Emphasis on Adipocyte Transformation: Anti-Inflammatory Agents to Prevent the Development of Cancer-Associated Adipocytes

Simple Summary Cancer cells that grow near adipose tissue inevitably exchange signals with adipocytes. The dynamic crosstalk between these two cell types facilitates the alteration of their cellular properties. The transformation of normal adipocytes into cancer-associated adipocytes (CAAs) provides a pro-tumorigenic niche for rapid tumor progression. However, the cancer-originated signals that mediate adipocyte transformation remain largely unknown. In this review, we discuss various inflammatory signals amplified in both obese and peritumoral adipose tissue. These inflammatory signals could mediate adipocyte transformation to phenotypes similar to CAAs by promoting adipocyte dedifferentiation and lipolysis. Epidemiological studies indicate a higher efficacy of nonsteroidal anti-inflammatory drugs in obese patients with cancer. Therefore, delivering anti-inflammatory agents can be a plausible therapeutic strategy to ameliorate the activation of tumor microenvironment components, especially adipocytes. Abstract Of the various cell types in the tumor microenvironment (TME), adipocytes undergo a dynamic transformation when activated by neighboring cancer cells. Although these adipocytes, known as cancer-associated adipocytes (CAAs), have been reported to play a crucial role in tumor progression, the factors that mediate their transformation remain elusive. In this review, we discuss the hypothesis that inflammatory signals involving NF-ĸB activation can induce lipolysis and adipocyte dedifferentiation. This provides a mechanistic understanding of CAA formation and introduces the concept of preventing adipocyte transformation via anti-inflammatory agents. Indeed, epidemiological studies indicate a higher efficacy of nonsteroidal anti-inflammatory drugs (NSAIDs) in obese patients with cancer, suggesting that NSAIDs can modulate the TME. Inhibition of cyclooxygenase-2 (COX-2) and prostaglandin production leads to the suppression of inflammatory signals such as NF-ĸB. Thus, we suggest the use of NSAIDs in cancer patients with metabolic disorders to prevent the transformation of TME components. Moreover, throughout this review, we attempt to expand our knowledge of CAA transformation to improve the clinical feasibility of targeting CAAs.


Introduction
Adipocytes are the primary constituents of adipose tissue that store energy in the form of lipids. Adipocytes sustain tissue homeostasis by secreting diverse endocrine signals, including adipokines, lipids, and exosomes [1]. Under certain pathological conditions, adipocytes undergo dynamic alterations in their physical properties. For instance, obesity and diabetes accompany adipose tissue inflammation, which induces adipocyte hypertrophy and impaired metabolism [2]. Moreover, there is growing evidence that the interaction between adipocytes and cancer cells modulates adipocyte characteristics, causing them to Figure 1. Oncogenic role of peritumoral and obese adipose tissue. Adipocyte characteristics are highly dependent on the environmental condition. At the invasive tumor front, adipocytes are exposed to various signals from cancer cells and are transformed into immature adipocytes. In parallel, under excessive consumption of nutrients, hypertrophic adipose tissue drives adipocyte death and inflammation. Both peritumoral and inflammatory adipose tissues exhibit downregulation of PPARγ and CEBPα expression, presenting the population of dedifferentiated adipocytes. Activation of lipolysis via key lipases, such as ATGL and HSL, leads to the secretion of various lipid metabolites that support tumor progression. Dedifferentiated adipocytes also provide an altered array of adipokines, which are considered pro-inflammatory and tumorigenic. The induction of adipocyte dedifferentiation and lipolysis contributes to cancer cell survival, growth, epithelial-mesenchymal transition (EMT), chemoresistance, and immune surveillance. Abbreviations: TNFα, tumor necrosis factor α; IL-6, interleukin-6; IL-1β, interleukin-1β; PGE2, prostaglandin E2; HGF1, human gingival fibroblast -1; AdipoQ, adiponectin; CCL2, C-C motif chemokine ligand 2; CCL5, C-C motif chemokine ligand 5; MCP-1, monocyte chemoattractant protein-1; Resistin, adipose tissue-specific secretory factor; MMP11, matrix metallopeptidase 11; EMT, epithelial-mesenchymal transition.

