Comparative Effectiveness and Safety of Biosimilar and Legacy Drugs

Comparative Effectiveness and Safety of Biosimilar and Legacy Drugs

In Canada and worldwide there is a need for updated independent real-world comparative effectiveness and safety data related to biologic drugs including biosimilar drugs. Biosimilar drugs hold potential to improve access to needed therapies at reduced cost enabling savings to be reallocated to other needs. However updated real-world evidence on comparative effectiveness and safety of biosimilar drugs is lacking. Investigators aim to demonstrate feasibility of creating network of clinical cohorts and other resources to provide real-world information on use of biosimilar drugs in Canada.

The core revolves around clinical datasets but investigators will complement with other data sources. Investigators will review data from National Prescription Drug Utilization Information System database that contains prescription claims-level data collected from publicly financed drug benefit programs in different provinces to conduct an environmental scan of the use of biosimilars and respective legacy drugs and other anti-Tumor Necrosis Factor agents covered by provincial drug plans from 2014-2017. Initial analysis will help to confirm that use of biosimilars is lower than corresponding legacy drugs.

Biologic drugs are relatively new and expensive drugs; biosimilar medicines are similar to original biologic drugs but cost less. If patients receive biosimilar drugs rather than originator biologics healthcare systems may be able to save money. Those savings can be used for other health care needs to benefit more Canadians. However investigators do not have detailed information on safety and effectiveness of these biosimilar drugs. The aim of study is to compare safety and effectiveness of biosimilar drugs to originator biologic drugs. Investigators will study patients with inflammatory rheumatic diseases (RA and AS) and Inflammatory Bowel Disease (CD and UC) and across Canada on these drugs. Primary focus is on patients without history of biologic drug use but investigators will also study patients switching to biosimilar drug from an originator biologic drug. Investigators will measure how long patients stay on treatment, if patients require new treatment, if the patients' disease control improves and occurrence of side effects such as infection that could be related to these drugs.

Research question: What is the comparative effectiveness and safety of biosimilar drugs versus the equivalent legacy drugs?

Primary aims:

To compare, in biologic-naive patients, new users of biosimilar drugs versus new users of the equivalent legacy drugs:

1. Frequency of discontinuation of the initial therapy
2. Persistence on the initial therapy (time until drug discontinuation)
3. Frequency of patients starting or increasing prednisone or other immunosuppressive drugs
4. Frequency of and time to discontinuation of treatment due to ineffectiveness
5. Frequency of and time to clinical remission/induction of response
6. Frequency of and time to serious adverse events

Secondary aims

To describe in biologic-naïve patients, new users of biosimilar drugs or the equivalent legacy drugs:

1. Change in disease activity over time
2. The frequency of, and time to, long-term outcomes (sustained remission, erosions in RA, radiographic progression in AS, and endoscopic mucosal healing scores in CD and UC).
3. Change in quality of life measures over time

Exploratory aims:

1. To describe the above outcomes in patients switching from the legacy drug, to the biosimilar.
2. To describe the above outcomes in patients switching from the biosimilar to the legacy drug Investigators will be working with two types of cohorts; more specifically, new cohorts collecting data prospectively and established cohorts that will share retrospective and prospective data as per Data Transfer Agreements. Data from these cohorts will be analyzed together.

