David T. Gilbertson

Background and Aims: The ASCEND-D/ID trial reported safety and efficacy of daprodustat in patients receiving dialysis [1,2]. Rates and types of clinical outcomes vary between incident and prevalent dialysis patients. This study assessed events, of which most were described as safety events in ASCEND-D/ID trial, among incident dialysis patients on erythropoiesis-stimulating agents (ESAs) and prevalent dialysis patients in the US Medicare population. Method: This is a retrospective, observational cohort study using United States Renal Data System data. The study included incident dialysis-dependent (IDD) patients from 2017–2019 and, separately, prevalent dialysis-dependent (PDD) patients as of January 1, 2018. For IDD patients, the index date was the dialysis initiation date; for PDD patients, the index date was January 1, 2018. To establish the cohorts, we further required patients to have at least 6 continuous months of Medicare Parts A and B coverage prior to, and be aged ≥18 years on, the index date. For IDD patients, we also required ESA use on the index date to better support comparability to the PDD patients. Patients with a history of kidney transplant or who had cancer during the year prior to, or who had a hospitalization for heart failure (HF), myocardial infarction (MI), or a stroke in the 4 weeks before, the index date were excluded. Patients were followed from index date (for IDD patients, follow-up started from index date plus 60 days to ascertain dialysis modality stability) to the earliest date of death, loss of Medicare coverage, kidney transplantation, or December 31, 2019. Events assessed were cardiovascular (CV: all-cause death, HF, MI, stroke, thromboembolic events [TEE], major adverse CV events [MACE: composite of death, MI, stroke], and MACE+HF) and non-CV (cancer, gastric erosions, ocular events, seizures, serious infection) events, derived from medical claims in the follow-up period using ICD-10 diagnosis, Healthcare Common Procedure Coding System, and Current Procedural Terminology codes in the follow-up period. HF, MI, stroke, and serious infection events were defined as inpatient events. TEEs included vascular access thrombosis (VAT), deep venous thrombosis (DVT), and pulmonary embolism (PE). First event rates were calculated separately for IDD and PDD cohorts and expressed as number of events per 100 patient-years. Results: The IDD cohort contained 23,909 patients; the PDD cohort, 230,055. Median follow-up time was 12.6 months for the IDD cohort and 24.0 months for PDD cohort. Mean ages of the IDD and PDD cohorts were 70.3 and 63.3 years, respectively. Compared with the PDD cohort, patients in the IDD cohort were more likely to be female (48.2% vs 43.7%), of non-Hispanic White race (61.6% vs 40.4%), less likely be of lower-income status (33.0% vs 47.4% for Medicare/Medicaid dual enrollment and 38.6% vs 57.0% for Part D low-income subsidy) and with higher proportion of comorbidities. First event rates are shown in the Table for the IDD and PDD cohorts. Compared with the PDD cohort, patients in the IDD cohort had higher rates of all-cause mortality (17.3 vs 16.8 per 100 patient-years), stroke (1.8 vs 1.4), VAT (109.4. vs 67.2, hemodialysis patients only), and MACE (20.9 vs 20.3); however, they had lower rates of MI (3.1 vs 3.5), HF (7.1 vs 8.9), DVT (3.2 vs 4.0), and MACE+HF (25.6 vs 26.9). For non-CV events, patients in the IDD cohort had lower event rates for nearly all events except malignancies. Conclusion: ESA-treated IDD patients had higher event rates for some CV events but lower rates for most non-CV events compared with PDD patients. As expected, however, rates of VAT were higher in IDD patients than PDD patients. These background event rates in real-world populations with wide ESA use provide context when considering potential alternative treatments for anemia. Limitation of this descriptive study includes the reporting of crude (unadjusted) rates, and therefore should be interpreted with caution. Funding: GSK (study 217316).

