Clinical Trial Intelligence

Which Endpoints Are Measured Most Often in European Heart Failure Phase II & III Trials?

8 July 2026

Across 46 European CTIS heart-failure Phase II and Phase III trial records authorised in 2024–2026, the most frequent endpoint family was heart-failure hospitalisation, HF events, urgent HF visits, or clinical worsening, measured in 30 records (65.2%). Mortality or cardiovascular death appeared in 28 records (60.9%), while KCCQ/quality-of-life and safety/tolerability endpoints each appeared in 21 records (45.7%). Phase III trials were dominated by hard clinical outcomes, while Phase II trials relied more on safety, NT-proBNP, imaging, functional capacity, and exploratory biomarker packages.

46
Phase II/III CTIS trial records
30 / 46
HF event / worsening endpoints
28 / 46
Mortality / CV death endpoints
24 / 28
Phase III records with HF event endpoints

HF events and mortality are the dominant endpoint families

HF hospitalisation, HF events, urgent visits, or clinical worsening appeared in 30 of 46 records (65.2%). Mortality, cardiovascular death, or all-cause death appeared in 28 of 46 records (60.9%), ahead of KCCQ/quality of life and safety/tolerability, each present in 21 of 46 records (45.7%).

Trial records measuring each endpoint family, % of 46
HF hospitalisation / HF event / clinical worsening65.2%
Mortality / CV death / all-cause death60.9%
KCCQ / quality of life45.7%
Safety / tolerability45.7%
NT-proBNP / BNP biomarkers39.1%
Renal function / kidney biomarkers37.0%
Cardiac imaging / structure / function34.8%
Endpoint family counted once per trial record when present in primary, secondary, or other/exploratory endpoint text.
Interpretation

For European CTIS heart-failure submissions, the most reusable endpoint architecture is still outcomes-first: HF hospitalisation/clinical worsening plus death/CV death. Patient-reported health status, safety, natriuretic peptides, renal function, and imaging are frequent but usually supportive rather than dominant.

Primary endpoints concentrate on death and HF events; secondary endpoints broaden the package

Primary endpoints most often measured mortality/CV death in 23 of 46 records (50.0%) and HF event or clinical-worsening outcomes in 22 of 46 records (47.8%). Secondary endpoints repeated these outcomes but added KCCQ in 19 records (41.3%), safety in 15 records (32.6%), imaging in 15 records (32.6%), and NT-proBNP/BNP in 14 records (30.4%).

Records containing endpoint family by endpoint position
Endpoint family Primary Secondary Exploratory / other
Mortality / CV death23 / 4624 / 460 / 46
HF event / worsening22 / 4622 / 461 / 46
KCCQ / quality of life5 / 4619 / 461 / 46
Safety / tolerability10 / 4615 / 461 / 46
NT-proBNP / BNP5 / 4614 / 460 / 46
Renal function7 / 4614 / 462 / 46
Cardiac imaging6 / 4615 / 461 / 46
Exploratory includes structured “other endpoints” and endpoint entries explicitly labelled exploratory.
Interpretation

For CTIS protocol design, mortality/CV death and HF events are the most defensible primary endpoint anchors. KCCQ, NT-proBNP, imaging, renal function, NYHA class, and 6MWD are more often used to support mechanism, health status, and clinical relevance rather than carry the primary claim.

Phase II measures safety and signals; Phase III measures events and survival

Among 18 Phase II records, safety/tolerability was the leading endpoint family at 12 of 18 records (66.7%), followed by NT-proBNP/BNP in 9 of 18 (50.0%) and cardiac imaging in 8 of 18 (44.4%). Among 28 Phase III records, HF event/clinical-worsening endpoints appeared in 24 of 28 (85.7%) and mortality/CV death in 23 of 28 (82.1%).

Top endpoint families by phase
Phase II: 18 records
Safety / tolerability66.7%
NT-proBNP / BNP50.0%
Cardiac imaging44.4%
KCCQ / QoL38.9%
6MWD / function38.9%
Phase III: 28 records
HF event / worsening85.7%
Mortality / CV death82.1%
KCCQ / QoL50.0%
Renal function50.0%
MACE / MI / stroke35.7%
Phase II and Phase III are based on the CTIS phase file slices used for the heart-failure cohort.
Interpretation

The operational burden shifts sharply between phases. Phase II submissions need dense safety, biomarker, imaging, and functional-readout capture, while Phase III submissions need adjudicated event workflows, hospitalisation capture, mortality follow-up, and KCCQ/renal support endpoints.

