Sex, Permanent Drug Discontinuation, and Study Retention in Clinical Trials

Background: Women are underrepresented across cardiovascular clinical trials. Whether women are more likely than men to prematurely discontinue study drug or withdraw consent once enrolled in a clinical trial is unknown. Methods: Eleven phase 3/4 TIMI (Thrombolysis in Myocardial Infarction) trials were included (135 879 men and 51 812 women [28%]). The association between sex and premature study drug discontinuation and withdrawal of consent were examined by multivariable logistic regression after adjusting for potential confounders in each individual trial and combining the individual point estimates in random effects models. Results: After adjusting for baseline differences, women had 22% higher odds of premature drug discontinuation (adjusted odds ratio [OR adj ], 1.22 [95% CI, 1.16–1.28]; P <0.001) compared with men. Qualitatively consistent results were observed for women versus men in the placebo arms (OR adj , 1.20 [95% CI, 1.13–1.27]) and active therapy arms (OR adj , 1.23 [95% CI, 1.17–1.30)]; there was some evidence for regional heterogeneity ( P interaction <0.001). Of those who stopped study drug prematurely, a similar proportion of men and women in the active arm stopped because of an adverse event (36% for both; P =0.60). Women were also more likely to withdraw consent compared with men (OR adj , 1.26 [95% CI, 1.17–1.36]; P <0.001). Conclusions: Women were more likely than men to prematurely discontinue study drug and withdraw consent across cardiovascular outcome trials. Premature study drug discontinuation was not explained by baseline differences by sex or a higher proportion of adverse events. Future trials should better capture reasons for drug discontinuation and withdrawal of consent to understand barriers to continued study drug use and clinical trial participation, particularly among women.

Abstract 15140: Older Participant Perspectives on Permanent Study Drug Discontinuation in an Ongoing Primary Prevention Trial of Statins: A Qualitative Study

Introduction: Study drug discontinuation is commonplace in clinical trials of older populations and poses a major challenge to trial investigators. Little is known about why older participants discontinue the study drug. This study aimed to understand factors contributing to permanent study drug discontinuation among participants aged ≥70 years within an ongoing primary prevention trial of statins by tapping into their experiences and perceptions. Methods: Trial participants who had permanently discontinued the study drug within 2 years of randomization were purposively sampled based on age (<75 and ≥75 years) and sex to participate in semi-structured phone interviews between March 2019 and February 2020. Interviews were audio-recorded, transcribed and analyzed thematically. Results: Thirty participants took part (21 females; mean age, 77 years) and three themes were identified from the data. Perceived adverse events (AEs) and their impact on daily living (mobility, functional capacity, quality of life, etc.) were identified as the major factors leading to the participants permanently discontinuing their study drug. Muscle symptoms were the most commonly reported AE. Selected participant quotes which describe symptoms and their impact are presented in Table. For some, a challenging life circumstance further lowered their tolerance to the perceived AEs thus making discontinuation more likely. A few discontinuations were attributed to other factors (e.g. concerns about possible side effects, GP advice, unrelated illness). Conclusion: Among healthy older participants enrolled in a statin trial, perceived AEs and their related impact were key factors contributing to the permanent study drug discontinuation. Addressing anticipated participant-reported AEs and their concerns about drug-related side effects at trial entry and offering timely medical assistance and support when AEs occur may be useful to reduce drug discontinuation rates.

684. Cardiac Safety in Adults with Community-Acquired Bacterial Pneumonia (CABP) Treated with Lefamulin (LEF) or Moxifloxacin (MOX): Analysis of Lefamulin Evaluation Against Pneumonia (LEAP) 1 and LEAP 2 Study Results

Background Preclinical data suggest potential effects of LEF on cardiac interval parameters. We therefore assessed LEF cardiac safety from the LEAP 1/2 trials. Methods In LEAP 1, PORT III–V patients received LEF 150mg IV q12h for 5–7 days or MOX 400mg IV q24h for 7 days, with optional IV-to-oral switch (600mg LEF q12h or 400 mg MOX q24h). In LEAP 2, PORT II–IV patients received oral LEF 600mg q12h for 5 days or oral MOX 400mg q24h for 7 days. Patients with known QT prolongation or on medication with potential to prolong the QT interval were excluded as per MOX label. After 5 minutes of rest in the supine position, triplicate 12-lead ECGs were obtained within a 5-minute interval at Screening in both studies, on Days 1/3 in LEAP 1 (predose and ≤15 minutes after first IV dose), and on Days 1/4 in LEAP 2 (predose and 1–3 hours after first oral dose), and sent to a central ECG reader for adjudication. Results Of 1,282 randomized/treated patients (n = 641/group), 1,274 had baseline (BL) and post-BL ECG data (n = 636 LEF, n = 638 MOX). Consistent with the resolution of infection, ECGs revealed mean reductions of 7–8 beats/minute for both groups in both studies. The largest mean change in QTcF from BL to post-BL was on Day 3 in LEAP 1 (13.6 and 16.4 msec with IV LEF and MOX, respectively) and on Day 4 in LEAP 2 (9.3 and 11.6 msec with oral LEF and MOX, respectively). The proportion of patients meeting potentially important post-BL QTcF values/changes was comparable between treatment groups (table). In the standardized MedDRA query of Torsade de pointes/QT prolongation (broad), the most common treatment-emergent adverse event was ECG QT prolonged (n = 4 LEF, n = 5 MOX). All events were nonserious and mild or moderate in severity. 6 events were considered study drug related (n = 4 LEF, n = 2 MOX). 5 events led to study drug discontinuation (n = 2 LEF, n = 3 MOX). In 2 patients with cardiovascular disease, 1 had ventricular arrhythmia on Day 20 (18 days after last LEF dose) and 1 had cardiac arrest on Day 18 (9 days after last MOX dose); both events were fatal and considered unrelated to study drug by investigator. Conclusion Mild prolongation of the QTcF interval was seen with LEF and MOX, with somewhat smaller effects seen with LEF. Given the small effect, LEF is unlikely to pose a clinically significant risk of ventricular proarrhythmia with appropriate precautions and use. Disclosures All authors: No reported disclosures.

