Network meta-analysis

HbA1c below 7 percent

Type 2 diabetes (adults); background therapy: one or more non insulin glucose lowering drugs

weeks 36 to 64 (target week 52)Not stated by the trialsEach dose separatelyOdds ratioSeparate part 5 of a disconnected network

Randomized trials

Treatments

2,398

Participants

1,766

Events

In plain language

Across 2 trials with 2,398 participants, 0 of 3 treatments had a clearly better result than tirzepatide 10 mg subcutaneous once weekly on HbA1c below 7 percent, 1 a clearly worse one, and for 2 the evidence could not tell them apart from tirzepatide 10 mg subcutaneous once weekly. The most favorable estimate was for Tirzepatide 15 mg subcutaneous once weekly: odds ratio 1.17 (95% credible interval 0.51 to 2.60).

Higher values favor the treatment. "Clearly" means the 95% credible interval excludes no difference. These are averages across trials; they do not say how any one person will respond. Not medical advice.

Network of trials

Each circle is a treatment; each line joins treatments that were randomized against each other in at least one trial.

Node area shows participants; line width shows trials comparing two treatments directly. Blue: GLP-1 receptor agonist containing treatments; grey: comparators. Hover over or tab to a node or line for details.

Effects compared with Tirzepatide 10 mg subcutaneous once weekly

Odds ratio. Higher values favor the treatment.

Versus Tirzepatide 10 mg subcutaneous once weekly, log scale. Blue: Bayesian random effects posterior median with 95% credible interval (thick) and 95% prediction interval (dashed). Amber diamond: frequentist estimate with 95% confidence interval. Hover over or tab to a row for its values.

Effects of each treatment versus Tirzepatide 10 mg subcutaneous once weekly
Treatment	Bayesian OR (95% CrI)	95% prediction interval	Frequentist OR (95% CI)	Verdict
Tirzepatide 15 mg subcutaneous once weekly TIRZ_SC_15MG_QW	1.17 (0.51 to 2.60)	0.35 to 3.82	1.22 (0.76 to 1.95)	Unclear
Tirzepatide 5 mg subcutaneous once weekly TIRZ_SC_5MG_QW	0.50 (0.22 to 1.05)	0.15 to 1.63	0.53 (0.35 to 0.80)	Unclear
Insulin glargine INS_GLARGINE	0.13 (0.06 to 0.33)	0.04 to 0.49	0.13 (0.09 to 0.19)	Worse

Ranking

How likely each treatment is to be among the best. Rankings ignore the size of differences and their certainty, so read them with the effects above.

League table

Every treatment against every other: odds ratio of the row treatment versus the column treatment, posterior median and 95% credible interval. Green: row clearly better; red: row clearly worse.

Every treatment in a row compared with every treatment in a column
Row versus column	TIRZ_SC_10MG_QW	TIRZ_SC_15MG_QW	TIRZ_SC_5MG_QW	INS_GLARGINE
TIRZ_SC_10MG_QW	TIRZ_SC_10MG_QW	0.86 0.38 to 1.94	2.00 0.95 to 4.49	7.67 3.06 to 17.49
TIRZ_SC_15MG_QW	1.17 0.51 to 2.60	TIRZ_SC_15MG_QW	2.31 1.05 to 5.29	9.04 3.33 to 19.97
TIRZ_SC_5MG_QW	0.50 0.22 to 1.05	0.43 0.19 to 0.95	TIRZ_SC_5MG_QW	3.88 1.40 to 8.36
INS_GLARGINE	0.13 0.06 to 0.33	0.11 0.05 to 0.30	0.26 0.12 to 0.71	INS_GLARGINE

Heterogeneity, fit and convergence

Bayesian random effects (primary)

Heterogeneity (tau): 0.287 (95% CrI 0.072 to 1.016), log scale
Residual deviance: 6.6 for 7 data points
DIC: 12.4 (pD 5.7)
Largest R-hat: 1.003
Smallest ESS: bulk 2,884, tail 2,732
Divergent transitions: 0
Convergence: All criteria met
Priors: PR02 N(0, 5^2); PR05 N(0, 2.5^2); PR09 log normal(-2, 1.58^2) on tau^2 (Biological markers (dichotomised), pharm vs pharm)

Frequentist check (netmeta)

Method: inverse variance, REML, OR
Tau: 0.000
I squared: 0%
Q (total): 1.39 on 2 df, P = 0.499
Q (between designs): 1.39, P = 0.499

Inconsistency

Do trials comparing two treatments directly agree with the indirect evidence through other treatments?