Wnt and Notch-Signaling to Induce CAA Transformation
Adipocyte dedifferentiation is a reverting process of mature adipocytes, which provides plasticity to transform adipocytes into fibroblast-like progenitor cells [7,9]. PPARγ and C/EBPα, the key regulators of adipogenesis and adipocyte differentiation, cooperatively provide the enzymes required for insulin sensitivity, lipogenesis, and lipolysis [46]. When affected by cancer cells, adipocytes undergo the downregulation of enzymes associated with adipogenesis, initiating adipocyte dedifferentiation ( Figure 2). To date, two distinct pathways (Wnt and Notch) have been proposed to explain adipocyte dedifferentiation induced by cancer cells [10,47,48]. Oncogenic role of peritumoral and obese adipose tissue. Adipocyte characteristics are highly dependent on the environmental condition. At the invasive tumor front, adipocytes are exposed to various signals from cancer cells and are transformed into immature adipocytes. In parallel, under excessive consumption of nutrients, hypertrophic adipose tissue drives adipocyte death and inflammation. Both peritumoral and inflammatory adipose tissues exhibit downregulation of PPARγ and CEBPα expression, presenting the population of dedifferentiated adipocytes. Activation of lipolysis via key lipases, such as ATGL and HSL, leads to the secretion of various lipid metabolites that support tumor progression. Dedifferentiated adipocytes also provide an altered array of adipokines, which are considered pro-inflammatory and tumorigenic. The induction of adipocyte dedifferentiation and lipolysis contributes to cancer cell survival, growth, epithelial-mesenchymal transition (EMT), chemoresistance, and immune surveillance. Abbreviations: TNFα, tumor necrosis factor α; IL-6, interleukin-6; IL-1β, interleukin-1β; PGE2, prostaglandin E2; HGF1, human gingival fibroblast -1; AdipoQ, adiponectin; CCL2, C-C motif chemokine ligand 2; CCL5, C-C motif chemokine ligand 5; MCP-1, monocyte chemoattractant protein-1; Resistin, adipose tissue-specific secretory factor; MMP11, matrix metallopeptidase 11; EMT, epithelial-mesenchymal transition.
Cancer-mediated Wnt/β-catenin signaling in preadipocytes mainly suppresses adipogenesis [9,10,47]. Wnt3α and Wnt5α, the ligands of canonical Wnt signaling, are transferred to adipocytes during the cocultivation of cancer cells. Administration of Wnt3α and Wnt5α induces the c-Jun and activator protein-1 (AP1) signaling pathways while suppressing adipogenic signals [9,10]. Additionally, Li et al. reported that mechanical pressure could dedifferentiate adipocytes through Wnt signaling activation [47]. The rapid expansion of solid tumors results in intense physical stress on adjacent stromal cells. Adipocytes, under such stress, undergo dedifferentiation to acquire a mesenchymal stem cell (MSC) phenotype. Compression-induced dedifferentiated adipocytes (CiDAs) generated by mechanical pressure activate canonical Wnt/β-catenin signaling. The co-injection of cancer cells and CiDAs into mice resulted in the enhanced growth of cancer cells [47].
A study on liposarcoma (LPS) showed that adipocyte dedifferentiation could also be mediated by active Notch signaling. Herein, mice with constitutively active Notch1 were generated in an adipocyte-specific manner (Ad/N1ICD) [48]. The Ad/N1ICD adipocytes developed an impaired lipid metabolism pathway and underwent a loss of lipid messenger required for PPARγ activity. Owing to PPARγ ligand deficiency, these adipocytes underwent downregulation of fatty acid oxidation, lipid uptake, and aerobic respiration. This led to adipocyte dedifferentiation and exhibited an enriched human LPS gene signature. Conversely, PPARγ ligand supplementation and Notch inhibition re-differentiated adipocytes and suppressed the LPS transformation of mature adipocytes in mice [48].
