Heart failure preserved in obese patients: semaglutide helps alot

 A recent study found that patients with heart failure with preserved ejection fraction (HFpEF) and obesity had significant heart failure symptom improvement with semaglutide, in the STEP-HFpEF trial (see chf preserved semaglutide helps NEJM2023 in dropbox, or DOI: 10.1056/NEJMoa2306963)

 

Details:

-- 529 patients having HFpEF and a BMI of at least 30 were randomized to weekly semaglutide 2.4 mg versus placebo, between March 2021 and March 2022

    -- semaglutide was initiated at 0.25 mg weekly for the first four weeks, with escalation monthly to achieve a maintenance dose of 2.4 mg per week

-- patients were recruited for this double-blind RCT from 96 sites in 13 countries in Asia, Europe, and North and South America

-- 56% women; 96% white; median age 69

-- median body weight 105.1 kg; BMI 37 (349 participants, 66%, had a BMI of at least 35)

-- median CRP level 3.8 mg/L, median NT-ProBNP 451 pg/ml

-- median LVEF 57%: 45-50% in 14%, 50-59% in 43%, at least 60% in 43%

-- atrial fibrillation 52%, hypertension 82%, coronary artery disease 20%, hospitalization in the prior 12 months for heart failure 15%,

-- medications: beta blockers 79%, diuretics 81% (loop diuretics 62%), renin-angiotensin system blockers (including ARNI) 80%, mineralocorticoid receptor antagonists 35%, SGLT-2 inhibitors 4%

-- Inclusion criteria: left ventricular ejection fractions of at least 45%; BMI at least 30; NYHA functional class II, III, or IV (NYHA class II in the cohort was 66%, class III and IV 34%); Kansas City Cardiomyopathy Questionnaire clinical summary score (KCCQ-CSS) <90 points (median KCCQ-CSS score was 59 points); a 6-minute walk distance of at least 100 m (median 6-minute walk distance was 320 m); and clinical evidence of heart failure based on direct invasive measurement (15% of the group), elevated NT-proBNP levels with echocardiographic abnormalities (72%), or hospitalizations for heart failure within the past 12 months (13%)

    -- KCCQ-CSS score is a standardized 23-item instrument that quantifies heart failure related symptoms (frequency, severity, and recent changes), physical function, quality-of-life, and social function. Results range from 0-100 with higher scores indicating fewer symptoms

-- exclusion criteria: included a history of diabetes, or a change in body weight of more than 5 kg within the 90 days before screening

-- the following results in the study were considered significant changes over the year of follow-up: 6-minute walk of at least 30 m; KCCQ-CSS at least 15, at least 10, or at least 5-point changes; the number and timing of heart failure events

-- dual primary endpoints: change from baseline in the KCCQ-CSS and change in body weight

-- secondary endpoints: change in the 6-minute walk distance, hierarchical composite endpoint that included death, heart failure events, and differences in change in the KCCQ-CSS and 6-minute walk distance; and change in CRP level

  

Results:

-- premature discontinuation of treatment occurred in 42 participants (16%) in the semaglutide group and 42 (15.8%) in the placebo group

-- 96% of each group completed the trial

-- 84% on semaglutide achieved the 2.4 milligram dose, and 98% in the placebo group

 

-- Dual primary endpoints:

    -- change in KCCQ-CSS from baseline to week 52:

        -- semaglutide group 16.6 points, placebo group 8.7 points, estimated difference 7.8 points (4.8-10.9), p<0.001

            -- secondary endpoint of degree of change:

                -- > 5-point increase in KCCQ-CSS score: 75.3 points versus 63.7 points

                -- > 10-point increase: 63.4 point versus 48.5 points

    -- change in body weight from baseline to week 52:

        -- semaglutide group -13.3%, placebo group -2.6%, estimated difference -10.7% (-11.9% to -9.4%), p<0.001

            -- secondary endpoint of amount of weight loss:

                -- >10% reduction: semaglutide 65.9%, placebo 9.5%

                -- >15% reduction: semaglutide 43.9%, placebo 2.1%

                -- >20% reduction: semaglutide 24%, placebo 0.4%

-- confirmatory secondary endpoints:

    -- change in 6-minute walk distance from baseline to week 52:

        -- semaglutide group 21.5 m, placebo group 1.2 m, estimated difference 20.3 m (8.6-32.1 m), p<0.001

    -- change in CRP level, from baseline to week 52:

           -- semaglutide group -43.5%, placebo group -7.3% (estimated difference by ratio of 0.61 (0.51- 0.72), p<0.001

