diff --git a/agenta-backend/agenta_backend/services/db_manager.py b/agenta-backend/agenta_backend/services/db_manager.py
index 99080a13f7..306218cc57 100644
--- a/agenta-backend/agenta_backend/services/db_manager.py
+++ b/agenta-backend/agenta_backend/services/db_manager.py
@@ -1973,7 +1973,11 @@ async def list_human_evaluations(app_id: str):
     """
 
     async with db_engine.get_session() as session:
-        base_query = select(HumanEvaluationDB).filter_by(app_id=uuid.UUID(app_id))
+        base_query = (
+            select(HumanEvaluationDB)
+            .filter_by(app_id=uuid.UUID(app_id))
+            .filter(HumanEvaluationDB.testset_id.isnot(None))
+        )
         if isCloudEE():
             query = base_query.options(
                 joinedload(HumanEvaluationDB.user.of_type(UserDB)).load_only(UserDB.id, UserDB.username),  # type: ignore