Both Wnt and Notch signaling are closely linked with cellular development, differentiation, and homeostasis [49,50]. Furthermore, mounting evidence supports that Wnt and Notch signaling are also involved in the pathogenesis of chronic inflammatory diseases, including cancer-mediated inflammation [49][50][51][52]. Exposure to inflammatory cytokines, such as IL-6 and TNFɑ, could upregulate Wnt signaling and abrogate the differen- Cancer-mediated Wnt/β-catenin signaling in preadipocytes mainly suppresses adipogenesis [9,10,47]. Wnt3α and Wnt5α, the ligands of canonical Wnt signaling, are transferred to adipocytes during the cocultivation of cancer cells. Administration of Wnt3α and Wnt5α induces the c-Jun and activator protein-1 (AP1) signaling pathways while suppressing adipogenic signals [9,10]. Additionally, Li et al. reported that mechanical pressure could dedifferentiate adipocytes through Wnt signaling activation [47]. The rapid expansion of solid tumors results in intense physical stress on adjacent stromal cells. Adipocytes, under such stress, undergo dedifferentiation to acquire a mesenchymal stem cell (MSC) phenotype. Compression-induced dedifferentiated adipocytes (CiDAs) generated by mechanical pressure activate canonical Wnt/β-catenin signaling. The co-injection of cancer cells and CiDAs into mice resulted in the enhanced growth of cancer cells [47].
A study on liposarcoma (LPS) showed that adipocyte dedifferentiation could also be mediated by active Notch signaling. Herein, mice with constitutively active Notch1 were generated in an adipocyte-specific manner (Ad/N1ICD) [48]. The Ad/N1ICD adipocytes developed an impaired lipid metabolism pathway and underwent a loss of lipid messenger required for PPARγ activity. Owing to PPARγ ligand deficiency, these adipocytes underwent downregulation of fatty acid oxidation, lipid uptake, and aerobic respiration. This led to adipocyte dedifferentiation and exhibited an enriched human LPS gene signature. Conversely, PPARγ ligand supplementation and Notch inhibition re-differentiated adipocytes and suppressed the LPS transformation of mature adipocytes in mice [48].
Both Wnt and Notch signaling are closely linked with cellular development, differentiation, and homeostasis [49,50]. Furthermore, mounting evidence supports that Wnt and Notch signaling are also involved in the pathogenesis of chronic inflammatory diseases, including cancer-mediated inflammation [49][50][51][52]. Exposure to inflammatory cytokines, such as IL-6 and TNFα, could upregulate Wnt signaling and abrogate the differentiation of mature adipose cells [53]. Notch signaling could also be regulated by inflammatory cytokines such as IL-1β and TNFα [50,54]. Collectively, the driver of adipocyte transfor-Cancers 2023, 15, 502 6 of 18 mation can also be regulated by diverse environmental conditions inflicting inflammatory signals on adipocytes.

Adipocyte Transformation via Cancer-Derived Inflammatory Factors
Tumor initiation and progression lead to local and systemic inflammation, which significantly disrupts tissue homeostasis [55]. In the TME, oncogenic and apoptotic signals recruit immune cells to the tumor expansion site [56,57]. In addition, enhanced ROS, hypoxia, and acidity in the TME present multiple inflammatory signals through cytokines, chemokines, growth factors, inflammasomes, exosomes, and metabolites [8,43,58]. Particularly in adipose tissue-rich environments, cancer cells attempt to fully utilize adipocytes through active crosstalk and maintain a favorable TME ( Figure 3) [5,12,22,23,43,59]. tiation of mature adipose cells [53]. Notch signaling could also be regulated by inflammatory cytokines such as IL-1β and TNFɑ [50,54]. Collectively, the driver of adipocyte transformation can also be regulated by diverse environmental conditions inflicting inflammatory signals on adipocytes.
The generation of CAAs can be mediated by inflammatory signals derived from cancer cells [60][61][62]. Inflammatory factors, such as transforming growth factor β (TGFβ) and TNFα, cooperatively inhibit the expression of genes associated with adipocyte maturation [60]. These molecules were found to be transported into adipose stromal cells at the invasive tumor front of the in vivo models. The transcriptional factors that maintain the adipocyte phenotype are downregulated in adipocytes cocultured with breast cancer cells [60]. Similarly, conditioned media (CM) from T47D breast cancer cells inhibited adipocyte Tumor cells exchange diverse signals with TME compartments to construct a favorable environment. In the adipose-rich environment, adipocytes undergo a transformation when influenced by cancer-derived secretory factors. Cancer cells and adipocytes also communicate via exosomes, which contain pro-inflammatory factors, miRNA, and metabolites. Juxtacrine signal also activates Notch signaling of adipocytes to initiate dedifferentiation. Abbreviation: TME, tumor microenvironment; ROS, reactive oxygen species; TNFα, tumor necrosis factor α; IL-11, interleukin-11; MCP1, monocyte chemoattractant protein1; PGE2, prostaglandin E2.