    -- Change in the hierarchical endpoint (composite that included death from any cause, number and timing of heart failure events, change in KCCQ-CSS score, difference of at least 30m in 6-minute walk distance) over the year:

        -- semaglutide 60.1, placebo 34.9, estimated difference by ratio of 1.72 (1.37-2.15), p<0.001

 

-- Exploratory endpoints:

    -- percent reduction in NT-proBNP from baseline to week 52:

        semaglutide -20.9% versus placebo -5.3%

    -- >15 point improvement in KCCS-CSS: 50.6% versus 35.9%

    -- number of adjudicated heart failure events (hospitalization or urgent care visits):

       -- semaglutide one event, placebo 12 events

 

-- Adverse events:

    -- serious adverse events: semaglutide 35 participants (13.3%)

 

Commentary:

-- HFpEF is very common, occurring in more than half of all heart failure cases in the US, with increasing prevalence over time. A majority of these patients are obese.

-- other studies have found that comparing obese versus non-obese patients with HFpEF, those who are obese tend to have worse hemodynamic and clinical features, increased symptom burden, worse functional capacity, and more impaired quality of life

 

-- results of this study: there was a clear benefit of semaglutide in both heart failure related symptoms/physical limitation/quality of life as well as weight loss. And, notably (and unexpectedly) semaglutide had only half of the reported serious adverse events versus placebo. Also, the CRP levels decreased significantly, which might presage a decrease in future cardiovascular events, as per the JUPITER study in patients with low LDL levels but mildly increased CRP levels (see cad jupiter trial NEJM2008 in dropbox, or DOI: 10.1056/NEJMoa0807646)

-- obesity (and specifically visceral/abdominal obesity) is associated with an array of problems, including insulin resistance/metabolic syndrome, inflammation, and left ventricular dysfunction

-- patients with obesity and HFpEF have increased exercise pulmonary wedge pressures, higher inflammatory markers, more pronounced hypertension, increased plasma volume, and reduced venous capacitance. obesity is also associated with lower levels of BNP: BNP deficiency is found in 20-35% of outpatients with HFpEF (decreased production and increased clearance) leading to reduced capacity for vasodilation and natriuresis (see doi.org/10.1093/eurheartj/ehac121)

    -- one unexpected finding in the study was the dramatic decrease in NT-proBNP with semaglutide (-difference of -15.6% vs placebo) in patients who lost weight: prior studies have found that NT-proBNP is lower with higher BMIs and that NT-proBNP levels actually increase with weight loss. Not sure what this dramatic decrease in NT-proBNP means. perhaps that the actual improvement in cardiac function/hemodynamics is actually even more impressive than the measured NT-proBNP might suggest?

-- of note, the increase of KCCQ-CSS, a strong marker quantifying heart failure symptoms/physical functioning/quality of life, improved 7.8 points, which was more than what was found for SGLT-2 inhibitors, spironolactone or sacubitril/valsartan (the mainstays of current therapy).  Of note, studies have shown that heart failure symptom control is at least as important to patients as preventing death.

-- and, the improvement in 6-minute walking distance is an important objective marker of semaglutide benefit, and it is greater than that found by exercise training in the HF-ACTION study for patients with either reduced ejection fraction or with preserved EF (doi: 10.1016/j.jacc.2021.07.014)

-- GLP-1 receptor agonists likely work through multiple mechanisms, including weight loss, perhaps improvement of A1c levels in this cohort of people without diabetes (even those with A1c levels < 6.5, "prediabetic" range have a higher likelihood of cardiovascular disease), but a few other points:

    -- this current study suggests that weight loss may be an important factor in improving HFpEF effects, in this case as mediated by semaglutide, and is consistent with some other studies. And this suggests that HFpEF may really be very different than HF with reduced ejection fraction: there is not just a continuum of “heart failure” stratified by different values of the ejection fraction, but physiologically and therapeutically, these are a very different entity

