Karen Ivester

Developing novel, safer treatments that may be curative for many individuals living with cancer depends not only on continued use of existing products but on the clinical and regulatory success of the newest treatments-including promising developments focused on targeted/immunotherapy combinations and immune checkpoint blockade therapies which are demonstrating that immunity is the key to long-term responses. Rigorous evaluations in clinical trials to assess efficacy and safety in patients are critical to the development of these highly sensitive targeted/immunotherapy combinations. New molecular entity (NME) selection, protocol development, patient population, and principal investigator (PI) and site selection are key areas in which to focus to establish a foundation for the successful execution of an early phase oncology trial.

	A substantial number of NMEs move through Phase I into Phase II; however, progression from Phase I through approval each year is very low. In 2014, approximately 64% of drugs moved from Phase I to Phase II and 10.4% moved from Phase I through approval.

	From 2005 through 2013, FDA’s Center for Drug Evaluation and Research (CDER) has averaged approximately 25 novel new drug approvals per year. These include drugs for all diseases and all indications. In 2014, 41 novel new drugs were approved-six in total for oncology. These drugs were approved under the FDA accelerated approval program, which allows early approval of a drug for serious or life-threatening illnesses that offer benefit over current treatment. Once accelerated approval is granted, these drugs must undergo additional testing. These recent approvals in oncology were based on a “surrogate endpoint” (e.g., a laboratory measure) or other clinical measure considered to predict the clinical benefit of a drug.

	Given the cost of drug development (which now exceeds $2.5 billion

), the selection of those molecules that have the highest potential for success is crucial. There is more at stake than the financial cost-we must consider the patient population, the PIs and the sites. They are all finite and the demands placed upon them are seriously impacting the future of clinical trials-especially early phase clinical trials.

Protocol development and optimization of design 

	Nearly 60% of protocols are amended during the trial, according to the 

Tufts Center for the Study of Drug Development

. In order to reduce or avoid costly protocol amendments, oncology sponsors must view early phase protocol development holistically and assist our sponsors in optimizing their protocol development. Important questions to consider include: 

		Has the early work been done (toxicology, animal studies, targeted starting dose established, appropriate formulation and manufacturing stability and scalability evaluated)? Performing a gap analysis can assist the client in identifying potential issues early on; therefore, an early evaluation by regulatory can be of added value.

		Has the sponsor identified biomarkers for the mechanism of action (MOA)? Has the patient population been selected based on these biomarkers and the MOA? Are assays validated?

		What is the turnaround time for any procedures or assessments and how will this impact patient enrollment? We must keep in mind these patients have been diagnosed or their disease has progressed and may also be aggressive. Asking them to wait four to six weeks may not be acceptable for them or for their treating surgical or medical oncologist.

		Are all of the protocol-defined procedures and assessments appropriate for collecting data that will support a new drug application or investigational medicinal product dossier? Is all of the information critical in a Phase I or Phase IIa (proof of concept, efficacy, or mechanism of action study) where the intent is to inform early go/no-go decisions? If it isn’t critical to inform the decision (e.g., not a critical variable) and not critical for safety, then there is a need to provide a rationale for collecting the variable, entering it into the data collection system and monitoring it. Significant costs lie in the collection of unnecessary information in early phase clinical research and this is an area where protocol and electronic case report forms (eCRFs) can be optimized and improved considerably. As sponsors, researchers, and contract research organizations (CROs) gain expertise in early phase research, this will greatly improve and reduce sponsor and CRO costs as well as reduce site burden in data collection.

	First-in-man studies for many candidate chemotherapies are constructed to identify the maximum tolerated dose and dosing schedule. Yet, technologies have yielded investigational agents that are designed to act with greater precision to inhibit cancer cell growth or promote cancer cell death. For sponsors of these newer targeted molecular agents, trial protocols may require an optimal biological dose endpoint rather than a more traditional maximum tolerated dose (MTD) endpoint. Consequently, the protocol will need to clearly define how to determine the recommended Phase II dose, and describe new assays or procedures to

 measure biologic endpoints, as well as to capture traditional patient safety assessments.