The generation of CAAs can be mediated by inflammatory signals derived from cancer cells [60][61][62]. Inflammatory factors, such as transforming growth factor β (TGFβ) and TNFα, cooperatively inhibit the expression of genes associated with adipocyte maturation [60]. These molecules were found to be transported into adipose stromal cells at the invasive tumor front of the in vivo models. The transcriptional factors that maintain the adipocyte phenotype are downregulated in adipocytes cocultured with breast cancer cells [60]. Similarly, conditioned media (CM) from T47D breast cancer cells inhibited adipocyte differentiation. CM treatment downregulated the expression of C/EBPα, PPARγ, and adipocyte protein 2 (AP2); such phenomena could be abrogated using neutralizing antibodies against TNFα and IL-11 [61]. Thus, cancer-derived inflammatory cytokines, such as IL-11, TGFβ, and TNFα, can stimulate the reversion of mature adipocyte phenotypes.
Cancer cachexia is prevalent among patients with advanced cancer, and its symptoms involve severe loss of adipose tissue mass. Inflammatory cytokines, including TNF-α and IL-6, are responsible for cancer cachexia and are relevant to adipocyte transformation. Adipocytes exposed to such signals experience increased lipolysis, decreased lipogenesis, impaired lipid deposition, and browning [63]. Anti-IL-6 receptor antibodies could inhibit lipolysis and adipocyte browning in cachectic mice [62], and the depletion of IL-6 from tumor cells could prevent lipolysis in cocultivated adipocytes [64,65]. Additionally, cancerderived IL-1β also triggers adipocyte transformation and adipose tissue cachexia [66].
Tumor cells also manipulate adipocytes to take full advantage when metastasizing into the adipocyte-rich bone marrow. Metastatic prostate carcinoma cells interact with the adipocytes in the bone to activate a pro-survival mechanism that allows rapid growth and escape from chemotherapy [67,68]. IL-1β secreted from cancer cells can sufficiently regulate the pro-inflammatory phenotype of adipocytes via the upregulation of COX-2 and monocyte chemoattractant protein-1 (MCP1). Since COX-2 activation increases prostaglandin E2 (PGE2) synthesis, modified adipocytes support cancer cells by promoting clonogenic growth and apoptosis resistance [69].
In cancer, the expression of inflammatory signals, such as TNFα, IL-1β, and IL-6, are crucial indicators of sensitivity to chemotherapy and patient survival. However, their effect on proximal adipose tissue has been overlooked [58,70,71]. As individual elements can alter adipocyte properties, it can be assumed that cancer-adipocyte proximity results in adipocyte transformation.

Adipocyte Transformation via NF-kB-Mediated Inflammation
NF-kB is a signaling hub for multiple inflammatory responses and is activated by various signals derived from solid tumors or impaired adipose tissues ( Figure 4A). Active NF-kB mediates diverse signals to promote lipolysis and adipocyte dedifferentiation. In this review, we discuss the mechanisms by which NF-kB modulates adipocyte transformation and the attempts that have been made to suppress its activation.

Adipocyte Transformation via NF-ĸB-Mediated Inflammation
NF-ĸB is a signaling hub for multiple inflammatory responses and is activated by various signals derived from solid tumors or impaired adipose tissues ( Figure 4A). Active NF-ĸB mediates diverse signals to promote lipolysis and adipocyte dedifferentiation. In this review, we discuss the mechanisms by which NF-ĸB modulates adipocyte transformation and the attempts that have been made to suppress its activation.  The mitogen-activated protein kinases (MAPK) pathways are also involved in the inhibition of PPARγ and C/EBPs. High concentrations of C-C chemokine receptor type 2 (CCR2), TNFα, and IL-1α in obese patients activate p38 in adipocytes and inhibit C/EBPβ and PPARγ expression [88][89][90][91]. The pharmacological inhibition of MAPK stimulated adipocyte metabolism and adipogenesis. However, due to the lack of specificity, the effect of MAPK inhibitors on adipocytes should be evaluated more cautiously [91]. Another proinflammatory cytokine, IL-15, upregulates calcineurin to inhibit PPARγ and C/EBPα [92]. The calcineurin inhibitor FK-506 could restore PPARγ and C/EBP activity in the presence of IL-15 [92].
