Beyond conventional therapies: Gut microbiota modulation and macromolecular drugs in the battle against cardiometabolic diseases

Abstract

Cardiometabolic diseases (CMDs) represent an ongoing major global health challenge, driven by complex interactions among genetic, environmental, microbiome-related, and other factors. While small-molecule drugs and lifestyle interventions can provide clinical benefits, they are possible to be constrained by the limited druggability of key target proteins, the potential risks of off-target effects, and difficulties in maintaining long-term adherence. In recent years, gut microbiota modulation and macromolecular drugs have emerged as promising therapeutic strategies. Gut microbiota modulation (e.g., probiotics, synbiotics, or natural products) exerts systemic metabolic and immune effects, supporting a therapeutic approach targeting multiple diseases. Meanwhile, macromolecular drugs (e.g., peptides, antibodies, and small nucleic acids) offer precise, pathway-targeted interventions. Despite advancements, limitations remain in addressing ethical considerations in microbiota modulation and optimizing targeted delivery systems, all of which may hinder clinical translation. Here, we provide a comprehensive overview of therapeutic approaches for CMDs, with a focus on obesity, type 2 diabetes mellitus (T2DM), and atherosclerosis (AS). The review is structured around three key aspects: i) conventional therapies, including small-molecule drugs and lifestyle interventions; ii) emerging therapies encompassing gut microbiota modulation, macromolecular drugs, and their interactions; and iii) challenges and opportunities for comorbidity management, microbiota ethics, and artificial intelligence (AI)-driven therapeutic optimization. We hope this review enhances the understanding of small-molecule drugs, lifestyle interventions, gut microbiota modulation, and macromolecular drugs in the management of CMDs, thereby fostering medical innovation and contributing to the development of system-based comprehensive therapeutic paradigms.

Graphical abstract

Highlights

• •Summarize conventional therapies for CMDs: small-molecule drugs and lifestyle interventions.
• •Uncover strategies for gut microbiota modulation in CMDs through direct and indirect methods.
• •Macromolecular drugs represent novel and promising approaches for the treatment of CMDs.
• •Explore future directions in comorbidity management, microbiota ethics, and AI-driven therapeutic optimization.

1. Introduction

Cardiometabolic diseases (CMDs) cover a broad array of cardiovascular and metabolic conditions that, despite appearing distinct, are closely interconnected, such as obesity, type 2 diabetes mellitus (T2DM), and atherosclerosis (AS) [1]. The prevalence and disease burden of obesity, T2DM, and AS are as follows. Obesity rates have persistently increased, amplifying the risk for cardiovascular diseases (CVDs), T2DM, several cancers, and multi-organ metabolic dysfunction [2]. Diabetes poses a significant global health burden, affecting over 500 million people in 2021 and projected to impact more than 1.3 billion by 2050 [3]. As a widespread and escalating condition, diabetes is not only prevalent across all demographics but also a major contributor to CVDs, such as AS and stroke [3]. AS is the most common chronic illness worldwide and the dominant pathophysiological mechanism of CVDs [4]. Both the American Heart Association's Life's Simple 7 and Life's Essential 8 indices have been shown to correlate significantly with incident CVDs and exhibit comparable predictive performance [5]. Given its more concise structure and long-standing familiarity, Life's Simple 7 may offer a practical advantage for routine implementation by clinicians and wider adoption by the general population [5]. Although Life's Simple 7 health behaviors offer basic prevention strategies (Fig. 1) against cardiometabolic risk factors [6], the cumulative burden of CMDs, especially their complications associated with increased mortality, underscores the urgent need for more effective therapeutic interventions [7].

Fig. 1.

Risk factors associated with cardiometabolic diseases (CMDs) and contributions of Life's Simple 7 healthy behaviors: (A) risk factors and (B) 7 healthy behaviors.

Contemporary CMDs management relies on small-molecule drugs and lifestyle interventions. Despite advantages in oral bioavailability and cost-effectiveness, small-molecule drugs frequently suffer from pleiotropic effects and diminishing returns. For instance, metformin, the first-line T2DM therapy, induces gastrointestinal intolerance in 1.2%–5.0% of patients, necessitating treatment discontinuation [8]. Lifestyle interventions are the cornerstone of CMDs prevention and management. However, systemic barriers often constrain their effectiveness, including socioeconomic disparities, behavioral inertia, and limited access to personalized nutrition and exercise programs [9]. Targeted protein degradation technology that leverages ubiquitin-proteasome and autophagy-lysosomal systems is an emerging modality exhibiting the potential for metabolic diseases modulation [10]. Although it has promising potential for selective protein elimination, this relatively nascent technology faces several significant challenges, including poor membrane permeability due to high-molecular-weight compounds, restricted tissue distribution, potential development of proteasome-mediated resistance, and complexities in optimizing the degradation kinetics for metabolic targets [10]. Research has increasingly focused on two promising frontiers to address these restrictions and explore novel therapeutic pathways: gut microbiota modulation and macromolecular drugs [[11], [12], [13], [14], [15], [16], [17], [18]].

On one hand, the gut microbiota, a sophisticated and ever-evolving ecosystem comprising trillions of microorganisms, is crucial for maintaining metabolic homeostasis and overall health by influencing physiological pathways [11]. Over the past decades, significant advancements in microbiome research have illuminated gut microbiota's active involvement in metabolism and immune regulation, overturning earlier perceptions that considered it merely a passive entity [12]. Dysbiosis, an imbalance in gut microbial communities, is closely associated with many CMDs [13]. The mechanisms and standard features of these CMDs are, to some extent, influenced by gut microbiota [14]. Consequently, understanding the specific functions of the gut microbiota in metabolic homeostasis opens new avenues for preventing and treating these health concerns.

On the other hand, macromolecular agents have shown remarkable efficacy in both disease diagnosis and treatment. For example, a nine cyclic peptide (TCP-1) exhibits high specificity for early-stage colon cancer by binding tightly to keratin type II cytoskeletal 5 in tumor tissues, enabling precise diagnosis with minimal off-target effects [15]. Meanwhile, macromolecular drugs (such as peptides, proteins, and small nucleic acids) are also emerging as transformative therapies due to their high specificity for disease-related targets. Examples include glucagon-like peptide-1 receptor agonists (GLP-1 RAs) for glycemic and weight control, and proprotein convertase subtilisin/kexin type 9 inhibitors (PCSK9i) for hyperlipidemia management [16,18]. Furthermore, these agents enable precise modulation of disease-related pathways with a favorable safety profile, as clinical trials have demonstrated sustained low-density lipoprotein cholesterol (LDL-C) reduction with small interfering RNA (siRNA)-based inclisiran [17].

By focusing on obesity (characterized by energy imbalance), T2DM (marked by glucose dysregulation), and AS (driven by dyslipidemia and vascular injury), this review systematically addresses: i) conventional therapeutic strategies and their limitations; ii) emerging therapies, including gut microbiota modulation strategies, advances in macromolecular therapeutics, and the bidirectional interactions between gut microbiota and macromolecular drugs; and iii) future directions for therapeutic optimization with an emphasis on comorbidity management, gut microbiota ethics, and artificial intelligence (AI). This review offers a comprehensive overview of the latest developments in treating CMDs, which aims to enhance the understanding of the gut microbiota modulation and macromolecular drugs associated with CMDs and to promote further growth in this domain. It also tries to establish a way toward AI-assisted integrated treatment strategies that merge microbial ecology with drugs for improving cardiometabolic outcomes.

2. Conventional therapies in CMDs

Conventional therapeutic approaches primarily involve small-molecule drugs and lifestyle interventions. Obesity, characterized by excessive body fat accumulation and measured using body mass index (BMI), is a significant risk factor for metabolic disorders (e.g., T2DM and AS) [19]. T2DM is marked by chronic hyperglycemia mainly resulting from insulin resistance, which can lead to cardiovascular complications. In some cases, certain antidiabetic drugs (e.g., insulin, sulfonylurea, and thiazolidinediones (TZDs)) may induce obesity [20,21]. To the best of our knowledge, few studies have reported on the direct effects of AS on T2DM and obesity. This section examines the established treatments for obesity, T2DM, and AS and their limitations. Fig. 2 illustrates the primary small-molecule drugs for the three diseases, relationship among them, and possible limitations, which necessitate the exploration of more innovative therapies (such as gut microbiota modulation and macromolecular drugs) and develpment of comprehensive therapeutic strategies.

Fig. 2.

The common small-molecule drugs for obesity, type 2 diabetes mellitus (T2DM), and atherosclerosis (AS), their interrelationship, and drug-associated limitations. (A) Representative small-molecule drugs. (B) Possible limitations of small-molecule drugs that drive the exploration of comprehensive therapeutic strategies. DPP4i: dipeptidyl peptidase-4 inhibitors.

2.1. Small-molecule drugs

Obesity is a major contributor to several metabolic and cardiovascular diseases, including T2DM, hypertension, and dyslipidemia. These conditions promote the development of AS by driving chronic systemic inflammation, insulin resistance, and β-cell dysfunction [22,23]. Weight loss is associated with various metabolic changes, which are beneficial to overall health. Against the escalating obesity epidemic worldwide, strengthening cardiometabolic protection has become a pivotal focus in chronic diseases prevention and control [24]. Our previous review summarized four anti-obesity small-molecule drugs and discussed recent developments in the field [25,26]. First, orlistat (Xenical®), used as an adjunct to a reduced-calorie diet, is indicated for overweight or obese adults at 120 mg three times daily, taken during or up to 1 h after meals containing fat. It has an elimination half-life of about 1–2 h. The most frequent adverse events are gastrointestinal oily spotting, flatus with discharge, urgent bowel movements, and steatorrhea. Second, naltrexone-bupropion combines two components: naltrexone inhibits the effects of opioids/alcohol and curbs hunger and food cravings, whereas bupropion functions as an antidepressant and decreases appetite. Contrave® (naltrexone HCl 8 mg/bupropion HCl 90 mg extended-release) is indicated for adults with BMI ≥ 30 kg/m² or BMI ≥ 27 kg/m² with at least one weight-related comorbidity. Dosing is titrated over four weeks, starting with one tablet each morning in week 1, increasing to one tablet morning and evening in week 2, two tablets morning and one evening in week 3, and two tablets morning and from week 4 onward (32 mg/360 mg daily). Elimination half-lives are about 5 h (naltrexone) and 21 h (bupropion). The most common adverse reactions are nausea, constipation, headache, vomiting, dizziness, insomnia, dry mouth, and diarrhea. Third, phentermine-topiramate (QSYMIA®) is indicated for a BMI ≥ 30 or 27 kg/m² with related comorbidity such as T2DM and dyslipidemia. Side effects in adults include paraesthesia, dizziness, dysgeusia, insomnia, constipation, and dry mouth. The half-life is 20 h (phentermine) and 65 h (topiramate), with orally stepped dosing starting at 3.75 mg/23 mg. Finally, lorcaserin (Belviq®), formerly indicated for adults with BMI ≥ 30 kg/m² or BMI ≥ 27 kg/m² with comorbidity, was voluntarily withdrawn in February 2020 after trial data showed an increased incidence of cancer [27].

