Highlights
- •Limited data exist on the development of tuberculosis in patients receiving immune checkpoint inhibitors.
- •We investigated the occurrence of TB after starting ICIs in a cancer-patient cohort.
- •The incidence rate of TB was much higher than that of the overall South Korean population in 2018.
- •Our data showed that TB can develop in cancer patients receiving ICIs.
Abstract
Background
Limited data exist on the development of tuberculosis (TB) in cancer patients receiving immune checkpoint inhibitors (ICIs).
Method
s: We evaluated the development of TB in 1144 solid-cancer patients who started ICIs (pembrolizumab, nivolumab, or atezolizumab) between July 2014 and December 2018.
Results
A total of 1144 cancer patients were treated with ICIs. The median age of the patients at the start of ICI treatment was 62 years (interquartile range [IQR]; 53–69 years). Lung cancer (n = 796, 69.6%) was the most common cancer followed by melanoma (n = 115, 10.1%), and lymphoma (n = 85, 7.4%). Pembrolizumab (n = 612, 53.5%) was the most common treatment, followed by nivolumab (n = 474, 41.4%) and atezolizumab (n = 58, 5.1%). The median treatment duration with ICIs was 42 days (IQR; 18–154 days), and the median follow-up duration after initiating ICIs was 187 days (IQR; 70–342 days). Overall, three patients developed TB, two of whom received nivolumab and one who received pembrolizumab.
Conclusions
Our data showed that TB can develop in cancer patients receiving ICIs. However, due to the small number of study population, it is insufficient to draw accurate conclusions about the role of ICIs in the development of TB. Moreover, it is unclear whether the incidence of TB would be comparable with the incidence of TB in elderly cancer patients. Further studies are needed to evaluate whether diagnosis and treatment of latent TB infections before starting ICIs could be helpful in preventing the development of TB in these patients.
Keywords
Abbreviations:
CI (confidence interval), DRESS (drug reaction with eosinophilia and systemic symptoms), ICI (immune checkpoint inhibitors), IQR (interquartile range), LTBI (latent tuberculosis infection), PD-1 (programmed death-1), TB (tuberculosis), TNF (tumor necrosis factor)1. Brief communication (word count: 1119/1000)
Immune checkpoint inhibitors (ICIs) have antitumor activity against several types of malignances [
[1]
]. ICIs reinforce the host immune system in attacking tumor cells, and they revitalize antitumor immune responses by disrupting co-inhibitory T-cell signaling, such as pembrolizumab and nivolumab's effects on programmed death-1 (PD-1) and atezolizumab's effects on programmed death-ligand 1 [[2]
]. Thus, some immune-related adverse events, such as thyroiditis, pneumonitis, and organizing pneumonia have been reported with ICIs [3
, 4
, 5
]. Interestingly, the occurrence of tuberculosis (TB) after ICI use has been steadily increasing [6
, 7
, 8
, 9
, 10
, 11
, 12
, 13
]; however, no previous reports have studied the development of TB associated with the use of ICIs in a large cohort. Therefore, we investigated the occurrence of TB after starting ICIs among a large cancer-patient cohort.We retrospectively screened 1248 cancer patients who started ICIs (pembrolizumab, nivolumab, or atezolizumab) for palliative chemotherapy between July 2014 and December 2018, using the immunotherapy registry of Samsung Medical Center (a 1979-bed referral hospital in Seoul, South Korea). Patients who received concurrent cytotoxic agents along with ICIs (n = 30) or other types of ICIs (n = 74) involved in clinical trials were excluded. The final study population included 1144 patients. Patients received ICIs intravenously at 2 mg/kg or 200 mg every 3 weeks for pembrolizumab, 3 mg/kg every 2 weeks for nivolumab, or 1200 mg every 3 weeks for atezolizumab until malignancy progression, intolerable toxicity, or the patient's refusal to continue. Our Institutional Review Board approved this study (2019-03-154-001) and waived the requirement for informed consent.
Clinical characteristics of study patients, use of ICIs, and confirmed diagnosis, treatment modality and outcomes of pulmonary TB were collected retrospectively. The incidence of TB was calculated as the number of new cases occurring in the population in a given period of time divided by the total number of people in the same population during the same period of time and expressed in terms of 100,000 person-years.
