Clinical development and testing B. L e o n a r d H o l m a n , M D , a n d Alun G. J o n e s , P h D Since its beginning in the 1950s, medical imaging with radioactive compounds has been successful because academic centers have played a crucial role in the design and testing of new radiopharmaceuticals. Their success has come about through the close cooperation of radiopharmaceutical chemists and clinical investigators. But the process for new radiopharmaceutical development has changed over the years and has, perhaps, become less cordial to the rapid transfer of biomedical technology to clinical practice. The development and testing of the myocardial perfusion agent 99mTc-labeled MIBI (cardiolite) is an illuminating example of the process.

Myocardial Scintigraphy: The Beginning With the discovery by Zaret et al? that transient myocardial ischemia could be imaged after the injection of a radioactive isotope of potassium, a major aim of radiopharmaceutical chemistry research became the synthesis of a 99mTc-labeled myocardial perfusion agent. Planar images obtained with 43K and 129Cs w e r e marginal at best because of the poor physical characteristics of the radionuclides? The introduction of 2~ resulted in a significant improvement in the technique, 3 particularly with the advent of delayed imaging 4 and tomography. 5 Thallium 201 is far from the ideal radionuclide, 6 however, and with its superior physical characteristics, 99roTeremained the target radionuclide for myocardial imaging during the late seventies and early eighties. The first major breakthrough came in 1981. A 99mTc-labeled compound synthesized by Deutsch et al., 7 DMPE, showed excellent myocardial uptake in a number of animal species including the primate. Their results were promising enough to warrant considerable commercial investment in the development of a suitable kit formulation for clinical testing. What followed was totally unexpected. Despite the very good images obtained in primates as large as the baboon, the uptake of the DMPE complex in the human heart was marginal at best. As a result, the axiom that had previously guided radiopharmaceutical development, namely, that biodistribution in

Reprint requests: B. Leonard Holman, MD, Chairman, Department of Radiology, Brigham and Women's Hospital, 75 Francis St., Boston, MA 02115. J NUCL CARDIOL1995;2:71-4. Copyright 9 1995 by the American Society of Nuclear Cardiology. 1071-3581/95/$3.00 + 0 43/72/61802

animal models will reflect uptake in the human being, was shattered. A new axiom, that human beings are the only species able to provide pharmakokinetic data accurate enough to predict clinical utility, would shape the future course of radiopharmaceutical design. Adherence to this new axiom has become essential as the cost of new drug approvals has spiralled over the years. In 1981 the first 99mTc-labeled isonitrile complex was prepared as part of an unrelated study of the basic chemistry of technetium, s The stability of these pure organometallic compounds, their ready formation in aqueous media at all concentrations, and their lipophilicity and charge made the class an interesting study in pharmacokinetics. Several of these isonitriles, most notably hexakis (t-butylisonitrile) technetium (I), accumulated avidly in the normal myocardium. Could these results be replicated in the human subject?

RDCRs Versus INDs With the development of a new radiopharmaceutical that shows promising imaging characteristics in animal models, an important decision must be made. Can human testing begin under the institution's Radioactive Drug Research Committee (RDRC) or must an application for an investigational new drug (IND) be submitted to the Food and Drug Administration (FDA)? The RDRCs were created by the FDA in 1975 to approve and oversee certain types of radioactive drug research involving human subjects at the local level. The RDRC can be used only under specific conditions when the radioactive drug is generally recognized as safe and effective for research use. In general, this precludes the use of RDRC approval for studying radioactive drugs that are new chemical entities. 9'1~Although it would seem that this condition would negate the value of the RDRC for new radiopharmaceuticals, there is one important instance in which the RDRC can be used. When a nonradioactive pharmaceutical has been approved by the FDA for clinical use or when a sufficient body of evidence demonstrating the pharmaceutical's safety in human beings exists in the literature, and when the compound has been labeled with a radionuclide, it may meet the requirements for R D R C approval. Secondly, the research must be basic, involving studies of metabolism, human physiology, pathophysiology, or biochemistry. Initial human biodistribution studies are included within the scope of this criterion. The 71

