REPORT OF INTERNAL INVESTIGATION INTO THE DEATH OFA

VOLUNTEER RESEARCH SUBJECT

Internal Investigative Committee Membership

Chair: Lewis C. Becker, M.D., Robert L. Levy Professor of Cardiology, Chair JCCI-1

Members: Roy G. Brower, M.D., Associate Professor, Pulmonary Medicine

Ruth Faden, Ph.D., M.P.H., Professor Health Policy and Management, Executive Director Johns Hopkins Bioethics Institute

Roger A. Johns, M.D., Professor and Chair, Anesthesiology and Critical Care Medicine

Pamela A. Lipsett, M.D., Associate Professor, Surgery and Anesthesiology and Clinical Care Medicine, Co-Director, Surgical ICU

Brent G. Petty, M.D., Associate Professor, Clinical Pharmacology

Peter J. Pronovost, M.D., Ph.D., Associate Professor, Anesthesiology and Critical Care Medicine, and Surgery

Consultants: Paul S. Lietman, M.D., Ph.D., Professor of Medicine, Pharmacology &

Molecular Science, and Pediatrics

E. Regis McFadden, M.D., Professor of Medicine, Division of Pulmonary and Critical Care Medicine, Case Western Reserve University

Gary Hunninghake, M.D., Chief, Pulmonary and Critical Care Medicine, University of Iowa

        . Introduction

Ellen Roche, a 24-year-old technician in the Johns Hopkins Asthma and Allergy Center, was recruited as a normal volunteer in an NIH-funded study of asthma directed by Dr. Alkis Togias, Associate Professor of Clinical Immunology. The purpose of the study was to understand how bronchodilation is maintained in normal people by deep inspirations in the face of a bronchoconstrictive stimulus, specifically whether bronchial wall nerves and nitric oxide release play important roles in this process. Ms. Roche consented to participate in the study on 4/16/01 and underwent a series of methacholine challenges over a period of 2 weeks, in accord with the IRB-approved protocol. On 5/4/01, as part of the protocol, she received approximately 1 gm of hexamethonium by inhalation, given to produce ganglionic neural blockade. The next day she developed a dry cough and dyspnea on exertion (reported 5/7/01). On 5/7/01, she had flu-like symptoms and her FEV1 was reduced. On 5/9/01, she was febrile, a chest X-ray revealed streaky densities in the right perihilar region, and her arterial oxygen saturation fell to 84% after walking a short distance. She was admitted to The Johns Hopkins Bayview Medical Center (JHBMC) for observation, and developed progressive dyspnea and was transferred to the ICU on 5/12/01. Subsequently she developed pulmonary infiltrates and worsening hypoxia, was intubated and ventilated, suffered bilateral pneumothoraces, and presented a clinical picture of adult respiratory distress syndrome (ARDS). She died on 6/2/01 with progressive hypotension and multiorgan failure.

At the request of the Dean of the School of Medicine and the President of the University, an internal committee was convened to investigate the circumstances surrounding this tragic death and to render its best judgement as to cause. This document is a report of the committee’s investigation.

. Issues Addressed by the Committee

. Was the study designed to address an important scientific question?

a. Relation of hexamethonium study to the NIH R01 grant application HL61277

The study in which Ellen Roche participated was part of National Institutes of Health funded R01 grant HL61277 entitled "Lung Inflation in Airways Hyper-Responsiveness." The principal investigator of the grant is Dr. Solbert Permutt, Professor of Medicine. Dr. Alkis Togias is the co-investigator on the grant, and was in charge of the specific study in which she took part. This particular study involving hexamethonium was not part of the original R01 application, but was mentioned as a planned study in the non-competitive continuation applications submitted 11/15/99 and 9/27/00. In the renewal submitted 9/27/00, the description of the study was as follows: "We are awaiting IRB approval for a protocol in which a ganglionic blocker, hexamethonium, will be used by inhalation in healthy subjects with the hypothesis that it will suppress the bronchial protective effects of deep inspiration. If the hypothesis is correct, neural mechanisms will be implicated in this important lung function." According to Dr. Patricia Noel, the Program Director for the R01, the hexamethonium study was felt to be consistent with the original goals of the R01, but was not otherwise reviewed scientifically. The research protocol for this study was described in detail in RPN No. AAC00-07-26-02, entitled "Mechanisms of Deep Inspiration Induced Airway Relaxation," approved by the JHBMC IRB on 9/18/2000.

b. Scientific background and support/justification for the hexamethonium study:

The original R01 was based on the following premise: Dr. Permutt=s laboratory had published studies providing evidence that patients with asthma lack a function by which lung inflation protects the airways from smooth muscle induced obstruction. Dr. Permutt proposed that this defect is fundamental in asthma and may constitute a mechanism of airway hyper-responsiveness. The goals of the project were to better understand the role of lung inflation in asthma pathophysiology and to identify mechanisms which account for the lung inflation effect being dysfunctional. This work is important because asthma is a common and important disease with significant morbidity and mortality, particularly in urban populations. In the progress report for the second year of the proposal, the results of a study are described in which inspirations during methacholine inhalation challenge resulted in a bronchoprotective effect that was completely suppressed by allergen challenge. These findings supported the principal investigator's hypothesis that an element of the allergic reaction renders the bronchoprotective effect of lung inflation non-functional. This initial work led to the hexamethonium study.

The IRB-approved protocol describes the specific background and justification for performing the hexamethonium studies. The investigators had described beneficial effects of deep inspirations on human airways and two separate phenomena that are involved. One of these is bronchoprotection, the phenomenon where, following a period of quiet breathing, deep inspirations are able to protect the airways from spasmogenic stimuli. The other phenomenon is bronchodilation, where once airway constriction has occurred, deep inspirations are able to partially reverse it. In healthy humans, the investigators demonstrated that bronchoprotection is stronger than bronchodilation, whereas in asthmatics bronchoprotection is either absent or minimized, and bronchodilation is only minimally decreased. This suggests that the mechanisms of the two phenomena are different.

The investigators proposed that the bronchoprotective response may be mediated by non-adrenergic, non-cholinergic nerves (NANC) and the hexamethonium study was designed to address this issue. The role of non-adrenergic, non-cholinergic neurons in bronchial activity had been studied extensively both in animal and human studies. Prior to the early 1990's, the agent mediating NANC neurotransmission was unknown, but it is now recognized that the mediator released from the NANC nerve terminals is nitric oxide. The investigators decided to study the role of NANC nerves in bronchoprotection by blocking NANC neurotransmission through the use of a ganglionic blocker, specifically hexamethonium. Dr. Togias quoted five studies (Sterk, et al, 1995; O'Byrne, et al, 1985; Taylor, et al, 1985, Holtzman, et al, 1980; Ichinose, et al, 1988) which employed hexamethonium by inhalation in humans. The results of these studies were inconsistent in terms of the role of NANC neurons in bronchial reactivity. Further, none of these studies investigated the role of NANC neurons in the specific mechanisms of deep inspiration-induced bronchoprotection. The novelty of the hexamethonium study was further enhanced by serial measurements of nitric oxide in the exhaled breath. This provided the possibility of very strong evidence for the involvement of NANC nerves in bronchoprotection. Thus, the scientific basis for the proposed studies is well justified based on the previously published human studies and based upon the preliminary work of the investigators.

Could an alternative approach to inhaled hexamethonium have been used to address the scientific question in this study?

The investigators chose to use hexamethonium to block NANC neurotransmission in the lung based on the studies cited above from the 1980s. An additional and more specific approach to blocking NANC neurotransmission had also been utilized in humans prior to the submission of the human investigation protocol. L-NAME (L-NG-nitro-L-arginine methyl-ester) had been used extensively in the 1990's to block NANC neurotransmission in multiple organs including the lung in several animal studies. In 1992, Ellis and Undem (American Review of Respiratory Disease, 146:1543-7, 1992) demonstrated that another inhibitor of nitric oxide synthase, L-NG-nitro-L-arginine, inhibited non-adrenergic, non-cholinergic mediated relaxations of human isolated central and peripheral airway. In 1997 and 1998, two studies were published utilizing nebulized inhaled L-NAME to investigate the role of nitric oxide and bronchial reactivity in normal and asthmatic humans. The use of L-NAME alone, however, would not have allowed the investigators to differentiate nitric oxide derived from NANC neurons from nitric oxide derived from other sources, including bronchial epithelium and lung macrophages. Thus, the use of hexamethonium on a scientific basis to investigate the specific question of the role of NANC neurons in combination with the measurement of exhaled nitric oxide was the best approach and is, therefore, well justified for use in the protocol.

The question remains as to whether another ganglionic blocker or intravenous administration of hexamethonium could have been used. Other ganglionic blockers were available, such as trimethaphen or mecylamine; but these drugs have also been implicated in lung injury in studies from the 1950=s. Also, hexamethonium is the drug that was used in all of the previous related literature. While ganglionic blockade could have been achieved with intravenous use of hexamethonium, all prior related literature utilized hexamethonium by inhalation. Also, the PI stated to the committee that he felt administration by the inhaled route would block intrapulmonary ganglia more effectively and with less systemic effect of the drug.

In summary, the study design addressed an important scientific question; namely, the role of NANC neurons in mechanisms of bronchoprotection. A better understanding of this process was important in understanding the mechanisms involved in human bronchial asthma. The use of hexamethonium also appears well-justified scientifically based on previous animal and human studies in which inhaled hexamethonium was used to address the role of NANC neurotransmission. In none of these prior studies was a lung toxicity component of inhaled hexamethonium reported, although these were not rigorous Phase I studies. Direct inhibitors of nitric oxide synthase had been administered to humans, but they could not have answered the direct question of a role of NANC neurons due to their non-specificity. Neuronally selective nitric oxide synthase inhibitors, while available, had not been reported in human lung investigations. Thus, the rationale for utilization of a ganglionic-blocking drug, such as hexamethonium, was scientifically sound. It was also critical to study this bronchial protective response in normal humans to determine the mechanism of the response. Studies in normal subjects were required to determine whether alterations in this reflex pathway could play a role in human bronchial asthma.

