Badbir protocol v12_13032008finaleth

British Association of Dermatologists' Biological Interventions Register
Final BADBIR Protocol (Version 12 March 13th 2008)
Steering group members
Anthony Ormerod (Chair)
Alex Anstey
Prof. Jonathan Barker
David Burden
Robert Chalmers
David Chandler (PAA)
Prof Andrew Finlay
Prof. Chris Griffiths
Karina Jackson
Neil McHugh
Kevin Mckenna
Prof N Reynolds
Catherine Smith
Study Team
Principal Investigator
1. Background
Biological interventions using highly specific immuno-modulatory agents represent a new therapeutic
approach to the treatment of patients with severe psoriasis, especially those in whom other agents have failed, are contra-indicated or are for other reasons unsuitable. The scientific basis, mode of action, effectiveness and safety of these interventions have been more rigorously tested than many standard psoriasis therapies but the evidence is based on short term clinical trial interventions, commonly 3 -6 months; there is, however, some limited data up to three years with certain of these agents and considerably more experience from use in other diseases with others (infliximab and Psoriasis tends to be a lifelong illness, most commonly starting before 40 years of age and often presenting initially in childhood or early adulthood. Patients with severe disease are known to have a significantly increased mortality, particularly from cardiovascular disease (Mallbris et al.;Wong et al.;Gladman et al.). They tend to require interventions over long periods of their life and many of these expose them to toxic and potentially fatal side effects. For photochemotherapy this includes squamous carcinoma and melanoma; for methotrexate, haematopoietic failure, cirrhosis and pulmonary fibrosis (rare in psoriasis patients); and for ciclosporin, renal impairment hypertension and its consequences. Paul et al noted a doubling of the incidence of malignancies in 1252 patients treated with ciclosporin due to a higher (six fold) incidence of squamous carcinoma, particularly in patients treated with PUVA and more than two years of ciclosporin (Paul et al.). With acitretin there may be the development of skeletal hyperostoses, hyperlipidaemia and its consequences, and hepatotoxicity. Side effects such as BADBIR Protocol Version 12 March 13th 2008
nausea, vomiting, headache, hair loss, myopathy etc. can prevent the use of an agent in some patients. The long term effects and relative risks with each of the modalities or combinations of these modalities are poorly studied and poorly understood. Retrospective cross sectional studies have been carried out in large populations of patients with severe psoriasis. A cohort of 8991 patients hospitalised for psoriasis (Mallbris et al.) showed that patients with severe disease, as indicated by frequent admission and earlier age of onset, is associated with an increased risk of cardiovascular death (SMR 2.62; 95% CI 1.91-3.49). Olsen (Olsen, Moller, and Frentz) reported on 6910 patients with psoriasis and found an increase in cancer of the larynx (RR2.8) and pharynx (RR2.9) in men and colon RR (1.6) and kidney (RR 2.3) in women. In a community-based study of more than 100,000 people aged over 65 years, Gelfand (Gelfand et al.) found there to be an increased incidence of lymphoma amongst the 2718 patients with psoriasis (Relative rate 2.95; CI 1.83-4.76): only 1.5% of these patients received ciclosporin and the cohort pre- dated the widespread use of this drug, and the finding pertained even when methotrexate patients and those developing mycosis fungoides were excluded. Boffetta (Boffetta, Gridley, and Lindelof) reported increased cancer risks in a cohort of 9773 patients with psoriasis (SIR 1.37; 95% CI 1.28 - 1.47), most notably squamous carcinoma of the skin (2.64), vulva (3.24) and penis (4.66). Interestingly, malignant melanoma was reduced in incidence (SIR 0.32; 95% CI 0.10-0.74). In addition several malignancies associated with smoking and alcohol were increased. A similar Finnish study by Anna Hannuksela- Svahn et al (Hannuksela-Svahn et al.) examining 5687 patients who had been hospitalised for psoriasis revealed an increased incidence of Hodgkin's disease (RR 3.3; CI 1.4-6.4) and squamous carcinoma of the skin (SIR 3.2; 95% CI 2.3-4.4), non Hodgkin's lymphoma (SIR 2.2; 95% CI 1.4-3.4) and laryngeal carcinoma. Melanoma incidence was reduced (SIR 0.8; CI 0.3-1.6). Margolis (Margolis et al.) studied 1101 patients with severe psoriasis requiring second line therapy and 16519 patients with less severe disease. They used patients with severe eczema, hypertension or organ transplantation as controls. They found a similar incidence of cancer in severe psoriasis patients to that found in the organ transplants (RR 2.12; 95% CI 1.8-2.5) with males and older patients having the greatest risk. The risk ratio for lymphoma was 7.95 (95% CI 4.94-12.79). Non-melanoma skin cancer accounted for most other malignancies in their patients but the sample was of insufficient power to compare differences between treatments. The increased risk in the non-severe psoriasis patients was only slightly increased (RR 1.13; 95% CI 1.03-1.25). Whether these effects are a consequence of disease severity or the use of therapies cannot be ascertained. Excess mortality related to alcohol and smoking is also found to be associated with severe psoriasis (Poikolainen, Karvonen, and Pukkala). Overall SMR was 1.62 (95% CI 1.52 -1.71) for men, and for women 1.54 (95% CI 1.43-1.64). For causes related to alcohol the SMR for men was 4.46 and for women 5.6. Similar ratios have been found for patients with psoriatic arthritis (SMR 1.59 for males and 1.65 for females) (Wong et al.). Potentially, disease modification can have beneficial effects on disease associated co-morbidity. This has been established for low dose methotrexate (Prodanowich et al.) and for TNF blockers in rheumatoid arthritis (Jacobsson et al.) BADBIR Protocol Version 12 March 13th 2008
Thus psoriasis itself is associated with health risks that may relate to disease severity and may alternatively be modified by interventions with immunosuppressive and UV based therapies. The disease is a long term condition for which optimal long term management has little evidence to guide the clinician. We do not know whether powerful but toxic interventions lead to a net benefit or a net
2. Rationale for the Establishment of a Biological Interventions for Psoriasis
Register

