People With Lowest Physical Functioning Scores Showed Greatest Improvement After Tirzepatide Treatment

The SURMOUNT‐1, ‐3, and ‐4 phase 3, multicenter, randomized, placebo‐controlled, double‐blind trials evaluated the efficacy and safety of tirzepatide for chronic weight management [22].

Adult participants (≥ 18 years old) with obesity (body mass index [BMI] ≥ 30 kg/m² or ≥ 27 kg/m² and ≥ 1 ORCs) and a history of ≥ 1 self‐reported unsuccessful dietary attempt to lose body weight were included in these trials [22]. The detailed study design, eligibility criteria, and endpoints for these trials have been published previously [15, 16, 17].

All trials were conducted in accordance with the good clinical practice guidelines and the principles of the Declaration of Helsinki. Each of the participating sites received approval from an independent ethics committee or institutional review board. All participants provided written informed consent prior to trial participation.

In SURMOUNT‐1, 2539 participants were randomly assigned (1:1:1:1) to receive tirzepatide (5, 10, or 15 mg) or placebo for 72 weeks [15]. In SURMOUNT‐3, 579 participants who achieved ≥ 5.0% weight reduction after a 12‐week intensive lifestyle modification program were randomly assigned (1:1) to receive tirzepatide maximum tolerated dose (MTD: 10 or 15 mg) or placebo for 72 weeks [16]. In SURMOUNT‐4, 670 participants who attained tirzepatide MTD during the 36‐week lead‐in period were randomly assigned (1:1) to continue receiving tirzepatide MTD or switch to placebo for an additional 52 weeks to assess maintenance of weight reduction [17]. In all three SURMOUNT (1, 3, and 4) trials, tirzepatide was administered subcutaneously once weekly. All participants received study treatment as an adjunct to lifestyle counseling (500 kcal/day deficit diet and ≥ 150 min of physical activity/week) [22].

Patient‐reported physical function was assessed through Short Form‐36 Version 2 Health Survey acute form (SF‐36v2) Physical Functioning domain score and the Impact of Weight on Quality of Life‐Lite‐Clinical Trials Version (IWQOL‐Lite‐CT) Physical Function composite score (Table ) in SURMOUNT‐1, ‐3, and ‐4 trials. In this post hoc analysis, participants were categorized into quartiles (Q) based on baseline SF‐36v2 Physical Functioning scores and the IWQOL‐Lite‐CT Physical Function scores within each study. The lower the quartile, the higher the degree of physical function limitation. S2

The study outcomes assessed by baseline quartiles of SF‐36v2 Physical Functioning domain scores and the IWQOL‐Lite‐CT Physical Function composite scores included (i) duration of obesity and presence of ORCs at baseline; (ii) efficacy measures (change from baseline in body weight, waist circumference, and BMI); and (iii) PRO measures (change from baseline in SF‐36v2 domain scores; and IWQOL‐Lite‐CT Total score and Physical Function and Psychosocial composite scores). Additionally, the correlation between weight reduction and improvement in SF‐36v2 Physical Functioning scores was assessed by baseline quartiles of SF‐36v2 Physical Functioning scores.

This post hoc analysis utilized data from Weeks 0 to 72 in SURMOUNT‐1 and SURMOUNT‐3 and from Weeks 0 to 88 in SURMOUNT‐4. SURMOUNT‐1 was the primary focus of the current analysis given its largest trial sample size. SURMOUNT‐3 and SURMOUNT‐4 evaluated the consistency of the primary findings. As tirzepatide MTD arm in SURMOUNT‐3 and SURMOUNT‐4 was predominantly 15 mg, only results from the 15 mg arm in SURMOUNT‐1 are presented for efficacy and PRO measures. For SURMOUNT‐4, only results of the tirzepatide arm are reported since participants in the comparator arm received both tirzepatide (Weeks 0–36) and placebo (Weeks 36–88).

Statistical analyses were conducted using the R Version 4.2.2 software. Efficacy and PRO measures were assessed in the efficacy analysis set (randomized population who received at least one dose of the study drug, excluding data after study drug discontinuation). All study outcomes were assessed among participants with baseline SF‐36v2 Physical Functioning score or IWQOL‐Lite‐CT Physical Function composite score.

