Gilead Sciences Leverages ePRO on TERISA Trial of 1st Antianginal Drug for Type 2 Diabetes

Gilead wanted to assess the efficacy of Ranolazine compared to placebo on reducing average weekly angina frequency in subjects with type 2 diabetes mellitus, coronary artery disease, and chronic stable angina who remain symptomatic despite treatment with one or two antianginal agents.

Gilead also wanted to evaluate the efficacy of Ranolazine compared to placebo on average weekly frequency of sublingual nitroglycerin (SL NTG) use.

Study Design

Eligible subjects entered a four-week, single-blind placebo run-in period and were provided a mobile eDiary from PHT Corporation which had built-in prompts for entry of daily angina episodes, daily number of sublingual nitroglycerine used, and a daily dyspnea score (on a scale from 1 to 5). Data collected on the eDiary was used in support of the primary endpoint for the trial and to assess efficacy.

Overview

Eight million people in the U.S. alone suffer from chronic stable angina, the chest pain and tightness associated with myocardial ischemia caused by coronary artery disease (CAD), yet many patients remain undertreated. Those with type 2 diabetes mellitus and CAD present an especially challenging group because their disease is often more extensive and difficult to manage. The TERISA trial, whose objective was to evaluate the antianginal efficacy of ranolazine (Ranexa), utlized the eDiary to collect patient reported outcomes (PROs).¹

Key Facts

• Ranolazine was proven effective as an antianginal medication.
• The eDiary was used to collect three daily questions about the subjects’ angina episodes and symptoms.
• The eDiary was used during screening to capture data that would determine a subject’s eligibility for randomization. Subjects were randomized based on the number of angina episodes recorded in the eDiary, and also had to achieve at least 85% compliance with.
• The eDiary enabled data collection including mid-study changes in 14 countries on three continents.
• Participants were 98% compliant with the eDiaries, including 97% compliance from people aged 75 and older.

Trial Goals

In previous trials, Ranolazine had been proven as an effective antianginal treatment in patients with clinically manifest CAD, both as monotherapy and in combination with other commonly prescribed medications as add-on therapy.

The TERISA study aimed to extend these findings in a specific subpopulation by demonstrating a reduction in weekly angina frequency and a decrease in sublingual nitroglycerin use in type 2 diabetes patients who also suffered from CAD and chronic angina, and who remained symptomatic despite treatment with one or two other therapies.

Selection of an ePRO Mobile Device vs. Paper

Previous Ranolazine studies used exercise treadmill parameters as the primary endpoint. While effective and easily verifiable, attaining the data was costly, provided numerous logistical challenges, and bore questionable clinical relevance, especially at the level of the individual patient. Other possible endpoints such as number of angina episodes and sublingual nitroglycerin use imparted concerns about using patient diaries to collect endpoint data.

Paper diaries have proven to be inaccurate although it is difficult to measure actual protocol compliance without time stamps. One study showed paper compliance to be as little as 11% when reported compliance was 90%.² There also is the potential, known colloquially as the “Parking Lot Syndrome,” for a patient to complete multiple entries in a paper diary just before their appointment, rather than daily for a stretch of time. Paper diaries can also be observed and analyzed only after collection and data are extracted, offering there little chance to asses a study while it is still ongoing.

For the scope of the TERISA trial, at 104 different sites across 14 countries in North America, Europe, and Asia diaries would have to be translated into numerous local dialects and languages. Diaries would need to be shipped globally, raising concerns about logistics costs and management.

Alternatively, using an electronic diary also prompted concerns. Many chronic angina sufferers are well above 50 years old. It was unclear if they would be able to operate a handheld electronic device or read the prompts on an electronic screen. Electronic diaries could also raise issues if software or trial updates were needed, or support issues if a device malfunctioned.

Solution

An eDiary system was chosen because it eliminated diary, data and logistics concerns. Observing the patient-reported number of angina episodes and sublingual nitroglycerin use over a period of time follows the increasing regulatory emphasis on patient-reported outcomes (PROs). It also provides a significantly more cost-effective trial design than using an exercise treadmill endpoint.

