Abstract
While biologics have proven to be effective in treating immune-mediated inflammatory diseases (IMIDs), the label of most biologics carries a warning regarding their use in patients with active cancer or a history of cancer. Consequently, there has been a reluctance to prescribe these medications to patients with a history of cancer, and little is known about the risk of new and/or recurrent cancer in patients with IMIDs treated with biologics and a previous cancer. This systematic review and meta-analysis searched PubMed, Embase, Cochrane Library, and Web of Science up to February 23, 2024, for studies on the risk of new and/or recurrent cancer following treatment with biologics in patients with IMIDs and a previous cancer compared to controls (PROSPERO; CRD42024516899). Controls included patients receiving non-biological or no therapy. Risk estimates were calculated with random-effects meta-analysis with I2 statistics to estimate between-study heterogeneity. In total, 20 articles comprising 4736 patients treated with biologics with 15,646 patient-years of follow-up were included. Treatment with tumor necrosis factor (TNF)-α inhibitors (TNFi), interleukin (IL)-12/23 inhibitors or vedolizumab did not appear to lead to increased risk of new or recurrent cancer in patients with IMIDs and a previous cancer compared to conventional systemics or no therapy. The results are reassuring to patients and physicians prescribing biologics to patients with IMIDs. However, further studies are needed especially on newer biologics and decision on initiation of biologics in patients with cancers should be based on an individual assessment.
Supplementary Information
The online version contains supplementary material available at 10.1007/s10238-025-01738-4.
Keywords: Adalimumab, Infliximab, Psoriasis, Mb Crohns, Colitis ulcerosa, Rheumatoid arthritis
Introduction
Since their introduction, some biologics have been used for treatment of several immune-mediated inflammatory diseases (IMIDs) by inhibition of inflammatory markers such as tumor necrosis factor (TNF), interleukin (IL)-17, IL-6, IL-12/23, IL-23 and the gut-selective α4β7-integrin [1–6]. While biologics have proven effective in the treatment of IMIDs including psoriasis, psoriatic arthritis (PsA), rheumatoid arthritis (RA) and inflammatory bowel disease (IBD) [7–10], concerns about a potential increased risk of cancer, particularly for TNF-α inhibitors (TNFi), have been raised [11, 12]. The label of most biologics carries a warning regarding their use in patients with active cancer or a history of cancer. Consequently, there has been a reluctance to prescribe these medications to patients with a history of cancer [13]. Ultimately, this might lead to confounding by indication where patients with higher risk of cancer recurrence being prescribed conventional systemics. However, studies examining the carcinogenic properties of biologics are conflicting. In a recent unpublished systematic review and meta-analysis of patients with IMIDs treated with biologics without a history of cancer, we did not find an increased risk of cancer overall or for individual cancers. Despite this, certain patients with severe disease, treatment resistance for conventional systemics, or corticosteroid dependence need other treatment options and are given biologics. Indeed, studies examining risk of new or recurrent cancers in patients with IMIDs and a previous cancer treated with biologics are accumulating [14–17]. Although these studies in general have not found increased risk of cancer recurrences following treatment with biologics [14, 15, 17], they may not by themselves be powered to identify such an increased risk. Therefore, we conducted a systematic review and meta-analysis examining the risk of new or recurrent cancer in patients with IMIDs and a previous cancer treated with various biologics.
Materials and methods
Search strategy and selection criteria
Before initiating the study, a study protocol was registered at the International Prospective Register of Systematic Reviews (PROSPERO; CRD42024516899). The Preferred Items for Systematic Review and Meta-Analysis (PRISMA) guidelines [18] were followed.
Three authors (DI, NL, and MBJ) independently screened four databases (PubMed, Embase, Web of Science, and Cochrane Library) from inception through February 23, 2024, based on a search string detailed in supplementary Table 1. As only biologics with IMIDs as the primary indication was included in this study, rituximab was not included in the search string. Articles were extracted and compiled into the web-tool Rayyan [19]. Duplicates were manually removed. All titles and abstracts were screened, and for relevant articles the full-text version was retrieved and assessed for eligibility.
Studies had to include patients with psoriasis, PsA, RA, ankylosing spondylitis, spondylarthritis, or IBD and a previous cancer undergoing treatment with biologics. (Supplementary Table 1).
The search was limited to original articles conducted on human participants with no restrictions on language, sex or geography. Case series and case reports were excluded. Studies had to report the number of new and/or recurrent cancers compared to a control group. Control groups consisted of patients treated with conventional therapies (e.g., methotrexate or thiopurines) or no immunosuppressive therapy.
Data analysis
Data was extracted and reviewed by three authors (DI, NL, and MBJ). Where possible, the following data were extracted; first author surname, year of publication, country, population source and study period, study design, initial cancer type, follow-up time, type of biologic or conventional systemic, and number of patients receiving therapy, age, number of males and females, number of patient-years (PYs), effect estimates, and adjustments for confounders.
The quality of cohort studies was evaluated using the Newcastle–Ottawa Scale (NOS) with a maximum of 9 points. Scores ≥ 7 were considered good quality, and scores < 7 were considered lower study quality.
The primary outcome was cancer recurrence (same as the initial cancer subtype) and/or development of a new cancer (different from the initial cancer subtype).
All meta-analyses were conducted using the meta package in R Statistical Software (version 4.2.0) [20, 21]. Extracted data from each study on the number of cancer events and the number of patients or PYs in each treatment group, as well as adjusted hazard ratios (HRs) were used to compare the risk of new and/or recurrent cancer in patients with IMIDs and previous cancer treated with biologics (stratified by target) to a control group (treated with DMARDs or no immunosuppressants).
