perspec tives
nature publishing group
commentaries Antibacterial Drug Development: Challenges, Recent Developments, and Future Considerations S Nambiar1, K Laessig1, J Toerner2, J Farley2 and E Cox2 A decline in antibacterial drug development coupled with emerging bacterial resistance has resulted in limited treatment options. One of the challenges facing antibacterial drug development is appropriate clinical trial designs. Noninferiority trials are appropriate to study new antibacterial drugs for the treatment of serious diseases; superiority trials can be challenging and cannot be the only acceptable trial design to study antibacterial drugs. Our efforts must continue to make new therapies available to meet patient needs. Over the past few decades, there has been a decline in antibacterial drug development, resulting in fewer new antibacterial drug approvals. Antibacterial drug development has not kept pace with evolving bacterial resistance to the existing therapeutic armamentarium of antibacterial drugs, thus limiting treatment options for patients. The reasons for this decline are multifactorial; they include economic issues, the inherent challenges of studying a therapeutic for an acute infectious disease, and scientific issues with clinical trial designs for antibacterial drugs. The US Food and Drug Administration (FDA) has undertaken several activities and initiatives to facilitate the development of antibacterial drugs so that therapeutic options are, and will continue to be, available to health-care providers to
treat their patients. Over the last several years, we have made a concerted effort to revise our guidance documents to provide durable recommendations for scientifically sound, ethical, and feasible clinical trials to facilitate the development and evaluation of new antibacterial drugs. In some instances, our recommendations for clinical trial designs were scientifically sound but not feasible because of significant scientific, practical, and economic challenges. Substantial progress has been made with updating clinical trial design recommendations, and additional work is under way. Our recommendations have been informed by comments from pharmaceutical sponsors, academia, and private citizens that we received in response to draft guidance documents. In addition,
we have convened meetings and engaged in public–private partnerships to discuss clinical trial designs and end points. These efforts have included the work of the Foundation for the National Institutes of Health (FNIH) Biomarkers Consortium, the Clinical Trials Transformation Initiative, the Brookings Institution, and public meetings, including workshops and advisory committee meetings. The FDA has regular interactions and exchanges of information with colleagues from the European Medicines Agency to discuss clinical trial designs, recognizing that most drug development programs are submitted to both US and European regulators. Colleagues from the Centers for Disease Control and Prevention and the NIH have also contributed to the scientific discussions supporting advances in clinical trial designs. Since 2010, we have issued 11 guidance documents describing recommended clinical trial designs for studying antibacterial drugs (http://www.fda.gov/Drugs/ GuidanceComplianceRegulatoryInformation/Guidances/ucm064980.htm). The guidance documents for treatment of serious infectious diseases describe active controlled trials because a placebo or no-treatment arm would not be ethical. In most circumstances, these trials are designed to show noninferiority of the test drug to an approved control drug. Noninferiority trials are an appropriate trial design to study new antibacterial drugs for the treatment of serious diseases for which a treatment effect for a clinically meaningful end point can be established and where no treatment, or delay in treatment, would be neither safe nor ethical.1 Almost all antibacterial drugs that we rely on today did not have a finding of superiority in their clinical trial evaluations before approval, including the
1Division of Anti-Infective Products, US Food and Drug Administration, Silver Spring, Maryland, USA; 2Office of Antimicrobial Products, US Food and Drug Administration, Silver Spring, Maryland, USA. Correspondence: S Nambiar ([email protected])
doi:10.1038/mt.2014.116 Clinical pharmacology & Therapeutics | VOLUME 96 NUMBER 2 | AUGUST 2014
147
perspec tives ones we rely on when faced with few or no options to treat a patient; most antibacterial drugs used today were approved on the basis of noninferiority trials or less sophisticated approaches used several decades ago. Many of the guidance documents we have issued since 2008 have included specific recommendations to improve the scientific quality of noninferiority clinical trials. Although it is always valuable to consider new approaches, it is also important not to discount established methods that can provide scientifically sound evidence, particularly when there are critical patient needs for new antibacterial drugs. Our experience has shown that welldesigned noninferiority clinical trials can reveal significant unanticipated findings. Valuable lessons learned from some of these clinical trials include a better understanding of the biology of the disease, important