The global burden of diabetes is substantial, with millions affected worldwide, and the prevalence of prediabetes exceeds that of individuals with diabetes, signifying a growing epidemic [28]. Diabetes, particularly T2DM, is a significant component of CMDs and is closely associated with conditions, including hypertension, obesity, and CVDs [28,29]. Endothelial dysfunction, inflammation, and oxidative stress are common hallmarks of diabetes mellitus and AS [30]. Understanding the interplay between diabetes and other cardiometabolic factors is crucial for effective management and prevention strategies [31]. The following are seven types of antidiabetic small-molecule drugs. i) Metformin is utilized daily by > 200 million patients, offering multiple benefits, including improved cardiovascular outcomes and glycemic control. It has been the first-line therapy for T2DM (especially before 2018) due to its efficacy and safety profile, primarily targeting the liver and gut, which can improve insulin resistance [31]. However, its application can be constrained by gastrointestinal side effects, and its effectiveness may vary according to genetic factors [31,32]. ii) As peroxisome proliferator-activated receptor γ (PPAR-γ) agonists, TZDs like rosiglitazone and pioglitazone also can ameliorate insulin resistance [31]. Chiglitazar sodium is a novel PPAR pan-agonist that has been included in the Guideline for the Prevention and Treatment of Diabetes Mellitus in China [33]. iii) Sodium-glucose cotransporter-2 inhibitors (SGLT-2i) constitute a novel class of oral hypoglycemic agents that lower blood glucose by inhibiting glucose reabsorption in the proximal renal tubules, consequently enhancing urinary glucose excretion, such as empagliflozin, dapagliflozin, canagliflozin, and ertugliflozin [34]. Unlike other glucose-lowering medications, the hypoglycemic effect of SGLT-2i is independent of insulin secretion, significantly reducing the risk of hypoglycemia [34]. Numerous extensive cardiovascular outcome trials have validated that SGLT-2i lowers glucose levels and provides significant cardiovascular and renal protective effects [25,27]. Its primary side effects encompass genitourinary infections, hypotension, and ketoacidosis [35]. iv) α-glucosidase inhibitors, e.g., acarbose, voglibose, and miglitol, delay carbohydrate absorption by inhibiting intestinal α-glucosidase enzymes, and they generally do not cause hypoglycemia [36]. v and vi) The remaining two classes act by increasing insulin secretion, either directly (sulfonylureas and glinides) or indirectly via glucose-dependent mechanisms (dipeptidyl peptidase-4 inhibitors (DPP4i)). This insulin-dependent mechanism explains their (sulfonylureas and glinides) distinctive side effect profiles, particularly the risk of hypoglycemia and weight gain in sulfonylureas. DPP4i (glucose-dependent), while generally well tolerated, are associated with mild side effects such as headache, nasopharyngitis, and gastrointestinal discomfort [36]. vii) Dorzagliatin, a glucose-dependent glucokinase (GCK) activator developed by Hua Medicine (Shanghai, China), was approved in China in 2022 for improving glycaemic control in adults with T2DM [37], highlighting the continued potential and value of small-molecule drugs development.

AS represents a predominant chronic health burden globally, accounting for significant morbidity and mortality across populations [4]. Established modifiable risk factors include hypertension, dyslipidemia, diabetes, obesity, and tobacco use, whereas non-modifiable risk factors include genetic predispositions and ageing. The disease frequently coexists with metabolic comorbidities, e.g., T2DM and obesity, creating synergistic pathways for vascular deterioration [38]. Oral pharmacotherapy for AS predominantly targets antiplatelet areas, regulating lipid metabolism and improving endothelial function. The combination of antiplatelet and cholesterol-lowering drugs (primarily aspirin combined with statins) remains the most common treatment for atherosclerotic diseases [4]. Statins are a cornerstone in managing AS by targeting 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase. They are extensively used in both primary and secondary prevention of atherosclerotic CVDs (ASCVDs), significantly reducing the risk of major cardiovascular events. Statins decrease plasma LDL-C and mitigate inflammation, critical contributors to AS progression. Their efficacy and safety have been well-documented across diverse patient populations, rendering them indispensable in cardiovascular care [39]. Non-statin small-molecule lipid-lowering drugs include bile acids (BAs) chelators such as cholestyramine, colestipol, colesevelam, and colestilan; fibrates such as fenofibrate, which primarily reduce triglycerides by activating PPAR-α; bempedoic acid, an oral ATP citrate lyase inhibitor; lomitapide, which reduces lipid levels by inhibiting the function of microsomal triglyceride transfer protein (MTP); muvalaplin, a novel oral small-molecule inhibitor that selectively disrupts the molecular interaction between apolipoprotein A (ApoA) and ApoB-100; and cholesteryl ester transfer protein (CETP) inhibitors, including torcetrapib, anacetrapib, dalcetrapib, evacetrapib, and the recently developed obicetrapib [40,41]. Additionally, ezetimibe lowers cholesterol levels by targeting and blocking the Niemann-Pick C1-like 1 (NPC1L1) protein [4]. The main antiplatelet agents used are aspirin, clopidogrel, and ticagrelor. Aspirin is commonly used for secondary prevention; however, it is not recommended for primary prevention due to risks such as gastrointestinal bleeding [42]. Some antiproliferative drugs, including rapamycin and its derivatives, are commonly used as coatings on drug-eluting stents and drug-coated balloons for interventional procedures. Other antiproliferative drugs, including methotrexate and sulfonyl purine, are effective in treating AS. However, these drugs are primarily limited to interventional therapies, as no oral antiproliferative drugs have been approved for clinical use in AS [4]. Previous clinical and animal studies have demonstrated that hydroxyurea-oral antiproliferative drugs can effectively treat AS by modulating gut microbiota and reducing NPC1L1 [4].

2.2. Lifestyle interventions

Although implementation and maintenance present significant challenges, lifestyle interventions remain basic to CMDs management [43]. Evidence supports the efficacy of various dietary approaches, particularly the Mediterranean diet, demonstrating substantial cardiovascular risk reduction [43]. Physical activity interventions benefit metabolic parameters, particularly combined aerobic and resistance training protocols [44]. Key challenges in lifestyle interventions include limited long-term adherence, maintenance rates below 25% after two years, socioeconomic barriers affecting implementation, personalized approaches considering individual variability, and access limitations to professional guidance and support systems [9].

Despite these diverse treatment options, conventional therapies for CMDs face potential hurdles [10]. Efficacy and safety debates persist, as dose-dependent effects often require sustained high dosing and risk toxicity [45]. Over 70% of human proteins, such as ApoC-III, lipoprotein (a) (Lp(a)), and PCSK9, etc., critical targets lack druggable binding pockets and/or appropriate chemicals, which limits drug design [10,46]. Resistance mechanisms further undermine efficacy, including compensatory feedback (statin-induced HMG-CoA reductase) and genetic adaptations [47]. Nevertheless, small-molecule drugs retain advantages such as oral bioavailability, cost-effectiveness, and well-established manufacturing pipelines, making them important to the treatment of obesity, T2DM, and AS. CMDs’ multifactorial nature necessitates multidrug regimens for lipid, glucose, and energy control, which increases interaction risk and lowers adherence [48].

3. Emerging therapies in CMDs

3.1. Interventions that modulate gut microbiota in CMDs

The gut microbiota plays crucial endocrine and paracrine roles through its metabolites, which act as bioactive signaling molecules affecting distant organ function [49]. Given the global epidemic of obesity, T2DM, and AS, there is growing interest in understanding how gut microbiota influences these conditions. Theofilis et al. [50] made a detailed discussion on the role of gut microbiota in the development and progression of CMDs. Interventions that modulate gut microbiota offer promising therapeutic potential for managing CMDs by improving key metabolic outcomes [51], which include fecal microbiota transplantation (FMT), engineered bacteria, live biotherapeutic products (LBPs), phage therapy, probiotics, prebiotics, synbiotics, postbiotics, diet, and natural products. This section focuses on direct and indirect strategies for modulating the gut microbiota to correct dysbiosis associated with the development of CMDs (Fig. 3).

Fig. 3.

Direct and indirect strategies for restoring cardiometabolic health by modulating gut microbiota dysbiosis. (A) Modulation strategies of gut microbiota. (B) Gut microbiota dysbiosis associated with the development of cardiometabolic diseases (CMDs). Dots represent the three CMDs focused on in this study. FMT: fecal microbiota transplantation; LBPs: live biotherapeutic products; NAFLD: non-alcoholic fatty liver disease; CKD: chronic kidney disease; CVDs: cardiovascular diseases.

3.1.1. Direct modulation for the gut microbiota

Direct modulation strategies involve introducing or engineering live microorganisms to reshape gut microbiota. FMT, transferring screened donor fecal material to a recipient's gastrointestinal tract, remains a foundational strategy for restoring gut microbial diversity and functional redundancy. FMT may help preserve residual β-cell function in individuals with type 1 diabetes mellitus (T1DM), potentially stabilizing endogenous insulin production by de Groot et al. [52]. This effect was associated with distinct gut microbial profiles. Similarly, Ng et al. [53] found that repeated FMT procedures could enhance the extent and duration of microbiota engraftment in obese individuals with T2DM. While FMT has shown promise as a metabolic intervention, its role in augmenting the therapeutic effects of metformin in T2DM management has not been widely investigated. Wu et al. [54] reported that FMT, with or without concurrent metformin use, significantly improved insulin resistance, BMI, and gut microbial composition in T2DM patients, primarily through successful colonization by donor-derived microbial species. However, reproducibility, donor-recipient compatibility, and safety challenges necessitate further optimization. Recent advancements include LBPs such as Seres Therapeutics 109 (SER-109), Rebiotix 2660 (RBX2660), and Clostridium P 101 (CP101), which are based on cultured bacteria or processed stool samples, offering an alternative to traditional FMT and promising more consistent therapeutic outcomes [55].