HFpEF, HFrEF, and HFmrEF show different endpoint emphasis

The most common sub-disease group was generic or unspecified heart failure in 21 of 46 records (45.7%), followed by HFpEF in 10 records (21.7%), HFrEF in 9 records (19.6%), and HFmrEF in 6 records (13.0%). HFpEF records split between HF outcomes and KCCQ at 5 of 10 each (50.0%), HFrEF most often used KCCQ in 6 of 9 (66.7%), and HFmrEF most often used HF events, mortality, and KCCQ at 4 of 6 each (66.7%).

Endpoint patterns in subgroups with 6 or more records
Subgroup Records Top endpoint family Frequency
Generic / unspecified HF21HF events and mortality14 / 21 each
HFpEF10HF events and KCCQ5 / 10 each
HFrEF9KCCQ6 / 9
HFmrEF6HF events, mortality, KCCQ4 / 6 each
Subgroups are non-exclusive because one trial record may include multiple HF phenotypes or comorbid populations.
Interpretation

HFpEF and HFmrEF programs retain heavy health-status endpoint use, while HFrEF records in this cohort frequently combine KCCQ with HF events and mortality. The smaller acute HF, pulmonary hypertension, obesity, CKD, and LV-hypertrophy subsets were visible but not large enough for robust endpoint-pattern comparison.

Small-molecule trials carry most of the outcomes-endpoint burden

Small molecules were the only large modality group, appearing in 31 of 46 records (67.4%). In small-molecule records, HF event/clinical-worsening endpoints appeared in 23 of 31 (74.2%) and mortality/CV death in 22 of 31 (71.0%). Peptide/protein/enzyme records were smaller at 5 of 46 (10.9%) and were led by safety endpoints in 3 of 5 records (60.0%).

Modality groups with 3 or more records
Small molecule
31 / 46 records
Top: HF events 23 / 31
Peptide / protein / enzyme
5 / 46 records
Top: safety 3 / 5
Other modality
3 / 46 records
Top: safety / NYHA / PK-PD 2 / 3 each
Modalities are counted non-exclusively when multiple modality labels are present in the CTIS record.
Interpretation

Small molecules define the core European HF endpoint benchmark in this cohort. Advanced modalities such as gene therapy, cell therapy, RNA, and oligonucleotides appeared, but their record counts were too small for stable modality-specific endpoint benchmarking.

CTIS submissions should plan around adjudicated events in Phase III and dense signal capture in Phase II

The 46 CTIS heart-failure records were distributed across 2024, 2025, and 2026 authorisation-year slices: 24 records in 2024 (52.2%), 17 in 2025 (37.0%), and 5 in 2026 (10.9%). Phase III represented 28 of 46 records (60.9%) and carried the strongest need for hospitalisation, death, recurrent-event, and adjudicated cardiovascular outcome capture.

Trial records by CTIS authorisation-year slice and phase
Year Phase II Phase III Total
202491524
202571017
2026235
CTIS year is based on the first-authorisation-year slice used in the cohort files.
Interpretation

For EU/CTIS submissions, Phase III heart-failure protocols should pre-specify event definitions, hospitalisation logic, death classification, recurrent-event handling, adjudication, and KCCQ timing clearly. Phase II protocols should justify biomarker, imaging, 6MWD, NYHA, PK/PD, and safety schedules because those endpoint families drive the early-signal package.

Definitions

HF event endpoints include heart-failure hospitalisation, urgent HF visit, clinical worsening, total HF events, first HF event, or recurrent HF events. KCCQ means Kansas City Cardiomyopathy Questionnaire. NT-proBNP/BNP means natriuretic-peptide biomarker endpoints. 6MWD/6MWT means six-minute walk distance/test. NYHA means New York Heart Association functional class. MACE means major adverse cardiovascular events, usually including cardiovascular death, myocardial infarction, stroke, and related ischemic events.