Serial QuantiFERON-TB Gold testing in patients with psoriasis treated with ustekinumab

Background There is increasing concern about the risk of latent tuberculosis infection (LTBI) reactivation during the use of biologics for psoriasis. Although ustekinumab had been documented with low risk of tuberculosis, the long-term follow-up of LTBI as determined by QuantiFERON-TB Gold (QFT-G) testing in patients treated with ustekinumab is limited. Objectives This study aims to use serial QFT-G testing as a screening method for detecting LTBI in patients with psoriasis from an intermediate tuberculosis burden country. Methods This retrospective review investigated 134 psoriatic patients in whom ustekinumab was prescribed for at least one year between 2010 and 2016 in National Taiwan University Hospital. All patients underwent annular QFT-G testing during ustekinumab therapy. Results Among the 134 enrolled patients, baseline LTBI rate was 13.4% (18/134). Indeterminate QFT-G result was noted in 5.2% (7/134) of patients and 71.4% (5/7) of them turn to be QFT-G negative during the next testing. 81.3% (109/134) of patients had a negative QFT-G at baseline and the seroconversion rate was 7.3% (8/109) in the serial QFT-G. All the patients in the conversion group were referred to a pulmonologist for evaluation and 81.5% (22/27) of them underwent chemoprophylactic therapy while on ustekinumab. No active TB infection was noted during further follow-up with or without chemoprophylaxis. Conclusions This study revealed that psoriatic patients receiving long-term ustekinumab therapy had a low QFT-G conversion rate (7.3%). The clinical significance of QFT-G conversion remains controversial and needs larger scale trials to investigate.


Introduction
Biologics have been increasingly used in the treatment of moderate to severe psoriasis due to their efficacy and safety compared to pre-existing treatment modalities. Anti-tumor necrosis factors (TNF), and ustekinumab (a monoclonal antibody against the p40 subunit of IL-12 and IL-23) are currently the most commonly used biologics for psoriasis. Biologics are also considered to be safer than conventional systemic agents except for increased granulomatous infections, mainly tuberculosis (TB). Screening and monitoring for both active and latent TB infection (LTBI) before and during treatment with TNF-blockers are considered to be the standard of care [1][2][3][4][5][6]. Among the different methods of LTBI detection, whole blood interferon-γ release assays (IGRAs) offer better sensitivity and specificity than tuberculin skin test (TST) in detecting LTBI (sensitivity 89% vs. 74%; specificity 98% vs. 81%) [7,8], especially in subjects who had received Bacilli Calmette-Guerin (BCG) vaccination previously [9,10]. There are two commercially available IGRAs, T-Spot and QuantiFERON-TB Gold In-Tube test (QFT-GIT; Cellestis Limited, Carnegie, Victoria, Australia), and the latter test is more commonly used due to its lower cost and ease to perform.
The validity of IGRAs in screening LTBI before initiating TNF-α antagonists had been well documented [11][12][13][14]. The prevalence of LTBI varies according to countries, and was 12/110 (11%) [15] and 10/101 (10%) [16] in Taiwan among TNF-blockers users for psoriasis. The optimal timing or need of IGRAs retest has not been standardized in otherwise asymptomatic biologic users, and seroconversion rate may be affected by the background TB prevalence, frequency of IGRAs retest, patient demographics, types of underlying diseases, types of biologics and concomitant medications. Serial QFT tests have revealed annual seroconversion rate between 7%~18% in TNF-blockers users in psoriasis [17][18][19][20], and it was 14.29% in Taiwan [16]. Risks of TB and TBLI reactivation are considered to be lower for using ustekinumab compared to TNF-blockers. However, there are limited studies regarding the serial IGRAs results among patients who received ustekinumab therapy. This study aimed to evaluate the serial QFT-GIT results in patients receiving long-term ustekinumab treatment in Taiwan, a country with an intermediate prevalence of tuberculosis.

