Application to Add Anti-TNFs to the World Health Organization’s Essential Medicines List 1. Summary statement of the proposal for inclusion, change or deletion

Tumour Necrosis-Factor-alpha (TNF-α) inhibitors, also known as Anti-TNF biologics, have proven to be an effective and safe treatment option for several chronic inflammatory autoimmune disorders, including rheumatology and gastroenterology. These drugs work to suppress the immune system by blocking TNF activity which is responsible for bodily inflammation and immune diseases. Although they improve outcomes, concerns have been raised regarding their high costs compared to other treatment options: due to their high costs, it is not usually recommended as a first line of treatment. This class remains absent in the current Essential Medicines List. Thus, this application proposes the addition of Anti-TNF to the Essential Medicines List (EML) and the Essential Medicines for Children (EMLc) for the treatment of Rheumatoid Arthritis, Ankylosing Spondylitis, Juvenile Idiopathic Arthritis, and Crohn’s disease, as part of the therapeutic options to deal with refractory cases. 2. Relevant WHO technical department and focal point 3. Name of organization(s) consulted and/or supporting the application Centre for Global Health - University of Ottawa. 4. International Nonproprietary Name (INN) and Anatomical Therapeutic Chemical (ATC) code of the medicine

This application pertains to immunosuppressants, the Tumor Necrosis Factor Alpha (TNF-α) Inhibitors classification, which contains 6 substances

International Nonproprietary Name Trade Name Anatomical Code Etanercept Enbrel L04AB01 Infliximab Remicade L04AB02 Afelimomab L04AB03 Adalimumab Humira L04AB04 Certolizumab pegol Cimzia L04AB05 Golimumab Simponi L04AB06

Due to the limited amount of evidence regarding the effectiveness and safety of Afelimomab, we do not propose its addition to the Core Essential Medicines List. 5. Formulation(s) and strength(s) proposed for inclusion (adult and pediatric)

We propose the following formulation for the use in pediatric and adult patients, which has full market availability:

INN Strength(s) Pediatric Adult Formulation Etanercept 0.8 mg/kg ✓ Subcutaneous injection 50 mg ✓ Subcutaneous injection Infliximab 3 mg/kg ✓ Intravenous injection 5 mg/kg ✓ ✓ Intravenous injection 10 mg/kg ✓ Intravenous injection Adalimumab 10 mg ✓ Subcutaneous injection 20 mg ✓ ✓ Subcutaneous injection 40 mg ✓ ✓ Subcutaneous injection 80 mg ✓ ✓ Subcutaneous injection 160 mg ✓ ✓ Subcutaneous injection Certolizumab pegol 200 mg ✓ Subcutaneous injection 400 mg ✓ Subcutaneous injection Golimumab 50 mg ✓ Subcutaneous injection 50 mg ✓ Intravenous injection

6. Whether listing is requested as an individual medicine or as representative of a class.

This application covers the group of medicines Tumour Necrosis-Factor-alpha (TNF-α) Inhibitors, including branded medicinal products and biosimilars. The biosimilars of interest are: ● Etanercept-szzs ● Infliximab-qbtx ● Infliximab-dyyb ● Infliximab-abda ● Adalimumab-atto ● Adalimumab-adbm

We propose a square box indication, highlighting the therapeutic equivalence of biosimilars to their biologics. These biosimilars have a similar structure and function as their biologic. They are have been assessed to confirm that there are no meaningful clinically differences regarding effectiveness and safety. This will provide choice to nations to ensure drugs are affordable and practical. The choice of Anti-TNF therapy should be determined by consultation between the medical professional and the patient.

7. Treatment details (requirements for diagnosis, treatment and monitoring)

Table 1 reports the current dosing regimen for Anti-TNF biologics in the treatment of Rheumatoid Arthritis, Ankylosing Spondylitis, Juvenile Idiopathic Arthritis, and Crohn’s disease in adult and pediatric patients. All professionals must have the appropriate knowledge and clinical skills to be aware of the diagnostic and treatment criteria. Generally, Anti-TNF biologics are prescribed when patients do not respond to conventional therapies and have a high risk for unfavourable prognosis. Anti-TNF biologics are to be prescribed by medical doctors and specialists in diagnosed patients. Subcutaneous injection Anti-TNF biologics may be administered by patients or their caregivers once they are instructed by a health professional on how to successfully administer it under the skin. Intravenous injections must be administered by a trained medical professionals, such as a nurse. All individuals who are prescribed Anti-TNF biologics should be monitored for the development of new infections or adverse reactions. However, it is crucial that these individuals are monitored for signs and symptoms of active tuberculosis during the course of treatment due to the possibility of a false negative test result. Those who have previously been identified as a chronic carrier of Hepatitis B should also be closely monitored for signs and symptoms of active infection during therapy and for the months following discontinuation. Individuals receiving Anti-TNFs, especially Infliximab, should be screened for cardiovascular events including new or worsening symptoms of heart failure (e.g., shortness of breath or swelling of the feet) as well as malignancies. Vaccinations should ideally be up to date in patients using Anti-TNF biologic therapy and the use of live vaccinations is not recommended. Use with other biological DMARDs such as Abatacept and Anakinra are not recommended due to the increased risk of serious infections.

Table 1. Dosing Regimen for Anti-TNF Biologics Drug Adult Dosage & Route of Pediatric Dosage & Route of Biosimilar Generic Administration Administration Name Etanercept Rheumatoid Arthritis & Ankylosing Juvenile Idiopathic Arthritis: Biosimilar Etanercept-szzs Spondylitis: Subcutaneous injection of 50 Children 2–17 years of age→ mg weekly. Subcutaneous injection of 0.8 mg/kg (maximum 50 mg) per week. Infliximab Rheumatoid Arthritis: Intravenous Crohn’s Disease: Children 6 years Biosimilar Infliximab-qbtx, injection of 3 mg/kg at 0, 2 and 6 weeks of age or older → Intravenous Infliximab-dyyb, and for induction regimen and every 8 weeks Infliximab-abda injection 5 mg/kg at 0, 2 and 6 following for maintenance regimen. Dosage may be increased up to 10 mg/kg weeks for induction regimen and and/or administered as often as once every 8 weeks following for every 4 weeks to patients who do not maintenance regimen. Considering respond to 3 mg/kg. discontinuing for patients who do Ankylosing Spondylitis: Intravenous not respond by week 14. injection of 5 mg/kg at 0, 2 and 6 weeks for induction regimen and every 8 weeks following for maintenance regimen.

Crohn’s Disease: Intravenous injection 5 mg/kg at 0, 2 and 6 weeks for induction regimen and every 8 weeks following for maintenance regimen. Dose of 10 mg/kg for patients who respond initially, but lose response. Considering discontinuing for patients who do not respond by week 14. Adalimumab Rheumatoid Arthritis: Subcutaneous Juvenile Idiopathic Arthritis: Biosimilar Adalimumab- injection of 40 mg once every other Children 4–17 years of age → atto and Adalimumab- week. Patients not receiving adbm Subcutaneous injection of 10 mg Methotrexate, may receive weekly doses of 40 mg. every other week for children Ankylosing Spondylitis: Subcutaneous weight 10 to <15 kg, 20 mg once administration of 40 mg once every other every other week for children week. weighing 15 to <30 kg and 40 mg Crohn’s Disease: Subcutaneous injection once every other week for children of 160 mg on day 1 (as four 40 mg

injections in one day or as two 40 mg ≥30 kg. injections per day for 2 consecutive Crohn’s Disease: For children >40 kg days), followed by 80 mg once 2 weeks subcutaneous injection of 160 mg on later (on day 15). Maintenance dosage of day 1 (as four 40 mg injections in one 40 mg once every other week on day 29 day or as two 40 mg injections per (2 weeks after the 80 mg dose). day for 2 consecutive days), followed by 80 mg once 2 weeks later (on day 15). Maintenance dosage of 40 mg once every other week on day 29 (2 weeks after the 80-mg dose). For children between 17 and <40 kg, all dosages are cut in half. Certolizumab Rheumatoid Arthritis: Subcutaneous pegol injection of 400 mg (as two 200 mg injections at separate sites) at 0, 2, and 4 weeks, followed by 200 mg every 2 weeks. 400 mg every 4 weeks may be considered for maintenance therapy. Ankylosing Spondylitis: Subcutaneous injection of 400 mg (as two 200 mg injections at separate sites) at 0, 2, and 4 weeks, followed by 200 mg every 2 weeks or 400 mg every 4 weeks. Crohn’s Disease: 400 mg (as two 200 mg injections at separate sites) at 0, 2, and 4 weeks for the induction regimen and patients who respond may receive additional 400 mg doses every 4 weeks for the maintenance regimen.

Golimumab Rheumatoid Arthritis & Ankylosing Spondylitis: Subcutaneous injection of 50 mg once monthly Sources: DynaMed Plus, 2018; Food and Drug Administration FDA, 2018a; Health Canada, 2018a

Rheumatoid Arthritis Diagnosis Early Rheumatoid Arthritis is defined as having Rheumatoid Arthritis for no more than 1 year of the diagnosed disease duration (Donahue et al., 2018). After this point in time, an individual progresses to advanced stage Rheumatoid Arthritis. It is important to separate these two conditions from one another as they warrant the need for different treatment approaches. The 2010 Rheumatoid Arthritis Classification Criteria, defined by the American College of Rheumatology and the European League Against Rheumatism, is the most standard and accepted means of a definite diagnosis of Rheumatoid Arthritis to date (Aletaha et al., 2010). As such, a Rheumatoid Arthritis case is based on i) the confirmed presence of synovitis in at least 1 small joint, ii) synovitis cannot better explained by an alternative diagnosis and iii) achievement of a total score of 6 or greater (of a possible 10) from the individual scores in 4 domains: number and site of involved joints (score range 0–5), serologic abnormality (score range 0–3), elevated acute- phase response (score range 0–1), and symptom duration (2 levels; range 0–1) (Aletaha et al., 2010). This updated classification criteria differs from the American College of Rheumatology 1987 criteria (Arnett et al., 1988), although also frequently used, as the 1987 criteria are suggested to predict a more erosive disease (Cornec et al., 2012).

Treatment Early treatment for Rheumatoid Arthritis is recommended, with the primary goal of sustained remission or the lowest disease activity possible (treat-to-target). The form of treatment depends on the level of disease activity - low, moderate or high. In general, the National Institute for Health and Care Excellence (NICE) (National Institute for Health and Care Excellence (NICE), 2018), European League of Associations in Rheumatology (EULAR) (Smolen et al., 2017), Canadian Rheumatology Association (Bykerk et al., 2012), and American College of Rheumatology (Singh et al., 2016) recommend beginning treatment as soon as possible, within 3 months of persistent symptoms, through the use of one conventional disease modifying anti- rheumatic drug (csDMARD). Methotrexate remains the initial drug of choice in all clinical practice guidelines, however, leflunomide, sulfasalazine or hydroxychloroquine are other options. If patients do not respond to csDMARDS and have poor prognostics factors, the practice guidelines recommend either a) an alternative csDMARD b) an increased dose of the original csDMARD or c) csDMARD double-or-triple therapy. Low-dose short-term glucocorticoids are recommended, in addition, if disease activity is moderate or high despite csDMARD use. In non- respondent patients, TNF or non-TNF biologics can be used as monotherapy or in combination with csDMARDS, most commonly methotrexate (please refer to table 1 for the dosing regimen of Anti-TNFs in Rheumatoid Arthritis treatment). Other treatment options include targeted synthetic DMARDS (tsDMARDs) or biosimilars.

Monitoring Active Rheumatoid Arthritis patients starting Anti-TNF therapy should be monitored as frequently as 1-3 months to evaluate the initial response to treatment (Royal College of Nursing (RCN), 2013; NICE, 2018; Bykerk et al., 2012). Those with low disease activity may be monitored at longer intervals, with recommendations of at least every 6 months (Bykerk et al., 2012; Holroud et al., 2018). According to practice guidelines, biologic therapy should only be continued if there is an adequate response 6 months following initiation, and if this response can

be adequately maintained. For those who do not respond, an alternative Anti-TNF medication can be considered (RCN, 2013). It is further recommended that adults who have maintained remission or low disease activity for one year without the use of glucocorticoids may benefit from reduced drug doses using a step-down approach (NICE, 2018). Therefore, systematic monitoring of disease activity may lead changes in treatment and result in more patients with lower disease activity (Fransen et al., 2005).

Ankylosing Spondylitis Diagnosis The most commonly cited classification criteria in the literature for the diagnosis of systematic Ankylosing Spondylitis are 1961 Rome, 1966 New York, or modified 1984 New York, European Spondyloarthropathy Study Group and Amor criteria. All tools include the criterion of inflammatory back or spinal pain (Aliling et al., 2016). The 1984 modified New York criteria has been the most widely used tool, however is not applicable for early Ankylosing Spondylitis as it does not consider those who have not yet developed radiographic sacroiliitis (Raychaudhuri et al. 2014). According to this criteria, a patient has definite Ankylosing Spondylitis if they possess at least one of the clinical criterion (inflammatory back pain, limited mobility of the lumbar spine, or limitation of chest expansion) plus a radiologic criterion (radiographic sacroiliitis of at least grade 2 bilaterally or grade 3–4 unilaterally) (Raychaudhuri et al. 2014). The Amor Classification Criteria for Spondyloarthopathy and the European Spondyloarthropathy Study Group Classification Criteria were later created to include patients with undifferentiated Ankylosing Spondylitis. These tools added additional criterion such as the presence of HLA-B27 (Amor criteria only) or a family history of Ankylosing Spondylitis (Aliling et al., 2016).

Treatment The goals of treatment for Ankylosing Spondylitis are to relieve symptoms, maintain spinal flexibility and normal posture, improve physical function and avoid structural damage or complications (Ward et al., 2016). A multidisciplinary approach is needed for the treatment of this disease with a combination of exercise, physical therapy and drug therapy. Practice guidelines have indicated that non-steroidal anti-inflammatory drugs (NSAIDs) are the first- line drug used to reduce pain and inflammation is (Rohekar et al., 2015; Ward et al., 2016; National Institute for Health and Care Excellence (NICE), 2017). However, there is a lack of data evaluating the long-term harms of NSAID therapy in Ankylosing Spondylitis (Kroon et al., 2015). Anti-TNF biologics, are used for those who do not respond well to NSAIDs (please refer to table 1 for the dosing regimen of Anti-TNF in Ankylosing Spondylitis treatment) (Rohekar et al., 2015; Ward et al., 2016; NICE 2017). The ASAS working group, ASAS European League Against Rheumatism and the Canadian Rheumatology Association/Spondyloarthritis Research Consortium of Canada have previously suggested the use of biologic therapy once a patient has a BASDAI with a cut off of ⩾4 for active disease activity (Braun et al., 2006; van der Heijde et al., 2016; Rohekar et al., 2015). It is unclear whether objective signs of inflammation, such as an MRI or raised C reactive protein, are needed prior to initiating Anti-TNF treatment. Patients with mainly peripheral symptoms may include a short-term glucocorticoid injection (van der Heijde et al., 2016;

Ward et al., 2016). If therapy fails, a different Anti-TNF should be considered. The ASAS European League Against Rheumatism indicate that non-TNF Biologics may be considered only for patients who are unresponsive to Anti-TNF therapy (van der Heijde et al., 2016).

Monitoring Disease monitoring of patients with Ankylosing Spondylitis should include patient history as well as clinical and laboratory tests according to clinical presentation and core set (Sieper et al., 2009). Individuals with Ankylosing Spondylitis on Anti-TNF therapy should be monitored for a response 12 weeks after the initiation of treatment and then at 12 week intervals (RCN, 2003; NICE, 2017). Anti-TNF therapy should be continued if a patient has a favourable treatment response 16 weeks after initiation, defined as either an absolute reduction by 2 in the BASDAI (0–10 scale) or a 50% relative reduction (RCN, 2003; Rohekar et al., 2015; NICE, 2017; van der Heijde et al., 2016).

Juvenile Idiopathic Arthritis Diagnosis The International League of Associations for Rheumatology specifies that the diagnosis of Juvenile Idiopathic Arthritis can be separated into seven distinct categories - oligoarticular, polyarticular (rheumatoid factor positive or negative), systemic, psoriatic, enthesitis-related and undifferentiated arthritis (Ravelli, 2016). In summary, Juvenile Idiopathic Arthritis is diagnosed if there is the presence of arthritis in at least one joint of unknown etiology, onsets before the age of 16 and persists for at least 6 weeks (Ravelli, 2016).

Treatment Juvenile Idiopathic Arthritis is a complex disease that requires a combined multi-disciplinary approach dependent on the subtype, current disease activity and prognosis (Smith, 2017). Treatment goals are to control and delay the progression of Juvenile Idiopathic Arthritis to prevent joint damage and to reduce pain (Smith, 2015; Ringold et al., 2013). For all Juvenile Idiopathic Arthritis subtypes, the American College of Rheumatology 2011 and 2013 updated clinical practice guidelines indicate that the standard first line of treatment is NSAID monotherapy (Beukelman et al., 2011; Ringold et al., 2013). If poor prognostic factors continue following 2 months of NSAID, glucocorticoid joint injections with adjunct NSAID, as needed, are recommended. The escalation of treatment includes methotrexate as an adjunct to NSAID or joint injections and then biologics, most commonly Anti-TNFs (please refer to table 1 for the dosing regimen of Anti-TNFs for Juvenile Idiopathic Arthritis). However, these guidelines are currently being updated to reflect new high-quality evidence regarding the effectiveness of biologic therapies (American College of Rheumatology (ACR), 2017). Currently, the American College of Rheumatology recommends initiating Anti-TNFs for patients with an AJC >4 irrespective of the MD global after a trial of an IL‐ 1 inhibitor or tocilizumab, or for patients with an AJC >0 irrespective of the MD global after a trial of both an IL‐ 1 inhibitor and tocilizumab sequentially (Ringold et al., 2013).

Monitoring It is suggested that patients receiving Anti-TNF therapy should have complete blood cell count, liver enzymes and serum creatinine tests every 3 to 6 months (Beukelman et al., 2011).

Crohn’s Disease Diagnosis Several endoscopic radiologic, histologic, and pathologic tests are all used (e.g., colonoscopy with biopsies, video capsule endoscopy, barium enema, small bowel imaging, cross-sectional imaging with MRI) in combination with clinical presentation to confirm focal, asymmetric, and transmural granulomatous inflammation of the gastrointestinal tract (Lichtenstein et al., 2018; American College of Gastroenterology (AMG), 2011; World Gastroenterology Organisation (WGO), 2015; Masser et al., 2018). Laboratory tests are complementary to assess Crohn’s disease severity and complications, such as blood samples to detect antibodies and markers of inflammation and stool specimen to test for fecal pathogens (Lichtenstein et al., 2018; American College of Gastroenterology (AMG), 2011). Active Crohn’s disease is defined in the literature by the presence of active clinical symptoms, and/or a Crohn's Disease activity index (CDAI) > 150, and/or Pediatric Crohn's Disease Activity Index (PCDAI) > 15, and/or a validated severity index indicating active disease (e.g. Harvey‐ Bradshaw Index, Van Hees Index, Perianal Crohn’s Disease Activity Index) (Lichtenstein et al., 2018; Masser et al., 2018). Remission of Crohn’s disease occurs at a CDAI of <150.

Treatment The two main goals in the treatment of Crohn’s disease are to first eliminate the symptoms, that is to induce clinical remission, and the second is to prevent future disease flare up, that is to maintain remission.Treatment for inducing remission in Crohn’s disease patients with first presentation or single inflammatory exacerbation often starts with short-term conventional glucocorticosteroid therapy (e.g., prednisolone, methylprednisolone or intravenous hydrocortisone) (NICE, 2012; Lichtenstein et al., 2018). Enteral nutrition may be offered as an alternative to a conventional glucocorticosteroid for patients, particularly children, who are concerned about growth and/or side effects (NICE, 2012; Lichtenstein et al., 2018). Budesonide or 5-aminosalicylate can be provided to those who do not have severe presentation and where treatment is contraindicated (NICE, 2012; Lichtenstein et al., 2018). For those who have two or more inflammatory exacerbations in 12 months or the glucocosteriod dose cannot be tapered, azathioprine or mercaptopurine can be added (NICE, 2012). The American Gastroenterological Association Institute does not recommend the use of Thiopurine immunosuppressant monotherapy to induce remission due to the delay in onset of action (Terdiman et al., 2013). If an individual has deficient thiopurine methyltransferase activity either azathioprine or mercaptopurine should not be added and methotrexate should instead be considered (NICE, 2012). The Anti-TNF biologics should be prescribed as monotherapy or combined therapy with immunosuppressants in severe Crohn’s disease patients who have not responded, are intolerant or have contraindications to the above conventional therapies (NICE, 2012; Terdiman et al., 2013; Lichtenstein et al., 2018; Feuerstein et al., 2017; Sadowski et al., 2009). Anti-TNF may be switched for those who do not respond to the first drug and the Anti-TNF planned course of treatment should continue until treatment failure, including surgery (NICE, 2012; Sadowski et al., 2009). Maintenance treatment for Crohn’s disease generally includes azathioprine or mercaptopurine monotherapy, especially for those who used corticosteroid to induce remissions (NICE, 2012; Terdiman et al., 2013; Lichtenstein et al., 2018). Methotrexate is provided to patients who used it to induce remission or do not tolerate/have contradictions to the first option (NICE, 2012; Lichtenstein et al., 2018). The American Gastroenterological Association,

American College of Gastroenterology and Canadian Association of Gastroenterology also recommend the use Anti–TNF Drugs to maintain corticosteroid or Anti–TNF induced remission (Terdiman et al., 2013; Lichtenstein et al., 2018; Sadowski et al., 2009) (please refer to table 1 for the dosing regimen of Anti-TNFs in Crohn’s disease treatment). Surgery is considered for patients who do not respond to medical treatment and/or have adverse prognostic factors (NICE, 2012). Although similar, The European Crohn’s and Colitis treatment guidelines differ according to Crohn’s disease location (Dignass et al., 2010).

Monitoring Treatment monitoring in patients with Crohn’s disease is crucial as therapy should only be continued if there is clear evidence of disease activity through clinical symptoms, biological markers and investigation (NICE, 2012). The objective evidence of improvement should be seen within 2–4 weeks and the maximal improvement should occur within 12–16 weeks. Anti-TNF therapy should only be continued beyond this time if there is a clear therapeutic response (Lichtenstein et al., 2018). Patients who further continue Anti-TNF therapy are encouraged to be monitored at least every 12 months to ensure treatment is still beneficial. Antibody levels to Anti-TNF agents can guide therapy, such as whether a dose can be minimized (WGO, 2015). Routine thiopurine methyltransferase activity and laboratory monitoring, including complete blood count should be performed for patients who started on thiopurine (Terdiman et al., 2013; Feuerstein et al., 2017). 8. Information Supporting Public Health Relevance

Epidemiological Information on Disease Burden Rheumatoid Arthritis The 2016 Global Burden of Disease study found that musculoskeletal conditions were the second highest contributor to global disability and contributed the greatest proportion of lost productivity in the workplace (World Health Organization (WHO), 2018a). Rheumatoid Arthritis, the most common inflammatory musculoskeletal condition, is a chronic autoimmune disease that can affect multiple joints, connective tissues, muscles, tensions and fibrous tissue (World Health Organization WHO, 2018b; Rudan et al., 2015). It is characterized by severe pain and deformity as a result (WHO, 2018a). If Rheumatoid Arthritis is not controlled early, damage may become permanent, leading to significant disability, diminished quality-of-life and increased mortality. In fact, research has found that the lifespan of individuals with Rheumatoid Arthritis is 3 to 12 years less than those who do not have Rheumatoid Arthritis (Donahue et al., 2018).