The raw data were pooled using a random-effects model (DerSimonian-Laird [22]) with inverse-variance weighting and Knapp-Hartung adjustments to produce pooled estimates of the crude risk ratio (RR) or incidence rate ratio (IRR). Since individual data was not available, the calculated outcomes were unadjusted. In case of zero-cells, a continuity correction of 0.5 was added. For studies reporting adjusted hazard ratios (HRs), the inverse-variance method was similarly applied to pool the adjusted estimates. Since cancer is a rare adverse event, we also pooled the effect sizes using the fixed-effect Mantel–Haenszel method without continuity correction [23].
Sensitivity analyses excluding influential studies were conducted for outcomes with high between-study heterogeneity (I2 > 50%). A priori-defined subgroup analyses were conducted to explore heterogeneity according to the disease, follow-up period (< 5 years vs. ≥ 5 years), and studies including ≥ 500 patients.
Role of the funding source
This study received funding from the LEO Foundation [grant number: LF-FE-24-700053]. The funder had no say in designing the experiments or writing the manuscript.
Results
We identified 14,451 records through database search (PubMed = 2021, Web of Science = 2611, Embase = 8862, and Cochrane Library = 957); 10,388 nonduplicate articles were screened by title and abstract. Of these, 68 studies were included for full-text assessment. In total, 20 studies published from 2010 to 2023 were included for qualitative assessment and data extraction (Fig. 1 and supplementary Table 2). The follow-up period ranged from 1.5 to 6.8 years. All studies were cohort studies. Most studies were conducted in RA (n = 10 studies) or IBD (n = 8 studies) with one study reporting separate data for RA and IBD, and 3 studies reported pooled estimates for different IMIDs. The initial cancer types were mixed (any cancer) in 13 (65%) studies, breast cancer in 2 (10%) studies, basal cell carcinoma in 2 (10%) studies, NMSC in 1 (5%) study, melanoma in 1 (5%) study, and head and neck cancer in 1 (5%) study. In total, 15 studies reported the risk of recurrent cancers, 8 studies reported risk of new cancer, and 19 studies reported risk of new or recurrent cancers combined.
Fig. 1.
Preferred items for systematic review and meta-analysis flowchart
In total, 4736 patients (n = 15,646 PYs) were treated with biologics: 4462 patients were treated with TNFi (n = 20 studies), 218 with vedolizumab ([VDZ]; n = 4 studies), and 56 with IL-12/23i (n = 3 studies). Seventeen (85%) studies compared treatment with biologics to disease-modifying anti-rheumatic drugs (DMARDs), and 3 (15%) compared to patients without any immunosuppressive therapy. Overall, 7 (35%) studies either did not adjust for confounders or did not report the respective adjustments (Supplementary Table 3). Generally, the initial cancer stage and relapse risk was evenly distributed among patients treated with biologics compared to patients receiving no immunosuppressive or conventional therapy. Comparing biologics to each other, similar distributions were found (Supplementary Table 4). Cancer severity or stage was reported in 9 studies with 5125 patients treated with biologics (9 studies, n = 1272 patients), systemics (6 studies, n = 3119 patients), a combination of biologics and systemics (2 studies, n = 103 patients) or no therapy (3 studies, n = 631 patients). Of these, TNFi was reported in 9 studies (n = 1064 patients), VDZ in 3 studies (n = 167 patients), and ustekinumab (IL-12/23i) in 2 studies (n = 41 patients). For patients treated with TNFi, data on stage was reported for 1035 cancers, and 111 (10.7%) were in situ, 482 (46.5%) were invasive, 325 (31.4%) were stage I, 70 (6.8%) were stage II, 21 (2.0%) were stage III, 23 (2.2%) were stage IV, and 3 (0.3%) were unknown stage. For patients treated with VDZ, stage was reported on 208 cancers, and 133 (63.9%) were stage I, 43 (20.7%) were stage II, 24 (11.5%) were stage IIII, and 10 (4.8%) were stage IV. For patients treated with IL-12/23i, stages were reported for 108 cancers, and 38 (35.2%) were stage I, 34 (31.5%) were stage II, 35 (32.5%) were stage III, and 1 (0.9%) was stage IV. For patients receiving conventional systemics (6 studies, n = 3119 patients), data on stage was reported for 2844 cancers, and 378 (13.3%) were in situ, 382 (13.4%) were stage I, 52 (1.8%) were stage II, 16 (0.6%) were stage III, and 19 (0.7%) were stage IV, 1918 (67.4%) were invasive, and 78 (2.7%) were unknown. For patients receiving no therapy (3 studies, n = 631 patients), data on stage was reported for 343 cancers, and 212 (61.8%) were stage I, 45 (13.1%) were stage II, 32 (9.3%) were stage III, and 54 (15.7%) were stage IV (Supplementary Table 4). All studies were considered high quality based on NOS (Supplementary Table 5).
Risk of recurrent cancer in patients with IMIDs treated with biologics and a previous cancer
In total, 14 studies (n = 2056 patients) examined risk of recurrent cancer during treatment with TNFi (11 studies, n = 1889 patients) or VDZ (3 studies, n = 167 patients) among patients with IMIDs. It was only possible to assess the risk of any recurrent cancer (regardless of subtype) due to the insufficient number of studies on subtypes of cancer.
Cancer recurrence during TNFi treatment
For recurrence of any cancer during treatment with TNFi, the RR was 1.09 (95% CI 0.78–1.52, p = 0.56, I2 = 51.6%) based on 11 studies comparing 1889 patients treated with TNFi of whom 162 had cancers recurrence to 8752 patients treated with conventional or no therapy of whom 444 had cancer recurrence (Fig. 2; Supplementary Fig. 1). Sensitivity analyses excluding influential studies revealed similar results. In a sensitivity analysis removing the studies where the comparator received no therapy, similar results were found (RR: 1.20 [95% CI 0.85–1.69], p = 0.25) based on 8 studies comparing 1565 patients with previous cancer treated with TNFi and 124 recurrent cancers to 8121 patients treated with conventional therapies with 376 recurrent cancers. In subgroup analyses stratified by IMID, the RR of recurrence of any cancer in patients with IBD treated with TNFi was 1.00 (95% CI 0.43–2.33, p = 0.99) and for RA was 1.12 (95% CI 0.48–2.60, p = 0.68) compared to conventional or no therapy. The results were not altered when studies were stratified based on follow-up time (< 5 years vs ≥ 5 years).