considerations about the activity of the drug at the site of infection, variations in physiology in the sick population that could potentially affect the pharmacokinetics of the drug, and dose-selection considerations. The interaction between daptomycin and surfactant was elucidated after a clinical trial did not demonstrate that daptomycin was noninferior to an active comparator in community-acquired pneumonia.2 Similarly, the clinical trials of doripenem in ventilator-associated bacterial pneumonia (VABP) and of tigecycline in hospital-acquired bacterial pneumonia/ VABP shed light on unanticipated limitations of these antibacterial drugs when treating patients with VABP.3,4 There are, however, potential pitfalls in noninferiority trials if they are not designed and conducted properly. The end point(s) in these trials should be well defined and reliable, and should be assessed at a time point at which treatment effect can be demonstrated. In the past few years, considerable progress has been made to define end points for antibacterial clinical trials. The FNIH has made important contributions to clinical trial designs and end points for acute bacterial skin and skin structure infections and community-acquired bacterial pneumonia, and has also helped to provide a path forward for development 148
of new antibacterial drugs.5 The end points developed through this process are being used by pharmaceutical sponsors in ongoing trials and were also used in completed acute bacterial skin and skin structure infections trials that were recently discussed at an Anti-Infective Drugs Advisory Committee meeting.6 Although superiority trials can provide clear evidence of efficacy of an antibacterial drug, demonstrating superiority of a new antibacterial drug over existing active antibacterial therapies is often very challenging. A patient’s infection and response to an infection are influenced not only by the effects of the antibacterial drug and the infecting bacteria but also by host factors such as immunity, inflammation, and tissue repair. Also, if demonstrating superiority is dependent on having a less effective comparator because resistance to existing therapies has developed, then we are in a situation where drug development (despite stewardship and infection-control efforts) has not kept up with emerging resistance such that we can now enroll patients without adequate treatment options in a clinical trial—a situation most would prefer to avoid. In this scenario, to conduct such a trial requires a sufficient prevalence of the microorganism(s) for which we do not have adequate treatment options at present. Once a new therapy becomes available, ongoing trials designed to show superiority over a standard-of-care regimen (a regimen likely to be suboptimal) will probably become unethical and may need to be stopped. Moving forward, trials would probably be designed as active controlled trials to show noninferiority against the new standard for care. Thus, the window of opportunity during which superiority trials against suboptimal comparator are ethical is likely time-limited. Ideally, antibacterial drug development should be ongoing to provide treatment options before we have a population of patients for whom there are few or no effective drugs because of the emergence of new resistance mechanisms. Recognizing that antibacterial drug development has not kept pace with the development of resistance to available therapies and patient needs, we have been working on recommendations for streamlined drug
development programs for antibacterial drugs to treat patients with serious infections who have few, if any, treatment options. The FDA has published a draft guidance on developing antibacterial therapies for patients with unmet medical need that describes streamlined approaches to developing antibacterial drugs for patients with few or limited treatment options (e.g., due to resistance to existing therapies).7 This guidance describes the characteristics of new antibacterial drugs appropriate for such development pathways such as those that operate via different mechanisms of action, are paired with resistance inhibitors, or have chemical modifications to their structures conferring stability in the setting of selected resistance mechanisms. These streamlined programs are intended to provide a pathway for development of antibacterial drugs for patients with the greatest need. Because these development programs are streamlined and approval would be based on smaller databases, indications for use would be limited to situations in which a patient has few or no therapeutic options. The rationale for accepting a more streamlined development program is consistent with Title 21 of the Code of Federal Regulations (312 Subpart E), which describes expediting the development, evaluation, and marketing of new therapies intended to treat persons with life-threatening and severely debilitating illnesses, especially when no satisfactory alternative therapy exists. With such development programs, there is recognition that physicians and patients are willing to accept greater risks or side effects from drugs that treat life-threatening and severely debilitating illnesses than for drugs that treat less serious illnesses. For drugs approved with a smaller safety database, significant safety findings may first become apparent in the postmarketing period. It is critical that adequate steps be in place to monitor for safety signals or concerns regarding efficacy that may become apparent once a drug is marketed. In addition to the efforts in the area of clinical trial designs, Title VIII of the Food and Drug Administration Safety and Innovation Act (FDASIA), “Generating Antibiotic Incentives Now” (GAIN),
