The most common cause of genetic kidney disease is the potentially life-threatening autosomal dominant polycystic kidney disease (ADPKD). The course of ADPKD is progressive, leading to kidney failure and the need for dialysis or kidney transplantation. Approximately 70% of patients with ADPKD require renal replacement therapy (RRT) by age 65. The psychosocial well-being and lifestyle of patients with ADPKD and related complications that include cardiovascular, hepatic, digestive, and neurologic disease, as well as symptoms that include pain, are severely impaired.
During the past two decades, there has been an increase in the number of randomized trials in patients with ADPKD. The choice of outcomes of trials in that patient population is challenging due to the lengthy natural history of the disease. Trials in the ADPKD population have often focused on surrogate outcomes such as kidney volume and function. Detailed empirical evidence of outcome reporting in trials is not available in the context of ADPKD. Bénédicte Sautenet, PhD, and colleagues and members of the SONG-PKD (Standardized Outcomes in Nephrology—Polycystic Kidney Disease) Initiative, conducted a systematic review to assess the range and variability of outcome domains and measures reported in trials of ADPKD. Results of the review were reported in the American Journal of Kidney Diseases [2020;76(2):213-223].
The researchers searched MEDLINE, EMBASE, the Cochrane Kidney and Transplant Specialized Register, Australian and New Zealand Clinical Trials Registry, the European clinical trials register, and ClinicalTrials.gov to identify all randomized controlled trials in patients with ADPKD up to October 1, 2019. For trials that included patients without ADPKD, only trials with at least 50% of patients with ADPKD were eligible to be included in the analysis.
Nine trial characteristics were extracted for each eligible trial: first author, year of publication, participating countries, sample size, mean age of participants, study duration, intervention type, primary outcome, and all discrete outcome measures. A discrete outcome measure was defined as any measurement or event reported separately for all trial arms.
A total of 68 trials involving 10,750 participants were identified. Of those, 36 (53%) were published, one (1%) was a protocol, and 31 (46%) were unpublished registered trials. The publication/registration year ranged from 2001 to 2019. Median duration of the trials was 18 months, and median sample size was 50 participants. Of the 68 trials, 86% (n=59) investigated a pharmacologic intervention, 5% (n=3) evaluated a surgical intervention, and 9% (n=6) examined a dietary intervention.
A total of 1413 outcomes were reported across the 68 trials. The median number of outcome measures (a different measurement, aggregation, metric, and time point) was 11 per trial. Exclusive of time points, the median number of outcome measures was seven per trial. The analysis classified the 1413 outcome measures into 97 outcome domains: 42% (n=41) were surrogate; 31% (n=30) were clinical; and 27% (n=26) were patient-reported outcomes.
The top six most frequently reported outcomes were: (1) kidney function, n=54, 79% of trials; (2) kidney and cyst volumes, n=43, 63% of trials; (3) blood pressure, n=27, 42% of trials; (4) proteinuria and albuminuria, n=22, 32% of trials; (5) unspecified adverse events, n=20, 29% of trials; and (6) pain (kidney, abdominal, and epigastric), n=16, 24% of trials. In <20% of trials, 90 outcome domains were reported. Mortality and kidney failure requiring RRT were reported in 9% (n=6) and 12% (n=8) of the trials, respectively.
Of the 41 surrogate outcomes, the three most frequent were kidney function (n=54; 79% of trials), which had 46 outcome measures (131 when timepoints were included); kidney and cyst volume (n=43; 63% of trials), which had 52 outcome measures (88 when timepoints were included); and blood pressure (n=28; 41% of trials), which had 30 outcome measures (65 when time points were included).
Of the clinical outcomes, the three most frequently reported were infection (n=10; 15% of trials), which had 21 measures (33 including time points); cardiovascular event (n=9; 13% of trials), which had six different outcome measures (14 including different time points); and end-stage kidney disease (n=8; 12% of trials), which had five (17 including time points) outcome measures.
Of the 26 patient-reported outcomes, the three most frequently reported were pain (kidney, abdominal, and epigastric; n=16; 24% of trials), which had 26 outcome measures (42 including time points); pain (other; n=11; 16% of trials), which had 11 (18 including time points) outcome measures; and diarrhea/constipation/gas/bloating (n=10; 15% of trials), which had nine (16 including time points) outcome measures.
Of the 68 total trials, 56 specified a primary outcome; of those, 47 specified a single primary outcome, and nine specified multiple primary outcomes. In 45 of the 47 studies specifying a single primary outcome, the majority of the reported outcomes were surrogate, compared with one with a clinical primary outcome and one with a patient-reported primary outcome. The two most frequently reported single primary outcomes were kidney and cyst volumes (n=17) and kidney function (n=13).
The researchers cited some limitations to the analysis results, including only assessing outcomes measures in the top three domains in each category, excluding non-English trials, and not analyzing all the reported measures.
In conclusion, the authors said, “The outcomes reported in ADPKD trials are varied, heterogeneous, and focused on surrogate outcomes related to progression to kidney failure requiring kidney replacement therapy (KRT). Clinical and patient-reported outcomes such as KRT requirement and pain are seldom reported. The development of a core outcome set for ADPKD will improve outcome reporting, which will be of relevance and importance to all stakeholders in clinical trials and ultimately will strengthen trial-based evidence to inform decision making in ADPKD based on the outcomes that matter to patients, caregivers, and health professionals.”
- Researchers conducted a systematic review of randomized controlled trials among patients with autosomal dominant polycystic kidney disease (ADPKD) to assess the range and variability of reported outcomes.
- The outcomes in ADPKD trials were both broad and highly variable.
- The three top outcome domains were surrogate, clinical, and patient-reported outcomes. Surrogate outcomes were the most widely reported.