Improved HbA_1c and Body Weight in GADA-Positive Individuals Treated With Tirzepatide: A Post Hoc Analysis of SURPASS

Abstract

Context

People with clinically diagnosed type 2 diabetes (T2D) but positive antiglutamic acid decarboxylase autoantibodies (GADA), referred to here as latent autoimmune diabetes in adults (LADA), may experience more rapid glycemic deterioration than those with T2D and may benefit from effective diabetes treatment with additional metabolic benefits.

Objective

This work aimed to assess glycated hemoglobin A_1c (HbA_1c) and body weight (BW) changes associated with tirzepatide in GADA-positive vs GADA-negative participants with a clinical T2D diagnosis.

Methods

Post hoc analyses based on pooled data from SURPASS 2-5, using mixed-model repeated measures from the efficacy analysis set, adjusting for study and baseline covariates including age, sex, baseline values, body mass index (BMI), and GADA status, were conducted on 3791 individuals. Intervention included tirzepatide (5, 10, 15 mg). Main outcome measure included change from baseline in HbA_1c at weeks 40 (SURPASS-2, -3, -5) and 42 (SURPASS-4) by GADA status.

Results

In participants with confirmed GADA status, 3671 (96.8%) were GADA negative and 120 (3.2%) were GADA positive (76 [63.3%] with low and 44 [36.7%] with high GADA levels). Baseline characteristics were similar between groups, except for slightly lower BMI in GADA-positive vs GADA-negative participants (mean [SD] BMI 32.2 [6.1] vs 33.6 [6.3]). At week 40/42, both groups achieved robust reductions in HbA_1c (−2.11% vs −2.32%) and BW (–9.2 kg vs −9.6 kg) (P < .001, both groups). HbA_1c reductions were greater in GADA-negative participants (estimated difference [95% CI]: 0.21% [0.03, 0.39]; P = .024) and BW reductions did not differ between groups (0.38 kg [−0.99, 1.75]; P = .588).

Conclusion

In this post hoc analysis, tirzepatide was associated with substantial reductions in HbA_1c and BW irrespective of GADA status in adults diagnosed with T2D, suggesting that tirzepatide may improve glycemic control in individuals with LADA.

Latent autoimmune diabetes in adults (LADA) is characterized by slow autoimmune damage of pancreatic β cells (1). The diagnosis is primarily made by the presence of glutamic acid decarboxylase autoantibodies (GADA) in adults with an absence of insulin requirement for at least 6 months since clinical onset (2). LADA accounts for approximately 2% to 12% of all patients with diabetes (3, 4) but has a phenotype similar to type 2 diabetes (T2D) in some cases and to type 1 diabetes (T1D) in other cases, and therefore may often remain undetected. The antihyperglycemic treatment for LADA may differ to those for T2D. Prior to the disease stage when insulin initiation is needed, the use of glucagon-like peptide-1 receptor agonists (GLP-1 RAs) could be considered due to efficient glycemic control, improvement in β-cell function, concomitant weight loss, and various other metabolic advantages (4).

As with all forms of diabetes, the treatment goals of LADA are to improve glycemic control, maintain β-cell secretion, reduce cardiovascular risk, and to prevent or delay further complications. While the treatment options that best achieve these goals in LADA remain under debate, not only insulin but also noninsulin therapies have been used to achieve target goals, depending on C-peptide levels. Different dipeptidyl peptidase 4 inhibitors and dulaglutide, a selective GLP-1 RA, have demonstrated efficacy (5, 6). While the clinical course in LADA can be varied, there have been reports of earlier insulin requirement and subtherapeutic responses to oral antihyperglycemic medications (4). Therefore, it may be of clinical value to understand whether newer antihyperglycemic medications such as tirzepatide are effective in improving glycemic control in individuals with LADA.

Tirzepatide, a once weekly glucose-dependent insulinotropic polypeptide (GIP) and GLP-1 RA, is approved for the treatment of T2D and obesity. In the phase 3 SURPASS clinical trial program for T2D, tirzepatide produced substantial reductions in glycated hemoglobin A_1c (HbA_1c) and body weight (BW), enabling many people with T2D to achieve normalization of glucose control (7-11). In this post hoc analysis, we evaluated the change from baseline in HbA_1c and BW associated with tirzepatide in participants with T2D from the SURPASS clinical trial program by GADA status.