No comparison here has both direct and indirect evidence, so inconsistency cannot be checked by node splitting; the between designs Q above is the global check.

Small study effects

A comparison adjusted funnel plot: each point is one trial comparison, centered on its own network estimate. Asymmetry can signal that small trials with unfavorable results are missing.

Fewer than ten trial comparisons, too few to judge funnel plot asymmetry.

Baseline characteristics of the trials

The baseline value of each effect modifier or prognostic factor that applies here, per trial: mean or percent over arms, weighted by the number randomized. A note under a value names a stand in (a median for a mean, 100 minus the male share, Asian subgroup rows, an all Asian site list); "not reported" means no source of the trial prints it.

Baseline effect modifiers and prognostic factors per trial
Trial	Baseline HbA1c (percent) effect modifier, prognostic factor; 2 of 2	Background glucose lowering therapy (stratum S0 to S4) effect modifier, prognostic factor; 2 of 2	Diabetes duration (years) effect modifier, prognostic factor; 1 of 2
SURPASS-4 S001287	8.52	S1	11.8
NCT03861039 S001385	8.56	S1	not reported

Comparability of the trials

Checked across designs (the set of treatments each trial compares) for eligibility, outcome definition and assessment, and the baseline factors above. Flags point to possible intransitivity; notes describe variation that is not tied to design.

FlagBaseline, Diabetes duration: not reported by 1 of 2 trials
NoteData handling, input conversions: count_from_percent 7

Comparability items per trial
Trial	time weeks	analysis set	definition	instrument	adjudicated	proportion method	intercurrent events (discontinuation)	intercurrent events (rescue)	blinding	funding	start year	Eligibility
SURPASS-4	52	modified_itt	not stated	not stated	not stated	not_stated	not stated	not stated	open_label	industry	2018	age 18 or older; HbA1c 7.5 to 10.5 percent; BMI 25 or more; increased cardiovascular risk; 1 to 3 oral agents
NCT03861039	52	modified_itt	not stated	not stated	not stated	not_stated	not stated	not stated	open_label	industry	2019	Japanese adults 20 y or older; BMI 23 or more; on OAM monotherapy

Trial data

What each trial contributes. Arm rows give the arm value with its standard error, or events over participants; contrast rows give the odds ratio on the analysis scale (log) against the trial's first row. Flags mark converted, combined or approximated inputs.

Trial data entering this analysis
Trial	Arm	Treatment	Week	Data	Events / N	Flags	Source
SURPASS-4	A1	TIRZ_SC_5MG_QW	52	arm, responders	264 / 326	count_from_percent	registry:outcomes
SURPASS-4	A2	TIRZ_SC_10MG_QW	52	arm, responders	283 / 321	count_from_percent	registry:outcomes
SURPASS-4	A3	TIRZ_SC_15MG_QW	52	arm, responders	303 / 334	count_from_percent	registry:outcomes
SURPASS-4	A4	INS_GLARGINE	52	arm, responders	496 / 978	count_from_percent	registry:outcomes
NCT03861039	A1	TIRZ_SC_5MG_QW	52	arm, responders	137 / 148	count_from_percent	registry:outcomes
NCT03861039	A2	TIRZ_SC_10MG_QW	52	arm, responders	143 / 146	count_from_percent	registry:outcomes
NCT03861039	A3	TIRZ_SC_15MG_QW	52	arm, responders	140 / 145	count_from_percent	registry:outcomes

Rerun this analysis yourself

Remove treatments or trials and refit the frequentist and Bayesian models in your browser, with the same methods as the published analysis.

Open in the NMA lab