The global prevalence of Rheumatoid Arthritis falls anywhere between 0.3% and 1.0% (WHO, 2018a). Although it is most common in developed countries, low-and-middle income countries are also largely affected by this disease (Rudan et al., 2015).

Ankylosing Spondylitis Ankylosing Spondylitis is a type of chronic inflammatory arthritis that primarily affects the spine and sacroiliac joints and ligaments. Research has shown that the burden of disease is similar to that of Rheumatoid Arthritis as it relates to pain, disability and well-being (Maxwell et al., 2015). However, individuals with Ankylosing Spondylitis have increased risk at developing articular

and extra-articular manifestations such as psoriasis, anterior uveitis, inflammatory bowel disease (Stolwijk et al., 2015). These manifestations further compound the negative health outcomes and impact prognosis.

It is estimated that the pooled global prevalence of Ankylosing Spondylitis is 0.18% (95% CI 0.15-0.23) (Stolwijk et al., 2016), with Europe (0.25; 95% CI 0.18 - 0.33) and North America (0.20; 95% CI 0.10 – 0.34) comprising the highest prevalence. The prevalence also highest in individuals who are HLA-B27 positive with a family member who has the disease (~20%) (Maxwell et al., 2015).

Juvenile Idiopathic Arthritis The World Health Organization guidelines for pharmacological treatments has acknowledged arthritis as a significant persistent pain for children in most parts of the world. As such, Juvenile Idiopathic Arthritis is a major public health concern of high significance (World Health Organization, 2012). It is the most common chronic rheumatic disease affecting children under the age of 16 years. Juvenile Idiopathic Arthritis is a term that describes several subtypes of inflammatory arthritis of unknown etiology that persist for at least 6 weeks in children.

There are limited epidemiological data for Juvenile Idiopathic Arthritis, likely due to lack of standard diagnostic criteria (Manners et al., 2002; Thierry et al., 2014; Smith, 2017). However, recent estimates indicate that the prevalence varies from 3.8 to 400/100,000 and after directly standardizing for age and gender, the pooled prevalence is 70.2 [62.9–78.1]/100,000 (Thierry et al., 2014). On the contrary, the annual incidence rate for Juvenile Idiopathic Arthritis varies from 1.6 to 23/100,000 and the standardized rate is 8.2 [7.5–9.0]/100,000 (Thierry et al., 2014). Approximately 50% of children have oligoarticular disease (involves 4 or fewer joints), 40% have polyarticular (involves 5 or 13 more joints), and ~10% have systemic symptoms along with arthritis (i.e., systemic arthritis) (ACR, 2017).

Crohn’s Disease Crohn’s Disease is a chronic autoimmune disorder characterized by severe inflammation of the gastrointestinal tract. Crohn’s disease can impact any part of the gastrointestinal tract, but most commonly occurs in the lower part of the small intestine and the colon. The inflammation may be so severe that symptoms manifest as a result and deep and destructive ulcers in the intestinal wall, abdominal infections/abscesses, strictured bowel and/or fistulas form. Most people will need surgery and/or drug treatment. Crohn’s disease is a lifelong systemic condition with deliberating symptoms that negatively affect an individual's quality-of-life. As such, it is associated with high morbidity, mortality, and substantial costs to the health-care system.

Crohn’s disease was traditionally considered a disease impacting developed countries. Although the incidence is still the highest in the westernized nations, it is greatly accelerating in Asia (8·4 100,000 per-person years in Western Asia, 3.91 100,000 per-person years in Southern Asia and 3.2 100,000 per-person years in Eastern Asia), South America (3.50 100,000 per-person years), and Africa (5.87 100,000 per-person years) (Figure 1) (Ng et al., 2017). The highest annual incidences of pediatric Crohn’s disease were 13.9/100, 000 in North America and 12.3/100,000 in Europe (Sýkora et al., 2018).

The overall burden of Crohn’s disease remains high as the prevalence exceeds 0.3% in North America, Oceania, and many countries in Europe. The highest reported prevalence values are reported in Europe (322 per 100, 000 in Germany) and North America (318.5 per 100, 000 in Canada) (Ng et al., 2017). The prevalence of Crohn’s disease has especially risen in the pediatrics population in the last 15 years. For example, from 1999 to 2010, the prevalence of Crohn’s disease in Canadian children under 16 years of age was 38.25 (95% CI 35.78 to 40.73) per 100,000 children (Benchimol et al., 2017). This prevalence increased significantly over time with an APC +4.56% (95% CI +3.71% to +5.42%) (Benchimol et al., 2017).

Figure 1. Crohn’s Disease Prevalence

Source: Ng et al., 2017

Assessment of Current Use

Anti-TNF biologics have become a wide-spread standard of care in the treatment of a number of chronic conditions. Data on the current use of these biologics are relatively high, despite their absence from the Essential Medicines List. For example, a previous study found that between 1999 and 2006, approximately 30% of Rheumatoid Arthritis Patients were treated with Anti- TNF biologic monotherapy in the United States alone (Yazici et al., 2008). The percentage of patients being treatment with Anti-TNF therapy in combination with methotrexate was even higher (approximately 36%). Although regimen modifications were frequent, the use of an Anti- TNF in conjunction with methotrexate had the longest duration of 5.1 (± 5.7) months, whereas biologic monotherapy had a treatment duration of 4.4 (± 6.2) months (p < 0.0001) (Yazici et al.,

2008). Monaco and colleagues (2014) argued that the sales of Anti-TNF biologics points towards the current use in the course of therapy. Anti-TNF biologics are the most profitable drug class with global sales over $US 25 billion per year. Adalimumab is the world’s best selling medicine with global sales over $US 10 billion per year. For the treatment of Rheumatoid Arthritis in 2014 alone, Anti-TNF Biologics comprised global sales of sales of $US 15.1 billion (Chaudhari et al., 2016). In addition, these drugs account for the largest portion of private plan drug spending in Canada in 2016, with the average prescription costing $2,349.90, with Infliximab comprising 38.1% of spending for inflammatory conditions, Adalimumab 28.3%, Etanercept 11.6% and Golimumab 4.8% (Express Scripts Canada, 2016). This number is expected to rise in the upcoming years as research on the effectiveness is growing.

Target Population Rheumatoid Arthritis The onset of Rheumatoid Arthritis begins most commonly between 20 to 40 years of age (WHO, 2018a). The prevalence and incidence of Rheumatoid Arthritis increases with age and peaks in the fifth decade (WHO, 2018b; Donahue et al., 2014). Research shows that at least twice as many women as men are affected (Cross, 2015; Donahue et al., 2014).

Ankylosing Spondylitis The onset of Ankylosing Spondylitis generally occurs in young adults between the ages of 20 and 30 years, however there is often a delay in diagnosis by approximately 5-6 years (Maxwell, 2015). The prevalence of Ankylosing Spondylitis is higher in males compared to females and those with or family history of the genetic marker HLA-B27 (Stolwijk et al., 2016).

Juvenile Idiopathic Arthritis The maximum age of onset for Juvenile Idiopathic Arthritis is 16 years of age with a symptom duration of at least 6 weeks (Thierry et al., 2014; Smith, 2017). In approximately 40 to 70% of cases, Juvenile Idiopathic Arthritis continues into adulthood (Smith, 2017).

Crohn’s Disease The peak onset of Crohn’s disease is 15 - 35 years of age, although it can be diagnosed at any age. Caucasians, people of Ashkenazi Jewish descent and South Asians develop Crohn’s at a higher rate than other ethnicities (Crohn’s and Colitis Canada, 2018). More females are affected by Crohn’s disease than males (Crohn’s and Colitis Canada, 2018).

Likely Impact on Treatment Rheumatoid Arthritis Early treatment for Rheumatoid Arthritis is recommended, with the primary goal of sustained remission or the lowest disease activity possible (treat-to-target). Addressing pain and inflammation have also been indicated as an integral part of care. Although disease modifying drugs are the first-line of treatment for Rheumatoid Arthritis, Anti-TNF biologics are encouraged for patients who do not respond adequately. Furthermore, they have been shown to reduce disease activity, retard joint erosions, and improve patients' quality of life. The ACR/EULAR has recognized that the early use of Anti-TNF therapy not only improves outcomes, but also drastically slows or stops the progression of structural damage (Aletaha et al., 2010). The safety

of these drugs have been well established and their effectiveness is evolving.

Ankylosing Spondylitis The optimal treatment of Ankylosing Spondylitis is to target symptomatic relief of pain and stiffness, improving physical function and avoiding structural damage. Research has proven that Anti-TNF drugs improve pain and functioning, but also increase the overall chances of partial remission of clinical symptoms (Maxwell et al., 2015). Therefore, Anti-TNFs have the potential to greatly improve the quality-of-life for patients. Early diagnosis of Ankylosing Spondylitis is a priority of care as Anti-TNF drugs are the most effective if used in the early stages of the disease (Raychaudhuri et al., 2014). As such, the earlier Anti-TNF treatment is initiated, the lower the radiographic progression (Rohekar et al., 2015).

Juvenile Idiopathic Arthritis Treatment goals are to control and delay the progression of Juvenile Idiopathic Arthritis to prevent joint damage and to reduce pain (Smith, 2015). Anti-TNF therapy has been shown to reduce disease activity in patients with active systemic Juvenile Idiopathic Arthritis (Smith, 2017). Early and aggressive therapy of Anti-TNF biologics yields the most optimal outcomes (Stoll et al., 2014).

Crohn’s Disease The primary goals in the treatment of Crohn’s disease are to induce remission to control symptoms, and once achieved, to maintain remission to decrease the frequency of disease flares. The Anti-TNF Biologics help to achieve this complete state of remission which allows the inflamed tissue to heal and retain its normal function. In doing so, a patient has reduced complications and an improved quality-of-life. Early treatments of Anti-TNF therapy are found to be more cost-effective and efficacious than late initiation. A 2016 cohort study found that patients who were treated with Infliximab or Adalimumab within 2 years of diagnosis had a reduced need for surgical resection and lower rates of dose escalation related to reductions in loss of response (Ma et al., 2016).

9. Review of benefits: summary of evidence of comparative effectiveness

Identification of clinical evidence

This application seeks to summarize the best-available evidence on effectiveness and safety of Anti-TNF Biologics for the treatment of Rheumatoid Arthritis (early and advanced stage), Ankylosing Spondylitis, Juvenile Idiopathic Arthritis, and Crohn’s disease. It includes systematic reviews using a network-meta analysis approach to compare with other available treatments for each disease. The Anti-TNF Biologics Etanercept (ETN), Infliximab (IFX), Adalimumab (ADA), Certolizumab pegol (CZP) and Golimumab (GOL) were compared to conventional therapy and/or placebo.

We searched The Cochrane Library for network meta-analysis covering each of the topics. When none were available from the Library, we did a search on Medline looking for network meta-

analysis. Complementary, we checked with topic experts for the most up-to-date summary on each area. For each target condition, we aimed to find the network meta-analysis that summarized the issues in the most relevant population types. When several were available, we consider the most adequate based on: - Inclusion of Anti-TNF therapies against the most commonly used therapies - The year of publication - Source of funding (we avoided studies funded by industry)

Summary of Available Data & Estimates of Comparative Effectiveness

The characteristics of included studies, outcome measures and results are described in detail for each disease below. Please see Appendix 1 for a complete summary of the data and comparative effectiveness of Anti-TNF biologics against different treatment options.

Rheumatoid Arthritis Two systematic reviews were included for Anti-TNFs in the treatment of early (Donahue et al., 2018) and advanced (Wells et al., 2018) Rheumatoid Arthritis.

The first systematic review assessed how different treatment strategies affected outcomes for patients with early Rheumatoid Arthritis (Donahue et al., 2018). This review included 16 RCTs comparing Anti-TNF biologics to csDMARD monotherapy (n=13) or combination therapy (n=3). One RCT compared TNF and non-TNF therapies. The majority of the included studies were rated as medium Risk of Bias (ROB). The total number of participants was 6908. The mean age range was 46 to 57 years with women comprising over one half of participants. The mean duration of disease varied from 2 to 12 months in the studies comparing Anti-TNF-therapies to csDMARD monotherapy with baseline Disease Activity Scores (DAS) ranging from a mean or median of 5.2 to 6.9 and Health Assessment Questionnaire Disability Index (HAQ) from 1.0 to 1.9. The mean disease duration was 4 to 6 months for combination therapy and the baseline mean DAS ranged from 2.5 to 5.6 and HAQ from 0.9 to 1.3. For the one RCT comparing Anti-TNF and non-TNF therapies, the mean disease duration ranged from 6.7 to 8.0 months with a baseline DAS of 6.2 and the median baseline HAQ of 1.7 to 1.8 (Donahue et al., 2018.

Overall, the results revealed that when Anti-TNF biologics were combined with Methotrexate (MTX), patients achieved a lower disease activity response (ACR50) and the strength of evidence was mostly moderate and low (ETN + MTX versus MTX RR, 1.49, 95% CI, 1.27 to 1.74; ADA + MTX versus MTX RR, 1.35, 95% CI, 1.15 to 1.59; CZP + MTX versus MTX RR, 1.20, 95% CI, 1.04 to 1.38) (Table 2). Infliximab plus MTX resulted in a significantly reduced disease activity response compared with csDMARD combination therapy. Network meta- analyses also indicated that the combination of Anti-TNF biologics plus MTX were favoured in comparison to biologic monotherapy. As such, the ACR50 response was significantly higher for Adalimumab plus MTX than Adalimumab monotherapy (RR, 1.52; 95% CI, 1.28 to 1.80; moderate evidence). The results were similar for the combination of Etanercept plus MTX versus Etanercept monotherapy (RR, 1.57; 95% CI, 1.23 to 2.02) (Donahue et al., 2018).

Table 2. ACR 50 Response of Anti-TNF Biologics Compared to Methotrexate for Early Arthritis

Intervention Comparison RR (95% CI) Strength of Evidence

ETN + MTX MTX 1.49 (1.27 to 1.74) Moderate IFX + MTX MTX Insufficient ADA + MTX MTX 1.35 (1.15 to 1.59) Low CZP + MTX MTX 1.20 (1.04 to 1.38) Low Source: Donahue et al., 2018

In addition, there were significantly higher remission rates for Anti-TNF biologics in combination with MTX compared to MTX monotherapy alone with a low level of evidence (ETN + MTX versus MTX RR, 1.77, 95% CI, 1.02, 3.08; ADA + MTX versus MTX RR, 2.04, 95% CI 1.15, 3.62; CZP & IFX) (Table 3). There was also a higher remission rate for Adalimumab plus MTX versus Adalimumab alone (RR, 1.87, 95% CI 1.06, 3.28 moderate evidence) (Donahue et al., 2018).

Table 3. Remission of Anti-TNF Biologics Compared to Methotrexate for Early Arthritis Intervention Comparison RR (95% CI) Strength of Evidence

ETN + MTX MTX 1.77 (1.02 to 3.08) Low ADA + MTX MTX 2.04 (1.15 to 3.62) Low Source: Donahue et al., 2018

The Anti-TNF biologics in combination with MTX produced significantly less radiographic progression (radiographic joint damage score) than MTX monotherapy (ETN + MTX versus MTX -0.81; 95% CI -0.98, -0.63; ADA + MTX versus MTX SMD, -0.99, 95% CI -1.17, -0.81; CZP + MTX versus MTX alone SMD, -0.38, 95% CI -0.53, -0.23) (mostly moderate and low SOE). There was also moderate evidence of significantly less radiographic progression for the Anti-TNF biologics in combination with MTX versus Anti-TNF monotherapy Adalimumab + MTX than Adalimumab alone (ETN + MTX versus ETN alone SMD -0.69; 95% CI -0.95, -0.4; ADA + MTX versus ADA alone SMD, -0.38; 95% CI -0.55, -0.21) and for Adalimumab monotherapy than MTX monotherapy (SMD, -0.61; 95% CI -0.78, -0.43) (Donahue et al., 2018).

The combinations of the Anti-TNFs Adalimumab (HAQ change -0.1 to -0.3 over 24 weeks to 2 years), Certolizumab pegol (HAQ change not consistently reported, but in favor of combination therapy, over 30 weeks to 1 year) and Infliximab (HAQ change not consistently reported, but in favor of combination therapy, over 30 weeks to 1 year) all plus MTX produced greater functional capacity than MTX monotherapy (all moderate or low SOE). The results for Etanercept were inconclusive (Donahue et al., 2018).

Evidence was insufficient to determine a difference in effectiveness for one biologic over another for ACR50 response, remission, radiographic changes or functional capacity (Donahue et al., 2018).

The systematic review for advanced Rheumatoid Arthritis included 98 unique RCTs (91 included in analysis), and 41 companion publications to determine the comparative clinical efficacy of different treatment options for Rheumatoid Arthritis (Wells et al., 2018). Of these, 61 studies were included to determine the efficacy of Anti-TNF biologics. Of the total number assessed for ROB (n=88), half were judged to have a high ROB and only 10 were considered to have a low ROB overall; the rest (39%) had an unclear ROB overall. The sample size of included studies ranged from 28 to 1,220 participants, with a median of 313 participants. The mean mean age of participants was 52.5 (range: 43.2 to 58.1) and 80.6% of the total number of participants were women (range: 43.3% to 100%). On average, participants had Rheumatoid Arthritis for 7.84 years (range: 0.94 to 13.0).

Etanercept (OR 3.95, 95% CrI 2.29 to 7.51), Infliximab (OR 3.00, 95% CrI 1.78 to 5.08), Adalimumab (OR 3.99, 95% CrI 2.84 to 5.62), Certolizumab pegol (OR 5.35, 95% CrI 3.42 to 8.67) and Golimumab (OR intravenous 2.90, 95% CrI 1.21 to 7.12; OR subcutaneous 6.00, 95% CrI 3.27 to 11.35) all produced greater ACR 50 responses when compared to MTX monotherapy (Table 4). The Anti-TNF biologics greater odds of achieving an ACR 50 response compared to a csDMARD in combination with MTX. Etanercept (OR 3.73, 95% CrI 1.98 to 7.04), Adalimumab (OR 3.78, 95% CrI 1.39 to 9.21), Golimumab (subcutaneous) (OR 5.67, 95% CrI 1.88 to 15.77), Certolizumab pegol (OR 5.05, 95% CrI 1.82 to 12.98) each in combination with MTX all achieved greater odds. With the exception of Infliximab, all the Anti-TNF biologics in combination with MTX produced a comparable ACR 50 response to csDMARD triple therapy. There were no significant differences in radiographic progression for any Anti-TNFs in combination with MTX compared to csDMARD double or triple therapy (Wells et al., 2018).

Table 4. ACR 50 Response of Anti-TNF Biologics Compared to Placebo plus Methotrexate for Late Arthritis Intervention Comparison OR (95% CrI)

ETN + MTX Placebo + MTX 3.95 (2.29 to 7.51) IFX + MTX Placebo + MTX 3.00 (1.78 to 5.08) ADA + MTX Placebo + MTX 3.99 (2.84 to 5.62)

CZP + MTX Placebo + MTX 5.35 (3.42 to 8.67)

GOL (SC) + MTX Placebo + MTX 6.00 (3.27 to 11.35)

GOL (IV) + MTX Placebo + MTX 2.90 (1.21 to 7.12)

Source: Wells et al., 2018

There were statistically significant higher odds of achieving remission among those who were treated with Anti-TNF biologics in combination with MTX compared to MTX monotherapy (ETN OR 2.86, 95% CrI 1.03 to 7.36; IFX OR 6.58, 95% CrI 1.98 to 23.28; ADA OR 8.87, 95%

CrI 3.57 to 27.29; CZP OR 9.85, 95% CrI 2.36 to 64.72; GOL subcutaneous OR 11.19, 95% CrI 4.19 to 33.16) (Table 5). The Anti-TNF biologics also produced more favourable odds of remission compared to a csDMARD plus MTX (ETN OR 4.73, 95% CrI 1.92 to 12.20; IFX OR 10.92, 95% CrI 1.89 to 72.19; ADA OR 14.96, 95% CrI 3.04 to 91.54; CZP OR 16.98, 95% CrI 2.35 to 158.00; GOL subcutaneous OR 18.85 95% CrI 3.58 to 103.80) (Wells et al., 2018).

Table 5. Remission of Anti-TNF Biologics Compared to Placebo plus Methotrexate for Late Arthritis Intervention Comparison OR (95% CrI)

ETN + MTX Placebo + MTX 2.86 (1.03 to 7.36) IFX + MTX Placebo + MTX 6.58 (1.98 to 23.28) ADA + MTX Placebo + MTX 8.87 (3.57 to 27.29)

CZP + MTX Placebo + MTX 9.85 (2.36 to 64.72)

GOL (SC) + MTX Placebo + MTX 11.19 (4.19 to 33.16)

GOL (IV) + MTX Placebo + MTX 18.85 (3.58 to 103.80)

Source: Wells et al., 2018

Certolizumab pegol with MTX, achieved a statistically significant improvement in the DAS 28 compared to MTX monotherapy (SMD –2.23, 95% CrI,–4.04 to –0.43). Infliximab (MD -0.24, 95% CrI –0.45 to –0.05), Adalimumab (MD –0.25, 95% CrI –0.34 to –0.15), Certolizumab pegol (MD –0.36, 95% CrI –0.50 to –0.22) and Golimumab subcutaneous and intravenous (MD –0.31, 95% CrI –0.43 to –0.19 and MD –0.28, 95% CrI –0.47 to –0.09, respectively) all in combination with MTX produced a significantly lower disability score compared to MTX monotherapy. The same Anti-TNF biologics also had higher physical health-related quality of life score compared to MTX monotherapy (IFX MD 4.58, 95% CrI 2.73 to 6.01; ADA MD 3.07 95% CrI, 0.73 to 5.45; Certolizumab pegol MD 5.07, 95% CrI 3.67 to 6.49; Golimumab subcutaneous MD 4.83, 95% CrI 3.03 to 6.76 and intravenous MD 3.65, 95% CrI 1.28 to 6.00). Intravenous Golimumab and Certolizumab pegol both in combination with MTX produced higher mental health-related quality of life than MTX monotherapy (MD 5.88, 95% CrI 2.18 to 9.71 and MD 3.60, 95% CrI 1.35 to 5.83, respectively). Patients treated with Etanercept, Adalimumab or Certolizumab pegol all in combination with MTX had lower pain scores than MTX monotherapy (SMD –0.72, 95% CrI –1.43 to –0.06, SMD –0.66, 95% CrI –1.15 to –0.30, SMD –1.58, 95% CrI –2.38 to –0.77, respectively). Certolizumab pegol with MTX produced a significantly lower fatigue score than MTX monotherapy (SMD 1.25, 95% CrI 0.17 to 2.36) (Wells et al., 2018).

Ankylosing Spondylitis One systematic review was included for Ankylosing Spondylitis which sought to assess the benefits and harms of Anti-TNF biologics in comparison to placebo, other drugs or usual care (Maxwell et al., 2015). Most included studies had low or unclear risk of bias. This review

contained 21 short-term (24-weeks or less) RCTs with a total of 3,308 participants. Participants were mostly male (65% to 80% in treatment groups, and 74% to 100% in control groups) and middle-aged (38 to 45 years in the treatment groups, and 39 to 47 years in the control groups). The mean disease duration in the treatment groups ranged from 8 to 16 years, and 10 to 17 years in the control groups.

