semaglutide improves PAD symptoms in diabetes

 A new drug company-sponsored study found that semaglutide, a GLP-1 receptor agonist, significantly improved walking capacity in patients with peripheral artery disease (PAD) and type II diabetes, in the STRIDE trial (see PAD semaglutide helps Lancet2025 in dropbox, or doi.org/10.1016/ S0140-6736(25)00509-4)

 

Details:

-- 792 patients were enrolled in this double-blind, randomized, placebo-controlled trial done in 112 outpatient clinical trial sites in 20 countries in North America, Asia and Europe, from 2020 to 2024

-- all participants were at least 18 years old and had both type II diabetes and peripheral artery disease (PAD), the latter characterized as having intermittent claudication but able to walk at least 200 meters, plus an ankle-brachial index of no more than 0.90 or a toe-brachial index of no more than 0.70. Patients also had to have a maximal walking distance on a flat treadmill limited to 600 meters because of claudication

-- participants were randomly assigned to receive subcutaneous semaglutide 1.0 mg once a week for 52 weeks, versus placebo

-- baseline testing: a flat treadmill at a fixed speed of 3.2 km/h (2 mph) and assessment of maximum walking distance

-- exclusion criteria included heart failure (NYHA class 3 to 4) or musculoskeletal disorder, or recently completed or planned revascularization

-- participants were contacted by the study team to ascertain medication adherence, dispense medication, and assess outcome and safety events at weeks 4, 8, 12, 26, 49, 52, and 57; functional testing was carried out at baseline, weeks 26, 52, and 57

-- rescue medications were defined as starting on cilostazol or pentoxifylline, or increasing the dose of either of these by at least 20%

-- median age 68, 25% female, 68% white/30% Asian/98% non-Latino

-- body weight 82 kg, BMI median 29, smoking status 25% current/47% previous/28% never

-- duration of diabetes 12 years, A1c 7.1%, blood pressure 134/76, eGFR 88, LDL 69 mg/dL

-- hypertension 88%, coronary artery disease 43%, coronary revascularization 32%, myocardial infarction 19%, chronic kidney disease 15%, stroke 7%, chronic heart failure 14% (all NYHA class I or II)

-- medications: metformin 80%, SGLT-2s 35%, sulfonylureas 22%, DPP-4s 18%, thiazolidinedione 5%, insulin 32%; statins in 83%, ezetimibe in 14%, PCSK9 inhibitors in 1%

 

-- primary endpoint: ratio to baseline of the maximum walking distance at week 52 measured on a constant load treadmill with 12 degrees of inclination

-- prespecified secondary endpoints: ratio to baseline of maximum walking distance at week 57 (5 weeks after the drug was discontinued), in order to assess the persistence of the effect; change in patient-Vascular Quality of Life Questionnaire-6 (VascuQol-6) score from baseline to week 52, and the ratio to baseline of pain-free walking distance at week 52

 

--median follow-up 13.2 months

 

Results:

-- baseline results:

    --median maximum walking distance was 185.5m and the pain-free walking distance was 114.5m, but this was on the constant load treadmill at baseline (the entry criteria to the study was at least 200m on a flat treadmill, but the study tests were based on a "load treadmill" with 12% inclination)

    -- Prior history of peripheral lower limb revascularization in 24%

    -- 65% of participants rated current limitations in walking as moderate or severe

    -- 63% rated the effect of PAD on their health-related quality of life as measured by the VascuQol-6 as moderate or severe, with a baseline score of 15.0

 

-- maximum walking distance at week 52, ratio to baseline (primary endpoint):

    -- semaglutide group: 1.21 (interquartile range, IQR 0.95-1.55)

    -- placebo group: 1.08 (IQR 0.86-1.36)

        -- estimated treatment ratio (ETR) 13% more with semaglutide, 1.13 (1.06-1.21), p=0.0004

    -- the median difference between the semaglutide and placebo groups was 26.4 m (11.8-40.9); and the mean improvement was 39.9 m (13.9-66.0)

