This Supplement Contains the Following Items

This supplement contains the following items:

1. Original protocol, final protocol, summary of changes.

2. Original statistical analysis plan, which had no changes made to it throughout the study.

Vapotherm Clinical Reseach Protocol Document Tracking

Protocol #: RP-EDPF2014001Dev

Study Title: Vapotherm High Velocity Nasal Insufflation compared to Non-invasive Positive Pressure Ventilation in the treatment of Respiratory Failure in the Emergency Department.

Applies to Protocol and corresponsing CRFs

Version / Date / Details / Changes
1.0 / 5-30-2014 / Original. Based on informed consent
1.1 / 6-17-2014 / Revised for waiver of consent at Memorial Hermann TMC
1.2 / 8-20-2014 / Revised for deferred consent at MHH
1.2 / 8-25-2014 / Changes accepted from MHH and saved for EHS

Study Protocol

Protocol Title:Vapotherm High velocity nasal insufflation compared to Non-invasive Positive Pressure Ventilation in the treatment of Respiratory Failure in the Emergency Department.

Protocol Number:RP-EDPF2014001Dev

Revision:Rev 1.0

Principal Investigator:Pratik Doshi, MD

Memorial Hermann, Texas Medical Center

(832)314-8062

Sponsor:Vapotherm, Inc

22 Industrial Drive

Exeter, NH 03833

1(603) 658-0011

Site Monitor:Charles Dunlap, RRT-NPS

(313) 808-0814

Confidential

Study site should keep protocol, all contents and related information confidential.

Protocol Approval

Investigator Statement

As Investigator of the study titled Vapotherm High Flow Therapy compared to Non-invasive Positive Pressure Ventilation in the treatment of Respiratory Failure in the Emergency Department(the “Study”), I agree to:

(i) conduct the Study in accordance with: this Investigator Agreement; the Study’s Protocol as approved by the IRB (the “Protocol”); all applicable laws and regulations; and any IRB or FDA conditions of approval;

(ii) await IRB approval for the Protocol before obtaining informed consents (if applicable);

(iii) ensure that all requirements for informed consent are met and not let any subject participate in the Study before obtaining that subject’s informed consent (if applicable);

(iv) not make modifications to the Protocol as supplied to me by Vapotherm, Inc. (the “Sponsor”), without first obtaining the written approval from the Sponsor and necessary IRB approval;

(v) provide the Sponsor with accurate financial information as required by appropriate regulations;

(vi) maintain Study documentation for the period of time as required by appropriate regulations; and

(vii) supply to the Sponsor, as part of this Investigator Agreement, my curriculum vitae.

INVESTIGATOR

Signature:______

Printed Name: ______

Date:______

Glossary of Definitions and Terms

I. Background and Significance

II. Overall Study Objective

III. Subject Selection

IV. Study Design

V. Patient Enrollment

VI. Study Procedures

VII. Data collection

VIII. Statistical Analysis

IX. Risks and Discomforts

X. Potential Benefits

XI. Monitoring and Quality Assurance

XII. Protocol Deviations

XIII. Adverse Event Reporting

XIV. Confidentiality

XV. References

APPENDIX A: Case Report Forms (CRFs)

Glossary of Definitions and Terms

High Flow Nasal Cannula (HFNC): Nasal cannula system that delivers flow rates of respiratory gas meeting or exceeding a patient’s normal spontaneous inspiratory flow demand. HFNC systems must maintain adequate heating and humidification of the delivered gas to protect the airway tissues from dryness.

High Velocity Nasal Insufflation (HVNI): The use of HFNC with specified parameters including flow rates, humidification and cannula interface dimensions to achieve a defined therapeutic effect.

Non-Invasive Positive Pressure Ventilation (NIPPV): Breathing assist where a mask is strapped tightly to a patients face and bi-level pressure is administered at an established frequency to support a patient’s minute ventilation.

Emergency Department (ED): Unit of the hospital were primary conditions are treated and triaged, the care is considered out-patient and billing is accomplished through the use of CPT codes assigned to Ambulatory Payment Classifications (APC)

Respiratory Failure: The inability to maintain sufficient arterial blood oxygen saturation and CO2 levels during unassisted spontaneous breathing.

