---

8
Availability, Safety, and Efficacy
of Drugs and Other Therapies

---

     This chapter reviews current and potential countermeasures for the chemical and biological agents. Discussion of chemical agents includes assessment of availability at both the first responder and local treatment facility level because of the need for rapid action in many cases. Treatment of victims of most of the biological agents being considered in this report is not so time dependent (in most instances there will not be any first responders involved), and discussion of availability therefore focuses on the existence and ease of purchase of required drugs and supportive equipment.

     The discussions and the respective tables that follow permit a detailed analysis of these chemicals and biological agents. In a world of infinite resources almost all of the antidotal interventions being pursued for these potential agents would be of scientific merit. Many of these interventions might substantially advance scientific thought, and most could play a limited role in improving care, but all will confront the problem of Investigational New Drug (IND) status and FDA approval in the face of a very low natural incidence and ethical barriers to controlled testing in human subjects. These investigations will be exceptionally expensive and it is not apparent that the commercial pharmaceutical industry would consider research in this domain profitable without military or governmental support. In our world of finite resources a more pragmatic approach has been chosen for suggestions for research and development of antidotal agents. In particular, interventions that have a demonstrated benefit and might be improved upon are favored over novel approaches that have yet to be shown efficacious in human patients.

     In addition, research and development recommendations are based on the premise that the most valuable treatments will be those that will be (or might be) useful even if a biological or chemical assault does not occur.

     A third consideration in making recommendations was based on the committee's view that prophylactic interventions will rarely, if ever, be appropriate. The decision as to whether prophylactic therapy is appropriate for any of these biological or chemical agents must be based on several issues: risk to personnel, potential benefit for the individual and society, and extent of societal expenditure. The committee's view is that these considerations preclude any prophylactic interventions for the entire population, at least for the biological or chemical agents under consideration in this report. Certain prehospital first responders might be considered for prophylaxis against specific biological or chemical agents, but the scientific evidence in favor of prophylaxis of this smaller but still very large population is limited, and the risks and expenditures would still be substantial. In making this recommendation the committee's focus is purely civilian, and it should not be construed as discouraging the development of prophylactic interventions for use by the armed forces. The differences with regard to military and civilian prophylaxis strategies are substantial, encompassing not only the simple contrast of known threats at known times for military forces as opposed to unknown threats at unknown times for the civilian population, but also the levels of organization and systemic preparedness required and available.

     Two general and important conclusions will become obvious to the reader as he or she proceeds through this chapter. The first is that with a few exceptions, drugs, antitoxins, and supportive medical equipment are generally available in small quantities (although two recent surveys [Dart et al., 1996; Skolfield, 1997] by poison control centers report that very few hospitals in their service areas carried sufficient amounts of all recommended antidotes). Proper planning and coordination among area medical and veterinary facilities might yield sufficient quantities of these drugs and other supplies for a multiple-victim incident, but few locales will have adequate supplies for a true mass-casualty event.

     The second general conclusion is that many of the vaccines and therapeutics described below are only available under Investigational New Drug (IND) applications to the Food and Drug Administration (FDA). Such products are generally produced in limited amounts and can be used only in a research setting and with the informed consent of the recipient (i.e., the patient or a proxy must provide informed consent, and the FDA must be contacted for an IND number for the patient before the manufacturer can provide the product). In some cases, a fully licensed FDA-approved product will emerge after the requisite evidence of safety and efficacy is accumulated. In the interim however, under current legal requirements, IND status will effectively preclude use in a mass-casualty situation. Furthermore, it will be difficult or impossible to collect the required evidence of efficacy for many INDs (randomized clinical trials in human patients), either because the disease is so rare that accumulating enough cases will take a very long time, or because the condition against which it is directed does not occur naturally (e.g., mustard poisoning). Earlier this year, FDA established rules making it easier to study investigational drugs and devices with patients in life-threatening situations and unable to give informed consent. However, these rules, which require extensive prior planning, are aimed at facilitating collection of efficacy data and do not directly address the mass-casualty situation, especially for terrorist acts involving chemical and biological agents.

     FDA recognized the difficulty IND status presented in potential mass-casualty situations during the Persian Gulf War and passed an interim rule waiving the requirement for the United States military to obtain informed consent in using two investigational products intended to provide protection against chemical and biological warfare agents (pyridostigmine bromide and botulinum toxoid vaccine). The FDA has recently solicited comments on the wisdom of revoking this interim rule as well as on the nature of the evidence that ought to be required when products cannot ethically be tested in humans (United States Food and Drug Administration, 1997).

     Discussion of chemical agents is based upon an approach that integrates local, state, and federal systems for the delivery and stockpiling of antidotes for mass casualty events. This approach emphasizes which agents must be available locally, how much and under whose jurisdiction. The principle being that a plan should be developed to deliver large quantities of antidotes to any part of our country in a rapid organized fashion. Research and development should focus on models of storage and methods for deployment and delivery in a timely fashion. First responders from Emergency Medical Services and Hazmat Services cannot be expected to make definitive decisions and in general will not be stocked for population antidotal care, although they should have access to personal antidotal material for high-risk toxins so as to effectively complete scene assessment and victim rescue.

     The treatment for nerve agent poisoning recommended by the U.S. military involves the use of three therapeutic drugs: atropine, pralidoxime, and diazepam. Nerve agents act by binding to the enzyme acetylcholinesterase, thereby blocking its normal function of breaking down the neurotransmitter acetylcholine following its release at neuronal synapses and neuromuscular junctions throughout the peripheral and central nervous systems. Acetylcholine accumulates and overstimulates synapses throughout the brain, nervous system, glands, and skeletal and smooth muscles. Death is usually caused by respiratory failure resulting from paralysis of the diaphragm and intercostal muscles, depression of the brain respiratory center, bronchospasm, and excessive bronchial secretions. Seizure activity also contributes to morbidity and mortality.

     Atropine sulfate is a drug that blocks muscarinic acetylcholine receptors, counteracting effects such as vomiting and diarrhea, excessive salivation and bronchial secretions, sweating, and bronchospasm. It is administered intravenously, if possible, in high doses at frequent intervals until signs of intoxication diminish. Pralidoxime chloride (2-PAM), a drug that reactivates the nerve agent-inhibited cholinesterase, is administered along with atropine. Diazepam, or another anticonvulsant, may be administered in severe cases to control seizures and thereby prevent seizure-induced brain damage.

     Appropriate adult doses of atropine sulfate, 2-PAM, and diazepam are packaged in autoinjectors issued to U.S. military personnel for self- or buddy-aid. A metered dose atropine methonitrate inhaler called the medical aerosolized nerve agent antidote (MANAA) has been approved by FDA and is being produced for DoD by 3M/Riker. However, it is intended for use, under medical supervision, as a supplement to injectable atropine, not as self/buddy aid. Except under special circumstances, utilization of these prepackaged autoinjectors should be limited to Hazmat and prehospital EMS staff for their own personal care and that of their coworkers. Consideration for use of these antidotes for the general public should be restricted to exceptional circumstances when patients cannot be expeditiously removed from the environment, decontaminated, and brought to an emergency department. None of these antidotes is ideally delivered intramuscularly, in the absence of intravenous fluids and control of the airway, or during a convulsion. If these are considered essential products for civilian care, the hospital emergency department is the ideal site for their use.

     In a nerve agent incident where a presumed exposed patient is to be decontaminated prior to transportation to an emergency department, it can be considered appropriate for prehospital medical personnel to utilize prepackaged antidotes (atropine sulfate, diazepam and pralidoxime chloride) if and only if:

1. There are signs and symptoms indicative of nerve agent poisoning, namely, meiosis, rhinorrhea, shortness of breath, fasciculations, or seizures.

2. There is an initial intelligence basis for suspecting the presence of a nerve agent at the scene or a high quality detection system that indicates the presence of a nerve agent at the scene.

3. A qualified physician with skills in medical toxicology is actively involved in the management of the patient.

4. The antidotes are utilized before or during decontamination and in no way delay transfer to a health care facility or casualty collection point.

     If transfer to a health care facility subsequent to decontamination will exceed 30 minutes, it may be appropriate to treat additional civilians at the scene. The committee is aware of no studies performed comparing central nervous system levels and benefits achieved by intravenous administration of these antidotes with those achieved by intramuscular injection performed 15—45 minutes earlier. Such a comparison would be an important consideration in deciding upon expedient prehospital treatment.