1374. Integrated Safety Summary of Omadacycline: A Novel Aminomethylcycline Antibiotic

Background Omadacycline (OMC) is a novel aminomethylcycline with activity against Gram-positive, many Gram-negative, anaerobic, and atypical pathogens. It is in clinical development as once-daily oral (PO) and intravenous (IV) monotherapy for acute bacterial skin and skin structure infections (ABSSSI) and community-acquired bacterial pneumonia (CABP). Cumulative safety results from Phase 3 clinical trials are reported. Methods This pooled safety analysis is based on 2,150 subjects: OASIS-1 (N = 645), OASIS-2 (N = 735) in ABSSSI; OPTIC (N = 770) in CABP. Comparators were linezolid (LZD) 600 mg IV then PO in ABSSSI (n = 689); moxifloxacin (MOX) 400 mg IV then PO in CABP (n = 388). Safety parameters included treatment-emergent adverse events (TEAEs), laboratory evaluations, vital signs, and electrocardiogram (ECG) findings. Results A total of 1,073 subjects received OMC: 705 received OMC IV then PO (ABSSSI, n = 323; CABP, n = 382); 368 received OMC PO only for ABSSSI. Overall, 60.6% were male and 91.6% white; mean age ranges were 44.7–45.1 and 60.9–62.1 years in ABSSSI and CABP studies, respectively. TEAEs were observed in 47.5% (OMC), 41.2% (LZD), and 48.5% (MOX) of subjects, with gastrointestinal events the most common TEAEs. Serious TEAEs were low (3.6% OMC, 1.9% LZD, 6.7% MOX). Nausea (14.9% OMC, 8.7% LZD, 5.4% MOX) and vomiting (8.3% OMC, 3.9% LZD, 1.5% MOX) were the most frequently reported TEAEs. Diarrhea was observed in 2.4% OMC, 2.9% LZD, and 8.0% MOX subjects, with no cases of Clostridium difficile in OMC-treated subjects. Most TEAEs were mild to moderate and did not result in study drug discontinuation (3.1% OMC, 1.5% LZD, 7.0% MOX); 4 OMC, 1 LZD, and 0 MOX subjects discontinued for nausea and vomiting. Frequency of hepatic TEAEs was similar for OMC, LZD, and MOX: 4.3% OMC, 4.1% LZD, and 4.5% MOX subjects had post-baseline ALT &gt;3× upper limit of normal. Vital signs and ECGs had comparable clinically notable values post-baseline in each treatment group. Known tetracycline class adverse events such as fungal infections were similar in all groups. Conclusion Pooled analyses demonstrate a favorable OMC safety profile, consistent with its tetracycline heritage. OMC was generally well tolerated in subjects with ABSSSI and CABP, with infrequent treatment discontinuations. Disclosures T. M. File Jr., BioMerieux: Consultant, Consulting fee; Curetis: Consultant, Consulting fee; Melinta Therapeutics: Consultant, Consulting fee; Merck: Consultant, Consulting fee; Motif Bio: Consultant, Consulting fee; Nabriva Therapeutics: Consultant and Investigator, Consulting fee and Research grant; Paratek Pharmaceuticals: Consultant, Consulting fee; Pfizer: Consultant, Consulting fee. T. Van Der Poll, Paratek Pharmaceuticals: Consultant, Consulting fee. P. McGovern, Paratek Pharmaceuticals: Employee, Salary. E. Tzanis, Paratek Pharmaceuticals: Employee, Salary.