    -- the on-treatment analysis of patients not requiring rescue medications or surgery found a consistent result with this

 

-- median maximum walking distance at week 57 (5 weeks after intervention stopped), ratio to baseline (the following are confirmatory secondary endpoints):

    -- semaglutide group: 1.16 (interquartile range, IQR 0.92-1.48)

    -- placebo group: 1.10 (IQR 0.87-1.40)

        -- estimated treatment ratio (ETR) 8% more, 1.08 (1.00-1.16), p=0.038

-- median VascuQol-6 total score at week 52, change from baseline

    --semaglutide group: 2.0 (0.0-4.0)

    -- placebo group 1.0 (-1.0 to 4.0)

        -- estimated treatment difference (ETD): 1.00 (0.48-1.52), p=0.011

-- median pain-free walking distance at week 52, ratio to baseline:

    -- semaglutide group 1.21 (0.92- 1.52)

    -- placebo group: 1.10 (0.86-1.44)

        -- ETR 11% more, 1.11 (1.03-1.20), p=0.0046

 

-- supportive secondary endpoints at week 52:

    -- geometric mean ankle-brachial index, ratio to baseline:

        -- semaglutide group: 1.06

        -- placebo group: 1.02

            -- ETR 5% more, 1.05 (1.02-1.09), p=0.0037

   --  mean change from baseline in body weight, in kilograms

        -- semaglutide group: -5.2 kg

        -- placebo group: -1.2 kg

            -- ETD -4.1 kg (-4.8 to -3.4), p<0.0001

    --mean change from baseline in hemoglobin A1c

        -- semaglutide group: -0.8%

        -- placebo group: 0.2%

            -- ETD -1.0 (-1.1 to -0.8), p<0.0001

-- of note, though there was a significant difference in BMI and the ability to perform the tests, these were weak relationships suggesting that weight loss was not the main driver

-- improvement of the physical functioning domain of the SF-36: 25% greater with semaglutide, ETD 1.25 (0.26-2.24), p=0.013

-- Exploratory post-hoc endpoints at week 52 (i will only highlight the estimated comparisons for semaglutide versus placebo):

    -- mean absolute improvement in maximum walking distance: ETD 39.9 m (13.9-66.0)

    -- mean absolute improvement in pain-free walking distance: ETD 29.8 m (11.6-48.0)

    -- composite of rescue treatment, all-cause death, and major adverse limb events: 54% better, HR 0.46 (0.24-0.84)

    -- composite of rescue treatment and all-cause death: 54% better, HR 0.46 (0.24-0.85)

    -- rescue treatment overall: 52% better, HR 0.48 (0.21-0.98)

    -- rescue medication: 3 in semaglutide group (1%), 8 in placebo group (2%)

    -- rescue revascularization: 8 in semaglutide group (2%) versus 13 in placebo group (3%)

    -- all-cause death: 4 (1%) in semaglutide group, 9(2%) in placebo group, HR 0.44 (0.12- 1.34), not statistically significant, but very low numbers

    -- major adverse limb events (acute or chronic limb ischemia requiring hospitalization): 6 (2%) in semaglutide group, 5 (1%) with placebo, HR 1.21 (0.36-4.20), similarly not statistically significant

-- Some notable findings in the above graphs: the benefit of semaglutide happens quite early on, and its benefit tends to increase over the course of the 52 week study for median walking distance and median pain (figures A and C); the benefit seems to plateau but not decrease for the five week period after the study ended (figure B)

 

-- overall adverse events: 53% in the semaglutide group and 46% in the placebo group

-- overall serious adverse events reported in 19% in the semaglutide group and 20% the placebo group

    -- serious adverse events, possibly or probably treatment-related:

        --semaglutide group: 5 (1%)

        -- placebo group: 6 (2%)

            -- most frequent were serious gastrointestinal events (2 events reported by 2 patients in the semaglutide group and 5 events reported by 3 in the placebo group)

            -- no treatment-related deaths

-- adverse events leading to dose reduction: 6% in semaglutide group and 1% in the placebo group

-- treatment discontinuation: 3% of participants in the semaglutide group and 3% in the placebo group

 

Commentary:

-- lower limb peripheral artery disease (PAD) is common, affecting more than 236 million individuals worldwide, with increasing prevalence as people are aging and more people having diabetes.