     An alternative to extensive field treatment by Hazmat, EMS, and MMST teams in a particular region might utilize Hazmat and MMST teams as a mobile stockpile system delivering large quantities of antidotes to the EMS teams/ambulances (and individual hospitals as patients move there). This approach will ensure that patient load at a given hospital will be matched by antidote supply, thus expediting therapy and avoiding delays in delivery from a single central stockpile. Decisions on antidote stockpiling and control will involve geographic (rural vs. urban), financial, and other legitimate but nonscientific determinations, but in the proposed procedure, first responders would draw on established supplies of antidotes prepared for disaster management to ensure that patients transported to local emergency departments arrived with sufficient antidotes to begin treatment. Simultaneous communication with Regional Poison Control Centers and Poison Control Center—Emergency Department linkages to local and state health departments would track stockpile usage and allow for coordination with more distant sources, such as the Centers for Disease Control and Prevention.

     In nonhuman primate studies, the combination of atropine and 2-PAM will protect against up to five times the LD₅₀ (the dose lethal to 50 percent of the population exposed)¹ of all known nerve agents except soman (GD). Soman is an exception, because 2-PAM acts by competitively binding to the organophosphate agent and thereby "reactivating" the acetylcholinesterase enzyme the agent had tied up. However, once the enzyme-agent complex has undergone an irreversible "aging" process, 2-PAM is unable to reactivate the enzyme. The aging process takes hours for VX and most of the G agents, but only minutes in the case of soman (GD). In most cases of domestic civilian terrorism involving soman intoxication, it will not be possible to administer 2-PAM this quickly. Additional limitations in the use of 2-PAM as an antidote in nerve agent toxicity include the fact that large doses may be necessary for protection and survival, but in such large doses 2-PAM itself can lead to significant side effects, most notably hypertension. In addition, because it does not readily cross the blood-brain barrier, 2-PAM is thought to have little action against the central nervous system effects of nerve-agent poisoning.

     Although 2-PAM and atropine sulfate have only limited efficacy against soman (GD), nonhuman primates given the peripherally acting carbamate pyridostigmine prior to exposure to the nerve agent and atropine sulfate and 2-PAM after exposure survived GD in doses up to 20 to 40 times the LD₅₀. Pyridostigmine appears to be without comparable benefit in treatment of sarin or VX, however. Like the nerve agents, carba-mates inhibit the enzymatic activity of acetylcholinesterase. In fact, carba-mate-enzyme binding precludes organophosphate-enzyme binding. Unlike the nerve agents, however, the carbamate-enzyme bond is freely and spontaneously reversible. As a result, it is possible to protect acetylcholinesterase from irreversible inhibition by nerve agent by use of the reversible carbamate inhibitor. The use of pyridostigmine by large numbers of military personnel for periods of 6—7 days during the Gulf War resulted in uncomfortable but not disabling side effects (primarily gastrointestinal and urinary) in more than half of those taking the drug (Dunn et al., 1997). In most cases these effects subsided after a day or two. Numerous controlled studies in humans, as well as years of use in the treatment of myasthenia gravis, support claims for the safety of pyridos-tigmine.

     The utilization of prepackaged diazepam for intramuscular use is a poor parenteral therapeutic delivery technique for this anticonvulsant. The diazepam is dissolved in propylene glycol and is poorly and erratically absorbed following intramuscular use. Although the intramuscular route is considered to be the least effective route for seizure control, lorazepam can be used intramuscularly and could be preferred to diazepam for EMS and Hazmat use. Lorazepam, however, has several disadvantages. From a financial perspective it is more expensive than diazepam. Lorazepam is not stable at high temperatures and therefore cannot be as easily stored as diazepam. Finally, without preloaded syringes or autoinjector packaging, intramuscular use will be difficult to accomplish efficiently while utilizing the protective clothing required at the scene.

     Organophosphate (OP) pesticides are widely used throughout the United States, and poisoning is common (Litovitz et al., 1997). Treatment is identical to that for nerve agents, and as a result, many emergency medical teams and most hospital emergency department staff have some familiarity with diagnosis and treatment of OP poisoning and have access to limited supplies of atropine and pralidoxime. However, multiple nerve agent victims may each need 10—50 milligrams (mg) of atropine sulfate, which would rapidly deplete supplies in receiving hospitals. Rural communities may be able to call on veterinarians, who sometimes hold substantial amounts of atropine to treat cattle or horses poisoned by organophosphate pesticides. They might also be sources of other drugs, resuscitation equipment. disinfectants, and other useful equipment and supplies (Schneider, 1987). The same general concern--treatment would be possible only for small numbers of patients--is also true with regard to availability of ventilators. As in many other disaster situations, intubated patients can be supported by bag valve mask ventilation until a ventilator is available. Bronchoconstriction and copious secretions are prominent effects of organophosphate poisoning, and therefore ventilation is likely to be required for up to several hours after exposure, even when appropriate drug therapy is available.

     Table 8-1 provides information on a number of treatments and prophylactic pretreatments in various stages of research and development. This table and those that follow contain the relativistic term "potential civilian utility" and employ a very liberal criterion in assessing products for such use. The accompanying text evaluates potential products in a more selective manner that emphasizes probability and priority. For example, various pralidoxime derivatives, such as Pro-2-PAM, P-2-S and the Hagedorn oximes such as HI-6, have been compared to 2-PAM. Although some of these products offer increases in efficacy under some circumstances, none are FDA approved and most have intrinsic formulation and stability problems. The committee recommends that no further investment be made in attempting to bring these or similar compounds to market and/or to establish stockpiles The potential cost appears far more substantial than the advantage they might provide over 2-PAM.

     Alternatives to atropine sulfate autoinjectors, such as the quaterary ammonium derivatives ipratropium bromide and atropine methonitrate, have the disadvantage of poor absorption across mucosae and the blood-brain barrier, resulting in prolonged local effects, but they have negligible systemic effects. Further comparison of inhaled atropine methonitrate, scopolamine, and ipratropium bromide with intramuscular atropine is indicated.

     Of the diverse agents with potential as catalytic and stoichiometric scavengers, most remain in the early research stages and, as is the case with other pretreatments, their potential utility in managing the consequences of a domestic civilian terrorist incident involving nerve agent is not clear. Further development of human butyrlcholinesterases by the DoD could provide a potential pretreatment for Hazmat and prehospital staff performing rescue in unsafe environments. The relationship between effectiveness and time of delivery relative to nerve agent exposure is essential in evaluating that possibility. Only exceptional intelligence and information-sharing will provide these first responders with the time likely to be necessary for effective use of these scavengers or any pretreatments requiring substantial lead times.

8-1 Atropine sulfate, pralidoxime, and diazepam autoinjectors and stockpiles of these drugs should be available both for onsite self/buddy use by emergency medical personnel and for delivery to hospitals or patient collection points with patients. In addition atropine, pralidoxime and diazepam should be readily available in large quantities from a stockpile controlled by the local health department to be brought to the site where EMS will bring the casualties. Studies of stockpile control and time necessary for the delivery to prehospital, hospital and health departments should be performed for each region. Specifically, a study should be designed to describe the most effective distribution system for a mass casualty event. These studies must emphasize an integrated analysis based on the potential of regional health, police, and fire personnel.

8-2 A comparison of the central nervous system levels and benefits achieved by intravenous administration of atropine and 2-PAM with those achieved by intramuscular injection performed 15—45 minutes earlier would provide important help in deciding upon the criteria for and amount of prehospital treatment to recommend.

8-3 Lorazepam should be investigated as an intramuscular anticonvulsant for use in the field by EMS and Hazmat personnel. A study of its stability at room temperature will be essential for its use in the field.

8-4 Needle, intravenous, and/or intravenous bag adapters should be designed to facilitate the intravenous delivery of antidotes currently prepared solely for intramuscular use. The prepackaged systems designed for intramuscular use can only be used intravenously with risk to the health professional giving the injections.

8-5 Further comparison of inhaled atropine methonitrate, scopolamine, and ipratropium bromide with intramuscular atropine is indicated.

8-6 New more effective anticonvulsants are needed for autoinjector applications. Anticonvulsants that are water soluble and effective in halting nerve agent-induced seizure activity as well as preventing recurrence of the seizure activity will improve the recovery of seriously poisoned casualties.

8-7 The committee recommends support of continuing research to develop catalytic scavenger molecules such as human butyrylcholinesterase and carboxyesterase as both potential pretreatments against anticholinesterases and as immediate postexposure therapies.

8-8 Research into the development of catalytic monoclonal antibodies against a broad spectrum of nerve agents may prove beneficial in the development of rapid diagnostic tests as well as in the development of potential new therapies.

     Included in this category of chemical agents are various forms of "mustard," an arsenical compound called Lewisite, and phosgene oxime. No evidence suggests that Lewisite or phosgene oxime has ever been used on the battlefield, but sulfur mustard (bis [2-chloroethyl] sulfide) has been used in several wars, most recently in the Iran-Iraq conflict, and it is considered the most likely to be used on the battlefield. An immediate precursor, thioglycol, has many industrial uses and is available commercially. A simple substitution reaction yields mustard. Sidell et al. (1997) is the primary source of the information presented in this section.