Fig. 2.
Forest plot of crude risk ratio for recurrence of any cancer for biologics vs conventional therapies. AS, Ankylosing Spondylitis; DMARD, Disease Modifying Anti-Rheumatic Drugs; IBD, Inflammatory Bowel Disease; IS, Immunosuppressants; MTX, Methotrexate; PsA, Psoriatic Arthritis; PsO, Psoriasis; RA, Rheumatoid Arthritis; RR, Risk Ratio; SpA, Spondyloarthritis; TNFi, Tumor Necrosis Factor-α Inhibitor; VDZ, Vedolizumab
When pooling adjusted HRs from 3 studies, the HR of recurrence of any cancer was 0.97 (95% CI 0.39–2.42) based on 3 studies with 1121 TNFi treated patients who developed 75 recurrent cancers to 7668 patients treated with conventional therapies with 323 recurrent cancers. In sensitivity-analysis based on studies with ≥ 500 patients, similar results were found.
Exposure-adjusted analysis demonstrated an IRR of 0.83 (95% CI 0.58–1.20, p = 0.28) based on 10,347 PYs and 150 recurrent cancer in the TNFi group, and 34,236 PYs and 430 recurrent cancers in the comparator group. In a sensitivity analysis excluding studies with no therapy in the comparator group, similar results were found (Table 1).
Table 1.
Effect estimates for recurrent cancer according to biological target
| Previous cancer | Effect size | Analysis | Sub-group | Studies, n | Biologic group | Conventional therapy group | Fixed effects | Random effects | Heterogeneity | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Sample size/person-years, n | Cancer events, n | Sample size/person-years, n | Cancer events, n | Effect size [95% CI] | p value | Effect size [95% CI] | p value | I2 [95% CI], % | Egger’s test | |||||
| Tumor Necrosis Factor-α inhibitor | ||||||||||||||
| Mixed | RR | Main | 11 | 1889 | 162 | 8752 | 444 | 1.05 [0.88; 1.25] | 0.6 | 1.09 [0.78; 1.52] | 0.56 | 51.6% [3.9%; 75.7%] | 0.79 | |
| Mixed | RR | Sensitivity | Influence analysis | 7 | 775 | 43 | 5070 | 144 | 1.25 [0.87; 1.79] | 0.22 | 1.33 [0.90; 1.96] | 0.12 | 0.0% [0.0%; 70.8%] | – |
| Mixed | RR | Sensitivity | Remove no therapy | 8 | 1565 | 124 | 8121 | 376 | 1.18 [0.97; 1.44] | 0.10 | 1.20 [0.85; 1.69] | 0.25 | 41.9% [0.0%; 74.3%] | – |
| Mixed | RR | Subgroup | Disease | – | – | – | – | Test for subgroup differences: | 0.10 | – | 0.76 | – | – | |
| Mixed | RR | IBD | 5 | 431 | 50 | 798 | 80 | 0.82 [0.60; 1.14] | 0.23 | 1.00 [0.43; 2.33] | 0.99 | 42% [0%; 79%] | – | |
| Mixed | RR | RA | 4 | 988 | 94 | 3554 | 257 | 1.14 [0.91; 1.43] | 0.27 | 1.12 [0.48; 2.60] | 0.68 | 73% [23%; 90%] | – | |
| Mixed | RR | Subgroup | Follow-up time | – | – | – | – | Test for subgroup differences: | 0.15 | – | 0.04 | – | – | |
| Mixed | RR | < 5 years | 3 | 154 | 18 | 400 | 23 | 1.52 [0.81; 2.84] | 0.19 | 1.52 [0.90; 2.57] | 0.07 | 0% [0%; 90%] | – | |
| Mixed | RR | > 5 years | 7 | 1515 | 123 | 7176 | 351 | 0.94 [0.77; 1.14] | 0.52 | 0.94 [0.58; 1.52] | 0.75 | 60% [7%; 82%] | – | |
| Mixed | IRR | Main | 9 | 10,347 | 150 | 34,236 | 430 | 0.83 [0.69; 1.01] | 0.06 | 0.83 [0.58; 1.20] | 0.28 | 42.3% [0.0%; 73.4%] | – | |
| Mixed | IRR | Sensitivity | Remove no therapy | 6 | 7644 | 112 | 29,628 | 362 | 0.89 [0.72; 1.11] | 0.32 | 0.88 [0.59; 1.31] | 0.44 | 38.4% [0.0%; 75.5%] | – |
| Mixed | HR | Main | Adjusted estimates | 3 | 1121 | 75 | 7668 | 323 | – | – | 0.97 [0.39; 2.42] | 0.89 | 38.6% [0.0%; 80.8%] | – |
| Mixed | HR | Sensitivity | > 500 patients | 3 | 1121 | 75 | 7668 | 323 | – | – | 0.97 0,39; 2.42] | 0.89 | 38.6% [0.0%; 80.8%] | – |
| Anti-α4β7-integrin | ||||||||||||||
| Mixed | RR | Main | 3 | 167 | 14 | 631 | 68 | 0.51 [0.30; 0.87] | 0.01 | 0.70 [0.07; 7.61] | 0.59 | 60.5% [0.0%; 88.7%] | – | |
| Mixed | IRR | Main | 3 | 1396 | 14 | 4608 | 68 | 0.45 [0.26; 0.80] | 0.006 | 0.63 [0.05; 7.83] | 0.50 | 59.7% [0.0%; 88.5%] | – | |
p values < 0.05 marked bold
NMSC, Non-Melanoma Skin Cancer; RR, Risk Ratio; HR, Hazard Ratio; IRR, Incidence Rate Ratio; IL, Interleukin
Cancer recurrence during VDZ treatment
For recurrence of any cancer during treatment with VDZ, the RR was 0.70 (95% CI 0.07–7.61, p = 0.59) based on 3 studies comparing 167 patients treated with VDZ with 14 recurrent cancers to 631 patients with no therapy and 68 recurrent cancers (Fig. 2). The IRR was 0.63 (95% CI 0.63–7.83, p = 0.50) (Table 1).