-- People with diabetes are twice as likely to develop PAD than those without diabetes; PAD has been reported to be the most common initial manifestation of cardiovascular disease in those with type II diabetes (see Verma S et al. Diabetes Ther 2024; 15: 1893–1961)

-- The effects of PAD are quite profound: heightened risk of other cardiovascular events as well as PAD-related amputation.

-- One important reason to detect PAD early is that patients may not be aware of having slowly progressive PAD for several reasons:

    -- the functional decline may be masked if it were slow enough that patients accommodate to the slowly progressive mild PAD changes

    -- some studies have suggested that even patients with severe degrees of disability do not acknowledge the problem

    -- patients with diabetes and PAD may well have other significant cardiorespiratory conditions that impair function and limit their ability to tease out the role of potential PAD vs their heart failure/frailty/etc

-- there are few current therapies that help patients with PAD. The strongest evidence seems to be for exercise therapy, including home-based programs.

    -- There are also medications, the best being cilostazol, which was approved in 1999 and is the only current guideline-recommended medical therapy in the US; however, half of patients treated with this discontinue treatment, and there is a black box warning for heart failure (a not so rare comorbidity). The only other medication is pentoxifylline which has been found in some studies to be relatively ineffective and is not recommended

-- this current study found that once weekly subcutaneous semaglutide at 1.0 mg versus placebo improved the maximum walking distance by 13%, with a median improvement of 26.4m and a mean improvement of 39.9m for patients on a graded treadmill with a 12% incline, the equivalent of walking up a steep hill. this improvement occurred early in the treatment and the improvement continued to increase over time

    -- it is important to note that these benefits did continue to improve over time; this is an important finding given that semaglutide would likely be given for the long-term In patients with diabetes, suggesting continuing PAD improvement in the future

    -- In addition, there were significant improvements in the quality of life and lower rates of needing rescue treatments, all in the setting of a generally tolerable medication without many severe side effects

    -- there was some weight loss in the group on semaglutide, 4.1 kg, likely not much because of the fact that the median BMI was 29 and the dose of semaglutide was not maximized. Of note, the weight loss achieved in the study was not a major factor in the measured improvements in PAD

-- the 12% incline reflects a 2 to 3 times greater change in what would be observed on a flat surface, as in the 6-minute walk test (the improvements in maximum walking distance in the study would reflect a 20-meter improvement in the 6- minute walk test!!)

-- the effects of semaglutide seem to be direct vascular effects given that the improvements in pain-free walking distance was accompanied with improvements in the ankle-brachial index

-- the ischemia related to PAD does promote inflammation, which is associated with endothelial dysfunction as well as having adverse effects on the micro-circulation and skeletal muscle metabolism, all affecting function

    -- semaglutide is an anti-inflammatory, as well as having a documented role in improving endothelial function: https://pmc.ncbi.nlm.nih.gov/articles/PMC10520195/ 

-- and pretty clearly the GLP-1 receptor agonist benefits the broad cardiometabolic Issues, which is a huge benefit over cilostazol for PAD therapy in patients with diabetes

-- one lingering concern with GLP-1 receptor agonists is the question about muscle loss/sarcopenia. Studies have suggested that the degree of muscle loss with GLP-1s is similar to that found with other methods leading to similar amounts of weight loss. This study found improved physical function on the treadmill with no evidence of detrimental effects on the muscle. There was unfortunately no information on the amount of exercise people were doing at baseline, but given how generally ill these patients were with their many medical problems, the likelihood is that it was not a whole lot

 

Limitations:

