Male hypogonadism and MASLD are bound together in a self-reinforcing metabolic cycle. IR and hyperinsulinemia suppress the HPG axis at multiple levels: IR directly impairs Leydig cell steroidogenesis, attenuates pituitary luteinizing hormone (LH) output, and, particularly in the context of chronic hyperinsulinemia, may disrupt hypothalamic gonadotropin-releasing hormone (GnRH) pulsatility, collectively impairing testicular testosterone synthesis [22–24]. Visceral adiposity further amplifies this suppression via estrogen-mediated negative feedback and direct pro-inflammatory cytokine (TNF-α, IL-6) signaling at the hypothalamus and pituitary [17, 20, 21]. In turn, hypogonadism promotes adipogenesis, reduces skeletal muscle mass, and exacerbates whole-body IR, creating a vicious cycle that accelerates hepatic lipid accumulation and MASLD progression (Figure 1) [17, 19, 20]. Understanding this reciprocal loop is essential to appreciating why TRT may offer a mechanistically distinct and potentially complementary strategy in MASLD management, alongside conventional pharmacotherapies.

At the hepatic level, TRT regulates lipid handling and steatosis progression. Activation of hepatic androgen receptors (ARs) suppresses DNL by downregulating core lipogenic enzymes, including acetyl-CoA carboxylase (ACC) and fatty acid synthase (FAS), thereby reducing triglyceride (TG) synthesis. In castration models, lipogenic transcriptional programs driven by SREBP-1c are significantly upregulated, promoting endoplasmic reticulum (ER) stress and lipid accumulation; TRT reverses these effects, restoring metabolic balance and alleviating steatosis [18, 33–38] (Figure 2).

Complementing this AR-dependent control, preclinical hypogonadism models indicate that TRT can activate hepatic energy-sensing pathways: in high-fat diet (HFD)-induced MASLD mice, TRT induces adenosine monophosphate-activated protein kinase α (AMPKα) phosphorylation at Thr172, leading to functional inhibition of ACC, enhanced mitochondrial fatty acid β-oxidation, and suppression of lipid droplet biogenesis through downregulation of perilipin-2 [32, 39]. These coordinated actions result in ~40% improvement in histologic steatosis scores, normalization of circulating TG and cholesterol, and attenuation of inflammatory mediators (decreased TNF-α and IL-6), consistent with earlier rodent studies demonstrating a reduction in inflammation without adverse glycemic perturbation [30, 38–41] (Figure 3). TRT further promotes hepatic lipid clearance by enhancing very low-density lipoprotein (VLDL) secretion, preventing intracellular TG retention, and reinforcing its antisteatotic profile [42]. The clinical outcomes observed in the LiFT trial, including ≥ 30% reductions in liver fat content by magnetic resonance imaging-proton density fat fraction (MRI-PDFF) and steatosis resolution in 53.8% of treated patients, are consistent with, though not mechanistically confirmatory of, the preclinically established AMPK pathway.

Beyond the liver, TRT may exert systemic metabolic effects that indirectly protect hepatic function, although responses vary across populations and trial designs. Some studies report reductions in adiposity-related measures and improvements in IR [36, 43]. Enhanced GLUT4 (glucose transporter type 4, encoded by Slc2a4) translocation in skeletal muscle and adipose tissue caused by TRT contributes to improved glucose disposal and reductions in homeostatic model assessment of IR (HOMA-IR), while AMPK activation may amplify these benefits by sensing ATP depletion in insulin-resistant states and coordinating cellular energy homeostasis [20, 32, 39, 44–46] (Figure 4). The resulting suppression of hyperinsulinemia-driven hepatic lipogenesis provides an additional mode of steatosis control [30, 36, 39, 44]. Meta-analyses corroborate these systemic effects, reporting mean hepatic fat reductions of ~6.66% and, after baseline adjustment, relative liver fat decreases of 38.3% following TRT [36].

Importantly, TRT also modulates inflammatory and fibrotic pathways central to MASLD progression. AMPK activation attenuates NLRP3 inflammasome signaling and suppresses Kupffer cell-mediated inflammatory responses, thereby reducing hepatocyte ballooning and improving liver biochemistry [i.e., decreased alanine aminotransferase (ALT)/aspartate aminotransferase (AST)] in preclinical HFD models, without evidence of fibrosis exacerbation [39, 47, 48] (Figure 5). These mechanistic insights are reflected clinically in the LiFT trial, where fibrosis improved in 80% of TRT-treated patients compared with 33% in placebo controls [49].

