The World Health Organization reported that in 2022 over 890 million adults (16% of the world’s population) were living with obesity, a number that has more than doubled since 1990. Increased body mass index (BMI), a measure of adiposity, is associated with increased cancer risk in multiple cancer types, including endometrial, colorectal, breast, and prostate cancers [8]. However, an inverse relationship between BMI and lung cancer risk has been observed, often attributable to residual confounding by smoking, as smoking is linked to lower body weight and reverse causality [9]. Despite this, body fat distribution, rather than overall obesity, has been shown to contribute to lung cancer development [10]. In 2014, 7.8% of all cancers (excluding non-melanoma skin cancers) were attributed to excess body weight among US adults aged 30 and older [11], with a significantly higher burden in women compared to men (10.9% vs. 4.8%) [12]. Contrary to associations with cancer incidence, studies have found that obesity is positively associated with improved immunotherapy efficacy and OS in non-small cell lung cancer (NSCLC) patients [1,13-17]. Nevertheless, we still have an incomplete understanding of the reciprocal interactions between inflammation and metabolism during obesity, and how these independent mechanisms interact to establish a tumor supportive microenvironment.   

The obesity-associated secretome, consisting of over 50 adipokines, cytokines, and chemokines which regulate both immunity and metabolism [18] , establishes a state of chronic, subclinical inflammation and such “smoldering” inflammation is a hallmark of cancer development [19]. Interestingly, obesity-associated systemic inflammation appears to be a stronger predictor of cancer risk than BMI itself. In conjunction, both adipocyte-produced factors and subsequent chronic inflammation drive metabolic dysfunction. Notably, metabolic function itself may also independently predict obesity-related cancer risk. In the longitudinal Framingham Heart Study, metabolically unhealthy overweight individuals had a 210% greater risk of developing cancer, whereas metabolically healthy overweight individuals had only a 47% increased cancer risk both compared to normal weight subjects [20]. These studies implicate chronic inflammation and metabolic dysfunction in cancer susceptibility, and suggest that common immunologic features of both, namely immunosuppression and T cell exhaustion may play a role in response to immune checkpoint inhibition (ICI). Supporting this, ICI targeting the PD-1/PD-L1 axis has yielded positive outcomes in obese NSCLC patients [21-23], with significantly longer PFS (3.7 vs. 2.8 months; HR: 0.79; 95 % CI: 0.64-0.98; P= 0.04) and OS (15.4 vs. 13.5 months; HR: 0.73; 95 % CI: 0.57-0.95; P= 0.02) in patients with a high BMI [22]. This is referred to as the obesity paradox of cancer immunotherapy response.  A more recent study of >12,000 advanced NSCLC patients who received ICI therapy (pembrolizumab, nivolumab, atezolizumab, ipilimumab), a U-shaped association was observed where higher BMI was linked with a lower risk of mortality that those with a lower BMI, supportive of this obesity paradox in NSCLC patients [24]. However, ICI treatment was not any better than standard chemotherapy in the overweight or obese patients [24]. Additional factors, such as visceral fat index, may further clarify the role of obesity as a predictive factor in ICI therapy [25]. Obesity-mediated inflammation increases adipocyte lipolysis and the subsequent release of soluble lipids. In conjunction, proinflammatory TNFα and IL-6, elevated glucose, insulin, fatty acids, and free lipids act locally to dysregulate metabolism in target tissues resulting in oxidative stress and contributing to cancer development and/or progression. Prior to tumor formation, obesity results in elevated leptin levels, with leptin-mediated activation of STAT3 contributing to T cell disfunction and PD-1 expression in humans, non-human primates and mice [26,27]. In the context of cancer initiation and progression, obesity and the associated leptin expression combine to promote tumor progression likely through overlapping mechanisms of immunosuppression, as well as metabolic and hormonal regulation [26,28]. Yet, obesity induced increases in T cell exhaustion/PD-1 expression prior to and during cancer progression may predispose individuals to respond to ICIs [26,29]. Therefore, it seems likely that obesity-dysregulated immunity resulting in and from chronic inflammation may be reverted via ICIs to restore effective antitumor immunity.

