolumab failure but control of tuberculosis pericarditis confirmed disease progression (E) and showed even stronger positivity for PD-L1 by immunohistochemistry (F).View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 3Tuberculosis specimen cohort of programmed death ligand 1 (PD-L1) expression. (A–C) Pericardial biopsy specimens after nivolumab treatment showed granulomatous inflammation (hematoxylin and eosin) (A) and strong positivity for PD-L1 (B). There were no malignant cells in the pericardium. These PD-L1–positive cells were also positive for CD68 (C), which is compatible with monocyte infiltration. (D–F) Cell block from pericardial effusion (D) after 2 weeks of antituberculosis treatment showed inflammatory cells that were PD-L1–negative (E) and CD68 positive (F).View Large Image Figure ViewerDownload Hi-res image Download (PPT) Pericardial effusion in patients with NSCLC during immunotherapy might result from disease progression or immune-related adverse effects. TB pericarditis had not been on the list of complications during previous reports of clinical trials for anti–PD-1 or anti–PD-L1 treatment. As we know, TB reactivation might occur simultaneously with malignancy or after chemotherapy, but little information was available in its relationship to immunotherapy. We were especially concerned that anti–PD-1 treatment would induce a serious TB complication. The pathophysiology of tuberculous pleurisy or pericarditis is usually attributed to a cell-mediated hypersensitivity response to tuberculin when tubercle bacilli gain access to the space.5Ellner J.J. Barnes P.F. Wallis R.S. Modlin R.L. The immunology of tuberculous pleurisy.Semin Respir Infect. 1988; 3: 335-342PubMed Google Scholar The tuberculous pericarditis that developed soon after anti-PD-1 treatment in our case implied restoration of host immunity and induced a hypersensitivity response. Another interesting issue is the role of TB reactivation in tumor progression and cancer response to immune checkpoint inhibitors. T-cell exhaustion has been demonstrated in humans with chronic infections or malignancies through the PD-1 pathway.6Wherry E.J. T cell exhaustion.Nat Immunol. 2011; 12: 492-499Crossref PubMed Scopus (2452) Google Scholar When TB reactivation develops in patients with lung cancer, upregulation of the PD-1 pathway in both diseases might co-promote each other to evade the host immunity. However, the dynamic expression of PD-L1 in the specimen cohort overrode this inference. That is, PD-L1 expression in lung cancer cells became more evident despite of control of TB and continuation of nivolumab. On the other hand, PD-L1 expression in the TB microenvironment was rapidly down-regulated once the infection was under control despite the presence of lung cancer. The expression of PD-L1 seemed to proceed by each disease course without significant interaction and to not be influenced by steroid or immune checkpoint inhibitor treatment. Our patient proved the theory of a hypersensitivity response to TB reactivation with pericardial tamponade after immune checkpoint inhibition. There is no interaction of the expression of PD-L1 between lung cancer and TB when the two diseases coexist. We suggest that tuberculous serositis be included in the differential diagnosis when new effusion develops in patients with lung cancer who have been receiving immune checkpoint inhibitors.