Collectively, NF-κB and its downstream pathways are crucial mediators of adipocyte transformation. Both direct and indirect deactivation of the NF-κB pathway can regulate lipolysis and adipocyte dedifferentiation. Therefore, we suggest that anti-inflammatory agents can effectively counter malignant TME formation by targeting adipocyte transformation.

Efficacy of Anti-Inflammatory Agents in Patients with Cancer
NSAIDs, such as aspirin, ibuprofen, mefenamic acid, celecoxib, piroxicam, sulindac, and diclofenac, act as blockers of the enzyme COX to inhibit PG synthesis [99]. Eicosanoids, the COX-derived PGs, are crucial mediators of inflammation. Exposure to PG derivatives, particularly PGE2, activates NF-κB to drive inflammation in adipocytes ( Figure 5) [93]. Although COX is constitutively expressed in various cell types, COX-2 in cancer cells is thought to promote the malignant behavior of cancer cells [100,101].
Cancers 2023, 15, x FOR PEER REVIEW 9 of 18 agents can effectively counter malignant TME formation by targeting adipocyte transformation.

Efficacy of Anti-Inflammatory Agents in Patients with Cancer
NSAIDs, such as aspirin, ibuprofen, mefenamic acid, celecoxib, piroxicam, sulindac, and diclofenac, act as blockers of the enzyme COX to inhibit PG synthesis [99]. Eicosanoids, the COX-derived PGs, are crucial mediators of inflammation. Exposure to PG derivatives, particularly PGE2, activates NF-κB to drive inflammation in adipocytes (Figure 5) [93]. Although COX is constitutively expressed in various cell types, COX-2 in cancer cells is thought to promote the malignant behavior of cancer cells [100,101]. To prevent systemic inflammation, NSAIDs are used to suppress COX-mediated prostaglandin production. Some NSAIDs alter adipocyte characteristics by non-selective inhibition of NF-κB and induction of PPARγ, which prevent adipocyte transformation by pro-inflammatory signals. Abbreviation: COX, cyclooxygenase; PGE2, prostaglandin E2; NF-κB, Nuclear factor kappa-lightchain-enhancer of activated B cells; NSAID, non-steroidal anti-inflammatory drug; PPARγ, Peroxisome proliferator-activated γ.
COX-2 is overexpressed in various cancers, such as pancreatic, prostate, ovarian, breast, lung, and colon cancers; this overexpression stimulates angiogenesis, metastasis, and the chemotherapy resistance of the tumor [102]. Production of PGE2 is increased in cancer cells and can stimulate cancer cell proliferation and invasion [103,104]. Additionally, the exchange of pro-inflammatory signals between cancer cells and TME components sustains the repeated activation of the NF-κB and STAT3 pathways to exacerbate tumor malignancy [105,106]. For these reasons, NSAIDs are considered plausible candidates for cancer therapy and prevention. Furthermore, the long-term use of NSAIDs reduced the incidence of colorectal, esophageal, breast, and lung cancers [107]. As persistent inflammation is coupled with cancer progression, NSAID use has become a reasonable strategy for managing chronic inflammation and preventing the activation of TME components [108].
Continuous use of NSAIDs results in a lower incidence and mortality in patients with colorectal and lung cancers [109,110]. In addition, ibuprofen and piroxicam use significantly Figure 5. Suppression of NF-κB-mediated adipocyte transformation via NSAID treatment. COX is the major enzyme that catalyzes the conversion of arachidonic acid into prostaglandins. PGE2 and other prostaglandins induce adipocyte transformation and tissue inflammation, involving NF-κB signaling. To prevent systemic inflammation, NSAIDs are used to suppress COX-mediated prostaglandin production. Some NSAIDs alter adipocyte characteristics by non-selective inhibition of NF-κB and induction of PPARγ, which prevent adipocyte transformation by pro-inflammatory signals. Abbreviation: COX, cyclooxygenase; PGE2, prostaglandin E2; NF-κB, Nuclear factor kappalight-chain-enhancer of activated B cells; NSAID, non-steroidal anti-inflammatory drug; PPARγ, Peroxisome proliferator-activated γ.
