Introduction
Psoriasis is a chronic multi-systemic inflammatory disease [1]. Anti-tumor necrosis factor (anti-TNF) agents such as adalimumab, infliximab, etanercept are effectively used for treatment and they are approved for moderate to severe psoriasis by the Food and Drug Administration [2]. These agents reduce inflammation in the body. However the increased risk of infection, especially mycobacterium tuberculosis infection is a major concern for patients receiving TNF inhibitor therapy [3]. Tumor necrosis factor (TNF) is a potent inflammatory cytokine that plays an important role in immunity to numerous bacterial infections, including Mycobacterium tuberculosis. TNF can support anti-tuberculosis immunity through the secretion of chemokines, up-regulation of adhesion molecules and the induction of macrophage apoptosis [4]. TNF is essential for granuloma formation and maintenance and increased clinical use of TNF antagonists has been associated with the increased rate of granulomatous infectious diseases, such as tuberculosis, histoplasmosis and several less common conditions [5, 6]. Blocking TNF-mediated immune responses has been linked to development of tuberculosis infection [4–6]. Psoriasis patients treated with TNF inhibitor therapy are at risk of reactivation of latent tuberculosis infection [7]. The majority of people infected with M. tuberculosis develop clinically latent infection in which mycobacteria are not transmitted. A 10% lifetime risk of reactivation exists during latent infection. It has been suggested that TNF-mediated formation and maintenance of the granuloma is fundamental in controlling M. tuberculosis infection. In 2004, voluntary data collected worldwide showed that the risk of reactivation of tuberculosis was statistically lower in patients who received etanercept (28/100,000 patients) than in those who received infliximab (54/100,000 patients) [8]. It has been proposed that the mechanisms of action between these anti-TNF molecules differ.
Aim
In this study, we aimed to determine the risk and frequency of tuberculosis in patients receiving TNF inhibitor therapy and whether there is a difference in tuberculosis risk between anti-TNF drug groups.
Material and methods
This study is based on data from the Turkish Psoriasis Registry (PSR-TR) and digital records of the Medical School of the Bezmialem Vakif University. Approval for the study was obtained from the local Ethics Committee of the Bezmialem Vakif University (Ethics committee approval no. E-54022451-050.05.04-116589). Clinical Trials Protocol Registration and Results System ID is NCT06563635. All demographic characteristics of 719 patients who received TNF inhibitors between January 2007 and January 2023 and periodic control tests including tuberculin test or QuantiFERON test, chest radiographies and routine biochemistry, haematological investigations were recorded. Duration of treatment, previous treatments, isoniazid (INH) prophylaxis and side effects were evaluated. Patients with suspected tuberculosis were consulted with chest disease and infectious disease departments and examined in detail. Characteristic features and details of treatment they received are shown in Table 1.
Table 1
Characteristics of patients receiving TNF inhibitor therapy
Statistical analysis
Categorical variables are presented as absolute numbers and percentages, continuous variables as means, standard deviations, minimum and maximum range. Only means are presented as their values essentially overlapped with those of medians. All data were analysed by the statistical package SPSS for Windows, version 26.0 (SPSS Inc., Chicago, IL, USA). A c2 test for trend was calculated to detect significant differences among categorical variables and t-test for continuous ones. P ≤ 0.05 was accepted to be significant.
Results
Patient characteristics
A total of 719 psoriasis patients treated with TNF inhibitor therapy between January 2007 and January 2023 were evaluated in this study.
315 psoriasis patients, including 183 (58%) males and 132 (42%) females were treated with etanercept. Mean age of 315 patients was 50.9 years. Mean height of 315 patients was 166 cm, mean weight was 78.4 kg and mean body mass index (BMI) score was 28.5 kg/m2.
176 psoriasis patients, including 95 (54%) males and 81 (46%) females were treated with infliximab. Mean age of 176 patients was 52 years. Mean height of 176 patients was 167 cm, mean weight was 86.5 kg and mean BMI was 31 kg/m2.