Concurrent with T cell dysregulation, adipocyte-derived leptins, TNFα, IL-6, and CCL2 promote myeloid cell proliferation, recruitment, and inflammatory (“M1”) macrophage polarization during obesity. Inflammatory macrophages express high level of iNOS, have increased phagocytic ability, and secrete a number of pro-inflammatory cytokines and chemokines which contribute to chronic inflammation [30]. In obese animals, inflammatory macrophages are recruited to dead or dying adipocytes forming crown-like structures (CLS), histologic hallmarks of proinflammatory processes in adipose tissues. Crown-like structures produce pro-inflammatory cytokines and acute-phase proteins and are associated with both cancer incidence and insulin resistance [31,32]. Interestingly, estrogen receptor beta (ERβ) knockout mice had increased CLS, with macrophage recruitment and inflammatory polarization driven by upregulation of HIF-1α and osteopontin. Delivery of an ERβ agonist (LY3201) reduced CLS numbers, HIF-1α and osteopontin, in part, through the upregulation of prolyl hydroxylase 2 to prevent HIF-1α-mediated macrophage activation [33].

Yet, chronic inflammation invariably drives reactive immunosuppression to mitigate tissue damage and as such, adipose tissues also contain varying proportions of anti-inflammatory, wound healing (“M2”) macrophages. These wound healing macrophages have reduced inflammatory functions resulting in decreased immunosurveillance and establish tissue microenvironments capable of supporting tumor growth and metastatic dissemination [34-36]. Macrophages exposed to byproducts of extracellular matrix remodeling in adipose tissues adopt a phenotype similar to that of tumor-associated macrophages (TAMs) which support cancer initiation and progression [37]. Over time, polarization of macrophages and/or monocyte precursors may promote the establishment of the pre-metastatic niche supporting tumor initiation and subsequent tumor progression. Interestingly, administration of leptin, with or without anti-PD-1 ICI, enhanced antitumor immunity and reduced tumor volumes in both obese and lean mice associated with repolarization of TAMs towards a pro-inflammatory phenotype [38]. In addition, TAMs in humans and mice express PD-1 which inhibits their ability to phagocytose tumor cells, and similar to T cells, leptin signaling upregulates PD-1 expression on macrophages via STAT1 [27,39]. Collectively, macrophage polarization by chronic obesity-derived inflammation may be reversed via ICI therapy resulting in increased responsiveness observed experimentally and clinically.

As illustrated by Jin et al., cachexia and pretreatment weight loss are detrimental to patients receiving immunotherapy [1]. Cachexia, the involuntary loss of >5 percent of someone’s weight, is a complex metabolic syndrome resulting in systemic inflammation. Cachexia results in the overexpression of pro-inflammatory cytokines, namely IL-1, IL-6, TNFα, and IFNγ which drive the body to a catabolic state promoting the breaking down of fat and muscle [40]. In previous studies on ICI efficacy focused on BMI, underweight patients were either excluded from the analysis or grouped into the “low BMI” category. However, it is now evident that just as being overweight can impact ICI efficacy, being underweight can also play a significant role by altering the immune landscape.

In pre-cachectic mice with colon cancer, IL-6 was found to reduce ketogenesis, impairing hepatic metabolism. In combination with the calorie deficient state, hypoketonemia increased the synthesis of glucocorticoids leading to suppression of anti-tumor immunity along with failure of immunotherapy [41]. Cachexia and resulting protein catabolism reduces circulating amino acids concentration, namely serine and arginine [42,43]. In mice, L-arginine was found to be essential in regulating multiple metabolic pathways in T cells, promoting T cell proliferation and differentiation, and most importantly, improving both survival and tumoricidal activity [44]. While the exact mechanism remains unknown, it was hypothesized that L-arginine promoted a central memory like T cell phenotype [44]. Serine was found to be essential in T effector cell proliferation, mainly by contributing to one-carbon intermediates and purine for cellular metabolism, and its deficiency in mice impaired their ability to effectively mount an immune response against bacterial pathogens [45]. Thus, cachexia and pretreatment weight loss may deprive the immune system of metabolic components required to mount an effective antitumor immune response to respond to ICI-mediated disinhibition. 