-- they started patients on fixed dose semaglutide 1.0 mg. The usual approach is to start at 0.25 and slowly increase that in increments to 2.0 mg if needed; this scaling up minimizes adverse effects, promotes adherence to the medication, and maximizes potential benefit. Starting at this higher dose of 1.0mg might lead to more adverse effects and understate the potential benefit. and the 2.0 mg semaglutid dose might have found even more improvement

    -- also, the findings in this study would likely be even better with the more potent tirzepatide, though it would be important to prove that in studies

-- many of these patients were on sulfonylureas (22%) and/or insulin (32%). These two medications can lead to insulin resistance, which is associated with incident peripheral artery disease (https://pmc.ncbi.nlm.nih.gov/articles/PMC3563259/), and many studies have found increased risk of atherosclerotic disease with these medications (hence the current appropriate push to use GLP-1 receptor agonists or SGLT2 inhibitors as first-line medications for diabetes, along with metformin). so the use of medications increasing insulin resistance should be avoided whenever possible, and inclusion in this study likely distorted the results somewhat

-- as with many patients with PAD, this group of patients who were quite sick with multiple comorbidities, which may limit the application of their results to patients who are less sick

-- the achieved LDL in this group was 69 mg/dL. Though this is acceptable per the guidelines, I personally would attempt to achieve a lower LDL given the multiple comorbidities of these patients, including 25% being current smokers: observational data suggests that LDLs even below 20 mg/dL are associated with clinical benefit with essentially no increased harm in patients with evident coronary artery disease: https://gmodestmedblogs.blogspot.com/2018/08/very-low-ldl-levels-benefit-without-harm.html , as well as the more recently released extension of the Fourier study: https://gmodestmedblogs.blogspot.com/2023/11/pcsk9-inhibitors-vs-statins-and-diabetes.html

-- though this study was done in many different countries and there were explicit attempts to include a diverse group of patients, the reality was that very few people were included other than white or Asian patients. This limits potential generalizability to other groups

-- this study had important inclusion criteria, most significantly people with type II diabetes. this is a very large group of persons with PAD, but it significantly decreases the generalizability of these results to those without diabetes

-- there is a training effect in a study such as this, where both groups are using treadmills regularly and both groups benefit from continued use of the treadmills. In clinical practice, many patients might well use the semaglutide without training, and the relative benefits in these patients might be somewhat different without the regular exercise

-- though the study was double-blinded, it may well have been evident to patients that they were on semaglutide, either by the associated adverse effects or the attendant weight loss; this could potentially increase the placebo effect

-- the patients in this group were in the mild to moderate stage of PAD. It is unclear what the benefits would be in patients who were more severely affected by their PAD

-- and, there is the bias of being in a study: both the selection bias of patients electing to be in the study and the bias conferred by having a study team that contacted them often (7 times during the course of the study). Both of these would be quite different in most non-study clinical settings and could affect generalizability to these settings

 

So,

-- this article presents a potentially very useful medical therapy for PAD in patients with diabetes, though it is clear that exercise is extremely important and should be part of the therapy for those with PAD

-- smoking is also an extremely important risk factor for the development of PAD and is an important target for us in clinical care

    -- 25% of the participants in this study were active smokers and this should lead to very active patient engagement in smoking cessation. these patients should also have more vigorous lipid control and other cardiovascular risk factors

-- it is also pretty clearly important that we clinicians should identify PAD early on so as to be effective in preventing the significant morbidity and mortality associated with progressive PAD

    -- this might mean more intensive questioning with very specific goals (e.g., asking the patients if they are able to walk one or two blocks as quickly as they used to be able to), checking the robustness of their peripheral pulses in patients more regularly than on annual exams, perhaps doing vascular assessments more regularly (there are relatively inexpensive handheld ultrasonic devices to measure the ankle-brachial index), more routinely doing a 6-minute walk test, and undoubtedly other potential ways.

geoff

---------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

to get access to all of the blogs:  go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them