     The name mustard apparently stems from the compound's smell, taste, and color rather than any chemical resemblance to the popular spice. At room temperature sulfur mustard is an oily liquid that is only slightly soluble in water and therefore very persistent in the environment. At higher temperatures it becomes a significant vapor hazard ("mustard gas"). It quickly permeates rubber and is readily absorbed by skin. eyes, airway, and gastrointestinal (GI) tract. It reacts within minutes with components of DNA, RNA, and proteins, severely compromising normal cell function. Acute local effects can be severe enough to require days to weeks of care, but mortality, usually from pulmonary insufficiency or superimposed infection, is low. No effective treatment of mustard damaged tissue is currently available. Immediate decontamination of exposed skin areas is the only means of preventing tissue damage from mustard. U.S. military publications recommend 0.5 percent sodium hypochlorite followed by soap and water, or the resin-based M291 and M295 decontamination kits. This task is made more difficult by the fact that clinical signs, including pain, are not evident for 2—12 hours, depending on the dose and tissue exposed. Eyes may be flushed with copious amounts of water. Skin, eye, and airway damage is treated similarly to thermal burns, and pain relief is provided by topical or systemic analgesics (Willems, 1991). Early intubation and oxygen therapy are recommended for patients with signs of airway damage.

     Lewisite (ß-chlorovinyldichloroarsine) was synthesized in 1918 for use as a weapon, and its clinical effects are similar to those of mustard in many respects, although the cellular mechanisms are believed to differ. However, unlike mustard, Lewisite liquid or vapor produces irritation and pain seconds to minutes after contact. Immediate decontamination may limit damage to skin or eyes, and intramuscular injections of a specific antidote, dimercaprol, or British antiLewisite (BAL) will reduce the severity of systemic effects. BAL has toxic effects of its own, however, and must be used with care.

     Phosgene oxime (dichloroformoxime) is a colorless crystalline solid with a melting point of approximately 37.7°C (100°F). In liquid or vapor form it is highly corrosive, and it penetrates clothing and rubber readily. The mechanism by which it damages tissue is unknown, but its effects are almost instantaneous and produce severe pain. Skin lesions are like those caused by a strong acid. There is no antidote; treatment will be similar to that for mustard.

     Table 8-2 provides information about ongoing work on potential countermeasures in various stages of development. All the entries are drawn from current work by DoD labs, primarily USAMRICD. Considerable basic research is devoted to better delineation of the mechanism(s) of action in order to develop protective and ameliorative interventions. Strategy to date has involved parallel investigation of intracellular scavengers, cell cycle inhibitors, calcium modulators, protease inhibitors, and antiinflammatory drugs. The rapid action of the vesicants, the lack of immediate pain in the case of mustard, and the attractiveness of an attack employing the vapor (ergo, pulmonary) route make decontamination an unsatisfactory strategy for civilians, and, as is the case with most of the agents being considered, pretreatments requiring any more than a few minutes lead time are not likely to be generally useful in coping with civilian terrorism incidents. Topically applied skin protectants offer the possibility of protection from trace amounts of agent-penetrating protective garments or surviving decontamination of equipment, but certainly could not be counted on to replace chemical-resistant clothing in areas known to be contaminated. The current research aimed at moderating or repairing vesicant injury is therefore extremely important, despite, or perhaps due to, the fact that most of the candidate drugs are still years away from licensing. Clinical testing of efficacy in humans is not possible, so early agreement with FDA on surrogate measures will be critical.

TABLE 8-2
Potential Additional Countermeasures for Vesicant Agent Poisoning

Antidote	Efficacy	Availability	Potential Civilian Utility	Stockpile
Topical skin protectant (TSP)	Passive protection	Goal is FDA license by FY00	Prehospital high-risk personnel	Health Dept.
Reactive TSP (decontaminates)	Proof of principle	Goal is FDA license by FY08	Prehospital high-risk personnel	Health Dept.
Nitric oxide synthase inhibitora Nitroarginine methylester (NAME) effective at high concentrations	Cell culture vs. sulphur mustard	Preclinical	Insufficient evidence	N/A
Combinations of dexamethasone, heparin promethazine, vitamin E, sodium thiosulfateb	Preliminary evidence in rats	FDA-approved drugs	Insufficient evidence	N/A
Sodium thiosulphate (i.v.)	Animal data shows effects up to 20 min post exposure	FDA-approved drug	Insufficient evidence	N/A
N-acetyl Cysteine	Protection from vapors in rats	Preclinical	Insufficient evidence	N/A
Calcium chelators	Protects skin cells in culture	Preclinical	Insufficient evidence	N/A
CO2 laser debridement	Mustard ion weanling pigskin	Preclinical	Insufficient evidence	N/A

^aSawyer et al., 1996; ^bVojvodic et al., 1985.

8-9 The lack of treatments for vesicant injury constitutes a serious shortfall in civilian medical preparedness, and the existing program of research into mechanism should be supplemented by an aggressive screening program focused on repairing or limiting vesicant injuries, especially airway injuries.

     The cyanide anion, CN^—, whether delivered in hydrocyanic acid or in a cyanogen such as cyanogen chloride, exerts its toxicity primarily by inhibiting mitochondrial cytochrome oxidase, which leads to lactic acidosis, cytotoxic hypoxia, seizures, dysrhythmias, respiratory failure, and death within minutes after inhalation or oral ingestion of a large dose (1 to 3 mg/kg of body weight). One antidote for cyanide poisoning is amyl nitrite, which converts hemoglobin to methemoglobin, which in turn competes effectively for cyanide with the mitochondrial cytochrome oxidase complex. Intravenous sodium nitrite is generally used for this purpose after an initial dose of the volatile amyl nitrite is given by inhalation. Cyanide is then removed from cyanomethemoglobin by intravenous sodium thiosulfate, which reacts with cyanide to form nontoxic thiocyanate. Gastric lavage with activated charcoal should be administered if cyanide is ingested. Supportive therapy includes intubation, correcting acidosis, and, if necessary, administering anticonvulsants. Cyanide is metabolized more readily than the other chemical agents, and as a result, if the initial dose is not so large as to kill the victim within minutes, supportive therapy may be sufficient for full recovery in a matter of hours.

     Amyl nitrite, sodium nitrite, and sodium thiosulfate are commercially available in standard doses in the Pasadena Cyanide Antidote Kit (formerly the Eli Lilly Cyanide Antidote Kit). Many poison control centers and emergency departments may have small quantities of such kits on hand. As in the case of nerve agents, a mass-casualty situation will quickly exhaust supplies. Pooling resources from the whole community could be beneficial, but only if communications and a mechanism for sharing pre-incident intelligence are already in place. As suggested above, the use of prehospital medical personnel to ensure that antidotes are delivered to the hospitals and/or casualty collection points with the affected patients is a potential means of matching antidote supply with patient load. As in the case of nerve agent incidents, the committee recommends that emergency medical teams responding to an event bring a substantial number of these antidote kits and move the kits with the patients to assure adequate therapy is available on emergency-department arrival. Only if substantial delay is expected due to unavailability of a local ED or if cyanide has been conclusively identified as the toxic substance involved should consideration be given to prehospital use of this antidote kit. The use of the cyanide antidote kit by prehospital personnel under uncertain circumstances would place civilians at risk for substantial delay in transport to a health care facility. This antidote kit is a three-part intervention requiring experienced clinical judgment prior to use and demanding intravenous access for the major components of therapy. Use of the kit demands substantial time, particularly for a heterogeneous population including children, the aged, and medically compromised individuals.

     Additional interventions not yet licensed or available for use are summarized in Table 8-3. Although a number of entries in the table have shown great promise in preclinical studies, pretreatment is not a viable option in the most probable civilian terrorism scenarios. The 8-amino-quiniline compound WR242511, for example, is reported to protect mice from a dose of cyanide 5 times the LD_50, but it must be administered 8 hours before cyanide exposure. The initial three entries therefore rate a higher priority than the remainder, which may eventually provide a measure of protection for military troops, cyanide industry workers, or first responders in the vicinity of large stores of cyanide, but will not be effective post exposure. Three postexposure treatments are currently approved drugs in Europe and might be licensed in the United States without great delay if there were a perceived market for them. The advantage of these newer drugs is their diminished acute risk/benefit in children and the physically compromised patients. On the other hand, the major difficulty with using them for current treatment protocol is the inability to treat the victim within minutes of exposure, a problem that will not be remedied by new drugs. In general, if the patient has survived to the time of arrival of clinical support, the probability of survival is great.