Background and Aims: The safety and efficacy of daprodustat in non-dialysis-dependent chronic kidney disease (CKD) patients have been reported in the ASCEND-ND trial [1]. However, CKD stage may affect both the need for anemia treatment as well as outcomes, including potential safety events for anemia treatment. This study describes cardiovascular (CV) and non-CV event rates by CKD stage among patients treated with erythropoietin-stimulating agents (ESAs) in the US Medicare population. Method: Using Medicare data for all fee-for-service enrollees, we conducted a retrospective cohort study of patients who had diagnosis of CKD stages 3, 4, or 5 and a first use of ESA in 2017–2019. CKD stage was identified by ICD-10 codes; ESA use was derived using Healthcare Common Procedure Coding System (HCPCS) codes. The date of first ESA use was the index date and CKD stage was defined as the most recent diagnosis on or before index date. We further required patients to have at least 12 continuous months of Medicare Parts A and B coverage (also used as a washout period to identify new ESA users) prior to, and be aged ≥66 years on, the index date. Patients with a history of kidney transplant or who had cancer during the year prior, and those who had a hospitalization for heart failure (HF), myocardial infarction (MI), or stroke in the 4 weeks before, the index date were excluded. Patients were followed from index date to the earliest date of death, loss of Medicare coverage, kidney transplantation or dialysis, or December 31, 2019. Both CV (all-cause death, HF, MI, stroke, thromboembolic events [TEEs], MACE [death, stroke, MI], and MACE+HF) and non-CV (cancer, gastric erosions, ocular events, seizures, and serious infections) events were ascertained, derived from Medical claims using ICD-10 diagnosis, HCPCS, and Current Procedural Terminology codes in the follow-up period. HF, MI, stroke, and serious infection were defined as inpatient events. First event rates were calculated separately by CKD stage, expressed as number of events per 100 patient-years during the follow-up period. Results: This study cohort included 50,206 patients; 22,332 had stage 3, 19,177 stage 4, and 8,697 stage 5 CKD at the index date. Median follow-up time was 11.5 months for patients with CKD stage 3, 8.9 months for patients with stage 4, and 3.7 months for patients with stage 5 CKD. Patients with higher (worse) CKD stages were slightly younger (mean ages for stages 3, 4, and 5 were 79.7, 79.1, and 76.8 years); with increasingly more males (42.9%, 45.1%, and 48.5%), increasingly more individuals of non-Hispanic Black race (14.2%, 15.7%, and 19.6%), with lower income (indicated by Medicare/Medicaid dual enrolment and Medicare Part D low-income subsidy), and lower proportion of comorbidities. Rates were higher for most CV events in patients with higher (worse) CKD stages (Table): death rates were 19.5, 22.1, and 28.1 per 100 patient-years for stages 3, 4, and 5, respectively; MI rates were 2.2, 2.9, and 3.2; HF rates were 12.5, 17.2, and 19.0; MACE rates were 22.0, 25.4, and 31.8; and MACE+HF rates were 31.0, 38.7 and 47.8, respectively. Rates of stroke were similar across CKD stages; but no pattern for TEE. For non-CV events, higher (worse) CKD stage had higher rates of first ocular events (17.2, 23.3, and 27.1) and seizures (2.3, 2.7, and 3.4), but lower rates of cancer (21.8, 14.2, and 10.2) and, within this, hematological malignancies (10.4, 4.11 and 2.3). Conclusion: Medicare-covered ESA-treated patients with higher CKD stage had higher rates for most events assessed, while malignancy rates were lower. These findings may provide important context when assessing US event rates in non-dialysis-dependent CKD patients. Further investigation could clarify some of the differences between stages, such as rates of malignancy. Rates were reported as crude (unadjusted) and so should be interpreted with caution. Limiting the analysis to ESA-treated patients may have resulted in different event rates than would be expected to occur in the general CKD population. Funding: GSK (study 217316).