Materials and Methods

Trial Design and Participants

The study designs, full inclusion and exclusion criteria, and primary results of the SURPASS-2, SURPASS-3, SURPASS-4, and SURPASS-5 clinical trials have been previously reported (8-11). Briefly, participants were randomly assigned to receive once-weekly tirzepatide (5 mg, 10 mg, or 15 mg), semaglutide (1 mg) (SURPASS-2), insulin degludec (SURPASS-3), insulin glargine (SURPASS-4), or a volume-matched placebo in a single-dose pen (SURPASS-5). Assignment to treatment group was determined by a computer-generated random sequence using the Eli Lilly and Company interactive web-response system. Key eligibility criteria included adults with T2D (HbA_1c of ≥7.0% or ≥7.5% to ≤10.5% at screening), a body mass index (BMI) of ≥23 or ≥25, and stable weight (±5%), and stable diabetes treatment during the previous 3 months with metformin greater than or equal to 1500 mg/day prior to screening (SURPASS-2 and ±SGLT-2i in SURPASS-3), 1 to 3 oral antihyperglycemic medications, which could include only metformin, sodium-glucose cotransporter-2 inhibitors (SGLT-2is), and sulfonylurea (SURPASS-4), or stable doses of once-daily insulin glargine (>0.25 U/kg/day or >20 U/day) with or without metformin (≥1500 mg/day) for 3 months prior (SURPASS-5). LADA as a subgroup was not investigated in these trials. GADA was not measured in SURPASS-1 (7), and therefore was excluded from this analysis.

The SURPASS clinical trials were conducted in accordance with the International Conference on Harmonisation Guidelines for Good Clinical Practice and the Declaration of Helsinki. All participants provided signed informed consent and protocols were approved by local ethical review boards. These trials are registered with ClinicalTrials.gov (NCT03987919, NCT03882970, NCT03730662, and NCT04039503).

Procedures

Participants who had a GADA concentration of 5 IU/mL or greater were categorized as GADA positive (ELISA [enzyme-linked immunosorbent assay], RSR Ltd, RRID: AB_3096313). GADA-positive participants were further divided into 2 subgroups based on GADA concentration (GADA-high levels >200 IU/mL, GADA-low levels ≤200 and ≥5 IU/mL) (12).

Efficacy measurements comprised change from baseline in HbA_1c and BW; fasting blood samples were collected at baseline and at weeks 4, 8, 12, 16, 20, 24, and 40/42 (SURPASS-2, -3, and -5 had measurements at week 40, whereas SURPASS-4 had measurements at week 42, therefore these time points were consolidated as week 40/42). Data from SURPASS 2 to 5 were pooled, therefore only common time points were evaluated (ie, HbA_1c and BW data at week 52 for SURPASS-3 and SURPASS-4 were not included in this analysis). Fasting C-peptide levels may help gauge β-cell function in people with LADA and guide therapeutic decisions, a potential need particularly for insulin therapy (12-15). Therefore, fasting C-peptide concentrations were assessed at baseline and at weeks 8, 16, 24, and 40 in SURPASS-2 and weeks 24 and 52 in SURPASS-4. Change from baseline in HbA_1c and fasting C-peptide based on the baseline fasting C-peptide categories described in a 2020 consensus statement (<0.3 nmol/L, 0.3-0.7 nmol/L, >0.7 nmol/L), in GADA-positive tirzepatide-treated participants were assessed (4). Additionally, assessments of β-cell function (homeostasis model assessment of β-cell function [HOMA2-B] calculated with fasting C-peptide) and insulin resistance (homeostasis model assessment of insulin resistance [HOMA2-IR] calculated with fasting C-peptide) were assessed in tirzepatide-treated participants from SURPASS-2 and SURPASS-4 (16). Level 2 hypoglycemic events of clinically significant blood glucose concentrations less than 54 mg/dL (<3 mmol/L) measured by any adequate methods by patients, caregivers, or medical professionals, including data obtained from central laboratories or from self-reported blood glucose measurements, or level 3 severe hypoglycemic events were assessed.

