Hematologic Malignancy Risk in Inflammatory Arthritis Patients Treated with TNF Inhibitors: The Real-Life Data from the HUR-BIO Registry.

Key Summary Points

Why carry out this study? The debate about the association between tumor necrosis factor inhibitors (TNFi) and hematologic malignancy (HM) is one of the concerns in the use of these drugs. In this study, we investigated whether the risk of HM was increased in patients using TNFi in our inflammatory arthritis cohort. What was learned from the study? We found that the overall risk of HM increased four-fold in patients with inflammatory arthritis patients treated with TNFi compared to the general Turkish population. However, the role of chronic inflammation in the development of HM in inflammatory arthritis should be kept in mind regardless of drug use.

Introduction

Tumor necrosis factor (TNF, cachectin, TNF-α) is a proinflammatory master cytokine with a large variety of biological activities: it modulates the production of inflammatory mediators, cell proliferation, cytotoxicity, and immune regulation [ 1 ]. Dysregulated TNF may contribute to various immune-mediated inflammatory diseases including inflammatory arthritis [ 2 ]. TNF inhibitors (TNFi) are a cornerstone of the treatment of inflammatory diseases such as rheumatoid arthritis (RA) and spondyloarthritis (SpA). Currently, there are five anti-TNF antibodies approved as a treatment option: three monoclonal agents (infliximab, adalimumab, and golimumab), PEGylated Fab’ fragment (certolizumab), and TNF receptor fusion protein (etanercept) [ 3 ]. Turkish reimbursement criteria for these agents require a non-response condition to conventional synthetic disease-modifying anti-rheumatic drugs (csDMARD). Although they have been widely used for more than 20 years, it is still debated whether treatment with TNFi agents is associated with increased hematologic malignancy (HM) incidence. Most observational studies have suggested that TNFi agents in patients with inflammatory arthritis, particularly RA, may be associated with the development of HM at a rate varying across studies [ 4 – 10 ]. In contrast to the accumulating evidence, there are also studies showing that the risk of HM is not increased in patients using TNFi [ 11 – 13 ].

HM risk with TNFi is still not completely understood, although results are conflicting. However, given the essential role of TNF-α in killing tumor cells via apoptosis by inducing natural killer cells and CD8 killer cells, it is possible that loss of such activity may play a role in malignancy [ 14 – 16 ].

The aim of this study is to evaluate the role of TNFi agents on the risk of overall hematologic malignancy in a large cohort of patients with RA, SpA, or PsA who were registered in the Hacettepe University Biologic (HUR-BIO) database, in comparison to the Turkish population-based care registry.

Methods

Study Design and Patient Selection

This study is a retrospective, observational cohort study of Hacettepe University Biological (HUR-BIO) prospective database of patients with inflammatory arthritis performed in the primary referral center for central Turkey. We conducted this study in patients with inflammatory arthritis, including rheumatoid arthritis (RA), spondyloarthritis (SpA), or psoriatic arthritis (PsA) who had at least one visit after initiation of an anti-TNF agent between 2005 and November 2021 and were registered in the HUR-BIO, which was set up in 2005 [ 17 ]. In the HUR-BIO database, the patient record is kept separately for each patient and does not contain duplicate patient records. Patients receiving non-TNF biologic drugs were also determined. Diagnosis of RA, SpA, or PsA was made according to physician decision or related classification criteria [ 18 – 21 ]. Treatment decisions by physicians were in accordance with the Turkish reimbursement criteria. All patients registered in the HUR-BIO database were manually scanned from the hospital medical records by the internal medicine resident (Zeynep Ozge Ozturk) for the presence of hematologic malignancies (HMs). Our study was compliant with the Helsinki Declaration and approved by Hacettepe University Ethical Committee (approval no. 2021/07-30). The final protocol and subject information and informed consent documentation were reviewed and approved by the Ethics Committee.

Data Collection

Data regarding HMs diagnosis were obtained from patient files according to the hematologists’ decision and/or bone marrow or lymph node biopsy results. Patients with a history of HM prior to anti-TNF or non-TNF biologic use were excluded from the statistical analysis. In addition, only one patient who developed HM while using non-TNF bDMARDs was not included in the SIR analysis.

We collected the following data: gender, age at the onset of inflammatory arthritis, type and duration of inflammatory arthritis, age at the onset and type of HMs, HM history before the TNF inhibitors, kind and duration of anti-TNF agents used until the development of HM, presence of secondary Sjögren’s syndrome, and death status. Age- and gender-specific HM incidence of the normal population was rendered from the 2017 Turkish National Cancer Registry (TNCR) data [ 22 ]. The death status of patients was obtained either by chart review or through the national death registration database.

Statistical Analyses

Standardized incidence rates (SIR) were calculated after adjustment for age and gender and compared with age- and gender-specific SIR values abstracted from the 2017 TNCR data. Person-years of follow-up for the patients with inflammatory arthritis registered in the HUR-BIO database were calculated from the date of TNFi initiation to the last follow-up. The observed number of cases is the number of all individuals diagnosed with HMs during the follow-up time. Expected cases represent the total number of patients that would have been reported to the cancer registry within the same period of follow-up as per the TNCR rates under the null hypothesis of no increased risk, given the age and gender structure. As the HUR-BIO registry includes adult patients, the expected number of cases under 18 years of age was not reported. The SIR was calculated by dividing the number of observed cases by the number of expected cases, a ratio greater than 1.00 indicated more cases observed than expected. The 95% confidence interval (CI) for the SIR was calculated independently using OpenEpi software version 3.01 and p values < 0.05 were considered statistically significant.