Background and Aims: The safety and efficacy of daprodustat in patients on dialysis were reported in ASCEND-D/ID trials [1,2]. Trials impose strict criteria for enrollment, making participants more narrowly selected than a broader population with disease who might benefit from treatment. We describe events, of which most were noted as safety events in ASCEND-D/ID trials, among general anemia “treatment eligible” (TE) patients on dialysis and patients similar to those in ASCEND-D/ID trials (“clinical trial-like” [CTL]) in the US. Method: Using United States Renal Data System data, we conducted a retrospective cohort study of incident dialysis-dependent (IDD) patients from 2017–2019 and, separately, point prevalent dialysis-dependent (PDD) patients as of Jan 1 2018. Dialysis initiation date was the index date for IDD patients and Jan 1 2018 for PDD patients. We also required patients to have ≥6 months of continuous Medicare coverage before, and be aged ≥18 years on, index date. Patients with a history of kidney transplant or who had cancer during the year prior, and those who had a hospitalization for heart failure (HF), myocardial infarction (MI), or stroke in the 4 weeks before, the index date were excluded. For IDD patients, we required use of ESAs at the index date to support comparability to the PDD patients. The CTL cohort was a subset of the TE cohort, approximating eligibility in the ASCEND-D/ID trials. Patients were followed from index date to the earliest date of death, loss of Medicare coverage, kidney transplantation, or Dec 31 2019. We assessed cardiovascular (CV: all-cause death, HF, MI, stroke, thromboembolic events [TEEs], major cardiovascular events [MACE: death, MI, stroke] and MACE+HF) and non-CV (cancer, gastric erosions, ocular events, seizures, serious infections) events, derived from medical claims in the follow-up period using ICD-10 diagnosis, Healthcare Common Procedure Coding System, and Current Procedural Terminology codes. HF, MI, stroke, and serious infections were defined as inpatient events. First event rates for TE and CTL cohorts were calculated separately for IDD and PDD patients and expressed as number of events per 100 patient-years (PYs). Results: Of IDD patients, 23,909 were in the TE and 6,962 in the CTL cohorts; corresponding numbers were 230,055 and 41,980 for PDD patients. Median follow-up was 12.6 and 13.8 months for the TE and CTL cohorts, respectively, in IDD patients; and 24 months for both TE and CTL cohorts in PDD patients. Mean ages were 70.3 and 71.0 years for the TE and CTL cohorts in IDD patients; and 63.3 and 64.1 years for the TE and CTL cohorts in PDD patients. For both IDD and PDD patients, the TE and CTL cohorts had similar demographics, but the TE cohorts had a slightly higher proportion of patients with comorbidities. First event rates are shown in the Table. TE cohorts had higher rates than CTL cohorts for most CV events (mortality, 17.3 vs 14.0 per 100 PYs in IDD and 16.8 vs 15.1 in PDD; MACE; 20.9 vs 17.5 in IDD and 20.3 vs 18.6 in PDD; MACE+HF, 25.6 vs 21.8 in IDD, 26.9 vs 25.1 in PDD). In IDD patients, HF was more common in the TE than the CTL cohort (7.1 vs 6.1) but was similar in PDD patients. MI, stroke, and TEEs (vascular access thrombosis [in hemodialysis patients only], deep venous thrombosis, and pulmonary embolism) had similar rates in the TE and CTL cohorts for both IDD and PDD patients. Serious infections were higher in the TE than the CTL cohort in both IDD and PDD patients. In PDD patients, gastric erosions were higher in the TE than the CTL cohort (40.0 vs 38.5) while ocular events and retinal hemorrhage were lower in the TE cohort. Conclusion: Rates for most CV events were higher in TE cohorts, comprised of patients eligible for anemia treatment, compared with CTL cohorts mimicking trial participants. Rates of non-CV events did not show a consistent pattern. As expected, rates of VAT were higher in IDD than PDD patients. These findings further contextualize real-world event rates of interest in dialysis populations. Reported rates were crude (unadjusted) and should be interpreted with caution. Funding: GSK (study 217316).

This is an Early Access article. Please select the PDF button, above, to view it.

Purpose: Secondary hyperparathyroidism (SHPT) is common among dialysis patients, and calcimimetics are a mainstay of treatment. This study assessed whether cinacalcet use is associated with gastrointestinal bleeding in a large hemodialysis cohort.Methods: A linked database of clinical records and medical claims for patients receiving hemodialysis in a large dialysis organization, 2007–2010, was used. A nested case–control study was performed among patients aged ≥18 years who had received hemodialysis for ≥90 days, had Medicare Parts A, B, and D coverage for ≥1 year, and had clinical evidence of SHPT (parathyroid hormone >300 pg/mL). Cases were those who experienced death or hospitalization caused by gastrointestinal bleeding. Each case was matched to up to four controls. Exposure was measured by any cinacalcet use, current use, past use, cumulative exposure days, and cumulative dosage. Conditional logistic models were used to assess the association.Results: Of 48 437 patients included, 2570 experienced gastrointestinal bleeding events (2498 non‐fatal, 72 fatal), and 2465 (2397 non‐fatal, 68 fatal) were matched to 9500 controls; 17.2% of cases and 15.8% of controls had cinacalcet exposure and 11.1% of both cases and controls had current use. The adjusted odds ratios (95% CI) of gastrointestinal bleeding for any use, current use, and past use of cinacalcet were 1.04 (0.91–1.19), 0.97 (0.83–1.13), and 1.22 (0.99–1.50), respectively, with no use as the reference.Conclusion: The results do not suggest an elevated risk of gastrointestinal bleeding resulting in hospitalization or death for hemodialysis patients exposed to cinacalcet.