References:

1. Ellis JL, Undem BJ. Inhibition by L-NG-nitro-L-arginine of nonadrenergic-noncholinergic-mediated relaxations of human isolated central and peripheral airway. Am Rev Respir Dis 1992; DEC; 146(6):1543-7.

2. Taylor DA, McGrath JL, O'Connor, BJ, Barnes PJ. Allergen-induced Early and Late Asthmatic Responses Are Not Affected by Inhibition of Endogenous Nitric Oxide. Am J Respir Crit Care Med 1998; 158:99-106.

3. Sterk PJ, Daniel EE, Zamel N, Hargreave FE. Limited Maximal Airway Narrowing in Nonasthmatic Subjects: Role of Neural Control and Prostaglandin Release. Am Rev Respir Dis 1985; 132:865-760.

4. O'Byrne PM, Thomson NC, Latimer KM, Roberts RS, Morris MM, Daniel ED, Hargreave, FE. The effect of inhaled hexamethonium bromide and atropine sulphase on airway responsiveness to histamine. J Allergy Clin Immunol 1985; 76(1):97-103.

5. Holtzman, MJ, Sheller JR, Dimeo M, Nadel JA, Boushey HA. Effect of Ganglionic Blockade on Bronchial Reactivity in Atopic Subjects. Am Rev Respir Dis 1980; 122(1):17-25.

6. Ichinose M, Inque H, Miura M, Takishima T. Nonadrenergic Bronchodilation in Normal Subjects. Am Rev Respir Dis 1988; 138:31-34.

7. Mackay TW, Fitzpatrick MF, Douglas NJ. Non-adrenergic, non-cholinergic nervous system and overnight airway calibre in asthmatic and normal subjects. Lancet 1991; Nov. 23, 338 (8778):1289-92.

8. Altiere RJ, Szarek JL, Diamond L. Neural control of relaxation in cat airways smooth muscle. J Appl Physiol 1984; Nov; 57(5):1536-44.

9. Maniscalco M, Sofia M, Smith A, Demoncheaux EA, Mormile M, Faraone S, Higenbottam T. Lack of effect of nitric oxide inhibition on bronchial tone and methacholine-induced bronchoconstriction in man. Respir Med 1997 Jul; 91(6):335-40.

2. Was the scientific and safety review of the study protocol by the JHBMC IRB appropriate?

Dr. Togias submitted the RPN application entitled AMechanisms of Deep Inspiration B Induced Airway Relaxation@ on 7/25/00 with co-investigators Alvin Sanico, M.D., and Robert Brown, M.D. to the JHBMC IRB.

The original application form submitted to the IRB indicated that marketed drugs would be used in the study, but no investigational new drugs would be administered. (It should be noted, however, that the form does not ask for whether non-FDA approved drugs without IND numbers would be used.)There was no indication that hexamethonium was no longer FDA approved, or that marketing approval had been withdrawn in 1972. The application described hypotension as a risk of hexamethonium. The previously reported lung toxicity was unknown to the principal investigator before Ms. Roche became ill, and this information was not included in the submission to the IRB. The dose of hexamethonium to be inhaled was based on the effective, and apparently safe, use of similar doses in 4 previously published studies from the 1980=s (cited and supplied with the application; see Section II1.). One additional study used inhaled hexamethonium in a smaller dose. All of these studies had been approved by local IRBs. However, the application did not explicitly point out that the information on safety and efficacy was derived from only 20 research subjects. Phase I safety data were not available for inhaled hexamethonium. The 4 published papers did not mention pulmonary toxicity, but at the same time did not specifically say that there was none. Even in the absence of any pulmonary adverse events in those 20 subjects, there was uncertainty regarding the true risk of inhaled hexamethonium. Small clinical trials give uncertain estimates for even frequent adverse events, and may miss even relatively common toxicity.

The principal investigator subsequently stated to the investigation committee that he had performed a standard PubMed search for potential hexamethonium toxicity and consulted standard, current edition textbooks of pharmacology and pulmonary medicine before submitting the application to the IRB. None of these sources mentioned hexamethonium-related pulmonary toxicity. Although the PI made a good-faith effort to uncover previous reports of hexamethonium-related toxicity, the vagaries of performing such a search are illustrated vividly by the following performed as part of this investigation: (1) None of the last four editions of Goodman and Gilman, or the current edition of Fishman’s Pulmonary Medicine mentioned this toxicity, but an older edition of Fishman’s text did; (2) During PubMed searches, "hexamethonium inhalation lung injury" gave 0 hits, "hexamethonium inhalation" gave 42 hits (but none referring to pulmonary toxicity), "hexamethonium lung" yielded 3 useful articles, "hexamethonium lung toxicity" gave 4 hits, but 0 useful articles, "hexamethonium lung hypersensitivity" gave 16 hits with 3 useful articles, and "hexamethonium lung fibrosis" gave 3 hits and 2 useful articles; (3) the Google search engine and Yahoo uncovered the pneumotox web site (see below), but LookSmart and GoTo.com did not; (4) the Micromedex data base had lung toxicity as the first adverse effect of hexamethonium.

The pulmonary toxicity associated with oral, intramuscular, and/or subcutaneous hexamethonium administration for hypertension was first reported in 1953 (1) and between then and 1960 several articles including individual case reports and small series of autopsied cases were published (1-12). In 1970 a review article (13) listed six references from the 1950’s.

The pulmonary pathology was described in various ways, such as "identical pathologically with ‘acute interstitial fibrosis of the lungs’ but differing from Hamman-Rich in that ‘the disease was rapidly progressive to death within a week without development of cor pulmonale’ (2) or ‘a mixed intra-alveolar and interstitial fibrosis associated with preservation of the normal alveolar elastic pattern’(3) or ‘fibrinous pneumonitis’ (7). In all the reports reviewed the hexamethonium had been administered for many weeks and in most cases many months before the pulmonary manifestations appeared. In no case was the duration of treatment less than a month and in no case was the hexamethonium delivered by inhalation. Many, but not all of these cases, had uremia, as a comorbid condition, which can produce similar diffuse alveolar damage by itself. According to Dr. Fred Askin, the lung pathology described in these reports is non-specific in type.

Thus, while evidence of an association between hexamethonium and severe pulmonary disease does exist, we have discovered no report of this lung pathology arising after inhalation of hexamethonium and it is not at all clear that such pulmonary pathology could be produced by one or a small number of doses of hexamethonium bromide delivered by inhalation..

1. Morrison B. Parenteral hexamethonium in hypertension. Brit M J 1953; 1:1291-1299.

2. Morrow JD, Schroeder JA, Perry HM. Studies on the control of hypertension by Hyphex II. Toxic reactions and side effects. Circulation 1953; 8:829-839.

3. Corcoran AC, Dustan HP, Taylor RD, Page IH. Management of hypertensive disease. Am J Med 1954; 17:383-394.

4. Doniach I, Morrison B, Steiner RE. Brit Heart J 1954; 16:101-108.

5. Robillard R, Riopelle JL, Adamkiewicz L, Tremblay G, Genest J. Pulmonary complications during treatment with hexamethonium. Canad M A J 1955; 72:448-451.

6. Goldsmith HJ, Beaven DW, Kambert HP. Severe hypertension treated with hexamethonium bromide in retard medium. Lancet 1955; 1:371-374.

7. Perry HM, O’Neal RM, Thomas WA. Pulmonary disease following chronic chemical ganglionic blockade. Am J Med 1957; 22:37-50.

8. Hildeen T, Krogsgaard AR, Vimtrup Bj. Fatal pulmonary changes during the medical treatment of malignant hypertension. Lancet 1958; 2:830-832.

9. Hilden T, Krogsgaard AR, Vimtrup Bj. Svaere lungeforandringer under medikamentel behandling af malign hypertension. Nord Med 1958; 6:353-356.

10. Heard BE. Fibrous healing of old iatrogenic pulmonary oedema ("hexamethonium lung"). J Path Bact 1962; 83:159-164.

11. Petersen AG, Dodge M, Helwig FC. Pulmonary change associated with hexamethonium therapy. Arch Int Med 1959; 103:285-288.

12. Rokseth R, Storstein O. Pulmonary complications during mecamylamine therapy. Acta Med Scand 1960; 167:23-27.

13. Rosenow EC. The spectrum of drug-induced pulmonary disease. Ann Int Med 1972; 77:977-991.

Dr. David Lepay (Director, Division of Scientific Investigations at the FDA) spoke to a consultant for the investigation committee on 7/3/2001. Dr. Lepay provided clarification on why hexamethonium was withdrawn from the market. He indicated that there had been two companies that had formulations on the market. The FDA had been charged to review all drugs approved and grandfathered to determine if they should stay on the market (the DESI review). The National Academy of Sciences assisted the FDA in this review by setting up review panels. The panels classified the drugs as effective or not effective for the marketed indications. The panels believed the drug was ineffective in the treatment of hypertension. FDA reviewed the panel’s recommendation, which was to remove the two hexamethonium products from the market. One company immediately withdrew its products on 3/30/72. The other company discussed the recommendation with FDA and was unsuccessful in its discussions. FDA removed the product from the market on 8/5/72. The results of the DESI review were published in the Federal Register on 5/21/71. Dr. Lepay made it clear that neither product was removed from the market due to safety concerns. They were removed because of lack of effectiveness and not because they were excessively toxic.

