Abstract
Background:
Tirzepatide (TZP) is a novel drug for type 2 diabetes mellitus (T2DM), but the gastrointestinal (GI) adverse events (AEs) is a limiting factor in clinical application. Therefore, this study systematically evaluated the GI AEs of TZP for T2DM.
Methods:
Clinical trials of TZP for T2DM were retrieved from eight databases published only from the establishment of the database to February 2023. Revman5.3 and TSA0.9.5.10 Beta were used for meta-analysis and trials sequential analysis (TSA).
Results:
Meta-analysis showed that compared with placebo, total GI AEs, nausea, decreased appetite, constipation and vomiting were significantly higher in all dose groups of TZP (P < .05), while abdominal pain and abdominal distension were comparable (P > .05). TSA showed that the differences in total GI AEs, nausea, decreased appetite and constipation were conclusive. Compared with insulin, nausea, diarrhea, vomiting and decreased appetite were significantly increased in all doses of TZP (P < .05), and dyspepsia was significantly increased with TZP 15 mg (P < .05). TSA showed that these differences were all conclusive. Compared with GLP-1 RA, decreased appetite was significantly higher with TZP 5 mg, total GI AEs, decreased appetite and diarrhea were significantly higher with TZP 10 mg (P < .05), while nausea, vomiting, dyspepsia and constipation were significantly different in all dose groups, abdominal pain were not significantly different (P < .05) and TSA showed no conclusive results in this group.
Conclusion:
The GI AEs of TZP were significantly higher than those of placebo and insulin, but comparable to GLP-1 RA. Nausea, diarrhea and decreased appetite are very common GI AEs of TZP, and the incidence is positively correlated with dose. GI AEs of TZP decrease gradually over time, so long-term steady medication may be expected to reduce GI AEs.
Keywords: gastrointestinal adverse events, meta-analysis, tirzepatide, trial sequential analysis, type 2 diabetes mellitus
1. Introduction
Type 2 diabetes mellitus (T2DM) is a chronic metabolic disease characterized by persistent hyperglycemia and dysregulation of the progressive insulin-glucose feedback mechanism.[1] Globally, the prevalence of diabetes has been steadily increasing and has reached pandemic proportions.[2] Epidemiology shows that the prevalence of diabetes among people aged 20 to 79 globally in 2021 is about 10.5% and is expected to rise to 12.2% by 2045.[3] T2DM is the most common type of diabetes, accounting for about 90% of diabetes.[4] In the course of T2DM, retinopathy, kidney disease and cardiovascular disease may be complicated, resulting in significant economic losses for individuals, families, health systems and society.[5,6] Effective glycemic control and weight loss are important treatments for T2DM.[7] Although insulin and glucagon-like peptide-1 receptor agonist (GLP-1 RA) play a good role in the glycemic control of T2DM. Unfortunately, insulin has the risk of increasing weight and inducing hypoglycemia,[8,9] while GLP-1 RA does not achieve the ideal effect of hypoglycemic and weight loss.[10,11] These factors limit the clinical efficacy of insulin and GLP-1 RA. Therefore, the treatment of T2DM is in urgent need of a drug that can effectively reduce glycemic and weight loss with additional benefits.
Tirzepatide (TZP) is a new drug for T2DM, which has the dual activation of gastric inhibitory polypeptide receptor (GIPR) and GLP-1 RA.[12] Meta-analysis showed that TZP was significantly superior to insulin and GLP-1 RA in reducing glycemic and weight loss.[13,14] There is also evidence that TZP has additional benefits in reducing cardiovascular risk and blood lipid levels,[15,16] which means that TZP may have broad application prospects.[17] It is of concern that TZP is accompanied by a significant increase in gastrointestinal (GI) adverse events (AEs) while giving play to the therapeutic effect, and thus leads to an increase in the percentage of patients who terminate treatment early.[18] In fact, mild to moderate GI AEs are the major treatment related AEs of TZP, including nausea, vomiting, diarrhea and loss of appetite.[19] These AEs may limit the clinical application of TZP to some extent. In order to identify the specific risks of TZP on different GI AEs, this study will evaluate the GI AEs of TZP treatment for T2DM by using meta-analysis and trial sequential analysis (TSA).
2. Materials and methods
This study strictly followed the preferred reporting items of systematic evaluation and meta-analysis methods.[20]
2.1. Literature search
China National Knowledge Infrastructure (CNKI), SinoMed, VIP, Wanfang, Embase, PubMed, the Cochrane Library and Web of Science databases were searched for clinical trials on TZP for T2DM from the establishment of the database to February 2023. English subject covered tirzepatide, type 2 diabetes mellitus and Chinese subject covered tirzepatide, erxingtangniaobing (the Chinese name of T2DM). On the basis of subject words, Chinese free words were expanded by CNKI and SinoMed, and English free words were expanded by MeSH database. Then the subject words and free words were merged for retrieval. The final search strategy was “(diabetes mellitus[Title/Abstract] OR diabetes[Title/Abstract] OR diabete[Title/Abstract]) AND (tirzepatide[Title/Abstract]) “ (in English), “ SU=((tangniaobing) AND (tirzepatide)) “ (in Chinese).
2.2. Inclusion and exclusion criteria
2.2.1. Inclusion criteria.
Type of data: randomized controlled trial; Subjects: meeting the basic diagnosis of T2DM[21]; Intervention measures: Patients in the experimental group received TZP and patients in the control group received placebo or other hypoglycemic agents. Outcome indicators: GI AEs such as nausea, diarrhea, decreased appetite, vomiting, dyspepsia, constipation, abdominal pain and abdominal distension were used as endpoints.
2.2.2. Exclusion criteria.
Studies such as reviews, animal experiments, case reports and other studies. Repeated published studies. Studies published in abstract form. Studies with incomplete or unclear data.
