Switching Among Biologic Therapies in Rheumatoid Arthritis
Increasing healthcare expenditures have led governments and their agencies across the globe to seek ways to limit healthcare spending and develop strategies to maximize health gains from fixed expenditures. The cost of RA has increased substantially with the introduction of effective biologic therapies, which target proinflammatory cytokines such as TNF and IL-6, and lymphocytes such as B or T cells. With annual costs that are considerably higher than those of older RA drug therapies, biologic therapies have gained particular attention from healthcare decision-makers who must evaluate whether the additional costs of biologics are worth the added clinical benefits.
One approach taken by payers to limit expenditures on biologic therapy in RA is the selection of the lowest price drug as the first-choice product. Relative prices of individual biologic agents vary by country, and acquisition costs fluctuate owing to discounts and patient access schemes. To address ongoing fiscal pressures, payers may suggest that less expensive agents be substituted for current treatments even in patients with an adequate response. Another example of nonclinical switching involves the substitution of a subcutaneously administered biologic agent for an intravenously administered agent (or vice versa) in patients who have responded successfully to the initial therapy, owing to patient preference or economic pressure.
Switching biologic therapies in patients with RA for economic or other nonclinical purposes is based, at least in part, on the assumption that the agents are equivalent in terms of safety and effectiveness. In the following sections, the evidence will be reviewed from studies in which biologic therapies have been switched for nonclinical reasons and the assumption of biologic equivalence will be examined.
Evidence to support switching among biologic therapies in patients with a therapeutic response to initial treatment is very limited. Based on a literature search, three studies were identified that included patients with RA who were established on a TNF antagonist (infliximab in all three studies) before switching to another anti-TNF agent for nonclinical reasons (Table 3). Laas et al. reported outcomes in 49 patients who were transitioned from infliximab to etanercept therapy in an observational study conducted in Finland. In the subgroup of 23 patients who had demonstrated a good response to infliximab and were switched for nonclinical reasons (e.g., preference for outpatient treatment), a nonsignificant increase in DAS28 of 0.4 was observed after the switch (Figure 2); patients switched owing to infliximab failure and adverse events showed statistically significant decreases in DAS28 of -1.2 (p = 0.023) and -1.3 (p = 0.048), respectively.
(Enlarge Image)
Figure 2.
Changes in 28-joint Disease Activity Score during infliximab and etanercept treatment by reason for switching therapy. The horizontal dashed line represents the cut-off level for DAS28 remission (DAS28 <2.6).DAS28: 28-joint Disease Activity Score.Reproduced from [35] with permission from Springer Science+Business Media.
In a 32-week, open-label study conducted in Ireland, Walsh et al. evaluated 19 patients with RA who had demonstrated a clinical response to hospital-administered infliximab (defined as DAS28 improvement >1.2 within 3 months of treatment) after they were switched to home-administered adalimumab for economic reasons. Mean DAS28 improved from 4.1 at baseline (i.e., at penultimate infliximab infusion) and 3.9 at the time of the switch to 3.0 after 16 weeks of adalimumab therapy (p < 0.005 vs both baseline and switch; Figure 3). No significant changes were observed in tender or swollen joint counts or in functional/quality-of-life assessments, including HAQ scores. The latter finding may be an important consideration for health economics, as cost–effectiveness models are driven by changes in HAQ, which was reduced from the time of switching to the final visit from 1.1 to 1.0 (Figure 3). Although all patients met response criteria for enrollment in the study, at the time that the therapies were switched, nearly three-quarters of patients had moderate (DAS28 3.2–5.0) or high (DAS28 >5.0) disease activity; therefore, they could be considered inadequate responders to TNF-antagonist therapy.
(Enlarge Image)
Figure 3.
Changes in clinical and functional/quality-of-life assessments before, during and after patients were switched from infliximab to adalimumab therapy. (A) DAS28 (mean ± SEM); (B) HAQ Disability Index (mean ± SEM).*p < 0.005 versus baseline and time of switch.DAS28: 28-Joint Disease Activity Score; HAQ: Health Assessment Questionnaire;SEM: Standard error of the mean.Adapted from [36] with permission from Oxford University Press.