After initial review of Dr. Togias’ RPN by the IRB members, Dr. Gary Briefel, the IRB Chair, wrote to the principal investigator with questions raised in the review. One question was: AIf the hexamethonium is not an FDA approved product, the protocol should describe the source of the hexamethonium and how it will be made safe for human use.@ Dr. Briefel confirmed to the investigation committee that he and other members of the IRB were aware that hexamethonium was not a marketed drug. It should also be noted that considerable expertise existed on the IRB to review this particular RPN. Members included Dr. Judith Stiff, Associate Professor of Anesthesiology and Critical Care Medicine, and Vice-Chair of the IRB; Dr. Darrell Abernethy, Clinical Director, Gerontology Research Center of the National Institute on Aging; Dr. N. Franklin Adkinson, Professor of Medicine, Division of Clinical Immunology; Dr. Philip Norman, Professor of Medicine, Division of Clinical Immunology; Dr. W. Lowell Maughan, Professor of Medicine and Chief of the Cardiology Division at JHBMC; and John Guarente, Pharm. D., a clinical pharmacist at JHBMC.

In his response to the IRB, the principal investigator did not specify the FDA status of hexamethonium, but he indicated that the drug would be provided from the pharmaceutical/chemical company, Fluka. He provided a certificate of analysis verifying that the substance was 99.6% pure by argentometry. The contaminants were said to be inorganic salts. The PI indicated that the hexamethonium would be suspended in isotonic saline under a chemical hood and the solution passed through an 0.2 micron sterile filter. A limulus test was to be done to exclude endotoxin contamination. The PI did not describe the labeling on Fluka’s hexamethonium bromide product ("For laboratory use only. Not for drug, household, or other use."), but he did not have the product in hand at the time of his response to the IRB. IRB was apparently satisfied with these responses, as well as the consent form. Approval for the study was granted at the convened meeting on 9/18/00 (application was assigned number AAC00-07-26-02). The IRB did not ask the investigator to submit an IND application for inhaled hexamethonium, or ask him to obtain an opinion from the FDA about whether it was exempt from the requirement to obtain an IND number. However, the decision not to pursue an IND occurred on a long background of uncertainty regarding FDA guidelines for non-FDA approved drugs in physiologic (non-therapeutic) studies of lung function. Dr. Briefel and Dr. Thomas Hendrix, Chair of the JCCI, had written jointly to Dr. Robert Temple, Associate Director for Medical Policy at the FDA on 10/24/97 for guidance on this issue. The letter stated "on occasion, protocols are submitted where biological agents, such as histamine or bradykinin are to be administered to subjects. The administration is usually topically on the nasal mucosa, or for inhalation as a mist into the lungs. The agents are given as part of studies examining how these biological agents might mediate disease processes or as bronchoprovocation tests to measure what effects the biological agents have on other mediators of disease, such as inflammation. . . . In none of the research projects is the biological agent being tested for safety or efficacy, nor are they being proposed for development of drugs for the treatment of disease . . . The biologic agents are not prepared by pharmaceutical companies; rather, they are obtained from chemical supply companies." The letter goes on to ask, "the question has been raised in the IRBs reviews as to whether or not such use of biologic agents in this class of protocols requires submission of an IND application to FDA before the agents are administered. In the past, we have had investigators speak to FDA staff to discuss this issue. They have received conflicting opinions and responses which the FDA staff have not put in writing. . . . We request, therefore, written guidance on how to determine whether an IND application is required for this kind of research protocol. If written guidance is not possible, let us know to who questions of this sort may be directed on a case-by-case basis, so that investigators will be able to provide the IRBs with a written opinion from FDA prior to submitting the protocol for review."

After the FDA failed to respond, the letter was resent by Fax on 1/4/99 to Dr. Paul Goebel at the FDA. A letter was received on 2/9/99 from Theresa M. Martin, Executive Secretariat Staff, Center for Drug Evaluation and Research at the FDA, stating "your question is a very complicated one. Currently CDER’s Regulatory Policy Staff is working on a response to your original letter of October 24, 1997." In the continued absence of a response, Dr. Briefel sent another letter on 4/19/01 to Dr. David Lepay at the FDA, seeking an opinion about whether an IND was needed for inhaled bradykinin in a study of nasal glucocorticoids in asthma. A reply has not yet been received.

In the absence of guidance from the FDA, the IRB made a decision regarding the need for an IND. In this case, based on the 4 published human studies employing inhaled hexamethonium, the IRB decided that an IND was not needed.

As part of the post hoc investigation into the event, a call was placed on 6/29/01 to the FDA to discuss the IND issue. Ms. Sandy Barnes, Chief, Project Management Staff, Division of Pulmonary and Allergy Drug Products for the FDA, and the principal contact person for pulmonary products, stated that IND opinions are provided differently by different FDA Divisions. For the Pulmonary Division, the principal investigator requesting an opinion usually makes a preliminary phone contact. He/she is then asked to Fax information describing the study, the name of the drug, dosage and route of administration, and subject population to be studied. If it is felt that the drug is exempt, a letter is sent back to the investigator by the FDA. If the drug is not exempt from the requirement for an IND number, a full application to the FDA is required.

Although not recognized at the time of the submission to the IRB, the pulmonary toxicity associated with prolonged hexamethonium administration was uncovered by Dr. Togias through the internet (http://www.pneumotox.com/pneumotox/hexomethonium) on 5/9/01, the day Ms. Roche was admitted to the JHBMC with shortness of breath, cough, flu-like symptoms, a decline in pulmonary function, and a lung infiltrate. The IRB received a letter on 5/9/01 from the PI reporting the serious adverse event in Ms. Roche. This letter also stated that "in reviewing the pneumonitis literature, I have found that hexamethonium has been implicated in some cases of Bronchiolitis Obliterans Organizing Pneumonia (BOOP)." The letter also first disclosed that the first volunteer in the study had developed a cough for 8 days after hexamethonium exposure. The study was immediately put on hold by the IRB.

One way to obtain evidence of safety would be to submit an IND application to the FDA and secure approval of it. It is understandable why the IRB did not ask Dr. Togias to submit an IND application for inhaled hexamethonium, given the repeated inability to obtain a written response from the FDA on this matter and the uncertainty of interpretation of the criteria for requiring submission of an IND application. Because of the relative paucity of published safety data of inhaled hexamethonium (only a total of 20 subjects studied), the committee believes that the IRB should have asked Dr. Togias to obtain an opinion from the FDA about whether the proposed use of inhaled hexamethonium required an IND number or was exempt. The committee also believes that the IRB should have requested more details from Dr. Togias about exactly how the hexamethonium was going to be prepared, stored, and tested for purity prior to administration to research subjects, although it is doubtful that more specification of these procedural details would have prevented the fatal outcome.

 

3. Was the consent form appropriate?

The IRB-approved consent form outlined the research question and the fact the subject was to be studied as a volunteer with healthy lungs. The form stated that "hexamethonium is a medication that has been used during surgery, as a part of anesthesia; this is capable of stopping some nerves in your airways from functioning for a short period." Regarding the risks of hexamethonium, the form stated "this medication, when you breathe it, may reduce your blood pressure and may make you feel dizzy especially when you stand up. This effect may last up to 3 hours. During the visit you receive hexamethonium, you will be connected to a heart monitor and we will measure your blood pressure very often. You will also have an IV (a small tube in your vein) placed only as a precaution. If your blood pressure becomes too low, we may ask you to lay flat for as long as the problem persists or you may have to receive fluids through your IV." The consent form stated that "there is no benefit for you from doing this." It added "the information that we will get from this study may help us understand some of the problems behind asthma. For your time and effort, you will be compensated with up to $365, that is $25 for each of the first phase (30-40 minute) visits you complete and $60 after each of the second phase (4 hour) visits you complete."

The consent form did not say that hexamethonium was no longer used clinically and was not an FDA-approved drug. It did not mention the reported relationship between long-term hexamethonium use and pulmonary pathology, or discuss whether this information was relevant to the current study of inhaled hexamethonium. It did not disclose that the evidence of safety of inhaled hexamethonium was based on published reports involving a total of only 20 individuals that were not designed to evaluate safety. Finally, the consent form did not state the possibility of adverse events not previously recognized, and that some events could be serious, or that death could result from study participation.

The committee believes that the consent form accurately outlined the purpose of the study, the procedures that would be undertaken and the alternatives to participation. In the consent form, the statement that "hexamethonium is a medication that has been used during surgery" is misleading because FDA approval had been withdrawn in 1972. Hexamethonium given by inhalation had never been used clinically. Furthermore, the risks section of the consent form was inadequate and should have included information about lack of FDA approval, a statement that the safety of the drug was based on only 20 subjects reported in the literature, and a statement that a serious adverse event or death were possible risks of the study.

. Was any coercion involved in the recruitment of the research subject?

Ellen Roche was a research technician who had worked in the Johns Hopkins Asthma and Allergy Center since 1/4/99. She was employed in the basic science laboratory of Dr. John Schroeder, Assistant Professor of Medicine. Dr. Schroeder and Dr. Togias were in the same academic division in the Medical School, but did not work or publish together. Dr. Togias did not have any supervisory responsibility over Ms. Roche, did not evaluate her job performance, and did not participate in decisions regarding merit raises or promotional opportunities. She had participated in several research studies in the A&AC. In August 1999, she enrolled, and became an active participant, in Dr. Mark Liu’s protocol AAC 93-01-27-01 entitled "Screening Evaluation and Bronchoprovocation Studies of Asthma and COPD." One of the functions of Dr. Liu’s protocol was to provide a registry of individuals who were interested in participating in future studies conducted at the A&AC.