Results from the systematic review suggest Anti-TNF biologics improve the clinical symptoms of Ankylosing Spondylitis. The Anti-TNF biologics were three to four times more likely to achieve an Assessment in SpondyloArthritis International Society (ASAS) 40 response by six months compared to placebo and the quality of evidence was high (ETN RR 3.31, 95% CrI 2.38 to 4.53; IFX RR 4.07, 95% CrI 2.80 to 5.74; ADA RR 3.53, 95% CrI 2.49 to 4.91; GOL RR 2.90, 95% CrI 1.90 to 4.23) (Table 6). The number needed-to-treat to receive this response ranged from 3 to 5. No significant difference were found for ASAS 40 response between the Anti-TNF biologics (Maxwell et al., 2015).

Table 6. ASAS 40 Response of Anti-TNF Biologics Compared to Placebo for Ankylosing Spondylitis Intervention Comparison RR (95% CrI) Strength of Evidence

ETN Placebo 3.31 (2.38 to 4.53) High IFX Placebo 4.07 (2.80 to 5.74) High ADA Placebo 3.53 (2.49 to 4.91) High

GOL Placebo 2.90 (1.90 to 4.23) High

Source: Maxwell et al., 2015

Moderate quality of evidence found that the Anti-TNF biologics were also significantly more likely than placebo to achieve ASAS partial remission (ETN RR 4.24, 95% CrI, 2.31, 8.09; IFX RR 15.41, 95% CrI, 5.09, 47.98; ADA RR 6.28, 95% CrI, 3.13, 12.78; GOL RR 5.18; 95% CrI, 1.90, 14.79) (Table 7) (Maxwell et al., 2015).

Table 7. ASAS Partial Remission of Anti-TNF Biologics Compared to Placebo for Ankylosing Spondylitis Intervention Comparison RR (95% CrI) Strength of Evidence

ETN Placebo 4.24 (2.31 to 8.09) Moderate IFX Placebo 15.41 (5.09 to 47.98) Moderate ADA Placebo 6.28 (3.13 to 12.78) Moderate

GOL Placebo 5.18 (1.90 to 14.79) Moderate

Source: Maxwell et al., 2015

In order to detect a minimally clinically important difference of 0.7 points for physical functioning, the number needed-to-treat ranged from 2 to 4. Etanercept (MD -1.09 SD -1.60 to - 0.56), Infliximab (MD -2.07 SD -2.71 to -1.35), Adalimumab (MD -1.57, SD -2.21 to -0.89) and Golimumab (MD -1.49 SD -2.27 to -0.69) all had significantly lower BASFI scores compared to placebo (high evidence). Low to moderate level evidence also suggested that the Anti-TNFs have a small impact on reducing spinal inflammation, however the clinical relevance was unclear (ADA 6% improvement, IFX 3% improvement and GOL 2.5% improvement) (Maxwell et al., 2015).

Juvenile Idiopathic Arthritis One systematic review was included for the treatment Juvenile Idiopathic Arthritis (Smith, 2017). This review included 100 full-text articles and conference abstracts which accounted for 67 distinct RCTs. However, there were 8 RCTs comparing Anti-TNF biologics.

This review found that the American College of Rheumatology (ACR) Pediatric (PEDI) 30 response was 1.91 (95% CrI, 1.28 to 2.59) times lower for patients treated with 0.4mg/kg of Etanercept compared to placebo. There were no significant differences between the Anti-TNF biologics and methotrexate in combination with placebo (Table 8) (Smith, 2017).

Table 8. ACR PEDI 30 Response Pairwise Comparisons for Juvenile Idiopathic Arthritis Intervention Comparison RR (95% CrI)

Placebo + MTX ETN (0.2mg/kg) 0.57 (0.24 to 1.37)

Placebo + MTX ETN (0.4mg/kg) 0.6 (0.26 to 1.06) ADA Placebo + MTX 1.33 (0.81 to 2.64)

ADA + MTX Placebo + MTX 1.44 (0.93 to 2.85)

Source: Smith, 2017

The number of active joints decreased for 0.2 mg/kg and 0.4 mg/kg of Etanercept (MD -11.23, 95% CrI -18.16 to -4.59 and MD -11.01, 95% CrI -14.59 to -7.52, respectively) and the number of joints with limited range of motion decreased for 0.4 mg/kg of Etanercept only (MD -5.15, CrI -9.5 to -0.8). There was some evidence that patients treated with Etanercept had a greater reduction in the number of active joints than those treated with the non-TNF biologic Abatacept (MD 8.17, 95% CrI 0.95 to 15.48 for 0.2mg/kg of ETN and MD 7.92, 95% CrI 3.51 to 12.36 for 0.4mg/kg) (Smith, 2017).

Crohn’s Disease One systematic review was included to compare the efficacy of therapies for induction and maintenance of remission in adult patients with Crohn’s disease (Hazlewood et al., 2015). A total of 15 trials were included that evaluated Anti-TNF therapy (Infliximab: 1 for induction and 2 for maintenance; Adalimumab: 4 for induction and 3 for maintenance; Certolizumab pegol: 4 for induction and 1 for maintenance) and 5 additional studies evaluated combination therapies with Infliximab. All but one study assessed remission using the Crohn’s disease Activity Index (CDAI) less than 150. Most of the included studies were assessed to have unclear risk of bias. Other limitations of this study have been identified in the literature including that different dosages of treatment were not factored into analyses and results were pooled for patients who were naive to biologics and for those failed past biologic treatment (Bonovas et al., 2015). Additional differences such as the varying classes of drugs, the outcome assessment measures and timing, the concomitant interventions and the varying baseline disease severities may further limit the applicability of the results (Bonovas et al., 2015). However, additional network meta- analyses have found similar effectiveness of Anti-TNFs against placebo in the induction and maintenance of remission for Crohn’s disease even after accounting for these differences. These studies were not included in this application given that the comparative effectiveness against conventional therapy was not conducted (Singh et al., 2018; Stidham et al., 2014; Singh et al., 2014). One additional study concluded that Anti-TNF monotherapy was the most effective therapy for post-operative prophylaxis, with large effect sizes relative to all other strategies including antibiotics, immundilators monotherapy, immundilators with antibiotics, budesonide (clinical relapse: RR, 0.02–0.20; endoscopic relapse: RR, 0.005–0.04) (Singh et al., 2015).

Compared to placebo, Infliximab (OR, 2.8 95% CrI 1.4–7.2), Infliximab plus azathioprine (OR 4.3 95% CrI 2.0–9.8) and Adalimumab (OR, 2.9 95% CrI 1.6–5.5) all had over a 99% probability of being superior at inducing remission in Crohn’s patients. These same drugs also proved to be superior than Azathioprine/6-mercaptopurine (OR, 2.3 95% CrI 1.3–5.0, OR 3.4 95% CrI 1.9-6.3, and OR 3.4 95% CrI 1.9-6.3) (Table 9). Infliximab plus azathioprine had 2.7 times more odds of inducing remission compared to methotrexate (95% CrI 1.9 - 6.3) (Hazlewood et al., 2015).

Table 9. Induction of Remission of Anti-TNF Biologics compared to Azathioprine/6- mercaptopurine for Crohn’s Disease Intervention Comparison OR (95% CrI)

IFX Azathioprine/6-mercaptopurine 2.3 (1.3 to 5.0) IFX + MTX Azathioprine/6-mercaptopurine 2.1 (0.67 to 7.9) IFX + Azathioprine Azathioprine/6-mercaptopurine 3.4 (1.9 to 6.3) ADA Azathioprine/6-mercaptopurine 2.4 (1.0 to 4.9) CZP Azathioprine/6-mercaptopurine 1.1 (0.58 to 2.0) Source: Hazlewood et al., 2015

For maintenance of remission, all Anti-TNFs drugs and combinations proved to be superior to placebo, with the exception of Infliximab plus methotrexate. Infliximab (OR 2.8, 95% CrI 1.8– 4.5), Infliximab plus azathioprine (OR 5.2, 95% CrI 2.8 - 11), Adalimumab (OR 5.1, 95% CrI 3.3–8.1) and Certolizumab (OR 2.0, 95% CrI 1.4–3.0) all had over 99% probability of being superior to placebo at achieving maintenance of remission for Crohn’s disease. Adalimumab, Infliximab and Infliximab plus azathioprine all had greater odds at achieving a maintenance of remission response compared to Azathioprine/6-mercaptopurine (OR, 2.9, 95% CrI 1.6–5.1, 1.6, 95% CrI 1.0–2.5 and OR 3.0, 95% CrI 1.7-5.5, respectively) (Table 10). Vedolizumab was inferior to both Infliximab plus azathioprine (OR 0.42, 95% CrI 0.17-0.92) and Adalimumab (OR, 0.42; 95% CrI, 0.22–0.85) (Hazlewood et al., 2015).

Table 10. Maintenance of Remission of Anti-TNF Biologics compared to Azathioprine/6- mercaptopurine for Crohn’s Disease Intervention Comparison OR (95% CrI)

IFX Azathioprine/6-mercaptopurine 1.6 (1.0 to 2.5)

IFX + MTX Azathioprine/6-mercaptopurine 1.5 (0.57 to 3.7)

IFX + Azathioprine Azathioprine/6-mercaptopurine 3.0 (1.7 to 5.5)

ADA Azathioprine/6-mercaptopurine 1.6 (1.0 to 2.5)

CZP Azathioprine/6-mercaptopurine 1.2 (0.65 to 1.9) Source: Hazlewood et al., 2015

Infliximab plus azathioprine as well as Adalimumab monotherapy demonstrated to have greater odds at inducing (OR 3.1 95% CrI 1.4–7.7 and OR 2.1 95% CrI 1.0-4.6, respectively) and maintaining (OR 2.6, 95% CrI 1.3-6.0 and OR 5.1, 95% CrI 3.3-8.1, respectively) remission relative to the Anti-TNF agent Certolizumab. Infliximab plus azathioprine had 2.6 times greater odds at maintaining remission compared to Infliximab monotherapy (OR 1.8, 95% CrI 1.0-3.8) (Hazlewood et al., 2015).

10. Review of harms and toxicity: summary of evidence on safety

Estimate of total patient exposure to date

Estimates on the total number of patients exposed to Anti-TNF biologics are currently not available. Anti-TNF biologics have been a primary focus for research and a comprehensive list of both short and long-term adverse effects have been well-established. For example, Etanercept alone has over four million patient-years of post-market exposure (Health Canada, 2018a).

Description of adverse effects/reactions and estimates of their frequency

It is of utmost importance to pool safety data across conditions in order to adequately understand the associated harms of drugs. A 2011 Cochrane Systematic Review assessed the potential adverse effects of Anti-TNF biologics Etanercept (39 RCTs), Infliximab (40 RCTs), Adalimumab (22 RCTs), Certolizumab pegol (6 RCTs) and Golimumab (8 RCTs) alone or in combination with other therapies (Singh et al., 2011). The median RCT duration was six months (range one to 63 months), the mean age of participants was 49.9 years and most were caucasian (85.4%). RCTs were included for the treatment of Rheumatoid Arthritis, Ankylosing Spondylitis, Crohn’s disease and others. Meta-analysis results are presented for each outcome measure in Table 11 below. All Anti-TNF biologics were compared to a placebo control.

Table 11. Safety Outcomes per Anti-TNF Biologic Network Meta-Analysis Results with Placebo Standard Meta-Analysis Results with control Placebo control

Treatment Serious Total Withdrawals Serious TB Lymphoma Congestive Adverse Adverse due to Infection Reactivation Heart Events Events Adverse Failure Events

Etanercept OR 1.24 OR 1.38 OR 1.30 OR 1.29 1.48 (0.06 to 2.40 (0.38 0.84 (0.05 (0.93 to (0.80 to (0.82 to 2.17) (0.72 to 36.93) to 15.31) to 14.26) 1.69 2.46) moderate 2.45) moderate high moderate Infliximab OR 1.15 OR 1.55 OR 2.34 OR 1.41 2.82 (0.65 to 3.00 (0.12 Not (0.85 to (1.01 to (1.40 to 4.14) (0.75 to 12.18) to 74.79) estimable 1.57) 2.35) moderate 2.62) moderate high moderate Adalimumab OR 0.96 OR 1.03 OR 1.35 OR 1.23 2.14 (0.33 to 0.95 (0.10 Not (0.74 to (0.67 to (0.82 to 2.22) (0.65 to 13.78) to 9.19) estimable 1.27) 1.54) moderate 2.40) moderate high moderate Certolizumab OR 1.57 OR 1.17 OR 1.32 OR 4.75 4.43 (0.50 to 0.33 (0.01 Not (1.06 to (0.71 to (0.69 to 2.69) (1.52 to 39.09) to 8.09) estimable 2.32) 1.95) moderate 18.45) moderate high high Golimumab OR 1.05 OR 1.24 OR 1.34 OR 1.11 3.04 (0.12 to Not 2.84 (0.11 (0.67 to (0.78 to (0.63 to 2.92) (0.45 to 75.13) estimable to 71.99) 1.69) 1.98) moderate 2.59) moderate high moderate *For moderate strength level of evidence, the 95% credible interval around the pooled effect includes both no effect and appreciable benefit or harm. Source: Singh et al., 2011

Compared to control, Certolizumab pegol was associated with a higher odds of serious adverse effects (OR 1.57, 95% CI 1.06 to 2.32) and serious infections (OR 4.75, 95% CI 1.52 to 18.45) and Infliximab was associated with higher odds of total adverse events (OR 1.55, 95% CI 1.01 to 2.35) and withdrawals due to adverse events (OR 2.34, 95% CI 1.40 to 4.14) (Singh et al., 2011).

Although uncommon, the most serious adverse reactions to Anti-TNFs, as indicated in the black boxed labels in product monographs/labels from the United States Food and Drug Administration (FDA) (2018a) and Health Canada (2018a), are: ● Serious infections (adult and pediatric) - Serious infections due to bacterial, mycobacterial, invasive fungal (disseminated or extrapulmonary histoplasmosis, aspergillosis, coccidiodomycosis), viral, parasitic, or other opportunistic infections have been reported. Sepsis, tuberculosis, candidiasis, legionellosis, listeriosis, and pneumocystis and hepatitis B relapse have also occurred. However, it is important to note that patients were treated with concomitant medications and their underlying disease itself could have posed the risk to infection. Treatment should not be initiated in patients with active infections and all patients should be up-to-date on their immunizations. ○ Etanercept: In controlled portions of trials for Rheumatoid Arthritis and Ankylosing Spondylitis, amongst others, the rate of serious infections was similar among placebo (0.8%), MTX monotherapy (3.6%) and Etanercept plus MTX (1.4%) (FDA, 2018a). ○ Infliximab: Clinical studies found that infections were reported in 36% of Infliximab patients, whereas infections were reported in 28% of placebo treated patients (Health Canada, 2018a). ○ Adalimumab: A total of 39 global controlled clinical trials found that the rate of infection for adult patients was 4.3 per 100-patient years in 7,973 Adalimumab patients compared to a rate of 2.9 per 100 in 4,848 control patients (FDA, 2018a). ○ Certolizumab pegol: The incidence of new infections in Rheumatoid Arthritis patients in controlled clinical studies was 0.91 per patient-year for those treated with Certolizumab pegol and 0.72 for placebo patients (FDA, 2018a; Health Canada, 2018a). There were more new cases of serious infections (i.e., tuberculosis, pneumonia, cellulitis and pyelonephritis) in the treatment group (0.06 per patient-year for Certolizumab Pegol versus 0.02 per patient- year for placebo). The incidence for Crohn’s patients was 38% for certolizumab pegol treated patients and 30% for placebo (FDA, 2018a). ○ Golimumab: Controlled portions of trials for Rheumatoid Arthritis and Ankylosing Spondylitis, amongst others, found that serious infections were observed in 1.4% of Golimumab patients compared to 1.3% of control patients (FDA, 2018a; Health Canada, 2018a). In Golimumab intravenous phase 3 trials, serious infections were 0.8% of treatment patients versus 0.4% of control (Health Canada, 2018a). Through week 16, the incidence of infections per 100 patient- years of follow-up was 5.7 (95% CI: 3.8, 8.2) for the Golimumab group and 4.2 (95% CI: 1.8, 8.2) for the placebo group (FDA, 2018a). Other controlled trials found that upper respiratory tract infections were the most common infection reported, as it occured in 12.6% of patients treated with Golimumab (incidence

per patient-year: 0.61; 95% CI: 0.55, 0.67) compared with 10.7% in control patients (incidence per patient-year: 0. 53; 95% CI: 0.44, 0.63) (FDA). ● Malignancies and/or Hepatosplenic T-cell lymphoma (adult and pediatric) - Very rare cases of cancer (e.g., skin, leukemia, breast, colorectal) have occured in patients and are more prominent in pediatric patients. People with more serious Rheumatoid Arthritis have up to a 2-fold higher risk, even in the absence of medication. There have also been cases of lymphoma and other malignancies. Very rare reports of Hepatosplenic T-cell lymphoma have been identified. However, several patients were also treated with concomitant medications (azathioprine or 6-mercaptopurine use for Crohn’s disease mostly in young adult males), thus the causal association is not clear. ○ Entarcept: In 6,543 rheumatology patients in controlled trials, the rate of lymphoma was 0.10 cases per patient years, 3-fold higher than the general population. Among 15,401 patients in open and controlled trials, the rate of leukemia was 0.03 per 100 patient-years and the rate of melanoma was 0.043 cases per 100 patient-years (FDA, 2018a; Health Canada, 2018a). ○ Infliximab: Of 5,780 patients treated with Infliximab in controlled and open-label trials, 5 patients developed lymphomas versus 0 out of 1600 control patients (Health Canada, 2018a). During controlled portions of trials, 14 out of 4019 Infliximab treated patients developed non-lymphoma malignancies (rate of 0.52/100 patient-year) versus 1 in 1,597 control patients (rate of 0.11/100 patient- years). ○ Adalimumab: 23 clinical trials of 8,764 patients found that the observed rate of lymphomas (95% CI) is 1.2 (0.9, 1.7) per 1,000 patient-years, a rate 3-fold higher than the general population (Health Canada, 2018a). ○ Certolizumab pegol: In controlled and open label trials, malignancies were observed at a rate 0.5 (95% CI 0.4, 0.7) per 100 patient-years among 4,650 Certolizumab pegol treated patients compared to a rate of 0.6 (95% CI 0.1, 1.7) per 100 patient-years among 1,319 placebo-treated patients (Health Canada, 2018a). ○ Golimumab: In phase 3 trials, the most frequently observed malignancies were basal cell carcinoma (19/2226, 0.9%), breast cancer (11/2226, 0.5%) and lung cancer (7/2226, 0.3%) (Health Canada, 2018a). During the controlled portions of the Phase 2 and Phase 3 trials for rheumatology conditions, the incidence of lymphoma per 100 patient-years of follow-up was 0.21 (95% CI: 0.03, 0.77) in the golimumab group compared with an incidence of 0 (95% CI: 0, 0.96) in the placebo group (FDA). There were no cases of lymphoma and malignancies in patients with Ankylosing Spondylitis treated with Intravenous Golimumab trials and the malignancies in patients with Rheumatoid Arthritis was similar to the general population (Health Canada, 2018a). ● Neurologic effects (adult and pediatric) (Adalimumab only) - Symptoms such as numbness or tingling, problems with vision, weakness in legs and dizziness. ○ Adalimumab: 21 controlled trials found the rate of new or exacerbated central nervous system and peripheral demyelinating diseases was less than 0.4 per 1,000 patient-years in 5,380 adalimumab treated patients and 0.7 per 1,000 patient-years in 3,337 control-treated patients (Health Canada, 2018a).

● Heart Failure (adult and pediatric) (Certolizumab pegol only) - Cases of worsening congestive heart failure and new onset have been reported. ○ Certolizumab pegol: 1 patient (0.1%) in placebo-controlled studies reported cardiac failure (0.11 per 100 patient-years) compared to no patients in the control group. For both controlled and open label studies, the incidence rate per 100 patient years of cardiac failure was 0.18 (Health Canada, 2018a). Available studies on the use of Anti-TNF biologics in pregnancy do not establish an association between their use and birth defects, but this has not been adequately studied (FDA, 2018a). These drugs are transferred across the placenta during the third semester of pregnancy and may affect immune response in the in-utero infant (FDA, 2018a). The exposure to a breastfed infant is estimated to be low. Generally, the safety of Anti-TNFs in pediatrics are similar to those of adults, but some adverse reactions (e.g., cancer) may be more common in children compared to their adult counterparts which may be due to concomitant medication. The safety of Certolizumab pegol in pediatrics has not been established (FDA, 2018a; Health Canada, 2018a). Adalimumab and Etanercept have not been studied for Juvenile Idiopathic Arthritis patients under 2 years or below 10 kg (FDA, 2018a). Adalimumab and Infliximab have not been studied for Crohn’s disease patients under 6 (FDA, 2018a). There may be more serious adverse events in geriatric populations who are taking Anti-TNFs, but it is noted that there is a higher incidence of infections and malignancies in this population to begin with (FDA, 2018a).

VigiAccess is the WHO Global Pharmacovigilance Database on various medicinal products collected by over 110 countries (VigiAccess 2018). The search identified 458,139 records for Etanercept, 141,148 for Infliximab, 417,265 for Adalimumab, 30,976 for Certolizumab pegol and 23,872 for Golimumab. The most common category for side-effects across all Anti-TNFs were general disorders and administration site conditions (ETN: 238,202, IFX: 39,709, ADA: 202,609, CZP: 12,369 and GOL: 8,205,). This category includes items such as injection site reactions (e.g., pain, erythema, swelling, bruising), drug ineffectiveness, pain, fatigue, malaise and pyrexia. Infections and infestations, musculoskeletal and connective tissue disorders, injury, poisoning and procedural complications and gastrointestinal disorders followed.

The FDA Adverse Events Reporting System provides an estimate of the total number of cases per Anti-TNF drug and per age group from 1998 to September 2018 (Figure 2 and 3, respectively) (Food and Drug Administration (FDA), 2018b). Out of all the total number cases, 431,678 were considered serious (including deaths) and 28,975 were death cases. The top five most commonly reported reactions for Anti-TNFs as a whole include injection site pain (9.2%), drug ineffective (7.36%), arthralgia (5.48%), injection site erythema (5.11%) and pain (4.42%). However, it is important to note that this report does not demonstrate causation, or in other words, that the drug caused the adverse event. Therefore, this data alone should not be used to identify the safety profile.

Figure 2. Estimated Total Number of Adverse Cases per Anti-TNF Drug According to FDA Adverse Events Reporting System

Source: FDA, 2018b

Figure 3. Estimated Total Number of Adverse Cases for Anti-TNF Drugs per Age Category

Source: FDA, 2018b

The Canada Vigilance adverse reactions online database provides information from 1965 to 2018. Without filtering the search criteria, 15,178 adverse reactions were reported for Etanercept, 51,170 for Infliximab, 29,474 for Adalimumab, 2,709 for Certolizumab pegol and 4,687 for Golimumab (Health Canada, 2018b).

The European Medicines Agency (2018a) reports the events for Anti-TNF biologics from December to present day. In this time frame, 89,382 events occurred from Etanercept, 75,273 from Infliximab, 52,299 from Adalimumab, 11,623 from Certolizumab pegol, and 10,440 from Golimumab.

