Severe Bullous Drug Eruption and Filgrastim (GNET)

Evaluating the Therapeutic Efficacy of Filgrastim in Severe Bullous Drug Eruptions (Lyell and Stevens-Johnson Syndromes)

Toxic epidermal necrolysis (TEN) including Stevens Johnson (SJS) and Lyell syndromes represent the most severe drug eruptions. It is an allergic disorder caused by cytotoxic T lymphocytes, specific of drugs, responsible for the destruction of keratinocytes by apoptosis. Regulatory T cell (CD25 high CD4+), normally responsible for controlling the activation of cytotoxic T lymphocytes, have altered function. Despite the progress made in the pathophysiological understanding of TEN, there is currently no effective treatment. The main symptom is bullous and skin peeling > 10% giving the appearance of great burns. The death rate is estimated between 30 and 40% due to visceral inflammatory injuries and bacterial superinfection. The risk of mortality is estimated during the initial treatment by calculating the SCORTEN (mortality>10% if SCORTEN>2, mortality>90% if SCORTEN>5). The morbidity is also very important (92% at 1 year), especially ophthalmologic with high risk of blindness... The therapeutic potential of G-CSF (Granulocyte-Colony Stimulating Factor) in TEN is supported by several observations. The G-CSF promotes skin healing. This has been shown in human burns, with a significant reduction in healing time under G-CSF. The mechanisms associate the growth factor effect on keratinocytes, macrophages stimulation and metalloprotease activity allowing tissue remodeling limiting sequels onset. Otherwise, healing altered in deficient G-CSF mice is corrected by the growth factor injection. The G-CSF is an immunomodulator whose activities appear to justify use in TEN : Polarization of immune response to Th2 non-cytotoxic (anti Th1), Preferential differentiation of naive LT (T lymphocytes) in regulator LT (CD25 high CD4+) and mobilization of regulator LT of the spinal cord to altered tissues. The G-CSF was used in a few cases of TEN with great efficacy. No data is available concerning sequels of SJS/TEN in treated patients. This clinical trial program, by providing proof of the efficacy of filgrastim in SJS/TEN, should allow progress in care of this serious toxics diseases. In the future, it could thus reduce the significant morbidity of these syndromes with a high rate of sequelae.

Toxic Epidermal Necrolysis (TEN), Stevens-Johnson syndrome (SJS), Lyell's syndrome, Ganulocyte-Colony Stimulating Factor (G-CSF), Filgrastim, Randomized controlled trial (RCT)

This is a single blind patient trial. The outcomes assessor of the primary endpoint outcome, the adjudication committee and the statistician, will also be blinded to the allocated treatment.

Standard symptomatic treatment of the SJS/NET (until the reepidermization of the patient) + ZARZIO@ (30 MU/0,5mL and/or 48 MU/0,5mL - solution of 20 ml diluted in GLUCOSE 5%) administrated by IV or subcutaneous route over a period of 5 consecutive days (1 injection per day during 30 minutes)

Standard symptomatic treatment of the SJS/NET (until the reepidermization of the patient) + 20 ml GLUCOSE 5% administrated by IV route over a period of 5 consecutive days (1 injection per day during 30 minutes)

Injection of ZARZIO 30 MU/0,5mL and/or ZARZIO 48 MU/0,5mL, over a period of 5 consecutive days (1 injection per day during 30 minutes - - day 1 : set up standard treatment). The filgrastim solution will be diluted in 20 mL of 5% Glucose. The dose of ZARZIO administrated depends of the patient's weight : 20 to < 30kg = 0,3 mL of ZARZIO 48 MU/0,5mL (subcutaneous route) 30 to < 60kg = 0,5 mL of ZARZIO 30 MU/0,5mL (by IV) 60 to < 90kg = 0,5 mL of ZARZIO 48 MU/0,5mL (by IV) 90 to < 120kg = 2x0,5 mL of ZARZIO 30 MU/0,5mL (by IV) 120 to 150kg = 0,5 mL of ZARZIO 30 MU/0,5mL + 0,5 mL of ZARZIO 48 MU/0,5mL (by IV) > 150kg = 2x0,5 mL of ZARZIO 48 MU/0,5mL (by IV)

Injection of 20 mL of Glucose 5% solution over a period of 5 consecutive days (1 injection per day during 30 minutes - day 1 : set up standard treatment). The dose given is equivalent to that used for filgrastim : 20 to < 30kg = placebo not available because the injection must be done subcutaneously so the blind cannot be respected. 30 to < 60kg = 20mL of Glucose 5% solution (by IV) 60 to < 90kg = 20mL of Glucose 5% solution (by IV) 90 to < 120kg = 20mL of Glucose 5% solution (by IV) 120 to 150kg = 20mL of Glucose 5% solution (by IV) > 150kg = 20mL of Glucose 5% solution (by IV)

Comparison between the 2 arms of the proportion of patients with arrest of progression of skin detachment defined by a peeled surface and/or bullous and/or associated at a NIKOLSKY sign determined according to the stable burn table - Evaluation done by the Outcomes Assessor and the adjudication committee.