LBPs represent a promising and emerging class of therapeutic agents that employ live microorganisms, such as bacteria, viruses, or fungi, to prevent and treat various diseases, such as cancer immunotherapy and metabolic disorders [56]. According to the U.S. Food and Drug Administration (FDA), LBPs defined as biological products that contain live organisms, such as bacteria, are intended for the prevention, treatment, or cure of diseases or medical conditions in humans; and are not vaccines. The European Pharmacopoeia similarly defines LBPs as medicinal products composed of live microorganisms (bacteria or yeasts) intended for human use, excluding fecal microbiota transplants and gene therapy products from this classification [57]. The recent FDA approvals of Rebyota™ and Vowst™ mark significant milestones in the rapidly evolving field of live microbiota-based therapies. These developments underscore the promising potential of future microbiota-based treatment strategies for various diseases and conditions. However, substantial challenges continue to impede progress in this area. Chief among them are the notable gaps in existing regulatory frameworks for developing and manufacturing LBPs, which require comprehensive expansion and refinement. In addition to regulatory hurdles, significant obstacles remain in optimizing and validating analytical methodologies critical for LBPs characterization, particularly in microbial identification, potency assessment, and bioburden evaluation [58].

Probiotics are also commonly regarded as alternative dietary supplements capable of modulating the composition and functional activity of the gut microbiota, thereby contributing to host health. Ecologic® Barrier (Winclove probiotics, Amsterdam, The Netherlands), a probiotic preparation, blend with Bifidobacterium bifidum W23, Bifidobacterium lactis W51, Bifidobacterium lactis W52, Lactobacillus acidophilus W37, Lactobacillus brevis W63, Lactobacillus casei W56, Lactobacillus salivarius W24, Lactococcus lactis W19, and Lactococcus lactis W58, demonstrated potential in preventing arterial stiffness in obese postmenopausal women. The study (Trial No.: NCT03100162) indicated that the multispecies probiotic supplementation achieved this by lowering trimethylamine N-oxide (TMAO) levels and obesity-related chronic inflammation while enhancing the body's total antioxidant capacity [59]. Furthermore, administration of Ecologic® Barrier has been shown to improve insulin resistance and reduce abdominal adiposity in patients with T2DM (Trial No.: NCT01765517) [60]. One major limitation of current strategies for modulating the gut microbiota lies in the lack of precise tools to target and alter specific species within complex microbial ecosystems. To address this challenge, two main approaches have been proposed: one involves introducing engineered bacteria that have been genetically modified outside the body into the gut, while the other leverages the natural diversity of the gut microbiota by using engineered bacteriophages to edit microbial populations directly in situ [61]. Phage therapy could target pathogenic bacteria selectively, offering a tailored alternative. These direct interventions aim to replenish or reprogram microbial communities but face standardization, storage, and antibiotic compatibility challenges [61].

3.1.2. Indirect modulation for the gut microbiota

Indirect strategies focus on stimulating the beneficial gut microbiota through substrates or dietary interventions. Prebiotics (inulin and galactooligosaccharides) selectively nourish commensal bacteria (Bifidobacteria), reducing inflammation and improving cardiovascular outcomes [61,62]. Synbiotics combine probiotics and prebiotics to enhance microbial survival and activity synergistically. For instance, yoghurt leverages live cultures (Lactobacillus) and fermentable substrates to amplify health benefits [62]. Postbiotics, including microbial metabolites such as short-chain fatty acids (SCFAs), enhance gut barrier function and immune modulation without requiring live microorganisms [61]. As we all know, dietary interventions, including Mediterranean or Dietary Approaches to Stop Hypertension (DASH) diets, promote microbial diversity and reduce CMDs risk by emphasizing plant-based fibres. Natural products have emerged as promising candidates for CMDs management, drawing considerable attention due to their multi-target therapeutic potential and well-documented historical efficacy. Table 1 [[63], [64], [65], [66], [67], [68], [69], [70], [71], [72], [73]] summarizes some examples of natural products, supported by evidence from direct or indirect animal and human studies, that ameliorate obesity, T2DM, and AS by modulating the gut microbiota. The unique identifier (NCT number) for each clinical trial is provided in the table, and detailed information could be accessed by searching for the respective identifier on the ClinicalTrials.gov website [64]. Fig. 4 illustrates the two-dimensional (2D) structures of these natural products, as obtained from the PubChem database. It should be noted that this table focuses on the role of natural products in ameliorating CMDs through gut microbiota modulation. For more systematic and detailed information, we recommend consulting the Traditional Chinese Medicine Integrated Database 2 (TCMID2) [63].

Table 1.

Some examples of natural products modulating gut microbiota to improve obesity, type 2 diabetes mellitus (T2DM), or atherosclerosis (AS): evidence from animal and human studies.

Natural productsa	Category	Target CMDs [64]	Gut microbiota related results	Herb medicine [63]
Apigenin	Flavonoid	Obesity-associated metabolic syndrome	Improve intestinal dysbiosis, Akkermansia ↑, Incertae_Sedis ↑, Faecalibaculum ↓, and Dubosiella ↓ [65]	Petroselinum crispum
Berberine	Alkaloid	Obesity (Trial No.: NCT06426966), T2DM (Trial No.: NCT02861261), and AS	TMAO biosynthesis ↓ in intestine (Trial NO.: ChiCTR-OPN-17012942) [66]	Coptis chinensis
Catechins	Flavanols	Obesity (Trial No.: NCT00692731)	Change the microbiota in terms of overall structure, composition, and protein functions [67]	Camellia sinensis
Celastrol	Pentacyclic triterpene	Obesity	Alter the gut microbiota composition [68]	Tripterygium wilfordi
Curcumin	Polyphenol	Obesity (Trial No.: NCT03864783) and AS (Trial No.: NCT02998918)	Firmicutes/Bacteroidetes ↓, Desulfovibrio bacteria ↓, Akkermansia ↑, and SCFAs-producing bacteria ↑ [69]	Curcuma longa L.
Eugenol	Phenylpropanoid	Obesity	Firmicutes ↑, Dubosiella ↑, Blautia ↑, unclassified_f_Oscillospiraceae ↑, and unclassified_f_Ruminococcaceae ↑, Desulfobacterota ↓, Alistipes ↓, Alloprevotella ↓, and Bilophila ↓ [70]	Eugenia caryophyllata Thunb.
Ginsenoside Rb1	Triterpenoid saponins	Diabetes-associated metabolic disorders	Parasutterella ↑, Alistipes ↓, f_Prevotellaceae_unclassified ↓, Odoribacter ↓, and Anaeroplasma ↓ [71]	Panax ginseng C.A. Mey.
Puerarin	Isoflavone aglycone	AS	Inhibition of Prevotella copri and its trimethylamine production [72]	Pueraria lobata
Quercetin	Flavonoid	T2DM	Proteobacteria ↓, Bacteroides ↓, Escherichia-Shigella ↓, and E. coli ↓ [73]	Allium macrostemon Bge.

^aThe natural products listed in this table are arranged alphabetically. ↑ and ↓ indicate an increase and a decrease. CMDs: cardiometabolic diseases; TMAO: trimethylamine N-oxide; SCFAs: short-chain fatty acids; E. coli: Escherichia coli.

Fig. 4.

The chemical structures of natural products for improving obesity, type 2 diabetes mellitus (T2DM), or atherosclerosis (AS).

Although multi-target mechanisms of these natural products present unique advantages (including comorbidity management), the inherent chemical complexity of these compounds, poor bioavailability, and their partially undefined mechanisms of action pose significant challenges for standardization, regulatory compliance, and integration into modern precision medicine frameworks [10,74]. Meanwhile, some traditional Chinese medicines are becoming increasingly difficult to obtain, such as bear bile. This highlights the urgent need to develop effective alternatives that can meet medical demands while also ensuring the protection of bears [75].

Briefly, two gut microbiota modulation strategies, direct (FMT, LBPs, engineered bacteria, phages therapy, and probiotics) and indirect (prebiotics, synbiotics, postbiotics, diet, and natural products), hold promise for correcting gut dysbiosis in obesity, T2DM, or AS, but advancing mechanistic insight, personalization, safety standards, and regulatory frameworks is essential for clinical translation.

3.2. Macromolecular drugs as an emerging frontier in CMDs

Recently, macromolecular drugs comprising therapeutic proteins, peptides, monoclonal antibodies (mAbs), small nucleic acid drugs, and gene therapies have garnered significant attention because of their advantages in targeting specificity, affinity, sustained efficacy, and potential safety for CMDs, compared to small-molecule drugs. Table 2 [76,77] illustrates the fundamental differences between macromolecular and small-molecule drugs. In this section, we elucidate the main macromolecular drugs following the four classifications and exhibit potent anti-obesity, anti-T2DM, or anti-AS effects through mechanisms including enhancing glycemic control, promoting weight loss, and reducing lipid-driven vascular inflammation, etc., offering new hope for addressing these diseases and their associated metabolic disorders. Table 3 [18,26,[78], [79], [80], [81], [82], [83], [84], [85], [86], [87], [88], [89], [90], [91], [92], [93], [94], [95], [96], [97], [98]] details the classification, representative drug examples, and recent studies involving one selected drug for treatment of obesity, T2DM, AS, and associated dyslipidemia. Further details of the recent studies, such as study subjects, study types, dosages, experimental designs, and pharmacological effects, are presented in Table S1 [18,[78], [79], [80], [81], [82], [83], [84], [85], [86], [87], [88], [89], [90], [91], [92], [93], [94], [95], [96], [97], [98]].

Table 2.

General differences between small-molecule drugs and macromolecular drugs^a [76,77].

Classification	MW	Size	Structure	Production	Dose route	Targets	Half-life	Immunogenicity	Bioavailability	Absorption	Dose interval	Target specificity	Storage
Small-molecule drugs	Low	Small (20–100 atoms)	Simple	Chemical synthesis or modification	Primally oral (tablets/capsules)	Intracellular (enzymes, receptors, and DNA)	Short (hours to days)	Rare	Typically, high	Passive diffusion/carrier-mediated transport	Frequent dosing (daily)	Low (more off-target effects)	Stable (room temperature)
Macromolecular drugs (mainly biologics)	High	Large (200–50,000 atoms)	Complex	Biotechnological production (e.g., mammalian cell culture)	Mostly parenteral (IV/SC; poor oral absorption)	Extracellular (cell-surface receptors and ligands)	Generally, long (days to weeks)	Common (risk of anti-drug antibodies)	Very low (if not administered parenterally)	Active endocytosis/lymphatic absorption	Sustained effects (weekly/monthly)	High (precision targeting)	Unstable (often requires cold chain)

^aThese are broad trends, not strict rules, such as frequent-dosing insulin, low-immunogenicity humanized antibodies, and highly selective small molecules. MW: molecular weight; IV: intravenous; SC: subcutaneous.

Table 3.

Classification, representative drug examples, and recent studies on macromolecular drugs for the treatment of obesity, type 2 diabetes mellitus (T2DM), atherosclerosis (AS), or associated dyslipidemia.

