14 CLINICAL STUDIES

The effectiveness of TZIELD was investigated in a randomized, double-blind, event-driven, placebo-controlled study (Study TN-10; NCT01030861) in 76 patients, 8 to 49 years of age with Stage 2 type 1 diabetes. Stage 2 type 1 diabetes was defined as having both of the following:

Two or more of the following pancreatic islet autoantibodies:
- Glutamic acid decarboxylase 65 (GAD) autoantibodies
- Insulin autoantibody (IAA)
- Insulinoma-associated antigen 2 autoantibody (IA-2A)
- Zinc transporter 8 autoantibody (ZnT8A)
- Islet cell autoantibody (ICA)
Dysglycemia on oral glucose tolerance testing

In this study, patients were randomized to receive TZIELD or placebo once daily by intravenous infusion for 14 days. Patients in the TZIELD group had a total drug exposure that was comparable to the total drug exposure achieved with the recommended total TZIELD dosage [see Dosage and Administration (2.4)]. The primary efficacy endpoint in this study was the time from randomization to development of Stage 3 type 1 diabetes diagnosis.

Baseline Patient Characteristics
In this study, 45% were female; 97% White, 1% Asian, and 1% reported multiracial background; 3% were Hispanic or Latino ethnicity; and 95% were from the United States. The median age was 14 years (72% were <18 years old) (Table 2).

Table 2. Baseline Age Characteristics of Adults and Pediatric Patients 8 Years of Age and Older with Stage 2 Type 1 Diabetes (Study TN-10)*
	TZIELD N=44	Placebo N=32
* Intent to treat (ITT) population
Age Group
≥ 18 Years	34%	19%
< 18 years	66%	81%
Pediatric Age Group Quartiles
8 to <11 years	21%	25%
11 to <14 years	27%	31%
14 to <18 years	18%	25%

Baseline Disease Characteristics
Table 3 displays the baseline disease characteristics in Study TN-10.

Table 3. Baseline Disease Characteristics of Adults and Pediatric Patients 8 Years of Age and Older with Stage 2 Type 1 Diabetes (Study TN-10)*
	TZIELD N=44	Placebo N=32
* Intent to treat (ITT) population † The glucose data are area under the time-concentration curve (AUC) values from the oral glucose tolerance test
Glucose, mg/dL†
median (min, max)	165 (115, 207)	154 (103, 200)
HbA1c, %
median (min, max)	5.2 (4.6, 6.1)	5.3 (4.3, 5.6)
HLA-DR4
Missing	5%	0
Absent	34%	34%
Present	61%	66%
HLA-DR3
Missing	5%	0
Absent	48%	53%
Present	48%	47%
HLA-DR3/DR4
Both DR3 and DR4	25%	22%
DR3 only	23%	25%
DR4 only	36%	44%
Missing	5%	0
Neither DR3 nor DR4	11%	9%
Autoantibodies Positive (N)
1	2%	0
2	27%	22%
3	25%	16%
4	27%	44%
5	18%	19%
Autoantibody Type Positive
GAD65	91%	88%
IAA	43%	34%
IA-2A	59%	75%
ICA	66%	88%
ZnT8	73%	75%
Abbreviations: HbA1c=hemoglobin A1c, SD=standard deviation, HLA = human leukocyte antigen, GAD65=Glutamic acid decarboxylase 65 (GAD) autoantibodies, IAA=Insulin autoantibody, IA- 2A=Insulinoma-associated antigen 2 autoantibody, ZnT8A=Zinc transporter 8 autoantibody, ICA=Islet cell autoantibody

Efficacy Results
In Study TN-10, Stage 3 type 1 diabetes was diagnosed in 20 (45%) of the TZIELD-treated patients and in 23 (72%) of the placebo-treated patients. A Cox proportional hazards model, stratified by age and oral glucose tolerance test status at randomization, demonstrated that the median time from randomization to Stage 3 type 1 diabetes diagnosis was 50 months in the TZIELD group and 25 months in the placebo group, for a difference of 25 months. With a median follow-up time of 51 months, therapy with TZIELD resulted in a statistically significant delay in the development of Stage 3 type 1 diabetes, hazard ratio 0.41 (95% CI: 0.22 to 0.78; p=0.0066) (Figure 1).

Study TN-10 was not designed to assess whether there were differences in the effectiveness between subgroups based on demographic characteristics or baseline disease characteristics.

Figure 1: Kaplan-Meier Curve of Time to Diagnosis of Stage 3 Type 1 Diabetes in Adult and Pediatric Patients Aged 8 Years and Older with Stage 2 Type 1 Diabetes by Treatment Group (Study TN-10)1

1: ITT population