Baseline clinical characteristics were compared across quartiles using analysis of variance (ANOVA) for continuous variables and chi‐square test for categorical variables. For efficacy and PRO measures, treatments were compared (tirzepatide versus placebo) using analysis of covariance (ANCOVA) within each quartile controlling for stratification factors and baseline value of the outcome, with the last observation carried forward for missing data imputation. Pearson's correlation (r) between weight reduction and improved physical function was calculated using pooled data from the tirzepatide treatment arm in SURMOUNT‐1, ‐3, and ‐4. The correlation thresholds were defined as < 0.25 for weak, 0.25–0.50 for mild, 0.50–0.75 for moderate, and > 0.75 for strong. The SURMOUNT‐1, ‐3, and ‐4 trials were not powered for the current post hoc analysis; thus, the reported p values should be regarded as exploratory rather than confirmatory.

Baseline demographics and clinical characteristics were generally similar across SURMOUNT‐1 (N = 2539), SURMOUNT‐3 (N = 579), and SURMOUNT‐4 (N = 670) trials (Table 1). The participants had a mean age of 44.9 to 47.7 years, with the majority being females (62.9% to 70.6%). The mean BMI ranged from 35.9 to 38.4 kg/m², mean waist circumference from 109.4 to 115.2 cm, and the mean body weight was between 101.9 and 107.3 kg. Participants had obesity for a mean duration of 14.4 to 15.5 years (Table 1).

Table 2 presents the quartiles of SF‐36v2 Physical Functioning scores and IWQOL‐Lite‐CT Physical Function composite scores at baseline. The baseline scores were comparable between the tirzepatide and placebo groups across quartiles within each trial. The distribution of the baseline SF‐36v2 Physical Functioning scores varied across the trials, with SURMOUNT‐3 demonstrating the highest degree of ceiling effect (i.e., achieving highest score): SURMOUNT‐4 (Q1, median, Q3) = 42.2, 49.9, 53.7; SURMOUNT‐1 = 46.0, 51.8, 55.7; and SURMOUNT‐3 = 49.9, 53.7, 57.6. Similarly, SURMOUNT‐3 had the highest degree of ceiling effect for the baseline IWQOL‐Lite‐CT Physical Function composite scores: SURMOUNT‐4 (Q1, median, Q3) = 40.0, 60.0, 80.0; SURMOUNT‐1 = 45.0, 65.0, 85.0; and SURMOUNT‐3 = 60.0, 75.0, 90.0.

Participants with lower baseline SF‐36v2 Physical Functioning scores had a longer duration of obesity (SURMOUNT‐1: Q1 to Q4 mean = 16.8 to 12.0 years, p < 0.001 for comparison across the quartiles; SURMOUNT‐3: 16.1 to 12.8 years, p < 0.05; SURMOUNT‐4: 16.8 to 13.7 years, p < 0.05) (Figure 1).

Similarly, participants who had lower IWQOL‐Lite‐CT Physi cal Function composite scores at baseline had a longer duration of obesity in SURMOUNT‐1 (Q1 to Q4 mean = 17.2 to 12.2 years, p < 0.001), SURMOUNT‐3 (17.3 to 13.9 years, p = 0.055), and SURMOUNT‐4 (17.1 to 13.5 years, p < 0.05) (Figure S1).

Lower baseline SF‐36v2 Physical Functioning scores and IWQOL‐Lite‐CT Physical Function composite scores were associated with a higher prevalence of ORCs including hypertension, anxiety/depression, osteoarthritis, asthma/chronic obstructive pulmonary disease (COPD), and OSA in SURMOUNT trials (Figure 2).

Tirzepatide treatment was associated with greater reductions in body weight, BMI, and waist circumference compared to placebo across all quartiles of SF‐36v2 Physical Functioning (Figure 3A) and IWQOL‐Lite‐CT Physical Function (Figure 3B) composite scores (p < 0.05).