Collectively, these interconnected pathways, now strengthened by AMPK-centered evidence, position TRT as a rational adjunctive therapy for a biologically defined subset of MASLD patients. Further studies are warranted to better delineate its indirect effects on fibrogenesis. Future phase 3 trials should evaluate potential synergistic strategies combining TRT with AMPK-modulating agents (e.g., metformin).

Beyond its metabolic and lipogenic effects, testosterone exerts significant anti-inflammatory actions that are directly relevant to MASLD pathobiology. Chronic low-grade hepatic and systemic inflammation is a defining and self-perpetuating feature of MASLD, contributing to disease progression from simple steatosis toward steatohepatitis, fibrosis, and cirrhosis [50]. Testosterone has been shown to suppress multiple pro-inflammatory mediators, including TNF-α, IL-1β, and IL-6, while promoting an anti-inflammatory cytokine milieu, effects mediated at least in part through AR-dependent modulation of NF-κB signaling [41]. In the context of male hypogonadism, the loss of these testosterone-mediated anti-inflammatory effects may therefore represent an independent, additive mechanism through which androgen deficiency accelerates MASLD progression, distinct from, yet synergistic with, the hyperinsulinemia-driven HPG axis suppression described above. This inflammatory axis further strengthens the biological rationale for TRT as a hepatoprotective intervention in hypogonadal men with MASLD and underscores the need for future trials to incorporate inflammatory biomarkers (e.g., high-sensitivity CRP, IL-6, TNF-α) as secondary endpoints alongside histological outcomes.

TRT, regardless of modality/formulation, consistently reduces hepatic steatosis in men with hypogonadism. Magnetic resonance imaging (MRI)-based assessment showed a 38.3% relative reduction in hepatic fat content over 40 weeks in T2D men treated with injectable testosterone undecanoate [61]. Similarly, in the LiFT program, oral testosterone undecanoate [(Lipocine Inc.) LPCN 1144] resulted in significant radiologic improvements. Notably, the LiFT Phase 2 trial included both hypogonadal and eugonadal men, and subgroup-specific effects were not fully delineated; therefore, the extent to which observed benefits are attributable to correction of hypogonadism versus pharmacologic testosterone exposure remains uncertain. After 16 weeks, MASLD resolved in nearly half of treated men, where resolution was defined as MRI-PDFF < 5%, with mean relative reductions in liver fat of approximately 50–55% across trials [49, 56]. In obese men with severe obstructive sleep apnea (OSA), TRT resulted in smaller but statistically significant reductions in computed tomography (CT)-quantified liver fat over 18 weeks, reflected by modest increases in hepatic attenuation relative to placebo [55]. Additional randomized evidence from a single-center trial in men with T2D and hypogonadism demonstrated a 1.7-fold reduction in hepatic fat fraction following six months of transdermal testosterone gel, accompanied by significant improvements in aminotransferases, GGT, IR, and visceral adiposity [57], reinforcing the consistency of steatosis reduction in metabolically high-risk populations. These trial-level findings are further supported by a 2025 systematic review and meta-analysis of nine interventional studies, which reported directionally consistent reductions in hepatic steatosis across TRT formulations and treatment durations, along with improvements in liver enzymes and, in some trials, MASLD or MASH resolution, albeit with substantial heterogeneity in study design, populations, and endpoints [36].

Two long-term observational reports [58, 59], likely derived from overlapping or related registry datasets, demonstrated sustained improvement with continued TRT exposure. Given similarities in study design, authorship, and patient populations, these reports should not be interpreted as independent cohorts, and their findings are best considered collectively rather than cumulatively. Across these studies, which followed large patient cohorts for 8–12 years, fatty liver index (FLI) declined from the high-risk range (~83–84) to intermediate-risk values (~66–67), accompanied by reductions in GGT, bilirubin, TGs, and cardiovascular mortality. Treatment interruptions were associated with partial reversal of hepatic and metabolic improvements, supporting a treatment-dependent effect, though these findings relied on surrogate indices rather than direct imaging or histological confirmation. A prospective observational study of hypogonadal men with severe obesity undergoing bariatric surgery reported lower steatosis and NAFLD activity scores, reduced hepatic TG content, and normalization of liver lipid profiles relative to untreated hypogonadal controls in TRT-treated individuals [37].