In patients with pancreatic ductal adenocarcinoma and cachexia, the upregulation of IL-20 induces tissue fibrosis and correlates with poor prognosis [46]. However, several studies treating cancer patients experiencing cachexia with cytokine inhibitors have shown limited weight recovery and no significant improvement in survival [47]. Concomitantly, immune cell subsets have been associated with cancer-associated cachexia. Myeloid-derived suppressor cells in a mouse model of 4T1-induced breast cancer accelerates the development of cachexia via greater energy expenditure and chronic inflammation [48]. Moreover, a recent pilot study revealed an association between increased regulatory T cells and reduced lean mass index across patients with heterogeneous types of cancers [49]. This suggests the possibility of other immune cells and factors playing coordinating roles in the immunometabolic response to cachexia.

By contrast, some immune cells can alleviate and even prevent weight loss in cancer-induced cachexia models. In a mouse model of colon cancer, delivery of exogenous IL-4 inhibited muscle loss via increased myogenesis and was associated with increased CD8⁺ T cell and macrophage infiltration into tumors [50]. Microglial cells, specialized macrophages in the brain, were shown to have a protective effect during pancreatic cancer associated cachexia in mice, and microglial depletion via CFS-1R antagonist resulted in increased anorexia, fatigue, and muscle catabolism [51]. The efflux of macrophages from adipose tissues observed in mice with hepatocellular carcinoma was shown to mitigate fat loss and delay tumor development [52]. Furthermore, CD4⁺CD44⁺ T cells help reduce muscle degradation in mice with Lewis lung carcinoma and cachexia, though further research is needed to investigate its specific antitumor effects [53].

Nearly 50% of lung cancer cases demonstrate cachexia, with rates increasing as the disease advances, primarily driven by sarcopenia (muscle wasting) rather than loss of adipose tissue [54].  A retrospective study of more than 8,000 patients from the Japanese Lung Cancer Registry found that males and those with comorbid lung conditions such as emphysema were more likely to be cachectic, and cachectic patients had worse outcomes from chemotherapy and radiotherapy [55]. When examining the impact of sarcopenia on anti-PD-1 efficacy, non-sarcopenic NSCLC patients had a 40% overall response rate versus only 9.1% in sarcopenic patients, and a higher 1-year PFS rate (38.1% vs. 10.1%) [56]. Similarly, in a prospective study of 83 NSCLC patients treated with PD-1 or PD-L1 inhibitors, those with cachexia at treatment initiation had worse OS and increased frequency of disease progression [57]. These clinical studies align with those reported in Jin et al. [1].

In lung cancer, tumor loss-of-function mutations in STK11/LKB1, a key regulator of the energy sensor AMP-activated protein kinase, have been shown to induce cachexia in preclinical models and was correlated with pretreatment weight loss in NSCLC patients [58]. Silencing of STK11/LKB1 in preclinical lung and colorectal cancer models was associated with an altered immune tumor microenvironment (TME) and increased cachexia-associated cytokines. Additionally, STK11/LKB1 mutations are linked to reduced T cell activity, lower PD-L1 tumor expression and resistance to ICI [59]. It is clear that these mutations play a critical role in promoting metabolic and inflammatory conditions that contribute to cancer cachexia in NSCLC patients, and targeting metabolic pathways in STK11 mutant tumors is an active area of research to help mitigate cachexia.

Another promising approach for cachexia management is targeting growth factor differentiation factor 15 (GDF-15), a stress-induced cytokine and known mediator of anorexia and weight loss. GDF-15 was identified in a proteomic profiling analysis of plasma samples from NSCLC patients with and without cancer cachexia. GDF-15 was significantly elevated in plasma from cachectic NSCLC patients, both early- and late-stage disease [60]. NSCLCs from those with cachexia also had a distinct transcriptomic profile with higher expression of pro-inflammatory signaling markers along with epithelial–mesenchymal transitional pathways [60]. The humanized monoclonal antibody to GDF-15, ponsegromab, was shown to inhibit serum GDF-15 in patients with cancer-associated cachexia leading to improved weight gain, appetite, and physical activity [61]. In the follow up phase 2 trial (NCT05546476) of 187 patients (40% with NSCLC), ponsegromab induced a significant dose-dependent weight gain along with improvements in appetite and physical activity, and decreased cachexia symptoms at higher doses [62,63]. These studies provide support for further development of GDF-15-targeting therapies for the treatment of cancer-associated cachexia.   