72

Holman and Jones Clinical development and testing

protocol design, however, cannot involve "immediate therapeutic, diagnostic or similar purposes or to determine the safety and effectiveness of the drug in humans for such purposes (i.e., to carry out a clinical trial). ''11 In addition, the amount of active ingredient or combination of active ingredients that will be administered must be shown to have no clinically detectable pharmacologic effect in human beings, the radiation exposure must be less than the amount prescribed in the regulations (similar to the maximum permissible exposure for occupational workers), the investigator must be qualified to conduct a proposed study and licensed to handle radioactive materials, and the rights of human subjects must be protected through proper review of the research by an institutional review board) ~ For human clinical trials that use a new radiopharmaceutical of which the safety and efficacy has not been established in human beings, a "Notice of Claimed Investigational Exemption" and IND must be submitted to the FDA. The IND is a major undertaking for the average academic center because it requires acute and subacute toxicity (including pathologic studies) in two species (one rodent and one nonrodent). This requirement, coupled with dosimetry data, can cost up to $100,000 if performed commercially. Since each change of the parent compound or the manufacturing procedure results in a new drug formulation, a new IND application may be required with each change along the way to a final product. For an academic center, this clearly presents a funding problem in the absence of industrial interest in the radiopharmaceutical. P h a s e I a n d !! o f t h e IND

After the submission of an IND to the FDA, as well as approval by the appropriate local committees (the institutional review boards and uSually the Radiation Safety Committee), and within 30 days of submission if the FDA does not comment, phase I and phase II studies can begin. Phase I studies are usually carried out in a limited number of persons to derive dosimetry data and determine basic pharmacologic information such as biodistribution, dosage range, and routes of administration of the radiopharmaceutical and its associated products. Phase II studies involve a limited number of subjects with a specific disease likely to occur in the clinical application of the radiopharmaceutical to determine whether there are any differences in dosimetry, safety, and biodistribution as a result of the disease. 12 Phase I and II studies demonstrated that, unlike previous 99mTC myocardial agents, 99mTc-labeled TBI

JOURNALOF NUCLEARCARDIOLOGY January/February 1995

accumulated in the human heart almost as avidly as 2roT1.13 Unfortunately, its other biologic characteristics were not helpful. Uptake in the lungs and liver were high and clearance from both organs slow. As a result, 2 to 4 hour delays were necessary before satisfactory images could be obtained. Nevertheless, a major barrier had been passed because these studies demonstrated that it was indeed possible to visualize the human heart with a complex containing 99mTC"

Identifying better isonitrile compounds turned out to be a complicated process. The range of possible substitutes on the isonitrile ligand meant that screening would involve hundreds of compounds. Of those synthesized, several appeared promising in animals and animal models. But, as we learned, animal experiments were only a rough guide to human biodistribution. The second member of the isonitrile class to be tested in humans, 99mTc-labeled CPI, was a total redesign. Rather than substituting inert pendant groups on the ligand, an ester moiety was chosen in an effort to improve the pharmacokinetics of the final complex. The concept was essentially that of a suicide agent: it was predicted that the ester (cationic) species would survive in the vasculature for sufficient time to allow localization in the myocardium where they would begin to hydrolyze. This hydrolysis would progressively generate more hydrophilic (anionic) species to promote clearance, particularly through the kidneys. While 99mTc-TBI was retained for a long time in the myocardium, 99mTc-CP] had a biologic half life of 2l/2 hours. 6 Reinjection of 99mTc-CPI was possible by 3 to 4 hours after initial injection because there was little redistribution into regions of transient ischemia. Clearance from the lung and the liver for this agent was also faster for 99mTc-CPI than for 99mTc-TBI. In 1986, a third agent 99mTc-MIBI was identified as having the most suitable imaging characteristics of all and was subsequently developed into a commercial productJ 4 To synthesize 99mTc-MIBI, another group was substituted for the final - C H 3 group in 99mTCTIBI. The result was a significant increase in heart activity compared with surrounding tissues, particularly after exercise. Furthermore, the fairly rapid clearance of 99mTc-MIBI from surrounding tissues coupled with a long biologic half-time in the heart resulted in high quality planar images of the myocardium and facilitated single photon emission computed tomography (SPECT) imaging. This agent was developed further during 1986-198715 and found promising enough to justify testing in the third phase of the IND process.