Evidence for modification of fibrosis with TRT remains exploratory. In a 2024 sub-analysis of the EARTH study, hypogonadal men with high baseline fibrosis risk [fibrosis-4 index (FIB-4) ≥ 1.30] demonstrated a modest but statistically significant reduction in FIB-4, from 1.98 ± 0.52 to 1.87 ± 0.60 (p = 0.0277), which was associated with improvements in waist circumference, body fat, and TG. In contrast, there was no significant change in FIB-4 in men with low baseline fibrosis risk [63], suggesting that any potential antifibrotic effect may be confined to individuals with pre-existing hepatic injury and may be mediated indirectly through improvements in adiposity and metabolic dysfunction, rather than direct hepatic remodeling.

The safety profile of TRT in men with MASLD appears to be generally favorable. A systematic review and meta-analysis found that adverse events were comparable between testosterone-treated and placebo groups, with no severe cardiac or hepatic events, including cholestasis and HCC, reported [36]. In the study of Apostolov et al. [61], TRT was well-tolerated with rare, serious adverse events that were not significantly different from placebo. In the EARTH study sub-analysis, TRT administered over 12 months was generally well-tolerated in Japanese hypogonadal men, with no serious adverse events attributable to the treatment reported [63].

Predictable androgenic effects have been documented, including prostate-related changes such as modest PSA elevations, but no study reported prostate cancer in either treatment or placebo groups [36]. PSA increases above 1 μg/L occurred in ~10% of testosterone-treated patients [43]; benign prostatic hyperplasia was not reported as an outcome in that source. In studies that directly reported hematological endpoints, TRT increased hemoglobin and hematocrit [55], with one patient experiencing an increase in hemoglobin to > 180 g/L [43]. Other MASLD-focused sources discuss polycythemia as a potential adverse event or focus on hepatic outcomes without reporting hematological parameters [36]. In the Maseroli observational study of men with severe obesity undergoing bariatric surgery, there were no hepatic concerns in TRT-treated hypogonadal men, though the study duration was relatively short (mean 30 weeks) and safety monitoring was not the primary endpoint [37].

Cardiovascular safety remains the most consequential area of uncertainty surrounding TRT. Concerns about its potential for adverse cardiovascular events have been longstanding and are well-documented in the literature [64–66]; however, short- to medium-term data have not demonstrated a significant increase in major adverse cardiovascular events, offering some reassurance for near-term clinical use [67]. Long-term registry data have further reported favorable outcomes, including significantly reduced overall mortality in TRT-treated patients compared with controls (7.8% vs. 15.2%, p = 0.035 in the Al-Qudimat cohort, with similar trends observed in the Yassin registry) [58, 59]. These findings contrast with meta-analytic evidence. Meta-analyses reported an approximately threefold increased risk of myocardial infarction with TRT [relative risk (RR) 3.12, 95% CI 0.34–29.01]; notably, this estimate was limited by small event numbers and wide CIs crossing unity, precluding definitive conclusions [36]. More recently, a real-world cohort study also demonstrated that long-term TRT was associated with increased cardiovascular risk [68], reinforcing the concern that short- to medium-term reassurance cannot be extrapolated to prolonged exposure. Taken together, these conflicting signals highlight the critical need for large-scale, prospective safety registries with extended follow-up. In parallel, pharmacological strategies aimed at augmenting endogenous testosterone production rather than exogenous replacement may represent an alternative avenue warranting further investigation, potentially offering a more favorable cardiovascular risk profile [69].

Formulation-specific data are reassuring. The LPCN 1144 oral testosterone prodrug study reported no serious adverse events, with only mild to moderate adverse events unrelated to treatment observed [56]. In the Phase 2 LiFT trial, LPCN 1144 demonstrated a favorable safety profile over 36 weeks, with adverse events distributed similarly between treatment and placebo groups [49, 56]. The small biopsy-based pilot study (n = 3) conducted over 52 weeks reported no safety concerns; however, the extremely limited sample size precludes meaningful conclusions regarding safety or efficacy [60]. In the Lee et al. [62] secondary analysis of the T Trials of older hypogonadal men (≥ 65 years), transdermal testosterone gel was generally well-tolerated over 12 months, with safety profiles consistent with the broader T Trials cohort, though this population excluded severe obesity [body mass index (BMI) > 37] and uncontrolled diabetes [hemoglobin A1c (HbA1c) > 8.5%]. Importantly, no studies reported hepatic adenoma or HCC in either treatment or control groups [36].