2.3. Literature screening, data statistics and risk of bias
2.3.1. Literature screening.
All relevant literature retrieved was entered into Endnote X9. Firstly, duplicate literature was eliminated based on the title and abstract. Secondly literature that did not match the study requirements was eliminated according to the inclusion and exclusion criteria. Finally, the full text was reviewed to determine the final included literature.
2.3.2. Data statistics.
The included literature was classified and organized, and basic characteristics were extracted and entered into the data statistics table.
2.3.3. Risk of bias.
The Cochrane Risk of Bias Assessment Tool was used to assess the risk of bias according to the required entries.
All work was carried out independently by two investigators, and any disagreement was adjudicated by a third investigator.
2.4. Statistical analysis
Meta-analysis was done using Revman 5.3. Dichotomous variables were used with relative risk and 95% confidence interval as effect sizes. Continuous variables used mean difference and 95% confidence interval as effect sizes. Fixed-effects model analysis and random-effects model analysis were selected based on heterogeneity.
TSA was performed using TSA0.9.5.10 Beta. Original results were conclusive if the cumulative Z value crossed the expected information value or TSA bound.
Publication bias was analyzed using Stata 15.0. If harbord regression showed P > .1, there was no publication bias.
3. Results
3.1. Literature search results
A total of 376 literatures were retrieved. A total of 188 duplicate literatures were excluded. After reading the title, abstract and full text, 180 literatures were removed. Finally, 8 literatures were included[22–29] (Fig. 1).
Figure 1.
The flow chart of literature search and screening.
3.2. Basic materials
A total of 8 clinical studies were included, with a total sample size of 7626 cases, including 5095 cases in the experimental group and 2531 cases in the control group. The study centers were distributed in 5 continents including North America, South America, Europe, Asia and Oceania. Six studies reported 1788 cases of “TZP vs placebo,” 2 studies reported 3432 cases of “TZP vs Insulin,” and 5 studies reported 2949 cases of “TZP vs GLP-1 RA” (Table 1).
Table 1.
Table of basic characteristics of the included studies.
| Author name | Location | Patient number | Treatment duration (wk) | Disease duration (yr) | Intervention | Number randomized | Age (yr) | Male/female |
|---|---|---|---|---|---|---|---|---|
| 2018 Frias[22] | Poland, Slovakia, Puerto Rico, USA | 264 | 26 | 8.9 | Tirzepatide 5 mg | 55 | 57.9 | 34/21 |
| 7.9 | Tirzepatide 10 mg | 51 | 56.5 | 30/21 | ||||
| 8.5 | Tirzepatide 15 mg | 53 | 56.0 | 22/31 | ||||
| 9.3 | Dulaglutide 1.5 mg | 54 | 58.7 | 24/30 | ||||
| 8.6 | Placebo | 51 | 56.6 | 29/22 | ||||
| 2021 Rosenstock[23] | India, Japan, Mexico, USA | 478 | 40 | 4.6 | Tirzepatide 5 mg | 121 | 54.1 | 56/65 |
| 4.9 | Tirzepatide 10 mg | 121 | 55.8 | 72/49 | ||||
| 4.8 | Tirzepatide 15 mg | 121 | 52.9 | 63/58 | ||||
| 4.5 | Placebo | 115 | 53.6 | 56/59 | ||||
| 2022 Dahl[24] | USA, Japan, Czech Republic, Germany, Poland, Slovakia, Puerto Rico, Spain | 475 | 40 | 14.1 | Tirzepatide 5 mg | 116 | 62 | 61/55 |
| 12.6 | Tirzepatide 10 mg | 119 | 60 | 72/47 | ||||
| 13.7 | Tirzepatide 15 mg | 120 | 61 | 65/55 | ||||
| 12.9 | Placebo | 120 | 60 | 66/54 | ||||
| 2021 Del Prato[25] | Argentina, Australia, Brazil, Canada, Greece, Israel, Mexico, Poland, Romania, Russia, Slovakia, Spain, Taiwan, USA | 1995 | 52 | 9.8 | Tirzepatide 5 mg | 329 | 62.9 | 198/131 |
| 10.6 | Tirzepatide 10 mg | 328 | 63.7 | 209/119 | ||||
| 10.4 | Tirzepatide 15 mg | 338 | 63.7 | 203/135 | ||||
| 2021 Ludvik[26] | Argentina, Austria, Greece, Hungary, Italy, Poland, Puerto Rico, Romania, South Korea, Spain, Taiwan, Ukraine, USA | 1547 | 52 | 8.5 | Tirzepatide 5 mg | 358 | 57.2 | 200/158 |
| 8.4 | Tirzepatide 10 mg | 360 | 57.4 | 195/165 | ||||
| 8.5 | Tirzepatide 15 mg | 359 | 57.5 | 194/165 | ||||
| 8.1 | Insulin degludec | 360 | 57.5 | 213/147 | ||||
| 2021 Frias[27] | USA, UK, Argentina, Australia, Brazil, Canada, Israel, Mexico, | 1878 | 40 | 9.1 | Tirzepatide 5 mg | 470 | 56.3 | 205/265 |
| 8.4 | Tirzepatide 10 mg | 469 | 57.2 | 238/231 | ||||
| 8.7 | Tirzepatide 15 mg | 470 | 55.9 | 214/256 | ||||
| 8.3 | Semaglutide 1mg | 469 | 56.9 | 225/244 | ||||
| 2022 Heise[28] | Germany | 117 | 28 | 10.2 | Tirzepatide 15 mg | 45 | 61.1 | 31/69 |
| 12.7 | Semaglutide 1 mg | 44 | 63.7 | 34/77 | ||||
| 11.0 | Placebo | 28 | 60.4 | 21/75 | ||||
| 2022 Inagaki[29] | Japan | 636 | 52 | 4.5 | Tirzepatide 5 mg | 159 | 56.8 | 113/71 |
| 5.1 | Tirzepatide 10 mg | 158 | 56.2 | 119/75 | ||||
| 5.1 | Tirzepatide 15 mg | 160 | 56.0 | 132/83 | ||||
| 5.0 | dulaglutide 0.75 mg | 159 | 57.5 | 117/74 |
3.3. Risk of bias assessment
Among the eight included studies, the biases in the remaining domains were low risk except for the intervention blinding of Del Prato 2021,[25] Ludvik 2021,[26] and Frias 2021,[27] which were high risk (Fig. 2).