In a 6-month, open-label study conducted in the UK, 14 out of 17 patients with stable RA receiving infliximab (3 mg/kg every 8 weeks), given a choice to change to treatment with adalimumab, agreed to switch. Adalimumab was started 8 weeks after the last infliximab infusion. Median DAS28 at 6 months was 3.6, which was similar to the preswitch score of 3.7 (Table 4). Of the 14 switchers, 11 patients demonstrated a good clinical response and continued receiving adalimumab long term. Among the three patients who did not continue adalimumab, two patients worsened after the switch and reverted to treatment with infliximab; one patient discontinued use of adalimumab, with a plan to restart infliximab, but was being managed with methotrexate and corticosteroids at last follow-up. As in the previously described Irish study, not all of the patients enrolled in this study would be considered infliximab responders, as their DAS28 scores were greater than 5.1, and outcomes were not analyzed by level of response before switching.
Justification for switching biologic agents in patients with RA for nonclinical purposes depends in part on these agents having comparable efficacy and safety profiles. As summarized in the following sections, salient differences between these agents have been detected that highlight their unique pharmacologic profiles ( Table 1 ).
Mechanism of Action/Other Properties Agents representing the major biologic classes used in the treatment of RA demonstrate differences in mode of action. Even the five currently approved TNF antagonists differ substantially in their mechanisms of action and kinetics, despite their shared inhibition of the cytokine TNF. In particular, etanercept, a fusion protein of human p75 soluble TNF receptor and human IgG1, appears to bind differently to cell-bound TNF than the other agents, which are all monoclonal antibodies or fragments of antibodies. Unlike the monoclonal antibodies, etanercept also binds lymphotoxin-α, another member of the TNF cytokine superfamily. Moreover, adalimumab, etanercept, golimumab and infliximab all have human IgG1 Fc, and are therefore capable of fixing complement and binding to Fc receptors; by contrast, certolizumab lacks an Fc component and does not exert Fc-mediated effects.
Although the potential clinical corollaries of the differences among the TNF antagonists in biologic structure, binding and Fc effects have not yet been well studied, the differences may contribute to the variability in these agents' efficacy and adverse event profiles seen in clinical trials and the practice setting. One pharmacologic feature, immunogenicity, has been well studied and has been shown to affect the therapeutic efficacy and tolerability of these agents. The presence of antibodies directed against adalimumab and infliximab has been associated with the lack or loss of clinical responses in RA in several clinical studies. In RA placebo-controlled trials, antidrug antibodies with neutralizing activity have also been detected in patients receiving certolizumab, and antibody formation was associated with reduced clinical efficacy. In the case of etanercept, antibodies identified using an ELISA assay method have been non-neutralizing and generally transient. In a 2009 study, de Vries et al. reported that no antibodies against etanercept were detected when using a two-site assay and antigen-binding test. Interestingly, in a recent study, patients previously treated with either infliximab or adalimumab and who had detectable antibodies to these agents showed a response to etanercept similar to that seen in biologic-naive patients. Patients initially treated with infliximab or adalimumab without antibodies and switched to etanercept therapy had a significantly lower response than those with antibodies. Positive antibody responses have been observed in RA clinical trials of golimumab; most antibodies were neutralizing, but a relationship between antigolimumab antibodies and clinical efficacy has not been confirmed owing to the small number of antibody-positive patients.
Efficacy & Safety in Randomized Controlled Trials Although generalizations are commonly made about the efficacy and safety of biologic therapy as a whole in the treatment of RA, individual agents within the biologic classes of therapy may have unique effects. No randomized controlled trials have directly compared TNF antagonists, and some meta-analyses have demonstrated equal effectiveness and safety based on indirect comparisons of findings from randomized controlled trials. However, the results of several recent meta-analyses suggest potential differences among these agents. Singh et al. conducted a network meta-analysis of Cochrane systematic reviews of biologics for RA that evaluated two major outcomes: benefit, which was defined as a 50% improvement in ACR criteria (ACR50) reported in 27 studies; and safety, which was assessed by the number of withdrawals due to adverse events reported in 29 studies. Compared with placebo, all of the biologics included in the meta-analysis were significantly more likely to provide an ACR50 response. In indirect comparisons of treatment benefit between biologics (Figure 4A), anakinra was associated with a significantly lower likelihood of patients achieving an ACR50 response than etanercept and adalimumab. In addition, significantly fewer withdrawals related to adverse events were found among etanercept-treated patients than among anakinra-, infliximab- or adalimumab-treated patients (Figure 4B).
(Enlarge Image)
Figure 4.
Findings of an indirect comparison between biologics for benefit (American College of Rheumatology 50% improvement criteria) and safety (number of withdrawals due to adverse events). In (A), a value of >1.0 indicates a benefit from drug #1; a value <1.0 indicates a benefit from drug #2. In (B), a value of >1.0 indicates a benefit from drug #2; a value <1.0 indicates a benefit from drug #1.Reproduced with permission from [5].
