NOTE: The MAESTRO-OL study was exploratory in nature and no primary efficacy and safety endpoint were defined in the clinical protocol. This and the other exploratory efficacy outcome measures posted were selected to be reported as a primary endpoints. All efficacy analyses were considered exploratory. The analyses of the exploratory efficacy endpoints focused on the absolute values and on the change from DB baseline to Week 16 in the DB study and to Month 6 and Month 12 in the OL study. For missing 6MWD values in the OL study, the following imputation rules were applied: If the reason for missing data was death, a distance of zero (0) meters was imputed for all 6MWD visits from the date of death. For any other reasons, the last available value was carried forward.

Class I: no symptoms with exercise or at rest. No limitation of activity. Class II: No symptoms at rest but slight limitation with ordinary activities causing symptoms (e.g. short of breath with climbing stairs). Class III: may not have symptoms at rest but activities greatly limited by shortness of breath, fatigue, or near fainting. Class IV: symptoms at rest and inability to carry out any physical activity without symptoms. Patients in class IV manifest signs of right heart failure. For missing WHO FC values in the OL study, the following imputation rules were applied: If the reason for missing data was death, class IV was imputed for all WHO visits from the date of death. For any other reasons, the last available value was carried forward.

The Borg dyspnea score rates the severity of dyspnea (difficult or labored breathing) on a scale from 0 ('Nothing at all') to 10 ('Very, very severe - maximal'). For missing Borg dyspnea index values in the OL study, the following imputation rules were applied: If the reason for missing data was death, a value of 10 was imputed for all Borg visits from the date of death. For any other reasons, the last available value was carried forward.

No imputation of missing data for SpO2 was applied. Oxygen saturation assessed by pulse oximetry: peripheral oxygen saturation (SpO2) at rest before the 6-minute walk test (6MWT)

No imputation of missing data for oxygen saturation as assessed by SpO2 was applied. Therefore out of 109 subjects 104 subjects were analyzed at month 6 and 92 subjects at month 12 in the DB-macitentan group. In the DB-placebo group out of 108 subjects 103 subjects were analyzed at month 6 and 84 subjects at month 12.