NATIENS: Optimal Management and Mechanisms of SJS/TEN

Background and Scientific Rationale Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN), are severe, life-threatening immunologically mediated adverse drug reactions representing the same disease across a spectrum of severity that affect 60,000 patients per year globally at an incidence of 5 cases per 1,000,000 in the United States. This study has been preceded by a planning phase to ensure testing and development of standardized infrastructure and operating procedures across sites. The scientific rationale for this study is the lack of evidence based treatment for SJS/TEN. The study will aim to establish the most clinically effective therapy for SJS/TEN, as there is currently no level 1A evidence for any treatment above aggressive supportive care which still has equipoise as the standard of care. A multi-centered, three-arm, double-blind double dummy randomized controlled trial with a planned enrollment of 267 patients over 6 years will be undertaken to evaluate which of supportive care, cyclosporine or etanercept leads to the shortest time to complete re-epithelialization. The controlled setting of the clinical trial provides the basis for which samples can be sequentially collected to answer important mechanistic questions. Samples collected in the course of the study will include DNA, RNA, plasma, serum, blister fluid cells and supernatant, sloughed epidermis and punch biopsies of skin. Samples will be biobanked to do immediate targeted high resolution HLA sequencing, cytokine profiling and single cell multidimensional analyses to identify biological markers that have the potential to predict risk and outcome of SJS/TEN. A pharmacokinetic study to assess the disposition of cyclosporine and etanercept will be done on a subset of patients. Samples will be biobanked for later analysis for other transcriptomic, proteomic whole genome studies. These mechanistic studies will allow us to gain important insights into the immunopathogenesis of SJS/TEN. Our study will be the first to examine in a blinded randomized controlled design both management and mechanisms of SJS/TEN. We anticipate that this will lead to new ways to prevent, diagnosis and treat SJS/TEN, and will create a roadmap and evidence-base for studies in serious immunologically mediated adverse drug reactions and other immunologically mediated diseases.

Interventions Supportive care is the current accepted standard of care of SJS/TEN. Systemic treatment of SJS/TEN and the preferred agents remain a matter of debate and contention. Observational studies performed over the last 15 years suggest that cyclosporine and etanercept may be promising treatments.

1. Study Hypotheses/Objectives 1.1. Hypotheses Our primary hypothesis is that both cyclosporine and etanercept will show benefit over supportive care alone and that etanercept 50 mg subcutaneously given at study day 1 and repeated at day 4. We further hypothesize that we will discover a number of genetic and biological predictors of both occurrence and prognosis of SJS/TEN including genetic associations (HLA and others), cytokine biomarkers that predict that course and likelihood of re-epithelialization and single cell multidimensional studies that will identify promising targets for prognosis and treatment.

1.2. Primary Objective(s) Our primary objective is to conduct a 25-site three-arm randomized double-blind controlled stratified multicenter phase III study to determine whether two therapeutic interventions - cyclosporine and etanercept will improve short-term outcomes associated with Stevens-Johnson Syndrome and toxic epidermal necrolysis (SJS/TEN) over harmonized best supportive care alone. Our hypothesis is that both etanercept and cyclosporine will show benefit over supportive care alone and that etanercept 50 mg subcutaneously on study day 0 and repeated 72 hours later on study day 3 for all subjects will show significant benefit over both cyclosporine and supportive care alone. The justification for a second dose is the short half-life of the drug and the usual twice a week dosing in other skin conditions such as psoriasis and a recent non-blinded randomized study of etanercept in SJS/TEN that used repeat dosing of etanercept.

Secondary Objective(s) Secondary objectives of this study are to determine the acute secondary outcomes and secondary clinical outcomes at 3 and 12 months (see secondary outcomes) and to determine the molecular and cellular mechanisms of SJS/TEN through collection of timed DNA, PBMC, RNA, blister fluid and blister fluid cells and skin. We hypothesize that patients will have significant sequelae identified at 3 and 12 months that will differ between treatment arms. We hypothesis that treatment interventions will significantly impact cytotoxic and cytokine signals and that these biomarkers will correlate with primary and secondary outcomes. We further hypothesize that significant genetic associations (in particular class I HLA) will be found in association with drug induced SJS/TEN.

2. Study Design 2.1. Description of Study Design This is a three-arm, randomized, double-blind double dummy phase III multicenter placebo-controlled trial comparing cyclosporine, etanercept and best supportive care for the treatment of SJS/TEN. A total enrollment of 267 subjects (89 per arm) over 24 site is planned. The sample size is based on a 90% power to determine the primary outcome and a minimum clinically important difference (MCID) of one day for full re-epithelialization of the skin. The study will enroll from multiple sites due to the low SJS/TEN incidence at individual sites and the need to identify patients with very specific inclusion criteria including those who have had no more than a 5 day window from onset of symptoms to presentation. We aim to complete enrollment over a period of 6 years.