JOURNAL OF NUCLEARCARDIOLOGY Volume 2, Number 1;71-74

P h a s e !11 o f t h e IND

"Phase III studies involve the study of sufficient numbers of patients by two or more investigators to establish safety and efficacy and directions for use for the particular dosage form of the RDP [radiopharmaceutical drug products] for each proposed indication. ''16 Phase III studies have become too large and expensive to be carried out by academic centers alone. INDs are generally submitted by an industrial sponsor, and phase III studies are carried out at multiple academic centers under the guidance and supervision of the pharmaceutical company. Furthermore, only a small number of phase III studies are carried out for each radiopharmaceutical with very strict attention to protocol design. Since the endpoint is FDA approval of the product, only one or two diagnostic strategies are usually tested in the phase III process. Among the many factors that go into a well-designed clinical study, the disease must be established by an alternate well-established method (often called a gold standard). Therefore the design of the study is often dictated by availability of this gold standard rather than identifying and pursuing the most valuable potential clinical applications of the diagnostic method.' In the case of 99mTc-MIBI, the availability of a gold standard coincided with the most important clinical applications of the method. Coronary artery disease can be detected with reasonable accuracy with coronary angiography and 99mTc-MIBI images can be compared quantitatively with 2~ myocardial scintigraphy. At the same time, coronary artery disease detection and evaluation were the most likely applications of the method. Such a correspondence was not the case for brain perfusion agents, for example, where the endpoint for phase III studies were detection of chronic stroke because computed tomography scanning was a well-established gold standard. ~7 After approval, however, the clinical applications of brain perfusion SPECT have included almost every neurologic and psychiatric disease except chronic stroke, is Since the FDA does not prohibit physicians from using radiopharmaceuticals (or other drugs) for medical indications other than those listed on the label (package insert), there is no incentive on the part of industry and the academic community to expand approval to other indications once initial FDA approval has been obtained. Consequently, the one or two approved indications of the radiopharmaceutical may not accurately reflect its most valuable uses in the clinic. Phase III studies are not cutting edge science.

Holman and Jones

73

Clinical development and testing

Clinical facilities frequently participate because it is exciting to be using a new radiopharmaceutical. But the protocols are geared for meeting the FDA requirements for new drug approval and seldom chart new ground and rarely ask fundamentally important scientific questions. On the other hand, the academic core facility offers the opportunity to develop and apply quantitative state-of-the-art methodology to assure that imaging and data acquisition standards are high, that they are reproducible at all sites, and that protocols are implemented in a consistent manner. In this way, all the data from each of the sites can be treated equivalently. Wackers and the group at Yale University, for example, developed excellent standards to assess 99mTc-MIBI images quantitatively and to compare them with angiography and 2roT1 imaging.15 The outcome of their work has formed the basis for a good deal of the quantitative approaches to myocardial scintigraphy that have been implemented clinically. While limiting the number of phase III protocols has reduced the opportunity for innovative clinical investigations, it has also reduced the number of poorly controlled studies. In the approval process, poorly designed studies usually result in bad data that translate to lost time and money. Even if the study is repeated, the bad data remain on record if they cannot be explained and often delay the approval process even after the properly conducted studies have been completed. Understandably, industry has severely limited the number of investigators participating in phase III studies for each new radiopharmaceutical. On completion of data collection and analysis of all phase III studies, the sponsor submits a final report, the NDA, to the FDA approval. The 99mTc-isonitrile story is a successful example of new drug development. Nevertheless, unsettling questions remain. 99mTc-MIBI is not an ideal myocardial imaging agent. Instantaneous myocardial extraction is not high and background activity (particularly from liver) can be a problem. Are there more favorable isonitriles for myocardial scintigraphy? Are there other isonitriles that might complement 99mTcMIBI for specific cardiac applications? Are there new clinical applications for other members of isonitrile series? It is unlikely that we will ever know the answers because the cost of screening a range of ligands with the same basic structure in human beings is prohibitive for academic centers and, once the new radiopharmaceutical has been approved, the costs for going through the approval process for a new radiopharmaceutical is unacceptably high even for a pharmaceutical company.