	The investigational brochure (IB) contains the information that will assist and guide the regulatory and safety advisory committees in assessing the risk/benefit of the NME. Early compound knowledge can also assist in the most critical variables to collect regarding safety, thereby reducing the collection of unnecessary data.

	Thoughtful design of an early stage trial protocol can help characterize biomarkers that will facilitate appropriate patient enrollment in follow-on advanced trials. Remember that most of the oncology drugs approved in 2014 were approved based on a surrogate endpoint or a predictor of clinical benefit.

Importance of adaptive design in early phase clinical trials

	Utilizing pharmacokinetic/pharmacodynamic (PK/PD) to guide dose escalation decisions and adaptive designs that enable adjustments to the study design and/or specific patient population as the trial progresses may increase the speed of the dose escalation and reduce patient exposure to doses that are not effective, as traditional designs often start with a dose well below animal toxicity. This lowest dose has no effect and the traditional method doesn’t allow reaching higher doses quickly.

	Interest in adaptive design study methods arises from the belief that these methods hold promise for improving drug development compared to conventional study design methods (such as 3 + 3 designs). Adaptive design approaches may lead to a study that provides the same information, but more efficiently, increases the likelihood of success, or provides more information regarding the drug’s effect, which may also lead to more efficient follow-on studies.

	The more progressive adaptive design algorithms permit a change in dose level after each patient is treated based on the accumulated responses of previously enrolled subjects. These algorithms lead to more dose-level changes, both increases and decreases of the dose, as the algorithm selects an exposure for each subject to the dose that will contribute the greatest amount of information towards the ultimate conclusion. By permitting escalation after each individual subject if that subject did not have a DLT, it is possible to reach the middle or higher end of the dose-response curve with fewer subjects at each of the prior levels.

	Adaptive designs allow for completing the study more rapidly than the traditional sequential fixed-size cohort design. CROs can assist sponsors in exploring the features of different study designs with regard to the balance of efficiency (study size) and subject safety. Study simulations with multiple combinations of escalation criteria, dose-step size, and hypothetical assumptions around relationships of exposure to severity and frequency of adverse events (AEs) may be useful in evaluating different designs. These simulations can assist in assessing the risks and selecting a design that offers improved efficiency without increasing risk excessively.

	Adaptively designed studies that enroll patients who are most likely to benefit could finish faster and consume fewer resources, which could yield economies in time to development, as well as cost and reduced burden on PIs and sites.

	Finally, in assessing the protocol development, is the imaging, procedures, and assessments in line with the standard of care (SOC) for the patient population, the disease indication, and the country/site in which the clinical trial is being conducted? This can vary significantly and, prior to site selection, feasibility, and the use of prescribing data can help determine the most appropriate country/site mix. Keeping imaging and disease assessments SOC will decrease costs and minimize regulatory delays from radiation committees at both the country and site levels.

Patient selection in early phase clinical trials

	Novel approaches to patient/subject selection can be used to ensure we are selecting the patients most likely to benefit from the NME. “Genotyping” tumors from patients is paving the way for targeted therapies for people living with cancer. The translational research and the technical capacity to screen large numbers of tumors have taken years and significant collaboration between oncologists and pathologists. Molecular profiles and tumor typing has identified the genetic abnormalities that activate and drive tumor growth. Understanding cancer development at the molecular genetic level, identifying mutations, and creating NMEs that target them are significantly improving outcomes for patients with lymphoma, breast, brain, GI, and lung cancers, as well as other indications. The process of extracting and purifying DNA and genotyping it using sophisticated software and assays can screen for hundreds of mutations that have been identified and linked with tumor growth.

	The identification of genetic mutations in tumors has been critical in the development of multiple treatments in oncology and now serves as the basis for personalized, targeted therapies as we have seen in adaptive clinical trials such as the I-SPY 2 TRIAL. This clinical trial was designed to treat patients with breast cancer, and the patients are assigned to treatment options (of which there are many in a single trial) based on the molecular characteristics (or biomarker signatures) of their disease.

	The genotype and mutations within specific tumors and indications are driving the patient population for targeted therapies. These innovative, genomically targeted therapies often provide good initial responses. For example, drugs that target a specific BRAF gene mutation in melanoma can shrink the tumors in about half of the patients. This approach has resulted in frequent, short-lived responses for multiple targeted therapies.