Statistical analysis was performed using the SPSS software (version 25.0; IBM Corporation, Armonk, NY, USA). The variables were not investigated to determine whether or not they are normally distributed due to the low number of cases with HM. The data of descriptive analysis were expressed as the median, interquartile range (IQR), minimum, and maximum. Categorical variables were given as numbers and percentages.

Results

Study Population

A total of 6139 patients were registered in the HUR-BIO database; 5355 (87.2%) patients used any anti-TNF agent at least once, and the diagnoses of these patients were as follows: 3218 (60.1%) SpA, 1579 (29.1%) RA, and 558 (10.4%) PsA; 3116 (58.2%) patients were female. The median (IQR) age was 46 (37–57) years for all patients receiving TNF inhibitors. In anti-TNF-treated patients, the total follow-up duration was 20,910 patient-years, and the median follow-up duration was 2.6 years.

A total of 784 (12.8%) patients received non-TNF biologic agents other than anti-TNF, and 732 (93.4%) patients were diagnosed with RA, 43 (5.5%) with SpA, and nine (1.1%) with PsA; 623 (79.5%) patients were female. The median (IQR) age was 59 (49–66) years for all patients using non-TNF biologics. In these patients, the total follow-up duration was 1513 patient-years, and the median follow-up duration was 1.2 years.

Since the HM history was before the diagnosis of rheumatic disease in two patients and prior to biologic drugs in two patients, four patients were excluded from the analysis (Fig. 1 ). Fig. 1 Flow chart of the development of hematologic malignancies (HM) in patients with inflammatory arthritis from the HUR-BIO database. anti-TNF anti-tumor necrosis factor, SpA spondyloarthritis, RA rheumatoid arthritis, PsA psoriatic arthritis, FL follicular lymphoma, DLBCL diffuse large B-cell lymphoma, LGL large granular lymphocytic leukemia, MM multiple myeloma, MDS myelodysplastic syndrome, MZL marginal zone lymphoma

Patients’ Characteristics with Hematologic Malignancy

Among all patients in the HURBIO registry, there were 14 patients with HM, and 13 patients using an anti-TNF agent [eight (61.5%) female] had HM on follow-up. Secondary Sjögren’s syndrome was present only in one patient with RA (case 8). In patients receiving TNF inhibitors with HM, the median age at the onset of inflammatory arthritis was 38 (range, 26–67), and the number of patients with SpA, RA, and PsA was seven, four, and two, respectively. The median duration of inflammatory arthritis was 17.7 years (range, 1–33). The median age at the diagnosis of HM was 55.5 (range, 38–76), and types of HM were presented in Table 1 . The median duration of the TNFi usage before the diagnosis of HM was 36 (range, 4–112) months. Five patients deceased on follow-up. Only one patient was diagnosed with large granular lymphocytic leukemia under the rituximab therapy. Figure 2 illustrated the cumulative number of hematologic malignancies in the function of time from onset of anti-TNF therapy. Table 1 Profiles of the patients with hematologic malignancies in inflammatory arthritis using anti-TNF therapy Case Sex Age at the onset of inflammatory arthritis (years) Type of inflammatory arthritis Age at the onset of HM (years) Type of HM Disease duration from the onset of RD to HM (years) Disease activity at the bDMARD initiation Usage of MTX before bDMARD Period from the onset of first bDMARD to HM (months) Type of bDMARD and the usage duration Prognosis 1 M 35 AS 54 Multiple myeloma 19 BASDAI:4 No 4 ADA: 3 months Alive 2 F 48 RA 70 Multiple myeloma 22 DAS-28: 7 Yes 13 ADA: 6 months Dead 3 M 35 RA 58 Diffuse large B-cell lymphoma 23 DAS-28: 6.2 Yes 18 ETA: 18 months Dead 4 M 36 AS 38 Marginal zone lymphoma 2 BASDAI:4.6 No 22 ETA: 22 months Alive 5 F 34 AS 39 Diffuse large B-cell lymphoma 5 BASDAI:5.1 No 48 IFX: 6 months ETA: 6 months Dead 6 F 67 RA 76 Follicular lymphoma 9 DAS-28: 6.4 Yes 65 ETA: 36 months Alive 7 M 39 AS 53 Follicular lymphoma 14 BASDAI:5 No 68 ETA 68 months Alive 8 F 56 RA 68 Follicular lymphoma 12 DAS-28: 5.6 Yes 74 ADA: 25 months Dead 9 F 36 PsA (spondylitis) 44 Plasma cell dyscrasia 8 BASDAI:5.8 Yes 85 IFX: 74 months ETA: 8 months CZP: 3 months Alive 10 F 44 PsA 55 Diffuse large B-cell lymphoma 11 DAS-28:4.5 Yes 96 ADA: 96 months Alive 11 M 37 AS 56 Follicular lymphoma 19 BASDAI:6.2 No 104 ADA: 62 months ETA: 42 months Alive 12 F 26 AS 48 Myelodysplastic syndrome 22 BASDAI:7.4 Yes 110 IFX: 90 months ADA: 14 months Dead 13 F 47 AS 57 Large granular lymphocytic leukemia 10 BASDAI:9.3 Yes 112 IFX: 46 months ETA: 14 months GOL: 52 months Alive 14 F 44 RA 50 Large granular lymphocytic leukemia 6 DAS-28: 6.8 Yes 49 RTX: 96 months Alive ADA adalimumab, AS ankylosing spondylitis, CZP certolizumab, ETA etanercept, GOL golimumab, HM hematologic malignancy, IFX infliximab, RA rheumatoid arthritis, RD rheumatic disease, RTX rituximab, PsA psoriatic arthritis Fig. 2 Cumulative number of hematologic malignancies in function of time from start of first anti-TNF therapy. anti -TNF anti-tumor necrosis factor

