This was a single-center, randomized, placebo-controlled, parallel-group study (EudraCT: 2015-004997-14) conducted in Germany between 24 February and 6 July, 2016. Investigators, site staff, and participants were blinded with respect to treatment with vericiguat or placebo; however, bioanalytical staff were not blinded.

Participants were randomized to one of two parallel groups of 16 subjects each. One group received vericiguat 10 mg and the other group received placebo once daily for 16 days (days 0–15; Fig. 1). Vericiguat or placebo was administered orally as film-coated tablets containing vericiguat 5 mg or corresponding placebo, which were identical in appearance (size, color, and shape), and with packaging and labeling designed to maintain blinding of the site staff, investigator, and participants.

Treatment schedule. aDuring the vericiguat/placebo treatment period, study treatment was administered in the clinic on in-house days and was administered by the subject at home on days 1–11

Vericiguat or placebo was administered within 30 min (in-house days: days 0, 12, 13, 14, and 15) or within 1 h (ambulatory days: days 1–11) after breakfast. Sildenafil 25 mg, 50 mg, and 100 mg were administered approximately 2 h after vericiguat/placebo on days 13, 14, and 15, respectively. For vericiguat at steady state, time to achieve maximum plasma concentration (Cmax) is approximately 4 h after intake [1, 2]. For sildenafil, time to achieve Cmax is approximately 1–2 h [5, 6]. Based on the administration scheme used in this study, it was expected that peak values for vericiguat and sildenafil would be reached at approximately the same time, maximizing the effect of co-administration on safety and pharmacodynamic parameters.

Sildenafil treatment was initiated on day 13 to ensure that it was administered at vericiguat hemodynamic steady state. Both sildenafil and its active metabolite have a terminal half-life of about 4 h [5, 6]; therefore, it was possible to administer ascending doses of sildenafil on consecutive study days.

Healthy white male participants were eligible for inclusion to the study if they were ≥ 40 years of age and ≤ 60 years of age and had a body mass index ≥ 18.0 kg/m2 and ≤ 29.9 kg/m2. Participants were excluded if they had a clinically relevant finding on the electrocardiogram (ECG), had systolic blood pressure (SBP) < 100 mmHg or > 145 mmHg, had diastolic blood pressure (DBP) < 60 mmHg or > 95 mmHg, or heart rate (HR) < 50 or > 95 beats per min (bpm). Regular use of medication within 4 weeks and any use of medication, herbal products, or vitamins within 14 days prior to the first study drug administration were prohibited. Regular daily consumption of more than ten cigarettes and intake of foods and beverages containing grapefruit within 14 days before study drug administration were exclusion criteria. A full list of inclusion and exclusion criteria is included in Table S1 of the Electronic Supplementary Material (ESM).

The safety evaluation included assessment of adverse events (AEs), clinical laboratory parameters, vital signs, and ECGs. Blood and urine samples for clinical laboratory parameters (hematology, clinical chemistry, coagulation, and urinalysis) were collected in the fasting state, at screening, prior to study drug administration on days 0, 12, 13, 14, and 15, prior to breakfast on day 16, and at follow-up. Vital signs and ECGs were recorded after 15 min of supine rest at screening, prior to and 5.5 h after study drug administration on days 0, 12 (ECG recorded pre-dose only), 13, 14, and 15, prior to breakfast on day 16, and at follow-up. The hemodynamic profile was also a measure of safety.

A hemodynamic profile including SBP, DBP, and HR, recorded after a supine phase of 15 min in seated and standing (after 2 min) positions, was measured over approximately 3.75 h, between 1.25 h and approximately 5 h after vericiguat/placebo administration on days 0, 12, 13, 14, and 15 (Fig. 2).

