[MUSIC] Welcome everyone to trial management and advanced operations. My name is Sheriza Baksh and I am an assistant scientist at the Center for Clinical Trials and Evidence Synthesis at Johns Hopkins University. In this lecture we will be discussing protocol events, how to minimize them, how to detect them and how to report them during the course of a trial. A protocol event can be an adverse event, serious adverse event also known as SAE or an unanticipated problem. We also classify protocol deviations and violations under this heading, each of these events has a potential impact on participants safety, and therefore require careful consideration in light of the potential benefits of the study results. After this lecture, you should leave with a better understanding of the following key points. Protocol events do not necessarily have to be associated with participants safety or data integrity for them to be recorded. Depending on the seriousness and impact on participants safety and ethical standards, there are different reporting mechanisms and timelines to both regulatory entities and trial oversight bodies such as an IRB. Categorization of protocol events are not limited to one type either. It is entirely possible that a single event could fall into two or more categories requiring two or more avenues for reporting. Data collection for protocol events are not limited to those that are related and unexpected. While investigators may know that a particular event is unrelated, future data, external data even may prove that to be untrue. Likewise, collecting data on expected protocol events can help us to better understand whether we are seeing an expected event more frequently than it would normally be expected within this particular population. This brings us to our final point regarding aggregating protocol event data. By analyzing safety data in a pre planned systematic manner, we can detect whether or not the risks inflicted on trial participants are greater than the intended benefits at these interim analyses. There are a few guiding principles behind the need to collect these types of data. The first of these is that research has a different objective than regular clinical care. While investigators are administering clinical care through the course of a clinical trial, they're doing so in a highly controlled environment with an investigational product that may or may not have been tested in this population. Because of the nature of the situation, enhanced safety monitoring is required. This is often in the form of active surveillance. As the name suggests, active surveillance involves study investigators ascertaining safety events through periodic data collection. This can include questionnaires administered to participants or through clinical assessments. Additionally, it is important to allow for the ad hoc data collection of adverse events throughout the trial. The some of these methods leads to a very comprehensive understanding of the participant risk throughout the trial. While developing questionnaires to collect these data, one common question investigators often have is whether or not to create pre populated list of expected adverse events. Their analytical benefits to doing so. However, one must allow for the collection of all adverse events on these forms. Additionally, if investigators are working with particularly vulnerable populations such as pregnant women, young children or older adults, data collection forms with pre populated adverse event list should be designed with this in mind by actively assessing participants for important adverse events in these specific populations. Study procedures must also thus be designed with these populations in mind in order to reduce the potential harm to those particular participants. On another note, safety data collected within the scope of a trial contributes to the overall knowledge base for a particular study intervention. Data collection of them should be done with the notion that these data may be used for post hoc analyses after the trial has concluded. Such as within a systematic review or regulatory oversight for particular chemical entity. Therefore, standardization of adverse event terms is integral to this process. One of the reasons we collect safety data throughout the course of the trial and not simply at the end, is to conduct interim analyses. These may be done at the behest of a regulatory agency or as part of a predefined analyses for as data safety monitoring board. With these aggregated data and related analyses, results important decisions around the conduct of the study can be made. These data are key to deciding whether or not a trial should continue or whether protocol changes are needed. Remember that we should be aiming to have clinical equipoise throughout the trial and not just at the onset. Part of this process is this continual risk benefit assessment. Study investigators must continually ask themselves whether or not the potential benefits of an intervention outweigh the risk reflected in the safety data that they are collecting. This must also be taken in light of external emerging evidence, including new therapeutics that may enter the market as I alluded to earlier. Safety monitoring is a continual process. There are many reasons for this, including required expedited reporting for some events and continual risk of benefit assessments. Because of this, it is important to design your study in a way that makes this type of data collection less burdensome and more transparent. There are many aspects of study design and study conduct to consider in the continual risk benefit assessment for trial. At the onset of this process, investigators must ensure that their protocols and standards of procedures are clear and interpretable in order to avoid protocol deviations. Protocol deviations have the potential to lead to safety events. Study investigators may elect to have a medical monitor for the trial. In such cases, they must also have explicit instructions for how medical monitors may adjudicate safety events and the course of action for when one does occur. Piloting this process may also be helpful for a group of investigators who may not have previously worked together when writing the protocol. It is also critical that the study team identifies