228 psoriasis patients, including 107 (46.9%) males and 121 (53.1%) females were treated with adalimumab. Mean age of 228 patients was 49.4 years. Mean height of 228 patients was 166 cm, mean weight was 80.3 kg and mean BMI was 29.1 kg/m2 (Table 1).
Chemoprophylaxis requirement
Patients were evaluated for latent tuberculosis infection risk before initiation of anti-TNF therapy with physical examination, anamnesis, periodic control tests including tuberculin test or QuantiFERON test, chest radiographies and routine biochemistry. 9-month isoniazid prophylaxis was used in patients with LTBI risk according to recommendations of the “Turkish Guidelines for the Management of Psoriasis with Biologic Agents” and “Turkey Psoriasis Treatment Guide-2016” [9, 10]. 358 patients of 719 (49.8%) patients required isoniazid prophylaxis. 152 of 315 patients receiving etanercept (48.2%), 108 of 176 patients receiving infliximab (61.4%) and 98 of 228 patients receiving adalimumab (43%) required isoniazid prophylaxis (Table 1).
Disease severity and anti-TNF exposure
315 psoriasis patients with a mean psoriasis area and severity index (PASI) score of 7.5 at the beginning of the therapy were treated with etanercept. Mean duration of the etanercept therapy was 28.6 months. 176 psoriasis patients with a mean PASI score of 10.5 at the beginning of the therapy were treated with infliximab. Mean duration of the infliximab therapy was 23.5 months. 228 psoriasis patients with a mean PASI score of 6.2 at the beginning of the therapy were treated with adalimumab. Mean duration of the adalimumab therapy was 26 months (Table 1).
Patients diagnosed with tuberculosis infection
A total of 719 psoriasis patients treated with TNF inhibitor therapy were evaluated in this study. Four psoriasis patients were diagnosed with tuberculosis. Three of them were diagnosed with pulmonary tuberculosis, one of them was diagnosed with tuberculosis peritonitis between June 2007 and January 2023. All patients with tuberculosis (1.27%, n = 4) had been treated with etanercept (Table 2). None of them had tuberculosis infection during the infliximab and adalimumab therapy.
Table 2
Characteristics of patients diagnosed with tuberculosis infection
A 54-year-old female patient was diagnosed with tuberculosis peritonitis at month 95 of the etanercept therapy. She had 9-month isoniazid prophylaxis at the beginning of the etanercept therapy. She had a positive QuantiFERON test result at month 89 of the therapy, she was diagnosed with tuberculosis peritonitis at month 95 of the therapy. A 59-year-old male patient was diagnosed with pulmonary tuberculosis at month 24 of the etanercept therapy. He had a positive QuantiFERON test for tuberculosis prior to the therapy, and therefore he had 9-month isoniazid prophylaxis for tuberculosis at the beginning of the etanercept therapy. A 40-year-old female patient was diagnosed with pulmonary tuberculosis at month 39 of the etanercept therapy. Her purified protein derivative (PPD) skin test for tuberculosis was 19 mm prior to the etanercept therapy, and therefore she had 1-month isoniazid prophylaxis before the etanercept therapy. A 54-year-old male patient was diagnosed with pulmonary tuberculosis. He had 12-month etanercept therapy, 12-month secukinumab therapy and 22-month certolizumab therapy. After certolizumab therapy, he did not have any treatment for 12 months, then he was diagnosed with pulmonary tuberculosis 58 months after starting the biological treatment. He had a positive QuantiFERON test and received 9-month isoniazid prophylaxis for tuberculosis at the beginning of biological treatment. The median time between the beginning of anti-TNF therapy and tuberculosis occurrence was 54 months (Table 2).