VOLUME 96 NUMBER 2 | AUGUST 2014 | www.nature.com/cpt
perspec tives is intended to encourage development of treatments for serious or life-threatening infections caused by bacteria or fungi. The term “qualified infectious disease product” (QIDP) refers to an antibacterial or antifungal human drug that is intended to treat serious or lifethreatening infections. For certain drugs designated as QIDP, GAIN provides an additional five years of exclusivity. An application for a drug designated as a QIDP is also eligible for priority review and designation as a fast-track product.8 The FDA has granted 47 QIDP designations representing 33 different antibacterial or antifungal drugs since the FDASIA was enacted in 2012. Antibacterial drug development is challenging on many fronts. Although significant progress has been made in the past few years on designing scientifically sound and feasible clinical trials, more work remains to be done to make new therapies available to meet patient needs. CONFLICT OF INTEREST The authors declared no conflict of interest. © 2014 ASCPT
1. US Food and Drug Administration. Guidance for industry: non-inferiority clinical trials. (March 2010). 2. Pertel, P.E. et al. Effects of prior effective therapy on the efficacy of daptomycin and ceftriaxone for the treatment of community-acquired pneumonia. Clin. Infect. Dis. 46, 1142–1151 (2008). 3. US Food and Drug Administration. Doribax (doripenem): drug safety communication— risk when used to treat pneumonia on ventilated patients (6 March 2014). 4. US Food and Drug Administration. FDA drug safety communication: FDA warns of increased risk of death with IV antibacterial Tygacil (tigecycline) and approves new boxed warning (27 September 2013). 5. Talbot, G.H. et al. Progress on developing endpoints for registrational clinical trials of community-acquired bacterial pneumonia and acute bacterial skin and skin structure infections: update from the Biomarkers Consortium of the Foundation for the National Institutes of Health. Clin. Infect. Dis. 55, 1114– 1121 (2012). 6. US Food and Drug Administration. March 31, 2014: Anti-Infective Drugs Advisory Committee meeting (12 February 2014). 7. US Food and Drug Administration. Guidance
for industry: antibacterial therapies for patients with unmet medical need for the treatment of serious bacterial diseases (July 2013). 8. Federal Register. Establishing a list of
qualifying pathogens under the Food and Drug Administration Safety and Innovation Act (12 June 2013).
See article page 239
Use of Internet Search Logs to Evaluate Potential Drug Adverse Events S Sarntivijai1 and DR Abernethy1 Internet search logs provide an abundant source of data that can be explored for purposes such as identifying drug exposure– adverse event relationships. The methodology to rigorously conduct such evaluations is not well characterized, and the utility of such analyses is not well defined. In this issue, White and colleagues propose an approach using Internet search logs for this purpose and compare it to parallel analyses conducted using the US Food and Drug Administration’s spontaneous reporting database. It is gratifying to see that various approaches are under evaluation to improve detection and interpretation of clinical adverse drug events. At present, a mainstay for postapproval monitoring for adverse drug events is the US Food and Drug Administration’s Adverse Event Reporting System (FAERS).1 The strengths and weaknesses of the FAERS have been discussed extensively. Weaknesses include under- or overreporting of adverse events as a result of the voluntary nature of the system and the burden of completing reports. The latter may be less of an issue with Internet search logs, but that is unknown at this time. It should be noted that a FAERS report contains considerable detail about the patient, the drug, the underlying clinical
condition, comorbid illness and concurrent drug exposures, and the temporal relationship between drug exposure and adverse event. This information would be lacking from Internet search log data. When a disproportionately strong association between an adverse event and a drug exposure is noted during ongoing analysis of FAERS data, it can be viewed as generation of a hypothesis. Such a finding leads to careful examination of other data sources, the published literature, and the information in the reports themselves. Evaluation of the reports can lead to a reevaluation of the association if it becomes clear that there are duplicate reports or there is evidence of stimulated reporting (e.g., medical–legal activity, heightened awareness due to public-
1US Food and Drug Administration, Silver Spring, Maryland, USA. Correspondence: DR Abernethy (darrell. [email protected])
doi:10.1038/clpt.2014.115
Clinical pharmacology & Therapeutics | VOLUME 96 NUMBER 2 | AUGUST 2014