The primary purpose of establishing a "biologicals" registry for psoriasis is to follow a large cohort of
patients treated with biological agents so that their long-term safety can be monitored. This long-term safety data cannot be determined from short-term clinical trials in selected patients. A subsidiary aim will be to collect information on their long-term efficacy. In the UK, three agents (infliximab, efalizumab and etanercept) are licensed for treatment of psoriasis and two are undergoing technology approval by NICE (efalizumab and etanercept). These agents are free from the traditional end organ toxicities of existing systemic agents but have other side effects such as infusion reactions, chills, injection site reactions and development of antinuclear antibodies (infliximab and etanercept); additional rare side effects include thrombocytopenia (efalizumab), rebound or flare (efalizumab), serious infection e.g. tuberculosis (infliximab and etanercept), cardiac failure and demyelinating disease (infliximab and etanercept). They are likely to offer considerable benefits in safety and quality of life for those with more severe disease but questions remain regarding long term safety and rare side effects. Other biological agents are being evaluated for psoriasis (e.g. adalimumab) and when these become licensed they could be integrated into the register. We need to have a better understanding of the advantages and disadvantages of these agents for maintaining suppression of severe psoriasis over years and of how these compare with existing agents. Such understanding will inform the place of each treatment in long-term treatment strategies, e.g. in what sequence should they be given and should they be used in combination with existing drugs or with each other? Although some of these questions might be answered by carefully designed randomised controlled trials, there will inevitably be many uncertainties about the "real world" use of these therapies and much could be learned from a registry. Establishing a registry of all UK patients exposed to biological therapy for psoriasis and a control group given conventional therapies including ciclosporin, methotrexate and acitretin will help to answer these questions. In a prospective follow-up study of 1380 patients treated with PUVA alone over a 20 year period, Stern, Nichols and Vakeva observed a five fold increase in the relative risk for melanoma (Stern, Nichols, and Vakeva). This was in an American population where there is a higher background incidence of melanoma than in the UK. Latency was as long as 10-15 years and the crude incidence BADBIR Protocol Version 12 March 13th 2008
was only of 8 more melanomas than expected. A similar follow up of 4799 Swedish patients treated with PUVA failed to show an increase in the risk of melanoma following systemic PUVA (Lindelof et al.). The proposed register will probably include a number of patients who have received large cumulative doses of PUVA in the past. Current UK practice, however, is to rely more on narrow band UVB for which the carcinogenic risks are thought to be less but where there is a lack of long-term Comparable databases have been set up in Europe and USA for anti-TNFα therapies in other indications and initial findings are starting to be reported. In the Stockholm register for inflammatory bowel disease, with 217 infliximab treated patients (Ljung et al.), the risk of adverse events was increased in elderly patients with severe inflammatory bowel disease and lymphoma had a 1.5% incidence. A Spanish register of 1,540 patients treated with infliximab (86%) and etanercept (14%) showed a 1.1-1.9% incidence of tuberculosis in patients treated for rheumatoid arthritis (Gomez-Reino The American national register for rheumatic disease studied by Wolfe (Wolfe and Michaud) included 18,572 patients with RA and showed an overall SIR for lymphoma in patients treated for RA with anti TNFα therapies of 2.9 (95% CI 1.7-4.9), but this may be due to patients with more severe disease being represented in these treatment cohorts and the authors could not establish a causal relationship between RA treatment and the lymphomas observed.
3. Methods