Discussion
Anti-tumor necrosis factor-a (TNF-a) immunotherapy has revolutionized the treatment of inflammatory diseases, including psoriasis and psoriatic arthritis [11]. M. tuberculosis infection in humans is thought to present either as active TB or as LTBI (latent Mycobacterium tuberculosis infection) defined by the presence of immunological responses to mycobacterial antigens in absence of clinical symptoms of the disease [12]. LTBI subjects are thought to present a life-long risk of reactivation of the infection, and 5–15% of them develop active TB during their lifetime. In patients receiving anti-TNF-a treatment, the relative risk of TB infection varies from 1.5 to 17 [13]. A major challenge in tuberculosis control is the diagnosis and treatment of latent tuberculosis infection. The risk of TB infection is higher with anti-TNF-a monoclonal antibody therapy than with soluble TNF-a receptor therapy [3]. In 2008, Wallis reported 248 cases of infliximab-associated and 39 cases of etanercept-associated TB infections, which were recorded from January 1998 to March 2003 in the Adverse Events Reporting System (AERS) database of the Food and Drug Administration [14]. Three studies demonstrated that infliximab and adalimumab have an increased incidence of TB compared to etanercept [8, 13, 15]. Structural and functional differences could explain the higher TB incidence with monoclonal versus soluble inhibitors. Even though monoclonal antibodies demonstrated an increased incidence of TB, the risk of TB reactivation with etanercept therapy is not negligible.
Previous reviews of active tuberculosis following initiation of biologic therapy mainly included patients with rheumatic diseases, who had frequently used concomitant immunosuppressive treatment. Active TB occurs in approximately 0.1–0.2% of psoriatic patients receiving biologic therapy [16]. Studies suggest that prophylaxis therapy can prevent TB reactivation in individuals who receive biologics [17].
Most cases of TB develop soon after treatment initiation and correspond to reactivation of a latent TB infection. Tubach et al. and Carmona et al. suggested that anti-TB chemoprophylactic treatment is efficient for preventing TB [15, 18]. The data regarding tuberculosis during anti-TNF treatment mostly originate from rheumatology- and gastroenterology-based studies. However, there are several differences compared to psoriasis patients both in terms of the genetic background, comorbidities and concomitant use of disease-modifying medications. Indeed, the risk of serious infections was found to be significantly higher in patients with rheumatoid arthritis and Crohn’s disease compared to those with psoriasis (relative risk: 4.6, 5.1, and 1.3, respectively) [19]. According to the World Health Organization (WHO) Global Tuberculosis Report 2022, 1.6 million people died from TB, and 10.6 million people had TB in 2021, with 98% of these cases originating from low- and middle-income countries [20]. According to national TB surveillance data, the TB incidence (per 100,000 persons) was reported to have declined from 29.4 in 2005 to 14.1 in Turkey in 2018 [21]. Although TB mostly affects the lungs, extrapulmonary TB is not uncommon due to anti-TNF treatment. Elbek et al. evaluated the risk of tuberculosis in 240 patients receiving anti-TNF-a from December 2005 to December 2007 in Turkey [22]. They reported 2 patients developing tuberculosis (incidence 833/100,000). Ergun et al. investigated TB in 370 patients with psoriasis receiving anti-TNF treatment (192 etanercept, 163 infliximab, and 113 adalimumab, mean treatment duration was 12.7 ±11.8 months). They reported 4 (1.08%) patients who developed tuberculosis, three pulmonary and one gastrointestinal [23].
Infliximab, adalimumab, and certolizumab pegol are anti-TNF monoclonal antibodies [24]. Etanercept functions as a soluble TNF receptor fusion protein that blocks TNF-a and is effective in inflammatory diseases [25]. This different nature of anti-TNF drugs may cause the different risk of tuberculosis infection.
Saliu et al. suggested that infliximab and adalimumab significantly inhibited T cell activation and reduced the proportion of tuberculosis-responsive CD4 cells by 70% and 50%, respectively, while etanercept produced no significant effect. Saliu et al. also claimed that the antiTNF antibodies, infliximab and adalimumab, inhibit T cell activation and IFN-g production in vitro, whereas the soluble TNF receptor etanercept has no effect. However, interleukin-10 production was equally suppressed by all three anti-TNF drugs. This study was also consistent with that anti-TNF-a monoclonal antibodies caused greater suppression of immunity against tuberculosis than soluble TNF receptor fusion molecule [3, 26].