Summary of available data (appraisal of quality, summary of results) and comparative safety

The characteristics of included studies are described above in the review of benefits section. The outcome measures and results for safety outcomes are described in detail for each disease below. Please see Appendix 2 for a complete summary of the data and comparative safety of Anti-TNF biologics against different treatment options.

Rheumatoid Arthritis The network meta-analysis for early Rheumatoid Arthritis found that there were no significant differences in serious adverse events or discontinuations attributable to adverse events between MTX monotherapy and any of the Anti-TNF biologics - Etanercept, Infliximab, Adalimumab, and Certolizumab pegol (low strength of evidence) (Donahue et al., 2018). Infliximab plus MTX also did not differ from csDMARD combination therapies (low strength of evidence). Anti-TNF therapy with a csDMARD did not differ significantly in serious adverse events or discontinuations attributable to adverse events compared to TNF biologic monotherapy (moderate strength of evidence). There was insufficient evidence to determine whether there was a difference in adverse events between older and younger patients taking Etanercept.

The systematic review for advanced Rheumatoid Arthritis found that there were no significant differences in serious adverse events or withdrawals attributable to adverse events between the Anti-TNF biologics (Etanercept, Adalimumab, Infliximab, Certolizumab pegol and subcutaneous Golimumab) in combination with MTX and MTX monotherapy (Wells et al., 2018). However, Etanercept in combination with MTX had lower odds of withdrawals attributable to adverse events compared to a csDMARD in combination with MTX (OR 0.33, CrL 0.11 to 0.89). There was insufficient evidence to detect any differences in Anti-TNF treatment comparisons for mortality, serious infections, tuberculosis, cancer, leukemia, lymphoma, congestive heart failure, major adverse cardiac events, and herpes zoster. A pairwise meta-analysis found no statistically significant difference in mortality for Infliximab combined with MTX and MTX monotherapy. Additional pairwise meta-analyses found that there were no differences in serious infections for patients treated with the Anti-TNFs Etanercept, Infliximab, Golimumab plus MTX versus MTX alone. The analysis for Adalimumab had insufficient evidence to demonstrate serious infections in Adalimumab plus MTX versus MTX monotherapy. A pooled estimate from two trials comparing Etanercept monotherapy and MTX combination therapy, found that were no significant differences in cancer and a pairwise meta-analysis found no significant differences between Infliximab plus MTX and MTX groups.

Ankylosing Spondylitis The pooled results for all Anti-TNF biologics demonstrate a moderate level of evidence that there is an increased risk of withdrawals due to adverse events compared to placebo (Peto OR 2.44, 95% CI 1.26 to 4.72; total events: 38/1637 in biologic group; 7/986 in placebo), with an absolute increase of 1% (95% CI 0% to 2%) (Maxwell et al., 2015). However, the results did not detect a difference in risk for serious adverse events (Peto OR 1.45, 95% CI 0.85 to 2.48). Entercept (25 and 50 mg) was the only Anti-TNF biologic that had an individual increase in withdrawals due to adverse events versus placebo (RR 3.65, 95% CI 1.27 to 11.79) with an absolute increased harm of 2% (95% CrI 0.2% to 8%). The effect of Etanercept compared to placebo for serious adverse events was uncertain. There was no certainty reported for adverse effects or withdrawals due to effect between either Adalimumab, Golimumab and Infliximab to placebo. The strength of evidence was moderate for all safety outcomes.

Juvenile Idiopathic Arthritis Although a network-meta analysis could be not conducted, the systematic review for Juvenile Idiopathic Arthritis found that biologics were safe in short-term use among both polyarticular course and active systemic patients (Smith, 2017). For polyarticular course, one RCT found that there were no serious adverse effects or withdrawals due to adverse effects occurred for high or low doses of Etanercept. Another RCT found no withdrawals due to adverse events occurred for Adalimumab with or without methotrexate and few withdrawals due to adverse events.

Crohn’s Disease Infliximab combined with azathioprine and Adalimumab monotherapy had >99% lower odds of total withdrawals compared to placebo (OR 0.27, 95% CrI 0.08 – 0.72 and OR 0.43, 95% CrI 0.26 - 0.69) (Hazlewood et al., 2015). Withdrawals for Infliximab and azathioprine combination therapy had significantly less odds than Azathioprine/6-mercaptopurine (OR 0.39, 95% CrI 0.14–0.98) and methotrexate (OR 0.29, 95% CrI 0.07–0.93). Adalimumab only produced less withdrawals compared to methotrexate (OR 0.45, 95% CrI 0.18–1.0).

Infliximab (OR 2.7, 95% CrI 1.6–4.7) and Infliximab plus azathioprine (OR 3.2, 95% CrI 1.6– 6.1) had significantly greater odds of withdrawals due to adverse events compared to placebo. Adalimumab had over a 99% probability of having less withdrawals due to adverse events than placebo (OR, 0.48, 95% CrI 0.31–0.74). The Anti-TNF biologics Certolizumab and Adalimumab had significantly less odds of withdrawals due to adverse events compared to Azathioprine/ 6- mercaptopurine (OR 0.23, 95% CrI 0.13–0.40 and OR 0.12, 95% CrI 0.06–0.24, respectively) and Methotrexate (OR 0.07, 95% CrI 0.01–0.28 and 0.04, 95% CrI 0.00–0.16, respectively). Infliximab monotherapy had significantly lower odds of withdrawals due to adverse events compared to methotrexate (OR 0.21, 95% CrI 0.02–0.93) (Hazlewood et al., 2015).

Anti-TNF comparisons indicate that Adalimumab and Infliximab plus azathioprine have significantly lower odds of total withdrawals than Certolizumab (OR 0.50, 95% CrI 0.24–0.96 and OR 0.32, 95% CrI 0.09–0.94, respectively). Adalimumab had lower odds of withdrawals due to adverse events than Certolizumab (OR 0.55, 95% CrI 0.32–0.93) and Infliximab (OR 0.18, 95% CrI 0.09–0.34). However, Infliximab plus azathioprine had significantly greater odds of withdrawals due to adverse events than Certolizumab (OR 3.6, 95% CrI 1.7–7.5) and as did

Infliximab monotherapy (OR 3.1, 95% CrI 1.7–5.8). Infliximab plus azathioprine had greater odds than Adalimumab (OR 6.5, 95% CrI 3.0–14) (Hazlewood et al., 2015).

Identification of variation in safety that may relate to health systems and patient factors

Given that several of the Anti-TNF biologics are administered by patients themselves subcutaneously, it is possible that proper protocols are not followed during injection. For example, poor hand hygiene or unsanitary practices may increase the risk for infections. Before initiating Anti-TNF therapy, patients must be screened for either active or latent tuberculosis infection. It is possible that appropriate screening tests (i.e., tuberculin skin test and chest X-ray) are not adequately performed in all patients. Likewise, patients may be unknown carriers of Hepatitis B or Tuberculosis and not monitored closely for infection throughout and months after therapy. Live vaccines should not be given concurrently, as they may also increase infections.

The concurrent administration of non-TNF and Anti-TNF biologics result in increased incidences of serious adverse events (e.g., malignancies and lymphoma). Prescribers are advised to exercise caution when using Anti-TNFs in patients with central or peripheral nervous system demyelinating disorders and heart conditions. It is possible that special care is not taken and the use of Anti-TNF biologics may exacerbate clinical symptoms. Thorough assessments are recommended prior to therapy for those who may be at high risk.

11. Summary of available data on comparative cost and cost- effectiveness of the medicine

The debate revolving around biologic treatment is whether the superior outcomes, as described above, outweigh the expensive monetary costs. However, future increases in inflammatory condition spending by insurers and payers will be driven by the increased utilization of Anti- TNF biologics.

Only economic evaluations that compared two or more Anti-TNF biologics against a comparator were included. The results below are presented for condition separately.

Prices for Anti-TNF Biologics were not available on international drug comparator guide. Therefore, the Anti-TNF prices from a range of settings are provided in Tables 12 - 16 below (Australian Government Department of Health The Pharmaceutical Benefits Scheme, 2018; Ontario Ministry of Health and Long-Term Care, 2018; Centers for Medicare and Medicaid Services, 2016; National Health Services, 2018).

Table 12. Costs of Etanercept Country Product Max quantity Cost

Australia General Schedule: 2 packs, 2 units, 3-5 DPMQ: $1049.55 Etanercept 25 mg repeats General Patient injection [4 vials] (&) Charge: $39.50 inert substance

diluent [4 x 1 mL syringes], 1 pack

S100 HSD Public: 1 pack, 1 unit, 0 DPMQ: $469.63 Etanercept 25 mg repeats General Patient injection [4 vials] (&) Charge: $39.50 inert substance diluent [4 x 1 mL syringes], 1 pack

S100 HSD Private: 1 pack, 1 unit, 0 DPMQ: $495.71 Etanercept 25 mg repeats General Patient injection [4 vials] (&) Charge: $39.50 inert substance diluent [4 x 1 mL syringes], 1 pack

General Schedule: 1 pack, 1 unit, 3-5 DPMQ: $1049.54 Etanercept 50 mg/mL repeats General Patient injection, 4 x 1 mL Charge: $39.50 syringes or injection devices

S100 HSD Public: 1 pack, 1 unit, 0 DPMQ: $939.25 Etanercept 50 mg/mL repeats General Patient injection, 4 x 1 mL Charge: $39.50 syringes or injection devices

S100 HSD Private: 1 pack, 1 unit, 0 DPMQ: $984.11 Etanercept 50 mg/mL repeats General Patient injection, 4 x 1 mL Charge: $39.50 syringes or injection devices or injection devices

Canada (Ontario) Enbrel 25mg/Vial Inj Drug benefit or unit Pd-Vial Pk price: 202.9300 Amount MOHLTC pays: 202.9300

Enbrel 50mg/mL Inj Drug benefit or unit Pref Syr or Enbrel price: 405.9850 SureClick 50mg/mL Amount MOHLTC Pref AutoInj pays: 405.9850

25 mg/vial, 50 mg Reimbursed through

prefilled syringe for the Exceptional subcutaneous Access Program injection (EAP)

United States Injection, etanercept, Average Spending (Medicare) 25 mg (Part B) Per Dosage Unit: $380.51 (Part B), Etanercept (Part D) $972.08 (Part D) Average Sales Price: $380.99 (Part B) Average Spending Per Claim: $1617.18 (Part B), $4563.82 (Part D)

Etanercept 4 Basic Price: 35750 United Kingdom 25mg/0.5ml solution for injection pre-filled syringes

Etanercept 50mg/1ml 4 Basic Price: 71500 solution for injection pre-filled syringes Sources: Australian Government Department of Health The Pharmaceutical Benefits Scheme, 2018; Ontario Ministry of Health and Long-Term Care, 2018; Centers for Medicare and Medicaid Services, 2016; National Health Services, 2018

Table 13. Costs of Infliximab Country Product Max quantity Cost

Australia S100 HSD Private: 1 pack, 1 unit, 0-1 DPMQ: $473.84 infliximab 100 mg repeat General Patient injection, 1 vial Charge: $39.50

S100 HSD Public: 5, packs, 5 units, 1 DPMQ: $2243.05 infliximab 100 mg repeat General Patient injection, 1 vial Charge: $39.50

S100 HSD Public: 1 pack, 1 unit, 0 DPMQ: $448.61 infliximab 100 mg repeat General Patient injection, 1 vial Charge: $39.50

Repatriation 1 pack, 1 unit, 2 DPMQ: $504.26 Pharmaceutical repeats General Patient Benefits: Charge: $6.40

infliximab 100 mg injection, 1 vial

Canada (Ontario) 100mg/10mL Reimbursed through intravenous infusion the Exceptional or 100 mg/vial Access Program (EAP)

United States Injection infliximab, Average Spending (Medicare) 10 mg (Part B) Per Dosage Unit: $80.18 (Part B), Infliximab (Part D) $812.27 (Part D) Average Sales Price: $81.66 (Part B) Average Spending Per Claim: $3935.71 (Part B), $5627.42 (Part D)

United Kingdom Pricing information not available

Sources: Australian Government Department of Health The Pharmaceutical Benefits Scheme, 2018; Ontario Ministry of Health and Long-Term Care, 2018; Centers for Medicare and Medicaid Services, 2016; National Health Services, 2018

Table 14. Costs of Adalimumab Country Product Max quantity Cost

Australia General Schedule: 1 pack, 2 unit, 0-5 DPMQ: $1269.60 Adalimumab 20 repeats General Patient mg/0.4 mL injection, Charge: $39.50 2 x 0.4 mL syringes or 40 mg/0.8 mL injection, 2 x 0.8 mL cartridges or syringes

S100 HSD Public: 1 pack, 2 unit, 0 DPMQ: $1151.87 Adalimumab 20 repeats General Patient mg/0.4 mL injection, Charge: $39.50 2 x 0.4 mL syringes or 40 mg/0.8 mL injection, 2 x 0.8 mL cartridges or syringes

S100 HSD Private: 1 pack, 2 unit, 0 DPMQ: $1199.16 Adalimumab 20 repeats General Patient

mg/0.4 mL injection, Charge: $39.50 2 x 0.4 mL syringes or 40 mg/0.8 mL injection, 2 x 0.8 mL cartridges or syringes

General Schedule: 1 pack, 1 unit, 2-5 DPMQ: $2454.79 Adalimumab 40 repeats General Patient mg/0.8 mL injection, Charge: $39.50 4 x 0.8 mL cartridges

General Schedule: 1 pack, 1 unit, 0 DPMQ: $3606.66 Adalimumab 40 repeats General Patient mg/0.8 mL injection, Charge: $39.50 6 x 0.8 mL cartridges or syringes

Canada (Ontario) Humira 40mg/0.8mL Drug benefit or unit Inj Sol-Pref Pen Pk or price: $769.9700 Pref Syr Pk Amount MOHLTC pays: $769.9700

Reimbursed through the Exceptional Access Program (EAP)

United States Injection, Average Spending (Medicare) adalimumab, 20 mg Per Dosage Unit: (Part B) $706.69 (Part B), $1965.01 Adalimumab (Part D) Average Sales Price: $815.10 (Part B) Average Spending Per Claim: $1675.12 (Part B), $4894.75 (Part D)

United Kingdom Adalimumab 2 Basic Price: 70428 40mg/0.4ml solution for injection pre-filled disposable devices or pre-filled syringes Sources: Australian Government Department of Health The Pharmaceutical Benefits Scheme, 2018; Ontario Ministry of Health and Long-Term Care, 2018; Centers for Medicare and Medicaid Services, 2016; National Health Services, 2018

Table 15. Costs of Certolizumab pegol Country Product Max quantity Cost

Australia General Schedule: 1 pack, 2 units, 2-5 DPMQ: $1014.61 certolizumab pegol repeats General Patient 200 mg/mL injection, Charge: $39.50 2 x 1 mL syringes or injection devices

General Schedule: 3 packs, 6 units, 0 DPMQ: $3007.41 certolizumab pegol repears General Patient 200 mg/mL injection, Charge: $39.50 2 x 1 mL syringes or injection devices

Canada (Ontario) 200 mg/mL prefilled Reimbursed through syringe the Exceptional Access Program (EAP)

United States Injection, Average Spending (Medicare) certolizumab pegol, 1 Per Dosage Unit: mg (Part B) $6.65 (Part B), $3448.44 (Part D) Certolizumab pegol Average Sales Price: (Part D) $6.78 (Part B) Average Spending Per Claim: $2543.74 (Part B), $4236.05 (Part D)

United Kingdom Cimzia 200mg/1ml Discount not solution for injection deducted pre-filled syringes Sources: Australian Government Department of Health The Pharmaceutical Benefits Scheme, 2018; Ontario Ministry of Health and Long-Term Care, 2018; Centers for Medicare and Medicaid Services, 2016; National Health Services, 2018

Table 16. Costs of Golimumab Country Product Max quantity Cost

Australia General Schedule: 1 pack, 1 unit, 3-5 DPMQ: $1318.17 golimumab 50 mg/0.5 repeats General Patient mL injection, 0.5 mL Charge: $39.50 syringe or injection device or 100 mg/mL

injection, 1 mL injection device

Canada (Ontario) 50 mg/0.5mL Pre- Reimbursed through Filled Syringe Or the Exceptional Auto-Injector, 100 Access Program mg/ mL Pre-filled (EAP) Syringe or Auto- Injector

United States Injection, golimumab, Average Spending (Medicare) 1 mg, for intravenous Per Dosage Unit: use (Part B) $24.20 (Part B), $7195.58 Golimumab (subcutaneous) and $370.73 (intravenous) (Part D) Average Sales Price: $24.62 (Part B) Average Spending Per Claim: $4238.06 (Part B), $4571.67 (subcutaneous) and $5424.48 (intravenous) (Part D)

United Kingdom Golimumab 1 Basic Price: 152594 100mg/1ml solution for injection pre-filled disposable devices

Golimumab 1 Basic Price: 76297 50mg/0.5ml solution for injection pre-filled disposable devices

Golimumab 1 Basic Price: 76297 50mg/0.5ml solution for injection pre-filled syringes Sources: Australian Government Department of Health The Pharmaceutical Benefits Scheme, 2018; Ontario Ministry of Health and Long-Term Care, 2018; Centers for Medicare and Medicaid Services, 2016; National Health Services, 2018

Rheumatoid Arthritis

It is estimated that the total one year costs for treating Rheumatoid Arthritis in a commercially insured United States population is $14,385, $14,601, $15,997 and $19,283 for Etanercept, Adalimumab, Golimumab and Infliximab, respectively (Curtis et al., 2015). The percentage of patients categorized as effectively treated, that is those who did not fail any treatments in 6 months, were 32.7% for Etanercept ($43,935 per effectively treated patient), 32.3% for golimumab ($49,589 per effectively treated patient), 27.7% for Adalimumab ($52,752 per effectively treated patient) and 19.0% for Infliximab ($101,402 per effectively treated patient) (Curtis et al., 2015). Previous systematic reviews have suggested that biologics may be cost- effective at a willingness-to-pay threshold between $50,000–100,000 per QALY in patients with insufficient treatment to csDMARD (Joensuu et al., 2015). Determining the cost-effectiveness of biologic pharmacological treatments has been identified as a top research priority for Rheumatoid Arthritis (Van der Velde et al., 2011).

A systematic review of 18 studies found that median incremental cost-effectiveness ratios (ICER) by biologic therapy were above $100,000 CAD/QALY (ADA: $111,000/QALY [range $106,000–$235,000/QALY], ETN: $124,000/QALY [range $62,000–$141,000/QALY], ADA + MTX: $127,000/QALY [range $78,000–$354,000/QALY], ETN + MTX: $105,000/QALY [range $63,000–$207,000/QALY], and IFX + MTX: $142,000/QALY [range $100,000– $169,000/QALY]) (Van der Velde et al., 2011). The same review found that the ICERs for biologic treatments in early Rheumatoid Arthritis were considerably lower (CAD range $75,000– $91,000/QALY) than advanced Rheumatoid Arthritis (CAD range $134,000–$378,000/QALY), and were cost-effective at a willingness-to-pay threshold of $100,000/QALY. The authors concluded that the most cost‐ effective approach for managing Rheumatoid Arthritis is to treat with a DMARD early, move to other DMARDs and if nonresponse continues, add a biologic before the late stages of Rheumatoid Arthritis occur. These findings parallel those of Schoels et al., (2010) in that Anti-TNF therapy appears most cost-effective only after DMARD failure and when used early in combination with methotrexate.

An independent economic evaluation using the Birmingham Rheumatoid Arthritis Model was carried out to determine the cost-effectiveness of Adalimumab, Etanercept and Infliximab in treating Rheumatoid Arthritis in the National Health Service (NHS) after conventional agents including methotrexate and have failed (Malottki et al., 2011). The HAQ DI was used in this model as an indicator for disease severity. Results revealed that compared with conventional DMARDs the ICERs were £34,300 (per QALY) for Adalimumab (0.02 £20,000/QALY and 0.30 £30,000/QALY), £38,800 for Etanercept (0.00 £20,000/QALY and 0.17 £30,000/QALY), £36,200 for Infliximab (0.02 £20,000/QALY and 0.24 £30,000/QALY) (Malottki et al., 2011). An earlier evaluation using the same simulation model found that the ICERs for Anti-TNFs were most cost-effective when used last in treatment (Chen et al., 2006). The base-case ICER was approximately £30,000 per QALY in early Rheumatoid Arthritis and £50,000 per QALY in late Rheumatoid Arthritis. As such, the ICER for Etanercept, Adalimumab and Infliximab was £24,000, £30,000 and £38,000 per QALY, respectively. When these drugs were used as first line monotherapy the ICERs were much higher for Adalimumab (£53,000) and Etanercept (£49,000). When used in combination with methotrexate for first line, the ICERs grew even higher (Chen et al., 2006). A US medicare cost-effective analysis predicted that Infliximab ($94,029) generates costs much higher than those for Etanercept ($81,181) or Adalimumab ($79,535) (Wailoo et al., 2008). In base-case scenario, the probability of cost-effectiveness at $60,000 was 0.00 for

Infliximab, 0.41 for Etanercept and 0.59 for Adalimumab and for 6 month biologic response was 0.00, 0.38 and 0.62, respectively (Wailoo et al., 2008). Another evaluation determined that the 6- month cost-per responder was $US 27,853 for Adalimumab (95% CI $19,284, $40,270) and $US 29,140 for Etanercept (95% CI $14,170, $61,030) (Liu et al., 2012).

Ankylosing Spondylitis

Corbett and colleagues (2016) developed a de novo decision model to assess the cost- effectiveness of five Anti-TNF biologics, Certolizumab, Golimumab, Adalimumab, Etanercept and Infliximab, versus conventional therapy (NSAIDs) for the treatment of Ankylosing Spondylitis. The analysis was based on costs from NHS and Personal Social Services and used a lifetime horizon of 60 years. Two rebound scenarios were considered for when patients fail to initial therapy: i) Rebound equal to gain where BASDAI and BASFI deteriorates the same amount by which it improves when the patient responded to therapy and, ii) Rebound back to conventional care where BASFI deteriorates to the level and subsequent trajectory it would have been had the patient had not initially responded to therapy. It is likely that a patient would fall somewhere in between these two scenarios.

Findings revealed that in the rebound equal to gain scenario, the ICER of Golimumab, Adalimumab, Etanercept, Certolizumab, Infliximab were £21,079, £21,170, £21,577, £23,133 and £40,576, respectively per additional QALY. Infliximab had the lowest probability of being cost-effective at a £20,000 (0.001) and £30,000 (0.089) per QALY threshold. On the other hand, Golimumab, Adalimumab and Etanercept had the highest probability of being cost-effective at £20,000 (0.427, 0.423 and 0.402, respectively) and £30,000 (0.841, 0.839 and 0.826, respectively) per QALY threshold. Certolizumab fell shortly behind at having a probability of being cost-effective at £20,000 (0.299) and £30,000 (0.761) (Corbett et al., 2016).

In the rebound conventional care scenario, the ICER of Golimumab, Adalimumab, Etanercept, Certolizumab, Infliximab were £36,554, £36,695, £37,322, £39,693 and £66,529, respectively per additional QALY. Infliximab still had the lowest probability of being cost-effective at a £20,000 (0.000) and £30,000 (0.001) per QALY threshold. Likewise, Golimumab, Adalimumab and Etanercept had the highest probability of being cost-effective at £20,000 (0.019, 0.017 and 0.017, respectively) and £30,000 (0.299, 0.293 and 0.275, respectively) per QALY threshold. Certolizumab fell shortly behind at having a probability of being cost-effective at £20,000 (0.011) and £30,000 (0.203) (Corbett et al., 2016).