study procedures and activities with the potential to harm the participants. They must set up risk mitigating procedures and enhanced safety surveillance around those activities. Finally, it is important to recognize that not all study risks will be known in the design phase. And that in order to comply with regulatory and ethical reporting requirements, study investigators must design data collection instruments to accommodate these. Protocol events are not just limited to those that occur during study visits. The study team should have procedures in place for detecting events between study visits in a systematic and or an ad hoc manner. This can be accomplished through telephone check ins instructing study participants to contact the study team in the event of anything adverse or by asking them about the interim period at each study visit. Again the goal here is to balance the risks and benefits between treatment arms throughout the study as much as possible. In order to do so, safety data collection must be continuous from the time of consent to study close out. Lastly, the study team should ensure that the data collection instruments are such that they are collecting all of the data needed for reporting protocol events. This includes setting up procedures to obtain those data that are not regularly collected. Protocol events are not mutually exclusive. As you see here in this simple Venn diagram, there can be overlap between unanticipated problems, adverse events and serious adverse events. Because of the study teams should be aware of reporting timelines relevant for their study. When outlining the procedures for data collection and form completion and finalization for each of these types of events, as there may be multiple avenues for reporting. We will now shift into a discussion of the different types of protocol events. The first and probably the most recognized of these is adverse events. One of the international standard setting bodies for regulatory procedures is the International Conference on Harmonization. They have developed a definition for an adverse event that has been adapted and adopted by national regulatory agencies around the world. Their definition states, that an adverse event is any untoward medical occurrence in a patient or clinical investigation subject, administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment. It is important to note that this definition was developed to cover both regulatory approved and unapproved products, despite being written for regulatory context. This definition has also been applied to the study of clinical procedures and behavioral interventions. Some regulatory bodies make a distinction between adverse events and adverse reactions. The key difference is that causality is assumed with adverse reaction, but it's not required for an adverse event. In an abundance of caution IRBs and regulatory agencies may request reporting of adverse events, especially for unapproved products. Adverse event data are collected without regard to causality, severity, relatedness, or expectedness. This is in light of these data being collected through active surveillance in a research context as opposed to a clinical one. Study investigators have a responsibility to ensure that their study is not inflicting undue harm to participants, and one of the ways of addressing this is assessing whether or not it is being done through data collection. Finally, adverse event data may be dose dependent so to speak. There may be situations within a trial where we see that adverse events are associated with a particular dose or intensity for an intervention. The only way to assess this would be through active safety surveillance. Adverse events have a few qualifying terms attached to them, one of these is relatedness. An adverse event can be related, possibly related or not related, and you see those definitions here on the slide. If an event is considered related, it is assumed to be associated with the intervention. An adverse event that is possibly related may have a temporal relationship with the administration of study treatment, meaning it occurred a logical time after administration. But there is room for doubt that the adverse event is purely due to the intervention. If the adverse event is clearly not related to the intervention for instance, a participant with a known biology in a study is looking at the effects of vitamin C on acne. Let's say this person got an anaphylactic response from a bee sting, then the event is still recorded but it's noted as not related. Another criterion for categorizing adverse events is expectedness. Determining whether or not an adverse event is expected or not should be considered within the context of both the intervention and the condition under study. For instance, in a study comparing one behavioral therapy to standard of care among Alzheimer's patients, a stroke in a participant would be considered expected. Because of the higher risk among Alzheimer's patients, even though it may not be expected with the intervention or the standard of care. As with any study trial or otherwise, it is important to have a clear understanding of the condition under study, whether it is related to your knowledge of the intervention or not. Adverse event data are often collected using standardized forms. These forms may be rooted in certain dictionaries such as metra or CTCAs. Metra is typically used for regulatory purposes and CTAs is commonly used in cancer trials. Some dictionaries may have a hierarchical structure allowing for the analysis of safety data at different levels of aggregation. Similarly, they may group adverse event terms for particular patient population or condition together. In the case of metra,, these are referred to as SMQs, standardized metro queries. Again, the goal of these categorization techniques is to facilitate analysis and allow for interpretable and comparable data across trials. This can be especially helpful to regulatory agencies who may see trials of the same chemical entity in different formulations. At the trial level however, quality assurance procedures for data collection should include a check of standardization of adverse of terms, to ensure that they are being collected, and data entered uniformly across trial participants at every