The least‐squares mean (LSM) percent changes in body weight (kg) with tirzepatide vs. placebo were consistent across baseline quartiles of SF‐36v2 Physical Functioning scores in SURMOUNT‐1 (Week 72: −20.1% to −22.8% vs. −2.1% to −3.3%), SURMOUNT‐3 (Week 72: −18.1% to −21.2% vs. 1.5% to 3.0%), and SURMOUNT‐4 (Week 88 [tirzepatide]: −24.0% to −27.9%) (Figure 3A). Similarly, the LSM changes in waist circumference were comparable across quartiles in SURMOUNT‐1 (tirzepatide vs. placebo: −17.2 to −20.2 cm vs. −2.7 to −4.3 cm), SURMOUNT‐3 (−14.6 to −17.1 cm vs. 0.4 to 1.4 cm), and SURMOUNT‐4 (−20.2 to −25.3 cm) (Figure 3A). The LSM changes in BMI also followed a similar pattern: SURMOUNT‐1 (−7.2 to −9.0 kg/m² vs. −0.8 to −1.3 kg/m²), SURMOUNT‐3 (−6.6 to −8.1 kg/m² vs. 0.7 to 1.0 kg/m²), and SURMOUNT‐4 (−8.8 to −11.2 kg/m²) (Figure 3A).

At Week 72, the LSM percent changes in body weight (kg) with tirzepatide vs. placebo were similar across baseline quartiles of IWQOL‐Lite‐CT Physical Function composite scores in SURMOUNT‐1 (−20.2% to −22.9% vs. −2.0% to −3.0%), SURMOUNT‐3 (−18.4% to −21.0% vs. 0.9% to 3.8%), and SURMOUNT‐4 (tirzepatide at Week 88: −24.1% to −28.4%) (Figure 3B). Similarly, the LSM changes in waist circumference were consistent across quartiles in SURMOUNT‐1 (tirzepatide vs. placebo: −17.1 to −21.1 cm vs. −2.8 to −4.4 cm), SURMOUNT‐3 (−14.3 to −16.9 cm vs. 0.1 to 1.6 cm), and SURMOUNT‐4 (tirzepatide: −19.4 to −25.6 cm). The LSM changes in BMI followed a similar pattern, with comparable reductions across quartiles with tirzepatide vs. placebo in SURMOUNT‐1 (−7.3 to −9.0 kg/m² vs. −0.6 to −1.2 kg/m²), SURMOUNT‐3 (−6.4 to −7.9 kg/m² vs. 0.4 to 1.3 kg/m²), and SURMOUNT‐4 (tirzepatide at Week 88: −8.7 to −11.7 kg/m²) (Figure 3B).

Despite similar changes in efficacy measures across baseline quartiles of physical function, improvements in physical function were more pronounced with tirzepatide among participants with lower baseline physical function. Particularly, participants with lower baseline SF‐36v2 Physical Functioning scores (Q1) achieved greater LSM improvements in IWQOL‐Lite‐CT Physical Function composite scores in SURMOUNT‐1 (Week 72 [tirzepatide vs. placebo]: Q1 = 38.0 vs. 18.8; Q4 = 10.3 vs. 4.1; p < 0.05 for tirzepatide vs. placebo in all quartiles), SURMOUNT‐3 (Week 72 [tirzepatide vs. placebo]: Q1 = 27.7 vs. 11.7; Q4 = 5.7 vs. −2.6; all p < 0.05), and SURMOUNT‐4 (tirzepatide; Week 88: Q1 = 42.5; Q4 = 12.6) (Figure 4A). Similarly, participants with lower baseline SF‐36v2 Physical Functioning scores achieved greater LSM improvements in these scores after tirzepatide treatment in SURMOUNT‐1 (Q1 = 12.5 vs. 8.3; Q4 = −0.8 vs. −1.2; all p < 0.05 except Q4), SURMOUNT‐3 (Q1 = 9.0 vs. 3.1; Q4 = −0.7 vs. −2.5; all p < 0.05), and SURMOUNT‐4 (tirzepatide: Q1 = 16.5; Q4 = 0.5) (Figure 4B). These findings for improvement in physical function were consistent for baseline quartiles of IWQOL‐Lite‐CT Physical Function composite scores (Figure 5).

Participants with lower baseline SF‐36v2 Physical Functioning scores achieved greater LSM improvements in IWQOL‐Lite‐CT Psychosocial composite scores in SURMOUNT‐1 (Week 72 [tirzepatide vs. placebo]: Q1 = 35.3 vs. 18.8; Q4 = 14.0 vs. 8.0; all p < 0.05), SURMOUNT‐3 (Week 72 [tirzepatide vs. placebo]: Q1 = 33.9 vs. 11.7; Q4 = 11.8 vs. −1.5; all p < 0.05), and SURMOUNT‐4 (tirzepatide; Week 88: Q1 = 45.1; Q4 = 17.3) (Figure 4A).

