ELIPSIS: developing tools to better understand VOC in SCD

In this issue of Blood, Pittman et al report on their use of an elegant and innovative at-home study design to prospectively follow patients with sickle cell disease (SCD) over 6 months. The study, Evaluation of Longitudinal Pain Study in Sickle Cell Disease (ELIPSIS), showed that patients could be monitored at home with an electronic patient-reported outcome (ePRO) tool, a mobile-based blood collection team, and a wearable actigraphy device to monitor activity and sleep.¹  Many clinical trials focus on vaso-occlusive crisis (VOC), but this study explored novel end points and used a home-based study design to address the difficulty in accurately capturing patient experiences over the long term.

One the largest previous efforts to track and understand pain was the Pain in Sickle Cell Epidemiology Study (PiSCES) performed by Smith et al.²  This 6-month study used paper-based diaries and asked participants to record daily pain levels, medication use, and health care use. PiSCES is widely referenced because of the impact of finding that 79% of VOCs are treated at home and, strikingly, only 3.5% of participants reported crisis with health care use. The recognition of the significant number of patients with pain at home has led to increased efforts to use innovative techniques to measure a patient’s pain experience at home.

Along with the advent of the digital era, there has been an increase in the number of ePRO tools to provide real-time data on patient experiences and avoid concerns with paper diaries such as contemporaneous completion (also known as the “parking lot effect,” in which responses are completed in the parking lot just before a visit).^3,4  Importantly, the US Food and Drug Administration has endorsed electronic capture of clinical trial source data, including PRO end points to improve documentation of source data in clinical trials.⁵  ELIPSIS provides details on the use of an ePRO developed on the basis of data from PiSCES, during which participants were asked to record daily pain, VOCs, medication use, and health care use. Interestingly, a mobile phlebotomy team was triggered when VOC was indicated. Although compliance in completing the daily entries decreased over the course of the study, participants completed on average 67% of daily entries. Similar compliance rates have been found in numerous ePRO studies.^5,6  Ultimately, Pittman et al found that the majority of VOC days (78.3%) and events (62.3%) recorded by patients were self-treated at home. These findings are similar to those in PiSCES and further validate the emphasis on understanding and monitoring pain at home.

One of the most intriguing aspects of this study is the use of a mobile-based phlebotomy team. Blood samples were performed in the participant’s home at baseline and repeated every 3 weeks until completion of the study. If participants reported a VOC pain crisis on their ePRO device, a series of VOC blood collections was also scheduled. This emphasis on decreasing the logistical difficulties of participating in a clinical trial is important and places the burden on the clinical research team instead of the patient. In addition to continuing to support patients through a variety of means such as transportation and meals, this focus on a patient-centric approach to clinical studies is critical.

Laboratory findings from ELIPSIS reinforced expected changes during VOC events compared with baseline, including increases in white blood cell adhesion, percentage of monocyte-platelet and neutrophil-platelet aggregates, inflammation markers, and coagulation markers. Several studies have previously documented changes during VOC events compared with baseline.^7,8  It is important, however, to appreciate that the investigators established a more robust baseline biomarker level for this comparison, because VOC events and pain are unpredictable. Using multiple blood draws throughout the course of the study when participants were not experiencing a VOC (via the mobile phlebotomy team) allowed for a more reflective baseline level as opposed to a single sample drawn at baseline.

Finally, the team integrated the use of a wearable actigraphy device (ActiWatch 2) to objectively acquire data on participants’ activity and sleep patterns. In the age of proliferating consumer and research-grade wearable devices, more attention is focused on leveraging the passive acquisition of physiological data through wearables. Recent studies have found that it is feasible to use wearables to monitor patients,⁹  and that wearables have the potential for their data to predict outcomes such as pain.¹⁰  Pittman et al note that average daily activity was expectedly lower during health care use on VOC days compared with non-VOC days. Patients visiting the emergency department or those who have been admitted to the hospital are expected to be less active. However, daily activity during non-VOC days and any VOC days (Figure 4B in the Pittman et al study) clearly shows differences in average activity during hour 13 to hour 23, corresponding to afternoon and evening hours. This granular decrease in activity during specific times of the day with VOC provides another window into the patient experience at home.

It is encouraging to see the level of detail in the study by Pittman et al regarding use of an ePRO, a mobile-based phlebotomy team, and wearable devices. The investigators set out to evaluate tools and potential end points for clinical trials in SCD. In doing so, they have demonstrated not only feasibility, but also the importance of a patient-centric home-based clinical study design and use of innovative strategies to gather data as well as help participants in clinical studies.