clinical site. This includes mundane checks such as spelling to more consequential ones such as terminology. By standardizing these terms and matching to an adverse event dictionary, investigators are enabling the possibility of cross communication of safety data with other trials that may be using different adverse event dictionaries. One translating tool was developed by Odyssey, the observational health data sciences and informatics group called Atlas. While developed for observational data, this tool is helpful in translating across adverse event dictionaries. Oftentimes it may be necessary to collect information on the grade of the adverse event. This is typically an indicator for the severity of the event and is scaled from 1-5. Some reporting bodies may request the reporting of higher grade adverse events only. Hence, it is important to assess this at the time of data collection. Grading can be interpreted as a rough threshold for reporting however, severity is not the same as seriousness. Seriousness of an event is used to determine whether or not expedited reporting is necessary. For instance, a participant may have had a debilitating migraine. However, that may not necessarily be classified as serious and therefore would not warrant expedited reporting despite the severity of the particular symptom. We will now shift over to serious adverse events. The US Food and Drug Administration, defines a serious adverse event as one that is serious and should be reported to FDA when the patient outcome is death, life threatening, hospitalization. Either initial or prolonged disability or permanent damage, congenital anomaly or birth defect. Required intervention to prevent permanent impairment or damage such as with devices, or other serious important medical events. That last qualifier does leave some wiggle room for investigator judgement, and this is an effort at greater sensitivity in reporting. Again, this definition is similar for many other regulatory agencies in the US NIH. Serious adverse events require expedited reporting. So it is important to set up study procedures that allow for rapid data collection and adjudication regarding the event. This may mean contacting the participant or their emergency contact for additional information about what occurred. With this added data collection procedure it is important to maintain privacy protections in both the collection and reporting of these data. There may be subsequent follow up needed as well to assess if or how the essay was resolved These should also be recorded on the data collection instruments. Finally, it may be necessary to unmask the medical monitor or certain members of the investigative team while collecting data around the SA. It is important to document these unmasking is and the extent to which they occurred. The final type of protocol event is unanticipated problems. This is a bit of a catchall for everything that study investigators could not anticipate in the planning and design of the trial. When conducting a trial in the United States. The Office of Human Research Protection OHRP is the standard setting body for this type of event. They have defined it as any incident experience or outcome that meets all of the following criteria. Unexpected in terms of the nature severity of frequency related or possibly related to participation in the research. And this can mean the study procedures, the study intervention or simply being part of that population. As well as suggesting that the research places participants or others at a greater risk of harm. And this can mean anything from physical, psychological, economic or social harm than was previously known or recognized. This is intended as a catch all for administrative, ethical and clinical breaches within the context of the study as it relates to participants safety. It may have implications or causes that are outside the scope of the trial as well. So it's important to note that there may be unanticipated problems that occur at the institutional level that do affect participants in your study. There may be certain circumstances when it may be difficult to determine whether a particular event is unexpected. And whether it is related or possibly related to the clinical trial in the event that an event meets those criteria. The Office of Human Research Protection provides some suggested methods for addressing the event to avoid any future occurrences. Generally in the reporting of unanticipated problems a corrective plan is also included. Some strategies include protocol changes that are implemented immediately and this can even be before IRB approval in an effort to mitigate any participant risk. There could also be modification of the inclusion and exclusion criteria to reduce any of the newly identified risks. There might be added monitoring activities to those that are already in existence. Investigators might also choose to pause enrollment until there's further investigation about what occurred, why it occurred and how it occurred. They could also choose to suspend part or all of the research procedures that are currently experienced by the enrolled subjects. The investigators could also modify the informed consent documents to reflect this new information about the perceived risk. And finally, the investigators could provide additional information about these new risks to those who are already enrolled. Before we go on I wanted to take a minute to take a note on protocol deviations and protocol violations. These two terms are often used interchangeably, but there is a bit of a difference between the two. A protocol deviation is when an event diverges from the IRB approved protocol and this is generally not considered to be serious. Protocol violation, on the other hand, is quite serious, and it is also a divergence from the IRB protocol. And this is something that could potentially affect the data quality or integrity might even invalidate what you have on your informed consent documents. Or it might impact participant safety or their general protections. Protocol deviations and violations may also overlap with the other protocol events that we have discussed. They may or may not be subject to expedited reporting if they result in an essay. So be very careful about how you collect these data as well. [MUSIC]