3.1 Aims
The primary purpose of establishing a "biologicals" register for psoriasis is to ascertain whether there is an importantly increased risk of serious adverse events following the introduction of these agents in the treatment of psoriasis compared to that expected from a conventionally treated cohort with comparable disease severity. This assessment is to include potential adverse effects, which have not been detected in the relatively short-term clinical trials and those which are theoretically or currently perceived as important. Specifically this includes cancer especially lymphoma, non-melanoma skin cancer especially squamous cell carcinoma, demyelinating disease and tuberculosis. A subsidiary aim will be to collect information on the long-term efficacy of these therapies. A number of subsidiary questions will also be addressed which include the evaluation of differences between these agents, multiple agents concurrently or in sequence in terms of serious adverse effects. Further, it is proposed that the register will seek to identify all available data on patients who become pregnant on treatment and to follow up the outcome of those pregnancies. BADBIR Protocol Version 12 March 13th 2008
The BADBIR will also correct for the influence of potential confounders on these outcomes such as psoriasis severity, alcohol and cigarette smoking; non-biological concomitant or previous therapy; and This initial proposal is based on outcomes to be ascertained within 5 years of start of treatment though it is accepted that longer term follow up may be required for serious adverse events with a greater The results will inform clinical practice for long-term management of this chronic, often lifelong 3.2 Design
This is a prospective cohort study consisting of two cohorts comparing patients treated with biological interventions to a control group with similar disease characteristics but exposed only to non-biological systemic therapies. The comparison group would include patients treated with PUVA, methotrexate, ciclosporin and acitretin. The protocol will be submitted for MREC approval. Analysis will take into account switching from the control group to a biologic agent or from one biologic agent to a different The register will be modelled on the existing British Society for Rheumatology Biologics Register, BSRBR, and run on a similar platform and co-located at Manchester University. Staff of the BSRBR will be partners in running this new register. BADBIR will promote registration of biological therapies and of controls to all dermatologists prescribing these interventions on all patients they treat that satisfy the inclusion criteria and that consent to take part. The register aims to recruit all patients receiving each agent until the required cohort size has been attained. Numbers required need to be achievable and sufficient to enable worthwhile comparisons to be made. It is anticipated that 2000- 4000 will be required in each biological intervention and 4000 controls. Recruitment for one agent would cease if the 4000 patient target is reached. Following registration, for the duration of the study, BADBIR will approach the dermatologists to update the records of all patients whether or not they continue on therapy. This will be captured primarily as web-based data entry. Dermatologists will be able to view data on their patients and add to this without unnecessary repetition. Paper forms will be available as a substitute for those unable to The co-ordinating centre will mail patients with paper forms to gain additional information on their quality of life, drinking and smoking habits, medication and any health care problems according to the BADBIR Protocol Version 12 March 13th 2008
Where responses from patients or physicians are delayed there will be repeated reminders and phone calls if necessary to ensure the most complete data possible is obtained. When formal follow-up of the last patients entered in the register is complete, BADBIR will continue to link the register to the national cancer register and to the death register. Patient data will need to be acquired and stored with patient specific information. This will be pseudonymised (e.g. patient
3.2.1 Exposed cohort
Inclusion criteria
1. Patients commencing treatment with a biological agent in the previous six months for
their psoriasis
2. Age 16 years or older
3. Willingness to give informed consent for long term follow-up and access to all medical records. To reduce bias between this and the active intervention group BADBIR will also collect at baseline the reasons for treatment with the chosen agent, whether the patient is either intolerant or contraindicated or failed to respond to other therapies. With support of the BAD, external validity will be maintained by urging involvement of all dermatologists in the registration process. BAD guidelines and guidance from NICE will all state that patients treated with biologic therapy should be registered. Failure to do so can be construed as not complying with normal clinical practice. The study will be restricted to the United Kingdom and the Republic of Ireland and will be co-
ordinated by a steering group acting on behalf of the BAD. 3.2.2 Non-exposed cohort
Many patients with similar disease severity will continue to be treated with traditional interventions. The severity of disease requiring a systemic intervention is likely to compare quite closely with that of those exposed to biological interventions. Most frequently a decision to use biological therapy will be based more on unsuitability or unresponsiveness to existing therapy than on disease severity. There are likely to be differences, for example in the responsiveness to standard agents, compared to patients in the biological group; these cannot be quantified other than by fully documenting previous systemic treatment for psoriasis. These random heterogeneous effects should be similar over a large BADBIR Protocol Version 12 March 13th 2008
The controls will be recruited across all contributing centres, with participants encouraged to register one control for every patient registered. This will ensure high recruitment of controls and reduce the Analysis will take into account switching from one group to another or to different biologics such that the person years of follow-up in the control group switch to the biological group if biological therapy is Inclusion criteria
1. Patients initiating or switching conventional therapy with PUVA, ciclosporin, methotrexate, fumaric acid esters or acitretin for therapy of their psoriasis.
2. If not switching therapy, patients must have severe psoriasis meeting the severity criteria for biological therapy as in the BAD guideline (rule of 10s) 3. Aged 16 years or older
4. Informed consent to participate in long-term follow-up and access to all medical records. Exclusion criteria
1. Patients must never have been exposed to biological therapy Note: If a patient was subsequently started on biological therapy, then he/she would switch from the control cohort to the biological cohort as the design is to include all eligible patients in that cohort.