In total, 1144 patients who received ICIs were analyzed (Table 1). The median age of patients at the start of ICIs was 62 years (interquartile range [IQR] 53–69 years), and 69% were male. Lung cancer (n = 796, 69.6%) was the most common malignancy, followed by melanoma (n = 115, 10.1%) and lymphoma (n = 85, 7.4%). Pembrolizumab (n = 612, 53.5%) was the most common treatment, followed by nivolumab (n = 474, 41.4%) and atezolizumab (n = 58, 5.1%). The median treatment duration with ICIs was 42 days (IQR 18–154 days), and the median follow-up period after starting ICIs was 187 days (IQR 70–342 days). No patients received immunosuppressive medication including TNF inhibitors and mycophenolate. There were 7.4% (85/1144) of patients, who were diagnosed with lymphoma and received high dose steroids. However, none of these patients were diagnosed with TB during the study.
Table 1Characteristics of study patients receiving immune checkpoint inhibitors.
| Characteristics | Total (n = 1144) |
|---|---|
| Sex, male | 786 (69.0) |
| Age, years | 62 (53–69) |
| Type of cancer | |
| Lung cancer | 796 (69.6) |
| Melanoma | 115 (10.1) |
| Lymphoma | 85 (7.4) |
| Gastric cancer | 73 (6.4) |
| Head and neck cancer | 35 (3.1) |
| Thymic cancer | 32 (2.7) |
| Mesothelioma | 5 (0.4) |
| Sarcoma | 3 (0.3) |
| Overall duration of immunotherapy, days | 42 (18–154) |
| Median cycles of chemotherapy | |
| Pembrolizumab (n = 612, 53.5%) | 4 (2–9) |
| Nivolumab (n = 474, 41.4%) | 3 (2–8) |
| Atezolizumab (n = 58, 5.1%) | 3 (1–5) |
| Median follow up duration after starting ICIs, days | 187 (70–342) |
The data are presented as either n (%) or median (interquartile range). ICIs, immune checkpoint inhibitors.
During the study period, three patients with advanced lung cancer developed pulmonary TB at 22.0, 14.0, and 7.0 months after starting ICIs (Table 2); two had received nivolumab and one received pembrolizumab. The overall incidence rate of TB was 394.4 cases (95% confidence interval [CI] 100.3–1073.4) per 100,000 person-years. No patients developed nontuberculous mycobacterial infections.
Table 2Characteristics of patients diagnosed with pulmonary tuberculosis after treatment with ICIs.
| Characteristics | Case 1 | Case 2 | Case 3 |
|---|---|---|---|
| Age/Sex | 63/male | 79/male | 59/female |
| Type of cancer | adenocarcinoma, lung | squamous cell carcinoma, lung | adenocarcinoma, lung |
| Previous rounds of chemotherapy | 3 | 1 | 6 |
| Type of ICI | nivolumab | pembrolizumab | nivolumab |
| Time interval between initiation of ICI and diagnosis of TB, months | 22 | 14 | 7 |
| Time interval between last dose of ICI and diagnosis of TB, months | 1 | 6 | 6 |
| Duration of ICIs before diagnosis of TB, months | 21 (41 cycles) | 8 (14 cycles) | 1 (3 cycles) |
| Lymphocyte counts, before the start of ICIs and at the time of TB diagnosis,/mcgL | 1880 → 2089 | 1448 → 30 | 1778 → 412 |
| Evaluation of cancer response after initiating ICI | partial response | stable disease | progressive disease; changed to cytotoxic chemotherapy |
| Other immune-related adverse events | Yes (drug-induced thyroiditis at 20 months after initiating ICI) | Yes (drug-induced thyroiditis at 7 months, and biopsy-proven membranous glomerulonephritis at 11 months after initiating ICI) | No |
| Use of steroids before diagnosis of TB | No | Yes (oral prednisolone 30 mg once daily for a month at 11 months after initiating ICI due to membranous glomerulonephritis) | Yes (a cumulative dosage of 128 mg dexamethasone at 2 months after initiating ICI to prevent toxicity from cytotoxic chemotherapy) |
| Diagnosis of TB | positive culture for Mycobacterium tuberculosis in bronchoalveolar lavage fluid | detection of Mycobacterium tuberculosis complex using Xpert MTB/RIF assay of sputum | positive culture for Mycobacterium tuberculosis in sputum |
| Radiological findings at the time of diagnosis of lung cancer | a 51-mm-sized mass in the left upper lobe with ipsilateral mediastinal lymphadenopathy | a 31-mm-sized mass in the right middle lobe with segmental atelectasis | a 36-mm-sized mass in the right middle lobe without mediastinal lymphadenopathy |
| Radiological findings at the time of diagnosis of pulmonary TB | pneumonic consolidation in the right upper lobe without cavity; no change of primary tumor in left upper lobe | air-fluid level and wall thickening of bullae in left upper lobe; no change of primary tumor in the right middle lobe | nodular infiltration with hematolymphangitic metastasis and pleural seeding; increased size of metastatic lymphadenopathy in the mediastinum and left supraclavicular area |
| Treatment of pulmonary TB | INH, RIF, EMB, and PZA | INH, RIF, EMB, and PZA | INH, RIF, EMB, and PZA |
| Prognosis | DRESS syndrome developed after initiating anti-TB medication; patient is still undergoing nivolumab treatment | died 14 months after initiating ICI due to peritonitis | died 11 months after initiating ICI due to progression of cancer |
ICIs, immune checkpoint inhibitors; TB, tuberculosis; DRESS, drug reaction with eosinophilia and systemic symptoms.