Sánchez-Moya et al. found that of the 793 psoriasis patients exposed to biological treatment, 20.5% (163) were diagnosed with latent tuberculosis infection before starting biologic therapy. These patients are at greater risk for tuberculosis [27].
Ai et al. found that etanercept is least likely to cause active tuberculosis in rheumatoid arthritis patients receiving TNF-a antagonists. Preventive treatment for latent tuberculosis reduced the tuberculosis risk by 65% [28]. However, it does not prevent all tuberculosis infections.
Murdaca et al. found that infliximab appears to be more responsible for the increased risk of infection than etanercept [29].
Yun et al. studied 112 arthritis patients treated with TNF antagonists in Korea. Latent tuberculosis infection treatment was indicated in 41 (37%) patients. Latent tuberculosis infection was diagnosed in one-third of Korean arthritis patients before initiating anti-TNF therapy [30]. In our study, 358 of 719 psoriasis patients (49.8%) required isoniazid prophylaxis before anti-TNF therapy.
Zhang et al. in their review and meta-analysis found that TNF-a treatment was associated with an increased occurrence of tuberculosis compared to control groups. But the type of drugs was not associated with statistically significant differences in the risk of tuberculosis between patients treated with TNF-a antagonists and control groups [31].
Souto et al. found no significant difference of tuberculosis risk in patients with chronic immune-mediated inflammatory diseases treated with TNF-a antagonists in their meta-analysis of randomized controlled trials and studies [32].
An overview of studies on the occurrence of tuberculosis during anti-TNF treatment is shown in Table 3.
Table 3
Overview of studies on the occurrence of tuberculosis during anti-TNF treatment
| Authors | Main findings |
|---|---|
| Wallis et al. [14] | Wallis et al. reported 248 cases of infliximab-associated and 39 cases of etanercept-associated TB infections, which were recorded from January 1998 to March 2003 in the Adverse Events Reporting System (AERS) database of the Food and Drug Administration |
| Elbek et al. [22] | Elbek et al. evaluated the risk of tuberculosis in 240 patients receiving anti-TNF-a, from December 2005 to December 2007 in Turkey. They reported 2 patients developing tuberculosis (incidence 833/100,000) |
| Ergun et al. [23] | Ergun et al. investigated TB in 370 patients with psoriasis receiving anti-TNF treatment (192 etanercept, 163 infliximab, and 113 adalimumab, mean treatment duration was 12.7 ±11.8 months). They reported 4 (1.08%) patients who developed tuberculosis, 3 pulmonary and 1 gastrointestinal |
| Athimni et al. [33] | Athimni et al. evaluated 82 patients with chronic rheumatic disease using TNF inhibitor therapy (37 men and 45 women). The mean age was 42 ±3.4 years. At inclusion, no patient had a medical history of tuberculosis. Two (2.4%) cases of active tuberculosis occurred under the biologic (infliximab) |
| Coşkunol et al. [34] | Coşkunol et al. evaluated retrospectively 265 (51.0%) patients with ankylosing spondylitis (AS), 175 (33.7%) with rheumatoid arthritis, 35 (6.7%) with Crohn’s disease (CD), 10 (1.9%) with ulcerative colitis (UC), 21 (4.0%) with psoriatic arthritis, 14 (2.7%) with psoriasis vulgaris. Active tuberculosis development was observed in 5 patients (4: pulmonary, 1: extrapulmonary; 3: UC, 2: AS) who all received anti TNF-a treatment (0.96%), infliximab |
| Hernanz et al. [35] | A retrospective review was performed of patients with TB-related uveitis developed under anti-TNF therapy at a single tertiary referral centre from 2014 to 2019 by Hernanz et al. Three patients with presumed TB-associated uveitis while under treatment with anti-TNF therapy for systemic and/or ocular conditions were identified |
| Mankia et al. [36] | Mankia et al. reported 5 cases of active TB occurring in 703 patients treated with anti-TNF therapy over a 10-year period in a central London hospital. Four patients were receiving adalimumab and 1 patient etanercept at the time of TB diagnosis. All patients had normal chest radiographs prior to anti-TNF treatment. They emphasised that a normal chest radiograph does not exclude latent TB |