Risk of new cancer in patients with IMIDs treated with biologics and a previous cancer
In total, 10 studies (n = 175 patients) examined the risk of developing a new cancer during treatment with TNFi (7 studies, n = 1108 patients) or VDZ (3 studies, n = 167 patients) in patients with IMIDs and previous cancer. Meta-analyses could only be conducted for the development of NMSC or all subtypes.
New cancer during TNFi treatment
For any new cancer during treatment with TNFi, the RR was 1.02 (95% CI 0.70–1.50, p = 0.88) based on seven studies comparing 1108 patients with previous cancer treated with TNFi with 137 new cancers to 5285 patients with previous cancer treated with conventional therapies or no therapy with 595 new cancers. In sensitivity analyses excluding studies with no therapy in the comparator group, similar results were found yielding a RR of 0.97 (95% CI 0.33–2.84, p = 0.92) based on 4 studies comparing 784 patients treated with TNFi developing 137 new cancers to 5285 patients treated with conventional therapies developing 595 new cancers.
The IRR was 0.84 (95% CI 0.63–1.32, p = 0.36) comparing 5955 PYs of TNFi and 134 new cancers to 22,222 PYs of conventional or no therapy and 588 new cancers. In sensitivity analyses excluding 3 studies with no therapy in the comparator group, similar results were found.
For new NMSC during treatment with TNFi, the IRR was 0.91 (95% CI 0.59–1.41, p = 0.44) based on 3 studies comparing 3165 patients with a history of NMSC treated with TNFi of whom 165 had new cancer to 14,041 patients with a history of NMSC treated with conventional or no therapies of whom 658 had new cancer. Pooling adjusted HRs (n = 3 studies) revealed a HR of 1.10 (95% CI 0.42–2.92, p = 0.70) for new NMSC during treatment with TNFi.
New cancer during VDZ treatment
For any new cancer during treatment with VDZ, the RR was 0.44 (95% CI 0.33–0.60, p = 0.0073) based on 3 studies comparing 167 VDZ treated patients developing 11 new cancers to 631 patients treated with conventional or no therapies developing 91 new cancers.
In exposure-adjusted analysis, the IRR of new cancer was 0.39 (95% CI 0.25–0.62, p = 0.0122) based on 3 studies comparing 1396 PYs (n = 11 new cancers) for VDZ to 4608 PYs (n = 91 new cancers) for conventional or no therapy (Table 2).
Table 2.
Effect estimates for new cancer according to biological target
| Previous cancer | Effect size | Analysis | Sub-group | Studies, n | Biologic group | Conventional therapy group | Fixed effects | Random effects | Heterogeneity | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Sample size/person-years, n | Cancer events, n | Sample size/person-years, n | Cancer events, n | Effect size [95% CI] | p value | Effect size [95% CI] | p value | I2 [95% CI], % | Egger’s test | |||||
| Tumor Necrosis Factor-α inhibitor | ||||||||||||||
| Mixed | RR | Main | 7 | 1108 | 137 | 5285 | 595 | 1.07 [0.89; 1.28] | 0.50 | 1.02 [0.70; 1.50] | 0.88 | 35.2% [0.0%; 72.6%] | – | |
| Mixed | RR | Sensitivity | Remove no therapy | 4 | 784 | 95 | 4654 | 504 | 1.12 [0.90; 1.40] | 0.30 | 0.97 [0.33; 2.84] | 0.92 | 55.9% [0.0%; 85.4%] | – |
| Mixed | IRR | Main | 6 | 6955 | 134 | 22,222 | 588 | 0.84 [0.68; 1.02] | 0.08 | 0.84 [0.53; 1.32] | 0.36 | 33.7% [0.0%; 73.3%] | – | |
| Mixed | IRR | Sensitivity | Remove no therapy | 3 | 4252 | 92 | 17,614 | 497 | 0.83 [0.65; 1.05] | 0.12 | 0.78 [0.16; 3.68] | 0.56 | 52.7% [0.0%; 86.4%] | – |
| NMSC | IRR | Main | 3 | 3165 | 165 | 14,041 | 658 | 0.92 [0.78; 1.10] | 0.35 | 0.91 [0.59; 1.41] | 0.44 | 19.0% [0.0%; 91.6%] | – | |
| NMSC | HR | Main | Adjusted estimates | 3 | – | – | – | – | – | – | 1.10 [0.42; 2.92] | 0.70 | 68.7% [0.0%; 90.9%] | – |
| Anti-α4β7-integrin | ||||||||||||||
| Mixed | RR | Main | 3 | 167 | 11 | 631 | 91 | 0.44 [0.24; 0.81] | 0.0087 | 0.44 [0.33; 0.60] | 0.0073 | 0.0% [0.0%; 89.6%] | – | |
| Mixed | IRR | Main | 3 | 1396 | 11 | 4608 | 91 | 0.39 [0.21; 0.74] | 0.0039 | 0.39 [0.25; 0.62] | 0.0122 | 0.0% [0.0%; 89.6%] | – | |
p values < 0.05 marked bold
NMSC, Non-Melanoma Skin Cancer; RR, Risk Ratio; HR, Hazard Ratio; IRR, Incidence Rate Ratio; IL, Interleukin
Risk of new or recurrent cancer in patients with IMIDs treated with biologics and a previous cancer
In total, 19 studies (n = 2341 patients) examined the risk of developing new or recurrent cancer combined for TNFi (13 studies, n = 2067 patients), VDZ (4 studies, n = 218 patients), and IL-12/23i (3 studies, n = 56 patients) among patients with IMIDs and a previous cancer. Meta-analyses could be conducted for new or recurrent cancers combined for any cancer or NMSC.