Macromolecular drugs	Classification	Drug examplesa [26]	Recent studies
Proteins and peptides	GLP-1 RAs short-acting	Beinaglutide, exenatide, and lixisenatide	[78]
	GLP-1 RAs long-acting	Albiglutide, dulaglutide, exenatide extended-release suspension, liraglutide, PEX-168, and semaglutide	[79,80]
	GLP-1R/GIPR or GCGR co-agonists	Cotadutide, mazdutide, and tirzepatide	[81]
	GLP-1R/GIPR/GCGR triple agonists	HM15211 and retatrutide	[82]
	Insulin	URAIs: FAsp, URLi, and Afrezza;	[83,84]
		RA: Asp, glulisine, and lispro;
		Short-acting: regular human insulin;
		Intermediate-acting: NPH insulin;
		Long-acting: detemir, degludec, glargine, and icodec;
		Premixed insulin;
		Biphasic insulin analogs: IDegAsp
	Recombinant human leptin analogs	Chimeric leptin analogs and metreleptin	[85]
	Selective peptide melanocortin-4 receptor agonist	Setmelanotide	[86]
	FGF21 analogs or receptor agonists	AP025, efruxifermin, GLP-1/FGF21 co-agonists, pegbelfermin, pegozafermin, and recombinant FGF21	[87]
mAbs	Anti-leptin receptor mAbs	Mibavademab	[88]
	PCSK9i	Alirocumab and evolocumab	[18]
	ANGPTL3i	Evinacumab, REGN-1001, and 14DH12	[89]
	Anti-IL-1β mAbs	Canakinumab and LY2189102	[90]
	Anti-TNF-α mAbs	Infliximab	[91]
Small nucleic acid drugs	siRNA-target PCSK9	Inclisiran	[92]
	siRNA-target Lp(a)	Lepodisiran, olpasiran, SLN360, and zerlasiran	[93]
	siRNA-target APOC3	Plozasiran	[94]
	siRNA-target ANGPLT3	Zodasiran	[95]
	ASOs-target Lp(a)	IONIS-APO(a)Rx (discontinued) and pelacarsen	[96]
	ASOs-target APOB	Mipomersen	[97]
	ASOs-target APOC3	ION-449, olezarsen, vupanorsen (discontinued), and volanesorsen (EMA)	[98]

^aThe examples listed in this table are arranged alphabetically. Bold font: U.S. Food and Drug Administration (FDA)-approved drugs; GLP-1 RAs: glucagon-like peptide-1 receptor agonists; GIPR: glucose-dependent insulinotropic polypeptide receptor; GCGR: glucagon receptor; URAIs: ultra-rapid-acting insulin analogs; FAsp: faster aspart; URLi: ultra-rapid lispro; Asp: insulin aspart; NPH: neutral protamine hagedorn; IDegAsp: co-formulated insulin degludec and insulin aspart; FGF21: fibroblast growth factor 21; mAbs: monoclonal ntibodies; PCSK9i: proprotein convertase subtilisin/kexin type 9 inhibitors; ANGPTL3i: angiopoietin-like protein 3 inhibitors; IL-1β: interleukin-1β; TNF-α: tumor necrosis factor-α; siRNA: small interfering RNA; Lp(a): lipoprotein(a); APOC3: apolipoprotein C-III; ANGPLT3: angiopoietin-like protein 3; ASOs: antisense oligonucleotides; IONIS: Ionis Pharmaceuticals; APOB: apolipoprotein B; EMA: European Medicines Agency.

3.2.1. Proteins and peptides

GLP-1 RAs are a class of glucose-lowering agents that mimic the action of endogenous GLP-1, a pluripotent incretin hormone which is involved in multiple physiological actions of organs and tissues, such as heart, vessels, stomach and brain [99] (Fig. 5). They function as synthetic analogs by selectively activating GLP-1R, enhancing glucose-dependent insulin secretion, suppressing glucagon release, slowing gastric emptying, and promoting satiety. Structurally engineered to resist degradation by DPP4, these agents exhibit prolonged pharmacokinetic profiles compared with native GLP-1. Clinically, GLP-1 RAs are indicated for managing T2DM and obesity for specific agents (liraglutide and semaglutide). Emerging evidence also supports cardiovascular benefits, including a reduced risk of major adverse cardiovascular events in high-risk populations [25,84]. However, gastrointestinal side effects, including nausea and diarrhea, have been commonly reported [100]. Long-term administration of GLP-1 RAs may lead to tachyphylaxis, diminishing these effects over time [101]. Based on pharmacokinetic characteristics and administration frequencies, GLP-1 RAs can be categorized into two distinct classes: short- and long-acting agents [25], the details of which are listed in Table 3 and Table S1. Recently, GLP-1R/glucose-dependent insulinotropic polypeptide receptor (GIPR) co-agonists and GLP-1R/GIPR/glucagon receptor (GCGR) triple agonists have become promising therapeutic strategies for improving glycemic control and promoting weight loss. These novel agonists leveraged the synergistic effects of activating multiple incretin pathways to enhance insulin secretion, suppress glucagon release, and reduce appetite. Meta-analysis has revealed that GLP-1R/GIPR co-agonists and GLP-1R/GIPR/GCGR triple agonists exhibit superior efficacy than GLP-1 RAs alone. Patients without T2DM, those with a high BMI, and those undergoing longer treatment cycles demonstrated significantly greater weight loss and reduced waist circumference [82]. However, further research is warranted to fully elucidate these novel agents' long-term safety and efficacy profiles. FDA-approved GLP-1 RAs include exenatide, introduced in 2005 by AstraZeneca, which was one of the first GLP-1 receptor agonists approved for T2DM treatment. Other GLP-1 RAs, including albiglutide, approved in 2014 by GlaxoSmithKline, and dulaglutide, launched the same year by Eli Lilly, have also contributed to the therapeutic arsenal for T2DM management, albeit with varying side-effect profiles [10]. Recently, lixisenatide, approved in 2016 by Sanofi, and semaglutide from Novo Nordisk in 2017 have further advanced treatment for T2DM, with significant improvements in glycemic control and weight reduction [10]. Accordingly, GLP-1 RAs have emerged as pivotal therapeutic options for T2DM and obesity management, offering glycemic control, cardiovascular benefits, and weight reduction; meanwhile, novel co-agonists/triple receptor agonists have demonstrated enhanced efficacy in CMDs.

Fig. 5.

Interactions between the gut microbiota and macromolecular drugs. (A) Gut microbiota influencing proprotein convertase subtilisin/kexin type 9 (PCSK9), PCSK9 inhibitors (PCSK9i) regulating gut microbiota, and PCSK9-driven mechanism of low-density lipoprotein cholesterol (LDL-C) elevation. (B) Gut microbiota influencing glucagon-like peptide-1 (GLP-1), GLP-1 receptor agonists (GLP-1 RAs) regulating gut microbiota, and the actions of GLP-1 as well as GLP-1 RAs on the brain, stomach, pancreas, adipose tissue, heart, and vessels. BAs: bile acids; LDLR: LDL receptor; SCFAs: short-chain fatty acids.

Although current treatment guidelines for T2DM recommend a stepwise approach with incretin-based therapies as first-line injectable treatments, insulin therapy remains a viable treatment option for individuals with T2DM, particularly for those with a long history of the disease who experience insufficient glycemic control despite the use of other glucose-lowering interventions [83,84]. Developing insulin analogs has allowed for more precise glucose control, minimizing the risk of hypoglycemia and achieving better postprandial blood glucose levels. These modifications in the insulin structure, which include altering the amino acid sequence or introducing specific substitutions, enhance their pharmacokinetic and pharmacodynamic profiles. For instance, ultra-rapid-acting insulin aspart (a modified formulation of insulin aspart, containing excipients vitamin B3 (niacinamide) and L-arginine) and ultra-rapid-acting insulin lispro (based on insulin lispro, adding treprostinil and citrate salts) have been developed and introduced to more accurately mimic the endogenous insulin response to meals [102]. Afrezza, an inhaled formulation of human insulin, is also an ultra-rapid-acting inhaled insulin powder, which has recently demonstrated its safety profile in adult patients and offers a solution to overcome the suboptimal glycemic control commonly associated with the delayed absorption of subcutaneously administered rapid-acting insulin analogs [103]. Long-acting insulin analogs, including insulin glargine, insulin detemir, and new long-acting insulin icodec, provide a steady release of insulin, reducing the need for multiple daily injections and improving patient compliance, effectively mimicking the physiological basal insulin secretion [104]. These advancements in insulin therapy have improved glycemic control and potentially enhanced the overall quality of life of individuals with diabetes.

Leptin, another hormone involved in metabolic regulation, predominantly produced by adipose tissue, is essential for regulating energy balance and BMI [105]. Leptin deficiency or resistance often results in hyperphagia and reduced energy expenditure, leading to marked weight gain. In individuals with genetic mutations in leptin or its receptor (LEPR), this manifests as class 3 obesity. Similarly, patients with hypoleptinemia secondary to lipoatrophy, such as those with generalized lipodystrophy, often develop severe metabolic and hepatic complications [88]. On February 24, 2014, the FDA approved metreleptin as a replacement therapy combined with diet to treat complications of leptin deficiency in patients with congenital or acquired lipodystrophies [26]. Lipodystrophies are rare disorders of acquired or genetic origin characterized by generalized or partial loss of adipose tissue. This condition is associated with a heightened risk of severe metabolic complications, including insulin resistance, glucose tolerance abnormalities, hypertriglyceridemia, and atherosclerotic events, while metreleptin has therapeutic effects in managing these metabolic disturbances and improving outcomes in affected patients [85]. It is worth noting that setmelanotide, an eight-amino-acid cyclic peptide, received FDA approval in November 2020 (Imcivree®) for chronic weight management in patients aged ≥ 6 years with obesity due to proopiomelanocortin (POMC), PCSK1 or LEPR deficiency, and Bardet-Biedl syndrome [26,86]. Other protein analogs, including fibroblast growth factor 21 (FGF21) analogs (pegbelfermin), can potentially treat cardiovascular and metabolic diseases by enhancing insulin sensitivity, promoting weight loss, and reducing hepatic steatosis. While FGF21 studies have focused on nonalcoholic steatohepatitis, some findings underscore its therapeutic potential in modulating AS progression and glycemic control in T2DM and weight reduction in obesity [106]. Furthermore, preclinical research is needed to explore the potential synergistic effects of combining these analogs, which may provide a more comprehensive approach to addressing comorbidities. These therapies' safety profiles and long-term efficacy must be thoroughly evaluated through clinical trials to ensure that patients receive maximum therapeutic benefit while minimizing adverse effects.