The Estimate of Hematological Malignancies

SIR values for both sexes according to different age groups for both overall HMs and lymphomas were given for patients taking TNF inhibitors in Tables 2 and 3 . Because of the low number of index cases, SIR values were calculated for only two age subgroups. Patients using TNF inhibitors had an increased incidence for both overall HMs and lymphomas (SIR 4.23, 95% CI 2.35–7.05 for HMs; SIR 5.26, 95% CI 2.44–9.99 for lymphomas). When analyzed separately, the increased incidence was observed for both males and females. Ten of 13 patients with HMs were under the age of 65. In this age group, there was a higher incidence of HMs in both men (SIR 5.15, 95% CI 1.88–11.43) and women (SIR 4.76, 95% CI 1.74–10.55). In patients aged 65 years or older, there were no male patients with HMs. SIR was also calculated separately according to RA and SpA, but age and sex groups were not differentiated due to the low number of index cases. Although the overall SIR of HMs appeared to be increased in patients with RA receiving TNF inhibitors, it was not statistically significant (SIR 2.55, 95% CI 0.81–6.16, p = 0.09). SpA patients using TNF inhibitors had a nearly fivefold increased risk of overall HMs (SIR 4.75, 95% CI 2.32–8.72, p < 0.001). Table 2 Standardized incidence rates for different age cut-offs in both sexes for patients with hematologic malignancies Gender Age Observed/expected cases SIR 95% confidence interval p Total All ages a 13/3.07 4.23 2.35–7.05 < 0.001 Male 20–64 years ( n = 2059) ≥ 65 years ( n = 180) Overall ( n = 2239) 5/0.97 0/0.39 5/1.36 5.15 NA 3.67 1.88–11.43 – 1.34–8.14 0.003 – 0.015 Female 20–64 years ( n = 2667) ≥ 65 years ( n = 449) Overall ( n = 3116) 5/1.05 3/0.66 8/1.71 4.76 4.54 4.67 1.74–10.55 1.15–12.37 2.17–8.88 0.005 0.034 < 0.001 NA not applicable, SIR standardized incidence rates a Includes patients ≥ 18 years Table 3 Standardized incidence rates for different age cut-offs in both sexes for patients with lymphoma Gender Age Observed/expected cases SIR 95% confidence interval p Total All ages a 8/1.52 5.26 2.44–9.99 < 0.001 Male 20–64 years ( n = 2059) ≥ 65 years ( n = 180) Overall ( n = 2239) 4/0.5 0/0.17 4/0.67 8 – 5.97 2.54–19.3 NA 1.89–14.4 0.001 – 0.005 Female 20–64 years ( n = 2667) ≥ 65 years (n = 449) Overall ( n = 3116) 2/0.53 2/0.32 4/0.85 3.77 6.25 4.70 0.63–12.47 1.04–20.65 1.49–11.35 0.116 0.045 0.012 NA not applicable, SIR standardized incidence rates a Includes patients ≥ 18 years

In patients receiving non-TNF biologic agents, only one patient was diagnosed with HM. The SIR value could not be calculated due to the low number of patients.

Discussion

The review of the HUR-BIO database for the patients using TNF inhibitors revealed 5355 patients with inflammatory arthritis followed up in the study center between 2005 and 2021. Thirteen of these patients were diagnosed with HM after using anti-TNF therapy. The standardized incidence ratios indicated that the risk of hematologic malignancy in the inflammatory arthritis patients with anti-TNF therapy was more than four times higher compared to the Turkish National Cancer Registry Data.

The data from previous studies on the association between hematologic malignancies and the use of TNFi are conflicting. In one of the first, data from the global Swedish national registry (ARTIS) revealed the risk of HMs in 53,057 RA patients, 4160 of whom were using TNFi. This study showed that the risk of hematologic malignancy increased two times, and the risk of lymphoma increased approximately three times in patients using TNFi compared to the general population. However, these risks were not different from RA patients who did not use TNFi [ 4 ]. Similarly, results published from the same database in 2009 indicated an increased risk of lymphoma in RA patients receiving TNFi compared to the general population without any significant differences between TNFi-naive RA patients [ 23 ]. In another large cohort from the Lombardy Rheumatology Network (LORHEN) registry, RA patients using TNFi had a fourfold increased risk of hematologic malignancy and sixfold increased risk of lymphoma compared to the general population [ 5 ]. On the contrary, data published from the US showed that the risk of lymphoma was not increased in RA patients receiving TNFi [ 11 , 12 ]. In our cohort, RA patients using TNFi compared to the general Turkish population had a trend towards a higher incidence of overall HM, but it was not statistically significant (SIR 2.55, 95% CI 0.81–6.16). A separate comparison in RA patients using TNFi could not be calculated because no group received only csDMARD.

Compared to RA, data on overall hematologic malignancies or lymphoma risk in patients with spondyloarthritides are limited. In a study from the national Taiwan database, Chang et al. reported that patients with AS had a higher risk of overall hematological malignancies (SIR 2.10; 95% CI 1.32–3.19) than the general population [ 6 ]. Additionally, in a nationwide population-based study in Korea, AS patients were associated with a threefold increased risk of lymphoma and multiple myeloma [ 7 ]. In an Italian cohort including 399 inflammatory arthritis patients, 70% of whom received anti-TNF therapy, the risk of hematological malignancy was significantly increased fivefold in RA patients, while the twofold increase in AS or PsA patients was not statistically significant [ 8 ]. However, two studies found an increased risk of overall hematologic malignancies in patients with PsA. Harberg and colleagues represented that the rate of incident hematologic cancer were higher in patients with PsA who received drugs including DMARD, biologics, and immunosuppressant for PsA compared with those who did not (IRs, 1.6 [95% CI, 0.9–2.4] and 0.4 [95% CI, 0.3–0.7], respectively; IRR, 3.50 [95% CI, 1.92–6.36]) [ 9 ]. A large prospective cohort study from four Nordic countries (Sweden, Denmark, Iceland, and Finland) showed a higher risk of lymphoma in patients with PsA on TNFi compared with the general population (SIR 1.84; 95% CI 1.20–2.82) [ 24 ]. Conversely, it was shown that the risk of lymphoma was not increased in TNFi-exposed versus TNFi-naive AS and PsA patients from the ARTIS and Danish Biologics (DANBIO) data [ 13 ]. We found that the patients with SpA using TNFi, also including PsA, had a nearly fivefold increased risk of overall HM (SIR 4.75, 95% CI 2.32–8.72). We could not record further information because our database did not include TNF-naive patients, and the number of patients with HM was low for the separate SIR calculation according to types of HM.

In our study, in inflammatory arthritis patients, cumulatively, 13 cases of hematologic malignancies, eight of which were female and five males, have been observed. The results showed a statistically significant difference compared with the general population (SIR 4.23; 95% CI 2.35–7.05), matched for age and sex. Similarly, in Dreyer and colleagues’ study, which included arthritis patients using TNFi from the DANBIO database, the authors found that hematologic malignancy incidence was higher than expected, with the standardized incidence ratio 2.24 (1.33–3.79) for the Danish population. In addition, the risk of Hodgkin’s lymphoma was reported to be higher in patients with arthritis receiving TNFi than the general population (SIR 8.27; 95% CI 2.07–33.1) [ 10 ]. We found that the risk of lymphoma increased fivefold compared to the general population in our cohort (SIR 5.26; 95% CI 2.44–9.99). Also, although we showed an increased risk of HM in patients with SpA, we could not confirm this in RA patients. This result can be explained by the fact that the number of SpA patients is more than two times higher than the RA patients.

The mechanism by which TNF-α protects against cancer, and thereby how its inhibition may promote cancer, is not well understood. TNF-α induces the killing of tumor cells through apoptosis via interaction with death-domain regions in tumor cells by stimulating natural killer cells and inducing CD8 killer cells [ 14 – 16 ]. Given the essential role of TNF-α in apoptosis, it is possible that loss of such activity plays a role in malignancy. There are fundamental differences in the specificity of binding and the way they neutralize TNF-α between monoclonal antibodies (mAbs) and receptor fusion protein. The monoclonal antibodies which bind membrane and soluble TNF-α receptors may perform stronger effects on the function of T-cells and the growth of B-cell clones compared to the soluble receptor, which only neutralizes soluble TNF-α [ 25 ]. This was consistent with the result that the risk of lymphoma was higher with mAbs than with soluble receptor anti-TNF therapy in the RATIO registry collected 38 cases of lymphoma (SIR [95% CI], 4.1 [2.3–7.1] for infliximab and 3.6 [2.3–5.6] for adalimumab vs. 0.9 [0.4–1.8] for etanercept) [ 26 ]. There were eight patients diagnosed with lymphoma in our cohort. Because of the small numbers, we did not calculate the SIR for individual TNF therapies.

Although methotrexate, which is frequently used in the treatment of inflammatory diseases, has been associated with an increased risk of treatment-related lymphoma, especially in RA patients [ 27 ], this notion has not been sufficiently supported by large, observational population-based studies [ 28 – 30 ]. Additionally, in these patients, high inflammatory activity rather than csDMARD therapy has been reported as a major predictor of increased lymphoma risk [ 28 ]. In our study, we could not adjust for the inflammatory activity or disease severity. However, since all patients used methotrexate and/or sulfasalazine prior to TNFi therapy, the requirement for TNFi may presumably reflect the group of patients with more severe disease.

Similar to our study, studies showing that the use of TNFi increases the risk of developing HM have been mentioned above. However, there are some studies that show an increased risk of HM in inflammatory arthritis independent of TNFi use. In a large cohort of AS that excludes patients using TNFi, the increased risk of non-Hodgkin’s lymphoma, chronic lymphocytic leukemia, and multiple myeloma has been demonstrated [ 31 ]. In a study comparing PsA cohort with non-PsA cohort using the UK Clinical Practice Research Datalink, the rate of hematological cancer cases was found to be slightly higher in the PsA cohort [ 32 ]. High inflammatory activity rather than treatment has been reported as an important predictor of increased lymphoma risk in patients with RA [ 28 ]. It seems that chronic inflammation plays an important role in the development of HM.

Some strengths and limitations of our study need to be addressed. Firstly, we could not fully control the potential confounding of the disease due to the lack of a control group of patients receiving only csDMARD. It is a major confounder for increased HM incidence. Although this study represented the so far assessment of hematologic malignancy risk following initiation of TNFi therapy in patients with inflammatory arthritis, the small number of patients prevented performing risk analysis separately for both hematologic malignancies subgroups and individual TNF therapies. Additionally, patients diagnosed with hematologic malignancy out of our institution might have been overlooked, so our findings might be biased. However, the data of the patients who were lost to follow-up in our clinic were obtained from the personal health record system of the Republic of Turkey Ministry of Health by patients' permission, if patients could be reached. Therefore, a significant deviation in SRI rates is unlikely. Another limitation was that there might be more comorbidities in the study population than the general population. This may result in an overestimation of the risk of hematologic malignancy. The final limitation was the lack of adjustment for inflammatory activity or disease severity. Despite all these limitations, the major strength of our study was the inclusion of more than 6000 biologically treated patients with inflammatory arthritis in the HUR BIO database. In addition, the follow-up data of patients with hematological malignancies were complete.

Conclusions

In conclusion, we found an increased risk in the overall hematologic malignancy in TNFi-treated inflammatory arthritis patients compared to the general population. This result supports the literature data regarding the critical safety of TNF inhibitors in inflammatory arthritis patients. Although no specific screening recommendations for hematologic malignancy in TNFi receiving patients are available, closer collaboration between the rheumatologist and hematologist/oncologist may reduce the growing concern about HM risk, resulting in a better quality of life for all our patients. However, it is clear that large, observational population-based controlled studies are needed to specifically investigate the risk of developing HM in TNFi due to the limitations of our study.

Acknowledgements

We thank all rheumatology department at Hacettepe University Medical School for reporting to the HUR-BIO Registry.

Funding

No funding or sponsorship was received for this study or publication of this article.

Author Contributions

All authors were involved in the writing and reviewing process and have read and approved this manuscript. Additional contributions to this body of work were as follows: Emine Duran, Umut Kalyoncu and Ali Ihsan Ertenli: conception/design of the study, acquisition, and interpretation of data. Zeynep Ozge Ozturk and Emine Duran: acquisition of data. Emre Bilgin, Yahya Buyukasık, Omer Dizdar, Levent Kilic, Ali Akdoğan, Omer Karadag, Şule Apraş Bilgen, Sedat Kiraz: conception/design of the study and interpretation of data. Gozde Kubra Yardimci, Bayram Farisogullari, Zehra Özsoy, Gizem Ayan, Gullu Sandal Uzun, Mustafa Ekici, Erdinc Unaldi: interpretation of data.

Prior Presentation

Data included in this manuscript have partially been presented as a poster at the EULAR European Congress of Rheumatology, June 1-4, 2022, Bella Center Copenhagen, Copenhagen, Denmark: Duran E, Öztürk ZÖ, Bilgin E, et al. POS 1440 Hematological Malignancies and anti-TNF in inflammatory arthritis: The Real Life Data From the HUR-BIO Registry. Annals of the Rheumatic Diseases 2022;81:1064.

Disclosures

Emine Duran, Zeynep Ozge Ozturk, Emre Bilgin, Yahya Büyükaşık, Omer Dizdar, Gozde Kubra Yardimci, Bayram Farisogullari, Zehra Özsoy, Gizem Ayan, Gullu Sandal Uzun, Mustafa Ekici, Erdinc Unaldi, Levent Kilic, Ali Akdoğan, Omer Karadag, Şule Apraş Bilgen, Sedat Kiraz, Umut Kalyoncu, and Ali Ihsan Ertenli have nothing to disclose.

Compliance with Ethics Guidelines

The final protocol and subject information and informed consent documentation were reviewed and approved by the Ethics Committee of the Hacettepe University, Turkey (approval no. 2021/07-30). This study was conducted in accordance with the Declaration of Helsinki.