Table 8-3
Potential Additional Antidotes for Cyanide

Antidote	Efficacy	Availability	Potential Civilian Utility	Stockpile
Hydroxocobalamin (Vitamin B12a)a—e (+NaHS04) Kit	No methemoglobin Low toxicity High CN affinity	Short shelf life (France) Orphan drug	Post exposure	Health dept. emergency dept.
Dicobalt ethylene diamine tetraacetic acid (EDTA) (Kelocyanor)f	IV Risk: Cardiac Dysrhythmias angina, death	Europe: commercial USA: Experimental	Post exposure	N/A
4-Dimethylaminophenol (4-DMAP)g,h and similar molecules P-aminopropiophenone (PAPP),i P-aminoheptanophenone (PAHP), P-aminooctanoylphenone (PAOP)j	IV, IM Possible Mutagen Local tissue necrosis Marked methemoglobin Temperature , pain PAHP (safest?)	Germany	Post exposure	N/A
Stroma free methemoglobink—m	Experimental	No	Postexposure and Prehospital high-risk personnel	Health dept. Poison center
Superactivated charcoaln,o	For oral exposure	FDA approved	Postexposure and Prehospital high-risk personnel	Emergency dept.
-adrenergic antagonists (chlorpromazine;p phenoxybenzamine)	Mechanisms uncertain	FDA-approved drugs		N/A
8-aminoquinoline analogs of primaquineq (e.g., WR242511)	Methemoglobin formers Pretreatment	Varies with compound	Prehospital high-risk personnel	N/A
Alpha-ketoglutaric acidr—u	Direct binding of cyanide without methemoglobin formation Animal studies only	Experimental	Insufficient evidence	N/A

^aHall and Rumack, 1987; ^bCottrell et al., 1978; ^cBrouard et al., 1987; ^dBismuth et al., 1988; ^eBeregri et al., 1991; ^fHillman et al., 1974; ^gWeger, 1983; ^hBhattacharya, 1995; ⁱMarrs and Bright, 1987; ^jRockwood et al., 1992; ^kTen Eyck et al., 1983; ^lTen Eyck et al., 1986; ^mBreen et al., 1996; ⁿAndersen, 1946; ^oLambert et al., 1988; ^pPeterson and Cohen 1985; ^qSteinhaus et al., 1990; ^rDulaney et al., 1991; ^sBhattacharya and Vijayaraghavan, 1991; ^tHume et al., 1995; ^uNorris et al., 1990.

8-10 The organization of delivery and availability of adequate supplies of the cyanide antidote kit must be achieved. Studies of stockpile control and time necessary for the delivery to prehospital, hospital, and health departments should be performed for each region. Specifically, a study should be designed to describe the most effective response system for a mass-casualty event. These studies must emphasize an integrated analysis based on the potential of regional health, police, and fire personnel.

8-11 Further understanding of the risks and benefits of methemoglobin forming agents should be investigated.

8-12 A continued investigation into the benefits of hydroxocobalamin and stroma-free methemoglobin would be valuable and is an appropriate avenue of investigation.

8-13 Dicobalt ethylene diamine tetraacetic acid and the strong methemoglobin forming compounds 4-dimethylaminophenol and various aminophenones merit further investigation, but must be given a lower priority then hydroxocobalamin and stroma-free methemoglobin, which carry less risk of creating excessive and unpredictable levels of methemoglobin.

     Although phosgene is not currently believed to be a significant threat as a military weapon, it was used in World War I artillery shells, and, more importantly to the present discussion, it is still a widely used industrial chemical, over a billion pounds of which are produced in the United States. It is generally stored and transported as a liquid, but its low boiling point (7.5°C) means that it readily becomes a heavier-than-air gas. Pulmonary edema is the most serious consequence of inhalation--onset within 2 to 6 hr. is indicative of severe injury. Low concentrations may produce mild coughing, dyspnea, and a feeling of discomfort in the chest, although individuals may remain essentially asymptomatic for up to 72 hours. Physical exertion during this period may precipitate signs and symptoms. Rest is thus an essential component of patient management, along with airway management (control of secretions and bronchospasm) and oxygen therapy. There are no proven pharmacological interventions for pulmonary phosgene at present, so, as with the vesicants, rapid removal from the source and thorough decontamination is essential. Table 8-4 provides information on other treatments that have been or are being investigated.

Table 8-4
Potential Antidotes for Pulmonary Phosgene

Antidote	Efficacy	Availability	Potential Civilian Utility	Stockpile
Hexamethylene tetraminea (methenamine; urotropin, HMT)	Limited prophylactic effect; no convincing benefit after exposure	Yes	None	N/A
Cysteineb,c	Traps phosgene and converts to less harmful metabolites.	Yes	Insufficient evidence	Hospital emergency dept.
N-acetylcysteined	Evidence in rabbits only	Yes	Prehospital? Pre- and post-exposure	Hospital emergency dept.
Corticosteroidse	To decrease inflammation Limited evidence	Yes	Yes	Hospital emergency dept.
Aminophyllinef	Preexposure and postexposure Attenuates lipid peroxidation	Yes	Yes	Hospital emergency dept.

^aDiller, 1980; ^bBridgeman et al., 1991; ^cLailey et al., 1991; ^dScuito et al., 1995; ^eLorin and Kulling, 1986; ^fSciuto et al., 1997.

8-14 Protection of the pulmonary bronchi and bronchioles may be possible by the use of cytoprotective agents. Ongoing studies of pneumocytes, interstitial, and epithelial cells suggest that antiinflammatory agents, such as aminophylline, corticosteroids, and ibuprofen, may be useful. Further studies on the ability of N-acetylcysteine to limit the inflammatory cascade produced by effects of phosgene and its metabolic byproducts are also justified, as are additional studies of its systemic antioxidant effects.

     The following section on biological agents begins with a review of vaccine research followed by reviews of bacterial infections, rickettsia, viruses, and toxins that include agent specific R&D needs. The section concludes with nonspecific defenses against biological agents that show promise for the future.

     Vaccines are the cheapest and most effective defense against a large number of infectious diseases. Public health vaccine programs are the principal means of providing protection to at-risk populations against a growing list of natural infectious disease hazards. As vaccine technology continues to dramatically improve and new vaccines are developed and licensed, the list of vaccine-preventable diseases is increasing. Laboratory workers are provided protection against several highly hazardous bacteria and viruses which are considered to be potential biologic weapons through the use of vaccines in IND status under open protocols. Military populations can now be protected against the hostile use of several biologic threats by vaccination. The armed services have two licensed vaccines suitable for routine use if needed and have several more vaccines under development for protection of military populations against biologic threat agents (see Table 8-5).

TABLE 8-5
Vaccines Against Biologic Threat Agents

Agent or Disease	Stage of Development	Type	Civilian Utility
Anthrax	licensed	inactivated toxins	yes
Plague	licensed	inactivated bacteria	no
Tularemia	IND	live attenuated	no
EEE	IND	inactivated virus	no
WEE	IND	inactivated virus	no
VEE	IND	live attenuated	no
VEE	IND	inactivated virus	no
Botulism	IND	toxoids	yes
Q Fever	IND	inactive antigen	no
Smallpox	licensed	avirulent vaccinia virus	yes
Smallpox	IND	avirulent vaccinia virus	yes
Ebola/Marburg	preclinical	viral replicon/DNA	no
SEB	preclinical	toxoid	no
Ricin	preclinical	toxoid	no
Brucellosis	preclinical		no
Yellow Fever	licensed	live attenuated virus	no
Rift Valley fever	IND	inactivated virus	no
Rift Valley fever	IND	live attenuated virus	no
Junin virus	IND	live attenuated virus	no
Hantaan virus	IND	engineered vaccinia	no
Dengue	IND	live attenuated virus	no

     Vaccination has limited value as a primary defense for civilian populations for several compelling reasons. The risk of exposure to a biologic threat agents is very low and uncertain for the general population. Pre-exposure vaccination of an entire population is a huge and daunting task. Achieving a high level of vaccine coverage of the U.S. adult population has never been done and probably could not be done in the absence of an eminent and credible threat. The costs and risks of vaccination are far too great and far outweigh potential benefits in view of the current assessment of the potential threat. Finally, there is a long list of potential agents and only a few licensed vaccines. The spectrum of achievable protection at present includes only smallpox, anthrax, and plague.