The findings of Corbett et al. (2016) were similar to those of McLeod et al. (2007). As such, McLeod concluded that the ICERs of Etanercept and Adalimumab fell below an assumed willingness-to-pay threshold of £30,000 per Quality Adjusted Life Year (QALY) (Golimumab and Certolizumab were not analyzed). However, infliximab did not fall below the threshold, making it a less favourable treatment option (McLeod et al., 2007). Other comparative models suggest that the cost-effectiveness of Infliximab is less favourable due to the acquisition and administration of the drug (Kirchhoff et al., 2012).

Juvenile Idiopathic Arthritis

The 2008 total drug costs for the treatment of Juvenile Idiopathic Arthritis in Canada for Infliximab 3–5 mg/kg, Etanercept, and Adalimumab were $18,330 ($2,034 drug, preparation and administration cost per infusion), $18,966 ($170 drug cost per infusion) and $18,654 ($668 drug cost per infusion), respectively (Ungar et al., 2010). These costs were generated using a 40 kg child and included concomitant medications and monitoring.

A recent cost–utility decision-analytic model was developed using a Markov approach, to determine the cost-effectiveness of two Anti-TNF biologics, Adalimumab and Etanercept, versus methotrexate for the treatment of Juvenile Idiopathic Arthritis (Shepherd et al., 2016). The model incorporated disease flares to estimate the clinical effectiveness of treatment. The analysis was based on NHS and Personal Social Services and used a time horizon of 30 years and discount rates of 3.5% for costs and health benefits. Results for the base-case analysis, where it was assumed that no patients entered clinical remission off treatment, indicated that the ICER versus methotrexate monotherapy was £38,127 for Adalimumab and £32,256 for Etanercept per QALY gained.

Another study determined the cost-effectiveness of Anti-TNF biologic therapy in patients unresponsive to DMARDs (Ungar et al., 2010). At one year, the mean incremental Canadian costs per additional ACR Pedi 30 were $26,061 (95% CI $17,070, $41,834), $46,711 (95% CI $30,042, $75,787), and $31,209 (95% CI $16,659, $66,220) for Etanercept, Adalimumab, and Infliximab, respectively (Ungar et al., 2010). The same study found that if an individual was willing to pay $23,000 to gain one additional ACR Pedi 30 responder, the probability of cost- effectiveness would be equal (50%) for both Etanercept and methotrexate. That is, the monetary value of health improvement exceeds the cost compared to standard care with methotrexate. The willingness‐ to‐ pay points at which Adalimumab and Infliximab had a 50% probability of cost‐ effectiveness were $45,000 for and $27,500, respectively (Ungar et al., 2010). The 95% probability of Etanercept to demonstrate a net economic benefit would occur if an individual was willing to pay no more than $30,000 to gain a responder (Ungar et al., 2010).

Crohn’s Disease

One recent systematic review assessed the literature on the cost-effectiveness of Infliximab, Adalimumab, Golimumab and Certolizumab pegol for Inflammatory Bowel Disease (Huoponen et al., 2015). In total 17 studies were included in the review for Crohn’s Disease and all studies were conducted in North America or in Europe. The willingness-to-pay threshold for biologics in the treatment of Crohn’s Disease is estimated to be 35,000 €/QALY. Compared to conventional drugs treating fistulizing Crohn’s disease in patients with no previous treatment, the ICER ascended in excess of 400,000 €/QALY. Infliximab was dominant for newly diagnosed luminal Crohn’s disease. For patients with previous conventional treatment the ICERs for the biologics ranged from dominance to 549,335 €/QALY in comparison to conventional medical treatment. Adalimumab produced lower ICERs than Infliximab when both were compared to conventional medical treatment. However, when Infliximab was compared to Adalimumab, ICERs above 300,000 €/QALY were produced and Adalimumab was dominant in maintenance treatment. The use of Anti-TNF biologics for induction treatment resulted in lower ICERs than maintenance treatment. The ICER was estimated to be between 11,725 to 947,769 €/QALY using a lifetime horizon. The authors concluded that Anti-TNF biologics are considered cost-effective for the

induction of remission in severe Crohn’s Disease patients, while maintenance remains unclear. Similarly, a Markov model found that both Adalimumab and Infliximab were cost-effective, compared to maintenance treatment and standard care, for induction treatment in for thresholds up to £100,000 per QALY (Dretzke et al., 2011). Both were cost-effective options for severe disease, but only Adalimumab was cost-effective for moderate. Neither were cost-effective for maintenance therapy (Dretzke et al., 2011). These conclusions mirrored those of Tang et al. (2013).

Other research has compared the cost-effectiveness of initiating Infliximab or Adalimumab at < 2 years of Crohn’s diagnosis versus starting it later (Beilman et al., 2018). Results indicate that early initiation provides more cost-savings than late initiation. As such, the initiation of Infliximab costs $628,603 (CAD) for a gain of 12.23 QALYs. Early initiation resulted in a lifetime cost savings of $50,418 less compared than late initiation and an incremental effectiveness of 0.72 QALYs. The early initiation of Adalimumab costs $472,612 (CAD) for a gain of 11.92 lifetime QALYs. Early initiation resulted in a lifetime cost savings of $43,969 less compared than late initiation and an incremental effectiveness of 0.54 QALYs. Early initiation of Infliximab and Adalimumab had a 73.9 and 73.6% of being cost effective at a willingness-to-pay threshold of $50,000 per QALY, respectively. Therefore, the benefits of starting Anti-TNF therapy early extend well-beyond the health benefits as they also have the potential reduce the economic burden.

The mean cost per infusion of Infliximab (including administration) was £1,691 United Kingdom, compared with £1430 for the cost of 8‐ weeks’ treatment with adalimumab and £43 for standard care (Bodger et al., 2009). A Canadian cost-utility analysis determined that the costs for Crohn’s Disease usual care, adalimumab, and infliximab strategies were $17,107, $45,480 and $54,084, respectively (Blackhouse et al., 2012). The cost-utility of adalimumab compared to usual care was $193,305 per QALY and the incremental cost-utility of infliximab compared to adalimumab was $451,165 per QALY. The willingness-to-pay per QALY was between $193,305 and $451,164 for Adalimumab was $451,165 or higher. A cost-per responder analysis determined that at 12-month the cost per responder for Crohn’s disease patients was $116,291 (95% CI $71,637, $208,348) for Adalimumab and $125,169 (95% CI $60,532, $267,101) for Infliximab (Liu et al., 2012).

12. Summary of regulatory status and market availability of the medicines

The summary of the regulatory status and market availability of each Anti-TNF biologic in various settings is presented in Table 17.

Table 17. Regulatory Status and Market Availability of the Anti-TNF Biologics

Drug Generic US Food and European Australian Japanese Health Canada Name (Trade Drug Medicines Government, Pharmaceutical Name) Administration Agency Department of s and Medical (FDA) (EMA) Health, Devices Agency Therapeutic Goods Administration Etanercept Availability Availability Availability Not approved Availability (Enbrel) 1) 25 mg/0.5ml 1) 10 mg 1) 25 mg/0.5 ml 1) 50 mg/ml 2) 25 mg 2) 25mg/0.5ml (not marketed) 3) 25 mg 3) 50 mg/ml 2) 25 mg 2) 25 mg 3) 50 mg 3) 25 mg/ml Administration Administration 4) 50 mg/ml 4) 50 mg Subcutaneous Subcutaneous 5) 50 mg/ml Administration Administration Status Status Subcutaneous Subcutaneous Prescription for Prescription for Rheumatoid Rheumatoid Status Status Arthritis, Arthritis, Prescription Prescription for Juvenile Juvenile for Rheumatoid Idiopathic Idiopathic Rheumatoid Arthritis, Arthritis, Arthritis, Arthritis, Juvenile Ankylosing Ankylosing Juvenile Idiopathic Spondylitis Spondylitis Idiopathic Arthritis, Arthritis, Ankylosing First approved First approved Ankylosing Spondylitis 2000 1998 Spondylitis First approved First approved 2000 2000

* Lifmior is also approved (2017) as a generic of the centrally authorized product of Enbrel

Infliximab Availability Availability Availability Availability Availability (Remicade) 1) 100 mg 1) 100 mg 1) 100 mg 1) 100 mg 1) 100 mg

Administration Administration Administration Administration Administration Intravenous Intravenous Intravenous Intravenous Intravenous

Status Status Status Status Status Prescription for Prescription Prescription for Prescription for Prescription for adult Crohn’s for adult adult Crohn’s adult Crohn’s adult Crohn’s disease, pediatric Crohn’s disease, disease, disease, Crohn’s disease, disease, pediatric Rheumatoid pediatric Rheumatoid pediatric Crohn’s Arthritis, Crohn’s Arthritis, Crohn’s disease, Ankylosing disease, Ankylosing disease, Rheumatoid Spondylitis Rheumatoid Spondylitis Rheumatoid Arthritis, Arthritis, Arthritis, Ankylosing First approved Ankylosing First approved Ankylosing Spondylitis 2015 Spondylitis 1998 Spondylitis First approved First approved First approved 2000 2001 1999 Adalimumab Availability Availability Availability Availability Availability (Humira) 1) 80mg/0.8 ml 1) 20 mg/0.2 1) 80mg/0.8 ml 1) 40 mg/0.8 ml 1) 80mg/0.8 ml 2) 40 mg/0.8 ml ml 2) 40 mg/0.8 ml 2) 40 mg/0.4 ml 2) 40 mg/0.8 ml 3) 40 mg/0.4 ml 2) 40 mg/0.4 3) 40 mg/0.4 ml 3) 80mg/0.8 ml 3) 40 mg/0.4 ml 4) 20 mg/0.4 ml ml 4) 20 mg/0.4 ml 4) 20 mg/0.2 ml 5) 20 mg/0.2 ml 3) 40 mg/0.8 5) 20 mg/0.2 ml Administration 6) 10 mg/0.1 ml 6) 10 mg/0.2 ml ml 6) 10 mg/0.2 ml Subcutaneous 7) 10 mg/0.1 ml 4) 80 mg/0.8 Administration ml Administration Status Subcutaneous Administration Subcutaneous Prescription for Subcutaneous Administration adult Crohn’s Status Subcutaneous Status disease, Prescription for Status Prescription for Rheumatoid adult Crohn’s Prescription for Status adult Crohn’s Arthritis, disease, adult Crohn’s Prescription disease, Juvenile pediatric disease, pediatric for adult pediatric Idiopathic Crohn’s Crohn’s disease, Crohn’s Crohn’s Arthritis, disease, Rheumatoid disease, disease, Ankylosing Rheumatoid Arthritis, pediatric Rheumatoid Spondylitis Arthritis, Juvenile Crohn’s Arthritis, Juvenile Idiopathic disease, Juvenile First approved Idiopathic Arthritis, Rheumatoid Idiopathic 2016 Arthritis, Ankylosing Arthritis, Arthritis, Ankylosing Spondylitis Juvenile Ankylosing Spondylitis Idiopathic Spondylitis First approved Arthritis, First approved 2002 Ankylosing First approved 2004 Spondylitis 2003

First approved 2003

Certolizumab Availability Availability Availability Not approved Availability pegol 1) 200 mg 1) 200 mg/ml 1) 200 mg/ml 1) 200 mg/ml (Cimzia) 2) 200 mg/ml Administration Administration Administration Administration Subcutaneous Subcutaneous Subcutaneous Subcutaneous Status Status Status Status Prescription Prescription for Prescription for Prescription for for Rheumatoid Rheumatoid adult Crohn’s Rheumatoid Arthritis, Arthritis, disease, Arthritis, Ankylosing Ankylosing Rheumatoid Ankylosing Spondylitis Spondylitis Arthritis, Spondylitis Ankylosing First approved First approved Spondylitis First approved 2010 2009 2009 First approved 2008 Golimumab Availability Availability Availability Not approved Availability (Simponi & 1) Strength: 50 1) Strength: 50 1) Strength: 50 1) Strength: 50 Simponi Aria) mg/0.5ml mg/0.5ml mg/0.5ml mg/0.5ml 2) 100 mg/ml 2) 100 mg/ml 2) 100 mg/ml 2) 100 mg/ml

Administration Administration Administration Administration Subcutaneous Subcutaneous Subcutaneous Subcutaneous

Availability Status Status Availability 1) 50 mg/4.0 ml Prescription Prescription for 1) 50 mg/4.0 ml for Rheumatoid Administration Rheumatoid Arthritis, Administration Intravenous Arthritis, Ankylosing Intravenous Ankylosing Spondylitis Status Spondylitis, Status Prescription for Juvenile First approved Prescription for Rheumatoid Idiopathic 2009 Rheumatoid Arthritis, Arthritis Arthritis, Ankylosing Ankylosing Spondylitis First approved Spondylitis 2009 First approved First approved 2009 2011 Sources: FDA, 2018a; European Medicines Agency, 2018b; Australian Government Department of Health Therapeutic Goods Administration, 2018; Japanese Pharmaceuticals and Medical Devices Agency, 2018; Health Canada, 2018b

13. Availability of pharmacopoeia standards (British Pharmacopoeia, International Pharmacopoeia, United States Pharmacopoeia, European Pharmacopeia)

The Anti-TNF medicinal biological products are not mentioned in The International Pharmacopeia - Eighth Edition, 2018

For the European Pharmacopeia, Etanercept and Infliximab are in the index; the other anti-TNF are not.

We did not have access to the other Pharmacopoeia.

References

Aletaha D, Neogi T, Silman AJ, Funovits J, Felson DT, Bingham III CO, Birnbaum NS, Burmester GR, Bykerk VP, Cohen MD, Combe B. 2010 rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis & Rheumatism. 2010 Sep;62(9):2569-81.

Aliling J, Brent LH. Criteria and Disease Activity Measures in Axial Spondyloarthropathies. 2016.

Arnett FC, Edworthy SM, Bloch DA, Mcshane DJ, Fries JF, Cooper NS, Healey LA, Kaplan SR, Liang MH, Luthra HS, Medsger Jr TA. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis & Rheumatism: Official Journal of the American College of Rheumatology. 1988 Mar;31(3):315-24.

American College of Gastroenterology. Crohn’s disease. 2011. Available from http://patients.gi.org/topics/crohns-disease/

American College of Rheumatology (ACR). American College of Rheumatology (ACR) Juvenile Idiopathic Arthritis Guideline. 2017 Jun.

Australian Government Department of Health The Pharmaceutical Benefits Scheme. Pharmaceutical benefits scheme (PBS). 2018. Available from http://www.pbs.gov.au/pbs/home

Australian Government Department of Health Therapeutic Goods Administration. 2018. Available from https://www.tga.gov.au/.

Beilman CL, Kirwin E, Ma C, McCabe C, Fedorak RN, Halloran B. Early Initiation of Tumor Necrosis Factor Antagonist-based Therapy for Patients with Crohn’s Disease Reduces Costs Compared With Late Initiation. Clinical Gastroenterology and Hepatology. 2018 Jul 26.

Benchimol EI, Bernstein CN, Bitton A, Carroll MW, Singh H, Otley AR, Vutcovici M, El- Matary W, Nguyen GC, Griffiths AM, Mack DR. Trends in epidemiology of pediatric inflammatory bowel disease in Canada: distributed network analysis of multiple population- based provincial health administrative databases. The American journal of gastroenterology. 2017 Jul;112(7):1120.

Beukelman T, Patkar NM, Saag KG, Tolleson‐ Rinehart S, Cron RQ, DeWitt EM, Ilowite NT, Kimura Y, Laxer RM, Lovell DJ, Martini A. 2011 American College of Rheumatology recommendations for the treatment of juvenile idiopathic arthritis: initiation and safety monitoring of therapeutic agents for the treatment of arthritis and systemic features. Arthritis care & research. 2011 Apr;63(4):465-82.

Blackhouse G, Assasi N, Xie F, Marshall J, Irvine EJ, Gaebel K, Campbell K, Hopkins R, O’reilly D, Tarride JE, Goeree R. Canadian cost-utility analysis of initiation and maintenance

treatment with anti-TNF-α drugs for refractory Crohn's disease. Journal of Crohn's and Colitis. 2012 Feb 1;6(1):77-85.

Bodger K, Kikuchi T, Hughes D. Cost‐ effectiveness of biological therapy for Crohn’s disease: Markov cohort analyses incorporating United Kingdom patient‐ level cost data. Alimentary pharmacology & therapeutics. 2009 Aug;30(3):265-74.

Bonovas S, Moja L, Danese S. In the presence of conceptual heterogeneity, results of network meta-analysis comparing therapies in Crohn's disease need to be interpreted with caution. Gastroenterology. 2015 Jun 1;148(7):1483-4.

Braun J, Davis J, Dougados M, Sieper J, van der Linden S, van der Heijde D. First update of the international ASAS consensus statement for the use of anti-TNF agents in patients with ankylosing spondylitis. Annals of the rheumatic diseases. 2006 Mar 1;65(3):316-20.

Bykerk VP, Akhavan P, Hazlewood GS, Schieir O, Dooley A, Haraoui B, Khraishi M, LeClercq SA, Légaré J, Mosher DP, Pencharz J. Canadian Rheumatology Association recommendations for pharmacological management of rheumatoid arthritis with traditional and biologic disease- modifying antirheumatic drugs. The Journal of rheumatology. 2012 Aug 1;39(8):1559-82.

Centers for Medicare and Medicaid Services. CMS drug spending Medicare Part B and D programs. 2016. Available from https://www.cms.gov/Research-Statistics-Data-and- Systems/Statistics-Trends-and-Reports/Information-on-Prescription-Drugs/index.html

Chen YF, Jobanputra P, Barton P, Jowett SM, Bryan S, Clark W, Fry-Smith A, Burls A. A systematic review of the effectiveness of adalimumab, etanercept and infliximab for the treatment of rheumatoid arthritis in adults and an economic evaluation of their cost-effectiveness. Health technology assessment. 2006 Nov 1;10(42):1-266.

Corbett M, Soares M, Jhuti G, Rice S, Spackman E, Sideris E, Moe-Byrne T, Fox D, Marzo- Ortega H, Kay L, Woolacott N. Tumour necrosis factor-α inhibitors for ankylosing spondylitis and non-radiographic axial spondyloarthritis: a systematic review and economic evaluation. Health Technology Assessment. 2016 Feb 1;20(9):333.

Cornec D, Varache S, Morvan J, Devauchelle-Pensec V, Berthelot JM, Le Henaff-Bourhis C, Hoang S, Martin A, Chalès G, Jousse-Joulin S, Saraux A. Comparison of ACR 1987 and ACR/EULAR 2010 criteria for predicting a 10-year diagnosis of rheumatoid arthritis. Joint Bone Spine. 2012 Dec 1;79(6):581-5.

Chaudhari K, Rizvi S, Syed BA. Rheumatoid arthritis: current and future trends. 2016 Apr 15.

Crohn’s and Colitis Canada. 2018 impact of inflammatory bowel disease in Canada. 2018.

Cross M, Smith E, Hoy D, Carmona L, Wolfe F, Vos T, Williams B, Gabriel S, Lassere M, Johns N, Buchbinder R. The global burden of rheumatoid arthritis: estimates from the global burden of disease 2010 study. Annals of the rheumatic diseases. 2014 Feb 18.

Curtis JR, Chastek B, Becker L, Quach C, Harrison DJ, Yun H, Joseph GJ, Collier DH. Cost and effectiveness of biologics for rheumatoid arthritis in a commercially insured population. Journal of managed care & specialty pharmacy. 2015 Apr;21(4):318-29.

Dignass A, Van Assche G, Lindsay JO, Lémann M, Söderholm J, Colombel JF, Danese S, D'Hoore A, Gassull M, Gomollón F, Hommes DW. The second European evidence-based consensus on the diagnosis and management of Crohn's disease: current management. Journal of Crohn's and Colitis. 2010 Feb 1;4(1):28-62.

Donahue KE, Gartlehner G, Schulman ER, Jonas B, Coker-Schwimmer E, Patel SV, Weber RP, Lohr KN, Bann C, Viswanathan M. Drug Therapy for Early Rheumatoid Arthritis: A Systematic Review Update. 2018 Jul.

Dretzke J, Edlin R, Round J, Connock M, Hulme C, Czeczot J, Fry-Smith A, McCabe C, Meads C. A systematic review and economic evaluation of the use of tumour necrosis factor-alpha (TNF-α) inhibitors, adalimumab and infliximab, for Crohn's disease. Health Technology Assessment (Winchester, England). 2011 Feb;15(6):1.

DynaMed Plus. Ipswich (MA): EBSCO Information Services 1995 - 2018. 2018. Available from http://www.www.dynamed.com. Registration and login required.

European Medicines Agency. EurdaVigilance - European database of suspected adverse drug reaction reports. 2018a. Available from http://www.adrreports.eu/en/disclaimer.html

European Medicines Agency. Medicines. 2018b. Available from https://www.ema.europa.eu/en/medicines.

Express Scripts Canada. Drug trend report 2016.

Food and Drug Administration (FDA). Drugs @ FDA: FDA approved drug products. 2018a. Available from https://www.accessdata.fda.gov/scripts/cder/daf/.

Food and Drug Administration (FDA). FDA Adverse Event Reporting System (FAERS) Public Dashboard. 2018b. Available from https://www.fda.gov/drugs/guidancecomplianceregulatoryinformation/surveillance/adversedruge ffects/ucm070093.htm

Feuerstein JD, Nguyen GC, Kupfer SS, Falck-Ytter Y, Singh S. American gastroenterological association institute clinical guidelines committee. American Gastroenterological Association Institute Guideline on Therapeutic Drug Monitoring in inflammatory bowel disease. Gastroenterology. 2017;153:827-34.

Fransen J, Moens HB, Speyer I, Van Riel PL. Effectiveness of systematic monitoring of rheumatoid arthritis disease activity in daily practice: a multicentre, cluster randomised controlled trial. Annals of the rheumatic diseases. 2005 Sep 1;64(9):1294-8.

Hazlewood GS, Rezaie A, Borman M, Panaccione R, Ghosh S, Seow CH, Kuenzig E, Tomlinson G, Siegel CA, Melmed GY, Kaplan GG. Comparative effectiveness of immunosuppressants and biologics for inducing and maintaining remission in Crohn's disease: a network meta-analysis. Gastroenterology. 2015 Feb 1;148(2):344-54.

Health Canada. Drug product database. 2018a. Available from https://health- products.canada.ca/dpd-bdpp/index-eng.jsp.

Health Canada. Canada vigilance adverse reaction online database 1965-2018 Jul 3. 2018b. Available from https://www.canada.ca/en/health-canada/services/drugs-health- products/medeffect-canada/adverse-reaction-database.html

Holroyd CR, Seth R, Bukhari M, Malaviya A, Holmes C, Curtis E, Chan C, Yusuf MA, Litwic A, Smolen S, Topliffe J. The British Society for Rheumatology biologic DMARD safety guidelines in inflammatory arthritis-Executive summary. Rheumatology (Oxford, England). 2018 Aug 21.

Huoponen S, Blom M. A systematic review of the cost-effectiveness of biologics for the treatment of inflammatory bowel diseases. PLoS One. 2015 Dec 16;10(12):e0145087.

Japanese Pharmaceuticals and Medical Devices Agency. Review reports: Drugs. 2018. Available from https://www.pmda.go.jp/english/review-services/reviews/approved- information/drugs/0001.html

Joensuu JT, Huoponen S, Aaltonen KJ, Konttinen YT, Nordström D, Blom M. The cost- effectiveness of biologics for the treatment of rheumatoid arthritis: a systematic review. PloS one. 2015 Mar 17;10(3):e0119683.