4. Statistics, Sample size and statistical power (see also appendix 1)
The initial analyses will consist of comparisons in baseline status between the individuals in the
treatment cohorts. For the purposes of analysis (initially) follow-up time will be censored in both cohorts if there is switching to another class of biologic therapy and censored in the standard therapy group if there is switching to a biologic agent. The adverse events of interest are calculated per person time of follow-up, after the start of therapy. Depending on the events, separate analyses are undertaken (i) restricting consideration to time on drug, which include the period within 90 days of last injection and (ii) all person time following start of therapy. Standard time-dependent regression analyses will be undertaken to compare event rates between groups after adjusting for baseline and Sample size of the biologics exposure group will be limited or determined by external factors: BADBIR Protocol Version 12 March 13th 2008
1) NICE technology assessment 103 and recommendation of the agents with etanercept failure being a criteria for the use of efalizumab 2) NICE indicates the need for pharmacovigilance and recommends patients are registered in this After discussion with sponsors, industry, health administration and opinion leaders we feel that over 5 years collecting 4000 patients in each of the biological intervention arms and in the control cohort is likely to be achievable. The size of the comparison cohort will be under our control. However, it is difficult to anticipate the magnitude of rate differences for adverse events between the cohorts as patients from all groups are likely to have had prior exposure to immunosuppressive drugs. Cancer is likely to have a low incidence, which may also be increased by having severe psoriasis. Using crude incidence figures in psoriatic patients, approximated from previous studies (Hannuksela-Svahn et al.), (Boffetta, Gridley, and Lindelof) these would be as follows: Any adverse events with a frequency of up to 1 in 2,000 in the control group should be addressed within the power of the register. (See assumptions in appendix 1). Bold figures, above, indicate those outcomes which the register is powered to address to an increase risk of 3 or 4 fold over 5 years. Other potential adverse effects with biologicals have been shown to have a strong signal which would be detected by the register by virtue of the many fold increase in risk of a rare outcome e.g. Tuberculosis increased by a factor of 5 with anti-TNFα agents. The sample size required in each group for a 2 sided significance of alpha < 0.05 to be detected with 80% power has been determined in patient years. Grey shaded areas in appendix 1 indicate predictions within the scope of the register. BADBIR Protocol Version 12 March 13th 2008
Estimating the risk of rare adverse effects with a smaller signal, especially lymphoma will be facilitated by long-term linkage to the national cancer registry (in addition to the control group). The risk window for cancer being defined as once exposed always at risk. Where two biologicals have been used, the proportion of time spent on each will define its possible contribution to risk. Where the adverse event is rare or where a biological intervention is under-represented in the register, the numbers of patient's data can also potentially be increased by sharing data with other compatible registers such as those operating in Sweden, Italy, and Germany. The aim is to recruit 4,000 patients on conventional treatments and 2,000 to 4,000 on each biological intervention (depending on the uptake of these drugs in clinical practice). NICE guidelines indicate that in the UK etanercept should be used first and it is likely to be used much more than the others. 4,000 patients in each cohort, biologicals and conventional treatment would give an exposure of 12,000 patient years in each group. This would give power to detect at least a 3 or 4 fold increase in risk of events occurring at a frequency of 1 in a 1000 or 1 in 2000 patients. Rarer events would be detected if the relative risks were higher. nQuery advisor (version 5, JD Elashoff) was used to calculate the person years of follow-up required using a 95% confidence level and 80% power 1 to 1 ratio in each cohort. This would be sufficient to detect for example the risk of non melanoma skin cancer which is a particular concern in these patients who have been exposed to phototherapy.
5. Auditing the conduct of the study and research governance