	Resistance develops when tumors have multiple genomic defects that drive the disease. After the targeted therapy knocks out one driver, another driver can take over and activate tumor growth again. To combat resistance and relapse, cancer immunotherapy has found a role in combination with genomically targeted therapies. This immune checkpoint blockade therapy has resulted in an approach that treats the immune system which is capable of recognizing distinctive features of cancer cells and launching T-cells that target and shut down tumor-specific antigens at the peptide level. The first immune checkpoint blockade, ipilimumab (Yervoy®), has been approved for melanoma. A second immune checkpoint inhibitor showed that pembrolizumab (Keytruda

) is also effective in the treatment of melanoma, and the drug was approved in 2014.

	Collaboration between researchers who focus on targeted therapies and researchers who focus on immune checkpoint therapies will likely result in the development of targeted/immunotherapy combinations which will have “curative potential.”

	Patient selection, down to the genetic mutation level, is impacting early phase clinical trials in ways not previously anticipated. The institutions that have the capability to utilize genotyping, in mass, will be at an advantage to quickly 

identify patients with tumors that match the novel therapies in these clinical trials. As adaptive designs expand and we learn more about specific therapies and combination therapies for multiple indications, there will be more I-SPY 2-type clinical trials in which patients have their tumor genotyped initially and are then given combination(s) of treatment developed specifically for their disease.

	Currently, this means sites will need to identify patients for clinical trials that have these specific mutations. In reality, this translates to a lower enrollment rate in some instances, especially if there are rare or multiple genetic mutations in the targeted tumor types or indications. It becomes very important to research and learn more about the occurrence of each of the genetic mutations in various oncology indications in order to plan for the number of sites required to enroll the study.

	Working with feasibility teams to research the indication, frequency of mutation, and specific line of therapy for each therapy or combination therapy will be critical to the success of early phase oncology clinical trials as the targets become more specialized. While challenging, the potential for effective, long-lasting treatment outcomes in multiple indications is a reality.

	With the NME identified, a well-designed protocol in place and the patient population selected, attention turns to the selection and activation of appropriate clinical sites. Historical site data, specifically site enrollment patterns with similar oncology indications, are critical to choosing experienced and qualified sites. Knowledge of a site helps determine which facilities have reliable PIs and clinical research staff that both understand and can “bring to life” the complexities of Phase I clinical trial protocols, including:

		Recruitment of niche patients, often with specific genetic mutations/alterations

		Management of DLTs and participation in dose-escalation decisions

		Collecting, processing, and analysis of PK/PD samples

		Extensive biological specimens are collected, genotyped, and analyzed

		Commitment to timely data entry and query resolution

	Site efficiencies can also be created when a sponsor or contracted CRO is familiar with each site’s institutional contracting procedures, scientific and ethics review board practices, and document requirements. Detailed knowledge of local trial site compliance with federal, local, and its own institutional regulations to protect and care for human subjects is critical.

	Finally, the use of document exchange portals can accelerate overall trial timelines and increase efficiencies without sacrificing quality or endangering regulatory compliance.

Principal investigator burden and impact on clinical trials

	While the number of NMEs and clinical trials are increasing, the number of PIs are declining and many are withdrawing from clinical research altogether. The number of investigators has fallen significantly since 2008 and there is a high turnover rate among those filing 1572s.

	Thirty-five percent of investigators in the U.S. are not returning to conduct another clinical trial since initially submitting a 1572 in 2006. The numbers of investigators not returning to conduct clinical studies are even higher in other countries, as reflected below:

	The reasons given are system and organization, time involvement, resources, lack of clinical or scientific rationale for the research, lack of interest in the research topic, complexity of trials, excessive trial costs not covered by the sponsor, and disruption to clinical practice.

	Some of these barriers, such as ethics submissions, are essential; however, many of the system, organization, and other barriers are under the direct control of the sponsor and the Clinical Research Organization (CRO) partner.