For Dr. Togias’ hexamethonium study Dr. George Pyrgos, a postdoctoral Research Fellow, reviewed the registry database of normal volunteers to determine if there were potential subjects who could be enrolled. He reviewed the data on who appeared to meet inclusion criteria and were designated as "active" (about 20-30 were identified, most of whom worked at the A&AC). Dr. Pyrgos reviewed the potential individuals who might be eligible with Dr. Togias. After checking with Dr. Togias, Dr. Pyrgos called potential participants. The study was explained over the phone; and for those individuals still interested, Dr. Pyrgos brought the consent form to them for review. People who wanted to join the study were scheduled for a screening visit, and brought the consent form with them for final review and signing by Dr. Togias. Nine people signed the consent form and entered the screening phase of the protocol (8 worked at the A&AC). Only three of the nine who enrolled reached the point in the study where they were exposed to hexamethonium.

It is unknown why Ellen Roche chose to join the study. Dr. Togias stated to the Committee he believes that the two most likely reasons were (1) an altruistic desire to help people with asthma and (2) monetary compensation ($25 for each of the first phase visits and $60 for each of the second phase visits, (totaling $365).

Whatever her motivation, the committee does not believe that Ms. Roche was coerced to join the study. There was no evidence of supervisory coercion, since she did not work in the laboratories of any of the investigators, and her employment status was not influenced in any way by her participation. There was a financial incentive, but the amount offered was consistent with the rate provided in other studies. The compensation was intended to provide for the participant’s time and was not provided for accepting the risks of participation.

. Was the study carried out appropriately?

a) Who performed the experiment?

The study protocol was carried out by Dr. George Pyrgos, a postdoctoral Research Fellow in Clinical Immunology, under the direct supervision of Dr. Alkis Togias, Associate Professor of Medicine. Dr. Pyrgos was born in Baltimore, received his education and medical training in Greece (where he was first in his class in medical school), and did a transitional year of medical residency at the Frankford Hospital in Pennsylvania from July 1998 to July 1999. He was appointed a Research Fellow in Clinical Immunology at Johns Hopkins in July 1999 and was in his final year of fellowship. Dr. Pyrgos passed his ECFMG examination in February 1998, had taken the Department of Medicine CORE (Course on Research Ethics), and had completed the JHUSOM on line training course on human subjects research; but he was not licensed to practice medicine in the State of Maryland.

Dr. Togias, the Principal Investigator, was born in Greece and obtained his M.D. in 1983 from the National and Kapodistrian University Medical School in Athens. He was a Research Fellow in Clinical Immunology at Hopkins from 1983-1985, an intern and resident on the Osler Medical Service from 1985-1988, and a Senior Clinical Fellow in Clinical Immunology from 1988-1989. He then joined the Hopkins faculty and was promoted in 1996 to Associate Professor of Medicine in the Clinical Immunology Division. He is an experienced clinical researcher with over 50 publications, numerous previous or current research grants, and 52 studies on file with the IRB (36 inactive studies and 16 active studies). He has been licensed to practice medicine in the State of Maryland since 1988.

Dr. Pyrgos worked closely with Dr. Togias in the planning and execution of the research protocol. Dr. Pyrgos recruited the research subjects from a normal volunteer registry. He had direct contact with the subjects during the protocol, including performance of spirometry and methacholine challenges, insertion of the intravenous line, monitoring of blood pressure and heart rate, and administration of hexamethonium or saline by inhalation. However, Dr. Togias confirmed to this investigation committee that he made all important decisions during the protocol, including the obtaining of informed consent, deciding whether the subject met inclusion/exclusion criteria and should be studied that day, and deciding whether to proceed with each dose of inhaled hexamethonium based on changes in hemodynamics and symptoms. The protocol was carried out in the A&AC room 3B.45, and Dr. Togias was often "in and out," as his office was located five doors away in room 3B.65B. If Dr. Togias was not in the laboratory room, Dr. Pyrgos took the hemodynamic monitoring records to him to obtain approval for the next hexamethonium dose. While Dr. Pyrgos was out of the laboratory, the subject was watched by Dr. Nicola Scichilone, a research fellow (who has completed his fellowship and now returned to Italy) who was there to measure exhaled breath nitric oxide (NO) levels as part of the study.

The research protocol was part of a long-standing program of asthma research directed by Dr. Solbert Permutt, Professor of Medicine. Dr. Permutt was the Principal Investigator of the NIH grant funding the research (HL61277), and played a principal role in the development of the scientific ideas underlying the protocol, including the use of a ganglionic blocker. However, he was not directly involved in the particulars of the approval or execution of the protocol being discussed.

Although Dr. Pyrgos was not licensed to practice in Maryland, he functioned as a research Fellow and carried out the protocol under Dr. Togias’ direct supervision. It should be noted that Dr. Pyrgos did not have to be a licensed physician to perform in this capacity. He was not called on to treat patients or prescribe drugs. A trained research technician could have also performed the functions called for in the experiment. The committee believes that Dr. Pyrgos was well trained to execute the research protocol, performed competently, and was supervised properly by Dr. Togias.

b) Was the experiment conducted in accord with the approved protocol?

The committee reviewed in detail what is known about the chemistry of hexamethonium as a background for interpretation of the research protocol. The synthesis of hexamethonium bromide is known to result in hydrogen bromide as a byproduct, and upon storage, hexamethonium bromide breaks down to an alkene (toxicology unknown) with the cleavage of the hexane portion of the molecule and the release of tetramethyl ammonium and hydrogen bromide. If not removed, the hydrogen bromide should lead to an acidic solution when the hexamethonium bromide with hydrogen bromide is dissolved in water. Since the formulation prepared from the purchased product in this case was rather easily buffered with sodium bicarbonate, it probably did not contain much hydrogen bromide. The final results from the actual chemical analysis of the hexamethonium powder used in the experiment are not yet available to the Committee.

Hexamethonium bromide exists as a salt of the positively charged hexamethonium, a quaternary ammonium compound, and the negatively charged bromide ions. The drug will exist in its ionized form at all reasonable physiologic pH=s, certainly from pH 5 through pH 8. As a charged (ionized) molecule, it would be expected to cross biologic membranes and enter cells poorly. This would seem consistent with known data for hexamethonium chloride, which is poorly and erratically absorbed from the intestinal tract. There is, on the other hand, precedent for some other small positively charged molecules crossing membranes and therefore it is unclear how hexamethonium would be handled after pulmonary administration.

The committee could not find studies describing the pulmonary disposition of hexamethonium delivered by inhalation to either experimental animals or humans. The pharmacokinetics of its absorption into the lung is unknown. It is not clear how much remains in the alveoli and for how long.

Two of the 4 papers cited by Dr. Togias used hexamethonium bromide, one used hexamethonium chloride, and in one the formulation was not identified. As far as can be ascertained, no formulation of hexamethonium bromide was ever approved by the FDA and no formulation of any salt of hexamethonium was ever approved by the FDA and/or marketed for use by inhalation.

Several steps were required to prepare the study substance for administration to humans. Hexamethonium was obtained from a chemical company, Fluka US. Dr. Pyrgos, a postdoctoral fellow in Dr. Togias’ laboratory, would weigh 2 grams of the substance on a scale and add this to 10 ml of normal saline, and then place it in a sterile container. The solution was then pushed through a 0.2 micron Nalgene sterile filter under a chemical fumes hood. The filtered solution was to be placed in a refrigerator for no more than one hour prior to being given to subjects by a nebulizer. A new solution of hexamethonium was to be made within two hours prior to administration in each subject. Once the experiment was complete, the unused portion of the prepared solution was discarded.

Because the osmolarity of the solution was high (1000 mM), the actual practice was to add the hexamethonium to distilled water, rather than normal saline using sterile technique. For research subject #1, the solution was stored in a refrigerator and then placed in a DeVilbiss nebulizer and administered. Because subject #1 suffered a cough that was thought to be due to the acidic pH of the solution (the pH of a subsequently prepared solution was 4.7), the preparation was changed by adding sodium bicarbonate as described below. Dr. Togias did not notify The JHBMC Institutional Review Board of this adverse event, or of the changes in preparation.

For subjects #2 and #3, the hexamethonium was weighed and mixed with distilled water. This solution was then buffered with a drop of 0.5 normal sodium bicarbonate solution to achieve a pH between 6.8 and 7.3, measured by pH meter. This buffered solution was then passed through a 0.2 micron filter.

For subject #3, the hexamethonium solution was made, as described above, approximately one hour prior to administration via a Pari-Jet LC Plus nebulizer. The type of nebulizer used for delivery was changed between subjects #2 and #3, because the delivery of hexamethonium took over 1.5 hours using the DeVilbiss nebulizer, and subject #1 and #2 complained of the duration. The IRB-approved protocol did not specify a particular type of nebulizer. Both types of these nebulizers are used on multiple subjects with antiseptic cleaning between uses. After the nebulizers are washed with the cleaning solutions, they are hung on a rack to dry. Dr. Pyrgos selected a nebulizer from this rack for use on subject #3.

Once prepared, the hexamethonium solution was stored for no more than 1 hour in the laboratory refrigerator. The solution was clearly marked. A limulus test had been performed twice on solutions prepared before exposure of subject #1, and both tests were negative. The limulus test was not repeated for each solution prepared for use in the research subjects. The solutions were not cultured.

The changes noted above in the preparation of hexamethonium (use of distilled water instead of saline to control for the high osmolarity and buffering with sodium bicarbonate to neutralize the high acidity) were not reported to the IRB. Neither was the switch of the hexamethonium control solution from normal saline to hypertonic saline (done to better match the osmolarity of the hexamethonium and control inhalants). Nevertheless, it is likely that these changes would have been approved by the IRB since they were made for subject comfort and to improve the scientific rigor of the study.