Participants with lower baseline SF‐36v2 Physical Functioning scores achieved greater LSM improvements in IWQOL‐Lite‐CT Total score after tirzepatide treatment in SURMOUNT‐1 (Q1 = 35.6 vs. 18.4; Q4 = 12.4 vs. 6.7; all p < 0.05), SURMOUNT‐3 (Q1 = 31.9 vs. 11.1; Q4 = 10.0 vs. −1.9; all p < 0.05), and SURMOUNT‐4 (tirzepatide: Q1 = 43.3; Q4 = 15.3) (Figure 4A). Improvements in SF‐36v2 General Health were also more pronounced in participants with lower baseline SF‐36v2 Physical Functioning scores in SURMOUNT‐1 (Q1 = 7.0 vs. 2.9; Q4 = 1.9 vs. −0.7; all p < 0.05), SURMOUNT‐3 (Q1 = 6.6 vs. 0.5; Q4 = 0.7 vs. −2.9; all p < 0.05), and SURMOUNT‐4 (tirzepatide: Q1 = 9.7; Q4 = 3.3) (Figure 4B). These improvements in IWQOL‐Lite‐CT Total and Psychosocial composite scores, and SF‐36v2 General Health were consistent for baseline quartiles of IWQOL‐Lite‐CT Physical Function composite scores (Figure 5).

Participants with lower baseline physical function scores showed greater improvements in other SF‐36v2 domain scores (Role‐Physical, Bodily Pain, Vitality, Social Functioning, Role‐Emotional, and Mental Health) after tirzepatide treatment (Table ). S3

Based on quartiles of baseline SF‐36v2 Physical Functioning scores, a weak to mild correlation was observed between weight reduction and improved SF‐36v2 Physical Functioning scores among tirzepatide‐treated participants pooled from SURMOUNT‐1, ‐3, and ‐4. The strength of correlation decreased from Q1 to Q4 (Q1: r = −0.26; Q2: r = −0.20; Q3: r = −0.19; Q4: r = −0.04).

This is one of the first papers to look at physical functioning from a physiologically induced versus weight loss‐dependent mechanism. Although the degree of ceiling effect varied across the trials, the study findings and trends remained consistent, allowing a reasonable degree of generalization. Similarly, the use of the multiple SURMOUNT phase 3 trials enabled the cross‐study evaluation on whether the observed between‐quartile differences in efficacy measures were artifacts or systemic. We utilized both generic (SF‐36v2) and disease‐specific (IWQOL‐Lite‐CT) PROs to categorize participants based on their baseline physical function, and consistent results were observed across these PROs. This study found that the degree of physical dysfunction at baseline is a determinant of the extent of improvement. The results suggest that individuals with worse perceived physical functioning at baseline have greater potential for recovery, and the data confirm that such improvement is achievable, regardless of the initial degree of impairment. However, this is not always seen in other severe baseline conditions. For instance, in diabetes management, individuals with high HbA1c levels may achieve smaller relative improvements (the change of mean HbA1c level divided by mean HbA1c level before therapy) due to more complex physiological or behavioral barriers than those with moderately elevated levels [36]. Similarly, people with severe baseline physical impairments, like advanced joint degeneration, may experience less improvement in mobility after physical therapy compared to those with milder impairments due to their more limited recovery potential [37]. These findings are clinically relevant, as they underscore the importance of the observed inverse relationship and highlight the need for further investigation into its underlying mechanisms and predictors of improvement in physical functioning. A better understanding of these determinants can support more effective treatment decision‐making, particularly when enhancing physical function is a key objective.

As this was a post hoc analysis, the findings are hypothesis generating rather than confirming. The study relied on self‐reported physical functioning instead of objective measures of physical functioning. Measuring both actual physical functioning and reported physical functioning scores in future studies can provide a better understanding of the effect of obesity medications on physical function decline, a symptom commonly associated with obesity. Furthermore, integrating these measurements with the assessment of changes in muscle mass and muscle function can help elucidate the relationship between the decline in actual physical function and the decline in perceived physical function. Understanding this relationship is important because various factors, such as improved quality of life and mental health, can contribute to perceived improvements in physical functioning.