The following coordinated program will ensure quality control
a. Training of staff - including a program of training for nurses in PASI scoring and how to use the register. A coordinated program is underway. b. An on-line manual will be provided for dermatologists to send in quality data, including c. Quality checks will be made for data received (i.e. manual scanning for completeness, errors and then checks at data entry stage for inconsistencies). d. Selected serious adverse events (SAEs) will be checked against a set of
predefined validation criteria.
BADBIR Protocol Version 12 March 13th 2008

6. Summary Study flow chart
Data captured
Baseline
Follow up (months)
6, 12, 18, 24,
30, 36, 48, 60
(If applicable)
applicable)
Transaminase, Cholesterol, triglycerides Systemic treatments
*Pregnancy: Specific prompts in the consultant follow-up forms with additional questionnaires if yes to
follow specific outcome.
7. Baseline data.

This will necessarily be comprehensive to identify potential confounding factors.
A unique identifier will be assigned on registration of the patient. Ascertainment of data will be from a combination of patient interview, examination and examination of hospital medical records, performed by a doctor or trained deputy e.g. nurse. 7.1 Patient identification (separately stored for confidentiality)
• NHS number (Chi number Scotland) (health and care number Northern Ireland) • Hospital unit number if above not known BADBIR Protocol Version 12 March 13th 2008
7.2 To appear in the register
• Patient identification unique number 7.3 Psoriasis details
Chronic plaque with guttate Chronic plaque without guttate Erythrodermic Generalised pustular Localised pustular Nails Flexural (inverse) Scalp Acrodermatits continua of Hallopeau Family history of psoriasis in first degree relatives yes / no Has the patient a diagnosis by a rheumatologist of inflammatory arthritis? Y/N Patients with arthritis - HAQ score to be obtained via patient questionnaires every 6 months. 7.4 Baseline severity
• CAGE questionnaire for alcohol dependence 7.5 Baseline examination
6.6 Prior therapy
Has the patient previously received and total exposure (months) BADBIR Protocol Version 12 March 13th 2008
3. Oral retinoids 4. Hydroxycarbamide 5. Azathioprine 6. Mycophenolate mofetil 7. Fumaric acid esters 8. Infliximab (biologicals group only) 9. Etanercept (biologicals group only) 10. Efalizumab (biologicals group only) 11. Adalimumab(biologicals group only) 12. Alefacept (biologicals group only) 7.7 Risk factors for skin cancer
Residence in tropical/subtropical countries History of prior Neoplastic or pre-cancerous lesions:- Melanoma, Melanoma in situ (give site and date for each), SCC (give number), BCC (give number), yes tick for Keratoacanthoma, Actinic Keratosis, Bowen's Disease. 7.8 Co morbidity data
High blood pressure Angina Heart attack Stroke Epilepsy Asthma Chronic bronchitis/emphysema Peptic ulcer Liver disease Hepatitis Abnormal LFTs Renal disease Raised creatinine BADBIR Protocol Version 12 March 13th 2008
TB increased risk Demyelination Diabetes Thyroid disease Depression Non skin cancer Type free text, site free text, date Blood dyscrasia Immunodeficiency syndromes 1. Current, 2.ex-smoker or 3.never-smoked a) Number of cigarettes currently smoked b) Number of cigarettes smoked when a smoker Current alcohol intake: number of units per week (give examples on the form) 7.9 Concomitant medications