74

Holman and Jones Clinical development and testing

Need for Changing the Process

Ways must be found to adapt the FDA regulations to the reality of radiopharmaceutical screening. We now know that only human data is reliable for determining the potential of a new radiolabeled compound, particularly where a choice between several in a class is necessary. It is also clear that funding, whether federal or industrial, must be set aside for the preclinical testing of promising new agents. Without such support, the incentive for the synthesis and critical development of radiotracers is diminished and the academic role in new radiopharmaceutical development shrinks even further. Finally, a solution must be found to encourage innovative clinical research before FDA approval of the new radiopharmaceutical. Modifications in the interpretation of existing FDA regulations must permit innovative research on drugs under investigation without raising additional barriers to FDA approval. Such innovative clinical work does go on, but often not in the United States. However, without the opportunity for expanding clinical investigations of new radiopharmaceuticals soon after they are designed, the incentive for their development will wither. A flourishing academic role in radiopharmaceutical design and development is essential not only for the future of nuclear cardiology, but for all applications of radiotracers in medicine. We thank Barry S. Siegel and S. James Adelstein for their criticisms and suggestions and to Linda Morris for her assistance in preparation of the manuscript.

References 1. Zaret BL, Strauss HW, Martin ND, Wells HP, Flamm MD. Noninvasive regional myocardial perfusion with radioactive potassium. Study of patients at rest, with exercise and during angina pectoris. New Engl J Med 1973;288:809-12. 2. Nishiyama H, Sodd VJ, Adolph RJ, Saenger EL, Lewis JT, Gabel M. Intercomparison of myocardial imaging agents: 2~ 129Cs, 43K, and SlRb. J Nucl Med 1976;17:880-9. 3. Lebowitz E, Greene MW, Bradley-Moore P, et al. 2~ for medical use. J Nucl Med 1975;16:151-5.

JOURNALOF NUCLEARCARDIOLOGY January/February 1995 4. Pohost GM, Zir LM, Moore RH, McKusick KA, Guiney TE, Bcller GA. Differentiation of transiently ischemic from infarcted myocardium by serial imaging after a single dose of thallium-201. Circulation 1977;55:294-302. 5. Holman BL, Hill TC, Wynn J, Lovett RD, Zimmerman RE, Smith EM. Single-photon transaxial emission computed tomography of the heart in normal subjects and in patients with infarction. J Nucl Med 1979;20:736-40. 6. Holman BL, Sporn V, Jones AG, et al. Myocardial imaging with technetium-99m CPI: initial experience in the human. J Nucl Med 1987;28:13-8. 7. Deutsch E, Bushong W, Glavan KA, et al. Heart imaging with cationic complexes of technetium. Science 1981;214: 85-6. 8. Abrams M J, Davison A, Jones AG, Costello CE, Pang H. Synthesis and characterization of hexakis(alkyl isocyanide) and hexakis(aryl isocyanide) complexes of technetium(I). Inorg Chem 1983;22:2798-800. 9. Temple R. FDA Letter to radioactive drug research chairpersons. J Nucl Med 1985;26:815-6. 10. Swanson DP, Lieto RP. The submission of I.N.D. applications for radiopharmaceutical research: when and why. J Nucl Med 1984;25:714-9. 11. Anonymous. Radioactive new drugs; radioactive biologics (21 CRF Parts 1, 310, 312, 370). Federal Register. 1974 July 29; 39(Number 146):27538-45. 12. Siegel BA. Radiopharmaceuticals and FDA: a clinician's prospective. Med Instru 1981;15:355-60. 13. Holman BL, Jones AG, Lister-James J, et al. A new Tc-99mlabeled myocardial imaging agent, hexakis(t-butylisonitrile)technetium (I) [Tc-99m TBI]: initial experience in the human. J Nucl Med 1984;25:1350-5. 14. Mousa SA, Cooney JM, Williams SJ. Regional myocardial distribution RP-30 in animal models of myocardial ischemia and reperfusion. J Nucl Med 1987;28:620. 15. Wackers FJT, Berman DS, Maddahi J, et al. Technetium-99m hexakis 2-methoxyisobutyl isonitrile: human biodistribution, dosimetry, safety, and preliminary comparison to thallium-201 for myocardial perfusion imaging. J Nucl Med 1989;30:301-11. 16. Larson SM, Siegel BA, Robinson RG. Guidelines for the clinical evaluation of radiopharmaceutical drugs (Letter). J Nucl Med 1978;19:1359-62. 17. Holman BL, Hellman RS, Goldsmith SJ, et al. Biodistribution, dosimetry, and clinical evaluation of technetium-99m ethyl cysteinate dimer in normal subjects and in patients with chronic cerebral infarction. J Nucl Med 1989;30:1018-25. 18. Holman BL, Devous MD. Functional brain SPECT: the emergence of a powerful clinical method. J Nucl Med 1992;33:1888-904.