Schedule of hemodynamic profiling. Timepoints refer to vericiguat/placebo administration. Arrows indicate times of blood pressure/heart rate assessment. Sildenafil was administered only on days 13, 14, and 15, 2 h after vericiguat administration. *Assessments that contributed to the hemodynamic reference value (‘baseline’). BP blood pressure, h hours, HR heart rate, min minutes

Blood samples for pharmacokinetic assessment of vericiguat were collected pre-dose and at 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12, and 15 h post-dose on days 12, 13, 14, and 15, and on day 16 (24 h after the last dose of vericiguat/placebo). Blood samples for pharmacokinetic assessment of sildenafil were collected at 3, 4, 6, 8, 12, and 24 h after administration of vericiguat/placebo on days 13, 14, and 15.

Blood samples were collected by a member of the investigator’s team. The amount of blood planned to be drawn during the course of this study was less than 500 mL per subject within approximately 5–6 weeks. The approximate volume was indicated in the subject information sheet and consent form.

Quantitative analysis of vericiguat and sildenafil concentrations in plasma was performed using fully validated assays. The analyses were performed in accordance with the US Food and Drug Administration guideline on bioanalytical validation (2001) [9].

Vericiguat was determined in human lithium heparinized plasma after addition of the internal standard [13C, 2H4] vericiguat and automated protein precipitation using a mixture of acetonitrile and ammonium acetate with formic acid in Milli-Q-type water. Separation was achieved by means of a liquid chromatographic system (Column: Synergi Polar-RP, 75 × 4.6 mm, 4 μm; Analyst 1.6.1; AB Sciex, Framingham, MA, USA). For the mass spectrometric detection, a triple quadrupole mass spectrometer in positive TurboIonSprayTM ionization mode was applied. The calibration range was from 1.00 μg/L (lower limit of quantification [LLOQ]) to 1000 μg/L (upper limit of quantification [ULOQ]). Quality control accuracy (calculated as percent of nominal) and precision (coefficient of variation) were 93.88–101.33% and 1.24–2.58%, respectively. All samples were stored at − 20 °C and analyzed within 36 days after sample collection. The stability data indicated that the analyte is stable for this time period. The demonstrated stability is 183 days at − 20 °C.

Sildenafil concentration was determined in human EDTA K3 plasma after addition of the internal standard sildenafil-d8 and liquid-liquid extraction with methyl tert-butyl ether. Separation was achieved by means of a liquid chromatographic system (Column: ACE 3 C18, 50 × 4.6 mm, 3 μm; Analyst 1.6.1, AB Sciex). For the mass spectrometric detection, a triple quadrupole mass spectrometer in positive TurboIonSpray™ ionization mode was applied. The calibration range was from 0.200 μg/L (LLOQ) to 250 μg/L (ULOQ). Quality control accuracy and precision were 97.60–103.50% and 2.72–7.81%, respectively. All samples were stored at − 20 °C and analyzed within 38 days after sample collection. The stability data indicated that the analyte is stable for this time period. The demonstrated stability is 55 days at − 20 °C.

Pharmacokinetic parameters were calculated with WinNonlin (versions 5.3 or higher; Pharsight Corporation, Mountain View, CA, USA) and included area under the concentration versus time curve, Cmax, time to achieve Cmax, and elimination half-life.

Participants attended a follow-up visit within 7 days of the last study drug administration. During this visit, participants received a full physical examination including an ECG, assessment of blood pressure (BP) and HR, and blood and urine safety analyses. They were also questioned about AEs they had experienced.

Evaluation of the safety and tolerability of vericiguat 10 mg once daily, and of single doses of sildenafil 25 mg, 50 mg, and 100 mg co-administered with steady-state vericiguat, was the primary objective of the study. All subjects who received at least one dose of the study medication were included in the safety evaluation. The incidence and intensity of treatment-emergent AEs (TEAEs), serious AEs, and drug-related AEs were summarized according to the Medical Dictionary for Regulatory Activities, version 19.0. Adverse events were treatment-emergent if they started or worsened after the first administration of treatment, up to 30 days after the end of treatment. For each TEAE, the investigator reported whether they considered it related to the study medication.

All subjects with evaluable pharmacodynamic data and without major protocol deviations affecting the pharmacodynamic validity were included in the evaluation of pharmacodynamics. The hemodynamic reference values for changes in BP or HR (designated as ‘baseline’) were the mean of the four values of seated assessments taken in the 2 h after administration of vericiguat or placebo, which occurred prior to the planned time of administration of sildenafil on days 13, 14, and 15. The change in seated SBP, DBP, and HR at each timepoint within the hemodynamic profile after the sildenafil dose were analyzed by an analysis of covariance (ANCOVA) including effects for treatment (vericiguat or placebo), time, and treatment by time, and baseline SBP, DBP, or HR as a covariate. The ANCOVA was performed separately for each of the three sildenafil doses. In addition, the maximum decrease in seated SBP or DBP and the maximum increase in seated HR after administration of each sildenafil dose were analyzed by ANCOVA including an effect for treatment (vericiguat or placebo) and including baseline SBP, DBP, or HR as a covariate, performed separately for each dose of sildenafil. Point estimates (least squares means) and exploratory 90% CIs and p-values for the main effect across all timepoints (differences of ‘vericiguat + sildenafil’ and ‘placebo + sildenafil’) were calculated. For mean changes (using area under the concentration versus time curve) of SBP, DBP, and HR, point estimates and exploratory 90% CIs were calculated by treatment and dose of sildenafil. Blood pressure and HR measurements from the standing BP procedure were analyzed separately and presented as summary descriptive statistics.

All subjects with a valid vericiguat profile (≥ two-thirds of individual values > LLOQ) on day 12 and at least one valid profile of both vericiguat and sildenafil on days 13, 14, or 15 were included in the evaluation of vericiguat pharmacokinetics. The main pharmacokinetic parameters of vericiguat (area under the concentration versus time curve from 0 to 24 h [AUC(0–24)] and Cmax) were analyzed assuming log-normally distributed data. The logarithms of these parameters were analyzed using an analysis of variance (ANOVA) including effects for administration (vericiguat alone or in combination) and subject. Point estimates (least squares mean parameters), exploratory 90% CIs, and 95% prediction intervals for the ratios ‘vericiguat + 25 mg sildenafil/vericiguat alone’, ‘vericiguat + 50 mg sildenafil/vericiguat alone’, and ‘vericiguat + 100 mg sildenafil/vericiguat alone’ were back-transformed from logarithmic data to the original scale. For the calculation of 90% CIs and 95% prediction intervals, the intra-individual standard deviation from the ANOVA was used.

All subjects with at least one valid sildenafil profile (in the vericiguat group, along with a valid vericiguat profile; ≥ two-thirds of individual values were > LLOQ) were included in the evaluation of sildenafil pharmacokinetics. The pharmacokinetic parameters of area under the concentration versus time curve from 0 to 22 h (AUC(0–22)) and Cmax of sildenafil were analyzed assuming log-normally distributed data. The logarithms of these parameters were analyzed separately for each sildenafil dose using an ANOVA including a treatment effect. Based on these analyses, point estimates (least squares mean), exploratory 90% CIs, and 95% prediction intervals for the ratios ‘vericiguat + 25 mg sildenafil/placebo + 25 mg sildenafil’, ‘vericiguat + 50 mg sildenafil/placebo + 50 mg sildenafil’, and ‘vericiguat + 100 mg sildenafil/placebo + 100 mg sildenafil’ were calculated by re-transformation of the logarithmic results obtained from the ANOVA. Statistical analyses were performed using SAS release 9.2 (SAS Institute, Cary, NC, USA).

For treatment with vericiguat in comparison with placebo, a sample size of 24 subjects (12 per treatment group) was calculated to have 94% power to detect a 10-mmHg difference in the sildenafil-induced decrease in SBP. The common standard deviation was assumed to be 6.65 mmHg for all SBP measurements within a 4-h time window after dosing (including a Schellong test), as observed in a previous trial (BAY 1021189/15357), in which multiple doses of vericiguat 10 mg were administered once daily. This sample size estimation was applicable for a one-way ANCOVA with a significance level alpha = 0.05. The calculation was conducted using SAS® Proc Power (SAS Institute). Thus, a total of 32 subjects (16 subjects per treatment group) were to be treated to ensure that a minimum number of 24 subjects (12 subjects per treatment group) would complete study treatments and supply valid data.