Association between chronic stress-related amygdala metabolic activity and distant metastasis in colorectal cancer

Background: Chronic stress has been implicated in cancer progression through neuroendocrine and inflammatory pathways, but its role in colorectal cancer (CRC) remains uncertain. The amygdala, a key stress-responsive brain structure, demonstrates measurable metabolic activity on ¹⁸F-fluorodeoxyglucose positron emission tomography/computed tomography (¹⁸F-FDG PET/CT) and may serve as a surrogate imaging biomarker of chronic stress. This study aimed to investigate whether elevated amygdala metabolic activity (AmygA) is associated with distant metastasis in patients with CRC.

Patients and methods: This study included patients with newly diagnosed CRC who underwent pre-treatment ¹⁸F-FDG PET/CT and curative-intent surgery between January 2019 and December 2023. AmygA was defined as the ratio of maximum standardized uptake value (SUV_max) of the amygdala to the mean standardized uptake value (SUV_mean) of the ipsilateral temporal lobe. Receiver-operating characteristic curve analysis determined the optimal AmygA threshold for predicting distant metastasis, and multivariable logistic regression identified independent predictors.

Results: Seventy-six patients were analyzed, of whom 21 (27.6%) had distant metastasis. AmygA was significantly higher in patients with distant metastasis than in those without (1.17 ± 0.06 vs. 1.08 ± 0.06; p < 0.001). The optimal AmygA cutoff value for predicting distant metastasis was 1.159, yielding 71.4% sensitivity and 89.1% specificity (area under the curve = 0.844; p < 0.001). Univariable analysis identified advanced T stage, lymph node metastasis, elevated AmygA, increased spleen SUV_max, and higher serum tumor marker levels as significant variables. In prespecified parsimonious multivariable logistic regression models with bootstrap internal validation, elevated AmygA (> 1.159) remained independently associated with distant metastasis.

Conclusions: Elevated amygdala metabolic activity on pre-treatment ¹⁸F-FDG PET/CT, a surrogate marker of chronic stress, was independently associated with distant metastasis in CRC. AmygA might serve as a novel imaging biomarker for risk stratification and offer insight into stress-related neural mechanisms underlying metastatic progression.

Introduction

Colorectal cancer (CRC) is the third most commonly diagnosed malignancy and a major contributor to cancer-related deaths worldwide (1, 2). Although advances in therapy have improved outcomes for localized CRC, metastatic progression remains the primary determinant of poor prognosis (1, 3–5). Emerging evidence suggests that chronic stress, resulting from prolonged exposure to persistent stressors, plays a significant role in cancer progression through multiple biological mechanisms (6–10).

Chronic stress stimulates the hypothalamic-pituitary-adrenal (HPA) axis and the sympathetic nervous system, leading to elevated release of stress hormones, including glucocorticoids and norepinephrine. These hormones facilitate tumor development and progression by causing DNA damage, suppressing p53 tumor-suppressive activity, and modifying the tumor microenvironment. Additionally, chronic stress promotes systemic inflammation by increasing circulating pro-inflammatory cells and upregulating pro-inflammatory cytokines, thus fostering a pro-tumorigenic state. Furthermore, chronic stress facilitates the formation of neutrophil extracellular traps (NETs) through glucocorticoid-mediated pathways, promotes the recruitment of myeloid-derived suppressor cells (MDSCs) via the TAM/CXCL1–CXCR2 axis, and further supports tumor progression by creating a microenvironment conducive to metastasis. Empirical evidence suggests that chronic stress is associated with cancer risk and tumor aggressiveness across multiple malignancies, including stomach (11, 12), lung (13, 14), breast (15–17), endometrium (18), head and neck (19), and skin cancers (20, 21).