List drugs patient is taking - (predictive text on field) A specific prompt will be made for topical tacrolimus and pimecrolimus use. 7.10 Laboratory investigations
Basic blood results will be captured including: Haemoglobin, White cell count, Platelets, Creatinine, Transaminase (ALT), and where possible fasting Cholesterol and triglyceride. These will be recorded in the register at baseline and every 6 months
8. Follow up data

Recorded at 6 monthly intervals for 3 years and yearly for a further 2 years the following data will be

8.1 Consultant Follow up

• Have there been any changes to the patient's biological therapy? • If yes record drug, dose started and stopped • Other
• Any change in the patient's oral anti-psoriatic medication? • Anti-psoriatic drug treatment, dose, started and stopped? BADBIR Protocol Version 12 March 13th 2008
• Number of doses and total Joules (PUVA) for the episode • Has your patient experienced an adverse event or new illness? • Adverse event detail (allow for input of several) • Specific prompt for skin cancer and non-cutaneous cancer and tuberculosis or serious • Was patient on biological therapy at the time of onset of event? • Did this result in death, hospitalisation, loss of function, significant disability, congenital malformation or was in any other way life threatening? • Do you believe that there is a reasonable possibility that this event was related to the patient's • Was a yellow card filled in for the adverse reaction?
8.2 Current psoriasis severity

• PASI • Physicians global assessment (PGA) (See appendix 5) • Patients with a rheumatologist's diagnosis of inflammatory arthritis will have the HAQ assessed at baseline and subsequent visits 8.3 Vital status

• Contact details for person completing the form • Add any missing data from registration • Adverse events will be classified according to the new pharmaceutical standard MedDRA • If pregnancy occurs follow up to include outcome
8.4 Data acquired directly from patients at follow up

• Any new hospital referrals and reason Y/N
• If yes name of hospital consultant in charge and reason • Any new hospital admissions and reason Y/N • If yes name of hospital, consultant in charge and reason • Euroqol

9.0 Analysis of the data
9.1 Primary endpoints for evaluation
BADBIR Protocol Version 12 March 13th 2008
9.2 Hypotheses to test
Increased risk is related to the duration of therapy Baseline characteristics determine increased risk, especially prior therapy Certain longitudinal combinations of treatment carry higher risks In addition the benefits of therapy will be assessed using a variety of objective scores, PGA and PASI, and quality of life DLQI, Euroqol EQ-5D. 9.3 Analytic approach
The initial analyses will consist of comparisons in baseline status between the individuals in the treatment cohorts. For the purposes of analysis (initially) follow up time will be censored in both cohorts if there is switching to another class of biologic therapy and censored in the standard therapy cohort if there is switching to a biologic agent. The adverse events of interest are calculated per person time of follow up, following the start of therapy. Depending on the events, separate analyses are undertaken (i) restricting consideration to time on drug, which includes the period within 90 days of last injection and (ii) all person time following start of therapy e.g. malignancy. Time-dependent regression analyses will be undertaken to compare event rates between groups after adjusting for 9.4 Interim Analyses