Specific Outcomes

1. Discontinuation of initial therapy, measured as the number of patients who discontinued the initial treatment during the follow-up period.
2. Persistence on the initial therapy, defined as the time in months from cohort entry until drug discontinuation/switching.
3. Frequency of patients starting or increasing prednisone or other immunosuppressive drug.
4. Frequency of treatment failure, measured as the proportion of biosimilar or legacy drug patients who discontinue treatment due to ineffectiveness during follow-up.
5. Time to treatment failure, measured in months from drug initiation until treatment failure.
6. Frequency of clinical remission, measured by disease-specific measures including for RA, the Disease Activity Score (DAS28), and Simple Disease Activity Index (SDAI); the Spondylitis Disease Activity Score (ASDAS), the Assessment in SpondyloArthritis International Society (ASAS) response criteria, Bath Ankylosing Spondylitis Functional Index (BASFI), Stoke Ankylosing Spondylitis Spine Score (mSASSS) and Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) in AS; Crohn's Disease Activity Index (CDAI), Harvey-Bradshaw Index (HBI) for CD and Mayo or Partial Mayo Score (PMS) for UC.
7. Serious adverse events (SAEs), measured as the proportion of patients in each group presenting with an AE that is fatal or life-threatening, requiring or extending a patient's hospitalization, resulting in persistent or significant disability or incapacity, inducing a congenital anomaly or birth defect, or otherwise be considered important by the physician.
8. Long-term outcomes: frequency of sustained remission, presence of erosions in RA and radiographic progression in AS, and improvement or normalization of C-reactive protein and fecal calprotectin.
9. Patient-reported change over time in the EuroQol-5D (EQ-5D), the Health Assessment Questionnaire Disability Index (HAQ-DI), the Short Inflammatory Bowel Disease Questionnaire (SIBDQ), Visual Analogue Scale (VAS) and the veterans RAND 12 (VR-12).

OBJECTIVES

The primary objectives are to compare, in biologic-naive patients, new users of biosimilar drugs versus new users of the equivalent legacy drugs:

1. Frequency of discontinuation of the initial therapy

2. Persistence (time until drug discontinuation) for the initial therapy

3. Frequency of patients starting or increasing prednisone or other immunosuppressive drugs

4. Frequency of and time to discontinuation of treatment due to ineffectiveness

5. Frequency of and time to

   1. Induction and maintenance of response
   2. Clinical remission and sustained clinical remission

6. Frequency of and time to serious adverse events.

Methods

Enrolment:

Investigators are working with pan-Canadian, prospective cohorts of patients with inflammatory rheumatic diseases (rheumatoid arthritis, RA, and ankylosing spondylitis, AS) and Inflammatory Bowel Disease (IBD) (Crohn's disease, CD, and Ulcerative Colitis, UC). Cohort members eligible for our study are patients starting treatment with any biosimilar or any legacy drug. Patients will be enrolled over an approximate 24 month-period and the follow-up period will be up to four and a half years.

Patients will be followed for a maximum period of four and a half years for the study including patients who permanently discontinue the biosimilar or equivalent legacy drug treatment. Data for the study will be entered at enrolment and then approximately every 3 months in the first year of follow-up and then every 6 months thereafter up to 24 months. To complement information collected by each cohort, patients' data obtained from administrative health databases, including hospitalizations, physician billing, and prescription drugs will also be collected. For those patients who reach the 24-month mark before the end of our study (Dec. 2022), investigators will continue to collect data at yearly intervals (with a final data collection at month 54) so that an exploratory analysis of outcomes beyond 24 months can be done. Off protocol visits for patient reported increases in disease activity will occur as clinically indicated, but no additional data are required for collection.

Statistical analysis

As mentioned earlier, the primary analyses will focus on biologic- naïve patients prior to starting one of the drugs under study. Secondary analyses of non-biologic naïve patients will be done, adjusting for number of prior biologics at baseline.

Baseline characteristics will be presented using descriptive statistics to compare the two treatment groups. All continuous data will be expressed as the median (with range or interquartile range, IQR) or mean (with standard deviation SD) when appropriate. Categorical data will be presented as counts and percentages. Further descriptive comparisons of follow-up data will include: mean and cumulative doses, augmentation/reduction of therapy, and initiation or changes in the dose of prednisone and DMARDs/immunosuppressive agents.

In univariate analysis, groups will be compared in terms of frequency of treatment discontinuation for any reason, treatment failure due to ineffectiveness, and the occurrence of AEs including SAEs. Clinical remission, disease progression, and patient-reported outcomes will be assessed at time points during follow-up. Disease damage will be assessed through presence of erosions/radiographic progression in RA and AS, and specific surgeries (joint replacements in AS and RA, abdominal surgery for CD and UC).