3.2.2. mAbs

mAbs are synthetic proteins that mimic natural antibodies generated by the immune system. mAbs offer several advantages, including high specificity, fewer side effects, and prolonged effectiveness. They have gained popularity with approximately 80 approved variants, expanding their therapeutic applications beyond oncology and infectious diseases [107]. PCSK9 is a serine protease primarily expressed in the liver and is essential for regulating LDL-C levels by promoting LDL receptor (LDLR) degradation in hepatocytes [18], as shown in Fig. 5. Elevated PCSK9 levels are associated with increased LDL-C and cardiovascular risk, leading to PCSK9i development, including alirocumab and evolocumab, which are mAbs that target PCSK9, preventing its interaction with LDLR and thereby lowering LDL-C. Recent studies have explored the potential of mAbs in AS. PCSK9i, including alirocumab and evolocumab, are mAbs that target PCSK9, leading to a significant drop in LDL-C and a lower CMDs risk. First, alirocumab, co-developed by Sanofi and Regeneron and approved in 2015, has been a game-changer in tackling hyperlipidemia. This drug targets the PCSK9 protein, and studies have indicated its effectiveness in lowering the risk of heart attacks and strokes, rendering it a crucial treatment for patients with heterozygous familial hypercholesterolemia (HeFH) [26]. Second, evolocumab, developed by Amgen and approved in 2015, targets the same mechanism and boasts comparable cardiovascular benefits [26]. A systematic review and meta-analysis of randomized controlled trials demonstrated a favorable safety profile of alirocumab and evolocumab in patients with dyslipidemia or ASCVDs [108].

In a significant development, Regeneron Pharmaceuticals introduced evinacumab in 2021, which represents an innovative therapeutic approach through its targeting of angiopoietin-like 3 (ANGPTL3), a protein that inhibits the activity of lipoprotein and endothelial lipase and thereby reduces the concentration of plasma lipids [89]. This medication is indicated as an adjunct to other LDL-C-lowering therapies for treating adult and pediatric patients aged ≥ 12 years with homozygous familial hypercholesterolemia (HoFH), a rare genetic disorder characterized by severely elevated LDL-C and premature ASCVDs. In 2023, the FDA revised the age indication, lowering it to five years [26]. The clinical trials by Gaudet et al. [89] demonstrated that evinacumab significantly reduces LDL-C levels in adolescent and adult patients with HoFH, irrespective of sex, and is generally well tolerated. Its efficacy and safety have been maintained over the long term. Third, anti-leptin receptors, a class I cytokine receptor family member, modulate the energy and metabolic balance by engaging leptin, an adipose tissue hormone. Altarejos et al. [88] developed mibavademab, a leptin receptor mAb. Mibavademab normalised body weight, glucose, and insulin sensitivity in leptin-deficient mice. In a phase I randomized, double-blind, placebo-controlled two-part study, it reduced body weight in individuals with low leptin levels. However, it did not affect those with higher baseline leptin levels. These findings suggest that mibavademab may effectively treat disorders linked to leptin deficiency, although its impact on individuals with normal leptin levels requires further investigation [88].

Over the past several decades, experimental studies have confirmed that inflammatory mediators, particularly interleukins (ILs), tumor necrosis factor-α (TNF-α), interferon (IFN)-γ, and various chemokines, play a role in the occurrence and development of AS [109]. mAbs targeting inflammatory mediators have also demonstrated therapeutic potential in complications of AS or myocardial injury occurring after myocardial infarction. The Canakinumab Anti-Inflammatory Thrombosis Outcomes Study (CANTOS) trial showed that the anti-IL-1β antibody canakinumab (150 mg) significantly reduced the cardiovascular event risk in patients with post-myocardial infarction (hazard ratio = 0.85, P = 0.021), likely by suppressing IL-1β-driven systemic inflammation (37% reduction in high-sensitivity C-reactive protein (hs-CRP), independent of lipid-level lowering) [90]. Regarding TNF-α, an in vitro study indicated that infliximab inhibited TNF-α/IL-1β-serum amyloid A-induced inflammatory responses in human coronary endothelial cells; however, developing anti-drug antibodies may neutralize its efficacy and potentially exacerbate AS by enhancing immune complex deposition [91]. Although anti-inflammatory antibodies exhibit potential for cardiovascular protection, limitations include the potential compensatory activation of alternative pathways when targeting single inflammatory factors and increased infection risk due to long-term immunosuppression (higher infection rates in CANTOS). Future strategies may include combination therapies or small-molecule anti-inflammatory drugs to optimize their efficacy and safety.

3.2.3. Small nucleic acid drugs

Small nucleic acid drugs, mainly siRNAs, antisense oligonucleotides (ASOs), microRNAs (miRNAs), and aptamers, selectively target specific genes or their corresponding messenger RNAs (mRNAs), enabling sophisticated modulation of gene expression and translational regulation [110]. Over the past few decades, research has revealed that miRNAs play key roles in regulating macrophage activation, lipid metabolism, and the development of hyperlipidemia, which is summarized in the recent review [111]. Herein, we focus on the roles of siRNA and ASOs in managing lipid metabolism disorders.

siRNA molecules bind to complementary mRNA sequences and induce their degradation, effectively silencing the expression of specific genes. The advent of inclisiran and related siRNA therapies signals a paradigm shift in lipid management. Inclisiran, a prime example of siRNA therapeutics, exemplifies the potential for long-lasting therapeutic effects with infrequent dosing regimens. Developed by the Medicines Company, later acquired by Novartis, and approved by FDA in 2021, inclisiran functions as a PCSK9i delivered through RNA interference [10]. PCSK9 is a key regulator of LDLR degradation. By targeting PCSK9 mRNA for degradation, inclisiran effectively increases LDLR availability in hepatocytes, thereby enhancing LDL-C clearance from circulation. The pathophysiological hallmark of familial hypercholesterolemia encompasses profound elevations in LDL-cholesterol parameters, accompanied by an augmented predisposition to premature CVDs, with risk magnification reaching 10 to 20-fold above baseline population levels [97]. Clinical trials have demonstrated inclisiran's remarkable ability to significantly lower LDL-C levels, positioning it as a valuable therapeutic option for patients with ASCVDs or HeFH, conditions characterized by elevated LDL-C and increased CMDs risk [92]. In the ORION-8 study, inclisiran maintained a safety profile consistent with earlier trials. It was reported to be well-tolerated with a safety profile similar to the placebo. Injection site reactions were reported in 5.9% of participants, a slightly lower incidence than the 8.2% observed in the inclisiran groups of the combined ORION-9, -10, and -11 trials [112]. Recent advancements have expanded the therapeutic applications of siRNA beyond PCSK9, opening new avenues for targeting novel lipid-related pathways. Among these emerging targets, Lp(a) has garnered significant attention as a heritable, distinct, and increasingly recognized etiological factor in CVDs [96]. Lp(a) is a complex composed of the ApoA protein and ApoB-100, and its elevation is associated with an increased risk of CMDs. For instance, Lp(a)-targeting agents (olpasiran, SLN360, and zerlasiran) silence LPA mRNA, significantly reducing plasma Lp(a) levels. The OCEAN(a)-DOSE trial demonstrated that olpasiran, administered every 12–24 weeks, achieved dose-dependent Lp(a) reductions of 70%–97% in patients with ASCVDs or baseline Lp(a) > 150 nmol/L, offering a paradigm shift for managing elevated Lp(a) [93,113]. The phase II (ARCHES-2) study demonstrated that zodasiran (targeting ANGPTL3) exhibits a favorable safety profile in patients with mixed dyslipidemia. Treatment-emergent adverse events were balanced between treatment and placebo groups, largely reflecting comorbidities and underlying conditions in the study population. Laboratory safety evaluations revealed no clinically significant alterations, with stable platelet counts and minimal glycated hemoglobin (HbA1c) variations [95]. Due to these safety concerns, mipomersen is generally reserved for patients with HoFH who have not achieved adequate lipid control with other therapies, and it is administered under close medical supervision. Severe hypertriglyceridemia (sHTG) are associated with an amplified risk profile encompassing ASCVDs, nonalcoholic steatohepatitis, and the development of acute pancreatitis among affected individuals [94]. In the SHASTA-2 trial, plozasiran targeting APOC3 significantly reduced triglyceride levels [94]. As siRNA pipelines expand to encompass targets, novel RNA-based therapeutics are poised to redefine precision medicine in CMDs risk management. By combining durable efficacy with patient-centric dosing, these therapies address the unmet needs of lifelong disease management, heralding a new era in molecularly targeting CMDs.

ASOs are short, single-stranded nucleic acids that selectively bind target RNA via Watson-Crick hybridization, with their mechanism determined by RNA target regions and chemical design. Optimal binding sites include terminal sequences, loops, hairpins, and bulges of ≥10 bases [110]. A notable example in the lipid-lowering therapeutic space is mipomersen (Kynamro®), an ASO developed by Genzyme Corporation and approved by the FDA in 2013 [114]. Mipomersen employs a distinct mechanism of action, targeting APOB mRNA in the liver. ApoB-100 is a critical structural component of LDL, VLDL, and chylomicrons, essential in lipid metabolism [114]. By binding to APOB mRNA, mipomersen reduces ApoB-100 synthesis and lowers LDL-C and other atherogenic lipoproteins, which is approved as adjunctive therapy for HoFH refractory to conventional treatments (e.g., statins) [97]. Although mipomersen has demonstrated efficacy in lowering lipid levels in patients with HoFH, its use is often limited by adverse effects, including injection site reactions, elevations in liver enzymes (alanine aminotransferase and aspartate aminotransferase), and potentially hepatic steatosis [114]. Table 3 and Table S1 demonstrate the encouraging advancement of ASOs targeting Lp(a) and APOC3 in lipid modulation, with pelacarsen exhibiting significant Lp(a)-lowering effects in patients with CVDs and volanesorsen effectively reducing triglycerides in patients with familial chylomicronemia syndrome.

Nevertheless, the potential adverse effects that remain to be validated arise from off-target effects. Chemical modifications of oligonucleotide drugs are the primary means of reducing off-target effects. Although these drugs offer the advantage of long-lasting effects, prolonged suppression may pose safety concerns, as both lipids and blood pressure exhibit natural physiological rhythms in normal organisms.

3.2.4. Others

Innovative nucleic acid therapies encompass a range of modalities, including gene-editing technologies, most notably clustered regularly interspaced short palindromic repeats (CRISPR)-associated protein 9 (Cas9) and mRNA vaccines, such as BNT162b2 (Pfizer-BioNTech).