Figure 2.
Risk of bias summary.
3.4. Total GI AEs
3.4.1. Tirzepatide vs placebo.
Meta-analysis showed that compared with the placebo group, the total GI AEs were significantly higher in the TZP dose groups (Table 2).
Table 2.
Meta-analysis and trial sequential analysis results of tirzepatide vs placebo for total gastrointestinal adverse events.
| Outcome | Tirzepatide arm | Comparator arm | I 2 | RR (95% CI) | P | TSA | Grade |
|---|---|---|---|---|---|---|---|
| TZP 5 mg vs Placebo | |||||||
| GI AEs[22,23] | 64/176 (36.4%) | 27/166 | 2 | 2.24 (1.51, 3.33) | <.0001 | Yes | High |
| TZP 10 mg vs Placebo | |||||||
| GI AEs[22,23] | 76/172 (44.2%) | 27/166 | 69 | 3.08 (1.31, 7.25) | .01 | Yes | Moderate |
| TZP 15 mg vs Placebo | |||||||
| GI AEs[22,23] | 85/174 (48.9%) | 27/166 | 82 | 3.59 (1.16, 11.17) | .03 | No | Low |
AEs = adverse events, GI = gastrointestinal, TSA = trial sequential analysis, TZP = tirzepatide.
3.4.2. Tirzepatide vs insulin.
None of the studies included in the literature for comparison with insulin mentioned total GI AEs.
3.4.3. Tirzepatide vs GLP-1 RA.
Meta-analysis showed that compared to the GLP-1 RA group, total GI AEs of each dose were no significant differences (Table 3).
Table 3.
Meta-analysis and trial sequential analysis results of tirzepatide vs GLP-1 RA for total gastrointestinal adverse events.
| Outcome | Tirzepatide arm | Comparator arm | I 2 | RR (95% CI) | P | TSA | Grade |
|---|---|---|---|---|---|---|---|
| TZP 5 mg vs GLP-1 RA | |||||||
| GI AEs[22,27,29] | 213/684 (31.1%) | 217/682 | 54 | 0.97 (0.63, 1.50) | .89 | No | Low |
| TZP 10 mg vs GLP-1 RA | |||||||
| GI AEs[22,27,29] | 246/678 (36.3%) | 217/682 | 0 | 1.14 (0.99, 1.31) | .06 | No | Moderate |
| TZP 15 mg vs GLP-1 RA | |||||||
| GI AEs[22,27,29] | 257/683 (37.6%) | 217/682 | 75 | 1.43 (0.87, 2.34) | .15 | No | Low |
AEs = adverse events, GI = gastrointestinal, GLP-1 RA = glucagon-like peptide-1 receptor agonist, TSA = trial sequential analysis, TZP = tirzepatide.
3.5. Very common GI AEs
3.5.1. Tirzepatide vs placebo.
Meta-analysis showed that compared with placebo group, the risk of nausea and decreased appetite increased in each dose group of TZP. That of diarrhea with TZP 10 mg was also significantly increased. There was no significant difference in diarrhea with TZP 5 mg and TZP 15 mg (Table 4).
Table 4.
Meta-analysis and trial sequential analysis results of tirzepatide vs placebo for very common gastrointestinal adverse events.
| Outcome | Tirzepatide arm | Comparator arm | I 2 | RR (95% CI) | P | TSA | Grade |
|---|---|---|---|---|---|---|---|
| TZP 5 mg vs Placebo | |||||||
| Nausea[22–24] | 40/292 (13.7%) | 13/286 | 0 | 2.98 (1.63, 5.44) | .0004 | Yes | High |
| Diarrhoea[22–24] | 41/292 (14.0%) | 23/286 | 50 | 1.79 (0.84, 3.83) | .13 | No | Very low |
| Decreased appetite[22–24] | 24/292 (8.2%) | 4/286 | 0 | 5.86 (2.05, 16.71) | .001 | Yes | High |
| TZP 10 mg vs Placebo | |||||||
| Nausea[22–24] | 48/291 (16.5%) | 13/286 | 23 | 3.62 (2.01, 6.53) | <.0001 | Yes | High |
| Diarrhoea[22–24] | 44/291 (15.1%) | 23/286 | 46 | 1.88 (1.17, 3.04) | .01 | No | Moderate |
| Decreased appetite[22–24] | 36/291 (12.4%) | 4/286 | 0 | 8.93 (3.23, 24.70) | <.0001 | Yes | High |
| TZP 15 mg vs Placebo | |||||||
| Nausea[22–24,28] | 76/339 (22.4%) | 20/314 | 74 | 3.25 (1.24, 8.49) | .02 | Yes | Moderate |
| Diarrhoea[22–24,28] | 65/339 (19.2%) | 29/314 | 58 | 1.92 (0.97, 3.79) | .06 | No | Moderate |
| Decreased appetite[22–24,28] | 64/339 (18.9%) | 11/314 | 41 | 4.52 (2.55, 8.00) | <.00001 | Yes | Moderate |
TSA = trial sequential analysis, TZP = tirzepatide.
