Trial eligibility associates with survival but not other treatment responses in pulmonary arterial hypertension

Background: Over the prior three decades, treatments for pulmonary arterial hypertension (PAH) have evolved to now include vasodilator and activin signalling inhibitor therapies. The key trials used to justify licensing of these therapies have strict patient enrolment criteria. Here, we investigate if treatment outcomes differ by clinical eligibility for PAH patients when analysed using three large international cohorts. Methods: We interrogated the UK National Cohort Study of Idiopathic and Heritable PAH (IPAH/HPAH), UK ASPIRE, and French PH registries. We investigated a population of 4,689 PAH patients for eligibility from six major PAH drug trials (SERAPHIN, GRIPHON, AMBITION, TRITON, PULSAR, and STELLAR). Subsequently, we investigated differences in short-term and long-term outcomes between eligible and non-eligible patients. Results: From a total of N=4,689 PAH patients, 49% (range 28% - 85%) of registry patients enrolled would not have been eligible to participate in clinical trials. Reasons for exclusion from trials varied but a 6-minute walk distance (6MWD) out of range for trial enrolment was most common. Trial eligibility did not associate with significant differences in clinical worsening in any analysis or differences in 6MWD except in the GRIPHON analysis (eligible 42 metres versus ineligible 26 metres, FDR = 0.0005). Trial-eligible populations, however, had significantly better long-term survival than trial-ineligible populations across the three cohorts; the UK cohort, ASPIRE registry and French PH registry. Conclusions: In three large PAH registries, many patients receiving approved, commissioned treatments would not have been eligible for the corresponding foundational licensing studies. The preservation of treatment response as measured by regulatory approved endpoints, 6MWD and time to clinical worsening strongly supports less restrictive trial entry criteria. The implications for the observed differences in long-term mortality between trial-eligible and trial-ineligible populations need further consideration.