The eDiary eliminated concerns created by a paper diary around data quality. eDiary completion times are logged and verified, enabling accurate data collection and contemporaneous compliance. The eDiary used in the TERISA trial resulted in 98% compliance of all 927 trial participants. There was 97% compliance among the 101 trial participants who were 75 years of age or older. The eDiary was designed to include daily reminders to encourage patient completion. The TERISA trial also featured retroactive “look-back” entries for patients to complete the previous day if it had been missed, but only for the past 24 hours – a feature not feasible in paper designs. The eDiary allowed for mid-study parameter changes to be implemented without any impact on data collection or accuracy. Devices were translated into local languages. Frequent data transmission allowed for real-time upload to the eDiary vendor’s servers and integration and transmission to the Medidata Rave eCRF database. This reduced data clarification forms and increased the speed of the time from last patient, last visit (LPLV) to final data transfer.

Lessons Learned

Compliance – Of the 77,968 possible entries, 76,418 (or 98%) of available diaries were completed. Despite concerns about elderly populations, in this study and others, this population demonstrates high and stable compliance.³ Further, high diary compliance can also be attributed to the study design which included randomization criteria of at least 85% compliance over a three week screening and the ability to enter episodes which occurred over the last 24 hours. Lastly to ensure compliance, if a diary had not been transmitted since the previous evening, email alerts were sent to sites and monitors.

Increased Event Rate – Prompting a patient to respond daily can generally lead to more accurate data, since the data is captured in real-time. Studies in oncology have shown that capturing patient symptoms on a daily basis provides a more complete and accurate representation of the patient symptom profile than the intervallic collection of data.⁴

Retroactive Diary Design – This element enabled additional data collection with a limited window of recall to the past 24 hours. Data showed 0.02% duplicated data entries. Future designs would prevent this possibility.

eDiary Software Update Timing – In some instances, sites obtained automatic downloads for software upgrades prior to IRB approval. Due to a protocol amendment affecting eligibility criteria, the vendor had to manage the timing manually so that sites received new study software only after IRB was approved for the protocol amendment.

Conclusions

Gilead Sciences successfully completed their TERISA trial and collected the needed regulatory-compliant patient-driven primary endpoint data. The result:

• An efficacious trial to support Gilead Sciences’ work to benefit angina sufferers
• Confirmation that eDiaries are an effective and economical way to capture angina endpoints such as episodes and dyspnea in angina patients vs. treadmill or paper diaries
• Evidence that global patients including elderly patients demonstrate high and stable compliance with eDiary completion at 98%
• Retrospective entry of up to 24 hours can increase compliance
• Upgrades and mid study changes need to be timed carefully with IRB approvals

Chart Text:
The frequency distribution of angina episodes collected electronically on handheld devices showed a normal (Gaussian) distribution with an R2 of 0.999. An important aspect of this finding is the bell shaped symmetrical frequency distribution curve is characteristic of many real world physical and biological phenomena and is known to describe random data variation.

1 Kosiborod M et al, JACC 2013 May 21;61(20):2038-45.

2 Stone, et al., 2002 “Patient non Compliance with Paper Diaries” British Medical Journal, BMJ2002;324:1193

3 Basch, E., M. McCabe, A. Iasonos, D. Schrag, H. I. Scher, R. Farquhar, J. Speakman, P. Fearn, D. Artz, M. G. Kris, A. Culkin, and A. Barz. “Long-Term Toxicity Monitoring Via Electronic Patient-Reported Outcomes In Patients Receiving Chemotherapy.”
Journal of Clinical Oncology 25, no. 34 (2007): 5374-5380. Gary ST, Otero AV, Dallabrida SM, “Effective Use of Electronic Patient Reported Outcomes (ePRO) in Breast Cancer.” PHT Corporation, March 2013. In press.

4 Basch, Ethan, Alexia Iasonos, Tiffani McDonough, Allison Barz, Ann Culkin, Mark G Kris, Howard I Scher, and Deborah Schrag. “Patient Versus Clinician Symptom Reporting Using The National Cancer Institute Common Terminology Criteria For Adverse Events: Results Of A Questionnaire-based Study.” The Lancet Oncology 7, no. 11 (2006): 903-909.
Basch, E., M. G. Kris, G. Heller, X. Jia, D. Schrag, A. Culkin, S. Goldfarb, T. Atkinson, A. Iasonos, M. Appawu, M. Fruscione, L. Sit, and A. Barz. “Adverse Symptom Event Reporting By Patients Vs Clinicians: Relationships With Clinical Outcomes.” JNCI Journal of the National Cancer Institute 101, no. 23 (2009): 1624-1632.