Gene editing, a transformative technology enabling precise insertion, deletion, or site-specific mutagenesis of target genes, has emerged as a powerful tool for addressing the genetic underpinnings of CMDs [115]. Among genome-editing platforms, the CRISPR/Cas9 system, derived from prokaryotic adaptive immunity, where CRISPR-Cas9 modules defend against foreign nucleic acids, has garnered significant attention owing to its unparalleled efficiency and accuracy [116]. The CRISPR/Cas9 system operates through a complex of Cas9 endonuclease and single-guide RNA (sgRNA), which directs Cas9 to induce double-strand DNA breaks (DSBs) at predetermined genomic loci. These DSBs are subsequently repaired through homology-directed repair or error-prone non-homologous end-joining, enabling targeted gene knock-in, knockout, or correction [117]. Current delivery strategies for CRISPR/Cas9 fall into three categories: Cas9-sgRNA ribonucleoprotein complexes, mRNA encoding Cas9 co-delivered sgRNA, and DNA plasmids expressing both components. This versatility has propelled its application across diverse cardiometabolic conditions. In AS, CRISPR-mediated permanent silencing of PCSK9, a gene regulating LDLR degradation, has demonstrated the potential to reduce cholesterol levels and attenuate plaque progression [118]. Similarly, CRISPR interference (CRISPRi) and CRISPR activation (CRISPRa) platforms allow for tunable gene regulation without introducing double-strand breaks, offering safer alternatives for long-term metabolic control [119]. For obesity-related metabolic dysfunction, Wang et al. [120] leveraged a catalytically inactive deactivated Cas9 (dCas9) system to upregulate uncoupling protein 1 expression in white adipose progenitor cells, driving their conversion into thermogenic brown fat-like cells. These engineered cells mitigated diet-induced obesity and improved metabolic syndrome in preclinical models. In diabetes research, CRISPR/Cas9 technology is widely employed for multiple purposes, including functional gene screening, mutation correction, and animal model generation. Gene screen studies, such as those targeting renalase (RNLS) and the autophagy receptor calcium binding and coiled-coil domain 2 (CALCOCO2), help identify genes involved in disease progression. Meanwhile, CRISPR has been used to correct pathogenic mutations in genes associated with monogenic diabetes, such as hepatocyte nuclear factor 1 alpha (HNF1A), HNF1B, HNF4A, GCK, and INS. Additionally, diabetic animal models have been generated through the introduction of mutations like PTPNR619W or by co-injecting Cas9 mRNA and sgRNAs to selectively disrupt genes such as the leptin receptor, insulin, and those involved in islet amyloid polypeptide (IAPP) production [121]. Therapeutic strategies also involve editing pancreatic β-cells or stem cells to enhance insulin secretion or to confer resistance to autoimmune-mediated destruction, providing promising avenues for future diabetes treatments [122]. ViaCyte and CRISPR therapeutics have initiated clinical trials to assess novel gene-edited stem cell therapies for T1DM. The first product, VCTX210 (Trial No.: NCT05210530), combines CRISPR/Cas9-edited allogeneic pancreatic endoderm cells (PEC210A) with a perforated implantable device that supports cell survival and immune evasion, aiming to restore insulin production in T1DM patients. Building on this, a second investigational therapy, VCTX211, is undergoing a phase I/II trial (Trial No.: NCT05565248) to evaluate its safety, efficacy, and tolerability in a similar patient population [122]. A significant improvement entails exosome-mediated delivery of CRISPR components, which enhances tissue specificity and minimizes off-target effects, offering a minimally invasive approach for rectifying the genetic defects underlying diabetes [123]. Despite these advances, several challenges persist, including optimizing delivery efficiency, ensuring long-term safety, and addressing ethical concerns. Future studies must prioritize the development of tissue-specific delivery systems and the rigorous evaluation of off-target effects in complex metabolic tissues. As such, integrating CRISPR with emerging technologies, including base editing or prime editing, may further refine the precision and expand the therapeutic potential of CMDs.

mRNA vaccines represent a highly effective and safe alternative to conventional vaccines, offering strong immunogenicity, rapid clinical development, and the potential for cost-efficient and scalable production [124]. The emergence of mRNA vaccine technology has opened innovative avenues for targeting autoimmune components in AS. Preclinical studies suggest that mRNA vaccines encoding ApoB-derived autoantigenic peptides (e.g., P210) may prevent atherosclerotic plaque formation by inducing expansion of IL-10-producing ApoB-specific regulatory T cells (Tregs), which modulate antigen-specific inflammatory responses in the arterial wall [[125], [126], [127]]. mRNA vaccines offer distinct advantages over conventional peptide-based approaches. By incorporating nucleoside analog modifications during in vitro transcription, mRNA constructs can mitigate Toll-like receptor-mediated non-specific inflammation, a common limitation of traditional adjuvants. Upon delivery, host cells translate mRNA into autoantigenic peptides, enabling antigen-presenting cells to prime CD4⁺ T cells toward a Tregs-dominant phenotype, thereby promoting immune tolerance. However, critical challenges remain, including the uncertain efficacy of these vaccines against established plaques and the need for rigorous safety validation in translational models. Future research should optimize mRNA design to enhance stability and targeted delivery while evaluating long-term immunomodulatory effects in advanced atherosclerotic lesions [125].

As mentioned above, macromolecular drugs have emerged as a promising tool in therapeutic applications. However, they also present limitations, including delivery barriers, limited stability, off-target effects, and the potential for immune activation [128]. One of the foremost limitations is poor oral bioavailability, particularly for siRNA and peptide-based therapeutics, due to degradation by gastrointestinal enzymes and poor membrane permeability. As a result, most of these agents require parenteral administration, which impacts patient compliance [128,129]. Moreover, immunogenicity remains a concern for biologics, especially mAbs, as repeated dosing may elicit anti-drug antibodies that reduce efficacy or provoke unwanted hypersensitivity reactions [130]. In parallel, achieving precise tissue- or cell-type-specific delivery continues to be a major obstacle, as systemic distribution of macromolecular drugs may lead to off-target effects or insufficient drug concentrations at the site of action [131]. To overcome these barriers, several innovative delivery strategies have emerged. Nanoparticle-based encapsulation systems, including lipid nanoparticles (LNPs) and polymeric carriers, have demonstrated improved siRNA stability, enhanced endosomal escape, and targeted tissue distribution [132]. In the case of aptamers, peptide, and protein-based drugs, PEGylation, the covalent attachment of polyethylene glycol (PEG) chains, has been widely used to increase molecular size, prolong circulation half-life, and reduce immunogenicity [133]. Additionally, ligand-directed delivery systems (e.g., N-acetylgalactosamine (GalNAc) conjugates for liver targeting) and exosome-based platforms are being developed to improve cellular uptake and specificity [134,135]. Although chemical modifications and advanced molecular designs have helped to mitigate some of these issues, a complete resolution remains an active area of ongoing research [128].

3.3. Interactions between gut microbiota and macromolecular drugs in CMDs

Interactions between gut microbiota and drugs can occur through various pathways, including microbial metabolites, changes in gut barrier integrity, and regulation of transporter and enzyme gene expression in the gastrointestinal tract [136]. Numerous studies have been conducted on the interactions between small-molecule drugs and gut microbiota [137]. This section primarily focuses on the interaction between gut microbiota and macromolecular drugs, representing a bidirectional regulatory axis with profound implications for CMDs management. It operates through two principal aspects: gut microbiota influencing macromolecular drug responses and macromolecular drugs regulating gut microbiota. Fig. 5 depicts the interactions between gut microbiota and macromolecular drugs, mainly PCSK9i and GLP-1 RAs, highlighting their mutual influences and their respective actions.

3.3.1. Gut microbiota influencing macromolecular drugs

The gut microbiome functions as a bioactive interface that significantly influences drug responses through microbial enzymatic drug modification, microbial metabolites regulating macromolecules, and commensal microbiota-mediated drug responses.

3.3.1.1. Microbial enzymatic drug modification

The interaction between gut microbiota and drugs is a complex and dynamic process significantly influencing drug pharmacokinetics. The gut microbiota expresses various enzymes that can chemically modify drugs, impacting their absorption, distribution, metabolism, and excretion (ADME), creating a complex interaction with pharmacological agents [138]. This enzymatic activity can transform drugs into metabolites with altered activities, toxicities, and lifespans within the body. The gut microbiota's role in drug metabolism is distinct from that of host enzymes, and understanding these microbial transformations is crucial for advancing personalized medicine and drug development [139]. Functional group analysis indicates that specific chemical structures, such as lactones, nitro, azo, and urea groups, make drugs more susceptible to microbial metabolism [140]. To the best of our knowledge, few classic studies analyze macromolecular drug metabolism by gut microbial enzymes; hence, we use small-molecule drugs as examples. Simvastatin, a lipid-lowering statin, is one such example. It is a prodrug with a lactone ring structure formed by an ester bond and a covalent C–C bond, rendering it inactive. Microbial β-hydroxy acid dehydrogenases convert simvastatin into its active metabolite (β-hydroxy acid), which enhances the inhibition of HMG-CoA reductase, a key enzyme in cholesterol biosynthesis. This conversion is crucial because it amplifies the drug's lipid-lowering effects, thereby reducing cardiovascular risk and mortality [140,141]. Researchers have systematically identified microbial gene products that metabolize drugs using high-throughput genetic screening and mass spectrometry (MS). These microbiome-encoded enzymes can directly and significantly affect intestinal and systemic drug metabolism, explaining the drug-metabolizing activities of human gut microbiota based on their genomic contents, linking the drug-metabolizing activities of the human gut microbiota to their genomic content. This insight elucidates how interindividual microbiome differences contribute to variability in drug metabolism, which has important implications for medical therapy and drug development [142]. Moreover, the gut microbiota's ability to chemically transform xenobiotics, including pharmaceuticals, is a key factor in drug pharmacokinetics. The chemical transformation of xenobiotics by the gut microbiota often results in metabolites with pharmacological properties that differ from those of the parent compound. This process is crucial for understanding drug efficacy and toxicity and developing strategies to manipulate the microbiome to improve drug responses. Integrating traditional and emerging technologies can facilitate progress toward a molecular understanding of gut microbial xenobiotic metabolism, ultimately guiding personalized medicine and improving drug discovery [139]. In addition to chemical transformation, the gut microbiota can modulate drug availability through bioaccumulation. Some microbiota can store drugs intracellularly without chemical modification, affecting drug availability and efficacy. This bioaccumulation can alter the composition of the gut microbiota and influence pharmacokinetics, side effects, and drug responses individually. Understanding these interactions is essential for predicting drug responses and optimizing therapeutic strategies [143]. Although most studies have focused on small-molecule drugs, the principle might extend to macromolecular drugs, including mAbs and peptides, where bacterial enzymes might break down or modify them. Understanding how gut bacteria alter or process drugs can offer new ways to adjust the microbiome and boost drug and biologics effectiveness [137].