Consenting adult subjects (≥18 years of age) who meet clinical criteria for diagnosis of SJS/TEN will undergo randomization stratified by baseline SJS/TEN prognosis (SCORTEN) and will be allocated 1:1:1 to cyclosporine plus best supportive care, etanercept plus best supportive care or best supportive care alone. Randomization will occur by a secure central online, computer-generated random randomization tool in REDCap using permuted blocks of size 3 and 6.

The active treatment period is from randomization until re-epithelialization is achieved. This is expected to be at most 21 days. The primary outcome is time (days) to full re-epithelialization of the skin. Subjects will be followed in hospital during their admission; patients with SJS/TEN in this study will be on active drug or placebo intravenously for at least 14 days. The plan will be the earliest discharge from hospital to occur at completion of the full 14 days of placebo versus active treatment if full re-epithelialization has occurred at that time but not until complete re-epithelialization. Death is a competing risk for the primary outcome (see statistical plan section 4.4). Subjects will be followed for 3 months and 12 months after complete re-epithelialization. Under circumstances where patients are inadvertently discharged prior to complete re-epithelialization appropriate and comparable follow-up to provide documentation of appropriate wound, eye and urogynecological care and of the timing of complete re-epithelialization at a clinical research center at supportive sites and/or the Hawthorne effect clinical trials services (see operations manual).

2.2. Stratification, Randomization, and Blinding/Masking

Randomization will occur by a secure central online, computer-generated random number system in REDCap. Randomization will be stratified by baseline SJS/TEN prognosis (SCORTEN) and will occur in randomly varying blocks of size 3 and 6. Subjects will be allocated 1:1:1 to cyclosporine plus best supportive care, etanercept plus best supportive care or best supportive care alone. Patients, treating physicians and outcome assessors will be blinded to the allocated treatment. All patients will receive supportive care, either with double placebo or with their study drug (cyclosporine or etanercept) plus placebo. The central randomization scheme will provide the site pharmacist with a unique drug code corresponding to either study drugs and matching placebos. These will be kept in identical, coded containers. Only the study central pharmacy and the site pharmacy who will prepare the preparations for blinding will be aware of the coding scheme. Masking will be carried out as follows:

- Cyclosporine: Study drug given as per protocol for 2 weeks. A single SC injection given on study day 0 of admission and study day 3.

- Etanercept: Study drug given as per protocol on study day 0 and 3. Receive IV normal saline for 2 weeks as to mimic the cyclosporine schedule.

- Supportive therapy: A single saline SC injection on study day 0 and study day 3 of admission and an additional 14 days of IV normal saline placebo to mimic the cyclosporine schedule 2.2.1. Procedure for Unblinding/Unmasking Unblinding must be approved by the study medical monitor unless an immediate life-threatening condition has developed, and the medical monitor is not available. The site investigator will notify the protocol chair(s) and the study statistical and clinical coordinating center of the unblinding event by the next business day. The emergency unblinding will also be reported to the Data and Safety Monitoring Board (DSMB). A full account of the event will be recorded, including the date, time of the unblinding, the reason or the decision to unblind and the name of the individual who made the decision and the names of the Medical Monitor and others who were notified. The reasons for unblinding of a participant's treatment will be included in the final study report. Unblinding the study due to an approved interim analysis, final analysis or study termination will required written approval from the institute sponsor (NIAID).

3. Selection of Participants and Clinical Sites/Laboratories 3.1. Rationale for Study Population Subject enrollment will reflect the population at large in our participating centers. All subjects who meet inclusion criteria will be screened for enrollment into the study. We will not discriminate based on sex, race, or ethnicity. Females, including those of childbearing potential, will be included in the trial.

Supportive care High-quality supportive care is always used in the treatment for SJS/TEN. We know it can decrease the chance of serious complications and death even in the absence of adjuvant medications to treat the condition.

There are no significant side effects associated with supportive care aside from possible discomfort with dressing changes.

3.5. Risks of Other Protocol Specified Medications Cyclosporine

The risks of venipuncture (bruising at the site and very low risk of infection) apply as cyclosporine is an intravenous medication. There is extensive experience with this medication in SJS/TEN for treatment of other conditions. In general, few side effects have been seen when used for SJS/TEN. When they have occurred, they have had no long-term impact on the person receiving the medication. Risks include:

Less likely (5% to 20%):

- Mild changes in kidney function: 1-6% had mild changes in their kidney function that required a change of dose; Serious infections were not more common than those treated with short-term use, as proposed in this study. Flu-like illness symptoms and a predisposition for lung infections have been experienced in 8- 10% of patients with long-term use (months to years).