149
nature publishing group
commentaries Antibacterial Drug Development: Challenges, Recent Developments, and Future Considerations S Nambiar1, K Laessig1, J Toerner2, J Farley2 and E Cox2 A decline in antibacterial drug development coupled with emerging bacterial resistance has resulted in limited treatment options. One of the challenges facing antibacterial drug development is appropriate clinical trial designs. Noninferiority trials are appropriate to study new antibacterial drugs for the treatment of serious diseases; superiority trials can be challenging and cannot be the only acceptable trial design to study antibacterial drugs. Our efforts must continue to make new therapies available to meet patient needs. Over the past few decades, there has been a decline in antibacterial drug development, resulting in fewer new antibacterial drug approvals. Antibacterial drug development has not kept pace with evolving bacterial resistance to the existing therapeutic armamentarium of antibacterial drugs, thus limiting treatment options for patients. The reasons for this decline are multifactorial; they include economic issues, the inherent challenges of studying a therapeutic for an acute infectious disease, and scientific issues with clinical trial designs for antibacterial drugs. The US Food and Drug Administration (FDA) has undertaken several activities and initiatives to facilitate the development of antibacterial drugs so that therapeutic options are, and will continue to be, available to health-care providers to
treat their patients. Over the last several years, we have made a concerted effort to revise our guidance documents to provide durable recommendations for scientifically sound, ethical, and feasible clinical trials to facilitate the development and evaluation of new antibacterial drugs. In some instances, our recommendations for clinical trial designs were scientifically sound but not feasible because of significant scientific, practical, and economic challenges. Substantial progress has been made with updating clinical trial design recommendations, and additional work is under way. Our recommendations have been informed by comments from pharmaceutical sponsors, academia, and private citizens that we received in response to draft guidance documents. In addition,
we have convened meetings and engaged in public–private partnerships to discuss clinical trial designs and end points. These efforts have included the work of the Foundation for the National Institutes of Health (FNIH) Biomarkers Consortium, the Clinical Trials Transformation Initiative, the Brookings Institution, and public meetings, including workshops and advisory committee meetings. The FDA has regular interactions and exchanges of information with colleagues from the European Medicines Agency to discuss clinical trial designs, recognizing that most drug development programs are submitted to both US and European regulators. Colleagues from the Centers for Disease Control and Prevention and the NIH have also contributed to the scientific discussions supporting advances in clinical trial designs. Since 2010, we have issued 11 guidance documents describing recommended clinical trial designs for studying antibacterial drugs (http://www.fda.gov/Drugs/ GuidanceComplianceRegulatoryInformation/Guidances/ucm064980.htm). The guidance documents for treatment of serious infectious diseases describe active controlled trials because a placebo or no-treatment arm would not be ethical. In most circumstances, these trials are designed to show noninferiority of the test drug to an approved control drug. Noninferiority trials are an appropriate trial design to study new antibacterial drugs for the treatment of serious diseases for which a treatment effect for a clinically meaningful end point can be established and where no treatment, or delay in treatment, would be neither safe nor ethical.1 Almost all antibacterial drugs that we rely on today did not have a finding of superiority in their clinical trial evaluations before approval, including the
1Division of Anti-Infective Products, US Food and Drug Administration, Silver Spring, Maryland, USA; 2Office of Antimicrobial Products, US Food and Drug Administration, Silver Spring, Maryland, USA. Correspondence: S Nambiar ([email protected])
doi:10.1038/mt.2014.116 Clinical pharmacology & Therapeutics | VOLUME 96 NUMBER 2 | AUGUST 2014
147
perspec tives ones we rely on when faced with few or no options to treat a patient; most antibacterial drugs used today were approved on the basis of noninferiority trials or less sophisticated approaches used several decades ago. Many of the guidance documents we have issued since 2008 have included specific recommendations to improve the scientific quality of noninferiority clinical trials. Although it is always valuable to consider new approaches, it is also important not to discount established methods that can provide scientifically sound evidence, particularly when there are critical patient needs for new antibacterial drugs. Our experience has shown that welldesigned noninferiority clinical trials can reveal significant unanticipated findings. Valuable lessons learned from some of these clinical trials include a better understanding of the biology of the disease, important