Taken together, the safety considerations surrounding TRT can be organized into three domains that are directly relevant to patient selection and clinical monitoring. With respect to cardiovascular risk, prior evidence has been inconsistent; however, the TRAVERSE trial provides reassurance in the short- to medium-term, demonstrating that TRT is noninferior to placebo with respect to major adverse cardiovascular events, albeit with a higher incidence of certain adverse events such as atrial fibrillation, pulmonary embolism, and acute kidney injury [67], a more recent real-world cohort study reported increased cardiovascular risk with long-term TRT exposure [68], and earlier registry data yielding apparently favorable mortality outcomes should be interpreted cautiously given the methodological limitations discussed above. Clinicians should therefore formally assess cardiovascular risk prior to initiation and exercise particular caution in men with pre-existing cardiac disease, uncontrolled hypertension, or a recent cardiovascular event—groups that were largely excluded from the trials reviewed here. With respect to hematological safety, erythrocytosis is a predictable and dose-dependent effect of TRT, and polycythemia is noted as a potential adverse event in MASLD-focused evidence [36]. Hemoglobin and hematocrit should be measured at baseline and monitored periodically during treatment, with dose reduction or temporary discontinuation considered if hematocrit exceeds 54%, consistent with established TRT monitoring guidance [16, 43]. With respect to prostate safety, MASLD-focused sources support a need for vigilance by reporting PSA-related signals [43], while baseline PSA assessment, digital rectal examination where clinically indicated, follow-up monitoring, and contraindications in men with known or suspected prostate or breast cancer should be grounded in established TRT guidance [16]. Regarding biochemical thresholds for treatment initiation, AUA guidelines recommend confirming hypogonadism with at least two morning TT measurements below 300 ng/dL in the presence of consistent symptoms, supplemented by calculated free testosterone and SHBG in cases where SHBG abnormality is suspected, a clinically important consideration given that obesity and IR, both prevalent in MASLD, suppress SHBG and may cause TT to underestimate bioavailable androgen [16].

Overall, while TRT demonstrates an acceptable safety profile in hypogonadal men with MASLD, monitoring for hematological changes, prostate parameters, and cardiovascular risk factors remains essential, and longer-term studies are needed to definitively establish safety across diverse patient populations.

Despite encouraging findings, the current evidence base is constrained by several important methodological limitations. Many studies rely on surrogate indices of hepatic steatosis and fibrosis, such as the FLI and FIB-4, which were originally developed for risk stratification rather than for monitoring treatment-induced changes, thereby limiting inference regarding true histological improvement [58, 59, 63]. While MRI-based techniques, particularly MRI-PDFF, allow accurate quantification of hepatic steatosis [56, 61], they do not capture the defining histological features of MASH, including lobular inflammation, hepatocellular ballooning, and fibrosis stage. In addition, most RCTs are of relatively short duration (generally under one year), which is insufficient to meaningfully assess fibrosis progression or regression, processes that typically evolve over years [55–57, 63]. Conversely, longer-term observational studies reporting sustained or durable improvements lack randomization and are therefore susceptible to selection bias and residual confounding [37, 58, 59]. Furthermore, substantial heterogeneity in study populations, TRT formulations, outcome definitions, biopsy availability, and follow-up duration complicates interpretation, limits generalizability, and constrains pooled inference in meta-analyses [36]. TRT is not included in current American Association for the Study of Liver Diseases (AASLD) and European Association for the Study of the Liver (EASL) guideline recommendations for MASLD [2, 11]; this absence should be interpreted in light of the current evidence landscape rather than as a conclusion explicitly attributed to those guidelines.

Large, well-powered Phase 3 RCTs incorporating standardized, clinically meaningful histological endpoints remain a critical unmet need in this field. Future trial designs should adopt regulatory-grade endpoints, specifically MASH resolution without worsening of fibrosis and ≥ 1-stage fibrosis improvement, alongside sufficient follow-up duration to adequately assess long-term disease modification. In this regard, studies such as the biopsy-based TEREPINS trial represent meaningful advances toward establishing the therapeutic efficacy of TRT in MASLD. Beyond trial design, biomarker-driven patient enrichment strategies will be essential for identifying subgroups most likely to derive benefit. Specifically, candidate enrichment criteria include baseline FIB-4 ≥ 1.30 and SHBG levels, together providing a more targeted and mechanistically informed approach to patient selection [63].