The amygdalae are key components of the limbic system located in the medial temporal lobes and are essential for regulating endocrine, autonomic, and behavioral responses to psychological stress. Previous studies have demonstrated that amygdala metabolic activity (AmygA) can be quantitatively assessed using ¹⁸F-fluorodeoxyglucose positron emission tomography/computed tomography (¹⁸F-FDG PET/CT) with excellent reproducibility (18, 19, 22, 23). Increased AmygA has been reported to predict chronic stress and to be associated with cancer progression (18, 19), highlighting its potential as an imaging biomarker of stress-related oncogenic activity. These findings suggest AmygA may serve as an imaging biomarker reflecting chronic stress burden.

Despite a growing body of evidence linking chronic stress to cancer pathogenesis, its relationship with CRC remains inconclusive. A large-scale prospective cohort study conducted in 2017 failed to establish a definitive association (24), but recent findings suggest that chronic stress may contribute to colorectal cancer progression by enhancing glycolysis and inducing dysbiosis (25, 26). Therefore, we investigated whether elevated AmygA measured on ¹⁸F-FDG PET/CT is associated with distant metastasis in patients with CRC.

Patients and methods

Patients

From January 2019 to December 2023, 76 patients with newly diagnosed colorectal cancer who underwent pre-treatment ¹⁸F-FDG PET/CT imaging and curative-intent surgery were retrospectively analyzed.

Patients with a history of other malignancies, autoimmune or chronic inflammatory diseases, neurologic disorders (dementia or stroke), or psychiatric conditions (mood or psychotic disorders) were excluded. Those with a prior history of brain surgery before ¹⁸F-FDG PET/CT or with active infection or systemic inflammatory comorbidities were also excluded. To minimize staging bias, 33 patients who had received chemotherapy before surgery were excluded, as their ¹⁸F-FDG PET/CT was performed prior to treatment whereas pathological staging was assessed after treatment, potentially leading to treatment-related downstaging. Ultimately, 76 patients met the inclusion criteria and were included in the final analysis (Figure 1).

Figure 1

Figure 1. Patient flow diagram. ¹⁸F-FDG PET/CT; ¹⁸F-fluorodeoxyglucose positron emission tomography/computed tomography.

This study was approved by the Institutional Review Board of Korea University Anam Hospital (Approval No. 2024AN0550). Given the retrospective nature of the study, the board waived the requirement for informed consent.

Tumor staging

Tumor staging was performed according to the American Joint Committee on Cancer (AJCC) 8th edition (27). Pathologic T and N staging were confirmed by histopathological analysis of resected specimens following definitive surgery. M staging, indicating the presence of distant metastasis, was radiologically determined using contrast-enhanced CT of the chest, abdomen, and pelvis with supplemental MRI or ¹⁸F-FDG PET/CT imaging.

Data collection, anthropometric, and laboratory assessments

Body mass index (BMI) was calculated by dividing weight in kilograms by the square of height in meters (kg/m²). Blood samples were obtained within the week before surgery. C-reactive protein (CRP) levels were measured using a chemiluminescence immunoassay (Beckman Coulter, Brea, CA, USA). Serum carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA 19-9) levels were assessed by radioimmunoassay (Gamma Pro, Kaien, Seoul, Korea). Smoking status was categorized as ‘yes’ for individuals with any history of smoking and ‘no’ for never-smokers (28, 29). Alcohol consumption was classified as “yes” for individuals who consumed alcohol at least three times per week and “no” otherwise; each serving corresponded to 355 mL (12 fl oz) of beer or an equivalent amount of alcohol (28, 30).