a Number of previous rounds of cytotoxic chemotherapy prior to use of ICI.
b Based on RECIST criteria (Response evaluation criteria in solid tumors).
c INH, isoniazid; RIF, rifampin; EMB, ethambutol; PZA, pyrazinamide.
To our knowledge, this is the first report on TB development associated with ICI use in a large cancer-patient cohort. The incidence rate of TB was 394.4 cases per 100,000 person-years, much higher than that seen in the overall South Korean population in 2018 (51.5 cases per 100,000 person-years) [
[14]
,- KCDC
2018 annual report on the notified tuberculosis in Korea, KCDC.
https://www.cdc.go.kr/board.es?mid=a20602010000&bid=0034&act=view&list_no=143392
Date: 2018
Date accessed: October 12, 2019
[15]
] but similar to two previous studies that included cancer patients (361.3 per 100,000 person-years, excluding lung cancer and 307 per 100,000 person-year, including lung cancer) [[16]
,[17]
]. However, the observation period of our study patients was much shorter than in the previous studies (median follow-up period of 6.1 vs. 19.2 vs. 29.0 months), likely because our patients had more advanced disease. Therefore, our data may underestimate the overall incidence rate.Immunomodulatory drugs, including immunosuppressants, steroids, and anti-tumor necrosis factor (TNF), are well-known risk factors for developing TB. Recent studies from South Korea have reported that the incidence of TB among patients receiving anti-TNF alpha therapy was >1000 cases per 100,000 person-years [
[18]
,[19]
]. However, adequate therapy for latent TB infections can significantly reduce the risk of TB occurrence in such patients [[19]
], which highlights the importance of screening for latent TB infections in cancer patients receiving immunotherapy. Additionally, due to side effects that can occur when treating latent TB infections and cancer simultaneously, active screening for and treatment of latent TB infections before immunotherapy is a treatment strategy worth considering.Previously, ICIs were suggested to upregulate the immune response not only to cancer but also to microorganisms such as HIV and chronic hepatitis B infections [
[20]
]. Conversely, some studies have shown the activation of TB after initiation of ICIs [[8]
,[21]
]. In PD-1-knockout mice, Mycobacterium tuberculosis-specific CD4 T-cell responses are increased, and high levels of inflammatory cytokines are expressed in the lungs [[21]
]. Similarly, PD-1 staining is less strong in TB granulomas after the use of ICIs, and activated T cells recruit permissive monocytes or neutrophils to the TB granuloma, which may lead to cavitation of lung tissue [[8]
]. The mechanism for these paradoxical reactions is unclear, but it appears to develop due to a disruption of immune system homeostasis. Further studies are needed to explore the pathophysiology behind the development of TB after immune checkpoint inhibition.In this study, two of the patients later diagnosed with TB had previously received high-dose steroid therapy either due to severe immune-related adverse events related to ICIs (three months before developing TB) or to prevent toxicity due to cytotoxic chemotherapy (two months before developing TB). Therefore, we could not rule out whether the development of TB was caused by the high-dose steroid medications. Nevertheless, attention should be paid to the development of TB after ICIs, since serious adverse events after the administration of ICIs can require steroid therapy, which might reactivate TB.
In conclusion, our data showed that TB can develop in cancer patients receiving ICIs. However, the retrospective nature of this study conducted in a single referral hospital limits the generalizability of these findings. In particular, due to the small number of study population, it is insufficient to draw accurate conclusions about the role of ICIs in the development of TB. More data on the incidence rate of TB in different settings are required. Moreover, it is unclear whether the incidence of pulmonary TB would be comparable with the incidence of pulmonary TB in elderly cancer patients, therefore national study with appropriate sample size is needed to establish whether ICIs might be responsible to development of TB on patients with cancer. Further studies are also needed to evaluate whether diagnosis and treatment of latent TB infections before starting ICIs could be helpful in preventing the development of TB in these patients.
Funding
This research was supported by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI) funded by the Ministry of Health & Welfare, Republic of Korea (HI19C0137). The sponsor had no role in the design of the study, the collection and analysis of the data, or the preparation of the manuscript.