3.3.1.2. Microbial metabolites regulating macromolecules

Microbial metabolites may amplify drug effects by regulating macromolecules, mainly GLP-1 and PCSK9. The gut microbiota, comprising trillions of microorganisms in the gastrointestinal tract, is increasingly acknowledged for its function in host metabolism by producing metabolites, such as secondary BAs and SCFAs. Secondary BAs, including deoxycholic acid, are produced by gut microbiota, including Clostridium scindens, through 7α-dehydroxylation. These secondary BAs significantly modulate host metabolism by activating various receptors, including Takeda G protein-coupled receptor 5 (TGR5) [144]. TGR5 activation has been demonstrated to augment the secretion of GLP-1 from enteroendocrine L cells in the gut. This process is crucial because GLP-1 is an incretin hormone that stimulates insulin secretion, significantly contributing to glucose homeostasis and energy metabolism [145]. TGR5 agonists have been identified as potential therapeutic agents [146]. These microbiota-modulated pathways exhibit tissue-specificity: intestinal TGR5 activation predominates in energy expenditure regulation, whereas pancreatic β-cell TGR5 modulates insulin granule exocytosis, highlighting the necessity for spatiotemporally targeted interventions [147]. Consequently, the synergistic interplay between TGR5 activation by secondary BAs and GLP-1 RAs may present a promising strategy to enhance the therapeutic efficacy of metabolic treatment, including T2DM and obesity. The therapeutic implications of such microbial-host interactions extend beyond BAs metabolism. In cultured mouse intestinal L cells, l-tryptophan upregulated the mRNA levels of genes involved in GLP-1 production, thereby enhancing GLP-1 secretion [148]. A seminal study by Yoon et al. [149] utilized advanced analytical techniques, including fast protein liquid chromatography (LC) and LC-MS, to identify a novel protein, P9, secreted by Akkermansia muciniphila, a bacterium known for its beneficial effects on metabolic health. P9 is pivotal in enhancing GLP-1 secretion and promoting thermogenesis in murine models fed a high-fat diet. These effects collectively improve glucose homeostasis, suggesting a mechanistic link between microbe-derived macromolecules and metabolic regulation. Furthermore, the study revealed that P9 interacts with intercellular adhesion molecule 2, underscoring the potential of harnessing gut microbiota-derived macromolecules to develop novel interventions for CMDs [149]. Another study confirmed that SCFAs, BAs, and lipopolysaccharide (LPS) by gut microbiota can also promote GLP-1 production [150]. Moreover, butyrate, a SCFA generated by Faecalibacterium prausnitzii, engages in a multimodal signalling cascade to potentiate incretin effects by increasing anorexigenic gut hormones and plasma levels of GLP-1, gastric inhibitory polypeptide (GIP), and peptide tyrosine-tyrosine (YY) [151]. The clinical translation of these mechanisms demonstrates dose-responsive efficacy. A four-week sodium butyrate intervention (4 g/day) in patients with T2DM elevated postprandial GLP-1 by 38% (95% confidence interval: 24%–52%) while reducing homeostatic model assessment for insulin resistance by 19% (P = 0.03), paralleling improvements in β-cell function [151]. Following that, this section examines the studies related to PCSK9. PCSK9 facilitates gut-heart communication by modulating circulating LDL-C levels, indicating a strong connection between PCSK9 upregulation, the gut microbiome, and AS [49]. Research has shown that LPS induces a significant elevation in PCSK9 mRNA expression in the liver, with a 2.5-fold rise observed at 4 h and a 12.5-fold rise at 38 h post-exposure. This response is highly sensitive, as even 1 μg of LPS can trigger half of the maximal PCSK9 induction. Additionally, LPS enhances PCSK9 expression levels in the kidneys [152]. Although direct studies linking gut microbiota to PCSK9 expression through BAs are limited, the farnesoid X receptor (FXR) pathway has been implicated. For instance, Sayin et al. [153] demonstrated that gut microbiota-derived secondary BAs activate FXR, influencing lipid metabolism, possibly by modulating PCSK9, though this mechanism remains to be clarified.

3.3.1.3. Commensal microbiota-mediated drug responses

The role of commensal microbiota in modulating mucosal immunity is a burgeoning area of research, particularly in the context of biological drug efficacy. Bifidobacterium, a prominent genus in the gut microbiota, exerts significant immunomodulatory effects that can influence the effectiveness of biological drugs. Research has revealed that Bifidobacterium enhances dendritic cell function, leading to improved CD8⁺ T cell priming and accumulation in the tumor microenvironment, thereby augmenting the efficacy of biological medicines (anti-PD-L1) therapy [154]. This suggests that Bifidobacterium can enhance the efficacy of other biological drugs by modulating immune responses. Furthermore, the interaction between gut microbiota and immune checkpoint inhibitors has been well-documented. The gut microbiome's composition can significantly affect the efficacy of these therapies, and manipulating the microbiota has been proposed as a strategy to improve therapeutic outcomes [155]. This is particularly relevant for biological drugs, as the microbiome's influence on immune modulation can alter drug efficacy. Antibiotic administration disrupts the gut microbiota, adversely affecting the efficacy of immune checkpoint inhibitors. A study involving patients receiving ipilimumab, an immune checkpoint inhibitor, indicated that antibiotic administration was associated with significantly lower clinical benefits, underscoring the essential role of gut microbiota in drug efficacy [156]. A pilot study (52 patients) showed the efficacy of GLP-1 RAs, including liraglutide or dulaglutide, correlates with distinct shifts in β-diversity among patients with T2DM, particularly between treatment responders and non-responders, indicating microbiota-mediated variability in glycemic outcomes [157]. After applying the random forest algorithm, 17 microbial signatures were selected. Among these, Bacteroides dorei and Roseburia inulinivorans, both known for their immunomodulatory properties, along with Lachnoclostridium spp. and Butyricicoccus spp., showed positive associations with glycemic improvement. In contrast, Prevotella copri, a species linked to insulin resistance, as well as Ruminococcaceae spp., Bacteroidales spp., Eubacterium coprostanoligenes, Dialister succinatiphilus, Alistipes obesi, Mitsuokella spp., Butyricimonas virosa, Moryella spp., and Lactobacillus mucosae, were negatively associated with glycemic reduction. Even after accounting for baseline HbA1c and C-peptide concentrations as clinical confounders, Bacteroides dorei, Lachnoclostridium sp., and Mitsuokella multacida showed significant associations [157]. The gut microbiota also could significantly modulate the immune response to mRNA vaccines, affecting their efficacy. Studies have identified specific bacterial taxa associated with improved immunogenicity. Ng et al. [158] conducted a prospective observational study involving adults vaccinated with either the mRNA vaccine BNT162b2 or the inactivated vaccine CoronaVac (Sinovac Biotech), revealing that Bifidobacterium adolescentis was persistently higher in individuals exhibiting high neutralizing antibodies to CoronaVac, indicating its role in enhancing vaccine efficacy. Another study reported that Faecalibacterium prausnitzii was associated with robust and persistent antibody responses post-BNT162b2 vaccination, indicating its role in maintaining long-term immunity [159]. Safety is also affected because the gut microbiota can modulate immune reactions that lead to adverse events. Ng et al. [158] identified correlations between gut microbiota composition and vaccine-related adverse events, including fever and diarrhea, particularly in mRNA vaccine recipients. This is potentially linked to dysbiosis, where certain diets (high protein and animal fat) increase pro-inflammatory bacteria, including Bacteroides, potentially exacerbating side effects. In summary, understanding these interactions and the role of gut microbiota in drug efficacy could lead to improved therapeutic strategies and outcomes.

3.3.2. Macromolecular drugs influencing gut microbiota

Macromolecular drugs function as microbial ecosystem modulators, shaping the gut microbiota, specifically, GLP-1 receptor agonists, PCSK9 inhibitors, and mRNA vaccines, each exhibiting distinct effects.

3.3.2.1. GLP-1 RAs

Studies in mice and humans have demonstrated that GLP-1 RAs, mainly used in managing T2DM and partly in managing obesity, can modulate the gut microbiota by increasing the abundance of certain bacteria, including Escherichia coli (E. coli) and altering microbial diversity. In preclinical mouse models, administration of liraglutide, a GLP-1 RA, has been established to rapidly elevate cecal E. coli levels, as measured by caseinolytic protease B expression, within 16 h, mediated by sympathetic nervous system activation and norepinephrine release into the intestinal lumen [101]. This shift is accompanied by broader microbial changes, including a significant reduction in Bacteroidetes and a trend toward increased Actinobacteria at the phylum level, suggesting an impact on overall microbial diversity [101]. In the meanwhile, these microbial alterations may influence GLP-1 RAs efficacy by influencing incretin secretion and glucose homeostasis, potentially through SCFAs production or immune modulation. Using 16S ribosomal RNA (rRNA) amplicon sequencing, a preliminary longitudinal study (337 subjects) investigated gut microbiota changes in newly diagnosed T2DM patients following 1 and 48 weeks of dulaglutide treatment. No significant changes were observed after 1 week; however, after 48 weeks, the gut microbiota composition changed markedly, with a notable decrease in microbial abundance. Additionally, fasting glucose, fasting C-peptide, HbA1c, and BMI were closely associated with gut microbiota profiles. These findings suggest that modulation of gut microbiota may be one of the mechanisms by which dulaglutide exerts its therapeutic effects in T2DM [160]. Zhang et al. [161] employed 16S rRNA sequencing and discovered that the prevalence of SCFAs-producing bacteria, including Bacteroides, Lachnospiraceae, and Bifidobacterium, significantly increased in the gut microbiota of liraglutide-treated diabetic rats. Moreover, studies in diet-induced obese mice have revealed that liraglutide and GLP-1/GLP-2 co-agonists induce phylogenetically similar gut bacterial changes, reducing diversity and altering low-abundance species, which may contribute to improved metabolic profiles, including weight loss and glucose tolerance [162]. This comprehensive understanding emphasizes the need for further research to harness gut microbiota for optimizing GLP-1 RAs therapy in CMDs, integrating pharmacomicrobiomics and multi-omics profiling.

3.3.2.2. PCSK9i

PCSK9i have been demonstrated to influence gut microbiota composition, acting as microbial ecosystem engineers. A recent human study by Caparrós-Martín et al. [163] evaluated the effect of alirocumab on statin-treated patients with elevated Lp(a) and observed no significant change in α- or β-diversity of the gut microbiota after 24 weeks of treatment. However, there was a notable increase in the relative abundance of the phylum Proteobacteria and the genus Escherichia-Shigella, along with a significant elevation in fecal BAs levels, particularly secondary BAs. These changes were positively correlated with reductions in plasma LDL-C levels, suggesting that PCSK9i may enhance BAs metabolism, potentially mediated by microbial shifts. The interplay between the gut microbiota and PCSK9 represents a bidirectional regulatory axis with significant implications for managing CMDs, particularly hypercholesterolemia and AS.