¹⁸F-FDG PET/CT

All patients fasted overnight before undergoing ¹⁸F-FDG PET/CT imaging to maintain blood glucose levels below 180 mg/dL. PET/CT imaging was performed approximately 60 minutes after intravenous injection of 5.29 MBq/kg of ¹⁸F-FDG, covering the region from the skull vertex to the proximal thighs. A PET/CT scanner (Gemini TF, Philips Medical Systems, Cleveland, OH, USA) incorporating a lutetium–yttrium oxyorthosilicate full-ring PET detector with time-of-flight capability and a 16-slice helical CT scanner was used. The procedure began with a CT scan (120 kVp, 50 mA, 4 mm slice thickness), which was performed for attenuation correction, and was followed by a PET scan conducted with a spatial resolution of 4.4 mm and an axial field of view of 18 cm. The PET scan covered nine-bed positions, with an acquisition time of one minute per position. Images were reconstructed using a three-dimensional row-action maximum likelihood algorithm with iterative reconstruction techniques.

Image analysis

Two board-certified nuclear medicine physicians independently reviewed ¹⁸F-FDG PET/CT images on a commercially available workstation (MIM Software, version 7.2.10, GE Healthcare, Cleveland, OH, USA). Both readers were blinded to all clinical and pathological information.

Quantitative analysis of ¹⁸F-FDG uptakes was performed using standardized uptake values (SUVs). Bilateral amygdalae were delineated on axial PET/CT images using established anatomical landmarks. The anterior boundary was defined as the inferior margin of the lateral ventricles adjacent to the thalamus, and the posterior boundary as the crux of the fornix located anterior to the basilar artery. The internal capsule defined the lateral and inferior boundaries. Circular 15-mm regions of interest (ROIs) were manually placed over each amygdala, and the maximum SUV (SUV_max) within each ROI was recorded. The average of bilateral SUV_max values was designated as amygdala SUV_max. Amygdala metabolic activity (AmygA) was defined as the ratio of amygdala SUV_max to the mean SUV (SUV_mean) of the ipsilateral temporal lobe while maintaining alignment with the same plane where the amygdala ROIs were located (18, 22, 23, 31).

Elevated metabolic activity in the spleen and bone marrow on ¹⁸F-FDG PET/CT is a recognized surrogate marker of increased myelopoietic activity and systemic inflammation (32, 33). ROIs were drawn over the entire spleen and axial slices of vertebral bodies from L3–L5 to assess metabolic activity in spleen and bone marrow (BM). The average maximum standardized uptake values (SUV_max) across these ROIs were calculated and recorded as spleen SUV_max and BM SUV_max, respectively, in accordance with previously validated protocols (18, 22, 34). Interobserver correlation coefficients of the measured SUV indices were greater than 0.9.

Statistical analysis

Results are presented as means ± standard deviations unless otherwise specified. Categorical variables were compared using the Chi-squared (χ²) test or Fisher’s exact test, as appropriate. The normality of continuous variables was evaluated using the Shapiro–Wilk test. The Student’s t test was used to analyze parametric data, and the Mann–Whitney U test was used for nonparametric data. Receiver-operating characteristic (ROC) curve analysis was performed to determine the optimal cutoff value of AmygA for predicting distant metastasis. Area under the curve (AUC), sensitivity, and specificity were calculated. To identify factors independently associated with distant metastasis, both univariable and multivariable logistic regression analyses were performed. Given the limited number of metastatic events, multivariable inference was based on prespecified parsimonious logistic regression models. AmygA was analyzed as a dichotomous variable using a ROC-derived cutoff. Internal validation was performed using bootstrap resampling (1,000 resamples), and bias-corrected and accelerated (BCa) 95% confidence intervals were obtained for multivariable odds ratios. Spearman’s rank correlation coefficients were used to evaluate associations between AmygA and systemic inflammatory markers. Statistical analysis was performed using MedCalc software version 23.2.1 (MedCalc Software Ltd, Ostend, Belgium) and SPSS software version 17.0 (SPSS Inc, Chicago, IL, USA). Statistical significance was accepted for p values < 0.05. The Bonferroni correction was applied to adjust for multiple comparisons.

Results

Patient characteristics

Seventy-six CRC patients of mean age 66.6 ± 13.0 years (range, 34–88 years) were included. The cohort consisted of 44 men (57.9%) and 32 women (42.1%). The mean body mass index (BMI) was 23.3 ± 3.9 kg/m² (range, 13.7–35.6 kg/m²).