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

References

1. Tracey D Klareskog L Sasso EH Tumor necrosis factor antagonist mechanisms of action: a comprehensive review Pharmacol Ther 2008 117 2 244 279 10.1016/j.pharmthera.2007.10.001 18155297 Tracey D, Klareskog L, Sasso EH, et al. Tumor necrosis factor antagonist mechanisms of action: a comprehensive review. Pharmacol Ther. 2008;117(2):244–79. 18155297 10.1016/j.pharmthera.2007.10.001

2. Silva LC Ortigosa LC Benard GJI Anti-TNF-α agents in the treatment of immune-mediated inflammatory diseases: mechanisms of action and pitfalls Immunotherapy 2010 2 6 817 833 10.2217/imt.10.67 21091114 Silva LC, Ortigosa LC, Benard GJI. Anti-TNF-α agents in the treatment of immune-mediated inflammatory diseases: mechanisms of action and pitfalls. Immunotherapy. 2010;2(6):817–33. 21091114 10.2217/imt.10.67

3. Monaco C Nanchahal J Taylor P Anti-TNF therapy: past, present and future Int Immunol 2015 27 1 55 62 10.1093/intimm/dxu102 25411043 PMC4279876 Monaco C, Nanchahal J, Taylor P, et al. Anti-TNF therapy: past, present and future. Int Immunol. 2015;27(1):55–62. 25411043 10.1093/intimm/dxu102 PMC4279876

4. Askling J Fored CM Baecklund E Haematopoietic malignancies in rheumatoid arthritis: lymphoma risk and characteristics after exposure to tumour necrosis factor antagonists Ann Rheum Dis 2015 64 10 1414 1420 10.1136/ard.2004.033241 PMC1755232 15843454 Askling J, Fored CM, Baecklund E, et al. Haematopoietic malignancies in rheumatoid arthritis: lymphoma risk and characteristics after exposure to tumour necrosis factor antagonists. Ann Rheum Dis. 2015;64(10):1414–20. 10.1136/ard.2004.033241 PMC1755232 15843454

5. Pallavicini FB Caporali R Sarzi-Puttini P Tumour necrosis factor antagonist therapy and cancer development: analysis of the LORHEN registry Autoimmun Rev 2010 9 3 175 180 10.1016/j.autrev.2009.07.006 19647103 Pallavicini FB, Caporali R, Sarzi-Puttini P, et al. Tumour necrosis factor antagonist therapy and cancer development: analysis of the LORHEN registry. Autoimmun Rev. 2010;9(3):175–80. 19647103 10.1016/j.autrev.2009.07.006

6. Chang CC Chang CW Nguyen PA Ankylosing spondylitis and the risk of cancer Oncol Lett 2017 14 2 1315 1322 10.3892/ol.2017.6368 28789346 PMC5529950 Chang CC, Chang CW, Nguyen PA, et al. Ankylosing spondylitis and the risk of cancer. Oncol Lett. 2017;14(2):1315–22. 28789346 10.3892/ol.2017.6368 PMC5529950

7. Lee E, Han K-D, Song Y-W, et al. Cancer risk in patients with ankylosing spondylitis: a nationwide population-based dynamic cohort study from Korea. In: Arthritis & rheumatology. Hoboken : Wiley; 2019.

8. Fantò M Peragallo MS Pietrosanti M Risk of malignancy in patients with rheumatoid arthritis, psoriatic arthritis and ankylosing spondylitis under immunosuppressive therapy: a single-center experience Intern Emerg Med 2016 11 1 31 40 10.1007/s11739-015-1270-0 26099279 Fantò M, Peragallo MS, Pietrosanti M, et al. Risk of malignancy in patients with rheumatoid arthritis, psoriatic arthritis and ankylosing spondylitis under immunosuppressive therapy: a single-center experience. Intern Emerg Med. 2016;11(1):31–40. 26099279 10.1007/s11739-015-1270-0

9. Hagberg KW Li L Peng M Rates of cancers and opportunistic infections in patients with psoriatic arthritis compared with patients without psoriatic arthritis J Clin Rheumatol 2016 22 5 241 247 10.1097/RHU.0000000000000364 26886439 PMC4966919 Hagberg KW, Li L, Peng M, et al. Rates of cancers and opportunistic infections in patients with psoriatic arthritis compared with patients without psoriatic arthritis. J Clin Rheumatol. 2016;22(5):241–7. 26886439 10.1097/RHU.0000000000000364 PMC4966919

10. Dreyer L Mellemkjær L Andersen AR Incidences of overall and site specific cancers in TNFα inhibitor treated patients with rheumatoid arthritis and other arthritides—a follow-up study from the DANBIO Registry Ann Rheum Dis 2013 72 1 79 82 10.1136/annrheumdis-2012-201969 22945500 Dreyer L, Mellemkjær L, Andersen AR, et al. Incidences of overall and site specific cancers in TNFα inhibitor treated patients with rheumatoid arthritis and other arthritides—a follow-up study from the DANBIO Registry. Ann Rheum Dis. 2013;72(1):79–82. 22945500 10.1136/annrheumdis-2012-201969