Kirchhoff TD, Mittendorf T, Schmidt RE, Jablonka A, Merkesdal S. Cost–effectiveness of TNF- α inhibition in active ankylosing spondylitis: a systematic appraisal of the literature. Expert review of pharmacoeconomics & outcomes research. 2012 Jun 1;12(3):307-17.

Kroon FPB, van der Burg LRA, Ramiro S, Landewé RBM, Buchbinder R, Falzon L, van der Heijde D. Non-steroidal anti-inflammatory drugs (NSAIDs) for axial spondyloarthritis (ankylosing spondylitis and non-radiographic axial spondyloarthritis). Cochrane Database of Systematic Reviews 2015 , Issue 7 . Art. No.: CD010952. Lichtenstein GR, Loftus EV, Isaacs KL, Regueiro MD, Gerson LB, Sands BE. ACG Clinical Guideline: management of Crohn’s disease in adults. The American journal of gastroenterology. 2018 Apr;113(4):481.

Liu Y, Wu EQ, Bensimon AG, Fan CP, Bao Y, Ganguli A, Yang M, Cifaldi M, Mulani P. Cost per responder associated with biologic therapies for Crohn’s disease, psoriasis, and rheumatoid arthritis. Advances in therapy. 2012 Jul 1;29(7):620-34.

Ma C, Beilman CL, Huang VW, Fedorak DK, Kroeker KI, Dieleman LA, Halloran BP, Fedorak RN. Anti-TNF therapy within 2 years of Crohn's disease diagnosis improves patient outcomes: a retrospective cohort study. Inflammatory bowel diseases. 2016 Jan 28;22(4):870-9.

Malottki K, Barton P, Tsourapas A, Uthman AO, Liu Z, Routh K, Connock M, Jobanputra P, Moore D, Fry-Smith A, Chen YF. Adalimumab, etanercept, infliximab, rituximab and abatacept for the treatment of rheumatoid arthritis after the failure of a tumour necrosis factor inhibitor: a systematic review and economic evaluation. 2011 Mar 15.

Manners PJ, Bower C. Worldwide prevalence of juvenile arthritis why does it vary so much?. The Journal of rheumatology. 2002 Jul 1;29(7):1520-30.

Maaser C, Sturm A, Vavricka SR, Kucharzik T, Fiorino G, Annese V, Calabrese E, Baumgart DC, Bettenworth D, Borralho Nunes P, Burisch J. ECCO-ESGAR Guideline for Diagnostic Assessment in Inflammatory Bowel Disease. Journal of Crohn's and Colitis. 2018 Aug 23.

Maxwell LJ, Zochling J, Boonen A, Singh JA, Veras MM, Ghogomu ET, Jandu MB, Tugwell P, Wells GA. TNF‐ alpha inhibitors for ankylosing spondylitis. Cochrane Database of Systematic Reviews. 2015, Issue 3, Art. No.: CD005468.

McLeod C, Bagust A, Boland A, Dagenais P, Dickson R, Dundar Y, Hill RA, Jones A, Mota RM, Walley T. Adalimumab, etanercept and infliximab for the treatment of ankylosing spondylitis: a systematic review and economic evaluation. Gray Pub.; 2007 Aug 1.

Monaco C, Nanchahal J, Taylor P, Feldmann M. Anti-TNF therapy: past, present and future. International immunology. 2014 Nov 19;27(1):55-62.

National Health Services (NHS). NHS prescription services. 2018 Nov 1. Available from https://www.nhsbsa.nhs.uk/sites/default/files/2018- 10/Drug%20Tariff%20November%202018.pdf

National Institute for Health and Care Excellence (NICE). Crohn’s disease: management. NICE guideline.; 2012 Oct 10.

National Institute for Health and Care Excellence (NICE). Rheumatoid arthritis in adults: management. NICE guideline.; 2018 Jul 11.

National Institute for Health and Care Excellence (NICE). Spondyloarthritis in over 16s: diagnosis and management.; 2017 Feb 28.

Ng SC, Shi HY, Hamidi N, Underwood FE, Tang W, Benchimol EI, Panaccione R, Ghosh S, Wu JC, Chan FK, Sung JJ. Worldwide incidence and prevalence of inflammatory bowel disease in the 21st century: a systematic review of population-based studies. The Lancet. 2017 Dec 23;390(10114):2769-78.

Ontario Ministry of Health and Long-Term Care. Drugs funded by Ontario drug benefit (ODB) program: E-formulary. 2018. Available from https://www.formulary.health.gov.on.ca/formulary/

Raychaudhuri SP, Deodhar A. The classification and diagnostic criteria of ankylosing spondylitis. Journal of autoimmunity. 2014 Feb 1;48:128-33.

Ravelli A. Disease classification. In Handbook of Juvenile Idiopathic Arthritis 2016 (pp. 17-24). Adis, Cham.

Ringold S, Weiss PF, Beukelman T, DeWitt EM, Ilowite NT, Kimura Y, Laxer RM, Lovell DJ, Nigrovic PA, Robinson AB, Vehe RK. 2013 update of the 2011 American College of Rheumatology recommendations for the treatment of juvenile idiopathic arthritis: recommendations for the medical therapy of children with systemic juvenile idiopathic arthritis and tuberculosis screening among children receiving biologic medications. Arthritis & Rheumatism. 2013 Oct;65(10):2499-512.

Rohekar S, Chan J, Shirley ML, Haroon N, Chandran V, Bessette L, Mosher D, Flanagan C, Keen KJ, Adams K, Mallinson M. 2014 Update of the Canadian Rheumatology Association/Spondyloarthritis Research Consortium of Canada treatment recommendations for the management of spondyloarthritis. Part II: specific management recommendations. The Journal of rheumatology. 2015 Feb 15:jrheum-141001.

Royal College of Nursing (RCN). Assessing, managing and monitoring biologic therapies for inflammatory arthritis: guidance for rheumatology practitioners. Musculoskeletal Care 2003;1:135-40.

Rudan I, Sidhu S, Papana A, Meng SJ, Xin–Wei Y, Wang W, Campbell–Page RM, Demaio AR, Nair H, Sridhar D, Theodoratou E. Prevalence of rheumatoid arthritis in low–and middle–income countries: A systematic review and analysis. Journal of global health. 2015 Jun;5(1).

Sadowski DC, Bernstein CN, Bitton A, Croitoru K, Fedorak RN, Griffiths A. Canadian Association of Gastroenterology Clinical Practice Guidelines: The use of tumour necrosis factor- alpha antagonist therapy in Crohn’s disease. Canadian Journal of Gastroenterology and Hepatology. 2009;23(3):185-202.

Schoels M, Wong J, Scott DL, Zink A, Richards P, Landewé R, Smolen JS, Aletaha D. Economic aspects of treatment options in rheumatoid arthritis: a systematic literature review informing the EULAR recommendations for the management of rheumatoid arthritis. Annals of the rheumatic diseases. 2010 May 1:annrheumdis126714.

Shepherd J, Cooper K, Harris P, Picot J, Rose M. The clinical effectiveness and cost- effectiveness of abatacept, adalimumab, etanercept and tocilizumab for treating juvenile idiopathic arthritis: a systematic review and economic evaluation. 2016 Apr 20.

Sieper J, Rudwaleit M, Baraliakos X, Brandt J, Braun J, Burgos-Vargas R, Dougados M, Hermann KG, Landewe R, Maksymowych W, Van Der Heijde D. The Assessment of SpondyloArthritis international Society (ASAS) handbook: a guide to assess spondyloarthritis. Annals of the rheumatic diseases. 2009 Jun 1;68(Suppl 2):ii1-44.

Singh JA, Saag KG, Bridges Jr SL, Akl EA, Bannuru RR, Sullivan MC, Vaysbrot E, McNaughton C, Osani M, Shmerling RH, Curtis JR. 2015 American College of Rheumatology guideline for the treatment of rheumatoid arthritis. Arthritis & rheumatology. 2016 Jan;68(1):1- 26.

Singh JA, Wells GA, Christensen R, Ghogomu ET, Maxwell LJ, MacDonald JK, Filippini G, Skoetz N, Francis DK, Lopes LC, Guyatt GH. Adverse effects of biologics: a network meta‐ analysis and Cochrane overview. Cochrane database of systematic reviews. 2011(2).

Singh S, Fumery M, Sandborn WJ, Murad MH. Systematic review and network meta‐ analysis: first‐ and second‐ line biologic therapies for moderate‐ severe Crohn's disease. Alimentary Pharmacology & Therapeutics. 2018 Jun 19.

Singh S, Garg SK, Pardi DS, Wang Z, Murad MH, Loftus Jr EV. Comparative efficacy of biologic therapy in biologic-naive patients with Crohn disease: a systematic review and network meta-analysis. InMayo Clinic Proceedings 2014 Dec 1 (Vol. 89, No. 12, pp. 1621-1635). Elsevier.

Smith CA, Toupin-April K, Jutai JW, Duffy CM, Rahman P, Cavallo S, Brosseau L. A systematic critical appraisal of clinical practice guidelines in juvenile idiopathic arthritis using the appraisal of guidelines for research and evaluation II (AGREE II) instrument. PLoS One. 2015 Sep 10;10(9):e0137180.

Smith C. Efficacy and Safety of Pharmacological and Non-Pharmacological Interventions in Juvenile Idiopathic Arthritis: A Series of Systematic Reviews and Network Meta-Analyses (Doctoral dissertation, Université d'Ottawa/University of Ottawa). 2017.

Smolen JS, Landewé R, Bijlsma J, Burmester G, Chatzidionysiou K, Dougados M, Nam J, Ramiro S, Voshaar M, van Vollenhoven R, Aletaha D. EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2016 update. Annals of the rheumatic diseases. 2017 Jun 1;76(6):960-77.

Stidham RW, Lee TC, Higgins PD, Deshpande AR, Sussman DA, Singal AG, Elmunzer BJ, Saini SD, Vijan S, Waljee AK. Systematic review with network meta‐ analysis: the efficacy of anti‐ TNF agents for the treatment of Crohn's disease. Alimentary pharmacology & therapeutics. 2014 Jun;39(12):1349-62.

Stoll ML, Cron RQ. Treatment of juvenile idiopathic arthritis: a revolution in care. Pediatric rheumatology. 2014 Apr 23;12(1):1.

Stolwijk C, van Onna M, Boonen A, van Tubergen A. Global Prevalence of Spondyloarthritis: A Systematic Review and Meta‐ Regression Analysis. Arthritis care & research. 2016 Sep;68(9):1320-31.

Stolwijk C, van Tubergen A, Castillo-Ortiz JD, Boonen A. Prevalence of extra-articular manifestations in patients with ankylosing spondylitis: a systematic review and meta-analysis. Annals of the rheumatic diseases. 2015 Jan 1;74(1):65-73.

Sýkora J, Pomahačová R, Kreslová M, Cvalínová D, Štych P, Schwarz J.Current global trends in the incidence of pediatric-onset inflammatory bowel disease. World journal of gastroenterology. 2018 Jul 7;24(25):2741.

Tang DH, Harrington AR, Lee JK, Lin M, Armstrong EP. A systematic review of economic studies on biological agents used to treat Crohn’s disease. Inflammatory bowel diseases. 2013 Jun 20;19(12):2673-94.

Terdiman JP, Gruss CB, Heidelbaugh JJ, Sultan S, Falck–Ytter YT. American Gastroenterological Association Institute guideline on the use of thiopurines, methotrexate, and anti–TNF-a biologic drugs for the induction and maintenance of remission in inflammatory Crohn’s disease. Gastroenterology. 2013 Dec;145(6):1459-63.

Thierry S, Fautrel B, Lemelle I, Guillemin F. Prevalence and incidence of juvenile idiopathic arthritis: a systematic review. Joint Bone Spine. 2014 Mar 1;81(2):112-7.

Ungar WJ, Costa V, Hancock‐ Howard R, Feldman BM, Laxer RM. Cost‐ effectiveness of biologics in polyarticular‐ course juvenile idiopathic arthritis patients unresponsive to disease‐ modifying antirheumatic drugs. Arthritis care & research. 2011 Jan 1;63(1):111-9. van der Heijde D, Ramiro S, Landewé R, Baraliakos X, Van den Bosch F, Sepriano A, Regel A, Ciurea A, Dagfinrud H, Dougados M, van Gaalen F. 2016 update of the ASAS-EULAR management recommendations for axial spondyloarthritis. Annals of the rheumatic diseases. 2017 Jan 13:annrheumdis-2016. van der Velde G, Machado M, Ieraci L, Witteman W, Bombardier C, Krahn M. Cost‐ effectiveness of biologic response modifiers compared to disease‐ modifying antirheumatic drugs for rheumatoid arthritis: A systematic review. Arthritis care & research. 2011 Jan 1;63(1):65-78.

VigiAccess. VigiBase adverse reactions database. 2018. Available from http://www.vigiaccess.org/

Wailoo AJ, Bansback N, Brennan A, Michaud K, Nixon RM, Wolfe F. Biologic drugs for rheumatoid arthritis in the Medicare program: a cost‐ effectiveness analysis. Arthritis & Rheumatism. 2008 Apr;58(4):939-46.

Ward MM, Deodhar A, Akl EA, Lui A, Ermann J, Gensler LS, Smith JA, Borenstein D, Hiratzka J, Weiss PF, Inman RD. American College of Rheumatology/Spondylitis Association of America/Spondyloarthritis Research and Treatment Network 2015 recommendations for the treatment of ankylosing spondylitis and nonradiographic axial spondyloarthritis. Arthritis & Rheumatology. 2016 Feb;68(2):282-98.

World Gastroenterology Organisation (WGO). Inflammatory bowel disease. Aug 2015. Available from http://www.worldgastroenterology.org/UserFiles/file/guidelines/inflammatory- bowel-disease-english-2015-update.pdf.

World Health Organization (WHO). Chronic rheumatic conditions. 2018a. Available from https://www.who.int/chp/topics/rheumatic/en/.

World Health Organization (WHO). Musculoskeletal conditions. Feb 15 2018b. Available from http://www.who.int/news-room/fact-sheets/detail/musculoskeletal-conditions.

World Health Organization. WHO guidelines on the pharmacological treatment of persisting pain in children with medical illnesses. 2012. Available from http://apps.who.int/iris/bitstream/handle/10665/44540/9789241548120_Guidelines.pdf?sequence =1.

Wells GA, Smith C, Hossain A, Karsh J, Singh J, Hazlewood G, Tugwell P. Drugs for the Management of Rheumatoid Arthritis: Clinical Evaluation. 2018 Mar.

Yazici Y, Shi N, John A. Utilization of Biologic Agents in Rheumatoid Arthritis in the United States. Bulletin of the NYU hospital for joint diseases. 2008 Apr 1;66(2).

Appendix 1 Tables 1 to 22 portray the summary of findings for Anti-TNF effectiveness outcomes with strength of evidence grades, if available.

Table 1. Effectiveness of Etanercept in the Treatment of Early Rheumatoid Arthritis Specific Outcome Result Strength of Evidence

Disease activity: Significantly improved response for TNF Moderate: Downgraded because Response biologic (ETN) + MTX than MTX alone medium level of study (RR, 1.49; 95% CI, 1.27 to 1.74) limitations

Disease activity: Significantly improved response for TNF Response biologic (ETN) + MTX than ETN alone (RR, 1.57; 95% CI, 1.23, 2.02)

Disease activity: Significantly higher for ETN vs. MTX Moderate: Downgraded because Response medium level of study limitations

Disease activity: TNF biologic (ETN) compared with non- Insufficient: Downgraded Response TNF biologic (RIT) because no ITT analysis; large CIs cross appreciable benefits or harms; and not enough events to meet optimal information size

Disease activity: Significantly increased remission for TNF Low: Downgraded because Remission biologics (ETN) + MTX than MTX alone medium level of study (RR, 1.77; 95% CI, 1.02, 3.08) limitations, and not enough events to meet optimal information size

Disease activity: TNF biologic (ETN) compared with non- Insufficient: Downgraded Remission TNF biologic (RIT) because no ITT analysis; large CIs cross appreciable benefits or harms; and not enough events to meet optimal information size

Disease activity: Significantly less radiographic progression Moderate: Downgraded because Radiographic for TNF biologic (ETN) plus MTX than medium level of study Changes MTX alone (SMD -0.81; 95% CI -0.98, - limitations 0.63)

Disease activity: Significantly less radiographic progression Moderate: Downgraded because Radiographic for TNF biologic (ETN) plus MTX than medium level of study Changes ETN alone (SMD -0.69; 95% CI -0.95, - limitations 0.43)

Functional Mixed results for TNF biologic (ETN) plus Low: Downgraded because Capacity MTX compared with MTX alone direction of effect varies, and large CIs

(Donahue et al., 2018)

Table 2. Effectiveness of Infliximab in the Treatment of Early Rheumatoid Arthritis Specific Outcome Result Strength of Evidence

Disease activity: Significantly improved response with TNF Low: Downgraded because Response biologic (IFX) + MTX than csDMARD medium level of study combination therapy (MTX + HCQ + limitations SSZ)

Disease activity: No significant difference between TNF Low: Downgraded because Response biologic (IFX) + csDMARD combination large CIs cross appreciable (MTX + PRED + HCQ + SSZ) and benefits or harms, and not csDMARD combination therapies alone enough events to meet optimal (MTX + PRED + HCQ + SSZ) information size

Disease activity: IFX + MTX compared with MTX alone Insufficient: Downgraded Response because not enough events to meet optimal information size; direction of effect varies; and large CIs cross appreciable benefits or harms

Disease activity: No significant difference between TNF Low: Downgraded because Remission biologic (IFX) + csDMARD combination large CIs cross appreciable (MTX + PRED + HCQ + SSZ) and benefits or harms, and not csDMARD combination therapies alone enough events to meet optimal (MTX + PRED + HCQ + SSZ) information size

Disease activity: Significantly increased remission for TNF Low: Downgraded because Remission biologics (IFX) + MTX than MTX alone medium level of study limitations

Disease activity: No significant difference between TNF Low: Downgraded because Radiographic biologic (IFX) + csDMARD combination large CIs cross appreciable Changes therapy (MTX + PRED + HCQ + SSZ) and benefits or harms, and not csDMARD combination therapies alone enough events to meet optimal (MTX + PRED + HCQ + SSZ) information size

Disease activity: IFX + MTX compared with MTX alone Insufficient: Downgraded Radiographic because not enough events to Changes meet optimal information size; direction of effect varies; and large CIs cross appreciable benefits or harms

Functional No significant difference between TNF Low: Downgraded because Capacity biologic (IFX) + csDMARD combination large CIs cross appreciable (MTX + PRED + HCQ + SSZ) and benefits or harms, and not csDMARD combination therapies alone enough events to meet optimal (MTX + PRED + HCQ + SSZ) information size

Functional Significantly greater improvement in TNF Low: Downgraded because Capacity biologic (IFX) plus MTX than MTX alone medium level of study limitations

(Donahue et al., 2018)

Table 3. Effectiveness of Adalimumab in the Treatment of Early Rheumatoid Arthritis Specific Outcome Result Strength of Evidence

Disease activity: Significantly improved response for TNF Moderate: Downgraded because Response biologic (ADA) + MTX than ADA alone high attrition (RR, 1.52; 95% CI 1.28, 1.80)

Disease activity: Significantly improved response with TNF Low: Downgraded because high Response biologic (ADA) + MTX compared with attrition MTX alone (RR, 1.35; 95% CI, 1.15 to 1.59)

Disease activity: ADA + MTX compared with csDMARD Insufficient: Downgraded Response combination (MTX + PRED + HCQ + because high attrition; not SSZ) enough events to meet optimal information size; and large CIs cross appreciable benefits or harms

Disease activity: TNF biologic (ADA) compared with non- Insufficient: Downgraded Response TNF biologic (RIT) because no ITT analysis; large CIs cross appreciable benefits or harms; and not enough events to meet optimal information size

Disease activity: Significantly higher remission for TNF Moderate: Downgraded because Remission biologic (ADA) + MTX than ADA alone high attrition (RR, 1.87; 95% CI 1.06, 3.28)

Disease activity: Significantly higher remission for TNF Low: Downgraded because high Remission biologic ADA + MTX than MTX alone attrition (RR, 2.04; 95% CI 1.15, 3.62)

Disease activity: ADA + MTX compared with csDMARD Insufficient: Downgraded Remission combination (MTX + PRED + HCQ + because high attrition; not SSZ) enough events to meet optimal information size; and large CIs cross appreciable benefits or harms

Disease activity: TNF biologic (ADA) compared with non- Insufficient: Downgraded Remission TNF biologic (RIT) because no ITT analysis; large CIs cross appreciable benefits or harms; and not enough events to meet optimal information size

Disease activity: ADA + MTX compared with csDMARD Insufficient: Downgraded Radiographic combination (MTX + PRED + HCQ + because high attrition; not Changes SSZ) enough events to meet optimal information size; and large CIs cross appreciable benefits or harms

Disease activity: Significantly less radiographic progression Moderate: Downgraded because Radiographic for TNF biologic (ADA) + MTX than ADA high attrition Changes alone (SMD, -0.38; 95% CI -0.55, -0.21)

Disease activity: Significantly less radiographic progression Moderate: Downgraded because Radiographic for TNF biologic (ADA) than MTX alone high attrition Changes (SMD, -0.61; 95% CI -0.78, -0.43)

Disease activity: Significantly less radiographic progression Low: Downgraded because high Radiographic for TNF biologic (ADA) + MTX than attrition, and large CIs cross Changes MTX alone (SMD, -0.99; 95% CI -1.17, - appreciable benefits or harms 0.81)

Functional Significantly greater improvement in TNF Moderate: Downgraded because Capacity biologic (ADA) plus MTX than ADA alone high attrition and for ADA than MTX

Functional Significantly greater improvement for TNF Moderate: Downgraded because Capacity biologic (ADA) + MTX than MTX alone high attrition

Functional ADA + MTX compared with csDMARD Insufficient: Downgraded Capacity combination (MTX + PRED + HCQ + because high attrition, and not SSZ) enough events to meet optimal information size (Donahue et al., 2018)

Table 4. Effectiveness of Certolizumab Pegol in the Treatment of Early Rheumatoid Arthritis Specific Outcome Result Strength of Evidence

Disease activity: Significantly improved response with TNF Low: Downgraded because high Response biologic (CZP) + MTX compared with attrition; large confidence MTX alone (RR, 1.20; 95% CI, 1.04 to intervals; and not enough events 1.38) to meet optimal information size

Disease activity: Significantly increased remission for TNF Low: Downgraded because high Remission biologics (CZP) + MTX compared with attrition; large CIs; and not MTX alone enough events to meet optimal information size

Disease activity: Significantly less radiographic progression Low: Downgraded because high Radiographic for TNF biologic (CZP) + MTX than attrition; large CIs; and not Changes MTX alone (SMD, -0.38; 95% CI -0.53, - enough events to meet optimal 0.23) information size

Functional Significantly greater improvement in TNF Low: Downgraded because high Capacity biologic (CZP) plus MTX than MTX attrition; large confidence alone intervals; and not enough events to meet optimal information size

(Donahue et al., 2018)

Table 5. Effectiveness of Etanercept in the Treatment of Advanced Rheumatoid Arthritis

Outcome Treatment Reference OR (95% CrI) RR (95% CrI) RD (95% CrI)