Regarding primary tumor locations, the most common sites were the sigmoid colon (30 patients, 39.5%) and rectum (20 patients, 26.3%). Histologically, most tumors were moderately differentiated (67 patients, 88.2%). Based on pathologic evaluation, T3 was the most prevalent stage and was observed in 43 patients (56.6%). Regarding nodal staging, 41 patients (53.9%) exhibited lymph node metastasis, whereas 35 patients (46.1%) showed no regional lymph node involvement (N0). Distant metastasis (M1) was identified radiologically in 21 patients (27.6%), while 55 patients (72.4%) showed no evidence of distant metastasis (M0) at diagnosis. According to the AJCC 8th edition staging system, 4 patients (5.3%) were stage I, 26 patients (34.2%) were stage II, 25 patients (32.9%) were stage III, and 21 patients (27.6%) were stage IV. Detailed patient characteristics are summarized in Table 1.

Table 1

Table 1. Patient characteristics.

Comparison of clinical and laboratory characteristics by distant metastatic status

Higher T stage and lymph node metastasis were significantly associated with distant metastasis. The proportion of patients with a T4 tumor was significantly higher in the distant metastasis-positive group than in the non-distant metastasis group (52.4% vs. 20.0%, p = 0.006), and lymph node metastasis was also more frequent in patients with distant metastasis (76.2% vs. 45.5%, p = 0.02). Histologic grade showed a trend towards a higher proportion of poorly differentiated tumors in the metastasis group (9.5% vs. 1.8%), but the difference was not statistically significant (p = 0.13). Also, patients with distant metastasis had a significantly higher spleen SUV_max value than those without distant metastasis (3.1 ± 0.4 vs. 2.8 ± 0.4, p = 0.02), whereas no significant difference was observed for primary tumor SUV_max or BM SUV_max.

Neither age, sex, BMI, comorbidities (hypertension, diabetes mellitus, or dyslipidemia), lifestyle factors (smoking or alcohol consumption), nor laboratory parameters (CRP, WBC, CEA, or CA 19-9) were significantly associated with distant metastasis. Detailed clinical and laboratory characteristics are summarized in Table 2. Normality test results and the corresponding test selection for continuous variables are provided in Supplementary Table 1.

Table 2

Table 2. Comparison of clinical and laboratory characteristics of colorectal cancer patients with or without distant metastasis.

The independent association between AmygA and distant metastasis

Representative axial ¹⁸F-FDG PET/CT images demonstrated AmygA in patients with distant metastasis (Figures 2A, B), and quantitatively, AmygA values were significantly higher in patients with distant metastasis (1.17 ± 0.06 vs. 1.08 ± 0.06, p < 0.001; Figure 2E). In contrast, neither amygdala SUV_max (p = 0.95; Figure 2C) nor temporal lobe SUV_mean (p = 0.18; Figure 2D) showed a significant difference. ROC curve analysis identified an optimal AmygA cutoff value of 1.159 for predicting distant metastasis, yielding a sensitivity of 71.4% and a specificity of 89.1% with an AUC of 0.844 (p < 0.001) (Figure 2F). We additionally performed subgroup ROC analyses of AmygA for discriminating distant metastasis, stratified by primary site, tumor stage, and inflammatory status. AmygA demonstrated good predictive performance for distant metastasis across all subgroups, with no significant differences in AUCs among the subgroups. The subgroup AUCs with bootstrap BCa 95% confidence intervals are summarized in Supplementary Table 3 and illustrated in Supplementary Figure 1.