3.5.2. Tirzepatide vs Insulin.
Meta-analysis demonstrated that compared with insulin group, the risk of nausea and diarrhea with each dose of TZP were significantly increased (Table 5).
Table 5.
Meta-analysis and trial sequential analysis results of tirzepatide vs insulin for very common gastrointestinal adverse events.
| Outcome | Tirzepatide arm | Comparator arm | I 2 | RR (95% CI) | P | TSA | Grade |
|---|---|---|---|---|---|---|---|
| TZP 5 mg vs Insulin | |||||||
| Nausea[25,26] | 80/687 (11.6%) | 29/1360 | 0 | 5.75 (3.68, 8.96) | <.00001 | Yes | Very low |
| Diarrhoea[25,26] | 96/687 (14.0%) | 58/1360 | 0 | 3.27 (2.34, 4.57) | <.00001 | Yes | Very low |
| TZP 10 mg vs Insulin | |||||||
| Nausea[25,26] | 134/688 (19.5%) | 29/1360 | 53 | 9.02 (4.62, 17.62) | <.00001 | Yes | Very low |
| Diarrhoea[25,26] | 125/688 (18.2%) | 58/1360 | 0 | 4.42 (3.24, 6.03) | <.00001 | Yes | Very low |
| TZP 15 mg vs Insulin | |||||||
| Nausea[25,26] | 161/697 (23.1%) | 29/1360 | 0 | 11.28 (7.44, 17.10) | <.00001 | Yes | Very low |
| Diarrhoea[25,26] | 130/697 (18.7%) | 58/1360 | 0 | 4.60 (3.40, 6.24) | <.00001 | Yes | Very low |
TSA = trial sequential analysis, TZP = tirzepatide.
3.5.3. Tirzepatide vs GLP-1 RA.
Meta-analysis proved that compared with GLP-1 RA group, the risk of decreased appetite with TZP 5 and 10 mg and diarrhea with TZP 10 mg was significantly increased. That of nausea with TZP 5, 10, and 15 mg, diarrhea with TZP 5 and 15 mg and decreased appetite with TZP 15 mg were no significant difference (Table 6).
Table 6.
Meta-analysis and trial sequential analysis results of tirzepatide vs glucagon-like peptide-1 receptor agonist for very common gastrointestinal adverse events.
| Outcome | Tirzepatide arm | Comparator arm | I 2 | RR (95% CI) | P | TSA | Grade |
|---|---|---|---|---|---|---|---|
| TZP 5 mg vs GLP-1 RA | |||||||
| Nausea[22,27,29] | 112/684 (16.4%) | 112/682 | 35 | 1.00 (0.79, 1.26) | .98 | No | Low |
| Diarrhoea[22,27,29] | 102/684 (14.9%) | 74/682 | 50 | 1.49 (0.94, 2.38) | .09 | No | Very low |
| Decreased appetite[22,27,29] | 68/684 (9.9%) | 35/682 | 50 | 2.19 (1.14, 4.19) | .02 | Yes | Low |
| TZP 10 mg vs GLP-1 RA | |||||||
| Nausea[22,27,29] | 132/678 (19.5%) | 112/682 | 77 | 1.25 (0.68, 2.31) | .47 | No | Very low |
| Diarrhoea[22,27,29] | 103/678 (15.2%) | 74/682 | 0 | 1.40 (1.06, 1.85) | .02 | No | Low |
| Decreased appetite[22,27,29] | 68/678 (10.0%) | 35/382 | 60 | 2.33 (1.11, 4.86) | .02 | No | Very low |
| TZP 15 mg vs GLP-1 RA | |||||||
| Nausea[22,27–29] | 168/728 (23.1%) | 125/726 | 56 | 1.37 (0.94, 1.99) | .1 | No | Very low |
| Diarrhoea[22,27–29] | 101/728 (13.9%) | 78/726 | 59 | 1.36 (0.78, 2.37) | .27 | No | Very low |
| Decreased appetite[22,27–29] | 114/728 (15.7%) | 66/726 | 89 | 2.06 (0.84, 5.03) | .11 | No | Very low |
GLP-1 RA = glucagon-like peptide-1 receptor agonist, TSA = trial sequential analysis, TZP = tirzepatide.
3.6. Common GI AEs
3.6.1. Tirzepatide vs placebo.
Meta-analysis revealed a significantly increased risk of constipation and vomiting in each dose group compared to the placebo group. There is also an increased risk of dyspepsia with TZP 5 mg and TZP 10 mg. However, there were no significant differences in dyspepsia with TZP 15mg, abdominal pain with TZP 5, 10, 15mg and abdominal distension with TZP 5, 10, 15mg (Table 7).
Table 7.
Meta-analysis and trial sequential analysis results of tirzepatide vs placebo for common gastrointestinal adverse events.