ACR 50 ETN Placebo + 1.76 (0.93 to 1.62 (0.94 to 0.07 (–0.01 to MTX 3.54) 2.75) 0.20) ACR 50 ETN + MTX Placebo + 3.95 (2.29 to 2.94 (1.98 to 0.22 (0.12 to MTX 7.51) 4.34) 0.38) ACR 50 ETN csDMARD 1.66 (0.72 to 1.53 (0.76 to 0.06 (–0.05 to + MTX 3.73) 3.10) 0.17) ACR 50 ETN + MTX csDMARD 3.73 (1.98 to 2.78 (1.61 to 0.21 (0.12 to + MTX 7.04) 4.98) 0.32) ACR 50 ETN MTX + 0.20 (0.05 to 0.36 (0.18 to –0.34 (–0.63 to SSZ + HCQ 0.77) 0.83) –0.05) ACR 50 ETN + MTX MTX + 0.45 (0.12 to 0.64 (0.37 to –0.19 (–0.48 to SSZ + HCQ 1.66) 1.39) 0.11) ACR 50 ETN Placebo + 4.10 (0.89 to 3.14 (0.90 to 0.21 (– 0.01 to csDMARD 23.63) 7.66) 0.62) ACR 50 ETN_STD + Placebo + 4.72 (1.40 to 3.45 (1.34 to 0.24 (0.03 to MTX csDMARD 16.87) 6.87) 0.54) Remission ETN Placebo + 1.40 (0.52 to 1.39 (0.52 to 0.01 (–0.01 to MTX 3.77) 3.53) 0.06) Remission ETN + MTX Placebo + 2.86 (1.03 to 2.73 (1.03 to 0.04 (0.001 to MTX 7.36) 6.38) 0.13) Remission ETN csDMARD 2.35 (0.78 to 2.30 (0.78 to 0.02 (–0.01 to + MTX 7.51) 7.20) 0.06) Remission ETN + MTX csDMARD 4.73 (1.92 to 4.46 (1.84 to 0.05 (0.02 to + MTX 12.20) 11.42) 0.12)

Outcome Treatment Reference SMD (95% CrI)

Disease ETN Placebo + –0.10 (–2.18 to Activity Score MTX 1.95) 28-Joint Count

Disease ETN + MTX Placebo + –0.98 (–2.72 to Activity Score MTX 0.75) 28-Joint Count Disease ETN Placebo 0.95 (–4.64 to Activity Score 6.44) 28-Joint Count Disease ETN + MTX Placebo 0.08 (–5.39 to Activity Score 5.45) 28-Joint Count Disease ETN Placebo + –1.88 (–5.79 to Activity Score csDMARD 1.98) 28-Joint Count Disease ETN + Placebo + –1.53 (–4.20 to Activity Score csDMARD csDMARD 1.14) 28-Joint Count Pain ETN + MTX Placebo + –0.72 (–1.43 to MTX –0.06) Pain ETN + MTX Placebo –1.33 (–2.55 to –0.11) Pain ETN + MTX MTX + –0.14 (–0.80 to SSZ + HCQ 0.57) Fatigue ETN + MTX Placebo + 0.47 (–0.64 to MTX 1.58) Radiographic ETN Placebo + –0.23 (–4.15 to Progression MTX 3.67) Radiographic ETN + MTX Placebo + -0.41 (–4.33 to Progression MTX 3.53) Radiographic ETN MTX + 0.04 (–5.11 to Progression SSZ + HCQ 5.22) Radiographic ETN + MTX MTX + –0.14 (–4.32 to Progression SSZ + HCQ 4.00)

Outcome Treatment Reference MD (95% CrI)

Health ETN + Placebo + –0.19 (–6.44 to Assessment csDMARD csDMARD 6.13)

Questionnaire Disability Index

(Wells et al., 2018)

Table 6. Effectiveness of Infliximab in the Treatment of Advanced Rheumatoid Arthritis Outcome Treatment Reference OR (95% CrI) RR (95% CrI) RD (95% CrI)

ACR 50 IFX + MTX Placebo + 3.00 (1.78 to 2.44 (1.63 to 0.17 (0.07 to MTX 5.08) 3.48) 0.28) ACR 50 IFX + MTX csDMARD 2.83 (0.97 to 2.30 (0.98 to 0.16 (–0.01 to + MTX 7.44) 5.28) 0.29) ACR 50 IFX + MTX MTX + 0.34 (0.08 to 0.53 (0.29 to –0.25 (–0.54 to SSZ + 1.29) 1.19) 0.05) HCQ Remission IFX + MTX Placebo + 6.58 (1.98 to 5.78 (1.94 to 0.12 (0.02 to MTX 23.28) 15.63) 0.33) Remission IFX + MTX csDMARD 10.92 (1.89 to 9.40 (1.82 to 0.12 (0.02 to + MTX 72.19) 52.36) 0.34) Outcome Treatment Reference SMD (95% CrI)

Disease Activity IFX + MTX Placebo + –0.78 (–3.06 to Score 28-Joint MTX 1.52) Count Disease Activity IFX + MTX Placebo 0.29 (–5.46 to Score 28-Joint 5.83) Count Disease Activity IFX + Placebo + –0.95 (–5.16 to Score 28-Joint csDMARD csDMARD 3.27) Count Radiographic IFX + MTX Placebo + –0.68 (–4.85 to Progression MTX 3.46) Radiographic IFX + MTX csDMARD –0.01 (–5.85 to Progression + MTX 5.90) Radiographic IFX + MTX MTX + –0.41 (–7.48 to

Progression SSZ + 6.62) HCQ Outcome Treatment Reference MD (95% CrI)

Health IFX + MTX Placebo + –0.24 (–0.45 to Assessment MTX –0.05) Questionnaire Disability Index Health-Related IFX + MTX Placebo + 4.58 (2.73 to Quality of Life, MTX 6.01) SF-36 Physical Component Score Health-Related IFX + MTX Placebo + 2.16 (-1.56 to Quality of Life, MTX 6.26) SF-36 Mental Component Score (Wells et al., 2018)

Table 7. Effectiveness of Adalimumab in the Treatment of Advanced Rheumatoid Arthritis

Outcome Treatment Reference OR (95% CrI) RR (95% CrI) RD (95% CrI)

ACR 50 ADA + MTX Placebo + 3.99 (2.84 to 2.96 (2.33 to 0.23 (0.16 to MTX 5.62) 3.72) 0.30) ACR 50 ADA + MTX csDMARD 3.78 (1.39 to 2.82 (1.27 to 0.22 (0.06 to + MTX 9.21) 6.16) 0.32) ACR 50 ADA + MTX MTX + SSZ 0.45 (0.11 to 0.64 (0.39 to –0.19 (–0.48 to + HCQ 1.61) 1.39) 0.10)

ACR 50 ADA + Placebo + 4.05 (1.24 to 3.11 (1.21 to 0.21 (0.02 to csDMARD csDMARD 13.53) 6.25) 0.50) Remission ADA + MTX Placebo + 8.87 (3.57 to 7.42 (3.34 to 0.16 (0.06 to MTX 27.29) 17.28) 0.37) Remission ADA + MTX csDMARD 14.96 (3.04 to 12.24 (2.77 to 0.16 (0.06 to + MTX 91.54) 63.80) 0.37)

Outcome Treatment Reference SMD (95% CrI)

Disease ADA Placebo + –1.80 (–6.18 to Activity Score MTX 2.74) 28-Joint Count Disease ADA + MTX Placebo –1.03 (–2.74 to Activity Score 0.66) 28-Joint Count Disease ADA Placebo + –0.73 (–3.29 to Activity Score MTX 1.84) 28-Joint Count Disease ADA + MTX Placebo 0.03 (–5.46 to Activity Score 5.40) 28-Joint Count Disease ADA + Placebo + –1.05 (–4.34 to Activity Score csDMARD csDMARD 2.20) 28-Joint Count Pain ADA Placebo + 0.02 (–1.32 to MTX 1.31) Pain ADA + MTX Placebo + –0.66 (–1.15 to MTX –0.30) Pain ADA MTX + SSZ 0.60 (–0.90 to + HCQ 2.14) Pain ADA + MTX MTX + SSZ –0.08 (–0.96 to + HCQ 0.77) Fatigue ADA + MTX Placebo + 0.39 (–0.21 to MTX 1.05)

Outcome Treatment Reference MD (95% CrI)

Health ADA + MTX Placebo + –0.25 (–0.34 to Assessment MTX –0.15) Questionnaire Disability Index

Health-Related ADA + MTX Placebo + 3.07 (0.73 to Quality of MTX 5.45) Life, SF-36 Physical Component Score

Health-Related ADA + MTX Placebo + 2.43 (–1.16 to Quality of MTX 6.11) Life, SF-36 Mental Component Score

(Wells et al., 2018)

Table 8. Effectiveness of Certolizumab Pegol in the Treatment of Advanced Rheumatoid Arthritis

Outcome Treatment Reference OR (95% CrI) RR (95% CrI) RD (95% CrI)

ACR 50 CZP + MTX Placebo + 5.35 (3.42 to 3.56 (2.66 to 0.30 (0.20 to MTX 8.67) 4.67) 0.41) ACR 50 CZP + MTX csDMARD 5.05 (1.82 to 3.37 (1.51 to 0.29 (0.12 to + MTX 12.98 7.46) 0.42) ACR 50 CZP + MTX MTX + SSZ 0.61 (0.15 to 0.77 (0.46 to –0.12 (–0.42 to + HCQ 2.26) 1.69) 0.18) ACR 50 CZP Placebo + 4.32 (0.82 to 3.25 (0.84 to 0.22 (– 0.02 to + csDMARD csDMARD 23.02) 7.53) 0.61) Remission CZP + MTX Placebo + 9.85 (2.36 to 8.08 (2.28 to 0.17 (0.03 to MTX 64.72) 27.89) 0.56) Remission CZP + MTX csDMARD 16.98 (2.35 to 13.52 (2.21 to 0.18 (0.04 to + MTX 158.00) 82.30) 0.57)

Outcome Treatment Reference SMD (95% CrI)

Disease CZP Placebo + –2.67 (–8.27 to Activity Score MTX 3.17) 28-Joint Count

Disease CZP + MTX Placebo + –2.23 (–4.04 to Activity Score MTX –0.43) 28-Joint Count Disease CZP Placebo –1.60 (–4.17 to Activity Score 0.97) 28-Joint Count Disease CZP + MTX Placebo –1.16 (–6.73 to Activity Score 4.21) 28-Joint Count

Pain CZP Placebo + –0.79 (–1.64 to MTX 0.02)

Pain CZP + MTX Placebo + –1.58 (–2.38 to MTX –0.77)

Pain CZP Placebo –1.40 (–1.98 to –0.81)

Pain CZP + MTX Placebo –2.19 (–3.46 to –0.88)

Pain CZP MTX + SSZ –0.21 (–1.30 to + HCQ 0.95)

Pain CZP + MTX MTX + SSZ –1.00 (–2.06 to + HCQ 0.16)

Fatigue CZP + MTX Placebo + 1.25 (0.17 to MTX 2.36)

Outcome Treatment Reference MD (95% CrI)

Health CZP + MTX Placebo + –0.36 (–0.50 to Assessment MTX –0.22) Questionnaire Disability Index Health-Related CZP + MTX Placebo + 5.07 (3.67 to Quality of MTX 6.49) Life, SF-36 Physical

Component Score Health-Related CZP + MTX Placebo + 3.60 (1.35 to Quality of MTX 5.83) Life, SF-36 Mental Component Score (Wells et al., 2018)

Table 9. Effectiveness of Golimumab in the Treatment of Advanced Rheumatoid Arthritis

Outcome Treatment Reference OR (95% CrI) RR (95% CrI) RD (95% CrI)

ACR 50 GOL (SC) + Placebo + 6.00 (3.27 to 3.80 (2.58 to 0.32 (0.19 to MTX MTX 11.35) 5.27) 0.48) ACR 50 GOL (IV) + Placebo + 2.90 (1.21 to 2.38 (1.19 to 0.16 (0.02 to MTX MTX 7.12) 4.19) 0.37) ACR 50 GOL (SC) + csDMARD 5.67 (1.88 to 3.59 (1.55 to 0.31 (0.12 to MTX + MTX 15.77) 8.13) 0.48) ACR 50 GOL (IV) + csDMARD 2.75 (0.74 to 2.25 (0.79 to 0.15 (–0.04 to MTX + MTX 9.17) 5.78) 0.37) ACR 50 GOL (SC) + MTX + SSZ 0.68 (0.16 to 0.82 (0.46 to –0.09 (–0.41 to MTX + HCQ 2.69) 1.83) 0.23) ACR 50 GOL (IV) + MTX + SSZ 0.33 (0.07 to 0.52 (0.23 to –0.25 (–0.57 to MTX + HCQ 1.49) 1.29) 0.09) Remission GOL (SC) + Placebo + 11.19 (4.19 to 8.93 (3.85 to 0.20 (0.08 to MTX MTX 33.16) 20.09) 0.40) Remission GOL (SC) + csDMARD 18.85 (3.58 to 14.75 (3.18 to 0.20 (0.08 to MTX + MTX 103.80) 70.16) 0.40)

Outcome Treatment Reference SMD (95% Crl)

Disease GOL (SC) Placebo + –2.02 (–7.70 to Activity Score MTX 3.76) 28-Joint Count

Disease GOL (SC) + Placebo + –1.26 (–3.80 to Activity Score MTX MTX 1.28) 28-Joint Count Disease GOL (IV) + Placebo + –1.03 (–3.54 to Activity Score MTX MTX 1.51) 28-Joint Count Disease GOL (SC) Placebo –0.96 (–3.51 to Activity Score 1.58) 28-Joint Count Disease GOL (SC) + Placebo –0.20 (–6.02 to Activity Score MTX 5.46) 28-Joint Count Disease GOL (IV) + Placebo 0.05 (–5.78 to Activity Score MTX 5.64) 28-Joint Count Fatigue GOL (SC) + Placebo + 0.54 (–0.25 to MTX MTX 1.33) Fatigue GOL (IV) + Placebo + 0.52 (–0.59 to MTX MTX 1.63)

Outcome Treatment Reference MD (95% CrI)

Health GOL (SC) + Placebo + –0.31 (–0.43 to Assessment MTX MTX –0.19) Questionnaire Disability Index Health GOL (IV) + Placebo + –0.28 (–0.47 to Assessment MTX MTX –0.09) Questionnaire Disability Index Health-Related GOL (SC) + Placebo + 4.83 (3.03 to Quality of Life, MTX MTX 6.76) SF-36 Physical Component Score

Health-Related GOL (IV) + Placebo + 3.65 (1.28 to Quality of Life, MTX MTX 6.00) SF-36 Physical Component Score Health-Related GOL (SC) + Placebo + 1.85 (–1.23 to Quality of Life, MTX MTX 4.93) SF-36 Mental Component Score Health-Related GOL (IV) + Placebo + 5.88 (2.18 to Quality of Life, MTX MTX 9.71) SF-36 Mental Component Score (Wells et al., 2018)

Table 10. Effectiveness of Etanercept in the Treatment of Ankylosing Spondylitis Outcome Result Strength of Evidence

ASAS40 Significantly improved ASAS40 response High for TNF biologics (ETN 25 mg twice weekly or 50 mg once weekly) versus placebo (RR, 3.31; 95% CrI, 2.38, 4.53) BASFI The mean BASFI in the ETN (25 mg twice High weekly or 50 mg once weekly) intervention group was 1.09 (SD: -1.60 to -0.56) lower than placebo ASAS Partial Significantly higher ASAS partial Moderate: Downgraded for Remission remission for TNF biologics (ETN 25 mg imprecision; fewer events than twice weekly or 50 mg once weekly) than 300 (a threshold rule-of-thumb) placebo (RR, 4.24; 95% CrI, 2.31, 8.09) and wide confidence interval. (Maxwell et al., 2015)

Table 11. Effectiveness of Infliximab in the Treatment of Ankylosing Spondylitis Outcome Result Strength of Evidence

ASAS40 Significantly improved ASAS40 response High for TNF biologics (IFX) versus placebo (RR, 4.07; 95% CrI, 2.80, 5.74)

BASFI The mean BASFI in the IFX intervention High group was 2.07 (SD: 2.71 to 1.35) lower than placebo BASFI A statistically significant mean difference of 1.50 (95% CI 0.94 to 2.06) in favour of infliximab was found in one study ASAS Partial Significantly higher ASAS partial Moderate: Downgraded for Remission remission for TNF biologics (IFX) than imprecision; fewer events than placebo (RR, 15.41; 95% CrI, 5.09, 47.98) 300 (a threshold rule-of-thumb) and wide confidence interval. MRI of Spinal The mean AS Spine MRI activity score in Moderate: Downgraded for Inflammation the GOL intervention group was 4.4 points imprecision; total population is (SD: 5.6 to 3.3) lower than placebo <400. MRI data available for 194/201 in infliximab group; 72/78 in placebo group. (Maxwell et al., 2015)

Table 12. Effectiveness of Adalimumab in the Treatment of Ankylosing Spondylitis Outcome Result Strength of Evidence

ASAS40 Significantly improved ASAS40 response High for TNF biologics (ADA) than placebo (RR, 3.53; 95% CrI, 2.49, 4.91) BASFI The mean BASFI in the ADA intervention High group was 1.57 (SD: -2.21 to -0.89) lower than placebo ASAS Partial Significantly higher ASAS partial Moderate: Downgraded for remission remission for TNF biologics (ADA) than imprecision; fewer events than placebo (RR, 6.28; 95% CrI, 3.13, 12.78) 300 (a threshold rule-of-thumb) and wide confidence interval. MRI of Spinal The mean SPARCC score in the ADA Moderate: Downgraded for Inflammation intervention group was 6.5 (SD: 13.06 to imprecision: total population is SPARCC score 0.06) lower than placebo <400. (Maxwell et al., 2015)

Table 13. Effectiveness of Golimumab in the Treatment of Ankylosing Spondylitis

Outcome Result Strength of Evidence

ASAS40 Significantly improved ASAS40 response High for TNF biologics (GOL) versus placebo (RR, 2.90; 95% CrI, 1.90, 4.23) BASFI The mean BASFI in the GOL intervention High group was 1.49 (SD: -2.27 to -0.69) lower than placebo ASAS Partial Significantly higher ASAS partial Moderate: Downgraded for Remission remission for TNF biologics (GOL) than imprecision; fewer events than placebo (RR, 5.18; 95% CrI, 1.90, 14.79) 300 (a threshold rule-of-thumb) and wide confidence interval. MRI of Spinal The mean AS Spine MRI activity score in Low: Downgraded for Inflammation the GOL intervention group was 3.4 points imprecision: total population is (SD: 7.7 to 0.90) lower than placebo <400 and MRI substudy (N = 60 for placebo and 50 mg golimumab arms) conducted at 10/57 participating sites; N = 216 in full RCT; readers were blinded but concerns regarding only modest level of agreement. Downgraded for concerns regarding missing data (12% did not have baseline and follow-up MRIs and imputed 7% of scores at week 14). (Maxwell et al., 2015)

Table 14. Disease Response - ACR PEDI 30 of Biologics in the Treatment of Juvenile Idiopathic Arthritis Placebo ETN ETN Placebo + ABT ADA ADA + (0.2mg/kg) (0.4mg/kg) MTX MTX

Placebo -- ETN 2.06 (0.9, -- (0.2mg/kg) 2.76) ETN 1.91 0.93 (0.71, -- (0.4mg/kg) (1.28, 1.87) 2.59)

Placebo + 1.14 0.57 (0.24, 0.6 (0.26, -- MTX (0.48, 1.37) 1.06) 1.94)

ABT 1.37 0.68 (0.35, 0.72 (0.37, 1.2 (0.54, -- (0.73, 1.56) 1.15) 2.92) 2.01)

ADA 1.55 0.76 (0.42, 0.81 (0.44, 1.33 (0.81, 1.12 -- (0.82, 1.74) 1.27) 2.64) (0.57, 2.28) 2.27)

ADA + 1.66 0.81 (0.5, 0.87 (0.53, 1.44 (0.93, 1.21 1.07 -- MTX (0.98, 1.86) 1.32) 2.85) (0.67, (0.73, 2.37) 2.39) 1.72) *Results are the median relative risk (95% credible interval) (Smith, 2017)

Table 15. Number of Active Joints For Biologics in the Treatment of Juvenile Idiopathic Arthritis Placebo ETN (0.2mg/kg) ETN (0.4mg/kg) ABT Placebo -- ETN (0.2mg/kg) -11.23 (-18.16, - -- 4.59)

ETN (0.4mg/kg) -11.01 (-14.59, - 0.23 (-5.52, 6.24) -- 7.52) ABT -3.10 (-5.88, -0.35) 8.17 (0.95, 15.48) 7.92 (3.51, 12.36) -- *Results are the median mean difference (95% credible interval); negative values indicate improvement (Smith, 2017)

Table 16. Number of Joints With Limited Range of Motion For Biologics in the Treatment of Juvenile Idiopathic Arthritis Placebo ETN (0.2mg/kg) ETN (0.4mg/kg) ABT Placebo -- ETN (0.2mg/kg) -5.32 (-14.6, 3.8) -- ETN (0.4mg/kg) -5.15 (-9.5, -0.8) 0.38 (-7.9, 8.2) -- ABT -1.15 (-4.7, 2.4) 4.22 (-5.7, 14) 3.91 (-1.6, 9.6) -- *Results are the median mean difference (95% credible interval); negative values indicate improvement (Smith, 2017)

Table 17. Physician's Global Assessment of Disease Activity 100 MM Vas For Biologics in the Treatment of Juvenile Idiopathic Arthritis Placebo ETN (0.2mg/kg) ETN (0.4mg/kg) ABT Placebo -- ETN (0.2mg/kg) -3.86 (-19.84, -- 12.34) ETN (0.4mg/kg) -4.1 (-11.56, 3.23) -0.07 (-14.59, 14.27) -- ABT -11.94 (-17.1, - -8.01 (-24.83, 8.85) -7.78 (-16.78, -- 6.79) 1.31) *Results are the median mean difference (95% credible interval); negative values indicate less disease activity (Smith, 2017)

Table 18. Functional Ability - Chaq Disability Index 0-3 For Biologics in the Treatment of Juvenile Idiopathic Arthritis Placebo ETN (0.2mg/kg) ETN (0.4mg/kg) ABT Placebo -- ETN (0.2mg/kg) -1.18 (-4.72, 2.4) -- ETN (0.4mg/kg) -0.69 (-3.17, 1.8) 0.48 (-2.05, 3) -- ABT -0.11 (-2.63, 2.35) 1.07 (-3.3, 5.38) 0.59 (-2.94, 4.07) -- *Results are the median mean difference (95% credible interval); negative values indicate better functional ability (Smith, 2017)

Table 19. Patient/Parent Assessment of Overall Well-Being, 100 MM Vas For Biologics in the Treatment of Juvenile Idiopathic Arthritis Placebo ETN (0.2mg/kg) ETN (0.4mg/kg) ABT Placebo -- ETN (0.2mg/kg) -4.74 (-22.2, -- 11.99) ETN (0.4mg/kg) -2.08 (-9.9, 5.63) 2.78 (-12.1, 18.07) -- ABT -6.04 (-11.32, - -1.32 (-18.91, 16.92) -3.93 (-13.27, 5.38) -- 0.83) *Results are the median mean difference (95% credible interval); negative values indicate an improvement (Smith, 2017)