The investigation committee considered whether any changes in the protocol could have increased the lung concentration of hexamethonium in subject #3. Hexamethonium exists in an ionized form over a very broad range of pH, so buffering of the solution should have had a minimal effect on tissue uptake. Changing the type of nebulizer used to deliver the drug resulted in a more rapid delivery of the hexamethonium. Although this could have resulted in a greater concentration of hexamethonium in the lung tissue compared to the first two subjects, the pharmacokinetics of inhaled hexamethonium are not known and any possible increase in lung tissue concentration in subject #3 cannot actually be verified.

The method of monitoring the total dose of hexamethonium given might have been improved upon. Hexamethonium causes ganglionic blockade and postural hypotension. The best way to assess adequate blockade would be to measure postural hypotension, not just changes in sitting blood pressure and heart rate. In several of the previous human inhalation studies of hexamethonium, blood pressure was determined either in the standing position or in moving from the supine to the standing position, looking at changes of postural response. In Dr. Togias’ study, blood pressure was measured in the sitting position. The principal investigator defined criteria for determining ganglionic blockade and termination of hexamethonium administration consisted of an increase in heart rate by 30 beats per minute, or reduction in the systolic blood pressure by 30 mm Hg. No subject exceeded these stopping rules during hexamethonium administration.

c) What were the results in the three research subjects?

Of the 9 subjects who signed the consent form and began the study, only 3 reached the point in the protocol calling for hexamethonium inhalation. The following outlines what transpired in each of these 3 subjects.

SUBJECT #1

 

April 9, 2001: Consent form signed

April 9, 2001: Methacholine-Bronchodilation

Baseline FEV1/FVC = 2.32/2.93

Methacholine 20 mg nebulization administered

Post-Methacholine FEV1/FVC = 1.83/2.35 (FEV1 reduced by 21%)

Four deep inspirations (DIs) administered

Post-DI FEV1/FVC = 1.93/2.50 (Bronchodilation = 17%)

No mention of adverse effects was noted for this experiment on April 9, 2001.

April 10, 2001: Methacholine-Bronchodilation

Baseline FEV1/FVC = 2.65/3.16

Methacholine 40 mg nebulization administered

Post-Methacholine FEV1/FVC = 1.74/2.26 (FEV1 reduced by 34%)

Four DIs administered

Post-DI FEV1/FVC = 2.30/2.98 (Bronchodilation = 61.5%)

No mention of adverse effects was noted for this experiment on April 10, 2001.

 

April 11, 2001: Methacholine-Bronchoprotection

Baseline FEV1/FVC = 2.68/3.19

Five DIs administered

Methacholine 40 mg nebulization administered

Post-Methacholine FEV1/FVC = 2.24/2.71 (Bronchoprotection = 53%)

No mention of adverse effects was noted for this experiment on April 11, 2001.

April 12, 2001: Methacholine-Hypertonic Saline-Bronchodilation

Baseline FEV1/FVC = 2.62/3.09

Saline nebulization administered

Post-saline FEV1/FVC = 2.71/3.24

Methacholine 40 mg nebulization administered

Post-methacholine FEV1/FVC = 2.36/2.97 (FEV1 reduced by 13%)

Four DIs administered

Post DI FEV1 = 2.47 (9% bronchodilation)

(The reduction in FEV1 after this 40 mg dose of methacholine was less than the required 20%. Subject was asked to return for additional testing at higher Methacholine dose.)

No mention of adverse effects was noted for this experiment on April 12, 2001.

April 18, 2001: repeat Methacholine-Hypertonic Saline-Bronchodilation

Baseline FEV1/FVC = 2.70/3.08

Saline nebulization administered

Post-saline FEV1/FVC = 2.64/3.01

75 mg methacholine nebulization administered

Post-Methacholine FEV1/FVC = 1.90/2.47 (28% decrease in FEV1)

Four DIs administered

Post-DI FEV1/FVC = 2.24/2.77 (bronchodilation = 46%)

No mention of adverse effects was noted for this experiment on April 18, 2001.

April 20, 2001: Methacholine-Hypertonic Saline-Bronchoprotection

Baseline FEV1/FVC = 2.65/3.10

Saline nebulization administered

Post-saline FEV1/FVC = 2.55/2.95

Five DIs administered

Methacholine 75 mg nebulization administered

Post-methacholine FEV1/FVC = 2.29/2.67 (bronchoprotection = 64%)

No mention of adverse effects was noted for this experiment on April 20, 2001

April 23, 2000: Methacholine-Hexamethonium-Bronchodilation

Baseline FEV1/FVC = 2.68/3.23

Baseline blood pressure recorded at 10:01 was 127/40. Additional blood pressures recorded before hexamethonium administration were 119/85 and 126/88. Hexamethonium was administered from ~ 10:30 to ~12:00 using the deVilbiss 646 nebulizer attached to the Rosenthal dosimeter. Investigators estimated that Subject #1 received 1.009 grams of hexamethonium on this date. During the hexamethonium administration, the lowest blood pressure recorded was 105/70, and the highest heart rate recorded was 94. The hemodynamic end-points for stopping the hexamethonium administration were not met, allowing the hexamethonium administration to continue to completion at approximately 11:40 a.m. Nitric oxide measurements in exhaled air were obtained seven times from 10:00 to 11:15.

Methacholine 75 mg nebulization administered

Post-methacholine FEV1/FVC = 1.20/1.68

Four DIs administered

Post-DI FEV1/FVC = 2.38/2.84 (bronchodilation = 77%.)

No mention of adverse effects was noted for this experiment on April 23, 2001.

April 25, 2001 Clinical Follow-up

Subject reported mild shortness of breath and a non-productive cough associated with deep inspiration. No other symptoms such as sputum production, fever, or chest pain were reported. Spirometry: FEV1/FVC = 2.33/2.74. These values were reduced by approximately 15% from the most recent baseline values. The patient was examined by Dr. Togias who reported no abnormal physical findings but recommended follow-up the next day.

April 26, 2001 Clinical Follow-up

The subject reported no change in symptoms. Spirometry: FEV1/FVC = 2.34/2.82. These values were virtually the same as those obtained on April 25 (reduced relative to the most recent baseline values obtained on April 23, the day that hexamethonium was administered).

April 27, 2001 Clinical Follow-up

The subject reported no changes in symptoms. The cough remained associated with physical activity and deep breathing, and she complained of dyspnea and coughing with exertion. Spirometry: FEV1/FVC = 2.47/2.97. These values were slightly improved relative to the previous two days but not yet back to the baseline obtained on April 23rd, before hexamethonium administration. A lung diffusing capacity measurement of 17.5 ml/minute/mm Hg was 81% of predicted, which is within the normal range. There was no explicit notation regarding subsequent follow-up.

April 30, 2001 Clinical Follow-up

A notation in the investigators’ log indicates that the volunteer was contacted and asked to repeat spirometry. The volunteer reported that the symptoms were improved but still present. No spirometry is reported on April 30th.

May 1, 2001 Clinical Follow-up

The subject reported that the symptoms were much better but that there was still some coughing. Dyspnea on exertion was reduced. Spirometry: FEV1/FVC = 2.39/2.70. These values were slightly lower than those obtained most recently on April 27th.

May 3, 2001 Clinical Follow-up

The volunteer reported complete resolution of symptoms. Spirometry: FEV1/FVC = 2.41/2.91. These values were not back to the baseline values obtained on the day that hexamethonium was administered, but they were similar to the values obtained on April 9, 2001 (2.32/2.93). With this objective information and the patient’s subjective improvement, the patient was advised that the study could continue in "a couple weeks if there is no change in broncho-protection as determined by methacholine and deep inspirations".

May 6, 2001 Clinical Follow-up

The subject was asked to return to the laboratory for physical examination. The examination was unremarkable. A chest radiograph on May 15, 2001 was reported as normal.

Dr. Togias did not immediately report the symptoms and reductions in spirometry experienced by Subject #1. After Subject #3 became ill, Togias reported this event in Subject #1 in his May 18, 2001 letter to Judith Stiff (JHBMC IRB). Togias explained to the Committee that he initially attributed this event in Subject #1 to a respiratory tract infection because many workers in the Asthma Center were experiencing a flu-like syndrome at the time. Togias added that under his direction, the pH of the hexamethonium solution was checked and found to be acidic (pH 4.7) In his interview with the Johns Hopkins internal committee, Dr. Pyrgos explained that one drop of a (0.5 M) sodium bicarbonate solution was required to raise the pH to approximately 6.8-7.5. This pH-adjustment was used for the hexamethonium used in Subjects #2 and #3.

 

SUBJECT #2

 

April 10, 2001: Consent form signed

April 10, 2001: Methacholine-Bronchodilation

Baseline FEV1/FVC = 2.98/3.65.

Methacholine 20 mg nebulization administered.

Post-methacholine FEV1/FVC = 1.44/1.87 (FEV1 reduced by 52%)

Four DIs administered

Post-DIs FEV1/FVC = 2.72/3.43 (Bronchodilation = 83%)

No mention of adverse effects was noted for this experiment on April 10, 2001.

 

April 11, 2001: Methacholine-Bronchoprotection

Baseline FEV1/FVC = 3.07/3.57

Five DIs administered

Methacholine 20 mg nebulization administered

Post-methacholine FEV1/FVC = 2.43/3.08 (Bronchoprotection = 60%)

No mention of adverse effects was noted for this experiment on April 11, 2001.