Length of Stay (LOS): The total time that a patient is admitted to a specific hospital unit, or the total time spent in the hospital.

Intubation: The placement of an endotracheal tube to facilitate mechanical ventilation.

Mechanical Ventilation (MV): Breathing assist that requires intubation and the delivery of forced positive pressure breaths at a set frequency.

Arterial or Venous Blood Gas Sample (ABG or VBG): Sample of patient’s blood used to assess respiratory status with respect to oxygen and CO2 content, as well as acid-base status.

Pulse Oximetry Reading (SpO2): Indirect measure of a patient’s arterial blood oxygen saturation using pulse oximetry technology.

Glasgow Coma Scale (GCS): A neurological scale representing a patient’s conscious state, ranging from 3 (very unconscious) to 15 (conscious).

Case Report Form (CRF): The form used to record pertinent patient data to address the study aim. CRFs do not contain patient names or medical record numbers; rather they will be coded with a patient number and the site principal investigator at each center will maintain the key. The CRFs are the property of Vapotherm.

Visual Analog Scale (VAS): A Likert scale of facial expressions ranging from a smiley face to a frowning face used to assess the subjects’ subjective level of dyspnea.

I. Background and Significance

High Flow Therapy is a technology coupled with a therapeutic application that impacts respiratory efficiency in such a way as to reduce a patient’s ventilatory work effort 1,2. Vapotherm has developed a delivery system (Vapotherm Precision Flow) that utilizes a patented humidification membrane technology to efficiently condition gas by heating to body temperature, and saturating with water vapor, and delivering through a water-jacket insulated delivery tube to maintain the energy state of the conditioned gas as it is delivered to the patient. Proper conditioning of respiratory gas, as defined above, allows for the administration of high flow rates without damage to the nasalmucosa3-5. The use of these high cannula flow rates of ideally conditioned respiratory gas result in improved ventilatory efficiency, and therefore reduced work of breathing in spontaneously breathing patients in respiratory failure. The mechanisms of action for HVNI are listed below.

1)High velocity nasal insufflation (HVNI) washes out the dead space in the nasopharynx and improves the fraction of alveolar gases with respect to carbon dioxide and oxygen6,7.

2)HVNI provides flow rates that approach or exceed inspiratory flow and therefore reduce inspiratory resistance associated with the nasopharynx, which reduces work of breathing6.

3)Warm and humidified gas improves pulmonary conductance and compliance8.

4)Warm and humidified gas through the nasal pharynx reduces the metabolic work associated with gas conditioning9.

5)Flushing the nasopharynx with high flow can provide mild positive distending pressure for lung recruitment 10-12.

6)The ideal humidification of the inspired gas has been shown to restore mucocilliary function and reduce symptoms of airway exacerbations13,14.

Since its introduction and FDA clearance in 2001, HFNC has seen rapid growth in the ICU environment based on clinical success and high patient tolerance of the therapy. It has consistently been used as a stop gap measure and as an alternative non-invasive ventilation support measure15 to treat patients in respiratory failure and potentially avoid intubation and mechanical ventilation. Moreover, recent unpublished case reports and recent studies have indicated that HVNI, using the Precision Flow device, is safe, very well tolerated and results in similar outcomes to non-invasive positive pressure ventilation (NIPPV)16. Vapotherm HVNI has distinct advantages over NIPPV in that it provides superior humidification and that it provides support without the need to secure a seal on the face, which potentially preserves the nasal and facial tissue 5,17. Patients using HVNI are supported with a nasal cannula instead of a facemask and are able to more easily communicate with the clinicians, take per os medications as well as consume foods and liquids, which is a relative contraindication during NIPPV use. Patients using NIPPV must be fitted carefully and taught to breathe on the therapy. Many are prone to feelings of claustrophobia and increased levels of anxiety, which are abated through delivery of HVNI with a simple nasal cannula interface. Preliminary data from an ED application assessment show that better patient tolerance of the therapy may make it easier and less work intensive for healthcare providers to support patients on HVNI as well as mitigating the risk for pulmonary aspiration seen with full-face NIV interfaces (manuscript in preparation). Thus HVNI can displace the use of non-invasive positive pressure ventilation (NIPPV).

