How the diagnosis of COVID-19 will be made – this will be either by PCR or radiologically. Certain sections have been clarified regarding fibrinogen monitoring and the action to be taken. In particular, it was unclear whether replacement therapy should be given when fibrinogen levels are <1.5 g/L or <1.0 g/L. This has been clarified. A drop in fibrinogen <1.5 g/L is not a stopping criteria for the trial and accordingly this has never been listed in the Stopping Criteria section (7.10) as the intention has always been that this was a reason for treatment discontinuation in an individual patient. The wording in the protocol has been clarified around this. Updated exclusion criteria no 7 as follows: ‘Patients considered inappropriate for critical care (prior decision re ceiling of care established e.g. being considered for palliative care)’. The new example given is more in line with the exclusion criteria. Information has been added around the rationale for not excluding patients who are on anticoagulants and to clarify and provide rationale as to why we will not exclude patients on anticoagulation at higher than standard doses. We have removed the following from the ISTH definition of major bleeding in the stopping criteria (so it will no longer be a stopping criteria for the trial): 'Bleeding causing a fall in haemoglobin level of 20 g/L or more, or leading to transfusion of two or more units of whole blood or red cells.' This is because the patients on ICU are regularly receiving blood transfusions, and even minor bleeding may cause a drop in Hb because of other complications. We have updated the stopping criteria around bleeding to add ‘that occur between start of treatment and 30 h after the last dose’ as events past that timepoint would not be considered related to the trial treatment and so would not be considered a reason to stop the trial.

Updated dosing schedule to allow dosing for a maximum duration of 14 days. The duration of therapy can be extended from five days to a maximum of 14 treatment days in total. This decision can only be made in the absence of any dose-limiting toxicity. Treatment discontinuation: Treatment may be stopped at any point in a patient if any of the following rules are met: - If plasma fibrinogen levels fall to <1.5 g/L, the further dosing of rtPA in the patient will be stopped. - Patient maintains saturation on room air that is either normal or normal for them for 48 hours as assessed by the investigator. - A trial stopping rule has been met. Re-treatment with rtPA after treatment discontinuation due to improvement in patient saturation: If there is a recurrence of COVID-19 ARDS related symptoms or a worsening of PaO2/FiO2 ratio, which the investigator considers could be related to treatment discontinuation, treatment can be restarted. The decision to re-treat will be made within 5 days (120 hrs) from the last dose of treatment. Treatment may not be restarted more than once. Treatment cannot be given for more than 56 doses. The rationale for restarting of treatment must be documented in the notes, along with the absence of toxicity. Re-treatment of patients previously consented and treated on the trial: For patients that have already been treated under the previous 3 day dosing schedule, if they remain eligible for the trial, they will be offered the option to continue treatment as above – the maximum number of days of extra treatment for these patients will be 11 days. Patients (or their legal representatives) must give consent to this re-treatment. The total duration of trial participation for these patients may exceed 28 days. Also updates made to the endpoints in relation to this extended treatment duration.

Use of historical matched controls: The standard of care arm will be replaced with a historical matched control arm. Historical controls will instead be recruited at a ratio of two controls to every one treatment arm patient. The study team agreed that if controls are matched according to ventilation type, disease severity, age and gender, this should be sufficient to identify controls and also allow for some degree of underlying matching of comorbidities. The identification of historical controls will begin immediately as it is permitted under the notices relating to the Health Service (Control of Patient Information) Regulations 2002. Under this notice, prior consent is not required in order to carry out this process. However, in order to retain the data in line with clinical trial regulations, consent will be sought from patients. If consent is not obtained by the time the notice expires or patients explicitly ask to not be included in the trial as historical controls, their data will be removed. Use of deceased patients as historical use of controls: We would also like to potentially include deceased patients as historical controls if they are a suitable match. The data monitoring committee highlighted the importance of allowing this because if they are not included, it may bias the control arm towards having a more favourable outcome than what standard of care actually provides. Given that GDPR and the data protection act do not apply to deceased people, the study team do not feel that consent is required in order to include deceased patients as historical controls. In addition, it may not be appropriate to contact next of kin for assent given that the patient will have died within the last few months. We would appreciate the RECs views on this and whether they feel it is acceptable to include these patients in the trial without the need to approach their next of kin for assent.

We have updated the consent section for historical control patients (section 6.4) following a discussion with CAG. Consent will not be sought from any historical controls patients because the use of their personal identifiable information is covered by the COPI notice. We therefore no longer plan to use the patient information sheet and consent form for historical controls (version 1.0 dated 18 Sep 20) as previously approved by REC. We have also made updates to the data to be collected for historical control patients to allow for a more complete comparison between treatment patients and control patients and to clarify that the data will be collected where available for 28 days from the date of admission to hospital, rather than ‘at a frequency that where possible matches the treatment arm patient that they have been matched to’ as it is not possible to collect for a total of 31 days due to the set-up of the database. We have also updated the criteria to be used for matching of control patients to allow for better matching and made some further clarifications. Other corrections/clarifications: Correction of WHO 7 point score criteria. To clarify that lung compliance will be calculated using tidal volume / (peak inspiratory pressure – PEEP) rather than 'mL/cm H₂O VT/( PIP – PEEP)'. Driving pressure removed from list of ventilation details to be collected for IMV patients as it is not required. Added the SOFA score to also be collected for control patients The Glasgow coma scale will be recorded based on the output data collected for the historical controls as this is required to calculate the SOFA. Clarification added around the medical history and COVID-19 admission history that is of particular relevance. Also clarified that PaO2/FiO2 will be a calculated field on the trial database derived from the fields for PaO2 and FiO2.

The major change is the addition of cohort 2, in which we plan to enrol patients for treatment with nebulised recombinant tissue plasminogen activator to firmly establish the safety profile. Preliminary review of the data has not demonstrated any significant drop in fibrinogen or systemic absorption of administration of 40 mg tPA over 24 hrs by inhalation for 3 to 14 days. As most clinicians are used to intravenous preparation, accrual of additional data was felt to be vital in preparation for a randomised control study. Cohort 2 will include up to 30 patients, with a minimum of 10 ventilated and 10 non ventilated patients receiving the intervention. The remainder 10 can be in either group. The dosing has been changed for pragmatic considerations. Ventilated patients will receive 20 mg three times a day instead of 10 mg four times a day. This is to account for the dead space created by the ventilator circuit and we consider will provide a more comparable dose delivery to the non-ventilated patients. Non-ventilated patients will receive 20 mg three times a day for 48 hrs to help build up the dose of tPA, followed by 20 twice a day. Patients can receive the drug for a maximum of 14 days with termination of treatment when patients achieve normal saturation on room air for 24 hrs. The collection of data for cohort 2 is similar to cohort 1, except the frequency of data points will be decreased to once a day. Multiple time points will be retained for the ratio of inspired oxygen and blood oxygen as it is a primary endpoint. Cohort 1 and cohort 2 will also be analysed separately. Co-enrolment into other clinical trials will be allowed, and documented. We have reduced the number of exclusion criteria for patients in Cohort 2 compared to the first cohort. We will also stop treatment in an individual patient if fibrinogen levels fall below 1.0gm/L (previously 1.5g/L) as we will allow patients with a lower baseline fibrinogen level to enter the trial.

PI change at Barnet Hospital