Given the complex and multifactorial pathophysiology of MASLD, combination therapeutic strategies targeting complementary mechanisms are a logical next step. Recently approved therapies, including resmetirom, a thyroid hormone receptor-β agonist [5–7], and semaglutide, a GLP-1 receptor agonist [8–10], primarily target hepatic lipid metabolism and systemic energy balance. TRT operates through AR-mediated transcriptional regulation and AMPK activation—mechanisms that are largely non-overlapping with those of resmetirom or semaglutide, though clinical validation of these pathways specifically in MASLD remains limited. This mechanistic distinctiveness positions TRT as a potentially complementary agent in combination regimens, particularly in metabolically high-risk men with hypogonadism and comorbid T2D or obesity. Of note, the convergence of TRT and metformin on AMPK signaling raises the possibility that co-administration could produce additive or synergistic benefits on hepatic lipid homeostasis and insulin sensitivity, an avenue that warrants dedicated prospective investigation [39, 48].

Looking ahead, integrating TRT with approved agents like resmetirom and semaglutide may offer a mechanism‑based, personalized approach to MASLD management in hypogonadal men. However, prospective clinical evidence is currently lacking, and dedicated combination trials are needed to establish efficacy and safety.

While preclinical studies provide strong mechanistic support for TRT in MASLD, translational validation in human tissue remains a critical gap. Experimental evidence demonstrates that testosterone activates AMPK signaling, suppresses lipogenic transcriptional programs (notably SREBP-1c), enhances mitochondrial β-oxidation, and attenuates both inflammatory and fibrotic signaling cascades [32, 39]. To confirm whether these mechanisms translate to the human liver, future studies should incorporate paired pre- and post-treatment liver biopsies, coupled with integrated multi-omics (transcriptomic, proteomic, and metabolomic) analyses. Such approaches would help verify pathway-level changes and identify potential biomarkers of therapeutic response. Particular attention should be paid to validating the effects of TRT on hepatic stellate cell (HSC) activation and fibrogenesis. This is especially important because clinical evidence in the TRT field currently lacks direct data on liver fibrogenesis.

Current evidence is largely derived from relatively homogeneous populations, limiting generalizability across diverse ethnic, geographic, and metabolic subgroups. Future studies should therefore prioritize inclusion of underrepresented populations to better characterize treatment effects across varying disease phenotypes and cardiometabolic risk profiles.

In addition, cost-effectiveness analyses are currently lacking, despite the rapidly expanding therapeutic landscape for MASLD, including TRT, resmetirom, and GLP-1 receptor agonists. Comparative studies evaluating the economic and clinical value of these therapies, both as monotherapy and in combination, are essential for informing clinical decision-making and healthcare policy development.

Finally, head-to-head comparisons across TRT formulations (oral, IM, and transdermal) in well-defined MASLD populations are needed to identify the optimal treatment strategies and facilitate integration into future clinical guidelines.

Building on these limitations, several priorities should guide the future development of TRT in MASLD. First, adequately powered Phase 3 RCTs with standardized histopathological endpoints and extended follow-up are urgently needed, consistent with contemporary MASLD regulatory and guideline expectations. Ongoing studies, including LPCN 1144 and the paired-biopsy TEREPINS trials, are critical next steps. Second, combination strategies incorporating FDA-approved agents such as resmetirom and semaglutide warrant systematic evaluation, given that TRT’s androgenic and AMPK-mediated mechanisms are largely non-overlapping with thyromimetic and incretin pathways [39]. Third, biomarker-driven stratification using SHBG, baseline FIB-4 (≥ 1.30) [63], and AR genetic variants may enhance therapeutic precision and optimize patient selection. Fourth, translational mechanistic studies incorporating paired pre- and post-treatment liver biopsies are needed to validate, in humans, the AMPK activation, lipogenic suppression, and antifibrotic stellate cell effects observed in preclinical models. Fifth, long-term safety registries should systematically track HCC incidence, cardiovascular events, and hematological complications, because existing meta-analytic and registry data provide incomplete coverage of these outcomes: cardiovascular signals are reported in registry datasets, whereas HCC incidence and hematological complications require more systematic capture [36, 58, 59]. Sixth, future trials should enroll more diverse populations and incorporate formal cost-effectiveness analysis, both of which are currently absent from the evidence base.