For the continuous variables, the change over time between the two groups will be analysed using linear mixed models for repeated measures. Fixed effects of time will be estimated, and diagnosis, sex and age will be included in the analysis as possible effect modifiers.

Kaplan-Meier curves will be used to compare time from cohort entry to: i) treatment discontinuation/switching for any reason, ii) treatment failure due to ineffectiveness; iii) first episode of clinical remission, iv) SAE v) disability. Additional Kaplan-Meier curves will be plotted to compare time under sustained remission (from first remission until loss of effectiveness).

Then the same outcomes will be assessed using multivariate survival analyses during follow-up. In these analyses, the main exposure will be modeled as binary time-fixed indicator of drug used at cohort entry, as well as time-dependent cumulative duration and time-dependent cumulative dose of the biosimilar and the legacy drug. Survival analysis (Cox regression model) will be conducted to evaluate the risk of any SAE during follow-up. Separate Cox regression analysis will be conducted to evaluate the risk serious infection (defined as those requiring a hospitalization, emergency visit, or intravenous antibiotics), malignancies, and congestive heart failure, during the observational period. All models (for effectiveness and safety analyses) will be adjusted for the baseline covariates: sex, age, race/ethnicity, education, smoking, comorbidities, disease duration, disease activity, non-biologic or biologic DMARDs drugs, steroids, NSAIDs and COXIBs.

KNOWLEDGE TRANSLATION (KT) PLAN The results of this project will be disseminated in part through traditional methods of dissemination, such as publication in peer reviewed and open access journals and abstracts submitted to national and international meetings. The advisory committee will also advise on the development, implementation and progress of KT activities in this project. Investigators will provide updates to Health Canada via the DSEN coordinating office in two formats: quarterly interim reports and a final report. Through quarterly interim reports investigators will share information on the status of the project, provide early results from preliminary analysis, as well as inform DSEN about potential modifications of project milestones and/or research protocols as needed. Investigators will also provide a final report, and make use of DSEN contacts with policy makers to ensure that results are adequately disseminated. Dissemination to a wide audience (researchers, policy makers, and other stakeholders) at DSEN-sponsored workshops (such as those held in the context of annual meetings of CADTH and the Canadian Association for Population Therapeutics [CAPT]) may also be possible.

A novel element of our KT plan will be the presentation of the preliminary and final results at stakeholder workshops. The presentation will help investigators gather information from a wide range of stakeholders (patients, physicians, and policy makers) regarding reasons for our results.

SUMMARY With this proof of concept project, investigators aim to demonstrate the feasibility of creating a network of clinical cohorts and other resources to provide real-world information on the use of biosimilar drugs in Canada. The core of the proposal revolves around the clinical datasets, but as investigators will complement that with analyses of national drug patterns using NPDUIS data. CAN-AIM is well-positioned to develop an effective program of research and surveillance related to biologic and biosimilar drugs.

Inclusion Criteria:

The study will include cohort members from both sexes, 18 years and older, with a clinical diagnosis of inflammatory rheumatic disease (either RA or AS), or IBD (CD or UC) who have given their informed consent. There are no disease activity criteria for entry. At the time of enrolment, patients in each disease group will be classified into the following treatment subgroups:

• Biologic-naïve, starting any biosimilar or the equivalent legacy drug;
• Patients switching to biosimilar or the equivalent legacy drug from an alternative biologic therapy;
• Patients switching to a biosimilar (or starting a new cycle with the equivalent legacy drug), successfully completed and exited a previous course of therapy with the equivalent legacy drug.

Exclusion Criteria:

• Under 18 years of age
• No clinical diagnosis of inflammatory rheumatic disease (either RA or AS), or IBD (CD or UC)
• Refused to participate or sign informed consent