Comparison of Time-to-First Event and Recurrent-Event Methods in Randomized Clinical Trials

Background: Most phase-3 trials feature time-to-first event end points for their primary and secondary analyses. In chronic diseases, where a clinical event can occur >1 time, recurrent-event methods have been proposed to more fully capture disease burden and have been assumed to improve statistical precision and power compared with conventional time-to-first methods. Methods: To better characterize factors that influence statistical properties of recurrent-event and time-to-first methods in the evaluation of randomized therapy, we repeatedly simulated trials with 1:1 randomization of 4000 patients to active versus control therapy, with true patient-level risk reduction of 20% (ie, relative risk=0.80). For patients who discontinued active therapy after a first event, we assumed their risk reverted subsequently to their original placebo-level risk. Through simulation, we varied the degree of between-patient heterogeneity of risk and the extent of treatment discontinuation. Findings were compared with those from actual randomized clinical trials. Results: As the degree of between-patient heterogeneity of risk increased, both time-to-first and recurrent-event methods lost statistical power to detect a true risk reduction and confidence intervals widened. The recurrent-event analyses continued to estimate the true relative risk (0.80) as heterogeneity increased, whereas the Cox model produced attenuated estimates. The power of recurrent-event methods declined as the rate of study drug discontinuation postevent increased. Recurrent-event methods provided greater power than time-to-first methods in scenarios where drug discontinuation was ≤30% after a first event, lesser power with drug discontinuation rates of ≥60%, and comparable power otherwise. We confirmed in several actual trials of chronic heart failure that treatment effect estimates were attenuated when estimated via the Cox model and that increased statistical power from recurrent-event methods was most pronounced in trials with lower treatment discontinuation rates. Conclusions: We find that the statistical power of both recurrent-events and time-to-first methods are reduced by increasing heterogeneity of patient risk, a parameter not included in conventional power and sample size formulas. Data from real clinical trials are consistent with simulation studies, confirming that the greatest statistical gains from use of recurrent-events methods occur in the presence of high patient heterogeneity and low rates of study drug discontinuation.

Abstract 078: Implications of Patient-Reported Treatment Satisfaction for Discontinuation of Rivaroxaban versus Warfarin in ROCKET-AF

Background: Patient-reported outcomes and satisfaction are important in both trials and clinical practice and may be associated with treatment adherence. Methods: ROCKET-AF was a randomized, double-blind trial of rivaroxaban versus warfarin for prevention of thromboembolism in patients with atrial fibrillation. In a substudy, we compared treatment satisfaction scores: Anti-Clot Treatment Scale (ACTS) and Treatment Satisfaction Questionnaire for Medication version II (TSQM II). Patient-driven discontinuation included stopping study drugs due to withdrawal of consent, non-compliance, or loss to follow-up. Rates of discontinuation were calculated for participants above and below the median scores for each scale. Results: Of 14,264 patients in ROCKET AF, 1,181 (8.3%; median age 75 years; 34% women) patients completed both the ACTS and TSQM II questionnaires 4 weeks after starting the study drug. Over a median follow-up of 1.6 years, 450 premature study drug discontinuations occurred, 116 (26%) patient-driven. Patients less satisfied with treatment by the ACTS Benefits and Burdens and TSQM II scales had higher rates of study drug discontinuation ( Table ). Conclusions: Patient-reported satisfaction was lower in patients with study drug discontinuation, suggesting that collecting patient-reported outcomes early in clinical trials may guide interventions that improve adherence and clinical outcomes.

Therapeutic care in metastatic renal cell carcinoma during the follow-up phase of the RECORD-1 phase III trial

e17531 Background: Following drug discontinuation for progression or adverse event in a clinical trial for relapsed or stage IV kidney cancer, supportive care including surgery, palliative radiotherapy, or bisphosphonates continue to be recommended by National Comprehensive Cancer Network (NCCN). However, published data on active therapeutic agents given to patients following study drug discontinuation in recent clinical trials is limited. Methods: World Health Organization Anatomical Therapeutic Chemical codes or therapeutic names, captured from the follow-up phase in a phase III clinical trial (RECORD-1) of patients with metastatic renal cell carcinoma (mRCC) patients, were used to describe antineoplastic therapies following discontinuation of study drug. Prior to trial, patients had progressed on at least one VEGFr-TKI therapy. Results: Of the 130 patients with follow-up after discontinuation of study drug, 78.5% received at least one of the following: corticosteroids, radiotherapy, protein kinase inhibitors, mTOR inhibitor, pyrimidine analogues, monoclonal antibodies, interferons, and investigational drugs. Among patients who received an active agent, nearly three-quarters (73.5%) utilized targeted therapy (protein kinase inhibitors, mTOR inhibitor, monoclonal antibodies). Conclusions: In a clinical trial setting with mRCC patients who have received several classes of systemic therapy, care delivered following study drug discontinuation often includes an active antineoplastic agent, despite the limited supportive evidence in this setting. While the placebo control with supportive care in a double-blind phase is acceptable to evaluate the efficacy and safety of a therapy for regulatory approval purposes, decision makers must also consider how these data may inform comparisons with the usual alternatives available to and used by physicians and patients in the non-trial setting. [Table: see text]