Outcomes

The primary outcome of this post hoc analysis was change from baseline in HbA_1c in tirzepatide-treated participants from SURPASS-2 through SURPASS-5 diagnosed with apparent T2D and who were positive for GADA vs negative for GADA. Secondary outcomes were change from baseline in BW in GADA-positive vs GADA-negative participants. Among those with detectable GADA, treatment outcomes in tirzepatide-treated GADA-high level vs GADA-low level participants were measured. Furthermore, C-peptide, HOMA2-IR, and HOMA2-B (both computed with C-peptide) were assessed. Insulin use throughout the duration of each trial and hypoglycemic events by GADA status were also assessed.

Statistical Analyses

All analyses were performed on the efficacy analysis set, comprising all randomly assigned participants who took at least one dose of study drug, excluding participants who discontinued study drug due to inadvertent enrollment and excluding data after initiation of rescue medication or premature discontinuation of study drug. Change from baseline in HbA_1c at 40/42 weeks was assessed in GADA-positive vs GADA-negative participants using mixed-model repeated measures from the efficacy analysis set, adjusting for study and baseline covariates including age, sex, HbA_1c, BMI, and GADA status. For incidence of hypoglycemia, a logistic regression model was used to assess the incidence rate between GADA-positive vs GADA-negative participants, and the model terms included study, baseline HbA_1c, and weight. There were no adjustments for multiplicity. Statistical test results were considered statistically significant at a 2-sided α level of .05. Statistical analyses were performed using SAS version 9.4, unless otherwise specified.

Results

Baseline Demographics and Clinical Characteristics

Overall, 3791 participants with confirmed GADA status were included in this analysis (3671 [96.8%] were GADA negative and 120 [3.2%] were GADA positive). Of the 120 participants who tested positive for GADA, 76 (63.3%) were GADA-low level and 44 (36.7%) were GADA-high level.

Baseline characteristics of all participants tested for GADA in the respective trials are shown in Table 1. Overall, baseline parameters were similar between the two groups, except for baseline BMI and serum triglyceride concentration, which were slightly lower in GADA-positive participants compared to GADA-negative participants (mean BMI [SD] 32.2 [6.1] vs 33.6 [6.3], mean triglycerides 169.2 [101.3] vs 188.1 [131.4] mg/dL). Among the GADA-positive participants, more participants with low GADA levels were male sex with lower baseline HbA_1c and fasting serum glucose and higher baseline triglyceride levels compared to those with high GADA levels.

Table 1.

Baseline demographics and clinical characteristics

Parameter	GADA-positive			GADA-negative (N = 3671)
	Total (N = 120)	GADA-high level (N = 44)	GADA-low level (N = 76)
Age, y	59.5 ± 10.92	59.7 ± 12.21	59.4 ± 10.19	58.9 ± 10.19
Sex
Female	55 (45.8)	24 (54.5)	31 (40.8)	1706 (46.5)
Male	65 (54.2)	20 (45.5)	45 (59.2)	1965 (53.5)
Duration of diabetes, y	10.1 ± 7.03	9.9 ± 6.81	10.2 ± 7.19	9.8 ± 7.02
HbA1c, %	8.5 ± 0.94	8.7 ± 0.91	8.3 ± 0.94	8.3 ± 0.96
Fasting serum glucose, mg/dL	172.5 ± 50.01	177.8 ± 51.67	169.5 ± 49.11	171.7 ± 50.95
Weight, kg	88.6 ± 18.35	87.2 ± 17.35	89.4 ± 18.97	93.3 ± 20.64
BMI	32.2 ± 6.08	31.8 ± 6.05	32.4 ± 6.13	33.6 ± 6.27
Systolic blood pressure, mm Hg	130.7 ± 15.90	129.5 ± 17.02	131.4 ± 15.29	132.5 ± 14.38
Diastolic blood pressure, mm Hg	77.7 ± 8.63	78.1 ± 10.52	77.4 ± 7.38	79.1 ± 9.27
Pulse rate, bpm	74.8 ± 9.60	75.7 ± 10.57	74.3 ± 9.02	74.3 ± 10.25
Use of OAM, yes	117 (97.5)	44 (100.0)	73 (96.1)	3614 (98.4)
Total cholesterol, mg/dL	169.0 ± 37.82	169.5 ± 38.99	168.7 ± 37.39	170.6 ± 42.62
HDL cholesterol, mg/dL	45.2 ± 11.62	46.9 ± 10.45	44.2 ± 12.20	43.8 ± 11.33
LDL cholesterol, mg/dL	90.2 ± 32.81	91.3 ± 36.57	89.6 ± 30.61	90.5 ± 34.76
Triglycerides, mg/dL	169.2 ± 101.28	156.3 ± 84.87	176.6 ± 109.51	188.1 ± 131.38

Data are mean ± SD or n (%) at baseline from the modified intention-to-treat population (efficacy analysis set). A total of 36 participants were reported to have “no GADA status.”