Figure 2

Figure 2. Representative axial ¹⁸F-FDG PET/CT images showing amygdala metabolic activity (AmygA) in patients without (A) and with (B) distant metastasis. Amygdalae are indicated by circular regions of interest. Box plot comparing Amyg SUV_max (C), Temporal SUV_mean (D), and AmygA (E) in patients with or without distant metastasis. Receiver operating characteristic (ROC) curve analysis of AmygA for predicting distant metastasis is shown in (F). The optimal AmygA cutoff value of 1.159 was determined using the maximum Youden index, defined as [sensitivity − (1 − specificity)]. M (-), n = 55; M (+), n = 21. M (-), without distant metastasis; M (+), with distant metastasis; SUV_max, maximum standardized uptake value; Amyg SUV_max, SUV_max of amygdala; SUV_mean, mean standardized uptake value; AUC, area under the curve.

Univariable logistic regression analyses were used to screen candidate variables. Multivariable inference was based on prespecified parsimonious models with bootstrap BCa 95% confidence intervals (Table 3). For transparency, the original full univariable screening and extended multivariable results are provided in Supplementary Table 2. Because T stage and lymph node status represent related measures of tumor burden, they were evaluated in separate prespecified models (Model 1 and Model 2, respectively), while spleen SUV_max was included as a single inflammatory surrogate covariate to limit model complexity given the limited number of metastatic events.

Table 3

Table 3. Prespecified parsimonious multivariable logistic regression models for predicting distant metastasis.

Univariable logistic regression analysis identified several variables significantly associated with distant metastasis in CRC patients, including advanced primary tumor stage (T4 vs. T1–T3; OR = 4.40, 95% CI: 1.49–12.98, p = 0.007), the presence of lymph node metastasis (OR = 5.38, 95% CI: 1.42–20.38, p = 0.005), elevated AmygA (> 1.159; OR = 14.40, 95% CI: 4.27–48.57, p < 0.001), increased spleen SUV_max (OR = 3.92, 95% CI: 1.18–13.08, p = 0.02), and higher serum levels of CEA (per ng/mL; OR = 1.01, 95% CI: 1.00–1.02, p = 0.04) and CA 19-9 (per U/mL; OR = 1.01, 95% CI: 1.00–1.01, p = 0.002) (Supplementary Table 2).

The prespecified parsimonious multivariable logistic regression models incorporating bootstrap BCa 95% confidence intervals showed that AmygA, dichotomized using the ROC-derived cutoff, remained independently associated with distant metastasis and was the only variable that met the Bonferroni-corrected significance threshold (Model 1: OR = 15.28, BCa 95% CI: 1.43–782.68, p < 0.001; Model 2: OR = 14.53, BCa 95% CI: 3.10–22.64, p < 0.001) after adjustment for tumor burden (T stage or lymph node status) and an inflammatory surrogate marker (spleen SUV_max) (Table 3).

Associations between AmygA and systemic inflammatory markers

No significant correlations were found between AmygA and systemic inflammatory markers, including BM SUV_max (r = 0.18, p = 0.13), spleen SUV_max (r = 0.06, p = 0.62), CRP (r = 0.13, p = 0.30), and WBC (r = −0.03, p = 0.81) (Table 4).

Table 4

Table 4. Spearman’s correlation coefficients for the relationship between AmygA and systemic inflammatory markers.

Discussion

This study indicates for the first time that chronic stress, as reflected by elevated AmygA, is independently associated with the presence of distant metastasis in CRC patients. Notably, AmygA retained its independent association with distant metastasis even after adjusting for established prognostic factors, including advanced T stage and regional lymph node involvement. The significantly higher AmygA observed in patients with metastatic disease suggests that chronic stress contributes meaningfully to the progression of metastasis.

Our findings are consistent with prior studies that have reported that elevated AmygA measured by ¹⁸F-FDG PET/CT serves as an imaging-based biomarker of chronic stress burden (18, 19, 22, 23, 35). Moreover, previous studies have reported significant associations between AmygA and adverse oncologic outcomes in malignancies, such as mortality and cancer progression in head and neck and endometrial cancers (18, 19). Our results extend these observations to CRC by indicating that elevated AmygA is associated with CRC progression and metastatic potential.