New or recurrent cancer during TNFi treatment
For new or recurrence of any cancer during TNFi treatment, the RR was 1.01 (95% CI 0.77–1.32, p = 0.95) based on 13 studies comparing 2067 patients treated with TNFi developing 308 new or recurrent cancers to 8855 patients treated with conventional or no therapies developing 1049 new or recurrent cancers.
In sensitivity analysis, similar results were observed when removing influential studies (3 studies, n = 861 patients) and studies with no active therapy in the control group (3 studies, n = 324 patients). In subgroup analyses, the RR of new or recurrence of any cancer following TNFi for IBD was 0.77 (95% CI 0.45–1.31, p = 0.25) compared to conventional or no therapies based on 6 studies including 509 patients (n = 102 cancers) for TNFi, and 824 patients (n = 200 cancers) for conventional or no therapy. For RA during TNFi, the RR of new or recurrence of any cancer was 1.19 (95% CI 0.75–1.89, p = 0.35) based on 5 studies with 1088 RA patients treated with TNFi developing 128 new or recurrent cancers to 3631 patients treated with conventional or no therapies developing 277 new or recurrent cancers. Subgroup analyses based on follow-up revealed similar results (Table 3).
Table 3.
Effect estimates for new or recurrent cancer according to biological target
| Previous cancer | Effect size | Analysis | Sub-group | Studies, n | Biologic group | Conventional therapy group | Fixed effects | Random effects | Heterogeneity | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Sample size/person-years, n | Cancer events, n | Sample size/person-years, n | Cancer events, n | Effect size [95% CI] | p value | Effect size [95% CI] | p value | I2 [95% CI], % | Egger’s test | |||||
| Tumor Necrosis Factor-α inhibitor | ||||||||||||||
| Mixed | RR | Main | 13 | 2067 | 308 | 8855 | 1049 | 1.04 [0.92; 1.17] | 0.52 | 1.01 [0.77; 1.32] | 0.95 | 60.0% [26.4%; 78.2%] | 0.73 | |
| Mixed | RR | Sensitivity | Influence analysis | 10 | 1206 | 122 | 4185 | 362 | 1.12 [0.92; 1.36] | 0.26 | 1.06 [0.71; 1.56] | 0.75 | 49.5% [0.0%; 75.6%] | – |
| Mixed | RR | Sensitivity | Remove no therapy | 10 | 1743 | 228 | 8224 | 888 | 1.13 [0.98; 1.30] | 0.087 | 1.08 [0.79; 1.48] | 0.60 | 52.9% [3.5%; 77.0%] | – |
| Mixed | RR | Subgroup | Disease | – | – | – | – | – | Test for subgroup differences: | 0.0091 | – | 0.098 | – | – |
| Mixed | RR | IBD | 6 | 509 | 102 | 824 | 200 | 0.77 [0.63; 0.95] | 0.014 | 0.77 [0.45; 1.31] | 0.25 | 55% [0%; 82%] | – | |
| Mixed | RR | RA | 5 | 1088 | 128 | 3631 | 277 | 1.14 [0.93; 1.39] | 0.22 | 1.19 [0.75; 1.89] | 0.35 | 56% [0%; 84%] | – | |
| Mixed | RR | Subgroup | Follow-up time | – | – | – | – | – | Test for subgroup differences: | 0.97 | – | 0.77 | – | – |
| Mixed | RR | < 5 years | 4 | 254 | 31 | 477 | 84 | 1.03 [0.70; 1.51] | 0.89 | 1.08 [0.36; 3.23] | 0.82 | 57% [ 0%; 86%] | – | |
| Mixed | RR | > 5 years | 7 | 1515 | 250 | 7176 | 890 | 1.02 [0.89; 1.16] | 0.79 | 0.97 [0.71; 1.33] | 0.84 | 64% [18%; 84%] | – | |
| Mixed | IRR | Main | 10 | 10,580 | 287 | 34,405 | 1023 | 0.83 [0.72; 0.95] | 0.0086 | 0.83 [0.62; 1.11] | 0.17 | 48.1% [0.0%; 74.9%] | 0.90 | |
| Mixed | IRR | Sensitivity | Remove no therapy | 7 | 7877 | 207 | 29,797 | 862 | 0.86 [0.74; 1.01] | 0.075 | 0.86 [0.65; 1.14] | 0.23 | 28.2% [0.0%; 69.0%] | – |
| Mixed | HR | Main | 8 | – | – | – | – | – | – | 0.84 [0.64; 1.10] | 0.16 | 22.9% [0.0%; 64.6%] | – | |
| Mixed | HR | Sensitivity | Remove no therapy | 5 | – | – | – | – | – | – | 0.88 [0.59; 1.32] | 0.43 | 44.2% [0.0%; 79.5%] | – |
| Mixed | HR | Sensitivity | > 500 patients | 3 | 1121 | 135 | 7668 | 788 | – | – | 0.96 [0.59; 1.55] | 0.72 | 5.6% [0.0%; 90.2%] | – |
| NMSC | IRR | Main | 3 | 3165 | 176 | 14,041 | 742 | 0.92 [0.78; 1.08] | 0.31 | 0.91 [0.60; 1.37] | 0.42 | 17.5% [0.0%; 91.4%] | – | |
| NMSC | HR | Main | 3 | – | – | – | – | – | – | 1.08 [0.39; 3.00] | 0.78 | 76.6% [23.3%; 92.8%] | – | |
| Anti-α4β7-integrin | ||||||||||||||
| Mixed | RR | Main | 4 | 218 | 28 | 657 | 166 | 0.46 [0.32; 0.66] | < 0.0001 | 0.47 [0.27; 0.84] | 0.026 | 0.0% [0.0%; 84.7%] | – | |
| Mixed | IRR | Main | 3 | 1396 | 25 | 4608 | 161 | 0.42 [0.27; 0.64] | < 0.0001 | 0.44 [0.13; 1.57] | 0.10 | 24.3% [0.0%; 92.1%] | – | |
| Mixed | HR | Main | 3 | – | – | – | – | – | – | 0.56 [0.08; 4.11] | 0.33 | 20.2% [0.0%; 91.7%] | – | |
| IL-12/23 inhibitor | ||||||||||||||
| Mixed | RR | Main | 3 | 56 | 7 | 474 | 88 | 0.71 [0.34; 1.49] | 0.37 | 0.71 [0.63; 0.81] | 0.0072 | 0.0% [0.0%; 89.6%] | – | |
p values < 0.05 marked bold
NMSC, Non-Melanoma Skin Cancer; RR, Risk Ratio; HR, Hazard Ratio; IRR, Incidence Rate Ratio; IBD, Inflammatory Bowel Disease; RA, Rheumatoid Arthritis; IL, Interleukin
For any new or recurrent cancer during treatment with TNFi, the IRR was 0.83 (95% CI 0.62–1.11, p = 0.17) based on 10,580 PYs and 287 new or recurrent cancers for TNFi and 34,405 PYs and 1023 new or recurrent cancers in the control group. When removing studies with no active therapy in the control group, similar results were found.