Interim analyses will be undertaken at appropriate intervals when 5000 person years of exposure have
been accumulated in any of the exposed groups. Such analyses will be a guide to the ultimate levels of recruitment and length of follow up required. Decisions as to the timing of publications and the need for continued follow up and/or recruitment can only be taken in the light of results from such analyses. A Data Monitoring and Ethics Committee (DMEC) has been established, analogous to a Data Safety & Monitoring Board established for major clinical trials. The DMEC will be independent of the principal investigators and also of any of the pharmaceutical industries involved, and will have the power to request interim analyses and advise on the timing and nature of any publications. The DMEC should include at least one epidemiologist, a dermatologist and a statistician. 10. Roles of interested parties
The BAD will seek funding and a generic contract with the pharmaceutical companies whose products are being monitored. The University of Manchester will be the sponsor of the study. BAD will have ownership of the data. The project will be steered by a steering group and data monitoring and ethics committee (DMEC) under the auspices of the BAD and will operate independently from direct industry BADBIR Protocol Version 12 March 13th 2008
involvement. The manufacturer of a product will have access to aggregated data from subjects exposed to their product but not to named individual data.
10. 1 Role of the Pharmaceutical companies
The goals of industry and the dermatological community are similar in seeking accurate estimates of
any increased risk of serious adverse events. It may also be a pre-requisite for drug licence approval, that a study such as the one proposed is established. It is accepted that it is beneficial that any study, such as the one proposed, should be independent of any direct industry involvement. Thus decisions on analyses, interpretation and publication should be independent of any industrial contribution. Industry can have a crucial role in stimulating registration after licensing, and also contributing their experience into the nature and type of data to be collected. Timely serious adverse event data will be shared with the relevant manufacturer according to agreed standardised protocols. Aggregated data relating to a particular product will be shared with industry in confidence, though individual identifiable patient data will not be released. A participant company has the option of requesting specific analyses and will be shown drafts of any publications, reports, abstracts or other material prior to submission for presentation or publication. They can ask for clarifications or amendments to such material but the final decision on these would rest with the principle investigators and the DMEC. All the principal investigators and members of the DMEC have to complete an annual 'Declaration of conflict of interests', which will be added to all publications. There will be an annual joint pharmaceutical companies meeting to discuss contractual issues and