11. Wolfe F Michaud K The effect of methotrexate and anti-tumor necrosis factor therapy on the risk of lymphoma in rheumatoid arthritis in 19,562 patients during 89,710 person-years of observation Arthritis Rheum 2007 56 5 1433 1439 10.1002/art.22579 17469100 Wolfe F, Michaud K. The effect of methotrexate and anti-tumor necrosis factor therapy on the risk of lymphoma in rheumatoid arthritis in 19,562 patients during 89,710 person-years of observation. Arthritis Rheum. 2007;56(5):1433–9. 17469100 10.1002/art.22579

12. Greenberg J, Strand V, Keystone E, et al. TNF inhibitors (TNF-I) and risk of malignancy in 8,072 RA patients followed over 15,495 patient years (abstract 282). In: American College of Rheumatology annual meeting. 2007.

13. Hellgren K Dreyer L Arkema EV Cancer risk in patients with spondyloarthritis treated with TNF inhibitors: a collaborative study from the ARTIS and DANBIO registers Ann Rheum Dis 2017 76 1 105 111 10.1136/annrheumdis-2016-209270 27147709 Hellgren K, Dreyer L, Arkema EV, et al. Cancer risk in patients with spondyloarthritis treated with TNF inhibitors: a collaborative study from the ARTIS and DANBIO registers. Ann Rheum Dis. 2017;76(1):105–11. 27147709 10.1136/annrheumdis-2016-209270

14. Kashii Y Giorda R Herberman RB Constitutive expression and role of the TNF family ligands in apoptotic killing of tumor cells by human NK cells J Immunol 1999 163 10 5358 5366 10.4049/jimmunol.163.10.5358 10553060 Kashii Y, Giorda R, Herberman RB, et al. Constitutive expression and role of the TNF family ligands in apoptotic killing of tumor cells by human NK cells. J Immunol. 1999;163(10):5358–66. 10553060

15. Katschinski DM Robins HI Schad M Role of tumor necrosis factor alpha in hyperthermia-induced apoptosis of human leukemia cells Cancer Res 1999 59 14 3404 3410 10416602 Katschinski DM, Robins HI, Schad M, et al. Role of tumor necrosis factor alpha in hyperthermia-induced apoptosis of human leukemia cells. Cancer Res. 1999;59(14):3404–10. 10416602

16. Screaton G Xu XN T cell life and death signalling via TNF-receptor family members Curr Opin Immunol 2000 12 3 316 322 10.1016/S0952-7915(00)00093-5 10781406 Screaton G, Xu XN. T cell life and death signalling via TNF-receptor family members. Curr Opin Immunol. 2000;12(3):316–22. 10781406 10.1016/s0952-7915(00)00093-5

17. Nikiphorou E Buch MH Hyrich KL Biologics registers in RA: methodological aspects, current role and future applications Nat Rev Rheumatol 2017 13 8 503 510 10.1038/nrrheum.2017.81 28569267 Nikiphorou E, Buch MH, Hyrich KL. Biologics registers in RA: methodological aspects, current role and future applications. Nat Rev Rheumatol. 2017;13(8):503–10. 28569267 10.1038/nrrheum.2017.81

18. Arnett FC Edworthy SM Bloch DA The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis Arthritis Rheum 1988 31 3 315 324 10.1002/art.1780310302 3358796 Arnett FC, Edworthy SM, Bloch DA, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum. 1988;31(3):315–24. 3358796 10.1002/art.1780310302

19. Aletaha D Neogi T Silman AJ 2010 rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative Arthritis Rheum 2010 62 9 2569 2581 10.1002/art.27584 20872595 Aletaha D, Neogi T, Silman AJ, et al. 2010 rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis Rheum. 2010;62(9):2569–81. 20872595 10.1002/art.27584

20. Rudwaleit M van der Heijde D Landewé R The development of Assessment of SpondyloArthritis international Society classification criteria for axial spondyloarthritis (part II): validation and final selection Ann Rheum Dis 2009 68 6 777 783 10.1136/ard.2009.108233 19297344 Rudwaleit M, van der Heijde D, Landewé R, et al. The development of Assessment of SpondyloArthritis international Society classification criteria for axial spondyloarthritis (part II): validation and final selection. Ann Rheum Dis. 2009;68(6):777–83. 19297344 10.1136/ard.2009.108233

21. Taylor W Gladman D Helliwell P Classification criteria for psoriatic arthritis: development of new criteria from a large international study Arthritis Rheum 2006 54 8 2665 2673 10.1002/art.21972 16871531 Taylor W, Gladman D, Helliwell P, et al. Classification criteria for psoriatic arthritis: development of new criteria from a large international study. Arthritis Rheum. 2006;54(8):2665–73. 16871531 10.1002/art.21972

22. Turkish National Cancer Registry (TNCR). https://hsgm.saglik.gov.tr/tr/kanser-istatistikleri/yillar/2017-turkiye-kanser-i-statistikleri.html . 2017.