Sex differences in cancer cachexia is an emerging area of research, with strong evidence that biological sex plays a significant role in how cancer cachexia develops and progresses. However, few studies have investigated the molecular mechanisms of sexual dimorphism in cancer cachexia [64]. In a recent study, Activin A, a protein involved in muscle breakdown, was identified as a driver of cachexia in a lung cancer murine model [65]. Interestingly, treatment with a decoy ligand targeting activin signaling, combined with an appetite-stimulating drug, was more effective in female mice, with benefits dependent on ovarian function. A separate study identified the myogenic microRNA miR-486 as a key marker of sex-specific differences in cancer-induced skeletal muscle defects and is regulated by the ERα signaling pathway [66]. These recent studies are beginning to shed light on the role of sex hormones in modulating cachexia outcomes, and it will be interesting to evaluate sex differences in response to the emerging therapeutic approaches discussed above.

It is increasingly appreciated that sex-linked genes and hormones play a pronounced role in differences in NSCLC presentation, molecular characteristics, and response to treatment between women and men [67-69]. There is a growing concern in the increasing incidence and mortality of NSCLC in women worldwide, especially in non-smoking and younger women [70-72]. The estrogen pathway is implicated in playing a role in lung cancer tumorigenesis, presentation, and treatment response [73]. Additionally, estrogens can alter the tumor immune microenvironment [73]. This has led to researchers questioning how biologic sex impacts immunotherapy response.

Sexual dimorphism in immunity has long been recognized to affect disease, with females generally possessing stronger innate and adaptive immune responses [74]. This is associated with increased response to vaccination and faster clearance of pathogens but also increased susceptibility to autoimmunity with females representing ~80% of all individuals with autoimmune disease [74,75].  By contrast, males exhibit higher mortality across all ages following infection and cancer. There are roughly 50 genes regulating immunity encoded on the X chromosome (e.g. TLR7, CXCR3, FOXP3, OGT) affecting cell differentiation, activation, pathogen/damage sensing, trafficking, proliferation and metabolism and incomplete epigenetic silencing of one of two X chromosomes in females via the long non-coding RNA Xist may result in deleterious overexpression [3,5,76,77].

The data regarding sex differences in ICI efficacy in NSCLC are mixed. A meta-analysis of 20 randomized control trials including 11,351 patients with solid tumors (mainly NSCLC and melanoma) found that males derive significantly improved response and survival compared to females when ICI therapies are administered as monotherapy compared to standard chemotherapy [78]. Interestingly, a follow up meta-analysis by these investigators found that among patients with NSCLC, women treated with combination chemotherapy and ICI had statistically superior OS compared to men, suggesting that alternate sex-specific therapeutic strategies may be required to effectively leverage antitumor immunity [79]. Subsequent retrospective and meta-analyses did not find sex differences [80,81]. In the study by Jin et al., females had a significantly lower risk of death compared to males, which was limited to first-line ICI monotherapy [1]. Of note, the previous studies did not use patient level data and did not directly compare males to females. Since female NSCLC patients generally have improved outcomes in NSCLC compared to males [82], these findings reported in Jin et al. may reflect the effect of a more favorable prognosis for females regardless of treatment, despite the relative benefit males may gain from ICI monotherapy compared to chemotherapy. In addition, immune-related adverse drug reactions to immunotherapy are more commonly reported in female patients [83], an observation also found in Jin et al. Presence of immune-related adverse events is associated with improved ICI efficacy and survival.

Cancer cells harbor oncogenic alterations that result in tumor-associated antigen expression by MHC class I and II molecules with subsequent immune recognition. Increased tumor mutational burden (TMB) and expression of neoantigens is associated with beneficial cancer prognosis. In a study of 151 NSCLC patients, high TMB was associated with increased PFS and OS in combination or mono-immunotherapy treated patients compared to those with low TMB [84]. At the same time, immune pressure drives selection cancer of cells/tumors with diminished expression of driver mutations. Likely due to more robust immunity, females and younger individuals develop tumors with poor expression of driver mutations, with younger females showing compounding effects of MHC-based selection [85]. Interestingly, females had less TMB which could explain worse responses to ICI compared to males in some studies [84].