| Outcome | Tirzepatide arm | Comparator arm | I 2 | RR (95% CI) | P | TSA | Grade |
|---|---|---|---|---|---|---|---|
| TZP 5 mg vs Placebo | |||||||
| Dyspepsia[22–24] | 20/292 (6.8%) | 6/286 | 0 | 3.08 (1.30, 7.31) | .01 | Yes | High |
| Vomiting[22–24] | 16/292 (5.5%) | 6/286 | 0 | 2.63 (1.04, 6.62) | .04 | No | Moderate |
| Abdominal distension[22] | 2/55 (3.6%) | 1/51 | 0 | 1.85 (0.17, 19.84) | .61 | No | Very low |
| Constipation[22–24] | 16/292 (5.5%) | 4/286 | 0 | 4.66 (1.48, 14.64) | .008 | Yes | High |
| Abdominal pain[22,24] | 1/171 (0.6%) | 1/171 | 0 | 0.93 (0.06, 14.44) | .96 | No | Low |
| TZP 10 mg vs Placebo | |||||||
| Dyspepsia[22–24] | 24/291 (8.2%) | 6/286 | 7 | 3.69 (1.58, 8.61) | .003 | Yes | High |
| Vomiting[22–24] | 20/291 (6.9%) | 6/286 | 0 | 3.31 (1.35, 8.09) | .009 | Yes | High |
| Abdominal distension[22] | 1/12 (8.3%) | 1/9 | 0 | 0.75 (0.05, 10.44) | .83 | No | Low |
| Constipation[22–24] | 20/291 (6.9%) | 3/286 | 0 | 5.79 (1.89, 17.80) | .002 | Yes | High |
| Abdominal pain[22,24] | 1//170 (0.6%) | 1/171 | 0 | 1.00 (0.14, 7.04) | 1.00 | No | Low |
| TZP 15 mg vs Placebo | |||||||
| Dyspepsia[22–24,28] | 18/339 (5.3%) | 8/314 | 0 | 1.94 (0.88, 4.25) | .1 | No | Moderate |
| Vomiting[22–24,28] | 39/339 (11.5%) | 7/314 | 0 | 5.18 (2.37, 11.34) | <.0001 | Yes | High |
| Abdominal distension[22,28] | 9/98 (9.2%) | 2/79 | 0 | 3.45 (0.77, 15.45) | .11 | No | Moderate |
| Constipation[22–24,28] | 24/339 (7.1%) | 3/314 | 0 | 5.61 (1.95, 16.13) | .001 | Yes | High |
| Abdominal pain[22,24,28] | 11/221 (5.0%) | 4/174 | 0 | 1.51 (0.49, 4.60) | .47 | No | Moderate |
TSA = trial sequential analysis, TZP = tirzepatide.
3.6.2. Tirzepatide vs insulin.
Meta-analysis displayed that compared with the insulin group, the risk of vomiting, decreased appetite with each dose of TZP and dyspepsia with TZP 15 mg was significantly increased. But no significant difference was found in dyspepsia with TZP 5 mg and TZP 10 mg (Table 8).
Table 8.
Meta-analysis and trial sequential analysis results of tirzepatide vs insulin for common gastrointestinal adverse events.
| Outcome | Tirzepatide arm | Comparator arm | I 2 | RR (95% CI) | P | TSA | Grade |
|---|---|---|---|---|---|---|---|
| TZP 5 mg vs Insulin | |||||||
| Decreased appetite[25,26] | 51/687 (7.4%) | 7/1360 | 0 | 14.70 (6.62, 32.64) | <.00001 | Yes | Very low |
| Vomiting[25,26] | 37/687 (5.4%) | 20/1360 | 0 | 3.79 (2.11, 6.79) | <.00001 | Yes | Very low |
| Dyspepsia[25,26] | 33/687 (4.8%) | 13/1360 | 58 | 7.92 (0.98, 63.86) | .05 | No | Very low |
| TZP 10 mg vs Insulin | |||||||
| Decreased appetite[25,26] | 73/688 (10.6%) | 7/1360 | 0 | 20.41 (9.14, 45.58) | <.00001 | Yes | Very low |
| Vomiting[25,26] | 61/688 (8.9%) | 20/1360 | 0 | 6.27 (3.65, 10.78) | <.00001 | Yes | Very low |
| Dyspepsia[25,26] | 59/688 (8.6%) | 13/1360 | 75 | 15.64 (0.95, 256.69) | .05 | No | Very low |
| TZP 15 mg vs Insulin | |||||||
| Decreased appetite[25,26] | 78/697 (11.2%) | 7/1360 | 0 | 21.09 (9.35, 47.53) | <.00001 | Yes | Very low |
| Vomiting[25,26] | 65/697 (9.3%) | 20/1360 | 0 | 6.57 (3.85, 11.24) | <.00001 | Yes | Very low |
| Dyspepsia[25,26] | 44/697 (6.3%) | 13/1360 | 51 | 9.87 (1.57, 62.03) | .01 | No | Very low |
TSA = trial sequential analysis, TZP = tirzepatide.
3.6.3. Tirzepatide vs GLP-1 RA.
Compared with GLP-1 RA group, meta-analysis showed that there were no significant differences in the risk of vomiting, dyspepsia, constipation and abdominal pain with each dose of TZP (Table 9).
Table 9.
Meta-analysis and trial sequential analysis results of tirzepatide vs glucagon-like peptide-1 receptor agonist for common gastrointestinal adverse events.
| Outcome | Tirzepatide arm | Comparator arm | I 2 | RR (95% CI) | P | TSA | Grade |
|---|---|---|---|---|---|---|---|
| TZP 5 mg vs GLP-1 RA | |||||||
| Vomiting[22,27,29] | 44/684 (6.4%) | 46/682 | 76 | 1.33 (0.38, 4.59) | .65 | No | Very low |
| Constipation[22,27,29] | 58/684 (8.5%) | 47/682 | 0 | 1.23 (0.85, 1.78) | .27 | No | Low |
| Dyspepsia[22,27,29] | 44/684 (6.4%) | 35/682 | 45 | 1.25 (0.82, 1.93) | .3 | No | Low |
| Abdominal pain[22,27] | 15/525 (2.9%) | 25/523 | 0 | 0.60 (0.32, 1.12) | .11 | No | Very low |
| TZP 10 mg vs GLP-1 RA | |||||||
| Vomiting[22,27,29] | 56/678 (8.3%) | 46/682 | 40 | 1.23 (0.84, 1.78) | .28 | No | Low |
| Constipation[22,27,29] | 55/678 (8.1%) | 47/682 | 54 | 1.25 (0.68, 2.31) | .47 | No | Very low |
| Dyspepsia[22,27,29] | 43/678 (6.3%) | 35/682 | 59 | 1.84 (0.64, 5.28) | .25 | No | Very low |
| Abdominal pain[22,27] | 21/520 (4.0%) | 25/523 | 0 | 0.85 (0.48, 1.48) | .56 | No | Very low |
| TZP 15 mg vs GLP-1 RA | |||||||
| Vomiting[22,27–29] | 82/728 (11.3%) | 51/726 | 74 | 1.94 (0.76, 4.95) | .17 | No | Very low |
| Constipation[22,27–29] | 51/728 (7.0%) | 55/726 | 0 | 0.92 (0.64, 1.33) | .66 | No | Low |
| Dyspepsia[22,27–29] | 58/728 (8.0%) | 49/726 | 76 | 1.07 (0.33, 3.42) | .92 | No | Very low |
| Abdominal pain[22,27,28] | 30/568 (5.3%) | 29/567 | 0 | 1.03 (0.63, 1.69) | .91 | No | Low |
GLP-1 RA = glucagon-like peptide-1 receptor agonist, TSA = trial sequential analysis, TZP = tirzepatide.