23. Askling J Baecklund E Granath F Anti-tumour necrosis factor therapy in rheumatoid arthritis and risk of malignant lymphomas: relative risks and time trends in the Swedish Biologics Register Ann Rheum Dis 2009 68 5 648 653 10.1136/ard.2007.085852 18467516 Askling J, Baecklund E, Granath F, et al. Anti-tumour necrosis factor therapy in rheumatoid arthritis and risk of malignant lymphomas: relative risks and time trends in the Swedish Biologics Register. Ann Rheum Dis. 2009;68(5):648–53. 18467516 10.1136/ard.2007.085852

24. Ballegaard C Hellgren K Cordtz R OP0005 incidence of overall and site-specific cancers in TNF inhibitor treated patients with psoriatic arthritis: a population-based cohort study from 4 Nordic countries 2019 London BMJ Publishing Group Ltd Ballegaard C, Hellgren K, Cordtz R, et al. OP0005 incidence of overall and site-specific cancers in TNF inhibitor treated patients with psoriatic arthritis: a population-based cohort study from 4 Nordic countries. London: BMJ Publishing Group Ltd; 2019.

25. Scallon B Cai A Solowski N Binding and functional comparisons of two types of tumor necrosis factor antagonists J Pharmacol Exp Ther 2002 301 2 418 426 10.1124/jpet.301.2.418 11961039 Scallon B, Cai A, Solowski N, et al. Binding and functional comparisons of two types of tumor necrosis factor antagonists. J Pharmacol Exp Ther. 2002;301(2):418–26. 11961039 10.1124/jpet.301.2.418

26. Mariette X Gottenberg JE Ravaud P Registries in rheumatoid arthritis and autoimmune diseases: data from the French registries Rheumatology (Oxford) 2011 50 1 222 229 10.1093/rheumatology/keq368 21148156 Mariette X, Gottenberg JE, Ravaud P, et al. Registries in rheumatoid arthritis and autoimmune diseases: data from the French registries. Rheumatology (Oxford). 2011;50(1):222–9. 21148156 10.1093/rheumatology/keq368

27. Smedby KE Askling J Mariette X Autoimmune and inflammatory disorders and risk of malignant lymphomas—an update J Intern Med 2008 264 6 514 527 10.1111/j.1365-2796.2008.02029.x 19017176 Smedby KE, Askling J, Mariette X, et al. Autoimmune and inflammatory disorders and risk of malignant lymphomas—an update. J Intern Med. 2008;264(6):514–27. 19017176 10.1111/j.1365-2796.2008.02029.x

28. Baecklund E Iliadou A Askling J Association of chronic inflammation, not its treatment, with increased lymphoma risk in rheumatoid arthritis Arthritis Rheum 2006 54 3 692 701 10.1002/art.21675 16508929 Baecklund E, Iliadou A, Askling J, et al. Association of chronic inflammation, not its treatment, with increased lymphoma risk in rheumatoid arthritis. Arthritis Rheum. 2006;54(3):692–701. 16508929 10.1002/art.21675

29. Solomon DH Kremer JM Fisher M Comparative cancer risk associated with methotrexate, other non-biologic and biologic disease-modifying anti-rheumatic drugs Semin Arthritis Rheum 2014 43 4 489 497 10.1016/j.semarthrit.2013.08.003 24012043 Solomon DH, Kremer JM, Fisher M, et al. Comparative cancer risk associated with methotrexate, other non-biologic and biologic disease-modifying anti-rheumatic drugs. Semin Arthritis Rheum. 2014;43(4):489–97. 24012043 10.1016/j.semarthrit.2013.08.003

30. Bernatsky S Clarke AE Suissa S Hematologic malignant neoplasms after drug exposure in rheumatoid arthritis Arch Intern Med 2008 168 4 378 381 10.1001/archinternmed.2007.107 18299492 Bernatsky S, Clarke AE, Suissa S. Hematologic malignant neoplasms after drug exposure in rheumatoid arthritis. Arch Intern Med. 2008;168(4):378–81. 18299492 10.1001/archinternmed.2007.107

31. Alehashemi S Ward MM Risk of hematologic malignancies in elderly patients with ankylosing spondylitis: a cohort study and systematic review [published correction appears in Mayo Clin Proc. 2023 Feb 13] Mayo Clin Proc 2023 98 1 100 110 10.1016/j.mayocp.2022.06.030 36470752 PMC9822846 Alehashemi S, Ward MM. Risk of hematologic malignancies in elderly patients with ankylosing spondylitis: a cohort study and systematic review [published correction appears in Mayo Clin Proc. 2023 Feb 13]. Mayo Clin Proc. 2023;98(1):100–10. 36470752 10.1016/j.mayocp.2022.06.030 PMC9822846

32. Hagberg KW Li L Peng M Paris M Shah K Jick SS Rates of cancers and opportunistic infections in patients with psoriatic arthritis compared with patients without psoriatic arthritis J Clin Rheumatol 2016 22 5 241 247 10.1097/RHU.0000000000000364 26886439 PMC4966919 Hagberg KW, Li L, Peng M, Paris M, Shah K, Jick SS. Rates of cancers and opportunistic infections in patients with psoriatic arthritis compared with patients without psoriatic arthritis. J Clin Rheumatol. 2016;22(5):241–7. 26886439 10.1097/RHU.0000000000000364 PMC4966919