COX-2 is overexpressed in various cancers, such as pancreatic, prostate, ovarian, breast, lung, and colon cancers; this overexpression stimulates angiogenesis, metastasis, and the chemotherapy resistance of the tumor [102]. Production of PGE2 is increased in cancer cells and can stimulate cancer cell proliferation and invasion [103,104]. Additionally, the exchange of pro-inflammatory signals between cancer cells and TME components sustains the repeated activation of the NF-κB and STAT3 pathways to exacerbate tumor malignancy [105,106]. For these reasons, NSAIDs are considered plausible candidates for cancer therapy and prevention. Furthermore, the long-term use of NSAIDs reduced the incidence of colorectal, esophageal, breast, and lung cancers [107]. As persistent inflammation is coupled with cancer progression, NSAID use has become a reasonable strategy for managing chronic inflammation and preventing the activation of TME components [108].
Continuous use of NSAIDs results in a lower incidence and mortality in patients with colorectal and lung cancers [109,110]. In addition, ibuprofen and piroxicam use significantly reduced the inflammatory potential in breast and colorectal cancers [111,112]. Moreover, patients with familial adenomatous polyposis (FAP) show decreased recurrence and lower polyp numbers when treated with sulindac [113,114]. The use of celecoxib in rat cancer models demonstrated a 90% tumor regression and a 25% reduction in the number of solid tumors [115]. Another study showed that ibuprofen inhibits cell proliferation in mouse and human colorectal cells. A 40-82% tumor regression and decreased tumor-induced angiogenesis were achieved by treatment with ibuprofen alone or in combination with the standard antineoplastic agents 5-fluorouracil or irinotecan [116]. Similarly, aspirin has demonstrated apoptotic and anti-proliferative effects in the HeLa cells. A synergistic anticancer effect for aspirin was observed when combined with doxorubicin, cooperatively inducing cell-cycle arrest, growth inhibition, and apoptosis in vitro and in vivo [117].
Numerous epidemiological studies support the notion that NSAID use benefits patients with cancer. A study involving 10,280 colorectal cancer cases showed a 27% reduction in colorectal cancer risk (odds ratio (OR) = 0.73; 95% confidence interval (CI): 0.54-0.99) [118]. A subsequent study investigated 2,118 women who had a female sibling with breast cancer. The use of non-COX-inhibiting NSAIDs did not correlate with reduced breast cancer risk among postmenopausal women. However, for premenopausal women, non-aspirin NSAIDs and aspirin reduced the risk by 34% (hazard ratio (HR) = 0.66; 95% CI: 0.50-0.87) and 43% (HR = 0.57; 95% CI: 0.33-0.98), respectively [119]. Another casecontrolled study of 1,736 breast cancer patients in Spain reported a 24% reduction in breast cancer risk (OR = 0.76; 95% CI: 0.64-0.89) in those who use non-aspirin NSAIDs [120,121]. According to an investigation of 819 patients with prostate cancer, NSAID use significantly reduced prostate cancer risk (OR = 0.48, 95% CI: 0.28-0.79) [122]. Furthermore, an investigation of 7776 patients with ovarian cancer revealed that only aspirin use reduced the risk by 9% (OR = 0.91; 95% CI: 0.84-0.99) [123]. Other epidemiological data on NSAID use and the risk of pancreatic, prostate, bladder, and renal cancers remain controversial and limited. However, due to the conflicting consequences of the use of NSAIDs, it remains unclear whether they should be widely implemented against multiple types of cancer. In addition, epidemiological data imply that the anti-tumor effect of NSAIDs varies depending on their dose and duration, as well as the cancer types [124][125][126].

Anti-Inflammatory Agents in Obese Patients with Cancer
Some cancer types are surrounded by a large population of adipocytes, and they tend to rely on environmental cues when developing their malignant behavior. Obesity-related cancers can be greatly affected by the alteration of the TME during chemotherapy. Strikingly, NSAID use in obese patients resulted in a better prognosis after cancer treatment [127][128][129][130][131][132]. Particularly, patients with inflammatory adipose tissue may benefit from suppressing the chronic inflammation caused by peritumoral adipocytes. For instance, a daily dose of aspirin was more effective in patients with colorectal cancer who had a higher BMI (25-29 kg/m 2 and >30 kg/m 2 ). Unlike the somewhat increased risk with NSAID use in patients with normal weights (BMI < 25 kg/m 2 ), obese individuals (BMI > 30 kg/m 2 ) experienced a 56% reduction in risk upon regular NSAID use [129]. Similarly, in a case-controlled study involving 5,078 women, those who used NSAIDs regularly had a significantly lower risk of breast cancer incidence (OR = 0.78; 95% CI: 0.69-0.89) [127]. Another meta-analysis of 7,120 women with endometrial cancer showed that using aspirin more than once per week caused a 15% risk reduction among overweight and obese women (OR = 0.86; 95% CI: 0.76-0.98 and OR = 0.86; 95% CI: 0.76-0.97, respectively, for aspirin; OR = 0.87; 95% CI: 0.76-1.00 and OR = 0.84; 95% CI: 0.74-0.96, respectively, for non-aspirin NSAIDs) [131]. Interestingly, there was no correlation between aspirin use and cancer risk among women of normal weight.