Published AmygA distributions differ across disease populations. In an osteoporosis cohort, AmygA was 0.81 ± 0.16 in patients with osteoporosis and 0.61 ± 0.13 in those without osteoporosis (22). In endometrial cancer, postmenopausal patients with versus without lymph node metastasis showed AmygA values of 0.93 ± 0.08 versus 0.86 ± 0.06. In contrast, no difference was observed in the premenopausal subgroup, with AmygA values of 0.89 ± 0.07 versus 0.89 ± 0.05 (18). In our CRC cohort, AmygA was 1.17 ± 0.06 in patients with distant metastasis and 1.08 ± 0.06 in those without distant metastasis, which is numerically higher than the values reported in these prior cohorts. However, direct cross-study comparison is not appropriate because AmygA is a ratio metric and may reflect differences in both amygdala uptake and reference-region metabolism, as well as cohort composition and imaging-related factors. Consistent with this heterogeneity, other studies also defined “high” amygdala activity using distribution-based thresholds rather than a single universal numeric cutoff (19, 23). Accordingly, we interpret our cutoff of 1.159 as a cohort-specific threshold intended for interpretability rather than a universal value.

The mechanisms underlying the association between elevated AmygA and distant metastasis in CRC remain incompletely understood. Previous studies have suggested that chronic stress may influence cancer progression by upregulating systemic inflammation (6, 7, 18, 19). Furthermore, the pro-tumorigenic effects of an inflammatory microenvironment in CRC have been well established (36). The following cascade of stress-induced inflammatory upregulation represents a biologically plausible mechanism linking chronic stress to metastatic progression in CRC (6, 7, 37–40). Chronic stress stimulates both the HPA axis and the sympathetic nervous system, resulting in elevated secretion of stress-associated hormones, including norepinephrine and glucocorticoids. These neuroendocrine changes subsequently inhibit cell-mediated immune responses and enhance systemic inflammatory activity. In addition, stress-activated sympathetic axons innervating the bone marrow stimulate hematopoietic stem cell proliferation and mobilize inflammatory monocytes, which release proinflammatory cytokines that impair immune surveillance and promote tumor invasion and distant metastasis. A schematic overview of the proposed pathway is shown in Figure 3.

Figure 3

Figure 3. Proposed schematic linking chronic stress to elevated AmygA and metastatic potential. Chronic stress is reflected by elevated AmygA, an ¹⁸F-FDG PET/CT–derived imaging marker of stress-related neural metabolic activity. Chronic stress activates the HPA (hypothalamic–pituitary–adrenal) axis and the sympathetic nervous system, resulting in increased glucocorticoids and catecholamines such as norepinephrine. These neuroendocrine signals may impair anti-tumor immune surveillance and promote systemic and tumor-associated inflammatory remodeling, thereby facilitating a pro-metastatic environment.

In this study, spleen SUV_max, a well-established surrogate marker of systemic inflammation (41–44), was elevated in CRC patients with distant metastasis and significantly predicted metastatic disease by univariable analysis. To address potential confounding by systemic inflammation, spleen SUV_max was included as an inflammatory surrogate covariate in the prespecified multivariable models, and the association between AmygA and distant metastasis remained significant after this adjustment. Although AmygA was positively correlated with surrogate markers of systemic inflammation, including BM SUV_max, spleen SUV_max, and CRP, these correlations were not significant. Several studies have reported that AmygA is significantly correlated with systemic inflammation (18, 19, 22, 23). In a cardiovascular cohort (n = 293), AmygA showed significant correlations with spleen SUV_max (r = 0.47, p < 0.001), BM SUV_max (r = 0.40, p < 0.001), and CRP (r = 0.83, p = 0.02) (23), and in patients with endometrial cancer (n = 161) and head and neck cancer (n = 240), AmygA was significantly correlated with CRP and BM SUV_max (18, 19). Furthermore, in a health-screening cohort (n = 115), AmygA was significantly correlated with CRP (22). This discrepancy could be attributed to the smaller sample size in our study (n = 76) compared with previous studies (n = 115–293) (18, 19, 22, 23), which may have limited our ability to detect modest associations. Accordingly, our data are insufficient to draw definitive conclusions regarding the association between AmygA and systemic inflammation in CRC.