April 17, 2001: Methacholine-Hypertonic Saline-Bronchodilation

Baseline FEV1/FVC = 2.91/3.41

Saline nebulization administered

Post-saline FEV1/FVC = 2.87/3.42

Methacholine 20 mg nebulization administered

Post-methacholine FEV1/FVC = 1.73/2.09 (FEV1 reduced by 40%)

Four DIs administered

Post-DIs = FEV1/FVC = 2.39/3.07 (Bronchodilation = 74%)

No mention of adverse effects was noted for this experiment April 17, 2001.

April 19, 2001: Methacholine-Hypertonic Saline-Bronchoprotection

Baseline FEV1/FVC = 2.89/3.45

Saline nebulization administered

Post-saline FEV1/FVC = 2.81/3.37

Five DIs administered

Methacholine 20 mg nebulization administered

Post-methacholine FEV1/FVC = 2.45/3.12 (Bronchoprotection = 67%)

No mention of adverse effects was noted for this experiment on April 23, 2001.

April 27, 2001: Methacholine-Hexamethonium-Bronchodilation

Pregnancy test negative

Baseline FEV1/FVC = 2.91/3.40

Baseline blood pressure at 9:40 was 136/86 with pulse rate = 75. Blood pressure and pulse were recorded several additional times before beginning hexamethonium administration. The lowest blood pressure recorded before hexamethonium was 98/71 with pulse = 78 (at 9:42). Other blood pressures and pulse rates before hexamethonium were similar to the initial blood pressure and pulse. Hexamethonium administration was initiated at approximately 10:30 using the DeVilbiss 646 nebulizer attached to the Rosenthal dosimeter. After beginning the hexamethonium administration, the lowest blood pressure recorded was 120/86. The highest heart rate while the subject was sitting or supine was 86. The hemodynamic end-points for stopping hexamethonium were not met. Therefore, the full dose of hexamethonium was administered. The investigators estimated the subject received 0.79 grams of hexamethonium on this date. Several assessments for exhaled nitric oxide were recorded from 9:15 to12:10.

Post-hexamethonium FEV1/FVC = 3.05/3.21

Methacholine 20 mg nebulization administered

Post-methacholine FEV1 = 1.95/2.39

Four DIs administered

Post-DIs FEV1/FVC = 2.63/3.03 (Bronchodilation = 61%)

No mention of adverse effects was noted for this experiment on April 27, 2001.

In his letter to Dr. Judith Stiff (JHBMC IRB) on May 18, 2001 and in his interview with the Johns Hopkins internal committee, Dr. Togias explained that the hexamethonium nebulization equipment used in Subject #1 and the first hexamethonium administration in Subject #2 required a long and tedious session with many breaths of the nebulized solution. Togias explained that the subjects complained about this session. Therefore, a decision was made to change the nebulization equipment to a Pari-Jet nebulizer attached to a Duraneb 2000 compressor. This allowed completion of the nebulization in a shorter period of time and with fewer breaths. In a letter dated May 24, 2001, Dr. Stiff asked how this change could affect the dose of hexamethonium received by the subjects or the particle size of the nebulized solution. Togias explained in a letter of May 30, 2001 that the dosing used in the newer system should have delivered ~1 gram (the intended dose) to the subject=s mouth (assuming completion of the hexamethonium administration). Togias explained further that the deVilbiss 646 nebulizer delivers ~36% of particles with diameters of 1-5 microns. The Pari-Jet nebulizer delivers ~52% of particles in this range. Togias added that this difference could have resulted in more aerosol being delivered to the periphery of the lung with the Pari-Jet nebulizer.

May 1, 2001: Methacholine-Hexamethonium-Bronchoprotection

Baseline FEV1/FVC = 2.92/3.34

The baseline blood pressure and heart rate at 10:24 were 131/94 and 71, respectively. Hexamethonium administration was initiated at 10:26 using the Pari-Jet nebulizer attached to a Duraneb 2000 compressor. Two doses of hexamethonium were administered, one at 10:26 and the second at 10:45. Three assessments of exhaled nitric oxide were obtained from 9:43 to 10:55. At 10:55, orthostatic values of blood pressure and heart rate were 110/93 and 102, respectively. These orthostatic values did not require cessation of the hexamethonium. However, the investigators reported that there was ptosis and that the patient appeared tired. Therefore, the hexamethonium administration was stopped at this point. Subsequent blood pressures recorded after the hexamethonium was stopped at 10:55 were higher (121/80, 118/94, 126/90), and subsequent heart rates were lower (76, 99, and 87).

Post-hexamethonium FEV1 = 3.04/3.35

Five DIs administered

Methacholine 20 mg administered

Post-Methacholine FEV1/FVC = 1.76/2.20 (Bronchoprotection = -14%)

The subject was discharged from the laboratory in satisfactory condition.

The subject was asked to return to the laboratory on May 16, 2001 by Dr. Togias, who conducted a physical examination. The examination was normal. Specifically, the lungs were clear to auscultation and to forced expiration. A chest radiograph obtained May 15, 2001 was normal.

SUBJECT #3

 

April 16, 2001: Consent form signed

April 16, 2001: Methacholine-bronchodilation

Baseline FEV1/FVC = 2.97/3.43

Methacholine 20 mg nebulization administered

Post-methacholine FEV1/FVC = 1.33/1.57 (FEV1 reduced 55%)

Four DIs administered

Post-DIs FEV1/FVC = 1.81/2.26 (Bronchodilation = 29%)

Two puffs of albuterol were given to reverse the methacholine effects.

No mention of adverse effects was noted for this experiment on April 16, 2001.

April 18, 2001: Methacholine-bronchodilation

Baseline FEV1/FVC = 3.10/3.63

Methacholine 20 mg nebulization administered

Post-methacholine FEV1/FVC = 2.62/3.13 (FEV1 reduced by only 15%)

No mention of adverse effects was noted for this experiment on April 18, 2001.

April 19, 2001: Methacholine-bronchodilation

Baseline FEV1/FVC = 3.16/3.60

Methacholine 20 mg nebulization administered

Post-methacholine FEV1/FVC = 2.77/3.44 (FEV1 reduced by only 12%)

No mention of adverse effects was noted for this experiment on April 19, 2001.

April 23, 2001: Methacholine-bronchodilation

Baseline FEV1/FVC = 3.13/3.57

Methacholine 40 mg nebulization administered

Post-methacholine FEV1/FVC = 1.87/2.31 (FEV1 reduced 40%)

Four DIs administered

Post-DIs FEV1/FVC = 2.40/2.89 (Bronchodilation = 42%)

No mention of adverse effects was noted for this experiment on April 23, 2001.

April 26, 2001: Methacholine-bronchoprotection

Baseline FEV1/FVC = 3.02/3.51

Five DIs administered

Methacholine 40 mg nebulization administered

Post-methacholine FEV1/FVC = 1.64/1.92 (Bronchoprotection = -13%).

No mention of adverse effects was noted for this experiment on April 26, 2001.

May 1, 2001: Methacholine-bronchoprotection

Baseline FEV1/FVC = 2.93/3.42

Five DIs administered

Methacholine 40 mg nebulization administered

Post-Methacholine FEV1/FVC = 2.34/2.89 (Bronchoprotection = 50%)

No mention of adverse effects was noted for this experiment on May 1, 2001.

May 2, 2001: Methacholine-Hypertonic Saline-Bronchodilation

Pregnancy test negative

Baseline FEV1/FVC = 3.05/3.46

Saline nebulization administered

Post-saline FEV1/FVC = 2.96/3.29

Methacholine 40 mg nebulization administered

Post-Methacholine FEV1/FVC = 1.81/2.25 (FEV1 reduced 39%)

Four DIs administered

Post-DIs FEV1/FVC = 2.61/3.14 (bronchodilation = 70%)

Four assessments of exhaled Nitric Oxide from 14:15 to ~15:30.

No mention of adverse effects was noted for this experiment on May 2, 2001.

May 4, 2001: Methacholine-Hexamethonium-Bronchodilation

Baseline FEV1/FVC = 3.08/3.60

Several blood pressures and pulse rates were recorded beginning at 14:24 (the recorded automatic blood pressure monitor times were all off by one hour; i.e., "13.24" should be "14.24"). A series of breath nitric oxide measurements were made from 14:20 to about 14:30. The baseline blood pressure was 134/81 and heart rate 90, recorded at 14:34 on the monitor print out. Hexamethonium administration began at about 14:43 (it could not have begun later than this, based on back calculation from the start of the second hexamethonium dose). At 14:46 (according to the monitor print out), about 3 min into the 7 min hexamethonium dosing period, the blood pressure was 146/86 and the heart rate 79 beats/min. These values were written on the first line of the Vital Signs Monitoring Sheet as occurring at 14:40, but the monitor print out indicates that they occurred at 14:46. Although these values were written on the first line as "time zero", they do not indicate the true baseline, since hexamethonium dosing was already in progress. As noted above, the true baseline was 134/81, recorded at 14:34. The first dose finished by 14:50, and breath NO measurements were repeated. The second hexamethonium dose began at 14:55, when blood pressure was 142/78 and heart rate 82 beats/min.

Hexamethonium was administered with the Pari-Jet nebulizer with the Duraneb 200 compressor. The nebulization proceeded through completion of the anticipated dose (1 gram). Exhaled nitric oxide was assessed seven times between 14:20 and 16:15. During the approximately two hours of dosing, blood pressures and heart rates were recorded frequently. The lowest recumbent blood pressure recorded on the Vital Signs Monitoring Sheet was at 16:03 = 115/93. This was 19 mm Hg lower than the baseline blood pressure at 14:34. A lower systolic blood pressure of 98 mm/Hg was recorded, but this was an orthostatic measurement.