Authors' contributions
Conception and design: Y. Im, W-J. Koh, B.W. Jhun; Acquisition of data: J. Lee, S.J. Kim, S–H. Lee; Analysis and interpretation of data: Y. Im, W-J. Koh, B.W. Jhun, S–H. Lee; Writing, review and critical revision of the manuscript: All authors.
Declaration of competing interest
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Dr. Se-Hoon Lee received grants and personal fees from Merck Sharp & Dohme Corp (MSD), and personal fees from Novartis, AstraZeneca, Bristol-Myers Squibb (BMS), and Roche, not associated with the submitted work. Dr. Won-Jung Koh has served on Advisory Boards for Insmed Inc. and Johnson and Johnson, not associated with the submitted work. The authors have no additional conflicts of interest to declare.
References
- Immune checkpoint inhibitors: recent progress and potential biomarkers.Exp. Mol. Med. 2018; 50: 165
- The blockade of immune checkpoints in cancer immunotherapy.Nat. Rev. Cancer. 2012; 12: 252
- A systematic review of immune-related adverse event reporting in clinical trials of immune checkpoint inhibitors.Ann. Oncol. 2015; 26: 1824-1829
- Immune-related adverse events associated with immune checkpoint blockade.N. Engl. J. Med. 2018; 378: 158-168
- Immune-checkpoint inhibitors associated with interstitial lung disease in cancer patients.Eur. Respir. J. 2017; 501700050
- Tuberculosis following PD-1 blockade for cancer immunotherapy.Sci. Transl. Med. 2019; 11eaat2702
- Infectious complications associated with the use of immune checkpoint inhibitors in oncology: reactivation of tuberculosis after anti PD-1 treatment.Clin. Microbiol. Infect. 2018; 24: 216-218
- Implications of tuberculosis reactivation after immune checkpoint inhibition.Am. J. Respir. Crit. Care Med. 2018; 198: 1451-1453
- Anti-PD1 antibody treatment and the development of acute pulmonary tuberculosis.J. Thorac. Oncol. 2016; 11: 2238-2240
- Tuberculosis reactivation in a patient receiving anti-programmed death-1 (PD-1) inhibitor for relapsed Hodgkin's lymphoma.Acta Oncol. 2016; 55: 519-520
- Pericardial tamponade caused by a hypersensitivity response to tuberculosis reactivation after anti-PD-1 treatment in a patient with advanced pulmonary adenocarcinoma.J. Thorac. Oncol. 2017; 12: e111-e114
- Paradoxical response in a patient with non-small cell lung cancer who received nivolumab followed by anti-Mycobacterium tuberculosis agents.J. Infect. Chemother. 2019; 25: 54-58
- Severe pulmonary infections complicating nivolumab treatment for lung cancer: a report of two cases.Acta Clin. Belg. 2019; : 1-3
- 2018 annual report on the notified tuberculosis in Korea, KCDC.https://www.cdc.go.kr/board.es?mid=a20602010000&bid=0034&act=view&list_no=143392Date: 2018Date accessed: October 12, 2019
- Tuberculosis control in the Republic of Korea.Epidemiol Health. 2018; 40 (e2018036-2018030)
- Cancer-specific incidence rates of tuberculosis: a 5-year nationwide population-based study in a country with an intermediate tuberculosis burden.Medicine. 2016; 95e4919
- Solid-organ malignancy as a risk factor for tuberculosis.Respirology. 2008; 13: 413-419
- Risk of tuberculosis in patients treated with anti-tumor necrosis factor therapy: a nationwide study in South Korea, a country with an intermediate tuberculosis burden.Int. J. Rheum. Dis. 2015; 18: 323-330
- Efficacy of treatment for latent tuberculosis in patients undergoing treatment with a tumor necrosis factor antagonist.Ann. Am. Thorac. Soc. 2017; 14: 690-697
- Immune checkpoint blockade in infectious diseases.Nat. Rev. Immunol. 2017; 18: 91
- CD4 T cells promote rather than control tuberculosis in the absence of PD-1–mediated inhibition.J. Immunol. 2011; 186: 1598-1607
Article info
Publication history
Published online: December 14, 2019
Accepted:
December 12,
2019
Received in revised form:
December 2,
2019
Received:
July 3,
2019
Identification
Copyright
© 2019 Elsevier Ltd.
User license
Elsevier user license | How you can reuse 
Elsevier's open access license policy
Elsevier user license
Permitted
For non-commercial purposes:
- Read, print & download
- Text & data mine
- Translate the article
Not Permitted
- Reuse portions or extracts from the article in other works
- Redistribute or republish the final article
- Sell or re-use for commercial purposes
Elsevier's open access license policy