Rare (1% to 4%):

• Hypertension: 1-3% needed to have their blood pressure lowered. Very rare but serious (less than 1%):Severe liver or kidney problems may be permanent in those taking cyclosporine for greater than 1year. Patients receiving cyclosporine under the NATIENS protocol will undergo treatment for a short amount of time; therefore, the chance of this occurring in this study is extremely unlikely. Close monitoring for hepatic or renal dysfunction is built into the NATIENS study protocol.

Discontinuing the study medication in the case of imminently organ-threatening effects, type I allergy, or rare immune reactions (e.g., thrombotic thrombocytopenic purpura] in the case of cyclosporine) will be documented. Cyclosporine would be the only arm where this may occur as the protocol is for 2 weeks of therapy.

An investigator and research coordinator will be available 24 hours a day from both the Vanderbilt Coordinating Center and Ottawa Coordinating Center. The consent form has contact numbers for reaching study personnel. In the event an investigator or research coordinator is on vacation or out-of-town, appropriate coverage will be provided.

Safety of the participant will remain of primary importance. All patients will undergo intensive monitoring while in hospital in accordance with usual ICU and Burn Unit practices.

3.6. Risks of Study Procedures Venipuncture and Blood Samples: The protocol requires patients to have blood drawn for researchmpurposes. The risks of drawing blood are uncommon and may include bleeding, minor infection, and bruising. Blood draws may be painful. Some people may feel faint or dizzy. The amount of blood drawn represents a small percentage of the amount of the total blood volume and will not represent a significant risk to the patient. Samples are planned to align with blood samples taken in the course of usual clinical care to minimize the need for repeat venipuncture. Blood sampling is performed by trained nurses or phlebotomists with experience performing venipuncture. We have also aligned our study-related blood samples with those taken for usual patient care. Multiple tests are also performed on blood from the same sample tubes. This minimizes the volume of blood required to complete Aims 2 and 3 as well as significantly reduces the number of needle sticks.

Tissue Samples: Tissue samples will be taken to understand the disease process of SJS/TEN. A sample of healthy skin, where possible, will be taken for comparison. Where samples of normal skin can be taken, this will be discussed with the patient. Some patients have little healthy skin to sample. Risks primarily include minor bleeding, a small chance of infection, and a small scar.

All tissue sampling will be done by an experienced physician to reduce the risk of complications.

Salivary Swab: There are no significant risks associated with the saliva samples being collected. In those with oral mucosal involvement due to SJS/TEN there may be irritation or a small degree of discomfort while producing a sample.

Blister Fluid Collection: This process involves aspirating fluid from an intact blister. There is a very low risk of infection, but there is no discomfort with the procedure.

4. Investigational Agent /Device/Intervention (see investigators brochure, section 5) 4.1. Drug Accountability Under Title 21 of the Code of Federal Regulations (21CFR §312.62) the investigator will maintain adequate records of the disposition of the investigational agent, including the date and quantity of the drug received, to whom the drug was dispensed (participant-by-participant accounting), and a detailed accounting of any drug accidentally or deliberately destroyed.

Records for receipt, storage, use, and disposition will be maintained by the study site. A drug-dispensing log will be kept current for each participant. This log will contain the identification of each participant and the date and quantity of drug dispensed. All records regarding the disposition of the investigational product will be available for inspection. Any remaining product will be destroyed in accordance with pharmacy regulation.

4.2. Assessment of Participant Compliance with Investigational Agent All medications will be administered by the investigator and/or study staff in the Burn Unit. Administration of study medication(s) will be witnessed by the investigator and/or study staff. The total volume of study medication infused will be compared with the total volume prepared to determine compliance with each dose administered.

4.3. Toxicity Prevention and Management Etanercept: No modifications are permitted. Cyclosporine: The study is blinded and the adjustment and stopping rules that primarily apply to cyclosporine will be carried out by an unblinded pharmacists according to well defined rules. A table will be provided for anticipated severe drug interactions with cyclosporine and for patents where those drugs are necessary these patients may be excluded from the study. Non-serious medication side-effects such as diastolic hypertension or mild renal failure will be medically managed while still maintaining blinding. For example, standard adjustments for cyclosporine will be made based on serum creatinine and blood pressure monitoring. These orders will be written and carried out by an unblinded pharmacist. The adjusted dose will appear on the IV bag cyclosporine/placebo and the treating team will remain blinded as to the treatment arm of the patient. Creatinine will be monitored daily by a non-blinded pharmacist at each site for all patients. A standard dose-adjustment algorithm to maintain a target creatinine of <130% above baseline will be used. Down-dosing of 25-30% will be used if diastolic blood pressure >100 mm. If >3 dose adjustments, 48-hours apart fail to control rises in creatinine beyond this threshold, cyclosporine be stopped. Any study treatment will also be stopped in the presence of any grade 3-4 toxicity (e.g. leukoencephalopathy) or creatinine >150% of initial value and diastolic blood pressure >110 or systolic persistently > 160 mmHg which is a common indicator of cyclosporine toxicity.