	The most burdensome tasks identified by PIs and sites were:

		Completing contractual and regulatory documents

		Getting paid for clinical trial work on time

		Reporting serious adverse events (SAEs)

		Interacting with remote and on-site monitors

Articles

How to meet the rigorous safety and efficacy demands critical to evaluating newer targeted cancer therapies.

Overcoming Early Phase Oncology Challenges

	Developing novel, safer treatments that may be curative for many individuals living with cancer depends not only on continued use of existing products but on the clinical and regulatory success of the newest treatments-including promising developments focused on targeted/immunotherapy combinations and immune checkpoint blockade therapies which are demonstrating that immunity is the key to long-term responses. Rigorous evaluations in clinical trials to assess efficacy and safety in patients are critical to the development of these highly sensitive targeted/immunotherapy combinations. New molecular entity (NME) selection, protocol development, patient population, and principal investigator (PI) and site selection are key areas in which to focus to establish a foundation for the successful execution of an early phase oncology trial.

. In order to reduce or avoid costly protocol amendments, oncology sponsors must view early phase protocol development holistically and assist our sponsors in optimizing their protocol development.

	For example, has the early work been done (toxicology, animal studies, targeted starting dose established, appropriate formulation and manufacturing stability and scalability evaluated)? Performing a gap analysis can assist the client in identifying potential issues early on; therefore, an early evaluation by regulatory can be of added value.

	Has the sponsor identified biomarkers for the mechanism of action (MOA)? Has the patient population been selected based on these biomarkers and the MOA? Are assays validated?

	What is the turnaround time for any procedures or assessments and how will this impact patient enrollment? We must keep in mind these patients have been diagnosed or their disease has progressed and may also be aggressive. Asking them to wait four to six weeks may not be acceptable for them or for their treating surgical or medical oncologist.

	Are all of the protocol-defined procedures and assessments appropriate for collecting data that will support a new drug application or investigational medicinal product dossier? Is all of the information critical in a Phase I or Phase IIa (proof of concept, efficacy, or mechanism of action study) where the intent is to inform early go/no-go decisions? If it isn’t critical to inform the decision (e.g., not a critical variable) and not critical for safety, then there is a need to provide a rationale for collecting the variable, entering it into the data collection system and monitoring it. Significant costs lie in the collection of unnecessary information in early phase clinical research and this is an area where protocol and electronic case report forms (eCRFs) can be optimized and improved considerably. As sponsors, researchers, and contract research organizations (CROs) gain expertise in early phase research, this will greatly improve and reduce sponsor and CRO costs as well as reduce site burden in data collection.

	First-in-man studies for many candidate chemotherapies are constructed to identify the maximum tolerated dose and dosing schedule. Yet, technologies have yielded investigational agents that are designed to act with greater precision to inhibit cancer cell growth or promote cancer cell death. For sponsors of these newer targeted molecular agents, trial protocols may require an optimal biological dose endpoint rather than a more traditional maximum tolerated dose (MTD) endpoint. Consequently, the protocol will need to clearly define how to determine the recommended Phase II dose (RP2D), and describe new assays or procedures to measure biologic endpoints, as well as to capture traditional patient safety assessments.

	Patient selection, down to the genetic mutation level, is impacting early phase clinical trials in ways not previously anticipated. The institutions that have the capability to utilize genotyping, in mass, will be at an advantage to quickly identify patients with tumors that match the novel therapies in these clinical trials. As adaptive designs expand and we learn more about specific therapies and combination therapies for multiple indications, there will be more I-SPY 2-type clinical trials in which patients have their tumor genotyped initially and are then given combination(s) of treatment developed specifically for their disease.

	This leads to a lower proportion of experienced sites and a high turnover rates among new PIs. The resulting impact for sponsors is higher operational costs, including substantially higher site start-up costs in areas of site selection, qualification, and training.

	What can sponsors and CROs do to assist PIs and sites? It is important to:

		Streamline start-up activities (GCP training, contracting, essential document collection)

As cancer immunotherapies continue to emerge, the need for rigorous evaluation to assess the safety and efficacy of these products in clinical trials is critical. Doing so in the early phase setting requires a foundational focus in such areas as NME selection, protocol development, patient population, and investigator and site selection.