10.2 Role of BAD
BAD will be the owner of the data that emerge from the study. The study co-ordinator will report on a
quarterly basis to such committees or sub-committees that BAD deems appropriate. The membership of the DMEC will be subject to the approval of BAD. Boffetta, P., G. Gridley, and B. Lindelof. "Cancer risk in a population-based cohort of patients hospitalized for psoriasis in Sweden." J Invest Dermatol 117.6 (2001): 1531-37. ---. "Cancer risk in a population-based cohort of patients hospitalized for psoriasis in Sweden." J Invest De, Bandt M., et al. "Systemic lupus erythematosus induced by anti-tumour necrosis factor alpha therapy: a French national survey." Arthritis Res.Ther. 7.3 (2005): R545-R551. BADBIR Protocol Version 12 March 13th 2008
Gelfand, J. M., et al. "Lymphoma rates are low but increased in patients with psoriasis: results from a population-based cohort study in the United Kingdom." Arch.Dermatol 139.11 (2003): 1425-29. Gladman, D. D., et al. "Mortality studies in psoriatic arthritis: results from a single outpatient center. II. Prognostic indicators for death." Arthritis Rheum. 41.6 (1998): 1103-10. Gomez-Reino, J. J., et al. "Treatment of rheumatoid arthritis with tumor necrosis factor inhibitors may predispose to significant increase in tuberculosis risk: a multicenter active-surveillance report." Arthritis Rheum. 48.8 (2003): 2122-27. Hannuksela-Svahn, A., et al. "Psoriasis, its treatment, and cancer in a cohort of Finnish patients." J ---. "Psoriasis, its treatment, and cancer in a cohort of Finnish patients." J Invest Dermatol 114.3 Jacobsson, L. T., et al. "Treatment with tumor necrosis factor blockers is associated with a lower incidence of first cardiovascular events in patients with rheumatoid arthritis." J.Rheumatol. 32.7 (2005): 1213-18. Lindelof, B., et al. "PUVA and cancer risk: the Swedish follow-up study." Br J Dermatol 141.1 (1999): Ljung, T., et al. "Infliximab in inflammatory bowel disease: clinical outcome in a population based cohort from Stockholm County." Gut 53.6 (2004): 849-53. Mallbris, L., et al. "Increased risk for cardiovascular mortality in psoriasis inpatients but not in outpatients." Eur.J Epidemiol. 19.3 (2004): 225-30. Margolis, D., et al. "The risk of malignancy associated with psoriasis." Arch.Dermatol 137.6 (2001): Olsen, J. H., H. Moller, and G. Frentz. "Malignant tumors in patients with psoriasis." J.Am.Acad.Dermatol. 27.5 Pt 1 (1992): 716-22. Paul, C. F., et al. "Risk of malignancies in psoriasis patients treated with cyclosporine: a 5 y cohort study." J Invest Dermatol 120.2 (2003): 211-16. Poikolainen, K., J. Karvonen, and E. Pukkala. "Excess mortality related to alcohol and smoking among hospital-treated patients with psoriasis." Arch.Dermatol 135.12 (1999): 1490-93. Prodanowich, S., et al. "Methotrexate reduces incidence of vascular diseases in veterans with psoriasis or rheumatoid arthritis." J.Am.Acad.Dermatol. 52.2 (2005): 262-67. Stern, R. S., K. T. Nichols, and L. H. Vakeva. "Malignant melanoma in patients treated for psoriasis with methoxsalen (psoralen) and ultraviolet A radiation (PUVA). The PUVA Follow-Up Study." N.Engl.J Med. 336.15 (1997): 1041-45. Voss, A., A. Green, and P. Junker. "Systemic lupus erythematosus in Denmark: clinical and epidemiological characterization of a county-based cohort." Scand.J.Rheumatol. 27.2 (1998): 98-105. Wolfe, F. and K. Michaud. "Lymphoma in rheumatoid arthritis: the effect of methotrexate and anti- tumor necrosis factor therapy in 18,572 patients." Arthritis Rheum. 50.6 (2004): 1740-51. Wolfe, F., et al. "Tuberculosis infection in patients with rheumatoid arthritis and the effect of infliximab therapy." Arthritis Rheum. 50.2 (2004): 372-79. BADBIR Protocol Version 12 March 13th 2008
---. "Tuberculosis infection in patients with rheumatoid arthritis and the effect of infliximab therapy." Wong, K., et al. "Mortality studies in psoriatic arthritis: results from a single outpatient clinic. I. Causes and risk of death." Arthritis Rheum. 40.10 (1997): 1868-72. ---. "Mortality studies in psoriatic arthritis: results from a single outpatient clinic. I. Causes and risk of death." Arthritis Rheum. 40.10 (1997): 1868-72. BADBIR Protocol Version 12 March 13th 2008
Appendix 1 Statistical Power and numbers converted to patient years
Number pt years required in each cohort (controls and biologics)
Shading indicates likely power of the register in 5years (dark) to 10 years (light)
Using stat calc (epi info) 95% confidence level 80% power 1 to 1 ratio in each cohort Chart of accrual of patient years given scenario of 1000 per year patients registered on biologics or 500 per year registered runs to 14 years which may be relevant for longer term e.g. melanoma data. BADBIR Protocol Version 12 March 13th 2008

Source: http://www.badbir.org/Website%20data/Protocol/BADBIR%20Protocol%20v12.pdf