6.5 Premature Discontinuation of Investigational Agent Patients who discontinue any study medication will be analyzed by intent-to-treat analysis and followed for complications of SJS/TEN as per study protocol. Stopping for lack of efficacy will occur according to pre-described stopping rules. Study therapy may be prematurely discontinued for any participant for any of the following reasons:1. Grade 3 or 4 toxicity 2. Toxicity that has not responded to clinical management (see cyclosporine above); 3. Necessary medication that interacts with cyclosporine (as per protocol table) 4. Patient request. All patients are administered the drug in the ICU and are monitored around the clock for adverse signs and symptoms by the investigator and/or study staff. If adverse signs are observed, patients will be evaluated by physicians from various specialties including intensive care, and dermatology. Following evaluations, decisions will be taken to decrease, adjust, or stop the dosage. Each site will be individually managed for treating adverse reactions.

Study therapy may also be prematurely discontinued for any participant if the investigator believes that the study treatment is no longer in the best interest of the participant.

5. Other Medications 5.1. Concomitant Medications 5.1.1. Protocol-mandated 1% lidocaine with epinephrine at a dilution of 1:100,000 as local anesthesia at the time of obtainment of biopsy samples (for research biospies) 5.1.2. Other permitted concomitant medications All topical agents for wound and eye care are permitted as outlined in the respective supportive care protocols.

5.2. Prophylactic Medications Any non-interacting medications (as per cyclosporine protocol) may be administered at the discretion of the clinical team.

5.3. Prohibited Medications 1) Medications that interact with cyclosporine where there are suitable alternatives20,21; 2) immunomodulatory agents other than those used in this clinical trial

The following medications will be considered to be contraindicated during the course of the study. If patients are on these medications and an alternative is available it should be substituted:

Aliskiren, atorvastatin, bosentin, cholic acid, cladribine, conivaptan, crizotinb, dronedarone, elagolix, enalutamide, eplerenone, foscarnet, any new drug found to be a significant CYP3A4 inhibitor or p-glycoprotein/ABCB1 inhibitor). A complete list of interacting drugs, potentially interacting drugs is included in the operations manual.

5.4. Rescue Medications Cyclosporine increases blood pressure. Blood pressure will be managed by administering non-interacting blood pressure lowering drugs at the discretion of the treating physician. Cyclosporine could also decrease renal function. Patients will be managed as per standard of care protocols for patients with acute renal failure (see operations manual for detailed procedures).

6. Study procedures (for laboratory procedures and sampling schedule, see section 5, comprehensive laboratory plan) Enrollment The research study will be explained in lay terms to each potential research participant. The potential participant will sign an informed consent form before undergoing any study procedures. Once informed consent has been signed and completed by both the participant and key study personnel, the participant is considered enrolled in the study and will be assigned a unique participant number. Consent will be obtained at the time of clinical encounter or diagnosis. To initially identify patients for consent, medical records may be reviewed for suitability or the admitting service (e.g., emergency, intensive care unit, burn surgery) may identify suitable patients to the study team. Then, a health care professional from the research team of key personnel will discuss the study with prospective participants (or substitute decision-maker) or obtain permission for the research personnel to discuss the study. In rare cases where emergency treatment is required (e.g., altered level of consciousness, lack of capacity), enrollment may still occur; however, confirmation of consent will be required by the research team at the first opportunity after the patient is initially managed or restoration of capacity occurs. Consent will be obtained from a substitute decision-maker or the patient at the earliest possible moment and the subject will be reconsented when their capacity for medical decision making returns. A copy of the patient's consent will be kept in the patient's chart. It is available in English and Spanish. If consent is withdrawn at any time, it should be determined to what aspect of the study consent is being withdrawn (e.g., consent to the use of genetic samples or the study itself). In case of full withdrawal of consent, the treating team may modify any of the study procedures as per their medical judgment.