     Vaccines against biologic threat agents do, however, have some very important uses in the civilian response to the threat of biologic terrorism. Anthrax vaccine can be effectively used in conjunction with antibiotics to prevent the development of pulmonary anthrax in exposed individuals. Botulinum toxoid vaccines can be used to immunize plasma donors to produce specific immune globulins for therapy of botulism. Smallpox vaccine (vaccinia) would be essential to prevent further spread of the disease following diagnosis of the originally exposed individuals. Vaccines against several of the threat agents may be used to immunize personnel in public health and research laboratories who must work with live agents. Several vaccines in IND status have been used with FDA approval in at-risk laboratory personnel for many years with appropriate informed consent and a suitable open protocol. Consideration might also be given to immunization of response team personnel and selected medical personnel in high-risk areas such as autopsy suites and microbiology laboratories. Consideration should also be given to immunization of selected law enforcement and intelligence personnel in high-risk assignments such as the White House or one of the national-level rapid response teams. In both of these cases, high risk will be a highly subjective judgment, given the lead time of weeks to months required for immunization to achieve its full effect.

     The value of vaccines, even in the limited uses described above, justifies an accelerated research effort to improve the licensed vaccines, such as the anthrax, plague, and smallpox vaccines, and to complete the development process and seek licensure for those vaccines still in IND status. The DoD has an ongoing research program aimed at improving or completing the development of the vaccines on the above list. A major DoD procurement effort, the Joint Vaccine Acquisition Program, has awarded a contract for the development and manufacture of many of the vaccines in Table 8-5. Completion times to fulfill the contract call for most vaccines on the list to be available by 2005.

     One reason that DoD has taken the lead in the development of these vaccines is that the diseases involved are not common in the developed countries, which severely limits the profitability of the vaccines. This is only partially true in the case of potential drug treatments: antiviral drugs have proven to be highly specific, like vaccines, so development of a drug to treat Ebola virus, for example, is unlikely to be financially attractive to the private sector. Antibiotics, on the other hand, are generally effective against a wide variety of microbial infections, greatly increasing the potential market of any new product. Drug industry spending on developing new antibiotics has been spurred in recent years by the alarming rise in resistance to even the best of current drugs. Information on specific products is tightly controlled for proprietary reasons, but the industry trade group, the Pharmaceutical Research and Manufacturers of America, reports that there are some 125 new antibiotics currently in some stage of development. It seems safe to say that few if any of these are likely to be tested for efficacy against any of the biological agents being considered in this report. Some may nevertheless prove highly effective, and a modest program to screen new antiviral and antimicrobial drugs for activity against biological warfare agents would certainly be a worthwhile R&D investment.

     The remainder of this section describes current and potential countermeasures to each of the biological agents on our list and concludes with a description and evaluation of some DARPA-sponsored research into generic, or at least multiagent, countermeasures.

     Anthrax is primarily a disease of herbivorous animals, domesticated as well as wild, and humans usually become infected by contact with infected sheep, goats, cattle, pigs, or horses (or contaminated products, for example, wool). The causative agent is Bacillus anthracis, a bacterium that forms inert spores when exposed to oxygen. These spores are extremely hardy and may survive outside a living host for years. Infections begin when spores are inhaled, ingested, or enter the body through a skin wound. Germination then occurs and bacteria proliferate. Cutaneous infections produce ulceration at the site, along with fever, malaise, and headache, but mortality is very low with antibiotic treatment. Gastrointestinal infection also begins with fever, malaise, and headache; severe abdominal pain follows, and mortality may be as high as 50 percent. Although military biological weapons programs and speculation about bioterrorism have focused on inhalational infection, naturally occurring cases of inhalation anthrax are rare. In these cases, the initial nonspecific symptoms have been followed by increasingly severe respiratory distress, cyanosis, and shock. Nearly 100 percent of such cases are fatal if left untreated. Meselson et al. (1994) provided extensive documentation of a major outbreak of inhalational anthrax in Sverdlovsk, USSR, in 1979.

     A licensed vaccine with demonstrated efficacy against cutaneous anthrax is available from Michigan Biological Products Institute. This vaccine is the formalin inactivated filtrate from culture of nonencapsulated B. anthracis. The principal antigen is Protective antigen (PA), although Lethal factor (LF) and Edema factor (EF) may also be involved in protective immunity. It is administered in six intramuscular doses at 0, 2, and 4 weeks, 6, 12, and 18 months, and affords continued protection if followed by annual boosters. Franz et al. (1997) note that there are few data regarding efficacy against inhalational anthrax in humans, although the vaccine has been shown to provide protection in studies using rhesus monkeys. Although the stockpile is not intended for civilian use, the Department of Defense has approximately seven million doses in cold storage, one million of which are bottled and ready for use (Danley, 1997). Since the release of our interim report, the Secretary of Defense has announced plans to vaccinate all U.S. military personnel. This decision will ensure continued U.S. production capability, but will almost certainly draw down the inventory substantially.

     Penicillin is the recommended treatment of inhalational anthrax, but tetracycline, erythromycin, and chloramphenicol have been used with success (Friedlander, 1997). A variety of other antibiotics have shown in-vitro activity, and current military doctrine calls for initiating treatment with oral ciprofloxacin or doxycycline as soon as exposure to anthrax spores is suspected and introducing intravenous ciprofloxacin at the earliest signs of infection or disease (Franz et al., 1997). It is essential to start antibiotic therapy before or very soon after such signs appear, if a high mortality rate is to be avoided. Other therapies for shock, volume deficit, and adequacy of airway may be necessary. The vaccination series should also be administered to victims not immunized in the previous 6 months. Antibiotic treatment should be continued for at least 4 weeks (i.e., until at least three doses of vaccine have been received). Penicillin and especially streptomycin are rarely used anymore, and hospital pharmacies will have very limited supplies on hand, but Pfizer will still ship streptomycin overnight. Ciprofloxacin and doxycycline are prescribed far more often, but they are expensive, especially ciprofloxacin, which may limit supplies in any one locale.

     The utility, indeed necessity, of anthrax vaccination subsequent to exposure is unique among the biological agents on our list. Anthrax vaccine is thus an exception to the view expressed above that vaccination has limited value as a bioweapon defense for civilians. However, the current vaccine, made by outdated technology, has several disadvantages. It is an impure mixture of bacterial products. Antigen content is variable from lot to lot due to the manufacturing process and the inability to precisely quantify antigenic components. Guinea pig potency assays are only semi-quantitative. The requirement for multiple doses is a serious limitation, especially if the vaccine is needed for use in response to exposure of a civilian population.

     The current state of knowledge on anthrax pathogenesis and studies of experimental anthrax vaccines indicate that a second-generation vaccine can be developed that could provide protection equal to, or better than, the current vaccine and would require fewer doses. A very effective two-dose vaccine is an achievable goal that should be aggressively pursued through a program that combines research and product development. A single-dose vaccine is a challenging goal that may or may not be achievable.

     Research is needed to define the optimal antigenic composition of a new vaccine. A vaccine based on purified protective antigen alone may meet the requirements, but there is a possibility that it will not be the optimal formulation. Including other antigenic components including lethal factor and edema factor and possibly others may enhance efficacy. New adjuvants or new formulations, such as microencapsulation, and alternative delivery systems, such as an oral delivery formulation, should be explored.

     Recent publications by Jackson et al. (1998) and Pomerantsev et al. (1997) have raised the question of whether certain strains of anthrax, either deliberately engineered or selected from nature, can overcome the protective immunity generated by a vaccine that is composed principally or entirely of protective antigen. Variation in virulence among anthrax strains and variation in relative resistance to vaccine-induced immunity has been observed in vaccination-challenge experiments in animals, but the basis for the variation is unclear. Antigenic variation in protective factor has been postulated but not demonstrated. The existence of additional virulence factors other than the two plasmid-encoded toxins and the poly-D glutamic acid capsule is a matter of conjecture. The preliminary findings of multiple anthrax strains in the Sverdlovsk anthrax victims by PCR and DNA analysis (Jackson et al., 1998) has raised questions regarding the spectrum of protective immunity provided by current vaccines.

     Recently published studies (Pomerantsev et al., 1997) have shown that insertion of the cereolysine AB gene from B. cereus into a virulent strain of B. anthracis enables the anthrax organism to overcome immunity induced by a live attenuated vaccine strain. Insertion of the cereolysine AB gene into the vaccine strain restores its ability to protect against the modified virulent organism. Questions raised by these studies should be experimentally addressed, within the legal and ethical constraints accepted by the U.S. in this area. Most importantly, will inactivated vaccine containing only protective factor provide protection in man against anthrax strains containing the cereolysine AB gene? Are there additional virulence factors in anthrax strains related to genes homologous with the cereolysine AB gene? Answers to these questions will help guide the design of a second generation vaccine.

     Licensure of a second generation vaccine in the absence of any possibility to conduct a formal efficacy trial will require additional studies of the pathogenetic mechanisms and the correlates of protective immunity.

     A second-generation, highly effective, and easy to administer anthrax vaccine would substantially improve the nation's ability to protect both civilian and military personnel against the number one biological threat.