considerations about the activity of the drug at the site of infection, variations in physiology in the sick population that could potentially affect the pharmacokinetics of the drug, and dose-selection considerations. The interaction between daptomycin and surfactant was elucidated after a clinical trial did not demonstrate that daptomycin was noninferior to an active comparator in community-acquired pneumonia.2 Similarly, the clinical trials of doripenem in ventilator-associated bacterial pneumonia (VABP) and of tigecycline in hospital-acquired bacterial pneumonia/ VABP shed light on unanticipated limitations of these antibacterial drugs when treating patients with VABP.3,4 There are, however, potential pitfalls in noninferiority trials if they are not designed and conducted properly. The end point(s) in these trials should be well defined and reliable, and should be assessed at a time point at which treatment effect can be demonstrated. In the past few years, considerable progress has been made to define end points for antibacterial clinical trials. The FNIH has made important contributions to clinical trial designs and end points for acute bacterial skin and skin structure infections and community-acquired bacterial pneumonia, and has also helped to provide a path forward for development 148
of new antibacterial drugs.5 The end points developed through this process are being used by pharmaceutical sponsors in ongoing trials and were also used in completed acute bacterial skin and skin structure infections trials that were recently discussed at an Anti-Infective Drugs Advisory Committee meeting.6 Although superiority trials can provide clear evidence of efficacy of an antibacterial drug, demonstrating superiority of a new antibacterial drug over existing active antibacterial therapies is often very challenging. A patient’s infection and response to an infection are influenced not only by the effects of the antibacterial drug and the infecting bacteria but also by host factors such as immunity, inflammation, and tissue repair. Also, if demonstrating superiority is dependent on having a less effective comparator because resistance to existing therapies has developed, then we are in a situation where drug development (despite stewardship and infection-control efforts) has not kept up with emerging resistance such that we can now enroll patients without adequate treatment options in a clinical trial—a situation most would prefer to avoid. In this scenario, to conduct such a trial requires a sufficient prevalence of the microorganism(s) for which we do not have adequate treatment options at present. Once a new therapy becomes available, ongoing trials designed to show superiority over a standard-of-care regimen (a regimen likely to be suboptimal) will probably become unethical and may need to be stopped. Moving forward, trials would probably be designed as active controlled trials to show noninferiority against the new standard for care. Thus, the window of opportunity during which superiority trials against suboptimal comparator are ethical is likely time-limited. Ideally, antibacterial drug development should be ongoing to provide treatment options before we have a population of patients for whom there are few or no effective drugs because of the emergence of new resistance mechanisms. Recognizing that antibacterial drug development has not kept pace with the development of resistance to available therapies and patient needs, we have been working on recommendations for streamlined drug
development programs for antibacterial drugs to treat patients with serious infections who have few, if any, treatment options. The FDA has published a draft guidance on developing antibacterial therapies for patients with unmet medical need that describes streamlined approaches to developing antibacterial drugs for patients with few or limited treatment options (e.g., due to resistance to existing therapies).7 This guidance describes the characteristics of new antibacterial drugs appropriate for such development pathways such as those that operate via different mechanisms of action, are paired with resistance inhibitors, or have chemical modifications to their structures conferring stability in the setting of selected resistance mechanisms. These streamlined programs are intended to provide a pathway for development of antibacterial drugs for patients with the greatest need. Because these development programs are streamlined and approval would be based on smaller databases, indications for use would be limited to situations in which a patient has few or no therapeutic options. The rationale for accepting a more streamlined development program is consistent with Title 21 of the Code of Federal Regulations (312 Subpart E), which describes expediting the development, evaluation, and marketing of new therapies intended to treat persons with life-threatening and severely debilitating illnesses, especially when no satisfactory alternative therapy exists. With such development programs, there is recognition that physicians and patients are willing to accept greater risks or side effects from drugs that treat life-threatening and severely debilitating illnesses than for drugs that treat less serious illnesses. For drugs approved with a smaller safety database, significant safety findings may first become apparent in the postmarketing period. It is critical that adequate steps be in place to monitor for safety signals or concerns regarding efficacy that may become apparent once a drug is marketed. In addition to the efforts in the area of clinical trial designs, Title VIII of the Food and Drug Administration Safety and Innovation Act (FDASIA), “Generating Antibiotic Incentives Now” (GAIN),
