The United States Food and Drug Administration and the End of Antibiotics

0 Beth Israel Deaconess Medical Center and Harvard Medical School , Boston, Massachusetts 1 Wyeth-Ayerst Research , Pearl River , New York - SirOn behalf of the Clinical Affairs Committee (CAC) of the Infectious Diseases Society of America (IDSA), we want to express our appreciation and gratitude to Dr. Joiner and colleagues for assessing the adequacy of fellowship training in our specialty [1]. As the IDSA attempts to respond to the needs of the trainees who are beginning a dynamic, exciting and challenging career in infectious diseases (ID), we applaud the willingness of the Training Program Directors Committee to determine how graduates evaluate their experiencesinformation that was not sought in their initial report on fellowship training in ID [2] but was incorporated into the 1998 IDSA membership survey [3]. The Committees stated intention to incorporate the results of the survey into recommendations for changing the design of training programs is the best guarantee that fellowships will remain enlightening, educational, and pertinent preparations for careers in ID. In light of this commitment and the fact that the majority of IDSA members are now clinicians, we note with interest the finding that 190% of the fellows surveyed found their training lacking with respect to the business aspects of medicine, personnel management, billing, coding, practice marketing, and dealing with managed care contracts. We agree with the authors statement that it is rare for an academic ID program to have faculty with expertise in the business aspects of the discipline ([1], p. 261). However, the CAC strongly disagrees with their recommendation that this weakness be corrected by ceding the responsibility for our fellows education in business acumen to internal medicine program directors and departments. Do we allow pulmonologists to teach our fellows how to diagnose and treat pneumonia in an immunocompromised host? Shall we ask intensivecare specialists to instruct them on the use of antibiotics in critical care units? We believe this educational charge must be taken up by the IDSA and, specifically, the CAC. The solution lies within the IDSA: it is the clinicians in the private practice of ID. These physicians have been required to become experts in running and analyzing economically successful practices. They have been forced to learn the business of medicine to survive and prosper, and they are now willing and capable of transmitting their experience and insight to our fellows in training. Our committee firmly and enthusiastically recommends a collaborative effort between program directors and the CAC and IDSA leadership to strengthen and expand fellows training in the increasingly demanding but essential skills of mastering the business aspects of ID practice. The CAC also feels this cannot be accomplished in a 1-day symposium, such as Fellows Day at the Annual Meeting. The extent and complexity of this subject would best be taught through sustained and prolonged experience with clinicians in private practice. Structured rotations with private practice clinicians, guided by a formal curriculum, may be the best way to accomplish this goal. The CAC believes this is a viable and practical way to correct the critical deficiency of training programs that educate fellows in the business aspects of ID practice a shortcoming that has now been now confirmed by the results of both the membership and the training graduates surveys. Marvin J. Tenenbaum1 and Russell Petrak2 1North Shore Infectious Diseases Consultants, Port Washington, New York, and 2Clinical Affairs Committee, Metro ID Consultants, Hinsdale, Illinois SirFor nearly 2 decades, antibacterial research has been the Cinderella area in the pharmaceutical industry. The market for these productswhich are used to treat acute, not chronic diseaseis modest. There are many products available, including many generic products. For the most part, the market is growing only slowly and, except for the problem of resistance, the public is largely satisfied. Recently, when presenting proposals for Phase III trial designs for antibacterial compounds to the US Food and Drug Administration (FDA) and to European regulatory bodies, a number of companies, both small and large, were told that the designs for equivalence studies had to target a 10% delta for the lower limit Cure rate, % Delta, % No. of patients 90% power 80% power 10% delta 15% delta 10% delta 15% delta a Related to indication C. of the confidence interval. This requirement, which seems innocent and technical, threw the pharmaceutical industry into a panic and probably contributed to the recent withdrawals by Lilly and Bristol-Myers Squibb from the antibacterial discovery business. Why? Most clinical trials leading to the approval of antibacterial drugs are based on equivalence studies. The delta statistically defines the boundary for equivalence. In the past, on the basis of the points-toconsider document that the FDA published in 1992 [1], a step function for the delta has been used (table 1). These boundaries have meant that the study results must show that there is a 95% (1-sided 97.5% CI) probability that true cure rate for the new drug is not more than 10%20% lower than the cure rate for the approved drug. In most reasonably sized studies, the new drug has had to be as good as or better than the old drug to be successful. The European regulatory authorities and the FDA are now suggesting that a 10% delta be used routinely in drug development [2], and the FDA has now disclaimed the old step function on their web site [3]. It is not completely clear upon what data this suggestion is based, other than purely statistical considerations. In fact, a quick calculation will show that 2 independent trials that were successful at a 15% delta would result in approval of a drug that is inferior at the 10% delta only 2% of the time. The concern that the FDA has expressed is about something called biocreep. According to this concept, a slightly inferior experimental drug becomes the comparator for the next generation of compounds, and so on, until the experimental drugs of the future asymptotically approach the efficacy of placebo. However, one must wonder whether, for serious infections, this is any more than a theoretical concern, especially when the most recent approvals (for the drugs quinupristin-dalfopristin [Synercid; Aventis] and linezolid [Zyvox; Pharmacia]) have been based on comparisons with standard therapy using older agents. In fact, one of the regulatory agencys best weapons against biocreep is their control over the choice of comparators in clinical trials. The deltas used in the step-function of the points-to-consider document from 1992 [1] were chosen not on the basis of scientific reasoning, but on the basis of the size of the trial that would be required given the cure rate (Pharmaceutical Manufacturers of America, personal communication). The trial size requi (...truncated)