Key Points: Variables most predictive of transfusion were previous transfusion, hemoglobin, ferritin, and length of hospitalization at baseline. Our transfusion prediction risk score performed well. It could be further developed into a clinically useful tool. The score could allow clinicians to identify hemodialysis patients most likely to benefit from an anemia treatment to avoid transfusions. Background: In patients on dialysis with anemia, avoiding red blood cell transfusions is preferable. We sought to develop and validate a novel transfusion prediction risk score for patients receiving maintenance hemodialysis. Methods: This retrospective cohort study used United States Renal Data System data to create a model development cohort (patients who were point prevalent and on hemodialysis on November 1, 2012) and a validation cohort (patients who were point prevalent and on hemodialysis on August 1, 2013). We characterized comorbidity, inflammatory conditions, hospitalizations, anemia and anemia management, iron parameters, intravenous iron use, and vitamin D use during a 6-month baseline period to predict subsequent 3-month transfusion risk. We used logistic least absolute shrinkage and selection operator regression. In an exploratory analysis, model results were used to calculate a score to predict 6- and 12-month hospitalization and mortality. Results: Variables most predictive of transfusion were prior transfusion, hemoglobin, ferritin, and number of hospital days in the baseline period. The resulting c-statistic in the validation cohort was 0.74, indicating relatively good predictive power. The score was associated with a significantly increased risk of subsequent mortality (hazard ratios 1.0, 1.22, 1.26, 1.54, 1.71, grouped from lowest to highest score), but not with hospitalization. Conclusions: We developed a transfusion prediction risk score with good performance characteristics that was associated with mortality. This score could be further developed into a clinically useful application, allowing clinicians to identify patients on hemodialysis most likely to benefit from a timely, proactive anemia treatment approach, with the goal of avoiding red blood cell transfusions and attendant risks of adverse clinical outcomes.

Background: Large-scale studies risk generating inaccurate and missing data due to the complexity of data collection. Technology has the potential to improve data quality by providing operational support to data collectors. However, this potential is under-explored in community-based trials. The Aspirin in reducing events in the elderly (ASPREE) trial developed a data suite that was specifically designed to support data collectors: the ASPREE Web Accessible Relational Database (AWARD). This paper describesAWARDand the impact of system design on data quality.Methods: AWARD’s operational requirements, conceptual design, key challenges and design solutions for data quality are presented. Impact of design features is assessed through comparison of baseline data collected prior to implementation of key functionality (n = 1000) with data collected post implementation (n = 18,114). Overall data quality is assessed according to data category.Results: At baseline, implementation of user-driven functionality reduced staff error (from 0.3% to 0.01%), out-of-range data entry (from 0.14% to 0.04%) and protocol deviations (from 0.4% to 0.08%). In the longitudinal data set, which contained more than 39 million data values collected withinAWARD, 96.6% of data values were entered within specified query range or found to be accurate upon querying. The remaining data were missing (3.4%). Participant non-attendance at scheduled study activity was the most common cause of missing data. Costs associated with cleaning data in ASPREE were lower than expected compared with reports from other trials.Conclusions: Clinical trials undertake complex operational activity in order to collect data, but technology rarely provides sufficient support. We find theAWARDsuite provides proof of principle that designing technology to support data collectors can mitigate known causes of poor data quality and produce higher-quality data. Health information technology (IT) products that support the conduct of scheduled activity in addition to traditional data entry will enhance community-based clinical trials. A standardised framework for reporting data quality would aid comparisons across clinical trials.Trial registration: International Standard Randomized Controlled Trial Number Register,ISRCTN83772183. Registered on 3 March 2005.