Post-hexamethonium FEV1/FVC = 3.32/3.72

Methacholine 40 mg nebulization administered

Post-methacholine FEV1/FVC = 2.63/3.10 (FEV1 reduced by 21%)

Four DIs administered

Post-DIs FEV1/FVC = 2.59/3.19 (no bronchodilating effect)

The subject was monitored until 18:00, at which time orthostatic changes were reported to have resolved. The volunteer reported fatigue only and was discharged from the laboratory in satisfactory condition.

May 7, 2001: Clinical Follow-up

On May 7, 2001, Ms. Roche contacted the investigators and reported that symptoms began on 5/5/01. The subject complained that on the day after the last visit, she experienced cough with inspiration. The cough was non-productive, and she denied other symptoms. On May 7th she reported rhinorrhea and feeling hot (but had not checked her temperature). Spirometry: FEV1/FVC = 1.93/2.20. These values were reduced ~35% from baseline values. Dr. Pyrgos reported these findings to Dr. Togias (he was out of town, with Dr. Mark Liu covering), and Dr. Togias indicated that there should be follow-up with the subject the next day. Dr. Pyrgos reported to the committee that he had called Ms. Roche on May 8th at home and the subject indicated that she was not feeling well and was going to see her primary care provider (which she did and received a prescription for an anti-tussive medication). On May 9th, the symptoms had not resolved and the subject was asked by the investigator to return to the laboratory for further evaluation. A chest radiograph showed abnormalities consistent with pneumonitis, her temperature was 101 degrees Fahrenheit, and her oxyhemoglobin saturation decreased from 92% to 84% after walking a short distance. She was admitted to the hospital on May 9th.

The serious adverse event experienced by Subject 3 was reported to the JHBMC IRB in a letter from Dr. Togias on May 9, 2001. Dr. Togias explained that the protocol would be "placed on hold at this point." No further experiments were conducted.

 

 

 

. Was medical care provided promptly and appropriately?

Ms. Roche underwent a series of methacholine challenges over a period of two weeks, beginning on 4/16/2001. On 5/4/2001, as part of the protocol, she received hexamethonium (1 gram total dose) to produce parasympathetic blockade. The next day, she developed a dry cough and dyspnea on exertion. She returned to the laboratory of Dr. Togias on 5/7/01. She was evaluated, reported flu-like symptoms, and her FEV-1 was reduced. She was evaluated by her local primary care source on 5/8/2001 and was given cough suppressant, which she used, and antibiotics, which she did not use. On 5/9/2001, with worsening symptoms, she was again seen in the laboratory of Dr. Togias. She was febrile and desaturated to 84% after walking a short distance. Her family was present with her at this visit and spoke to Dr. Togias.

Ms. Roche was evaluated on 5/9/2001at the JHBMC Emergency Department, where she was evaluated by the Medical/Pulmonary Services, and she was admitted to the hospital. Her parents had accompanied her to the hospital. Her chest x-ray demonstrated streaky densities in the perihilar region. She developed increasing exertional dyspnea and hypoxemia and was provided supplemental oxygen via nasal cannula. She was transferred to the PCU, and then was transferred to the Medical Intensive Care Unit over a series of days (5/10/2001 B 5/12/2001) for progressive pulmonary insufficiency. In spite of appropriate empiric antibiotics for possible community acquired pneumonia (CAP), her clinical condition worsened. On 5/12/2001, a chest computerized tomography (CT) demonstrated a ground glass appearance of the lungs, worse at the lung bases. Prior to this CT diagnosis, several diagnoses, including CAP due to typical and atypical pathogens, were considered, as well as possible pulmonary emboli. Appropriate medical therapy was employed for these possible diagnoses. Participation in the research study and exposure to hexamethonium were known and considered as a cause of her illness. Steroid therapy was initiated.

On 5/14/2001, Ms. Roche was intubated and mechanically ventilated due to progressive respiratory failure. The timing and management of her respiratory failure and progressive pulmonary infiltrates (an ARDS clinical picture) was consistent with state-of-the art critical care and included all of the principles and practices in the recently completed ARDS network trial using low tidal volume therapy and control of plateau pressure. Prone positioning was used to improve oxygenation. In addition to the intensivists directing her care in the ICU, infectious disease consultants, renal medicine consultants, and surgical consultants provided appropriate and comprehensive differential diagnoses and treatment. An extensive, thorough and comprehensive search for infectious pathogens was coordinated with the Infectious Disease team and discussions with the Center for Disease Control. Hantavirus, Q Fever, and Legionella were considered and ruled out. Additional common bacterial pathogens were also considered. A pneumothorax developed, was identified, and treated with a tube thoracostomy.

In spite of ongoing supportive management and continued excellent care, Ms. Roche developed worsening hypoxemia, requiring increasing ventilatory support and prolonged ICU care. She developed a catheter-related infection, which again was promptly identified and treated. She was managed with appropriate ventilatory management, and nitric oxide (NO) therapy was added to her management on 5/27/2001. Though she had an initial response to NO therapy, oxygenation worsened, and renal failure and progressive hypotension developed. Dialysis was initiated. Ms. Roche=s family considered additional therapeutic options, including ECMO, which were discussed. Given her worsening condition and unsupportable oxygenation, her family elected to withdraw support and Ms. Roche died on 6/2/2001.

In summary, Ms. Roche presented to the laboratory where she had participated in the inhalation of hexamethonium experiment on 5/9/2001. Dr. Togias recognized that she was unwell and brought her to the emergency department with her family for evaluation. She was admitted for oxygen therapy and monitoring. Dr. Togias excused himself from providing direct medical care, but he remained concerned and involved in the case (per the internal investigation committee interview with Dr. Togias.) All available medical treatment options were considered and utilized in this case. Documentation about these issues in the medical record was excellent.

Based on the interview with Dr. Togias, Ms. Roche’s family was kept appraised of her condition. References to discussions about her prognosis and outcome are documented in the medical record. Little additional information regarding family communication is available. Lastly, the report of the death of Ms. Roche to the IRB was done in a timely fashion. Release of information to the public was delayed at the request of the family.

. What was the cause of the serious adverse event?

We believe that we will never be certain about the cause of this subject's death. During her month-long admission to the hospital, multiple tests and cultures were performed in an effort to identify an infectious or other cause for her condition. The cultures were only positive for bacteria that normally inhabit the areas of the body from which they were taken and would not be expected to be responsible for infection. No specific disease process was identified through all of the other testing performed before her death, which included sending blood specimens to look for antibody evidence of leptospirosis (ELISA negative), recent infection by Mycoplasma pneumoniae (IgG by ELISA slightly positive at 1.19, with negative range up to 1.10; IgM antibody negative--results typical of long past infection with residual immunity), Q fever (complement fixation negative), hantavirus (both IgG and IgM negative), or Legionella pneumophila (IFA negative). These test results offer no basis for a diagnosis of any of the infections sought. Her autopsy did not show characteristic changes of any specific infection or other pathologic condition, just the evidence of widespread and extensive damage to the lungs seen with a variety of different disease processes. Specifically, there were diffuse areas of hemorrhagic patches in both lungs. These foci were worse in the upper lobes on both sides. There was extreme congestion, and the appearance of septation within the parenchyma. There was no gross evidence of infiltrates, consolidation, hilar lymphadenopathy, or pulmonary embolism. Overall, the gross picture was consistent with that of an acute and diffuse injury to the lungs. The microscopic examination of the lungs revealed extensive, diffuse loss of alveolar space with marked fibrosis and fibrin thrombi involving all lobes. There was also evidence of alveolar cell hyperplasia as well as chronic inflammation compatible with an organizing stage of diffuse alveolar damage. There was no evidence of bacteria, fungal organisms, or viral inclusions on routine or special stains. It should be noted that the absence of specific evidence of infection does not rule out the possibility that infection caused or contributed to the process; we simply can not identify a specific pathogen or suspicious family of pathogens based on the autopsy findings.

All this said, it is the overall opinion of the group that the inhaled hexamethonium phase of the experiment was either solely responsible for the subject's illness or played an important contributory role. While this conclusion can not be confirmed with certainty, the previous good health enjoyed by the subject and the proximity of the onset of her symptoms to the hexamethonium exposure suggested a causal relationship. Whether she had a particular underlying susceptibility to the injury induced by the hexamethonium phase of the experiment is unknown but possible. Whether such a susceptibility could have or should have been identifiable before she entered the study is extremely unlikely.

Conclusions

This section summarizes the major conclusions of the internal investigation. It should be noted that for some of the conclusions, the committee was divided, and this document consequently represents the opinion of the majority rather than a total consensus. In addition, it should be emphasized that although the facts in this case were relatively easily established, deciding what actions should or should not have been taken by the IRB or the PI at the time events were transpiring was often very difficult, and engaged several unsettled, important issues in research ethics. Some of the committee’s conclusions therefore represent value judgements, influenced considerably by hindsight.

1. Was the study designed to address an important scientific question?

The research study addressed an important scientific question about the role of bronchial nerves in human asthma. The study had solid scientific rationale and was well designed to answer the questions posed. The study required human subjects and could not have been adequately done in experimental animals. The research was of particular importance given the increasingly high prevalence of asthma in urban populations, including Baltimore.

2. Was the review of the study protocol by the JHBMC IRB appropriate?

The review of the research protocol by the JHBMC IRB followed its published standard procedures. Considerable relevant expertise existed on the IRB for the scientific review. Nevertheless, the majority of the committee concluded that an adequate evidence base did not exist for the IRB to be confident that inhaled hexamethonium was safe for use in research subjects. The four published papers provided to the IRB did not mention any pulmonary toxicity; but even in the absence of pulmonary adverse events in the 20 subjects, there was uncertainty regarding the true risk of inhaled hexamethonium. The committee believes that the IRB should have required more safety evidence for a non-FDA approved drug no longer in clinical use, and administered by a non-standard route.