Based on the experiences of staff at several medical centers that use HVNIas a routine respiratory intervention in the ED, its believed that HVNI can achieve the same clinical outcome for patients requiring NIPPV without the adverse effects, patient intolerance and protracted time to stability typical with NIPPV. Moreover, anecdotal observations indicate that intervention with HVNI may result in a reduced time to stability, decline in length of stay,as well as nursing and RT time to support patients. The equivalent clinical benefit achieved through a lower intensity and duration of care may offer both economic and quality of care benefits compared to NIPPV.

II. Overall Study Objective

The overall objective of this study is to determine if Vapotherm HVNI, when used to treat respiratory failure in the ED, is at least equivalent to the current standard of care for non-invasive ventilatory support, NIPPV. Moreover, this study will investigate the potential that HVNI has possible advantages over NIPPV, such as decreased time to physiologic stability from respiratory failure, and the ease of use as a first line intervention for respiratory failure in the ED environment.

The hypothesis is that HVNI via the Vapotherm Precision Flow will demonstrate clinical non-inferiority when compared to NIPPV with regard to treatment failure by way of animpact on ventilation indices and a lower intolerance rate,and have a positive association with hospital disposition and length of stay.

To test this hypothesis, the study will be conducted with the following specific aims:

Aim #1: Determine the efficacy of HVNI compared to NIPPV in treating respiratory failure. The primary endpointwill be treatment failurewithin 72 hrsas determined by intolerance, failure to adequately oxygenate or ventilate and medical deterioration. The data will be sub-analyzed for intubation within the ED and within the first 4hrs of treatment.

Aim #2: Evaluate the capability of HVNI, compared to NIPPV, to affect indices of ventilation. Thesecondary endpoint is the degree of physiologic improvementin blood oxygen and CO2 levels that signify a reduction in both hypoxemia and/or hypercapnia. These indices of ventilation will be blood gas/chemistry analyses taken at baseline (presentation) and the follow up periods 1 hr and 4 hr (if still on therapy); vital signs (heart rate, respiratory rate, perceived dyspnea/comfort level) recorded at baseline, 30min, 1 hr, 90min, and 4 hrs (if still on therapy) and at treatment failure/intubation (if applicable); and clinician perception of efficacy recorded at the end of the treatment period.

Aim #3: Evaluate the impact of HVNI compared to NIPPV on post ED discharge disposition and hospital stays for the first 72 hrs after presentation to the ED. Thetertiary endpoint will be to compare the treatment groups with respect to time spent in the ED, ICU, medical floors and outpatient step-down units. Disposition and length of stay in any unit is at the discretion of the medical team treating the patient.

It is believed that theseendpoints will provide evidence that Vapotherm HVNI is an effective therapy for the non-invasive ventilatory support of respiratory failure patients in the ED setting, which demonstrates at least equivalent clinical efficacy to NIPPV while having patient care and economic value superior to NIPPV.

III. Subject Selection

The subjects in this study are patients that present to the ED in respiratory failure, which is to be based on symptoms and not a presumed diagnosis. Patients will be approached for enrollment based on inclusion criteria, and maintained in the study as test results show that exclusion criteria are not met.

Inclusion Criteria

1. Adult patients (> 18 yrs of age)
2. Presentation with acute respiratory failure
3. Clinical decision to escalate therapy to non-invasive ventilatory support, or to maintain non-invasive ventilatory support if delivered to the ED on such.

Exclusion Criteria

1. Suspected drug overdose
2. Cardiovascular instability as demonstrated by hypotension relative to initial clinical presentation that requires immediate intervention
3. End stage cancer
4. Life expectancy < 6 months
5. Respiratory arrest or significant respiratory depression on presentation
6. Glasgow Coma Scale score < 9
7. Cardiac arrest on initial presentation
8. Need for emergent intubation
9. Known or suspected cerebrovascular accident
10. Known or suspected ST segment elevation myocardial infarction
11. Patients with increased risk of pulmonary aspiration
12. Agitation or uncooperativeness

IV. Study Design

This study will be a prospective, multi-center, randomized, parallel, controlled trial of non-invasive ventilatory support modalities, using a non-inferiority model. Patients in respiratory failure who fit the enrollment criteria will be randomized to one of the two groups. One group will receive HVNI and the other will receive NIPPV to support their work of breathing while clinical management will otherwise remain unchanged based on each center’s standard practices. The assigned therapy will be judged to have failed if the certain pre-specified criteria are met (see below). If the assigned therapy has failed and the patient is not in need of immediate intubation, patients will be permitted to crossover to the alternate therapy group for treatment. If they meet criteria for intubation, or are deemed to need intubation by the critical care team, they will undergo prompt intubation and be managed according to standard practice. Likewise, post emergency department discharge disposition and length of unit stay and therapy use will be determined at the discretion of the treating medical team.

The study design focuses on three time periods. First is the time period when the patient is in the ED, reflecting the patient’s need for emergency care. Second is the 4 hour window which allows adequate time for all front line therapies and medications to take effect, allowing physiologic indices to be meaningfully compared between the two support modalities. Lastly, the 72 hour time point represents the end of an acute event; patients whose symptoms are not abated beyond this time are judged to be suffering from a chronic and potentially progressive condition.

The study will involve two arms. Patients presenting with acute respiratory failure who are escalated to treatment with a non-invasive respiratory support modality will be randomized to one of the following:

• Vapotherm high velocity nasal insufflation (HVNI) administered as defined below (test arm)
• Non-invasive positive pressure ventilation (NIPPV) delivered via single-limb circuit with a full face mask using a pressure support mechanical ventilator system (e.g., Respironics Vision, V60) administered as defined below (control arm)

This study will be conducted in the emergency departments of at least three distinct hospitals. Primary and secondary outcomes will be completed in the first 72 hours; however, the study subjects’ disposition (or mortality) and length of stay data will be followed for 72hours or to discharge, whichever comes sooner. An intention-to-treat analysis model will be used with regard to the non-invasive ventilatory support therapy applied, such that if the non-invasive respiratory support therapy is removed (patient is weaned from the therapy), and non-invasive ventilatory support is again warranted later, the same therapy will be appliedunless the patient had crossed over to the alternate therapy, in which case that will be applied. All respiratory interventions will be tracked during the 72 hr window; disposition and mortality will be tracked for each patient’s entire length of hospital stay.

V. Patient Enrollment

Patients who present to the ED and meet the inclusion criteria, with no obvious signs of meeting any of the exclusion criteria, will be considered for enrollment into the study. Patients who appear to be candidates for enrollment will be approached for informed consent by research staff. Enrolled patients will be randomly assigned to either the HVNI or NIPPV treatment arm, and will otherwise be managed according to current clinical practice. As laboratory or diagnostic results are resolved, the appearance of any exclusion criteria would call for termination of the subject’s participation and subsequent treatment per the medical team’s discretion.

Sample Size:

Total N=204

(102 in each randomization arm: 102 using NIPPV (control arm) and 102using HVNI (test arm).

Subjects will be randomized to either the control or test arm based on a computer generated randomization schedule.

A sample size of 102 subjects in each arm (Total N= 204) was calculated using the following Failure Rates: Control intubation rate of 16.1% and Test rate of 16% . The Control rate was based on the NIPPV results from the Cochrane 2004 review18. From that review the intubation rates of the COPD patients in the top quartile of the studies with the largest number of subjects was used. Since there are no published studies for intubation rates on HVNI patients, a Test rate of 16 % was chosen based on unpublished experience among investigators that lead to the hypothesis that HVNI outcomes would be similar to NIPPV. At the halfway point of subject enrollment, the failure rates for the Control arm and Test arm will be assessed to determine if sample size assumptions were appropriate; thus it may be necessary to adjust the sample size. This will be done by a third party (not sponsor or investigators) so that possible bias will be avoided.

Assumptions used for sample size calculation:

Significance level = 0.05

Proportion of Control Arm = 16.1%

Proportion of Test Arm = 16%

Power = 90%

When the sample size in each group is 102, a two-group large-sample normal approximation test of proportions with a0.050 significance level will have 90% power to reject the null hypothesis that the test and the standard arenot equivalent (the difference in proportions is 0.15 or farther from zero in the same direction) in favor of thealternative hypothesis that the proportions in the two groups are equivalent, assuming that the expected difference inproportions is 0.001 (.161-.160) and the proportion in the standard group is 0.161