3.7. Assessment of publication bias
Harbord regression of nausea showed no significant publication bias (P = .41) (Fig. 3).
Figure 3.
Assessment of publication bias.
4. Discussion
A total of 8 clinical studies and 7626 sample sizes were included in this meta-analysis and TSA, which is the first publication to date to focus on GI AEs in the treatment of T2DM with TZP. TSA was carried out to make the results more reliable.
Meta-analysis showed that compared with placebo, nausea, diarrhea and decreased appetite are all more than 10%, which are very common AEs. The incidence of dyspepsia, vomiting, abdominal distension and constipation are all in the range of 1% to 10%, which are common AEs. The incidence of abdominal pain is 0.6% at TZP 5 mg, which is an occasional adverse event. In the 3 dose groups of TZP 5, 10, and 15 mg, the incidence of nausea is 13.7%, 16.5%, and 22.4%, diarrhea is 14.0%, 15.1%, and 19.2%, decreased appetite was 8.2%, 12.4%, and 18.9%, vomiting is 5.5%, 6.9%, and 11.5%, and constipation is 5.5%, 6.9%, and 7.1%, which means that these GI AEs induced by TZP may be dose-dependent. In addition, the incidence of dyspepsia with TZP did not increase with dose, being 6.8%, 8.2%, and 5.3% for 5, 10, and 15 mg, respectively. We speculate that the reduced incidence of dyspepsia is associated with the increased incidence of decreased appetite. Decreased appetite causes patient to consume less food, making indigestion masked. Notably, three very common adverse effects may be associated with drug-induced weight loss. The effects of antidiabetic drugs such as slowing gastric emptying and increasing satiety may be manifested as nausea and decreased appetite, both of which can lead to decreased food intake in patients.[30] Diarrhea, on the other hand, can lead to decreased absorption of nutrients. These three AEs may benefit weight loss in patients with T2DM.
In the comparison between TZP and insulin, nausea and diarrhea are very common AEs caused by TZP, and decreased appetite, vomiting and dyspepsia are common AEs. The risk of these AEs was significantly increased, except for dyspepsia with TZP 5mg and 10mg, suggesting that TZP showed a higher GI risk compared to insulin. Although previous meta-analyses have confirmed that TZP has better glycemic reduction and weight loss effects than insulin, clinicians also need to consider the potential GI safety issues of TZP, especially in some patients with underlying GI diseases.
Compared with GLP-1 RA, only TZP 5mg’s decreased appetite and TZP 10 mg’s decreased appetite and diarrhea were different, and the incidence of other AEs was not significantly different. These results are inconclusive when TSA displays the current information amount, indicating that TZP has good security. Nausea, diarrhea and decreased appetite are still very common AEs. We found that no dose-related GI AEs in the TZP 15 mg group were significantly different from those in GLP-1 RA, but there were differences in the medium and low doses, which may be caused by confounding factors. Interestingly, the results of the two included studies were quite different, with Frias 2018[22] suggesting that TZP had a significantly higher GI AEs than GLP-1 RA, while Frias 2021 supported a similar GI AEs. This contradiction may be related to the course of treatment. Frias 2018 is a short course of 26 weeks, while Frías 2021[27] is a long course of 40 weeks. The recent study published by Del Prato 2021,[25] pointed out that GI AEs induced by TZP would gradually decrease over time, which was consistent with the research results of Frias 2021. Therefore, long-term steady medication may be expected to reduce GI AEs.
Although this study strictly followed PRISMA guidelines for systematic review and meta-analysis, there are still some limitations in this study. First, the two studies of Del Prato 2021,[25] Ludvik 2021,[26] and Frías 2021[27] did not use intervention blinding, it may increase the risk of conduct bias, which may lead to a decrease in the confidence of the results. Second, the narrow inclusion criteria limited the readability of the results. Ludvik subjectively excluded patients with underlying diseases.[26] The inclusion of the sample was predominantly European-American, with a few Asians, but no African were included, which would have resulted in less generalizability of the results of this meta-analysis. Third, except Rosenstock 2021[23] and Inagaki 2022,[29] the other included subjects all used hypoglycemic agents and/or insulin. This implies that the results of the AEs of TZP were mainly based on the combination of drug, and the safety of TZP alone cannot be fully explained. Fourth, the subjects included in all the studies were overweight or obese people, so the effect of TZP on people with normal weight could not be determined.