8.1 Screening/Baseline Visit The purpose of the screening period is to confirm eligibility to continue in the study.

The following procedures, assessments, and laboratory measures will be conducted to determine participant eligibility:

1. Research staff will screen potential participants upon being notified of their admission to the ICU or Burn ICU based on the study inclusion/exclusion criteria. (so as written here - the subjects will be screened and consented here).

2. A diagnostic punch biopsy will be obtained from the patient on the day of admission by the institution's appropriate staff. (is this a clinical biopsy or a research biopsy)

3. The results of the diagnostic biopsy will be reviewed by research staff to determine eligibility to continue in the study within 2-3 days after admission.

Randomization: will occur by a secure central online, computer generated random number system. Randomization will be stratified by baseline SJS/TEN prognosis (SCORTEN) and will occur in randomly varying blocks of size 3 and 6. Subjects will be allocated 1:1:1 to cyclosporine plus best supportive care, etanercept plus best supportive care or best supportive care alone. Continuation of the patient in the study will be contingent on a compatible clinical picture in combination with supportive pathology. Since pathology can take 48 hours to come back however subject randomization will occur prior to the availability of the biopsy results.

Patients, treating physicians, and outcome assessors will be blinded to the allocated treatment. The central randomization scheme will provide the site pharmacists with a unique drug code corresponding to either study drugs or matching placebo. These will be kept in identical, coded containers. Only the study central and site pharmacy will be aware of the coding scheme. Masking will be carried out as follows: 1) Cyclosporine: study drug given as per protocol for 2 weeks. A single SC injection given on Day 1 of admission. With a second injection given on day 4 of admission.; 2) Etanercept: Study drug given as per protocol. Receive IV normal saline for 2 weeks as to mimic the cyclosporine schedule. A subcutaneous injection of etanercept given on study day 1 and study day 4; 3) Supportive therapy: A single saline SC injection on Day 1 and Day 4 to mimic etanercept and normal saline for 14 days to mimic the cyclosporine schedule.

8.2 Study Visits or Study Assessments

Also as listed in research plan, facilities and resources and detailed laboratory plan).

8.3 Unscheduled Visits (following discharge between 3 and 12 month follow-ups) Following discharge the patient will be discharged to the care of their clinical team that will generally include follow-up with dermatology and ophthalmology at a minimum and if disease activity increase or other concerns arise they will be asked to liaise with the clinical team however study personnel should also be contacted and the participant may be asked to return for an "unscheduled" visit.

9. Biospecimen Storage (see comprehensive laboratory plan) 10. Criteria for Participant and Study Completion and Premature Study Termination

a. Participant Stopping Rules and Withdrawal Criteria

Participants may be prematurely terminated from the study for the following reasons:

1. The participant elects to withdraw consent from all future study activities, including follow-up.

2. The participant is "lost to follow-up" (i.e., no further follow-up is possible because attempts to reestablish contact with the participant have failed). [use specific criteria to define "lost to follow-up"]

3. The participant dies.

4. The participant is withdrawn from the study.

5. Individual safety stopping rules as defined under the study drugs section.

Study stopping rules:

We plan two interim analyses to allow early stopping for efficacy. These analyses will take place when we have acquired outcome measures for one third and for two thirds of the total target sample. We will use the Haybittle-Peto rule applied to each of the three pairwise Mann-Whitney comparisons of time to re-epithelialization between arms. As with the primary analysis, patients who die prior to achieving re-epithelialization will be allocated the maximum observed time to re-epithelialization plus one, so as to ensure that death is treated as worse than any time to successful treatment. One arm will be declared inferior to another if the p-value for the difference is less than 0.001. No adjustment will be made to the final analysis for interim testing. If one arm is found to be inferior to another arm, the Data Safety Monitoring Board will be instructed to strongly consider stopping recruitment to that arm. If one arm is found to be inferior to both other arms, the Data Safety Monitoring Board will be instructed to immediately stop recruitment in the inferior arm. If the other two arms also differ with a p-value less than 0.001, the recruitment will be stopped in all arms. We will have no formal stopping rules for safety outcomes different from time to re-epithelialization, but the Data Monitoring Committee will be explicitly mandated to carefully consider any safety signals with a view towards stopping recruitment if any expected or unexpected safety signal emerges.

b. Participant Replacement If a participant withdraws or is withdrawn from the study, no further data or samples will be collected. The participant will not be replaced.

c. Follow-up after Early Study Withdrawal Participants who withdraw from the study will be taken out of the study and no further data or samples will be collected. Study drugs and interventions will no longer continue after the patient is withdrawn from the study. Data and samples collected during the time of participation in the study will still be kept for study analysis.

11. Safety Monitoring and Reporting (see data safety monitoring plan section 3.3)