Table 20. Inflammation ESR, MM/H For Biologics in the Treatment of Juvenile Idiopathic Arthritis Placebo ETN (0.2mg/kg) ETN (0.4mg/kg) ABT Placebo ETN -1.18 (-10.13, 7.7) (0.2mg/kg) ETN -1.18 (-7.52, 5.04) 0.02 (-6.34, 6.41) (0.4mg/kg) ABT -0.28 (-6.46, 6.08) 0.9 (-9.86, 11.73) 0.91 (-7.99, 9.83) *Results are the median mean difference (95% credible interval); negative values indicate an improvement (Smith, 2017)

Table 21. Biologics in The Induction of Remission For Crohn’s Disease Placebo Azathiopri MTX CZP IFX ADA IFX + IFX + ne/6- azathiop MTX mercaptop rine urine Azathiop 1.2 -- rine/6- (0.76– mercapto 2.1), 81% purine MTX 1.5 1.3 (0.49– -- (0.72– 2.9), 71% 3.2), 88%

CZP 1.4 1.1 (0.58– 0.89 -- (0.95– 2.0), 63% (0.40– 2.0), 96% 2.1), 38%

IFX 2.8 (1.4– 2.3 (1.3– 1.8 2.1 (0.98– -- 7.2), >99 5.0), >99% (0.69– 5.5), 97% % 6.4), 89%

ADA 2.9 (1.6– 2.4 (1.0– 1.9 2.1 (1.0– 1.0 -- 5.5), >99 4.9), 98% (0.76– 4.6), 98% (0.32– % 4.8), 92% 2.4), 53%

IFX + 4.3 (2.0– 3.4 (1.9– 2.7 (1.0– 3.1 (1.4– 1.5 1.4 -- azathiopr 9.8), >99 6.3), >99% 8.5), 98% 7.7), >99 (0.61– (0.59– ine % % 3.1), 4.2), 85% 79% IFX + 2.6 2.1 (0.67– 1.7 1.9 (0.56– 0.93 0.90 0.62 -- methotre (0.81– 7.9), 90% (0.43– 8.2), 84% (0.34– (0.22– (0.18– xate 11), 94% 8.9), 76% 2.6), 4.4), 2.4), 44% 44% 23% Vedolizu 2.0 (1.2– 1.6 (0.78– 1.3 1.4 (0.77– 0.70 0.67 0.47 0.75 mab 3.3), >99 3.2), 91% (0.53– 2.7), 89% (0.25– (0.33– (0.18– (0.17– % 3.2), 71% 1.5), 1.5), 1.1), 4% 2.7), 20% 15% 33% *Comparator, OR (95% CrI), probability intervention superior to comparator (greater odds of remission) (Hazlewood et al., 2015)

Table 22. Biologics in the Maintenance of Remission For Crohn’s Disease Placebo Azathiopri MTX CZP IFX ADA IFX + IFX + ne/6- azathiop MTX mercaptop rine urine Azathiop 1.7 (1.3– -- rine/6- 2.6), >99 mercapto % purine MTX 2.4 (1.1– 1.4 (0.58– -- 4.8), 98% 2.8), 78%

CZP 2.0 (1.4– 1.2 (0.65– 0.85 -- 3.0), >99 1.9), 72% (0.39– % 2.1), 34% IFX 2.8 (1.8– 1.6 (1.0– 1.2 1.4 (0.77– -- 4.5), >99 2.5), 98% (0.51– 2.6), 87% % 2.8), 65% ADA 5.1 (3.3– 2.9 (1.6– 2.1 5.1 (3.3– 1.8 -- 8.1), >99 5.1), >99% (0.96– 8.1) (0.94– % 5.0), 97% 3.4), 96% IFX + 5.2 (2.8– 3.0 (1.7– 2.2 2.6 (1.3– 1.8 (1.0– 1.0 -- azathiopr 11), >99 5.5), >99% (0.90– 6.0), 99% 3.8), 98% (0.48– ine % 6.1), 96% 2.5), 53%

IFX + 2.6 (0.96– 1.5 (0.57– 1.1 1.3 (0.44– 0.91 0.51 0.48 -- MTX 6.6), 97% 3.7), 79% (0.35– 3.5), 68% (0.41– (0.17– (0.16– 3.6), 54% 2.1), 41% 1.4), 9% 1.3), 8%

Vedolizu 2.2 (1.3– 1.3 (0.65– 0.91 1.1 (0.57– 0.77 0.42 0.42 0.85 mab 3.7), >99 2.3), 76% (0.39– 2.1), 59% (0.39– (0.22– (0.17– (0.29– % 2.3), 42% 1.5), 22% 0.85), 0.92), 2% 2.5), 38% 1% *Comparator, OR (95% CrI), probability intervention superior to comparator (greater odds of remission) (Hazlewood et al., 2015)

Appendix 2 Tables 1 to 15 portray the summary of findings for Anti-TNF safety outcomes with strength of evidence grades, if available.

Table 1. Safety of Etanercept in the Treatment of Early Rheumatoid Arthritis Specific Outcome Result Strength of Evidence Harms: SAEs and No significant differences between TNF Low: Downgraded D/C attributable to biologic (ETN) + MTX vs MTX alone (RR, because not enough AEs 0.81; 95% CI, 0.50, 1.29) events to meet optimal information size

Harms: SAEs and TNF biologic (ETN) compared with non- Insufficient: Downgraded D/C attributable to TNF biologic (RIT) because no ITT analysis; AEs large CIs cross appreciable benefits or harms; and not enough events to meet optimal information size

(Donahue et al., 2018)

Table 2. Table 3. Safety of Infliximab In the Treatment of Early Rheumatoid Arthritis Specific Outcome Result Strength of Evidence Harms: SAEs and No significant differences between TNF Low: Downgraded D/C attributable to biologic (IFX) + MTX vs MTX alone (RR, because medium level of AEs 3.03; 95% CI, 1.56, 5.90) study limitations

Harms: SAEs and No significant difference between TNF Low: Downgraded D/C attributable to biologic (IFX) + MTX or IFX + csDMARD because medium level of AEs combination (MTX + SSZ + HCQ + PRED) study limitations IFX + and csDMARD combination therapies (MTX csDMARD combination: + SSZ + HCQ + PRED) Downgraded because large CIs cross appreciable benefits or harms, and not enough events to meet optimal information size

(Donahue et al., 2018)

Table 3. Safety of Adalimumab in the Treatment of Early Rheumatoid Arthritis Specific Outcome Result Strength of Evidence

Harms: SAEs and No significant differences between TNF Low: Downgraded D/C attributable to biologic (ADA) + MTX vs MTX alone because high attrition; AEs (RR, 0.97; 95% CI, 0.29, 3.22) direction of effect varies; and large CIs cross appreciable benefits or harms

Harms: SAEs and TNF biologic (ADA) compared with non- Insufficient: Downgraded D/C attributable to TNF biologic (RIT) because no ITT analysis; AEs large CIs cross appreciable benefits or harms; and not enough events to meet optimal information size

Harms: SAEs and No significant differences between TNF Moderate: Downgraded D/C attributable to biologic (ADA) + MTX and ADA alone or because high attrition AEs between ADA and MTX

Harms: SAEs ADA + MTX compared with csDMARD Insufficient: Downgraded combination therapy (MTX + PRED + because high attrition; not HCQ + SSZ) enough events to meet optimal information size; and large CIs cross appreciable benefits or harms (Donahue et al., 2018)

Table 4. Safety of Certolizumab Pegol in the Treatment of Early Rheumatoid Arthritis Specific Outcome Result Strength of Evidence

Harms: SAEs and No significant differences between TNF Low: Downgraded D/C attributable to biologic (CZP) + MTX vs MTX alone (RR, because high attrition; AEs 1.08; 95% CI, 0.68, 1.73) large confidence intervals; and not enough events to meet optimal information size

(Donahue et al., 2018)

Table 5. Safety of Entarancept in the Treatment of Advanced Rheumatoid Arthritis Outcome Treatmen Reference OR (95% CrI) RR (95% CrI) RD (95% CrI) t Serious ETN Placebo + 1.10 (0.63 to 1.09 (0.65 to 0.005 (–0.02 to

Adverse MTX 1.85) 1.77) 0.04) Events Serious ETN + Placebo + 1.24 (0.73 to 1.22 (0.74 to 0.01 (–0.01 to Adverse MTX MTX 2.13) 2.02) 0.05) Events Serious ETN csDMAR 1.56 (0.51 to 1.52 (0.54 to 0.02 (–0.05 to Adverse D + MTX 5.61) 5.32) 0.06) Events Serious ETN + csDMAR 1.76 (0.65 to 1.71 (0.67 to 0.02 (–0.03 to Adverse MTX D + MTX 5.75) 5.42) 0.06) Events Serious ETN Placebo + 1.26 (0.19 to 1.25 (0.20 to 0.01 (–0.03 to Adverse csDMAR 8.74) 7.74) 0.12) Events D Serious ETN + Placebo + 2.35 (0.67 to 2.27 (0.68 to 0.03 (–0.01 to Adverse csDMAR csDMAR 9.82) 8.70) 0.12) Events D D

Withdrawal ETN Placebo + 0.79 (0.44 to 0.80 (0.45 to –0.01 (–0.02 to due to MTX 1.49) 1.47) 0.02) Adverse Events

Withdrawal ETN + Placebo + 0.68 (0.37 to 0.68 (0.38 to –0.01 (–0.02 to due to MTX MTX 1.22) 1.21) 0.01) Adverse Events

Withdrawal ETN csDMAR 0.39 (0.12 to 0.41 (0.14 to –0.04 (–0.16 to due to D + MTX 1.18) 1.18) 0.004) Adverse Events

Withdrawal ETN + csDMAR 0.33 (0.11 to 0.35 (0.13 to –0.04 (–0.16 to due to MTX D + MTX 0.89) 0.89) –0.003) Adverse Events

Withdrawal ETN MTX + 1.88 (0.40 to 1.86 (0.41 to 0.01 (–0.03 to due to SSZ + 12.21) 11.82) 0.04) Adverse HCQ Events

Withdrawal ETN + MTX + 1.61 (0.34 to 1.60 (0.35 to 0.01 (–0.04 to due to MTX SSZ + 9.95) 9.71) 0.03) Adverse HCQ Events

Withdrawal ETN Placebo + 3.46 (1.07 to 3.11 (1.07 to 0.10 (0.004 to due to csDMAR 13.18) 9.08) 0.31) Adverse D Events

Withdrawal ETN + Placebo + 1.65 (0.53 to 1.60 (0.54 to 0.03 (–0.03 to due to csDMAR csDMAR 6.03) 5.11) 0.15) Adverse D D Events (Wells et al., 2018)

Table 6. Safety of Infliximab in the Treatment of Advanced Rheumatoid Arthritis Outcome Treatment Reference OR (95% CrI) RR (95% CrI) RD (95% CrI) Serious IXF + Placebo + 0.62 (0.39 to 0.64 (0.40 to –0.02 (–0.03 to Adverse MTX MTX 1.03) 1.03) 0.001) Events Serious IFX + csDMAR 0.89 (0.25 to 0.89 (0.27 to –0.004 (–0.08 Adverse MTX D + MTX 3.62) 3.50) to 0.03) Events Withdrawal IFX + Placebo + 1.44 (0.69 to 1.42 (0.70 to 0.01 (–0.01 to due to MTX MTX 2.87) 2.71) 0.05) Adverse Events Withdrawal IFX + Placebo + 3.01 (0.09 to 2.82 (0.09 to 0.06 (–0.03 to due to MTX MTX 227.00) 29.48) 0.85) Adverse Events

Withdrawal IFX + csDMAR 0.70 (0.16 to 0.71 (0.19 to –0.02 (–0.15 to due to MTX D + MTX 2.74) 2.62) 0.04) Adverse Events

Withdrawal IFX csDMAR 1.44 (0.04 to 1.39 (0.04 to 0.02 (–0.14 to due to D + MTX 142.40) 22.49) 0.82) Adverse Events

Withdrawal IFX + MTX + 3.40 (0.70 to 3.28 (0.71 to 0.03 (–0.01 to due to MTX SSZ + 22.16) 20.91) 0.07) Adverse HCQ Events

Withdrawal IFX MTX + 8.03 (0.16 to 7.09 (0.17 to 0.08 (–0.03 to due to SSZ + 610.40) 104.80) 0.86) Adverse HCQ Events

(Wells et al., 2018)

Table 7. Safety of Adalimumab in the Treatment of Advanced Rheumatoid Arthritis Outcome Treatmen Reference OR (95% CrI) RR (95% CrI) RD (95% CrI) t Serious ADA + Placebo + 1.08 (0.46 to 1.08 (0.47 to 0.004 (–0.03 to Adverse MTX MTX 2.92) 2.68) 0.08) Events Serious ADA + csDMAR 1.53 (0.36 to 1.50 (0.39 to 0.02 (–0.06 to Adverse MTX D + MTX 8.29) 7.53) 0.10) Events

Serious ADA + Placebo + 2.17 (0.55 to 2.10 (0.56 to 0.03 (–0.02 to Adverse csDMAR csDMAR 11.04) 9.57) 0.14) Events D D Withdrawal ADA + Placebo + 1.27 (0.60 to 1.26 (0.61 to 0.01 (–0.01 to due to MTX MTX 2.75) 2.61) 0.05) Adverse Events

Withdrawal ADA + csDMAR 0.62 (0.14 to 0.63 (0.16 to –0.02 (–0.16 to due to MTX D + MTX 2.43) 2.35) 0.04) Adverse Events Withdrawal ADA + MTX + 2.97 (0.56 to 2.89 (0.58 to 0.03 (–0.02 to due to MTX SSZ + 20.37) 19.23) 0.07) Adverse HCQ Events

Withdrawal ADA + Placebo + 1.16 (0.24 to 1.15 (0.25 to 0.01 (–0.04 to due to csDMAR csDMAR 6.08) 5.10) 0.16) Adverse D D Events (Wells et al., 2018)

Table 8. Safety of Certolizumab Pegol In the Treatment of Advanced Rheumatoid Arthritis Outcome Treatmen Reference OR (95% CrI) RR (95% CrI) RD (95% CrI) t Serious CZP + Placebo + 1.30 (0.55 to 1.28 (0.56 to 0.01 (–0.02 to Adverse MTX MTX 3.41) 3.03) 0.10) Events Serious CZP + csDMAR 1.90 (0.42 to 1.83 (0.44 to 0.03 (–0.05 to Adverse MTX D + MTX 9.92) 8.78) 0.13) Events Withdrawal CZP + Placebo + 1.12 (0.45 to 1.11 (0.46 to 0.004 (–0.02 to due to MTX MTX 2.78) 2.63) 0.05) Adverse Events Withdrawal CZP + csDMAR 0.54 (0.12 to 0.56 (0.14 to –0.03 (–0.16 to due to MTX D + MTX 2.31) 2.23) 0.04) Adverse Events Withdrawal CZP + MTX + 2.64 (0.48 to 2.58 (0.50 to 0.02 (–0.02 to due to MTX SSZ + 20.02) 18.92) 0.07) Adverse HCQ Events Withdrawal CZP + Placebo + 1.47 (0.52 to 1.44 (0.53 to 0.02 (–0.02 to due to csDMAR csDMAR 4.78) 4.14) 0.13) Adverse D D

Events

(Wells et al., 2018)

Table 9. Safety of Golimumab in the Treatment of Advanced Rheumatoid Arthritis Outcome Treatment Reference OR (95% CrI) RR (95% CrI) RD (95% CrI)

Serious GOL (SC) Placebo + 2.10 (0.73 to 1.99 (0.74 to 0.05 (–0.01 to Adverse + MTX MTX 6.43) 5.13) 0.19) Events Serious GOL (SC) csDMAR 2.98 (0.61 to 2.76 (0.63 to 0.06 (–0.03 to Adverse + MTX D + MTX 17.50) 14.42) 0.21) Events Withdrawal GOL (SC) Placebo + 1.04 (0.35 to 1.04 (0.36 to 0.001 (–0.02 to due to + MTX MTX 3.05) 2.87) 0.06) Adverse Events

Withdrawal GOL (SC) csDMAR 0.52 (0.10 to 0.53 (0.11 to –0.03 (–0.16 to due to + MTX D + MTX 2.41) 2.31) 0.04) Adverse Events

Withdrawal GOL (SC) MTX + 2.45 (0.40 to 2.39 (0.41 to 0.02 (–0.03 to due to + MTX SSZ + 20.76) 19.45) 0.08) Adverse HCQ Events (Wells et al., 2018)

Table 10. Safety of Etanercept In the Treatment of Ankylosing Spondylitis Outcome Result Strength of Evidence

Withdrawals due Significantly increased withdrawals due Moderate: Downgraded for to adverse events to adverse events for TNF biologics (ETN imprecision; fewer events 25 mg twice weekly or 50 mg once than 300 (a threshold rule- weekly) than placebo (RR, 3.65; 95% CrI, of-thumb) and wide 1.27, 11.79) confidence interval. Serious adverse No significant difference in withdrawals Moderate: Downgraded for events due to adverse events for TNF biologics imprecision; fewer events (ETN 25 mg twice weekly or 50 mg once than 300 (a threshold rule- weekly) versus placebo (RR, 1.69; 95% of-thumb) and wide

CrI, 0.76, 3.72) confidence interval. (Maxwell et al., 2015)

Table 11. Safety of Infliximab In the Treatment of Ankylosing Spondylitis Outcome Result Strength of Evidence

Withdrawals due No significant difference in withdrawals Moderate: Downgraded for to adverse events due to adverse events for TNF biologics imprecision; fewer events (IFX) versus placebo (RR, 1.77; 95% CrI, than 300 (a threshold rule- 0.43, 8.46) of-thumb) and wide confidence interval. Serious adverse No significant difference in withdrawals Moderate: Downgraded for events due to adverse events for TNF biologics imprecision; fewer events (IFX) versus placebo (RR, 2.53; 95% CrI, than 300 (a threshold rule- 0.76, 11.09) of-thumb) and wide confidence interval. (Maxwell et al., 2015)

Table 12. Safety of Adalimumab In the Treatment of Ankylosing Spondylitis Outcome Result Strength of Evidence

Withdrawals due No significant difference in withdrawals Moderate: Downgraded for to adverse events due to adverse events for TNF biologics imprecision; fewer events (ADA) versus placebo (RR, 1.69; 95% than 300 (a threshold rule-of- CrI, 0.35, 10.84) thumb) and wide confidence interval. Serious adverse No significant difference in serious Moderate: Downgraded for events adverse events for TNF biologics (ADA) imprecision; fewer events versus placebo (RR, 0.92; 95% CrI, 0.26, than 300 (a threshold rule-of- 3.93) thumb) and wide confidence interval. (Maxwell et al., 2015)

Table 13. Safety of Golimumab In the Treatment of Ankylosing Spondylitis Outcome Result Strength of Evidence

Withdrawals due No significant difference in withdrawals Moderate: Downgraded for to adverse events due to adverse events for TNF biologics imprecision; fewer events (GOL) versus placebo (RR, 1.97; 95% than 300 (a threshold rule- CrI, 0.36, 17.51) of-thumb) and wide

confidence interval.

Serious adverse No significant difference in withdrawals Moderate: Downgraded for events due to adverse events for TNF biologics imprecision; fewer events (GOL) versus placebo (RR, 0.69; 95% than 300 (a threshold rule- CrI, 0.15, 3.32) of-thumb) and wide confidence interval. (Maxwell et al., 2015)

Table 14. Withdrawals in Anti-TNF Biologic Treated Patients For Crohn’s Disease Total Placebo Azathiopri MTX CZP IFX ADA IFX + IFX + withdrawa ne/6- azathiopri MTX ls mercaptop ne urine Azathiop 0.69 -- rine/6- (0.41– mercapto 1.1), purine 95% MTX 0.96 1.4 (0.62– -- (0.47– 3.3), 21% 1.9), 55% CZP 0.85 1.2 (0.67– 0.89 -- (0.56– 2.6), 27% (0.39– 1.4), 2.2), 76% 61% IFX 0.71 1.0 (0.55– 0.75 0.84 -- (0.39– 2.1), 47% (0.29– (0.37– 1.3), 1.9), 1.7), 88% 74% 68% ADA 0.43 0.62 0.45 0.50 0.60 -- (0.26– (0.32– (0.18– (0.24– (0.27– 0.69), >9 1.2), 92% 1.0), 0.96), 1.3), 9% 97% 98% 91% IFX + 0.27 0.39 0.29 0.32 0.39 0.64 -- azathiopr (0.08– (0.14– (0.07– (0.09– (0.11– (0.18– ine 0.72), 0.98), 0.93), 0.94), 1.1), 1.9), 99% 98% 98% 98% 96% 79%

IFX + 0.76 1.1 (0.29– 0.80 0.90 1.1 1.8 2.8 -- MTX (0.20– 4.5), 44% (0.17– (0.21– (0.33– (0.44– (0.59– 2.9), 3.6), 3.5), 3.5), 7.4), 16), 10% 66% 62% 56% 45% 20% Vedolizu 0.89 1.3 (0.65– 0.93 1.0 1.3 2.1 3.3 (1.1– 1.2 mab (0.51– 2.9), 24% (0.38– (0.50– (0.57– (1.0– 12), 2% (0.28– 1.6), 2.3), 2.1), 2.9), 4.6), 5.1), 66% 56% 45% 28% 2% 41% *Comparator, OR (95% CrI), probability intervention superior to comparator (lower odds of withdrawal) (Hazlewood et al., 2015)

Table 15. Withdrawals Due to Adverse Events in Anti-TNF Biologic Treated Patients For Crohn’s Disease WDAEs Placebo Azathiopri MTX CZP IFX ADA IFX + IFX + ne/6- azathiopr MTX mercaptop ine urine Azathiop 3.9 (2.4– -- rine/6- 6.4), mercapto <1% purine MTX 13 (3.2– 3.3 (0.78– -- 109), 29), 6% <1% CZP 0.88 0.23 0.07 -- (0.64– (0.13– (0.01– 1.2), 0.40), >99 0.28), > 78% % 99% IFX 2.7 (1.6– 0.71 0.21 3.1 -- 4.7), (0.44– (0.02– (1.7– <1% 1.2), 92% 0.93), 5.8), 98% <1% ADA 0.48 0.12 0.04 0.55 0.18 -- (0.31– (0.06– (0.00– (0.32– (0.09– 0.74), >9 0.24), >99 0.16), > 0.93), 0.34), > 9% % 99% 99% 99% IFX + 3.2 (1.6– 0.81 0.24 3.6 1.1 6.5 -- azathiopr 6.1), (0.47– (0.03– (1.7– (0.64– (3.0– ine <1% 1.4), 77% 1.1), 7.5), 2.0), 14), 96% <1% 32% <1%

IFX + 6.6 1.7 (0.12– 0.49 7.5 2.4 14 2.1 -- MTX (0.47– 52), 35% (0.02– (0.53– (0.18– (0.95– (0.15– 201), 8% 18), 226), 71), 411), 63), 29% 66% 7% 25% 3% Vedolizu 0.68 0.17 0.05 0.77 0.24 1.4 0.21 0.10 mab (0.40– (0.09– (0.01– (0.41– (0.12– (0.72– (0.09– (0.00– 1.1), 0.35), >99 0.23), > 1.4), 0.51), > 2.8), 0.50), >9 1.5), 93% % 99% 80% 99% 16% 9% 95% *Comparator, OR (95% CrI), probability intervention superior to comparator (lower odds of withdrawal) (Hazlewood et al., 2015)