Given the limitations of existing studies, we expect that future studies will continue to improve. First, a stratified study can be conducted to control relevant variables and stratified to explore GI AEs in patients with T2DM with different ages, disease courses and genders after taking TZP, so as to comprehensively evaluate the safety of TZP in different baseline populations. Second, research centers can be set up in Asian and African countries to further understand whether there are differences in adverse reactions to TZP among different ethnic groups, and to evaluate the potential GI risks of different ethnic groups taking TZP. Third, carry out the in-depth studies on the value of TZP in combination drugs. With the progress of T2DM, most patients need combined medication to achieve glycemic control. Therefore, it is of great significance to actively conduct clinical trials of TZP combined with other drugs to explore the safety of TZP in combined medication. Fourth, explore the efficacy and safety of TZP combined with probiotics. Probiotics can increase the richness of intestinal flora, reduce insulin resistance caused by inflammation, increase insulin secretion, reduce glucagon secretion, and thus reduce blood sugar.[31] Meanwhile, probiotics can also synthesize digestive enzymes, promote the absorption and digestion of nutrients, and reduce the occurrence of diarrhea and constipation.[32] This implies that combined probiotics may assist TZP in lowering glycemic and have the potential to reduce GI AEs of TZP. Therefore, we expect future studies to attempt to explore the value of TZP in combination with probiotics. Fifth, In the eight studies included in this article, the rate of adverse reactions with a treatment period of 52 weeks was lower than the rate of adverse reactions at 26 and 40 weeks. For the purposes of this piece the conclusion is that adverse reactions decrease with time on medication. However, due to the maximum duration of 52 weeks in the available studies, we are unable to assess the risk of GI adverse events beyond 52 weeks, after which drug adverse events may continue to decrease, or stabilize, or increase with the duration of therapy. Compared with the short course of treatment for diabetic patients who take medication for life, extending the course of treatment may be an effective means to confirm this conclusion.
5. Conclusion
The very common GI AEs of TZP are nausea, diarrhea and decreased appetite. Common GI AEs are dyspepsia, vomiting, abdominal distension, constipation and abdominal pain. The GI AEs of TZP were significantly higher than those of placebo and insulin, but comparable to GLP-1 RA. Except for diarrhea, the incidence of other AEs was positively correlated with the dose, and these AEs were gradually reduced with the passage of medication time. Long-term and stable administration of TZP may reduce the GI risks.
Author contributions
Conceptualization: Yunfeng Yu, Zhenjie Liu.
Data curation: Xinyu Yang.
Funding acquisition: Zhenjie Liu.
Methodology: Fei Zhang, Zhenjie Liu
Software: Xinyu Yang.
Supervision: Shuang Yin.
Writing – original draft: Keke Tong.
Writing – review editing: Shuang Yin, Gang Hu.
Abbreviations:
- AEs
- adverse events
- GI
- gastrointestinal
- GLP-1 RA
- glucagon-like peptide-1 receptor agonist
- T2DM
- type 2 diabetes mellitus
- TSA
- trial sequential analysis
- TZP
- tirzepatide
KKT, SY, and YFY contributed equally to this work.
This work is supported by Innovative project for graduate students of Hunan University of Chinese Medicine (2022CX159).
This study did not involve animal experiments or human studies and did not require ethical approval.
The authors have no conflicts of interest to disclose.
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
How to cite this article: Tong K, Yin S, Yu Y, Yang X, Hu G, Zhang F, Liu Z. Gastrointestinal adverse events of tirzepatide in the treatment of type 2 diabetes mellitus: A meta-analysis and trials sequential analysis. Medicine 2023;102:43(e35488).
Contributor Information
Keke Tong, Email: 498672459@qq.com.
Shuang Yin, Email: 154633296@qq.com.
Yunfeng Yu, Email: yuyunfeng1510@163.com.
Xinyu Yang, Email: 1164967851@qq.com.
Gang Hu, Email: 2944931756@qq.com.
Fei Zhang, Email: 3192911626@qq.com.
References
- [1].American Diabetes Association. 9 Pharmacologic approaches to glycemic treatment: standards of medical care in diabetes-2019. Diabetes Care. 2019;42(Suppl 1):S90–102. [DOI] [PubMed] [Google Scholar]
- [2].Aschner P, Karuranga S, James S, et al.; International Diabetes Federation's Diabetes Epidemiological Guide Writing Group. The International Diabetes Federation’s guide for diabetes epidemiological studies. Diabetes Res Clin Pract. 2021;172:108630. [DOI] [PubMed] [Google Scholar]
- [3].Sun H, Saeedi P, Karuranga S, et al. IDF Diabetes Atlas: Global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Res Clin Pract. 2022;183:109119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Maphumulo SC, Pretorius E. Role of circulating microparticles in type 2 diabetes mellitus: implications for pathological clotting. Semin Thromb Hemost. 2022;48:188–205. [DOI] [PubMed] [Google Scholar]
- [5].Gu JC, Wu YG, Huang WG, et al. Effect of vitamin D on oxidative stress and serum inflammatory factors in the patients with type 2 diabetes. J Clin Lab Anal. 2022;36:e24430. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Kocbek S, Kocbek P, Gosak L, et al. Extracting new temporal features to improve the interpretability of undiagnosed type 2 diabetes mellitus prediction models. J Pers Med. 2022;12:368. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Kim J, Kwon HS. Not control but conquest: strategies for the remission of type 2 diabetes mellitus. Diabetes Metab J. 2022;46:165–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Prasathkumar M, Becky R, Anisha S, et al. Evaluation of hypoglycemic therapeutics and nutritional supplementation for type 2 diabetes mellitus management: an insight on molecular approaches. Biotechnol Lett. 2022;44:203–38. [DOI] [PubMed] [Google Scholar]