A way to obtain evidence of safety would have been to have Dr. Togias submit an IND application to the FDA for review. Although the legal requirement is unclear, the committee believes that the IRB should have asked Dr. Togias to obtain a written opinion from the FDA about the need for an IND for inhaled hexamethonium for safety reasons. In view of the relatively limited safety data available in the literature, it is not clear that inhaled hexamethonium would have automatically been exempt from IND requirements, despite the fact that it was not going to be marketed for any therapeutic purpose. However, other inhaled drugs for pulmonary studies had been considered exempt by the FDA in the past. Both external consultants thought that the need for an IND was far from clear, and that there was adequate precedent for approving the protocol without one.

3. Was the consent form appropriate?

The committee believes that the consent form should not have been approved by the IRB. Although the consent form accurately outlined the purpose of the study, the procedures, the benefits and the alternatives, it was inadequate in the description of the research risks, and should have indicated that (a) hexamethonium was not FDA approved and was no longer used clinically, (b) hexamethonium as an inhalant was used only experimentally, never clinically, (c) the safety of inhaled hexamethonium was not known with certainty since it was based on published reports involving a total of only 20 people, and (d) there was a possibility of serious adverse event or death related to study participation. In addition, the committee was divided on whether the PI should have known about the published case reports linking prolonged IV hexamethonium use with pulmonary disease, and, therefore, whether this information should have been included in the consent form. The committee agreed that the PI had made a good-faith effort to discover potential toxicities through standard searches of the current medical literature and textbook reviews. The committee uniformly believes that had the information about pulmonary toxicity been available, it should have been included in the consent form and put into an appropriate context.

4. Was any coercion involved in the recruitment of the research subject?

The committee believes that Ms. Roche was not coerced to join Dr. Togias’ study. Although Ms. Roche worked in the Asthma and Allergy Center, she did not work in the laboratory of the investigators who conducted this research. She was an active member of an Asthma Study Registry and had participated previously in asthma studies. There is no evidence that any condition of her employment was affected in any way by her participation in the study. The financial incentive might have been a factor in her joining, but was modest and within the guidelines approved by the IRB and offered by other research studies at Hopkins.

. Was the study carried out appropriately?

(a) Who performed the study?

The study was carried out by a postdoctoral Research Fellow, Dr. George Pyrgos, under Dr. Togias’ direct supervision. Dr. Togias personally made all of the important decisions in the execution of the research protocol.

(b) Was the experiment conducted in accordance with the approved protocol?

A description of the proposed preparation of the hexamethonium was approved by the IRB, but the actual preparation used for Ms. Roche was modified without IRB approval. Use of distilled water as a diluent and buffering with sodium bicarbonate were not presented to the IRB, nor was the use of hypertonic saline instead of normal saline for the hexamethonium control. However, it is likely that the IRB would have approved the changes in protocol, since the modifications were made to improve subject comfort and make the control solution more comparable to the hexamethonium solution in terms of osmolarity. Both the committee and the external consultants thought these unreported protocol changes (cited by the FDA) were unlikely to have had a significant bearing on the outcome.

(c) What were the results in the three research subjects?

Dr. Togias did not inform the IRB that Subject #1 developed a cough and exertional dyspnea lasting 1 week following exposure to hexamethonium until after Ms. Roche developed a significant adverse pulmonary complication and was admitted to the hospital. He explained that the adverse event in Subject #1 was not an "unexpected and serious adverse event," because it was self-limited and required no treatment and therefore did not require immediate reporting to the IRB. He further explained that he thought the symptoms were related to a URI which was prevalent in the A&AC or possibly to the low pH of the hexamethonium solution (discovered by testing another preparation). Dr. Togias also said that he did not seriously consider the possibility of hexamethonium toxicity. Subject #2 was exposed to hexamethonium while Subject #1 still had pulmonary symptoms. Subject #2 did not report any symptoms. Subject #3 (Ms. Roche) was exposed 1 day after the symptoms disappeared in Subject #1. The committee believes that Dr. Togias should have considered the possibility of hexamethonium-induced pulmonary toxicity in Subject #1. Most members believe that he should have reported the event promptly to the IRB, delayed the exposure of the next subject to hexamethonium until the symptoms resolved in subject #1, and searched more comprehensively for previous reports of hexamethonium-induced pulmonary toxicity.

6. Was medical care provided promptly and appropriately?

The committee believes that Ms. Roche received state-of-the-art medical care after having developed pulmonary symptoms and presenting for care at the JHBMC Emergency Department. Medical care was delivered promptly and appropriately at the JHBMC, independent of the research team. Liberal use was made of expert consultants and all treatment modalities known to be effective were employed. Despite intensive efforts, Ms. Roche’s clinical course was progressively downhill and ultimately fatal.

7. What was the cause of the serious adverse event?

The autopsy on Ms. Roche demonstrated diffuse alveolar damage but provided no specific etiologic diagnosis. The committee believes the precise cause of death is likely to remain uncertain. Nevertheless, the committee also believes that the death was most likely the result of participation in the hexamethonium phase of the experiment. This belief is based on the following considerations: (a) the timing between hexamethonium inhalation and the development of pulmonary symptoms, (b) the lack of any identifiable cause, including negative cultures and serological tests for a wide variety of infectious agents, (c) the development of pulmonary symptoms, although less severe, after hexamethonium in Subject #1, and (d) the previously reported association between prolonged intravenous hexamethonium administration and pulmonary toxicity in a few seriously ill patients.

 

IV. Materials Reviewed During Investigation

Original NIH grant application ALung Inflation and Airways Hyperresponsiveness,@ Solbert Permutt, M.D., Principal Investigator, submitted 1/29/97 (R01 HL61277).

Applications for continuation of above grant, including progress reports, dated 11/15/99 and 9/27/00.

Final Autopsy Report on Ellen Roche, Johns Hopkins Bayview Medical Center, 6/15/01.

Medical Record on Ellen Roche, Johns Hopkins Bayview Medical Center from 5/7/2001 to 6/2/2001.

Detailed research files on the 3 volunteers studied under RPN AAC00-07-26-02 and receiving inhaled hexamethonium, including signed consent forms, spirometry records, results of methacholine challenges, hemodynamic and spirometry records during inhalation of hexamethonium, followup notes, physical examination and spirometry tests.

Original application to the JHBMC IRB for the study entitled "Mechanisms of Deep Inspiration - Induced Airway Relaxation", Alkis Togias, M.D., Principal Investigator, dated 7/25/00, assigned RPN number AAC00-07-26-02.

Letter dated 8/10/00 from the IRB to Dr. Togias, asking questions raised in the review; and the 9/14/00 response from Dr. Togias to IRB, responding and providing an updated consent form. Letter from IRB to Dr. Togias indicating approval of the protocol and consent form, 9/18/00.

Letter from Dr. Togias to IRB, 4/26/01, requesting approval of a proposed advertisement for the study.

Letter from IRB to Dr. Togias approving the advertisement.

Letter from Dr. Togias to IRB, 5/9/01, reporting a serious adverse event in Ellen Roche, pointing out an association in the literature between hexamethonium and Bronchiolitis Obliterans Organizing Pneumonia (BOOP), reporting the occurrence of coughing as an adverse event in the first volunteer, and recommending that the study be placed on hold.

E-mail from IRB to Dr. Togias, 5/16/01, asking questions about the serious adverse event and inviting him to a meeting of the full IRB.

Letter from Dr. Togias to IRB, 5/18/01, replying to the questions.

Letter from IRB to Dr. Togias, 5/24/01, asking more questions and indicating that the PI failed to inform the IRB of protocol changes.

Letter from Dr. Togias to IRB, 5/30/01, responding to the questions.

Letter from Dr. Togias to IRB, 6/3/01, informing the IRB of Ellen Roche=s death.

Certificate of Analysis on hexamethonium, Fluka US, 10/30/00.

Note to the record by Judith Stiff, M.D., Vice Chair JH Bayview IRB, 6/13/01.

Four publications cited by Dr. Togias in which inhaled hexamethonium had been given to normal volunteers without adverse effects.

1. Holtzman MJ et al. Effect of ganglionic blockade on bronchial reactivity in atopic subjects. Am Rev Respir Dis 1980; 122:17-25.

2. O=Byrne PM et al. The effect of inhaled hexamethonium bromide and atropine sulphate on airway responsiveness to histamine. J Allergy Clin Immunol 1985; 76:97-103.

60783. Sterk PJ et al. Limited maximal airway narrowing in nonasthmatic subjects. Role for neural control and prostaglandin release. Am Rev Respir Dis 1985; 132:865-70.

60784. Ichinose M et al. Nonadrenergic bronchodilatation in normal subjects. Am Rev Respir Dis 1988; 138:31-34.

List of studies in which Ms. Roche participated.

Federal Register (3/19/87):

Regulations, Food and Drug Administration 21 CFR Parts 312, 314, 511, and 514. New Drug, Antibiotic, and Biologic, Drug Product Regulations.

Letter from OHRP to Johns Hopkins University, 6/15/01.

Interviews with:

Dr. Gary Briefel, Chairman JHBMC IRB

Dr. Alkis Togias, Associate Professor of Medicine, Principal Investigator

Dr. George Pyrgos, Research Fellow, Clinical Immunology

Dr. Darrell Abernethy, member of the JHBMC IRB

Textbooks:

Surgical Pathology of Non-Neoplastic Lung Disease, Major Problems in Pathology, Katzenstein, A-L, Askin FB, W.B. Saunders Co., Vol. 13, 1990.

Goodman and Gilman’s Pharmacological Basis of Therapeutics, McGraw Hill, 9th Edition, 1996.

3. Fishman’s Pulmonary Diseases and Disorders, McGraw Hill, 3rd Edition, 1998; 2nd edition 1998.