Although NSAID use as a generalized chemotherapeutic strategy remains uncertain, specific populations with adipose-rich cancer exhibit positive outcomes. Clinical studies support that obese patients are likely to benefit more from the protective effects of NSAID use than normal-weight patients (Table 1). Crude measures, such as BMI and waist circumference, are insufficient to predict a positive response to NSAIDs. Thus, a more precise method should be considered to anticipate the effects of NSAIDs on the TME, especially in the case of an adipose-rich environment.

Future Perspective
To date, only Wnt and Notch signaling have been suggested as the major mediators of CAA transformation. Further studies are needed to identify the key drivers of the dynamic conversion of adipocytes. As many inflammatory signals are intertwined, it is necessary to comprehend the complex network of the corresponding pathways. To date, no study has specifically targeted inflammation to mitigate CAA transformation, even though the oncogenic role of CAA is starting to be recognized. We have shown that NF-κB could regulate adipocyte properties through multiple pathways (Figure 4). Although single delivery of an NF-κB inhibitor has distinct effects on the downstream signaling [133], it should be considered which inhibitor accounts for preventing CAA transformation.
NSAIDs target COX, which is the limiting enzyme in PGE2 synthesis. As PGE2 is a potent inducer of NF-κB, NSAID use also regulates multiple NF-κB downstream signals with minimal side effects ( Figure 5). Numerous studies have demonstrated the inhibitory impact of NSAID use on cancer progression. In particular, we highlighted that patients with adipose-rich cancer benefited markedly from regular NSAID use. Furthermore, patients with a higher BMI showed better prognoses for breast, colorectal, and endometrial cancers (Table 1). It can be inferred that cancer patients with metabolic dysregulation or inflammatory adipose tissue may benefit from anti-inflammatory agents. This highlights the importance of evaluating an individual's TME when predicting therapeutic outcomes and efficacy.
There remains a need to identify standard CAA markers through which cancerderived cytokines drive CAA transformation. In addition, it is necessary to elucidate how adipocytes exposed to inflammatory signals share common and distinct features with CAAs. This will enable us to understand the resemblance between peritumoral and inflammatory adipose tissue, revealing the veiled connection between obesity and cancer. Furthermore, it is also essential to evaluate the condition of adipose tissue in patients with cancer to understand how the TME aids tumor growth. Currently, crude measures such as BMI and waist circumference are the only means used to evaluate adipose tissue condition; a more precise evaluation of patients' TME status will improve the efficacy of chemotherapy, including that based on NSAIDs.

Conclusions
CAAs contribute to the secretion of inflammatory signals, metabolic reprogramming, and ECM remodeling in cancer cells. Despite the critical role of CAAs in the TME, the cell-intrinsic and extrinsic factors that trigger adipocyte transformation remain largely unknown. Inflammatory adipose tissues impose signals on adipocytes that are akin to the inflammatory factors secreted by tumor cells into the TME. These inflammatory signals have been found to be sufficient to modulate adipocyte properties (Figure 1). The major characteristics of CAA: suppression of adipogenic potential and activation of lipolysis, are found to be more pronounced in the adipose tissue of obese cancer patients. This implies that inflammatory adipose tissue provides more malignant TME through enhanced adipocyte transformation. Particularly, patients with impaired adipose tissue may significantly benefit from the delivery of anti-inflammatory agents via reconstructing the TME. NSAID use in obese patients yields better cancer prognosis, especially in those who bear tumors in an adipose-rich environment. This phenomenon may further explain the strong correlation between obesity and cancer progression.