8-15 A vigorous national effort is needed to develop, manufacture, and stockpile an improved anthrax vaccine. This will both benefit the armed forces and enhance the ability to protect the civilian population. The ongoing DoD effort should be supported and accelerated by a well-coordinated complementary DHHS program.

     Brucellosis is another disease of domesticated animals and usually occurs in humans as a result of ingestion of unpasteurized dairy products. Person-to-person transmission is very rare. The infectious agent is one of six species of the Brucella bacterium. Although nonsporulating, brucellae are aerobic organisms viable for long periods outside a host. Its ready transmission by the aerosol route led the United States to experiment with weaponizing Brucella during World War II, although the resulting bombs were never used. Fever, chills, and body aches occur in nearly all cases and regardless of route of infection. Brucellae disseminate widely and may cause disease in nearly any organ system, so additional signs and symptoms vary widely. Although rarely fatal, brucellosis can be debilitating for weeks or months if not treated. See Hoover and Friedlander (1997) for additional information.

     There is no approved Brucella vaccine for humans.

     According to Franz et al. (1997), patients should be treated with combinations of antibiotics because treatment with a single antibiotic causes poor response or relapse. Usually, a combination of doxycycline and rifampin is given orally for six weeks. Trimethoprim-sulfamethoxazole can be substituted for rifampin, although relapse rates may be as high as 30 percent (Franz et al., 1997). The recommended treatment for bone and joint infections, endocarditis, and central nervous system disease is streptomycin or another aminoglycoside, and therapy should be extended.

     All of the current R&D on brucellosis located by the committee focuses on development of a vaccine. As noted above, the committee considers it unlikely that a vaccine could be usefully employed for protection from a domestic terrorist attack and therefore considers such R&D a low priority for improving civilian medical capability. Antibiotic treatment, though not simple, is possible with current products. USAMRIID conducts assays of second- and third-generation antibiotics as they come on the market, using all of the bacterial threat agents in animal models.

     No action is required at this time.

     Plague is well known as the cause of the Black Death, which devastated the population of Europe in the fourteenth century. The infectious agent is Yersina pestis, a nonsporulating bacillus maintained in nature in fleas, most notably the rat flea. In humans the bite of an infected flea leads to a high fever, chills, and headache, often accompanied by nausea and vomiting. Six to eight hours later, very painful swelling of one or more lymph nodes (a bubo, hence bubonic plague) develops. Without treatment, septicemia will develop in 2 to 6 days, with a mortality rate of 33 percent. Inhalation of Y. pestis aerosol will lead to pneumonic plague (extensive, fulminant pneumonia with bloody sputum), which is almost always fatal if not treated within 24 hours of symptom onset. Patients in terminal stages of pneumonic or septicemic plague may develop large subcutaneous hemorrhages, which may have given rise to the name "Black Death." Additional information is available in McGovern and Friedlander (1997).

     A licensed, killed whole-cell vaccine is available. Although some epidemiologic evidence supports the efficacy of this vaccine against bubonic plague, its efficacy against aerosolized Y. pestis has not been established.

     Plague pneumonia is almost always fatal if treatment is not initiated within 24 hours of the onset of symptoms. Streptomycin is administered intramuscularly for 10 days (2 doses each day). Gentamicin can be substituted for streptomycin. Plague meningitis and cases of circulatory compromise are treated with chloramphenicol given intravenously. Intravenous doxycycline administered for 10 to 14 days is also effective.

     With the exception of four projects examining the mechanism of Y. pestis virulence factors, the very small amount of current R&D on plague located by the committee focuses on development of a second-generation vaccine. As noted above, the committee considers it unlikely that a vaccine could be usefully employed for protection from a domestic terrorist attack and therefore considers such R&D a low priority for improving civilian medical capability. Antibiotic treatment, though not simple, is possible with current products. USAMRIID conducts assays of 2nd- and 3rd-generation antibiotics as they come on the market, using all of the bacterial threat agents in animal models.

     No action is required at this time.

     Tularemia results from infection by the insect-borne bacterium Francisella tularensis. In North America, the tick is the principal reservoir, and the rabbit is the vertebrate most closely associated with transmission. As few as 10 organisms can give rise to a clinical infection in humans (Saslaw et al., 1961a, 1961b), and transmission may be via inhalation, in-gestion, or, most commonly, through breaks in the skin. The disease is characterized by fever, localized ulceration, enlarged lymph glands, and, in about 50 percent of patients, pneumonia. Without treatment with antibiotics, patients may have a prolonged illness with malaise, weakness, and weight loss persisting for months. Treatment with appropriate antibiotic drugs reduces the duration and severity of the disease, and overall mortality is quite low (1 to 2 percent).

     The United States Army Medical Research and Material Command is the IND holder for a live attenuated tularemia vaccine that appears to be effective against inhalational exposure.

     Streptomycin is administered intramuscularly in two divided doses daily for approximately 10 to 14 days. Gentamicin is also effective. Tetracycline and chloramphenicol are also effective but tend to be associated with significant relapse rates (Franz et al., 1997). See Evans and Friedlander (1997) for additional information.

     The committee could find no active U.S. research on tularemia. Given the possibility of effective treatment with current antibiotics and the recent increase in antibiotic development to counter resistance in many more common pathogens, tularemia research is not a high priority. USAMRIID conducts assays of 2nd and 3rd generation antibiotics as they come on the market, using all of the bacterial threat agents in animal models.

     No action is required at this time.

     Q fever is an incapacitating but rarely fatal disease caused by the rickettsia-like agent Coxiella burnetti. A large number of mammalian species can serve as host for C. burnetti, but humans are apparently the only hosts in which infection results in a disease. Although the organism cannot grow or replicate outside host cells, inhalation of a single organism can result in disease. The usual route of human infection is through contact with domestic livestock, but this may be very indirect contact, because the agent can assume a spore-like form that is extremely resistant to heat, desiccation, and many standard antiseptic treatments, allowing the organism to survive on inanimate surfaces for weeks or months. Human infection is usually the result of inhalation of infected aerosols, and signs and symptoms appear 10 to 40 days after exposure, sometimes abruptly and sometimes very gradually. There is no characteristic set of signs and symptoms, although fever and chills are nearly universal. Headache, fatigue, muscle aches, anorexia, and weight loss are common. Fatalities from Q fever are very rare, and although malaise and fatiguability may persist for months, most other effects last only 2 to 3 weeks. For additional information see Byrne (1997).

     Q fever vaccines in the United States are still investigational, although an effective vaccine, Q-Vax, is licensed in Australia.

     The most common treatments for Q fever are tetracyclines. Macrolide antibiotics, such as erythromycin and azithromycin, are also effective. Other agents used to treat Q fever include quinolones, chloramphenicol, and trimethoprim-sulfamethoxazole. Clinical experience with these drugs is limited. Treatment is most effective when administered during the 10- to 40-day incubation period.

     The committee could locate only two current U.S. studies of C. burnetti. Both of the NIH-funded grants are exploring genes and gene products thought to be involved in pathogenesis. Given the possibility of effective treatment with a wide selection of current antibiotics and the recent increase in antibiotic development to counter resistance in many more common pathogens, Q-fever research is not a high priority. USAMRIID conducts assays of 2nd- and 3rd-generation antibiotics as they come on the market, using all of the nonviral threat agents in animal models.

     No action is required at this time.

     Until very recently, smallpox was an important cause of morbidity and mortality in the developing world. The causative agent of smallpox is variola, one of a family of large, enveloped deoxyribonucleic acid (DNA) poxviruses. Unlike many of the agents discussed above, the variola virus thrives only in human hosts, and as a result, aggressive case finding and vaccination programs (using the closely related but nonpathogenic vaccinia virus) are thought to have eradicated smallpox. The last known cases occurred in 1978. Concerns about its use as a weapon persist, however, because variola virus is highly stable and retains its infectivity for long periods outside the host, and because enough is known of its sequencing that biotechnology might be used to create variola or a pathogenic variation of variola. Although characteristic pustular skin lesions provided the name for this disease, and virus can be recovered from scabs throughout convalescence, smallpox is infectious by aerosol as well. It is transmitted more easily than any of the other agents being considered in this report, and its use in a terrorist attack would pose the threat of a global epidemic. Regardless of route of transmission, clinical manifestations begin with fever, malaise, headache, and vomiting, and the infection is a systemic one that produced mortality rates of 20 to 30 percent in unvaccinated populations (McClain, 1997).

     Individuals who were vaccinated during the WHO smallpox eradication campaign in the 1970s were considered to have immunity to smallpox for at least 3 years, but protection diminishes over time. The only vaccine still available in the United States is a live vaccinia virus manufactured by Wyeth-Ayerst Laboratories (now Wyeth-Lederle Vaccines and Pediatrics, and no longer manufacturing the vaccinia vaccine). The CDC holds the entire remaining stock (approximately 6 million doses). Vaccination causes a pustule and local reaction on intradermal administration. Adverse reactions include encephalitis and other neurological disorders, generalized vaccinia, and vaccinia necrosum. Virus is transmissible to nonvaccinees and hazardous to individuals with eczema or immunosuppressive disorders.