Abbreviations: BMI, body mass index; GADA, glutamic acid decarboxylase autoantibodies; HbA_1c, glycated hemoglobin A_1c; HDL, high-density lipoprotein; LDL, low-density lipoprotein; OAM, oral antihyperglycemic medication.

Reduction in Glycated Hemoglobin A_1c During Treatment With Tirzepatide

Significant reductions from baseline in HbA_1c were observed both in GADA-positive and GADA-negative participants over time (P < .001 vs baseline for both groups at all time points) (Fig. 1A). At week 40/42, both groups achieved significant HbA_1c reductions with slightly greater reductions in HbA_1c observed in GADA-negative participants compared to GADA-positive participants (−2.32% vs –2.11%, estimated between GADA subgroup differences [95% CI, 0.21% [0.03%-0.39%]; P = .024) (Fig. 1B). Significant reductions from baseline in HbA_1c were observed in GADA-positive participants treated with tirzepatide, irrespective of having high or low GADA levels over time (Fig. 2A). At week 40/42, HbA_1c reductions with tirzepatide were −2.01% in GADA-positive participants with high GADA levels and −2.17% in GADA-positive participants with low GADA levels. Reductions in HbA_1c significantly differed between GADA-positive participants with high GADA levels vs GADA-negative participants (estimated difference: 0.32% [0.01%-0.63%]; P = .046) but did not differ statistically in GADA-low levels vs GADA-negative participants (0.15% [−0.07% to 0.38%]; P = .180) (Fig. 2B).

Figure 1.

Change from baseline in HbA_1c and body weight by GADA status data are LSM (±SE) change from baseline over time and at the primary end point (efficacy analysis set). Change from baseline in HbA_1c at 40/42 weeks was assessed in GADA-positive vs GADA-negative participants using mixed-model repeated measures from the efficacy analysis set, adjusting for study and baseline covariates including age, sex, HbA_1c, BMI, and GADA status. A, Change from baseline in HbA_1c over time with actual HbA_1c values at week 40/42 in parentheses. B, Change from baseline in HbA_1c at week 40/42. C, Change from baseline in body weight over time with actual body weight values at week 40/42 in parentheses. D, Change from baseline in body weight at week 40/42. **P less than .001 change from baseline. BMI, body mass index; ETD, estimated treatment difference; GADA, glutamic acid decarboxylase autoantibodies; HbA_1c, glycated hemoglobin; LSM, least squares mean.

Figure 2.

Change from baseline in HbA_1c and body weight over time in GADA-positive participants by GADA level subgroup (high, low) data are LSM (±SE) change from baseline over time and at the primary end point (efficacy analysis set). Change from baseline in HbA_1c at 40/42 weeks was assessed in GADA-positive vs GADA-negative participants using mixed-model repeated measures from the efficacy analysis set, adjusting for study and baseline covariates including age, sex, HbA_1c, BMI, and GADA status. A, Change from baseline in HbA_1c over time with actual HbA_1c values at week 40/42 in parentheses by GADA level subgroup. B, Change from baseline in HbA_1c at week 40/42 by GADA level subgroup. C, Change from baseline in body weight over time with actual body weight values at week 40/42 in parentheses by GADA level subgroup. D, Change from baseline in body weight at week 40/42 by GADA level subgroup. **P less than .001 change from baseline. BMI, body mass index; ETD, estimated treatment difference; GADA, glutamic acid decarboxylase autoantibodies; HbA_1c, glycated hemoglobin; LSM, least squares mean.

GADA-positive participants from SURPASS-2 and SURPASS-4 achieved improved glycemic control with tirzepatide treatment regardless of baseline fasting C-peptide level (Table 2).

Table 2.

Fasting C-peptide at baseline and change in glycated hemoglobin A_1c in GADA-positive participants

Baseline fasting C-peptide subgroup, nmol/L	Pooled tirzepatide No.	Baseline fasting C-peptide, nmol/L	Change in HbA1c at primary end point, %
<0.3	10	0.19 (0.1 to 0.3)	−2.00 (−2.9 to 0.9)
≥0.3 to ≤0.7	22	0.54 (0.3 to 0.7)	−2.40 (−4.1 to −0.1)
>0.7	32	1.09 (0.7 to 2.8)	−2.15 (−4.9 to −0.5)

Data are median (min, max) baseline C-peptide and change in HbA_1c at the primary end point from GADA-positive participants from SURPASS-2 and SURPASS-4 (efficacy analysis set). C-peptide cutoffs are based on the current guidelines for management of LADA (4).

Abbreviations: GADA, glutamic acid decarboxylase autoantibodies; HbA_1c, glycated hemoglobin A_1c; LADA, latent autoimmune diabetes in adults; max, maximum; min, minimum.

Reductions in Body Weight During Treatment With Tirzepatide

Significant reductions from baseline in BW were observed both in GADA-positive and GADA-negative participants (P < .001; both groups at all time points) over time (Fig. 1C). At week 40/42, reductions in BW did not significantly differ between GADA-negative participants and GADA-positive participants (−9.6 kg [−10.2%] vs –9.2 kg [−10.4%], estimated treatment differences [95% CI, 0.38 kg [−0.99 to 1.75 kg]; P = .588) (Fig. 1D). Significant reductions from baseline in BW were observed in GADA-positive participants treated with tirzepatide, irrespective of having high or low GADA levels over time (Fig. 2C). At week 40/42, BW reductions with tirzepatide were −9.5 kg (−11.0%) in GADA-positive participants with high GADA levels and −9.1 kg (−9.8%) in GADA-positive participants with low GADA levels. Reductions in BW did not significantly differ between GADA-positive participants with high GADA levels vs GADA-negative participants (estimated difference: 0.1 kg [−2.3 to 2.4 kg]; P = .959) or between GADA-low level vs GADA-negative participants (0.5 kg [−1.2 to 2.2 kg]; P = .539) (Fig. 2D).

Insulin and Sulfonylurea Use During Planned Treatment Period

In SURPASS 2 to 5 during the planned treatment period, 12 (27.3%) participants from the GADA-high level subgroup and 8 (10.5%) participants from the GADA-low level subgroup used insulin, compared to 400 (10.9%) of GADA-negative participants (Table 3). Furthermore, 6 (13.6%) participants from the GADA-high level subgroup and 8 (10.5%) participants from the GADA-low level subgroup used sulfonylurea, compared to 579 (15.7%) of GADA-negative participants (see Table 3).

Table 3.

Summary of postbaseline insulin and sulfonylurea therapy during the planned study treatment

Parameter	GADA-positive(N = 120)		GADA-negative (N = 3680)	Total (N = 3800)
	GADA-high level (N = 44)	GADA-low level (N = 76)
Participants with ≥1 insulin therapy other than study drug	12 (27.3)	8 (10.5)	400 (10.9)	420 (11.1)
Insulins	12 (27.3)	8 (10.5)	400 (10.9)	420 (11.1)
Long-acting insulin	9 (20.5)	8 (10.5)	373 (10.1)	390 (10.3)
Short-acting insulin	1 (2.3)	0	14 (0.4)	15 (0.4)
Other	3 (6.8)	0	20 (0.5)	23 (0.6)
Participants with ≥1 sulfonylurea therapy	6 (13.6)	8 (10.5)	579 (15.7)	593 (15.6)

Data are n (%) from the modified intention-to-treat population (safety analysis set). “Long-acting insulin” includes insulin glargine, insulin detemir, and insulin degludec; “short-acting insulin” includes insulin aspart, insulin lispro, and other insulin and analogue injections that were fast-acting; and “other” includes insulin (unspecified), insulin human, and insulin porcine.

Abbreviation: GADA, glutamic acid decarboxylase autoantibodies.

Fasting C-peptide, β-Cell Function, and Insulin Sensitivity

Fasting C-peptide values at the primary end point of week 40/52 ranged from 0.44 to 0.66 nmol/L in SURPASS-2 and 0.83 to 1.24 nmol/L in SURPASS-4 with tirzepatide treatment (Supplementary Fig. S1) (17).

In a pooled analysis of SURPASS-2 and SURPASS-4, baseline HOMA2-B (calculated with fasting C-peptide) was significantly lower in GADA-positive participants compared to GADA-negative participants (P = .018) and over time significantly increased from baseline both in GADA-positive and GADA-negative participants during treatment with tirzepatide by 62.4% and 97.0%, respectively, (P < .001; both subgroups) (Supplementary Table S1) (17). Improvement in HOMA2-B was significantly greater in GADA-negative participants compared to GADA-positive participants at week 40/52 (estimate difference vs GADA-positive participants [95% CI], 21.3% [7.6%-36.8%]; P = .002).

Furthermore, baseline HOMA2-IR (calculated with fasting C-peptide) was significantly lower in GADA-positive participants compared to GADA-negative participants (P < .001) and significantly decreased from baseline both in GADA-positive and GADA-negative participants during treatment with tirzepatide by 27.2% and 16.9%, respectively (P < .001, both subgroups) (see Supplementary Table S1) (17). Improvement in HOMA2-IR was significantly greater in GADA-positive participants compared to GADA-negative participants at week 40/52 (14.3% [1.9%-28.2%]; P = .023).

Hypoglycemia

Overall, hypoglycemia (blood glucose <54 mg/dL or severe) was reported in 151 out of 3678 participants who were GADA negative compared to 12 out of 120 participants who were GADA positive (Supplementary Table S2) (17). Among the GADA-positive participants, hypoglycemic events were reported in 6 with high GADA levels and 6 with low GADA levels. Seven of these 12 participants were undergoing concomitant insulin therapy. Among them, 4 belonged to the high-GADA level group. None of these participants were undergoing concomitant sulfonylurea use. Among the 151 GADA-negative participants who reported hypoglycemic events, 121 were undergoing concomitant insulin and/or sulfonylurea use, The relative incidence rates (adjusted by patient-year of exposure) were significantly higher for GADA-positive participants compared to GADA-negative participants (relative rate 4.45 [95% CI], [2.20-9.02]; P < .001), and for those with high GADA levels compared to GADA-negative participants (9.04 [3.77-21.67]; P < .001). The relative rate of hypoglycemia for the participants with low GADA levels compared to GADA-negative participants was not statistically significant (1.83 [0.69-4.83]; P = .222). No cases of severe hypoglycemia were reported in GADA-positive participants.

Discussion

This post hoc analysis is the first demonstration of changes in HbA_1c and BW associated with novel once-weekly tirzepatide in GADA-positive participants who were previously diagnosed with T2D who likely have LADA. Tirzepatide was associated with marked reductions in HbA_1c and BW, irrespective of GADA status across the study population, suggesting that tirzepatide may be effective in improving glycemic control and weight management goals in LADA. In addition, the glycemic improvement observed with tirzepatide among GADA-positive individuals occurred regardless of GADA level or baseline fasting C-peptide concentration, which may indicate insulin production. Weight reductions with tirzepatide treatment observed in GADA-positive individuals is clinically relevant given the recommendations of the American Diabetes Association/European Association for the Study of Diabetes consensus report (18) and evidence supporting weight loss of 5% to 15% as a primary target in people diagnosed with T2D (19, 20).

In this analysis, the risk of hypoglycemia was slightly higher among GADA-positive individuals compared to GADA-negative individuals. This may be in part due to a higher prevalence of insulin use in GADA-positive individuals.

To date, selective GLP-1 RAs have demonstrated varied glycemic efficacies in LADA. For example, dulaglutide treatment was associated with a comparable improvement in HbA_1c in individuals with LADA compared to individuals without LADA (6). In contrast, exenatide or liraglutide treatment resulted in less HbA_1c reduction in LADA vs those with T2D (21). Regarding adults with new-onset T1D, a recent small-cohort report showed that semaglutide could replace or reduce insulin therapy, thus illustrating the continuum between adult-onset T1D and T2D, where C-peptide production exists even in the context of positive islet autoantibodies (22). These results suggest that incretin-based therapies, including tirzepatide, may have a place in the treatment regimens of people with LADA.

Our results confirm that the pancreatic β-cell function is impaired in LADA, as demonstrated by lower HOMA2-B in individuals with LADA compared to those with T2D. Nevertheless, tirzepatide treatment was associated with a robust increase in β-cell function also in GADA-positive individuals. Furthermore, fasting C-peptide levels remained stable over the course of the study in GADA-positive individuals, thus showing no sign of β-cell deterioration while on tirzepatide treatment during the 40/52-week period. The increase in β-cell function observed with tirzepatide treatment in this study is consistent with previous observations (23-26). Preclinical studies demonstrated an important role of GIP in intraislet coordination of α-cell and β-cell functions, which in turn may contribute to the improvement in β-cell function (27). Combined, these results suggest that increased β-cell function may contribute to the improved glycemic control associated with tirzepatide in people with LADA.