| Jauregui-Amezaga et al. [37] | Inflammatory bowel disease (IBD) patients from a single centre treated with anti-TNF therapy between January 2000 and September 2011 were identified through a database that prospectively records clinical data, treatments and adverse events. During the study period, 423 patients received anti-TNF therapy. Seven (1.65%) patients developed TB while under anti-TNF treatment. They concluded that in the IBD population under study, incidence of TB infection associated with anti-TNF therapy is higher than that reported in controlled trials and occurs early after treatment initiation |
| Kaptan et al. [38] | 665 patients given anti-TNF therapy in a single centre were included in this study by Kaptan et al. Complete data were obtained from the records of 389 patients. Seven (1.1%) patients were diagnosed with TB. In this cohort, all patients in whom tuberculosis developed had been treated for LTBI and all but one were initially tuberculin-positive. No risk factors have been identified. The current policy of treating tuberculin-positive patients with a 9-month INH regimen does not seem to be fully effective in preventing tuberculosis |
| Sartori et al. [39] | Sartori et al. evaluated retrospectively 5,853 patients with rheumatic diseases of the Public Health System from the Brazilian state, a high TB incidence area, who received prescriptions of TNF inhibitors agents between 2006 and 2016. Forty-three cases of TB were found (2.86 cases per 1000 person-years; 18 times higher than in the general population). Adalimumab and certolizumab users presented a higher risk for TB development compared to etanercept users |
| Lee et al. [40] | Lee et al. evaluated incidence and clinical features of Korean patients with inflammatory bowel disease (IBD) who developed active TB during anti-TNF therapy. Ten cases of active TB developed in patients treated with infliximab (n = 592) or adalimumab (n = 229) for UC (n = 160) or CD (n = 661). The incidence of active TB was 1.2% (10/821): 1.5% (9/592) and 0.4% (1/229) in patients receiving infliximab and adalimumab, respectively |
| Sánchez-Moya et al. [41] | One hundred and forty-four patients with moderate-to-severe psoriasis treated with anti-TNF agents were evaluated by Sánchez-Moya et al. All of them were screened for active TB or LTBI before therapy. A total of 42 (29%) patients were diagnosed with LTBI based on a positive TST or re-TST, and/or signs of past TB in the chest X-ray. All of them received chemoprophylaxis with isoniazid. One patient developed a primary active lymph node TB. |
| Valido et al. [42] | Valido et al. evaluated 942 patients with inflammatory rheumatic diseases who visited their day-care unit to undergo treatment with anti-TNF agents, 5 developed TB between 2006 and 2019 and 3 developed TB between 1999 and 2006, corresponding to 0.85% of the exposed patients |
We reviewed 719 psoriasis patients receiving anti-TNF treatment from 1 June 2007 to 2 January 2023. We identified 4 cases with TB, three pulmonary and one TB peritonitis. None of these patients had personal or family history of TB. All patients with tuberculosis (1.27%, n = 4) had been treated with etanercept. There were no TB cases in the groups receiving infliximab (176 patients) and adalimumab (228 patients). Although monoclonal antibodies demonstrated an increased incidence of TB, the risk of TB reactivation with etanercept therapy is not negligible. Tuberculosis developed in 4 patients using etanercept after a mean time of 54 months in our series but in 1 patient who used several TNF inhibitors it is not possible to attribute development of tuberculosis solely to etanercept. They all used chemoprophylaxis with INH at the beginning of etanercept treatment. None of our patients who developed TB had comorbidities such as diabetes, malignancy or HIV infection.