For new or recurrence of any cancer during treatment with TNFi using the reported HRs, the HR was 0.84 (95% CI 0.64–1.10, p = 0.16) compared to conventional or no therapy based on 8 studies. In sensitivity analysis excluding studies with no therapy in the control group, similar results were found. In sensitivity analysis of 3 studies including ≥ 500 patients and pooling the reported adjusted HRs, the HR was 0.96 (95% CI 0.59–1.55, p = 0.72) based on 3 studies comparing 1121 patients with 135 new or recurrent cancers to 7668 controls with 742 new or recurrent cancers.
For new or recurrent NMSC following treatment with TNFi, the IRR was 0.91 (95% CI 0.60–1.37, p = 0.42) based on 3 studies comparing 3165 PYs of TNFi and 176 new or recurrent cancers to 14,041 PYs of conventional or no therapies developing 742 new or recurrent cancers (Table 3).
New or recurrent cancer during VDZ treatment
For new or recurrence of any cancer during treatment with VDZ, the RR was 0.47 (95% CI 0.27–0.84, p = 0.026) based on 4 studies including 218 patients treated with VDZ developing 28 new or recurrent cancers, and 657 patients treated with conventional or no therapies developing 166 new or recurrent cancers.
In exposure-adjusted analysis of VDZ, the IRR was 0.44 (95% CI 0.13–1.57, p = 0.10) based on 3 studies comparing 1396 PYs of VDZ treatment developing 25 new or recurrent cancers to 4608 PYs of conventional or no therapies developing 161 new or recurrent cancers (Table 3).
New or recurrent cancer during IL-12/23i treatment
For any new or recurrent cancer during treatment with IL-12/23i, the RR was 0.71 (95% CI 0.63–0.81, p = 0.0072) based on 3 studies with 56 IL-12/23i treated patients developing 7 cancers to 474 patients undergoing conventional or no therapy developing 88 cancers (Table 3).
Discussion
In this systematic review and meta-analysis including 20 studies and 4736 patients treated with biologics, treatment with TNFi among patients with IMIDs and a previous cancer did not reveal a significantly increased risk of new or recurrent cancers when compared with conventional systemics or no treatment. However, VDZ was associated with a lower risk of both new or recurrent cancer compared with conventional systemics or no therapy.
The potentially reduced carcinomatous surveillance properties of biologics and relative contraindications of use of biologics in patients with active or previous cancers has led to hesitancy of prescribing biologics in patients with active cancer or a history of cancer. Still, patients with severe disease need treatment for disease control which in many cases would be best treated with biologics. While no RCTs will be conducted due to ethical considerations, real-world studies are highly important in assessing the safety of such treatment. Recently, in an unpublished systematic review we did not find an increased risk of cancer following treatment with biologics in RCTs or observational studies. Similarly, this systematic review and meta-analyses found no increased risk of new or recurrent cancer for treatment with TNFi, IL-12/-23i, or VDZ among patients with previous cancers. The finding of no increased risk of new or recurrent cancer or even protective properties of biologics in patients with a history of cancer might be attributed to channeling bias wherein patients with higher risk of new cancers or cancer recurrences are not prescribed biologics. Indeed, patients prescribed biologics were slightly younger than their controls. However, while the proportion of patients with high-stage initial cancers with high relapse risk was evenly distributed between patients treated with biologics compared to patients receiving conventional or no immunosuppressive therapy [24, 25], higher proportions of initial aggressive cancers were found among patients receiving TNFi or VDZ. Despite these differences, in studies adjusting for cancer recurrence risk, no increased risk of recurrent cancers was found [24, 25]. Additionally, studies accounting for age found similar results [26–28] and patients treated with biologics had in many cases more severe disease [14, 16] which is by itself associated with increased risk of cancer [29, 30]. Further, one study found that patients prescribed TNFi either alone or in combination with conventional systemics had cancers with higher risk recurrence potential and were more often smokers but had numerically—not statistically—lower risk of cancer recurrence compared to controls treated with conventional therapies [14]. Importantly, most meta-analyses could only be conducted on cancer as a combined group and the risk might differ according to cancer type. Indeed, one study found a numerically higher occurrence of melanoma relapses in patients treated with TNFi compared with conventional systemics (3 vs 0) [16], and one study found MTX and TNFi to be associated with an increased risk of NMSC in RA [31].