Another key component of glycemic homeostasis, insulin resistance, was also improved in people with LADA treated with tirzepatide. Tirzepatide was associated with significant reductions in the marker of insulin resistance, HOMA2-IR, and the reduction was as marked in people with LADA as in individuals with T2D. In a separate post hoc analysis, dulaglutide did not appear to meaningfully influence insulin resistance in people with LADA (6). The apparent insulin sensitization associated with tirzepatide in people with LADA may be partially due to a GIP receptor–mediated reduction in visceral fat content potentially leading to reduced insulin resistance (28).

Strengths of this study include incorporation of GADA antibody-level data and fasting C-peptide data to shed light on the effect of tirzepatide treatment in LADA. Limitations of this study include the exploratory nature of this post hoc analysis, the relatively small number of GADA-positive patients in this study, a lack of data on clinical course and other diabetes-autoantibodies needed to diagnose LADA in GADA-negative people, and a lack of long-term follow-up data.

In conclusion, once-weekly tirzepatide was associated with significant reductions in HbA_1c and BW, irrespective of GADA status in adults diagnosed with T2D, suggesting that tirzepatide may improve glycemic control in individuals with LADA. Future studies could help elucidate the long-term effect of the presence of GADA on the clinical progression of LADA and whether treatment with tirzepatide could delay or prevent insulin requirement in this population (29).

Abbreviations

BMI

body mass index

BW

body weight

GADA

glutamic acid decarboxylase autoantibodies

GIP

glucose-dependent insulinotropic polypeptide

GLP-1 RAs

glucagon-like peptide-1 receptor agonists

HbA_1c

glycated hemoglobin A_1c

HDL

high-density lipoprotein

HOMA2-B

homeostasis model assessment of β-cell function

HOMA2-IR

homeostasis model assessment of insulin resistance

LADA

latent autoimmune diabetes in adults

LDL

low-density lipoprotein

T1D

type 1 diabetes

T2D

type 2 diabetes

Funding

This work was supported by Eli Lilly and Company.

Author Contributions

J.S.P., I.P., and C.J.L. contributed to the study design. J.S.P., I.P., and C.J.L. provided medical oversight during the trial. M.L. was responsible for the statistical analyses. J.S.P., I.P., and C.J.L. are the guarantors of this work and, as such, take responsibility for the integrity of the data and the accuracy of the data analysis. All authors participated in interpretation of the data and critical review of the manuscript, had full access to all the data in the study, and approved of this manuscript to be submitted for publication.

Disclosures

A.L.P. has served on advisory boards for Abbott, Eli Lilly and Company, Vertex, and Medscape and received research funding from Abbott Diabetes Care and Insulet. R.B. has served as a speaker for Abbott, AstraZeneca, Eli Lilly and Company, Novo Nordisk, and Sanofi and has served on advisory boards for Abbott, Eli Lilly and Company, Novo Nordisk, Sanofi, and Vertex. M.L. is a contracted employee of Eli Lilly and Company. C.J.L., I.P., C.A.K., and J.S.P. are employees and shareholders of Eli Lilly and Company.

Data Availability

Lilly provides access to all individual participant data collected during the trial, after anonymization, with the exception of pharmacokinetic or genetic data. Data are available to request 6 months after the indication studied has been approved in the United States and European Union and after primary publication acceptance, whichever is later. No expiration date of data requests is currently set once data are made available. Access is provided after a proposal has been approved by an independent review committee identified for this purpose and after receipt of a signed data-sharing agreement. Data and documents, including the study protocol, statistical analysis plan, clinical study report, blank or annotated case report forms, will be provided in a secure data-sharing environment. For details on submitting a request, see the instructions provided at www.vivli.org.

Clinical Trial Information

Trials included in this post hoc analysis are registered with ClinicalTrials.gov [NCT03987919 (registered July 30, 2019-February 15, 2021), NCT03882970 (registered April 1, 2019-January 4, 2021), NCT03730662 (registered November 20, 2018-April 22, 2021), and NCT04039503 (registered August 30, 2019-January 13, 2021)].