VOLUME 96 NUMBER 2 | AUGUST 2014 | www.nature.com/cpt
perspec tives is intended to encourage development of treatments for serious or life-threatening infections caused by bacteria or fungi. The term “qualified infectious disease product” (QIDP) refers to an antibacterial or antifungal human drug that is intended to treat serious or lifethreatening infections. For certain drugs designated as QIDP, GAIN provides an additional five years of exclusivity. An application for a drug designated as a QIDP is also eligible for priority review and designation as a fast-track product.8 The FDA has granted 47 QIDP designations representing 33 different antibacterial or antifungal drugs since the FDASIA was enacted in 2012. Antibacterial drug development is challenging on many fronts. Although significant progress has been made in the past few years on designing scientifically sound and feasible clinical trials, more work remains to be done to make new therapies available to meet patient needs. CONFLICT OF INTEREST The authors declared no conflict of interest. © 2014 ASCPT
1. US Food and Drug Administration. Guidance for industry: non-inferiority clinical trials. (March 2010). 2. Pertel, P.E. et al. Effects of prior effective therapy on the efficacy of daptomycin and ceftriaxone for the treatment of community-acquired pneumonia. Clin. Infect. Dis. 46, 1142–1151 (2008). 3. US Food and Drug Administration. Doribax (doripenem): drug safety communication— risk when used to treat pneumonia on ventilated patients (6 March 2014). 4. US Food and Drug Administration. FDA drug safety communication: FDA warns of increased risk of death with IV antibacterial Tygacil (tigecycline) and approves new boxed warning (27 September 2013). 5. Talbot, G.H. et al. Progress on developing endpoints for registrational clinical trials of community-acquired bacterial pneumonia and acute bacterial skin and skin structure infections: update from the Biomarkers Consortium of the Foundation for the National Institutes of Health. Clin. Infect. Dis. 55, 1114– 1121 (2012). 6. US Food and Drug Administration. March 31, 2014: Anti-Infective Drugs Advisory Committee meeting (12 February 2014). 7. US Food and Drug Administration. Guidance
for industry: antibacterial therapies for patients with unmet medical need for the treatment of serious bacterial diseases (July 2013). 8. Federal Register. Establishing a list of
qualifying pathogens under the Food and Drug Administration Safety and Innovation Act (12 June 2013).
See article page 239
Use of Internet Search Logs to Evaluate Potential Drug Adverse Events S Sarntivijai1 and DR Abernethy1 Internet search logs provide an abundant source of data that can be explored for purposes such as identifying drug exposure– adverse event relationships. The methodology to rigorously conduct such evaluations is not well characterized, and the utility of such analyses is not well defined. In this issue, White and colleagues propose an approach using Internet search logs for this purpose and compare it to parallel analyses conducted using the US Food and Drug Administration’s spontaneous reporting database. It is gratifying to see that various approaches are under evaluation to improve detection and interpretation of clinical adverse drug events. At present, a mainstay for postapproval monitoring for adverse drug events is the US Food and Drug Administration’s Adverse Event Reporting System (FAERS).1 The strengths and weaknesses of the FAERS have been discussed extensively. Weaknesses include under- or overreporting of adverse events as a result of the voluntary nature of the system and the burden of completing reports. The latter may be less of an issue with Internet search logs, but that is unknown at this time. It should be noted that a FAERS report contains considerable detail about the patient, the drug, the underlying clinical
condition, comorbid illness and concurrent drug exposures, and the temporal relationship between drug exposure and adverse event. This information would be lacking from Internet search log data. When a disproportionately strong association between an adverse event and a drug exposure is noted during ongoing analysis of FAERS data, it can be viewed as generation of a hypothesis. Such a finding leads to careful examination of other data sources, the published literature, and the information in the reports themselves. Evaluation of the reports can lead to a reevaluation of the association if it becomes clear that there are duplicate reports or there is evidence of stimulated reporting (e.g., medical–legal activity, heightened awareness due to public-
1US Food and Drug Administration, Silver Spring, Maryland, USA. Correspondence: DR Abernethy (darrell. [email protected])
doi:10.1038/clpt.2014.115
Clinical pharmacology & Therapeutics | VOLUME 96 NUMBER 2 | AUGUST 2014
149