     Vaccination will give protection to an exposed individual if it is administered within a few days of exposure, regardless of time since any prior vaccination. There is no chemotherapeutic agent with proven effectiveness against smallpox, but Franz et al. (1997) suggest that preclinical tests against other poxviruses indicate that chemotherapy with cidofovir might be useful (see below). Vaccinia-immune globulin (VIG) may also be of use if given within the first week following exposure (preferably within 24 hours). VIG, which is prepared from the blood of repeatedly vaccinated persons, is available from the CDC in extremely minute quantities. Because almost no one is being vaccinated anymore, there is little prospect of producing a large stockpile of VIG.

     Smallpox presents a unique risk among the possible biologic weapons in that the secondary contamination risk (person-to-person transmission) is significant. This is in distinction of virtually all other candidate biologic weapons. Coping with a global pandemic produced by use of this weapon would require significant investment in research and development of vaccine and antiviral therapies as well as a significant investment of public health resources. A careful risk assessment as to the likelihood of this agent being employed would guide the appropriate response required.

     Vaccination is the only proven and feasible means of combating an epidemic that could result from deliberate release of smallpox virus in the U.S. population. Antiviral drugs may be of value in dealing with infected patients, but expanding ring vaccination is the proven means of eradicating foci of infection. Smallpox vaccine is therefore the second (and last) exception to the committee's strong preference for treatments in planning for terrorist incidents rather than preexposure prophylaxis. The current U.S. stockpile of vaccine is far less than would be needed in the event of such a contingency. There is no existing licensed manufacturing capacity for production of additional stocks of the current vaccine. Use of the current vaccine would entail a substantial risk of vaccine-induced complications, many of which would require treatment by vaccinia-immune globulin and antiviral drugs. The epidemic of AIDS substantially increases the risk, since generalized vaccinia is known to occur in vaccinated AIDS patients. Reestablishing manufacturing of the current product is not a recommended option in view of the undesirable characteristics of the product and the potential for improvement. An experimental vaccine has been under development by the U.S. Army and is included in the current DoD Joint Vaccine Acquisition Program. This candidate vaccine contains a virus derived from a previously licensed vaccinia strain (Connaught strain), has been produced in cell culture, and has progressed to phase one trials. Licensure of a new vaccine in the absence of any possibility to conduct efficacy trials will pose multiple research problems, especially those relating to correlates of vaccine-induced immunity and levels of protection. Although unlikely to solve all of the problems of traditional vaccinia vaccine, development and manufacture of adequate stockpiles of this vaccine is the current best option for dealing with the contingency of a release of smallpox virus.

8-16 The development, manufacture, and stockpiling of an improved smallpox vaccine for post-attack management of a potential epidemic should be given a high priority. DHHS agencies could assist the military development program by addressing research questions related to product development, such as correlates of immunity. An agreement with the DoD and the manufacturer of a new vaccine on purchase and stockpile of vaccine for civilian use may be an important incentive and an important factor in production planning.

     Recent research by scientists at NIH and USAMRIID on antiviral drugs against orthopox viruses, including variola, have shown some promising leads, including antivariola activity by at least three classes of compounds (Huggins et al., 1996; 1998). One of these includes a licensed drug, cidofovir (marketed as Vistide™), currently used to treat cytomegalovirus retinitis in AIDS patients. However, cidofovir is an intravenous preparation with substantial toxicity and would therefore be of limited value in the event of a terrorist release of variola virus. Retention of U.S. stocks of variola virus currently scheduled for destruction in 1999 would be of value to a drug discovery and development program. Pox viruses vary widely in their sensitivity to chemotherapeutic agents, and use of surrogates for variola such as monkeypox cannot be relied upon totally, although monkeypox is a serious emerging disease in central Africa, and development of a drug that is active against variola and monkeypox would provide an immediate benefit independent of terrorism. Similarly, development of an antipox drug that could treat vaccinia problems, such as eczema vaccinatum and generalized vaccinia in the immunosuppressed, could be of great value in the event we need to do large-scale vaccination in the future. Notwithstanding the possible additional benefits of focusing drug development on variola surrogates, direct in vitro testing against variola virus is important to be sure of usefulness in treating smallpox.

8-17 A major R&D program should be undertaken to exploit the previous studies to discover and develop new antismallpox drugs for therapy and/or prophylaxis.

     Although other viruses can also produce encephalitis, three closely related enveloped RNA viruses of the Alphavirus genus initially recovered from moribund horses in the 1930s are considered the primary candidates for weaponization: Venezuelan equine encephalomyelitis virus (VEE), eastern equine encephalomyelitis virus (EEE), and western equine encephalomyelitis (WEE). All could be inexpensively produced in quantity, are relatively stable, and are readily amenable to genetic manipulations that might confound defenses against them. Natural infections are acquired through mosquito bites, but these viruses are also highly infectious as aerosols. Victims develop an incapacitating combination of fever, headache, and fatigue, and the most severe of the three, EEE, results in case fatality rates of 50 to 75 percent. Survivors may be left with seizures, sensorimotor deficits, or cognitive impairment. See Smith et al. (1997) for additional information.

     A live attenuated vaccine for VEE (TC-83) is immunogenic in 80 percent of recipients, but it causes more than 20 percent of recipients to experience high fever, malaise, and headache serious enough to require bed rest. Inactivated vaccines for VEE, WEE, and EEE in humans also exist; they require multiple injections and have poor immunogenicity. All vaccines, including TC-83, are available only in IND status.

     No specific therapy exists for these alphavirus encephalitides and treatment is directed at management of specific symptoms (e.g., convulsions, respiratory infection, and high fever). Even treatment with virus-neutralizing antisera (antibody-containing serum from the blood of previously-infected patients or animals) will fail to stop progression of established encephalitis. Antimosquito precautions should also be implemented.

     As is the case with most of the other infectious diseases of concern to the biological defense program, the primary thrust of current R&D is on vaccine development, and several candidate vaccines using live attenuated VEE, WEE, and EEE viruses have been identified and tested in animal models at USAMRIID. Unlike the antibiotics used to treat bacterial infections, most antiviral drugs are highly virus-specific, so drugs like AZT and the protease inhibitors that have proven so successful in controlling HIV, for example, have not been useful against the viral encephali-tides. For the same reason, there is little incentive for drug companies to pursue an antiviral drug for VEE, WEE, or EEE, diseases for which there is essentially no market in the developed nations. The potential market for a broadly effective antiviral drug is huge, however, and DARPA is sponsoring a number of research projects aimed at structures or processes common to a number of different viruses. For example, scientists of enVision and Boston Biomedical Research Institute have isolated developmental proteins which regulate cell proliferation in animal fetal tissues (Barnea et al, 1995; 1996) and have begun testing them for activity against viruses (whose replication is inherently linked to the host cell).

      A project at the University of Wisconsin focuses on design of compounds that inhibit viral entry, intracellular transport, maturation, and release. Combinatorial chemistry and organic synthesis will be used to design compounds to prevent progression of infection by viruses with bioterrorism potential.

      Teams at the University of Texas Medical Branch and the University of Wisconsin are working jointly using combinatorial chemistry to design antiviral drugs that will act by inhibition of capsid-RNA interaction, polymerase activity, or glycoprotein attachment to cellular receptors. A group at the University of Alabama, Birmingham is also using combinatorial chemistry to design a drug to inhibit capsid-RNA interaction (Edberg and Luo, 1997; DeLucas, 1998), and scientists at The Scripps Research Institute are attempting to build antiviral antibodies that will enter infected cells and fight viruses at the intracellular level (McLane et al., 1995; LeBlanc et al., 1998).

     Finally, GeneLabs Technology is attempting to develop broad-spectrum antiviral drugs from a large library of chemical compounds by assaying for RNA binding and selecting for dimer molecules that bind dsRNA, but not DNA.

     All of these projects, funded as they are by DoD, begin with the aim of improving biological defense, so, although most are years away from a licensed product, effectiveness against viruses like VEE will be a central feature rather than an adventitious side effect.

8-18 Support for broad-spectrum antiviral drugs for treatment of VEE, WEE, EEE and other viruses considered biological terrorism threats should be considered a high priority.