        -- the conceptual framework for HFpEF as an LVEF >50 vs HFrEF being <40 promotes the concept that it’s all in the echocardiographic (or nuclear medicine) reading of LV function. It is certainly clear that HFrEF is related to poor cardiac ventricular contractility as a major (but not sole) cause of symptoms. But HFpEF does not have LV function being clearly so important: patients may not have evident cardiac contractility issues; the echo can appear pretty normal; these patients typically are older and have more comorbidities; and, as in this study, perhaps conceptually the issue is that this is a “metabolic disease”, as per the editorial (Pinto YM. N Engl J Med 2023: 389: 1145-46).  In this study, not only did weight decrease, but systolic blood pressure decreased 2.9mmHg, waist circumference (a marker of visceral obesity) by 9.1 cm, unfortunately there was no measure of lipids (though, of course, all of these metabolic parameters are intertwined). Perhaps that is why the best meds for HFpEF do have broader metabolic effects, such as mineralocorticoid receptor antagonists (eg spironolactone) or SGLT-2 inhibitors. it is also notable that the degree of elevation of systemic inflammation (CRP of 3.8 mg/L) was not so different from in the JUPITER trial mentioned above (CRP 4.2 mg/L), both reflecting a similar but mildly elevated level of inflammation; and GLP-1's seem to have anti-inflammatory effects (see The anti-inflammatory and immunological properties of GLP-1 Receptor Agonists - PubMed (nih.gov) )

        -- and, our conceptual framework for medications should be expansive. GLP-1 receptors do not seem to be isolated to the pancreas to stimulate insulin release when there is an oral glucose load. In fact, a small study found GLP-1 receptors in all 4 chambers of the human heart: https://pubmed.ncbi.nlm.nih.gov/29444223/ . this suggests that the myriad of GLP-1 benefits (decreasing weight and waist circumference, blood pressure, heart failure symptoms, lipids, inflammation, etc: all part of the "metabolic disease") may not just be from weight loss but perhaps also some direct effect on the heart

        -- one aside here: we often think about medication side-effects as being an annoyance unrelated to the medication but that they just happen. A perhaps more appropriate model is to see medications as having many effects in the body, perhaps related to receptors in many different parts of the body or other direct actions (eg changes in the microbiome, which can lead to dramatic body and brain effects). And “side-effects” are simply the broad effects of medications: some effects being helpful (better disease control, etc), and some being bad (toxicity to some or many other aspects of the body)

-- another tangential issue: a recent blog suggested that GLP-1's inhibits the same dopamine reward center as for alcohol, nicotine, opiates, cocaine (and likely others) as well as food consumption; they seem to help decrease alcohol addiction, but may also help with these other substance use disorders, which all stimulate the same reward center: https://gmodestmedblogs.blogspot.com/2023/09/glp-1-receptor-agonists-for-alcohol.html .

 

Limitatons:

-- this was only a one year study, so unsure about the durability of the benefit of semaglutide. However, the study did find that the trajectory of benefit for KCCQ-CSS, the 6-minute walk distance, and weight loss would suggest continued improvement. And the STEP study on weight loss with semaglutide 2.4 mg weekly found that weight continued to decrease for about 68 months, then plateaued until 104 months  https://gmodestmedblogs.blogspot.com/2022/11/obesity-semaglutide-continues-to-work.html .

-- the trial had limited participant diversity, with 96% being white, potentially limiting generalizability

-- this was a pretty small study, not powered to be able to assess important clinical outcomes, such as hospitalizations for heart failure or urgent care visits (very few events occurred)

-- the study predated widespread use of SGLT-2 inhibitors (only 4%), which could limit generalizability to the current situation

-- this study excluded patients with diabetes. it seems remarkably likely that these patients would benefit from GLP-1's, given their remarkable effect on diabetic control. And focusing on patients without diabetes ensures the outcome would be more generalizable to the overall population

-- also, the study did not include non-obese people with HFpEF. the effects of GLP-1's are still quite dramatic in non-obese patients with diabetes. it might well be the case that non-obese patients with HFpEF might still have much benefit from GLP-1's

 

So, an interesting study that brings up a few important clinical issues:

-- GLP-1 receptor agonists can be added to the list of meds for treating obese patients with HFpEF, and, from the above study, may have better overall results than spironolactone and SGLT-2's (per the KCCQ-CSS) . Though it would be really helpful to have specific head-to-head comparisons

-- there are fundamental differences conceptually for HFrEF vs HFpEF, the former resulting from left ventricular damage and decreased function, the latter more associated with metabolic risk factors

    -- this helps explain the differences in approaches to treatment

    -- it also opens the door further for aggressive treatment of these metabolic factors as a means to help with effective clinical HFpEF treatment

    -- this also may help push the development of new meds (and ?repurposing older ones) to treat HFpEF

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

  -- click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

  -- 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

if you would like access to the dropbox for articles, go to https://www.dropbox.com/scl/fo/vj803z91w1trd471h9fj8/h?rlkey=klpxdjpdhcdt3sahnpirzz730&dl=0

please feel free to circulate this to others. also, if you send me their emails (gmodest@bidmc.harvard.edu), i can add them to the list