- [9].Shafie AA, Ng CH. Cost-effectiveness of insulin glargine and insulin detemir in the 5basal regimen for naïve insulin patients with type 2 diabetes mellitus (T2DM) in Malaysia. Clinicoecon Outcomes Res. 2020;12:333–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].English WJ, Williams DB. Metabolic and bariatric surgery: an effective treatment option for obesity and cardiovascular disease. Prog Cardiovasc Dis. 2018;61:253–69. [DOI] [PubMed] [Google Scholar]
- [11].Min T, Bain SC. The role of Tirzepatide, dual GIP and GLP-1 receptor agonist, in the management of type 2 diabetes: the SURPASS clinical trials. Diabetes Ther. 2021;12:143–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Ryan DH. Next generation antiobesity medications: Setmelanotide, Semaglutide, Tirzepatide and Bimagrumab: what do they mean for clinical practice? J Obes Metab Syndr. 2021;30:196–208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Thomas MK, Nikooienejad A, Bray R, et al. Dual GIP and GLP-1 receptor agonist tirzepatide improves beta-cell function and insulin sensitivity in type 2 diabetes. J Clin Endocrinol Metab. 2021;106:388–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Bhagavathula AS, Vidyasagar K, Tesfaye W. Efficacy and safety of tirzepatide in patients with type 2 diabetes mellitus: a systematic review and meta-analysis of randomized phase II/III trials. Pharmaceuticals (Basel). 2021;14:991. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Sattar N, McGuire DK, Pavo I, et al. Tirzepatide cardiovascular event risk assessment: a pre-specified meta-analysis. Nat Med. 2022;28:591–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Ruotolo G, Roth KD, Milligan PL, et al. Effects of tirzepatide, a novel dual GIP and GLP-1 receptor agonist, on lipid profiling in patients with type 2 diabetes. Eur Heart J. 2020;41(Suppl_2):ehaa946. 3056. [Google Scholar]
- [17].Dutta D, Surana V, Singla R, et al. Efficacy and safety of novel twincretintirzepatide a dual GIP and GLP-1 receptor agonist in the management of type-2 diabetes: a Cochrane meta-analysis. Indian J Endocrinol Metab. 2021;25:475–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Karagiannis T, Avgerinos I, Liakos A, et al. Management of type 2 diabetes with the dual GIP/GLP-1 receptor agonist tirzepatide: a systematic review and meta-analysis. Diabetologia. 2022;65:1251–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [19].Baggio LL, Drucker DJ. Glucagon-like peptide-1 receptor co-agonists for treating metabolic disease. Mol Metab. 2021;46:101090. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [20].Shamseer L, Moher D, Clarke M, et al.; PRISMA-P Group. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ. 2015;350:g7647. [DOI] [PubMed] [Google Scholar]
- [21].American Diabetes Association. Standards of Medical Care in Diabetes-2022 abridged for primary care providers. Clin Diabetes. 2022;40:10–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Frías JP, Nauck MA, Van J, et al. Efficacy and safety of LY3298176, a novel dual GIP and GLP-1 receptor agonist, in patients with type 2 diabetes: a randomised, placebo-controlled and active comparator-controlled phase 2 trial. Lancet. 2018;392:2180–93. [DOI] [PubMed] [Google Scholar]
- [23].Rosenstock J, Wysham C, Frías JP, et al. Efficacy and safety of a novel dual GIP and GLP-1 receptor agonist tirzepatide in patients with type 2 diabetes (SURPASS-1): a double-blind, randomised, phase 3 trial. Lancet. 2021;398:143–55. [DOI] [PubMed] [Google Scholar]
- [24].Dahl D, Onishi Y, Norwood P, et al. Effect of subcutaneous tirzepatide vs placebo added to titrated insulin glargine on glycemic control in patients with type 2 diabetes: the SURPASS-5 randomized clinical trial. JAMA. 2022;327:534–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [25].Del Prato S, Kahn SE, Pavo I, et al.; SURPASS-4 Investigators. Tirzepatide versus insulin glargine in type 2 diabetes and increased cardiovascular risk (SURPASS-4): a randomised, open-label, parallel-group, multicentre, phase 3 trial. Lancet. 2021;398:1811–24. [DOI] [PubMed] [Google Scholar]
- [26].Ludvik B, Giorgino F, Jódar E, et al. Once-weekly tirzepatide versus once-daily insulin degludec as add-on to metformin with or without SGLT2 inhibitors in patients with type 2 diabetes (SURPASS-3): a randomised, open-label, parallel-group, phase 3 trial. Lancet. 2021;398:583–98. [DOI] [PubMed] [Google Scholar]
- [27].Frías JP, Davies MJ, Rosenstock J, et al.; SURPASS-2 Investigators. Tirzepatide versus Semaglutide once weekly in patients with type 2 diabetes. N Engl J Med. 2021;385:503–15. [DOI] [PubMed] [Google Scholar]
- [28].Heise T, Mari A, DeVries JH, et al. Effects of subcutaneous tirzepatide versus placebo or semaglutide on pancreatic islet function and insulin sensitivity in adults with type 2 diabetes: a multicentre, randomised, double-blind, parallel-arm, phase 1 clinical trial. Lancet Diabetes Endocrinol. 2022;10:418–29. [DOI] [PubMed] [Google Scholar]
- [29].Inagaki N, Takeuchi M, Oura T, et al. Efficacy and safety of tirzepatide monotherapy compared with dulaglutide in Japanese patients with type 2 diabetes (SURPASS J-mono): a double-blind, multicentre, randomised, phase 3 trial. Lancet Diabetes Endocrinol. 2022;10:623–33. [DOI] [PubMed] [Google Scholar]
- [30].Apovian CM, Okemah J, O’Neil PM. Body weight considerations in the management of type 2 diabetes. Adv Ther. 2019;36:44–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [31].Tao YW, Gu YL, Mao XQ, et al. Effects of probiotics on type II diabetes mellitus: a meta-analysis. J Transl Med. 2020;18:30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Kim SK, Guevarra RB, Kim YT, et al. Role of probiotics in human gut microbiome-associated diseases. J Microbiol Biotechnol. 2019;29:1335–40. [DOI] [PubMed] [Google Scholar]