     Viral hemorrhagic fever is a term indicating an acute febrile illness accompanied by circulatory abnormalities and increased vascular permeability. Similar diseases result from infection with any of about a dozen RNA viruses belonging to four different families: Arenaviridea (Lassa, Argentine, Bolivian, Venezuelan, Brazilian), Bunyaviridea (Rift Valley, Crimean-Congo, and Hantaan), Filoviridea (Marburg and Ebola), and Flaviviridea (Dengue and Yellow Fever). All of these diseases are thought to be transmitted to humans through contact with infected animal reservoirs or arthropod vectors (mosquito or tick). All are relatively stable and highly infectious as fine-particle aerosols. Patients generally present with high fever and some indication of vascular involvement: low blood pressure, flushing, or small subcutaneous hemorrhage. Progression of the disease typically involves bleeding from mucous membranes, signs of pulmonary, liver, or kidney failure, and shock. Mortality varies widely among the diseases, from 5 to 20 percent of symptomatic cases for most, but as high as 90 percent for Ebola virus. See Jahrling (1997) for further information.

     Vaccines are available for yellow fever (YF), Rift Valley fever (RVF), and Argentine hemorrhagic fever (Junin virus, JUN). Cross protection against Bolivian hemorrhagic fever may also be provided by the Junin vaccine. These are the only vaccines available for any of this set of diseases and will be effective only if personnel are immunized before exposure. Only yellow fever vaccine is licensed by the FDA; the others are used in the United States under IND protocols and can only be obtained through the CDC.

     Vaccines have no application in treatment of exposed targets. Intravenous administration of the antiviral drug ribavirin is recommended for therapy of infections with Lassa virus and with Hantaan and other Old World Hantaan-related viruses. Ribavirin may also be useful for treatment of infections by other arenaviruses and Crimean-Congo hemorrhagic fever (CCHF) virus, but data proving efficacy are lacking. Ribavirin for these infections is used under IND protocols. It is not thought likely to be effective against filoviruses, such as Ebola, or flaviviruses, such as YF or Dengue. There is no proven chemotherapeutic drug available. Human immune serum is efficacious for treatment of persons exposed to Junin virus. Some anecdotal evidence suggests that Ebola human convalescent serum may be effective in preventing death from Ebola virus, but no scientifically controlled studies have been reported. Case management includes careful monitoring of fluid and electrolytes and intravenous corrective therapy where needed.

     Hospitalization under barrier precautions (gloves and gowns, face shields, or surgical masks and eye protection, for all those coming within 3 feet of the patient) is usually adequate to prevent transmission of Ebola, Lassa, CCHF, and other hemorrhagic fevers, but isolation of the patient provides an added measure of safety and is preferred, if facilities are available. Disinfection of bedding, utensils, and excreta by heat or chemicals is recommended for all of the viral diseases under consideration. Quarantine, defined by Benenson (1995) as "restriction of the activities of well persons or animals who have been exposed to a case of communicable disease during its period of communicability, to prevent disease transmission during the incubation period if infection should occur," may be indicated following an act of bioterrorism. If the agent is already identified, the decision to quarantine should be made based on the known communicability of the agent. Quarantine, for instance is not recommended for those exposed to anthrax, but is recommended for those exposed to plague, if chemoprophylaxis is not available. If the agent is not identified, then quarantine should be considered. CDC has provided detailed instructions on the management of suspected hemorrhagic cases, including handling and laboratory testing of potentially infectious materials (Centers for Disease Control and Prevention, 1988, 1995b).

     CDC, USAMRIID, and NIH all support small programs of research on one or more of the hemorrhagic fever viruses, primarily basic research on mechanisms of pathogenicity and explorations of possible vaccine candidates. The collaboration between USAMRIID and the NIAID Drug Discovery Program that identified cidofovir as a potential smallpox treatment has also discovered a class of compounds (s-adenosyl homocysteine hydrolase inhibitors) that may be effective against filoviruses, such as Ebola. In a mouse Ebola model that produces 100 percent mortality within 7 days, treatment beginning on the day of exposure provided 100 percent protection, and treatment beginning 4 days after exposure saved 40 percent of infected mice (memo from John Huggins of USAMRIID to F Manning, May 21, 1998).

     Much of the discussion of antiviral drug therapy for the viral encephalitides is also applicable in the case of the hemorrhagic fever viruses. The apparent differences in the activity of ribavirin against the various viruses of this group further underlines the specificity of current antivirals and emphasizes the need for broad-spectrum compounds. In addition to the DARPA-sponsored work cited in that discussion, researchers at Inotek have developed a compound that inhibits the nitric oxide pathway at multiple sites. Although this drug has potential for broad application to infectious agents that cause oxidative damage as part of their pathogenesis, it is being tested for efficacy in the Ebola guinea pig model. Additionally, arenavirus anti-polymerase humanized monoclonal antibodies will be synthesized at the University of Wisconsin for evaluation as antiviral drugs and researchers at the University of Texas Medical Branch are attempting to find ways to inhibit the intracellular transcription factor NFkB to modulate cytokine effects that are associated with arenavirus pathogenicity.

8-19 Support for the discovery and development of antiviral drugs for treatment of viral hemorrhagic fevers and other viral diseases considered biological terrorism threats should be considered a high priority.

     Botulism, an often lethal form of poisoning associated with improperly canned or stored foods, is the result of neurotoxins produced by the spore-forming anaerobic bacterium Clostridium botulinum. The botulinum toxins are the most toxic substances known. Some have an estimated LD₅₀ of 1 nanogram/kilogram of body weight (Gill, 1982). Some in vitro work suggests that these neurotoxins act presynaptically to block the release of acetylcholine and perhaps other neurotransmitters (Habermann, 1989), but the exact mechanism is as yet unknown. It is known that whether ingested, inhaled, or injected, the clinical course is similar. Several hours to 1 or 2 days later, dry mouth, difficulty swallowing, and double vision may be reported, followed by a progressive muscle weakness culminating in respiratory failure from skeletal muscle paralysis. See Middlebrook and Franz (1997) for additional information.

     The currently available vaccine is a formalin-fixed supernatant from cultures of C. botulinum. It protects against botulinum toxin types A through E, but is available only as an IND product, with a license held by the CDC. A series of three vaccinations must be started 12 weeks before exposure, and 80 percent of recipients exhibit protective titers at 14 weeks. Yearly boosters are required to maintain protection. Although it is an IND product, the vaccine has been given to hundreds of people since its development in the 1950s.

     Foodborne botulism is treated with a licensed trivalent equine antitoxin (serotypes A, B, and E) that is available only from the CDC. There is no other approved therapy for airborne botulism, although animal studies show that botulinum antitoxin can be very effective if given before the manifestation of clinical signs of disease. Mechanical ventilation is invariably necessary due to paralysis of respiratory muscles, if antitoxin is not given before the onset of clinical signs (Shapiro et al., 1997).

     See Table 8-6. A despeciated equine heptavalent antitoxin that has been developed by the U.S. Army specifically for aerosol exposures to serotypes A—G has IND status (Franz et al., 1997). This antitoxin markedly reduces the chances of serum sickness by eliminating the species-specific antigens from the horse immunoglobin (the basic immunoglobulin molecule is altered by removing complement fixing (Fc) region with pepsin to produce a fragment labeled F(ab¹)₂). This F(ab¹)₂ antitoxin protected animals from inhaled toxin at doses of 10 times the LD₅₀ when given prior to exposure and was also fully protective when given after exposure, as long as it was given prior to the onset of clinical signs. Further DoD work focuses on developing a recombinant vaccine, with much less substantial investigations of monoclonal antibodies and drugs to inhibit toxin uptake into cells (metalloprotease inhibitors).

8-20 Recombinant vaccines, monoclonal antibodies, and antibody fragments all have potential benefit, but require extensive investigation. Advances in these areas would benefit unintentional botulism case care, and the effort could provide prophylaxis and early treatment for those exposed to a potential toxin in a foodborne epidemic. Investigation of new techniques for this disorder would thus have substantial societal benefit for a rare clinical occurrence that is also a possible biological warfare event.

8-21 Further investigation into the utility of botulinum immune globulin would offer an immediate post-exposure therapy. This would be a great advance, because current passive immunity is equine derived and probably only beneficial prior to the onset of symptoms. Preexposure active immunization might be beneficial for those with the potential to be placed at very high risk, such as Hazmat or MMST teams, but this appears to be a very low-probability occurrence.

     SEB is one of seven toxins produced by strains of the Staphylococcus aureus bacterium. Like botulinum toxin, it is most often associated with food poisoning. Unlike botulism neurotoxin, SEB appears to exert its effects through overstimulation of cytokine production by the immune system (Ulrich et al., 1997). Ingested SEB is incapacitating rather than lethal, with vomiting and nausea prominent. It is relatively stable in aerosol, however, and the consequences of inhalation may be much more severe, possibly even a fatal "toxic shock" syndrome involving high fever, a rapid drop in blood pressure, and multiple organ failure.