Comparison between the 2 arms of the time until the arrest of the SJS/NET progression defined by a peeled surface and/or bullous and/or associated at a NIKOLSKY sign determined according to the stable burn table - Evaluation done by the Outcomes Assessor and the adjudication committee.

Comparison between the 2 arms of the time until the complete re-epidermization defined by the disappearance of skin rashes and return to normal skin. Evaluation done by the Outcomes Assessor and the adjudication committee.

Comparison between the 2 arms of the number days of hospitalization corresponding to the duration of SJS/NET episode's hospitalization.

Number of clinical and biological adverse event leading to stop the filgrastim treatment (per patient, nature and grade of the AE (adverse event ) - repository CTCAE version 5.0)

Comparison between the 2 arms of the number of clinical and biological adverse event (by organ systems - repository CTCAE version 5.0)..

Study of the nature, cumulative doses and indications objectifying a corticosteroid prescription. A comparison between the 2 arms will be done.

Determination of the proportion of patients with at least one sequel among the following specialty : ophthalmology, stomatology/ORL(otorhinolaryngology), gastroenterology, gynecology, urology and psychiatry, researched during specialized consultations during the remote monitoring phase. A comparison between the 2 arms will be done.

Quality of life evolution assessed with Dermatology Life Quality Index (DLQI) or Children's Dermatology Life Quality Index (CDLQI). DLQI is calculated by adding the score of each question, resulting in a maximum of 30 and a minimum of 0. The higher the score, the more quality of life is impaired. CDLQI is used in the same way. It is documented in 5 minutes. Each question is illustrated by a cartoon based on the theme of the question, making it more fun for younger children. The scoring for each question is "Very much"[Score = 3], "Quite a lot"[2], "Only a little"[1], "Not at all"[0], "Blank"[0], and Q7 - "prevented school"[3]. The 10 individual question scores are summed to provide a total CDLQI score. The maximum possible score is 30, indicating maximum impact on QoL. A comparison between the 2 arms will be done.

Determination of the proportion of patients with high risk of developing PTSD measured with Impact of Event Scale Revised (IES-R) or Children's Revised Impact of Event Scale (CRIES). IES-R is a 22-item self-administered questionnaire that measures symptoms of intrusion, avoidance and over-arousal experienced during the past 7 days. Patients rate each item on a 0-5 point Likert scale. The total score, between 0 and 88, is obtained by adding all the ratings for each item: low risk (score less than 12), moderate (score between 12 and 34 inclusive) or at high risk (score strictly greater than 34) of developing PTSD. CRIES-13 is a revised scale to measure the impact of events on children from 8 years old. Items are rated on a 4-point scale, with scores of 0, 1, 3 and 5 points. The maximum score is 65. The score is obtained by adding the responses to each item. A score of 30 or more would be effective in screening for PTSD. A comparison between the 2 arms will be done.

Inclusion Criteria: Patient aged of 6 years old or more, presenting SJS/TEN, drug or infectious origin proofed and very strongly suspected (indirect certainty argument), confirmed by evaluator. SJS or TEN evolving since less than 7 days with a progression of the detachment or the eruption observed dating less than 48 hours. Patient and/or have right able to understand the objectives of the trial and having given their written consent to participate (parents for minors, have right for patients in immediate life-saving emergency). Patient registered with a social security scheme or benefiting from a similar scheme. Pregnancy test beta HCG negative for women of childbearing age Exclusion Criteria: Patient weighing less than 20kg Chronic myeloid pathology such as myeloid leukemia or AML (acute myeloid leukemia) Thrombophilia or thrombotic pathology in progress PNN (polymorphonuclear neutrophils) > 50.000 on the CBC (Complete Blood Count) during the inclusion visit Patient who received cyclosporine, anti-TNFalpha or intravenous immunoglobulins or lithium in the month prior the inclusion Pregnant or breastfeeding woman Patient under protective measure (safeguard measure, curatorship, guardianship) or deprived of liberty Patient in exclusion period after participation at other interventional clinical trial Known hypersensitivity to the active substance (FILGRASTIM) or to the one of the excipients (glutamine acid, sorbitol E420, Polysorbate 80) Patient presenting a known glucose intolerance or hereditary fructose intolerance Patient with a traumatic brain injury less than 24 hours Patient admitted with septic shock

Reference center for toxic bullous dermatoses and severe drug eruptions, Edouard Herriot Hospital, Hospices Civils de Lyon