3.3.2.3. mRNA vaccines

mRNA vaccines could alter gut microbiota composition, potentially affecting its diversity and function. Ng et al. [158] observed that vaccination with ChAdOx1 (an adenovirus-vectored vaccine, frequently compared with mRNA vaccines in studies) resulted in a gradual decline in α-diversity (community richness and microbial diversity) from baseline to post-vaccination, with significant differences identified through Wilcoxon signed-rank test for gut microbiota composition. This suggests that mRNA vaccines may similarly influence the microbiota, although direct evidence for BNT162b2 is less pronounced in the literature.

Taken together, these studies highlight the significant role of gut microbiota in modulating individual variability in drug responses, and suggest that macromolecular drugs may, in turn, alter the gut microbiota. The burgeoning field of pharmacomicrobiomics explores how microbiota variation affects therapeutic outcomes, including adverse drug reactions. This bidirectional interaction paradigm underscores the necessity of "microbiome-aware" drug development strategies integrating pharmacomicrobiomics, synthetic biology, and multi-omics profiling to optimize therapeutic outcomes in CMDs.

4. Challenges and opportunities for CMDs treatment

The future of CMDs management lies in integrating these advanced therapeutics within personalized medical frameworks, allowing healthcare providers to tailor treatments based on individual patient characteristics, including genetic factors, lifestyle, and comorbidities. The current therapeutic landscape of CMDs has revealed several critical needs for advancement.

4.1. Comorbidity management

Gut microbiota-related interventions, including probiotics, prebiotics, and synbiotics, have significantly improved the glucose and lipid profiles in metabolic diseases, including diabetes and obesity. These interventions reduce fasting glucose, HbA1c, and cholesterol levels while increasing high-density lipoprotein cholesterol (HDL-C) levels, indicating improved metabolic health [51,62]. In terms of macromolecular drugs, recent advances have highlighted dual-targeting peptides, which improve glycemic control and exert anti-atherogenic effects by modulating lipid metabolism and vascular inflammation. Tirzepatide, a GLP-1R/GIPR co-agonist, has demonstrated significant reductions in HbA1c and body weight, alongside promising cardiovascular outcomes in patients with obesity, T2DM, and AS [164]. Moreover, utilizing high-throughput metabolomics technology, researchers have discovered that macromolecular drugs can synergistically address multiple diseases by targeting essential metabolic enzymes (3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR) and dihydroorotate dehydrogenase (DHODH)) or regulating metabolic pathways (the glucocorticoid receptor) [165]. In parallel, emerging research supports the development of microbiota-diet-drug trilateral networks as a novel framework for integrated intervention. Dietary fibres, such as β-glucan and arabinoxylan, have been shown to enrich SCFAs-producing gut bacteria, enhancing GLP-1 secretion and insulin sensitivity [166]. When combined with pharmacological agents like metformin, such networks may exert synergistic effects on metabolic and vascular parameters. A study demonstrated that dietary modulation of the microbiome improved metformin responsiveness and reduced systemic inflammation, highlighting the translational potential of such integrative strategies [167].

Therefore, integrating the comorbidity concept with “one treatment for multiple diseases” is an exciting frontier in treating CMDs. This holistic approach addresses the pathophysiological overlap between diseases and offers practical solutions to the challenges of polypharmacy and complex disease management. There remains a need to better understand how these therapeutic strategies can be optimized for clinical use, paving the way for efficient treatments to address the growing burden of CMDs.

4.2. Gut microbiota ethical considerations

While gut microbiota-related therapies hold great promise, sustained interventions carry potential risks that must be carefully taken into account. Recent studies have demonstrated that long-term antibiotic exposure can irreversibly alter gut microbial composition, decreasing colonization resistance to invading harmful bacteria and facilitating the emergence of opportunistic pathogens [168]. While engineered probiotics achieve therapeutic effects, they may also induce unintended consequences, including harmful compound synthesis, microbiome dysbiosis, and even host genome modifications, via mechanisms such as horizontal gene transfer and ecological pressure shifts. The likelihood and severity of these effects often exceed the predictive capacity of existing models [169]. The potential risks associated with introducing LBPs, especially uncharacterized microbial species, require stringent ethical guidelines and clinical trials to ensure that these therapies do not disrupt the delicate balance of the gut ecosystem or cause unintended harm [170]. The FMT application has demonstrated that restoration of a healthy gut microbiome can significantly reduce the recurrence of infection. However, variability in donor stool preparations and potential risk, including pathogen transmission, highlight the need for standardized microbiome-based therapeutics [55].

As such, ethical frameworks for personalized microbiota therapies are still nascent. Current research prioritizes safety over ecological equity, failing to address disparities in access to advanced therapies or the commodification of microbial products. Interdisciplinary collaboration (e.g., clinical physicians, nutritionists, and microbiologists) is essential to establish standardized protocols for microbial modulation, rigorous, long-term safety monitoring, and inclusive policies that balance innovation with ethical responsibility. It is worth noting that the potential for gut microbiota-drug interaction in enhancing treatment efficacy is promising, but critical gaps persist. On the one hand, most research focuses on bacterial communities, neglecting archaea, fungi, and viruses, which constitute 10%–15% of the gut ecosystem. On the other hand, although pharmacomicrobiomics has identified some drug-microbe interactions (metformin's modulation of Akkermansia spp.), mechanistic insights into how macromolecular drugs, including GLP-1 RAs or mAbs, reshape microbial networks remain scarce.

4.3. AI-driven therapeutic optimization

AI is reshaping the management of CMDs by enhancing risk prediction and enabling personalized interventions [171,172]. Although preclinical models highlight microbial modulation as a therapeutic lever, clinical validation remains nascent, constrained by data heterogeneity, limited sample sizes, and the dynamic nature of host-microbiome-drug interactions [137]. The growing emphasis on AI and microbiome therapeutics reflects their potential to address the multifactorial nature of CMDs. Previous studies have established foundational insights into microbial biomarkers [173], computational drug-microbiome predictions [174], and digital health interventions [175] by using machine learning algorithms such as random forests and support vector machines. Cutting-edge computational frameworks synthesize multi-omics datasets, spanning microbial genomics, host metabolomics, and pharmacokinetic profiles, to predict drug-microbiota interactions and optimize treatment regimens. Tools like QIIME2 and MetaboAnalyst facilitate the analysis of microbial and metabolic signatures, enabling data-driven therapeutic optimization [176,177]. Deep learning models, including graph convolutional network and attention network for microbe-drug association prediction (GCNATMDA), achieve > 96% accuracy in identifying microbe-drug associations, revealing novel opportunities for drug repurposing and microbiome-directed therapies [174]. Similarly, machine learning analyses of gut microbiome signatures in diabetes have identified microbial biomarkers linked to glucose dysregulation, enabling targeted prebiotic interventions to enhance drug efficacy [173]. Deep-learning models have demonstrated the potential to improve the prediction accuracy of disease treatment by integrating microbial data. These models can identify microbial markers and enhance the use of microbial data for disease prediction, thereby supporting clinical diagnosis and treatment. The novel drug sensitivity prediction (NDSP) model can accurately predict drug sensitivity. It provides interpretable biological features that can facilitate the development of therapeutic optimization, which aims to customize treatments for individual patients based on their distinct biological profiles [178].

Despite its vast potential in therapeutic optimization, AI faces significant challenges and ethical concerns in the clinical translation of CMDs therapies [179]. First, existing models often rely on fragmented or small-scale datasets, limiting their generalizability across diverse populations. Second, while predictive algorithms, including the NDSP model, offer interpretable biological insights for therapeutic optimization [180], analogous frameworks for CMDs require validation in longitudinal, multiethnic cohorts. Third, the dynamic integration of real-time biosensor data with adaptive AI systems, although promising for refining treatment personalization, requires standardized protocols for data harmonization and ethical governance [178]. Lastly, AI applications in biotechnology face significant ethical challenges, particularly concerning data privacy vulnerabilities, inherent algorithmic biases, and opaque decision-making processes. These issues potentially compromise individual rights, perpetuate healthcare disparities, and undermine trust in AI-driven medical solutions [181].

Based on the above challenges, it highlights the urgent need to prioritize practical and scalable strategies that can bridge innovation with clinical implementation in CMDs management. First, developing standardized protocols for microbiota-drug interaction assays is essential to ensure reproducibility, safety, and cross-study comparability, particularly in light of the complex interplay between microbial metabolism and host pharmacodynamics. Second, building open-access, high-quality multi-omics, and pharmacomicrobiomic databases will facilitate the training of AI models and support more transparent, data-driven personalization of care. Third, biomarker-guided clinical trial designs should be adopted to stratify patient populations based on genetic, metabolic, or microbial profiles, allowing therapies to be tailored more precisely. Lastly, integrating ethical oversight into AI-driven and microbiome-based therapeutic optimization is critical. This includes ensuring data privacy and algorithmic fairness and promoting equitable access to novel interventions across diverse populations. Future research may transform CMDs management into a safer, smarter, and more inclusive field by focusing on these actionable directions.

5. Conclusions

CMDs, including obesity, T2DM, and AS, continue to impose a substantial global health burden. While conventional therapies based on small-molecule drugs and lifestyle interventions remain fundamental, their limitations, such as side effects, poor adherence, and the difficulty in addressing disease heterogeneity, underscore the urgent need for comprehensive strategies. Interventions targeting the gut microbiota, such as FMT, engineered bacteria, LBPs, phage therapy, probiotics, prebiotics, synbiotics, postbiotics, and diet or natural products, have shown considerable promise in improving various metabolic outcomes associated with CMDs. In parallel, macromolecular drugs, such as peptides, mAbs, and small nucleic acid drugs, enable precise and pathway-specific interventions. Emerging evidence of two-way interactions between these pharmaceuticals and the gut microbiome suggests opportunities for synergistic therapies. Nevertheless, challenges remain, particularly regarding targeted drug delivery, causal validation of microbiota-disease links, and ethical considerations. Moving forward, AI-driven technologies holds the potential to personalize and optimize therapeutic approaches. Ultimately, the integration of small-molecule drugs, gut microbiota modulations, and macromolecular drugs, alongside lifestyle interventions, may enhance CMDs management and pave the way for efficient, systems-based comprehensive medicine.

CRediT authorship contribution statement

Jingyue Wang: Writing – original draft, Methodology, Investigation, Conceptualization. Jing Qu: Writing – original draft, Methodology, Investigation, Conceptualization. Mengliang Ye: Writing – original draft, Methodology, Investigation, Conceptualization. Ru Feng: Resources, Investigation. Xiang Hui: Resources. Xinyu Yang: Conceptualization. Jingyu Jin: Methodology. Qian Tong: Writing – review & editing, Writing – original draft. Xianfeng Zhang: Writing – review & editing, Writing – original draft. Yan Wang: Writing – review & editing, Writing – original draft, Supervision, Project administration.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: