Patients

This study took place in 3 research centers and involved 48 subjects (Table 1). 243 subjects were screened after signing informed consent, and 195 did not meet the inclusion criteria. Forty-eight subjects were randomly assigned into three dose groups of HSK7653 (10, 25 and 50 mg) and placebo group using an interactive web response system. Forty-three subjects (89.6%, 43/48) completed 6 months of study, and drug compliance was 100% for the whole treatment period. In the initial study period, four subjects (2, 1 and 1 subject in the 10, 50 mg and placebo groups respectively) withdrew, 2 due to diabetic ketosis, and 2 for protocol violations in the extended study period, 1 subject (in the placebo group) withdrew due to poor blood glucose control and hyperglycemia (Fig. 1).

Fig. 1Safety

In this study, 38 subjects (79.2%, 38/48) experienced 137 incidences of treatment-emerged adverse events (TEAEs). The incidence of TEAEs in the 10, 25, 50 mg and placebo group were 58.3%, 91.7%, 83.3% and 83.3%, respectively. The highest incidence reported was positive urinary leukocyte (27.1%). Other TEAEs with an incidence > 5% included: prolonged QT interval (16.7%), abnormal T wave (12.5%), hypertriglyceridemia (10.4%), dyspepsia (10.4%), decreased T wave amplitude of ECG (8.3%), hyperlipidemia (8.3%), positive urine red blood cells (6.3%), hyperuricemia (6.3%), sinus bradycardia (6.3%), urinary tract infection (6.3%), hyperbilirubinemia (6.3%), dizziness (6.3%), and anemia (6.3%). Except one subject (25 mg dose group) with grade 3 upper respiratory tract infection, all the other TEAEs were grade 1 or 2, and 92.0% (126/137) of the TEAEs was classified as “symptom disappearance without sequelae”.

A total of 29 (60.4%, 29/48) subjects in this study experienced 93 adverse drug reactions (ADRs). The incidences of ADR in the 10, 25, 50 mg and the placebo group were 33.3%, 58.3%, 75.0% and 75.0%, respectively. The ADRs with HSK7653 or placebo incidence > 5.0% (HSK7653 dose group total vs placebo) included dyspepsia (11.1% vs 8.3%), hypertriglyceridemia (11.1% vs 0%),prolonged QT interval of ECG (11.1% vs 16.7%), white blood cells urine positive (8.3% vs 8.3%), electrocardiogram T wave abnormal (8.3% vs 8.3%), hyperlipidemia (8.3% vs. 8.3%), urinary tract infection (8.3% vs. 0%), decreased T wave amplitude of ECG (5.6% vs. 8.3%), positive urine red blood cells (5.6% vs. 8.3%), upper respiratory tract infection (5.6% vs. 0%), hyperbilirubinemia (5.6% vs. 8.3%), increased blood bilirubin (5.6% vs. 0%), anemia (5.6% vs. 8.3%) and dizziness (2.8% vs. 16.7%). Except for one subject with upper respiratory tract infection in the 25 mg dose group was classified as grade 3, and the rest of the ADRs were grades 1 or 2, 94.6% of the ADRs was classified as “symptom disappearance without sequelae”.

There were no serious adverse events/events, or adverse events requiring dose reduction, drug suspension or resumption, or posing a threat to life or death. In summary, HSK7653 was well-tolerated at doses of 10–50 mg, with no significant increase in TEAEs or ADRs compared to the placebo group (Supplementary Tables S1, S2).

EfficacyHbA1c

HbA1c changes comparing to the baseline at each visit are shown in Table 2. Figure 2 shows that on day 85, the level of HbA1c in the HSK7653 dose groups were significantly lower than that at the baseline. Compared with the placebo group (the HbA1c level decreased by 0.3% from baseline), 10 and 25 mg dose groups showed more significant reductions (1.0% and 1.3% respectively), but 50 mg dose groups showed the lowest reductions (− 0.05).

Table 2 HbA1c changes comparing with baseline at each visit pointFig. 2

HbA1c levels in each dose group before and after treatment

Glycated Albumin (GA) and Fasting Plasma Glucose

On day 43, GA levels in the 10 and 25 mg dose groups were significantly lower than the baseline (2.3% and 0.8%), while GA levels in the 50 mg and placebo group were significantly higher than the baseline (0.4% and 0.8%). With the increase of treatment time, the level of GA in 10 and 25 mg dose groups decreased continuously. The mean baseline GA value of 10 and 25 mg dose groups were 23.3% and 21.8%; on day 43 were 21.2% and 21.0%; on day 85 were 20.0% and 20.5%, which were lower than those in the 50 mg and placebo group. The baseline mean GA value in the 50 mg and placebo group were 23.2% and 23.7%; on day 43 were 23.6% and 24.5%; on day 85 were 24.2% and 22.8%.

On day 43, compared to baseline, fasting blood glucose levels were significantly lower in 10 and 25 mg dose groups (0.7 and 0.5 mmol/L); in 50 mg and placebo group were significantly higher (increased 0.8 mmol/L and 0.5 mmol/L). Both 10 and 25 mg dose groups showed significant hypoglycemic benefit at each checking point. The baseline mean fasting blood glucose values in 10 and 25 mg dose group were 10.1 and 8.7 mmol/L; on day 43 were 9.2 and 8.1 mmol/L; on day 85 were 9.2 and 8.5 mmol/L. The hypoglycemic effect lasted until day 85 after administration, which decreased by 0.7 and 0.2 mmol/L compared with the baseline.

PK/PDPharmacokinetics

After oral administration of HSK7653, drug absorption was relatively rapid, and the median Tmax of HSK7653 was about 1.0–3.0 h. In the dose range of 10–50 mg, PK characteristics of the first dose and the last dose were similar. Cmax and AUC of HSK7653 increased linearly with dose. The average CL/F, VZ/F, and t1/2 of HSK7653 were 0.7–0.8 L/h, 61.1–81.9 L, and 64.0–87.0 h, respectively. After oral administration of 10–50 mg HSK7653 every 2 weeks, the mean Rac_Cmax and CV% were 0.9–1.0 and 19.0–22.0%, respectively. The mean Rac_AUC and CV% were 1.0–1.1 and 10.0–18.0%, respectively. Under the current administration regimen, HSK7653 did not accumulated in body (Table 3). The average plasma concentration versus time curve is shown in Fig. 3.

Table 3 Summary of pharmacokinetic parameter of HSK7653 after single (first dose) or multiple (last dose) oral administration in each dose groupFig. 3

Average concentration versus time curves of subjects after single (Day 1–15) or multiple (Day 29–34) oral administration in each dose group. A Linear scale; B log-linear scale

PharmacodynamicsPlasma DPP-4 Inhibition Rate

After oral administration of HSK7653 every two weeks, the mean inhibition rate of DPP-4 in each HSK7653 dose group was significantly higher than that in the placebo group (Table 4a). There was no significant difference in DPP-4 inhibition rate among the HSK7653 dose groups after first and last doses, as shown in Fig. 4. The plasma concentration scatters diagram of DPP-4 inhibition rate of HSK7653 is shown in Fig. 5, when the plasma concentration of HSK7653 was 20.0 ng/mL or above, the DPP-4 inhibition rate in most subjects (99.8%) could reach more than 80.0%. AUECDPP-4 in HSK7653 dose groups was significantly higher than that in the placebo group and increased slightly with dose increase (Table 4b). There was no significant difference in AUECDPP-4 in each dose group after first and last doses (Table 4b).

Table 4 DPP-4 inhibition rate of each dose group of HSK7653 (A. Average DPP-4 inhibition rate at different time points; B. The area under the DPP-4 inhibition rate-time curve.)Fig. 4

DPP-4 inhibition rate-time curves of each dose group after first dose and last dose of HSK7653. Left: first dose (Day 1–15), Right: last dose (Day 29–43). Nominal time after first dose: time after planned first dose. Each color line represents the mean DPP-4 inhibition rate-time curve for one dose group, and the error bars represent the SD

Fig. 5

DPP-4 inhibition rate of HSK7653 plasma concentration scatter plot. Figure shows model fit line and 95% confidence band in red and light grey lines, respectively

Plasma GLP-1 Levels

The MTT test was carried out on day − 1 (baseline), day 42 (1 day before the fourth administration) and day 85 (2 weeks after the sixth administration), and the time courses of plasma GLP-1 levels are shown in Fig. 6. Before treatment, there was a slight difference in the AUECGLP-1 between the HSK7653 dose groups and the placebo group; the mean baseline values (CV%) of AUECGLP-1 for 10, 25, 50 mg and placebo were 15.5 (25.0), 13.7 (45.0), 21.2 (79.0), and 11.0 (26.0) pg*h/mL, respectively. After treatment with HSK7653, AUCEGLP-1 in the treatment groups were significantly higher than the placebo group. And the values of all the dose groups were similar. With the increase of dose, the median value showed a certain increasing trend. In each dose group, AUCEGLP-1 of HSK7653 (i.e., on day 42 and 85) was about twice as high as that at baseline. The mean value (CV%) of AUECGLP-1 on day 42 for 10, 25, 50 mg and placebo were 31.4 (53.0), 30.7 (45.0), 44.6 (60.0), and 12.2 (44.0) pg*h/mL, respectively. The mean (CV%) value of AUECGLP-1 on day 85 for 10, 25, 50 mg and placebo were 35.5 (80.0), 38.4 (60.0), 42.6 (38.0), and 15.0 (35.0) pg*h/mL, respectively.

Fig. 6

Plasma GLP-1 changes with time in each dose group in MTT test. Left: Day 1; middle: Day 42; right: Day 85

Serum Insulin

After multiple administration of HSK7653 (on day 85), the serum insulin levels were increased in each dose group compared with the placebo group. The mean AUECinsulin0–4 h and its change rate in each dose group were increased steadily with increasing treatment time. The AUECinsulin0–4 h values of the 10, 25, 50 mg and placebo groups on day − 1 were 73.1 ± 40.9, 96.3 ± 48.8, 65.9 ± 33.3, and 73.4 ± 35.8 mU*h/L, respectively; the AUECinsulin0–4 h values of the 10, 25, 50 mg and placebo groups on day 85 were 91.2 ± 41.8, 124.0 ± 107.0, 96.4 ± 34.7, and 91.1 ± 43.2 mU*h/L, respectively.

Serum C-peptide and Plasma Glucagon

Before administration, the level of plasma C-peptide in 25 mg dose group was the highest, followed by the 50 mg dose group and the placebo group, and the lowest in the 10 mg dose group. With the increase of treatment time, the level of C-peptide in the 10 mg dose group gradually increased to a similar level as the placebo group, while there was no significant change in the levels of C-peptide in the 25 and 50 mg dose groups compared with the placebo group.

Before administration, the level of the plasma glucagon in the 10 mg dose group was the highest, while in the 25 and 50 mg dose groups were lower than the placebo group. With the increase in treatment time, glucagon level in the 10 mg dose group was significantly higher than the placebo group at the beginning. It gradually decreased to below the level of the placebo group at 90–180 min after meal. The glucagon levels in both 25 and 50 mg dose groups were lower than the placebo group during treatment.