Although the findings are reassuring in that treatment with the TNFi, IL-12/-23i, and VDZ classes does not appear to lead to an increased risk of new cancers or cancer recurrences among patients with a history of cancer, certain limitations should be considered when interpreting the results. Differences in study design, patient populations, and follow-up periods introduce heterogeneity, thus potentially altering the pooled results. Most studies were conducted in patients with IBD or RA and the results might not be generalizable to other diseases with these drugs. Moreover, we were unable to consider the time from cancer diagnosis to initiation of biologics. Further, for most outcomes only the combined outcome measure “any cancer” were possible to assess and the findings might not be applicable to all cancer types. Lack of adjustment for confounders may influence the results and promote inadequate effect estimates, though meta-analysis on adjusted estimates did not reveal significant associations. Lastly, we cannot refute that the protective properties of the biologics are partly attributed to channeling bias. However, differences in the biologics and controls group were only marginally different.
Conclusion
In this systematic review and meta-analysis, treatment with TNFi, IL-12/23i, or VDZ do not appear to lead to increased risk of new or recurrent cancer among patients with IMIDs with a previous cancer when compared with conventional systemics or no treatment. However, further studies are needed especially on newer biologics and decision on initiation of biologics in patients with cancers should be based on an individual assessment.
Supplementary Information
Below is the link to the electronic supplementary material.
Author contributions
DI, NL and MJ screened articles based on title and abstract as well as full-text. DI, NL and MJ extracted data from eligible studies. DI and CS prepared data for and conducted meta-analysis. DI, LS, and NL initiated the study. JS, LS, and NL supervised and gave opinions during the process. JS provided opinions and viewpoint on specifically the articles involving IBD. All authors reviewed the manuscript.
Funding
Open access funding provided by Copenhagen University. This study received funding from the LEO Foundation [grant number: LF-FE-24-700053].
Data availability
Data is provided within the manuscript or supplementary information files.
Declarations
Conflict of interest
Mr. Isufi, Dr. Schwarz, and Dr. Jensen report no conflict of interest. Dr. Loft has been a paid speaker for Eli Lilly, Janssen Cilag, and Sandoz. Dr. Seidelin has received research grants from Takeda, Janssen, the Danish Research Council, and the Capital Region Denmark; and has served as the national coordinator of studies for AbbVie, Arena Pharmaceuticals, Ely Lilly, and Boehringer Ingelheim. Dr. Skov has been a paid speaker for AbbVie, Eli Lilly, Novartis, Pfizer, and LEO Pharma, and has been a consultant or has served on Advisory Boards with AbbVie, Janssen Cilag, Novartis, Eli Lilly, Boehringer Ingelheim, LEO Pharma, Takeda, Stada, UCB, Almirall, Bristol-Myers Squibb, and Sanofi and has received research and educational grants/funding from LEO Foundation, Kgl Hofbundtmager Aage Bang Foundation, Sanofi, Bristol-Myers Squibb, Janssen Cilag, and Almirall.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Sandborn WJ, Baert F, Danese S, et al. Efficacy and safety of vedolizumab subcutaneous formulation in a randomized trial of patients with ulcerative colitis. Gastroenterology. 2020;158(3):562-572.e12. 10.1053/J.GASTRO.2019.08.027. [DOI] [PubMed] [Google Scholar]
- 2.Nielsen OH, Ainsworth MA. Tumor necrosis factor inhibitors for inflammatory bowel disease. N Engl J Med. 2013;369(8):754–62. 10.1056/NEJMCT1209614. [DOI] [PubMed] [Google Scholar]
- 3.Geale K, Lindberg I, Paulsson EC, et al. Persistence of biologic treatments in psoriatic arthritis: a population-based study in Sweden. Rheumatol Adv Pract. 2020. 10.1093/RAP/RKAA070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Honap S, Meade S, Ibraheim H, Irving PM, Jones MP, Samaan MA. Effectiveness and safety of ustekinumab in inflammatory bowel disease: a systematic review and meta-analysis. Dig Dis Sci. 2022;67(3):1018–35. 10.1007/S10620-021-06932-4. [DOI] [PubMed] [Google Scholar]
- 5.Thatiparthi A, Martin A, Liu J, Egeberg A, Wu JJ. Biologic treatment algorithms for moderate-to-severe psoriasis with comorbid conditions and special populations: a review. Am J Clin Dermatol. 2021;22(4):425–42. 10.1007/S40257-021-00603-W. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Li P, Zheng Y, Chen X. Drugs for autoimmune inflammatory diseases: from small molecule compounds to anti-TNF biologics. Front Pharmacol. 2017. 10.3389/FPHAR.2017.00460. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Stidham RW, Lee TCH, Higgins PDR, et al. Systematic review with network meta-analysis: the efficacy of anti-tumour necrosis factor-alpha agents for the treatment of ulcerative colitis. Aliment Pharmacol Ther. 2014;39(7):660–71. 10.1111/APT.12644. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Stidham RW, Lee TCH, Higgins PDR, et al. Systematic review with network meta-analysis: the efficacy of anti-TNF agents for the treatment of Crohn’s disease. Aliment Pharmacol Ther. 2014;39(12):1349–62. 10.1111/APT.12749. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Sbidian E, Chaimani A, Guelimi R, et al. Systemic pharmacological treatments for chronic plaque psoriasis: a network meta-analysis. Cochrane Database Syst Rev. 2023. 10.1002/14651858.CD011535.PUB6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Aaltonen KJ, Virkki LM, Malmivaara A, Konttinen YT, Nordström DC, Blom M. Systematic review and meta-analysis of the efficacy and safety of existing TNF blocking agents in treatment of rheumatoid arthritis. PLoS ONE. 2012;7(1): e30275. 10.1371/JOURNAL.PONE.0030275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Wolfe F, Michaud K. Lymphoma in rheumatoid arthritis: the effect of methotrexate and anti–tumor necrosis factor therapy in 18,572 patients. Arthritis Rheum. 2004;50(6):1740–51. 10.1002/ART.20311. [DOI] [PubMed] [Google Scholar]