Sex steroid hormones impact the behavior of immune cells, influencing how they respond to ICI. For example, estrogen has been shown to modulate immune cell activity in ways that may enhance anticancer responses, while testosterone can suppress immune responses, potentially affecting treatment outcomes. Hormonal influences also shift with age, especially in postmenopausal women and older men, further complicating ICI treatment responses. The role of sex hormones in ICI response suggests that targeting estrogen or androgen pathways, could improve immunotherapy outcomes by reprogramming immune responses based on sex? [86-88].  The androgen receptor (AR) plays a role in suppressing the CD8⁺ T cell activity in the TME. AR regulates key immune pathways, including the Tcf7/TCF1 transcription factor, which is critical for the development and function of progenitor exhausted CD8+ T cells in tumors. This contributes to sex differences in cancer immunity, with male tumors showing higher AR activity and reduced immune responses compared to females. When AR was deleted from CD8⁺ T cells, male mice experienced the same level of tumor protection as female mice, indicating AR suppresses effective CD8⁺ T cell responses in males, suggesting that targeting AR could improve immunotherapy outcomes, particularly in male patients [87,88]. Estrogen enhances the activity of these CD8⁺ T cells, promoting more effective tumor cell killing in some cases. Estrogen signaling can also shape the TME by influencing immune cell infiltration and function [73].

Recently, Anobile et al. reported the effect of a cancer cell autocrine 17-β-estradiol (E2)/estrogen receptor alpha (ERα) signaling loop on NSCLC progression and response to ICI [89]. Here, aromatase generates endogenous E2, which in conjunction with EGFR-downstream effectors AKT and ERK1/2, activates ERα upregulating CD274/PD-L1 transcript and protein expression [89]. High ERa expression was a stronger predictor of response to ICI (pembrolizumab) than either gender or PD-L1 expression, with authors suggesting a four-biomarker signature (EGFR activity, E2, ERα, and PD-L1) to identify NSCLC patients most likely to benefit from ICI or ICI/E2-inhibition. NSCLC patient-derived xenografts possessing high E2/ERα/PD-L1 expression were more responsive to ICI (pembrolizumab), an effect that was further increased in female-derived xenografts. Finally, co-administration of the aromatase inhibitor letrozole augmented ICI efficacy, a finding that again was stronger in females. Collectively, the E2 pathway likely promotes both tumor intrinsic as well as extrinsic (immune-mediated) mechanisms of immunosuppression within the TME leading to increased tumor growth and decreased response to immunotherapy [90]. Notably, fulvestrant, an ER antagonist was found to reduce mesenchymal features of NSCLC and enhance immune cytotoxicity [91]. Thus, the combination of E2 inhibition with ICI may provide additional benefit for the treatment of a subset of refractory NCSLS patients.

The inconsistency of sex-differences in ICI efficacy among published studies highlights the need for further research to understand the mechanism by which sex influences ICI outcomes. Additionally, addressing the historical underrepresentation of women in immunotherapy clinical trials will be crucial to adequately perform rigorous sex-specific analyses that could inform sex-specific treatment strategies [77]. Incorporating sex-based factors into cancer cachexia basic mechanistic studies will also be important and could ultimately lead to a sex-tailored approach, such as hormone manipulation or metabolic reprogramming, to optimize immunotherapeutic approaches separately for both men and women, ultimately improving patient outcomes.

In summary, these findings highlight the inherent complex interplay between body weight, sex, and ICI efficacy in NSCLC patients as summarized in Figure 1. Body weight, whether obesity or cachexia, alters the metabolic and immune landscape to, in turn, affect response to ICI. While the impact of patient sex on ICI efficacy is still not fully understood, emerging evidence suggests that an estradiol/ER/PD-L1 signaling loop may play a critical role. Additionally, cancer cachexia may manifest differently in females, with hormonal factors and estrogen signaling influencing disease severity and response to therapy. As such, considering sex-specific factors when treating and managing cancer-associated cachexia may improve patient outcomes and response to ICIs.

The authors have no conflicts of interest.

The authors have no relevant financial disclosures or financial/material support.

The authors acknowledge the original authors of the publication “Male sex and pretreatment weight loss are associated with poor outcome in patients with advanced non-small cell lung cancer treated with immunotherapy: a retrospective study”. This commentary would not be possible without their input.

All authors contributed equally to writing, editing, and reviewing the manuscript.