The relative efficacy of the approved anti-TNF agents adalimumab, certolizumab, etanercept, golimumab and infliximab in adult patients with RA and an inadequate response to methotrexate was also recently estimated using a mixed treatment comparison model in a systematic review of 16 randomized controlled trials. Schmitz et al. reported some differences among these biologics, including greater efficacy of certolizumab versus adalimumab and infliximab and of etanercept versus infliximab and golimumab. Analysis of HAQ outcomes suggested that adalimumab, certolizumab, etanercept and golimumab were superior to infliximab, and that etanercept was superior to adalimumab.
Similarly, differences among TNF antagonists have also been observed in other independent analyses, including: the 2006 Health Technology Assessment report commissioned by NICE for appraisal of adalimumab, etanercept and infliximab for the treatment of adult patients with RA; the 2009 systematic review/meta-analysis of the efficacy and adverse events associated with adalimumab, etanercept, infliximab, anakinra, rituximab and abatacept conducted by the Australian Pharmaceutical Benefits Advisory Committee; and the meta-analysis of malignancy risk in clinical trials of adalimumab, etanercept and infliximab commissioned by the European Medicines Evaluation Agency. Although a detailed reporting of these comparative analyses is beyond the scope of this review, they are nonetheless noteworthy, as their outcomes reinforce the dissimilarity of the TNF antagonists and raise questions about the validity of nonclinical substitutions.
Evidence From Clinical Practice Evidence from several national observational registries has also demonstrated important differences among anti-TNF biologics, including in treatment retention rates. Survival rates on drug therapy are generally considered valuable markers of a therapy's efficacy and safety in the clinical practice setting. In the Czech National Registry, ATTRA, the survival rates of patients with RA treated with adalimumab and patients treated with etanercept were significantly higher than those of patients treated with infliximab during 4 years of treatment; no significant difference in survival rates was observed between etanercept and adalimumab. Similar findings were reported in the nationwide Danish DANBIO Registry, in which drug continuation rates were significantly higher with adalimumab and etanercept at 48 months than with infliximab. A study from the Swiss registry also showed significant differences in discontinuation among the three TNF antagonists, with the lowest retention reported for infliximab. Adherence rates were also significantly higher in etanercept-treated patients than in infliximab-treated patients with RA in a 5-year observational study in southern Sweden. In the Italian Lombardy Rheumatology Network registry, a significantly greater likelihood of survival was reported in patients receiving etanercept than in those receiving adalimumab or infliximab after 36 months.
Some reports in the literature have also suggested important distinctions among anti-TNF agents in specific adverse events, including in the rates of certain infections. A potentially higher incidence of granulomatous infections such as tuberculosis (TB) has been reported in patients treated with infliximab versus etanercept, and results from biologic therapy registries have provided additional support for this observation. According to recent registry data, the risk of TB is significantly higher in patients treated with either infliximab or adalimumab than in those treated with etanercept. Some registries have also reported a lower risk of other types of infections with the TNF receptor fusion protein etanercept compared with the monoclonal antibodies infliximab and adalimumab, including serious lower respiratory tract infection, herpes zoster and non-TB opportunistic infections. As TB and these other infections involve cell-mediated immunity, the cited differences in prevalence have been attributed to differences in the binding of these agents to membrane-bound TNF.
Generally, no significant differences among individual TNF antagonists in malignancy risk have been found; however, there are a few exceptions. In a recent report from the Anti-Rheumatic Therapy in Sweden registry, the incidence of malignancy in patients treated with etanercept was significantly lower in the first year of exposure than in patients treated with anti-TNF monoclonal antibodies, although this difference disappeared in later years. Researchers for the French enhanced pharmacovigilence program Research Axed on Tolerance of Biotherapies also reported a significantly lower risk of lymphoma with etanercept than with monoclonal antibodies, for which a biological explanation has been proposed.
Based on this evidence of potential differences among TNF antagonists in mechanism of action and other properties, as well as efficacy and safety profiles, these agents are clearly not interchangeable. Transferring among the agents in patients who are biologic-naive or biologic-experienced may have clinically meaningful consequences, including a reduced likelihood of response or greater risk of adverse events. Given the limited research on nonclinical switching and the strong evidence of dissimilarities among the biologic agents, caution is warranted when physicians are considering a change in therapy in patients with an adequate treatment response and no intolerable side effects.
Switching for Nonclinical Reasons
Increasing healthcare expenditures have led governments and their agencies across the globe to seek ways to limit healthcare spending and develop strategies to maximize health gains from fixed expenditures. The cost of RA has increased substantially with the introduction of effective biologic therapies, which target proinflammatory cytokines such as TNF and IL-6, and lymphocytes such as B or T cells. With annual costs that are considerably higher than those of older RA drug therapies, biologic therapies have gained particular attention from healthcare decision-makers who must evaluate whether the additional costs of biologics are worth the added clinical benefits.
One approach taken by payers to limit expenditures on biologic therapy in RA is the selection of the lowest price drug as the first-choice product. Relative prices of individual biologic agents vary by country, and acquisition costs fluctuate owing to discounts and patient access schemes. To address ongoing fiscal pressures, payers may suggest that less expensive agents be substituted for current treatments even in patients with an adequate response. Another example of nonclinical switching involves the substitution of a subcutaneously administered biologic agent for an intravenously administered agent (or vice versa) in patients who have responded successfully to the initial therapy, owing to patient preference or economic pressure.
Switching biologic therapies in patients with RA for economic or other nonclinical purposes is based, at least in part, on the assumption that the agents are equivalent in terms of safety and effectiveness. In the following sections, the evidence will be reviewed from studies in which biologic therapies have been switched for nonclinical reasons and the assumption of biologic equivalence will be examined.
Nonclinical Switching Studies
Evidence to support switching among biologic therapies in patients with a therapeutic response to initial treatment is very limited. Based on a literature search, three studies were identified that included patients with RA who were established on a TNF antagonist (infliximab in all three studies) before switching to another anti-TNF agent for nonclinical reasons (Table 3). Laas et al. reported outcomes in 49 patients who were transitioned from infliximab to etanercept therapy in an observational study conducted in Finland. In the subgroup of 23 patients who had demonstrated a good response to infliximab and were switched for nonclinical reasons (e.g., preference for outpatient treatment), a nonsignificant increase in DAS28 of 0.4 was observed after the switch (Figure 2); patients switched owing to infliximab failure and adverse events showed statistically significant decreases in DAS28 of -1.2 (p = 0.023) and -1.3 (p = 0.048), respectively.
(Enlarge Image)
Figure 2.
Changes in 28-joint Disease Activity Score during infliximab and etanercept treatment by reason for switching therapy. The horizontal dashed line represents the cut-off level for DAS28 remission (DAS28 <2.6).DAS28: 28-joint Disease Activity Score.Reproduced from [35] with permission from Springer Science+Business Media.
In a 32-week, open-label study conducted in Ireland, Walsh et al. evaluated 19 patients with RA who had demonstrated a clinical response to hospital-administered infliximab (defined as DAS28 improvement >1.2 within 3 months of treatment) after they were switched to home-administered adalimumab for economic reasons. Mean DAS28 improved from 4.1 at baseline (i.e., at penultimate infliximab infusion) and 3.9 at the time of the switch to 3.0 after 16 weeks of adalimumab therapy (p < 0.005 vs both baseline and switch; Figure 3). No significant changes were observed in tender or swollen joint counts or in functional/quality-of-life assessments, including HAQ scores. The latter finding may be an important consideration for health economics, as cost–effectiveness models are driven by changes in HAQ, which was reduced from the time of switching to the final visit from 1.1 to 1.0 (Figure 3). Although all patients met response criteria for enrollment in the study, at the time that the therapies were switched, nearly three-quarters of patients had moderate (DAS28 3.2–5.0) or high (DAS28 >5.0) disease activity; therefore, they could be considered inadequate responders to TNF-antagonist therapy.
(Enlarge Image)
Figure 3.
Changes in clinical and functional/quality-of-life assessments before, during and after patients were switched from infliximab to adalimumab therapy. (A) DAS28 (mean ± SEM); (B) HAQ Disability Index (mean ± SEM).*p < 0.005 versus baseline and time of switch.DAS28: 28-Joint Disease Activity Score; HAQ: Health Assessment Questionnaire;SEM: Standard error of the mean.Adapted from [36] with permission from Oxford University Press.
In a 6-month, open-label study conducted in the UK, 14 out of 17 patients with stable RA receiving infliximab (3 mg/kg every 8 weeks), given a choice to change to treatment with adalimumab, agreed to switch. Adalimumab was started 8 weeks after the last infliximab infusion. Median DAS28 at 6 months was 3.6, which was similar to the preswitch score of 3.7 (Table 4). Of the 14 switchers, 11 patients demonstrated a good clinical response and continued receiving adalimumab long term. Among the three patients who did not continue adalimumab, two patients worsened after the switch and reverted to treatment with infliximab; one patient discontinued use of adalimumab, with a plan to restart infliximab, but was being managed with methotrexate and corticosteroids at last follow-up. As in the previously described Irish study, not all of the patients enrolled in this study would be considered infliximab responders, as their DAS28 scores were greater than 5.1, and outcomes were not analyzed by level of response before switching.
Comparison of Biologic Therapies
Justification for switching biologic agents in patients with RA for nonclinical purposes depends in part on these agents having comparable efficacy and safety profiles. As summarized in the following sections, salient differences between these agents have been detected that highlight their unique pharmacologic profiles ( Table 1 ).
Mechanism of Action/Other Properties Agents representing the major biologic classes used in the treatment of RA demonstrate differences in mode of action. Even the five currently approved TNF antagonists differ substantially in their mechanisms of action and kinetics, despite their shared inhibition of the cytokine TNF. In particular, etanercept, a fusion protein of human p75 soluble TNF receptor and human IgG1, appears to bind differently to cell-bound TNF than the other agents, which are all monoclonal antibodies or fragments of antibodies. Unlike the monoclonal antibodies, etanercept also binds lymphotoxin-α, another member of the TNF cytokine superfamily. Moreover, adalimumab, etanercept, golimumab and infliximab all have human IgG1 Fc, and are therefore capable of fixing complement and binding to Fc receptors; by contrast, certolizumab lacks an Fc component and does not exert Fc-mediated effects.
Although the potential clinical corollaries of the differences among the TNF antagonists in biologic structure, binding and Fc effects have not yet been well studied, the differences may contribute to the variability in these agents' efficacy and adverse event profiles seen in clinical trials and the practice setting. One pharmacologic feature, immunogenicity, has been well studied and has been shown to affect the therapeutic efficacy and tolerability of these agents. The presence of antibodies directed against adalimumab and infliximab has been associated with the lack or loss of clinical responses in RA in several clinical studies. In RA placebo-controlled trials, antidrug antibodies with neutralizing activity have also been detected in patients receiving certolizumab, and antibody formation was associated with reduced clinical efficacy. In the case of etanercept, antibodies identified using an ELISA assay method have been non-neutralizing and generally transient. In a 2009 study, de Vries et al. reported that no antibodies against etanercept were detected when using a two-site assay and antigen-binding test. Interestingly, in a recent study, patients previously treated with either infliximab or adalimumab and who had detectable antibodies to these agents showed a response to etanercept similar to that seen in biologic-naive patients. Patients initially treated with infliximab or adalimumab without antibodies and switched to etanercept therapy had a significantly lower response than those with antibodies. Positive antibody responses have been observed in RA clinical trials of golimumab; most antibodies were neutralizing, but a relationship between antigolimumab antibodies and clinical efficacy has not been confirmed owing to the small number of antibody-positive patients.
Efficacy & Safety in Randomized Controlled Trials Although generalizations are commonly made about the efficacy and safety of biologic therapy as a whole in the treatment of RA, individual agents within the biologic classes of therapy may have unique effects. No randomized controlled trials have directly compared TNF antagonists, and some meta-analyses have demonstrated equal effectiveness and safety based on indirect comparisons of findings from randomized controlled trials. However, the results of several recent meta-analyses suggest potential differences among these agents. Singh et al. conducted a network meta-analysis of Cochrane systematic reviews of biologics for RA that evaluated two major outcomes: benefit, which was defined as a 50% improvement in ACR criteria (ACR50) reported in 27 studies; and safety, which was assessed by the number of withdrawals due to adverse events reported in 29 studies. Compared with placebo, all of the biologics included in the meta-analysis were significantly more likely to provide an ACR50 response. In indirect comparisons of treatment benefit between biologics (Figure 4A), anakinra was associated with a significantly lower likelihood of patients achieving an ACR50 response than etanercept and adalimumab. In addition, significantly fewer withdrawals related to adverse events were found among etanercept-treated patients than among anakinra-, infliximab- or adalimumab-treated patients (Figure 4B).
(Enlarge Image)
Figure 4.
Findings of an indirect comparison between biologics for benefit (American College of Rheumatology 50% improvement criteria) and safety (number of withdrawals due to adverse events). In (A), a value of >1.0 indicates a benefit from drug #1; a value <1.0 indicates a benefit from drug #2. In (B), a value of >1.0 indicates a benefit from drug #2; a value <1.0 indicates a benefit from drug #1.Reproduced with permission from [5].
The relative efficacy of the approved anti-TNF agents adalimumab, certolizumab, etanercept, golimumab and infliximab in adult patients with RA and an inadequate response to methotrexate was also recently estimated using a mixed treatment comparison model in a systematic review of 16 randomized controlled trials. Schmitz et al. reported some differences among these biologics, including greater efficacy of certolizumab versus adalimumab and infliximab and of etanercept versus infliximab and golimumab. Analysis of HAQ outcomes suggested that adalimumab, certolizumab, etanercept and golimumab were superior to infliximab, and that etanercept was superior to adalimumab.
Similarly, differences among TNF antagonists have also been observed in other independent analyses, including: the 2006 Health Technology Assessment report commissioned by NICE for appraisal of adalimumab, etanercept and infliximab for the treatment of adult patients with RA; the 2009 systematic review/meta-analysis of the efficacy and adverse events associated with adalimumab, etanercept, infliximab, anakinra, rituximab and abatacept conducted by the Australian Pharmaceutical Benefits Advisory Committee; and the meta-analysis of malignancy risk in clinical trials of adalimumab, etanercept and infliximab commissioned by the European Medicines Evaluation Agency. Although a detailed reporting of these comparative analyses is beyond the scope of this review, they are nonetheless noteworthy, as their outcomes reinforce the dissimilarity of the TNF antagonists and raise questions about the validity of nonclinical substitutions.
Evidence From Clinical Practice Evidence from several national observational registries has also demonstrated important differences among anti-TNF biologics, including in treatment retention rates. Survival rates on drug therapy are generally considered valuable markers of a therapy's efficacy and safety in the clinical practice setting. In the Czech National Registry, ATTRA, the survival rates of patients with RA treated with adalimumab and patients treated with etanercept were significantly higher than those of patients treated with infliximab during 4 years of treatment; no significant difference in survival rates was observed between etanercept and adalimumab. Similar findings were reported in the nationwide Danish DANBIO Registry, in which drug continuation rates were significantly higher with adalimumab and etanercept at 48 months than with infliximab. A study from the Swiss registry also showed significant differences in discontinuation among the three TNF antagonists, with the lowest retention reported for infliximab. Adherence rates were also significantly higher in etanercept-treated patients than in infliximab-treated patients with RA in a 5-year observational study in southern Sweden. In the Italian Lombardy Rheumatology Network registry, a significantly greater likelihood of survival was reported in patients receiving etanercept than in those receiving adalimumab or infliximab after 36 months.
Some reports in the literature have also suggested important distinctions among anti-TNF agents in specific adverse events, including in the rates of certain infections. A potentially higher incidence of granulomatous infections such as tuberculosis (TB) has been reported in patients treated with infliximab versus etanercept, and results from biologic therapy registries have provided additional support for this observation. According to recent registry data, the risk of TB is significantly higher in patients treated with either infliximab or adalimumab than in those treated with etanercept. Some registries have also reported a lower risk of other types of infections with the TNF receptor fusion protein etanercept compared with the monoclonal antibodies infliximab and adalimumab, including serious lower respiratory tract infection, herpes zoster and non-TB opportunistic infections. As TB and these other infections involve cell-mediated immunity, the cited differences in prevalence have been attributed to differences in the binding of these agents to membrane-bound TNF.
Generally, no significant differences among individual TNF antagonists in malignancy risk have been found; however, there are a few exceptions. In a recent report from the Anti-Rheumatic Therapy in Sweden registry, the incidence of malignancy in patients treated with etanercept was significantly lower in the first year of exposure than in patients treated with anti-TNF monoclonal antibodies, although this difference disappeared in later years. Researchers for the French enhanced pharmacovigilence program Research Axed on Tolerance of Biotherapies also reported a significantly lower risk of lymphoma with etanercept than with monoclonal antibodies, for which a biological explanation has been proposed.
Based on this evidence of potential differences among TNF antagonists in mechanism of action and other properties, as well as efficacy and safety profiles, these agents are clearly not interchangeable. Transferring among the agents in patients who are biologic-naive or biologic-experienced may have clinically meaningful consequences, including a reduced likelihood of response or greater risk of adverse events. Given the limited research on nonclinical switching and the strong evidence of dissimilarities among the biologic agents, caution is warranted when physicians are considering a change in therapy in patients with an adequate treatment response and no intolerable side effects.
Source...