- 12.Bongartz T, Sutton AJ, Sweeting MJ, Buchan I, Matteson EL, Montori V. Anti-TNF antibody therapy in rheumatoid arthritis and the risk of serious infections and malignancies: systematic review and meta-analysis of rare harmful effects in randomized controlled trials. JAMA. 2006;295(19):2275–85. 10.1001/JAMA.295.19.2275. [DOI] [PubMed] [Google Scholar]
- 13.Bernheim O, Colombel JF, Ullman TA, Laharie D, Beaugerie L, Itzkowitz SH. The management of immunosuppression in patients with inflammatory bowel disease and cancer. Gut. 2013;62(11):1523–8. 10.1136/GUTJNL-2013-305300. [DOI] [PubMed] [Google Scholar]
- 14.Axelrad J, Bernheim O, Colombel JF, et al. Risk of new or recurrent cancer in patients with inflammatory bowel disease and previous cancer exposed to immunosuppressive and anti-tumor necrosis factor agents. Clin Gastroenterol Hepatol. 2016;14(1):58–64. 10.1016/J.CGH.2015.07.037. [DOI] [PubMed] [Google Scholar]
- 15.Beaugerie L, Carrat F, Colombel JF, et al. Risk of new or recurrent cancer under immunosuppressive therapy in patients with IBD and previous cancer. Gut. 2014;63(9):1416–23. 10.1136/GUTJNL-2013-305763. [DOI] [PubMed] [Google Scholar]
- 16.Dixon WG, Watson KD, Lunt M, et al. Influence of anti–tumor necrosis factor therapy on cancer incidence in patients with rheumatoid arthritis who have had a prior malignancy: results from the British Society for rheumatology biologics register. Arthritis Care Res (Hoboken). 2010;62(6):755–63. 10.1002/ACR.20129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Biancone L, Orlando A, Kohn A, et al. Infliximab and newly diagnosed neoplasia in Crohn’s disease: a multicentre matched pair study. Gut. 2006;55(2):228–33. 10.1136/GUT.2005.075937. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med. 2009;151(4):264–9. 10.7326/0003-4819-151-4-200908180-00135/SUPPL_FILE/PRISMA_FIGURE-S1.DOC. [DOI] [PubMed] [Google Scholar]
- 19.Ouzzani M, Hammady H, Fedorowicz Z, Elmagarmid A. Rayyan—a web and mobile app for systematic reviews. Syst Rev. 2016. 10.1186/S13643-016-0384-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. https://www.R-project.org/. Accessed 3 Sept 2024.
- 21.Harrer M, Cuijpers P, Furukawa TA, Ebert DD. Doing meta-analysis with R: a hands-on guide. 1st ed. Chapman & Hall/CRC Press; 2021. [Google Scholar]
- 22.DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177–88. 10.1016/0197-2456(86)90046-2. [DOI] [PubMed] [Google Scholar]
- 23.Deeks J, Higgins J, Altman D, et al., editors. Chapter 10: Analysing data and undertaking meta-analyses | Cochrane Training. In: Cochrane; 2023. www.training.cochrane.org/handbook. Accessed 3 Sept 2024.
- 24.Hasan B, Tandon KS, Miret R, et al. Ustekinumab does not increase risk of new or recurrent cancer in inflammatory bowel disease patients with prior malignancy. J Gastroenterol Hepatol. 2022;37(6):1016–21. 10.1111/JGH.15806. [DOI] [PubMed] [Google Scholar]
- 25.Hong SJ, Zenger C, Pecoriello J, et al. Ustekinumab and vedolizumab are not associated with subsequent cancer in IBD patients with prior malignancy. Inflamm Bowel Dis. 2022;28(12):1826. 10.1093/IBD/IZAC035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Silva-Fernández L, Lunt M, Kearsley-Fleet L, et al. The incidence of cancer in patients with rheumatoid arthritis and a prior malignancy who receive TNF inhibitors or rituximab: results from the British Society for Rheumatology Biologics Register-Rheumatoid Arthritis. Rheumatology (Oxford). 2016;55(11):2033. 10.1093/RHEUMATOLOGY/KEW314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Raaschou P, Frisell T, Askling J. TNF inhibitor therapy and risk of breast cancer recurrence in patients with rheumatoid arthritis: a nationwide cohort study. Ann Rheum Dis. 2015;74(12):2137–43. 10.1136/ANNRHEUMDIS-2014-205745. [DOI] [PubMed] [Google Scholar]
- 28.Mercer LK, Green AC, Galloway JB, et al. The influence of anti-TNF therapy upon incidence of keratinocyte skin cancer in patients with rheumatoid arthritis: longitudinal results from the British Society for Rheumatology Biologics Register. Ann Rheum Dis. 2012;71(6):869–74. 10.1136/ANNRHEUMDIS-2011-200622. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Vaengebjerg S, Skov L, Egeberg A, Loft ND. Prevalence, incidence, and risk of cancer in patients with psoriasis and psoriatic arthritis: a systematic review and meta-analysis. JAMA Dermatol. 2020;156(4):421–9. 10.1001/JAMADERMATOL.2020.0024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Loft ND, Vaengebjerg S, Skov L. Cancer risk in patients with psoriasis: Should we be paying more attention? Expert Rev Clin Immunol. 2020;16(5):479–92. 10.1080/1744666X.2020.1754194. [DOI] [PubMed] [Google Scholar]
- 31.Scott FI, Mamtani R, Brensinger CM, et al. Risk of nonmelanoma skin cancer associated with the use of immunosuppressant and biologic agents in patients with a history of autoimmune disease and nonmelanoma skin cancer. JAMA Dermatol. 2016;152(2):164–72. 10.1001/JAMADERMATOL.2015.3029. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data is provided within the manuscript or supplementary information files.