Patients
The study was approved by the local investigational research bureau of National Taiwan University Hospital. Screening for LTBI is mandatory before and during biologics use for psoriasis in Taiwan. According to local risk management requirement of all biologics used for psoriasis, QFT-GIT has to be performed once a year in routine practice. Both serial IGRA testing and regular chest X-ray examination were performed. We retrospectively enrolled all psoriatic patients treated with ustekinumab from May 2010 to September 2016 in National Taiwan University Hospital, Taipei and Hsin-Chu, Taiwan. All patients who used ustekinumab and have received quantiferon test at least twice separated by at least one year during the study period were included. The time span between serial QFT for a single patient was at least 12 months respectively. We reviewed the medial records of these patients to evaluate clinical diagnosis, demographic information, medical history and laboratory data.

QuantiFERON-TB Gold In-Tube assay for latent tuberculosis infection
QFT-GIT was performed in all patients with ustekinumab use. Antigen with peptide cocktail simulating the proteins ESAT-6, CFP-10 and TB7.7 were used in the test. The interferon-γ (IFN-γ) values were calculated by subtracting the obtained value with nil antigens. A positive result was defined as IFN-γ! 0.35 IU/ml and positive control value (IFN-γ of mitogen minus nil antigens) ! 0.5 IU/ml. A negative result was defined as 0 < IFN-γ< 0.35 IU/ml and positive control (mitogen) value ! 0.5 IU/ml. Indeterminate result was defined as IFN-γ of nil antigen >8 IU/ml or positive control value < 0.5 IU/ml. Conversion of IGRA was defined as a negative IGRA at baseline and a positive IGRA at follow-up, while reversion of IGRA was defined as a positive IGRA at baseline and negative IGRA at follow-up.

Treatment and follow-up of LTBI
The diagnosis of LTBI in our study was defined as a positive QFT-GIT result and a negative chest X-ray or microbiological assay. All patients diagnosed with LTBI were referred to pulmonologists for evaluation and underwent further survey and chemoprophylactic therapy if recommended. The recommendation of tuberculosis screening and treatment in Taiwan was originally based on anti-TNF agents. However, no modification was made after the introduction of ustekinumab. For anti-TNF agents, an at least 4-week treatment with isoniazid (INH) before anti-TNF agent use was suggested in cases of latent TB detected by either IRGAs or tuberculin skin test. A 9-month INH is needed for latent TB. Since the tuberculosis risk of ustekinumab is supposed to be much lower than anti-TNF agents, and there is a risk of INH resistance for tuberculosis, the infection specialists will decide the appropriateness of INH prophylaxis according to their risk assessment results based on clinical symptoms, chest X ray findings, exposure sources, and sputum culture results. In clinical practice, ustekinumab is usually prescribed at the same time as INH initiation in cases without any other evidence of latent tuberculosis. In fact, in the registration trial of ustekinumab for psoriasis in Taiwan, ustekinumab was approved to be given at the same day as INH initiation [21].

Statistical analysis
Statistical analysis and graphs were done with standard spreadsheet software program using Microsoft Excel 2010 (Microsoft Corporation, Seattle, WA, USA) and Fisher exact test. Statistical significant was defined as p value < 0.05.

Patient characteristics
We enrolled 134 patients to this study in total. Of these patients, the mean age was 47.16 years (standard deviation((SD)) 13.14 years) and the male-to female ratio was 101/33. The mean follow-up months were 20.5 (SD 10.52) and 42 patients (32.1%) also had psoriatic arthritis. Most patients were biologics naïve, 24 patients (17.9%) had received at least 6 months of etanercept and 23 patients (17.2%) had received at least 6 months of adalimumab. During ustekinumab treatment, 17 patients (12.7%) also received concomitant immunosuppressant therapy or other biological agents. Among these agents, methotrexate was the most commonly used drug (n = 12, 9%). The demographic features are presented in Table 1.

Serial QuantiFERON-TB Gold In-Tube assay results
The serial QFT-GIT assay results are summarized in the study flowchart (Fig 1) and histogram (  We divided patient data according to (I) patients receiving only ustekinumab (n = 117) and (II) patients receiving another treatment as well (n = 17). The seroconversion patient number in each group was 7 and 1 respectively and showed no statistically significant difference ( Table 2).
Patients were also divided into two subgroups: (I) biologic-naïve group who had never received biologics agents before using ustekinumab (n = 97) (II) non-biologic-naïve group (n = 37). The seroconversion patient number in each group was 5 and 2 respectively and revealed no statistically significant difference (Table 3).
Totally 27 patients were diagnosed with LTBI in our study. All were referred to pulmonologists for evaluation. Among them, one patient had poor compliance and was lost to follow-up afterwards. 16 patients (59.2%) had received 9-month course of prophylactic therapy with 300 mg of daily isoniazid (INH) and had been follow-up for an average of 17.2 months. 5 patients (18.5%) had only received 1~3 months of INH prophylactic due to nausea, vomiting and elevated liver enzyme and received follow-up for an average of 11.4 months. The rest 5 patients (18.5%) didn't receive prophylactic therapy after pulmonologists' evaluation and had been followed up for an average of 29.2 months. During follow-up, no active TB infection was detected in our study.

Discussion
Biologic agents had become increasingly used in treating moderate to severe psoriasis and psoriatic arthritis since 2002. Opportunistic infections were one of the most important adverse effects during usage of biologic agents. Patients treated with TNF-blockers have increased risk of LTBI reactivation and need to undergo regular TB screening [1]. Regular screening for LTBI during TNF-blockers or ustekinumab treatment had been listed into guidance in many professional dermatologic associations including American Academy of Dermatology (AAD), the Japanese Dermatology Association (JDA) and the European Academy of Dermatology and Venereology (EADV) [5]. Among the screening methods, IGRA such as QFT-GIT was preferred due to higher sensitivity and specificity [7,8]. Risks of TB and LTBI reactivation are generally considered lower for ustekinumab compared to TNF-blockers. While previous studies indicate that patients with inborn errors of IL-12/ 23-IFN-γ-mediated immunity area are at higher risk for developing TB infection [22], cases of TB reactivation or infection during ustekinumab use remain limited. Across five phase III clinical trials of ustekinumab-treated patients with psoriasis, no cases of LTBI reactivation was observed in patients receiving concomitant INH prophylaxis [23]. However, reactivation of LTBI during ustekinumab and low dose oral steroid treatment had been reported in a patient who underwent prophylactic therapy with isoniazid [24]. The seroconversion rate from our study was 7.3%, which is lower than previous study (14.29%) conducted among psoriasis patients treated with anti-tumor necrosis factors in Taiwan [16]. Accuracy of IGRAs during biological therapy has been disputed, especially for patients undergoing anti-TNF blockers. For example, it was found that the development of positive QFT-GIT, observed in 8% of patients treated with infliximab, was not associated with pulmonary or extra-pulmonary tuberculosis [25]. In the other study, adalimumab (a TNF blocker) was found to significantly reduce the IFN-γ levels, potentially causing false negative QF results in a dose-dependent manner [26]. As for ustekinumab, an IL12/IL23 blocker, there are no prior reports of serial quantiferon results in patients receiving ustekinumab for psoriasis. Although, a prior study showed that treatment with ustekinumab resulted in no changes of serum TNF levels [27], maintenance of pulmonary Th1 effector function in chronic tuberculosis requires persistent IL-12 production [28]. Thus, we suspect the QFT seroconversion rate might reflect either a background tuberculosis exposure or from a prior false negative QFT test result at baseline before ustekinumab treatment, such as in the case report of latent tuberculosis reactivation with quantiferon conversion [29]. Weather the fluctuation of IFN-γ level correlated with clinical significance still remained controversial. Some reported that there is a biphasic emergence of TB infection in patients receiving TNF-blockers. Persistently high levels of released IFN-γ (during the first 3 months) or QFT conversion (after 20-24 months) strongly indicate the development of active tuberculosis in patients undergoing long-term anti-TNFα therapy [30]. Some advocate that dynamic changes occurring with serial IFN-γ release assay testing in patients treated with biologic therapy do not correlate with clinical outcome [31]. Scrivo et al proposed a concept of "zone of uncertainty" and re-defined the cut off value of QFT positive result. The zone of uncertainty was defined in the range of 0.2-0.5 IU/mL IFN-γ levels. Any results fluctuate within the uncertainty zone during repeated testing were considered doubtful conversions or doubtful reversions. As seen in Fig 3, some IFN-γ levels did fluctuate within the uncertainty zone. After excluding these doubtful conversions, seroconversion rate would be decreased to 3.7% in our study. In further study, Mancuso et al found that patients with negative TB responses near the cutoff (i.e., !0.25 but <0.35 IU/ml) were, 30 times more likely to convert to positive than those with responses further from the cutoff [32]. As the IGRAs are complicated laboratory tests that may require at least 126 measurements for a single test [33], the in-test variability may contribute to the zone of uncertainty. Use of this borderline zone may reduce the conversion numbers but would increase the number of patients with uncertain results. Further evaluation of risk, serial duplicate testing and longterm follow-up in clinical practice are needed to clarify the usefulness of the concept of "zone of uncertainty".
The QF seroconversion rate might reflect either a background tuberculosis exposure or from a prior false negative QFT test result at baseline. A background tuberculosis exposure may differ significantly in different areas and diseases. Taiwan is an intermediate tuberculosis burden country. Although the prior reports show QF conversion rate to be between 7%~18% in TNF-blockers users in psoriasis [17-20], the 7.3% percent is still lower than the experience of TNF blockers for psoriasis in Taiwan [16]. Moreover, the LTBI rate in our study was 13.4% (18/134), similar to previous studies: 12/110 (11%) [15] and 10/101 (10%) [16] in Taiwan among TNF-blockers users for psoriasis. All the patients with LTBI were referred to pulmonologists for evaluation and most patients had underwent 1~9-month course of INH prophylactic therapy while 5 patients (18.5%) didn't. During further follow-up, no patients from either INH prophylactic or non-prophylactic group developed active tuberculosis. Although chemoprophylactic therapy can efficiently decrease the LTBI reactivation rate in patients who underwent ustekinumab therapy [23], adverse effect such as liver toxicity and increased drug resistance had also been proposed. The possibility of pre-existing or newly emergent INH resistant Mycobacterium tuberculosis strains has also been considered. Thus INH monotherapy, although recommended, is not always prescribed by all referral pulmonologists, especially in low risk patients. Apart from IGRA, clinical vigilance which incorporates careful evaluation of the clinical symptoms and signs, the contact and exposure history, and the TB prevalence rate of the endemic area are needed to guide the final treatment decision.
There are several limitations in the study. First, this is a non-prospective study, which results in some variation in the exact QF retest time. However, most patients received annual tests according to the risk management requirement. Second, some patients received concomitant systemic therapy. However, methotrexate is mainly used and methotrexate has not been found to increase TB risk in prior study when used alone in psoriasis. Additionally, even with concomitant use of TNF blockers, none of the subjects developed tuberculosis in the study [34]. Third, the subjects who had long term follow-up beyond two years are still limited.
In summary, our study revealed that psoriatic patients had a lower QFT-GIT seroconversion rate (7.3%) after ustekinumab therapy compared to anti-tumor necrosis factors therapy. No active tuberculosis was detected during serial follow-up. However, a larger sample size with longer follow-up is still needed to fully assess the TB risks associated with ustekinumab therapy.