Non-pharmacologic strategies, such as yoga, exercise, and meditation, and pharmacologic approaches, such as β-adrenergic blockade, have shown promise for mitigating stress-induced biological changes (35, 45–48), and a recent meta-analysis showed that exercise might reduce CRC risk by 13–16% (49). Furthermore, a prospective randomized clinical trial demonstrated that exercise exerts promising effects on survival among stage II and III CRC survivors (50). Interestingly, we previously found that AmygA may reflect the beneficial effects of exercise on stress-related neurobiology (35). Thus, given its ability to assess stress-related neural activity noninvasively, AmygA may have potential as an imaging-based biomarker of the efficacy of stress-reduction interventions in CRC patients.

This study has several limitations. The modest sample size and single-center, retrospective design may limit the generalizability of the findings, which should be confirmed by larger, multicenter prospective studies. Although an association between elevated AmygA and distant metastasis in CRC was observed, causality could not be inferred because of the study’s observational nature. Moreover, given the limited number of metastatic events, effect estimates showed wide confidence intervals, warranting cautious interpretation and confirmation in larger cohorts. Additionally, one patient exhibited an extremely low BMI consistent with cachexia and markedly poor general status. Such extreme outliers may disproportionately influence estimates in small cohorts. Residual confounding is possible because medication use that may influence stress-related neural activity or tumor biology (e.g., β-blockers, systemic steroids, or antidepressants) was not systematically captured and direct HPA-axis or sympathetic nervous system biomarkers were not assessed. In addition, the limited spatial resolution of current ¹⁸F-FDG PET/CT systems hinders detailed assessment of other stress-related brain regions, such as the hippocampus and prefrontal cortex.

Conclusion

This study shows that chronic stress-related metabolic activity in the amygdala is closely linked to the development of distant metastases in patients with CRC. While the underlying mechanisms remain unclear, these findings offer insight into the connection between stress-related neural activity and cancer progression and highlight the potential of AmygA as a novel imaging-based biomarker of CRC aggressiveness.

Data availability statement

The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author/s.

Ethics statement

The studies involving humans were approved by the Institutional Review Board of Korea University Anam Hospital. The studies were conducted in accordance with the local legislation and institutional requirements. The ethics committee/institutional review board waived the requirement of written informed consent for participation from the participants or the participants’ legal guardians/next of kin because given the retrospective nature of the study, the board waived the requirement for informed consent.

Author contributions

HK: Writing – original draft, Formal analysis, Investigation, Data curation, Conceptualization. SH: Writing – original draft, Conceptualization, Investigation, Formal analysis, Data curation. CJ: Writing – original draft, Data curation, Visualization, Methodology. SK: Conceptualization, Validation, Supervision, Writing – review & editing. KP: Writing – review & editing, Funding acquisition, Supervision, Writing – original draft, Conceptualization, Validation, Formal analysis.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This research was supported by grants from Korea University Anam Hospital, Seoul, Republic of Korea (K2424151, K2515161), and Korea University (K2508361, K2509651). This research was also supported by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant number: RS-2025-25458758), and by Young Medical Scientist Research Grant through the Daewoong Foundation (DFY2509P).

Acknowledgments

Figure 3 was created using Biorender.com by the author HJK (https://BioRender.com/8d0rnly).

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fendo.2026.1747732/full#supplementary-material

Supplementary Figure 1 | Subgroup receiver-operating characteristic (ROC) curves of AmygA for discriminating distant metastasis, stratified by primary site, tumor stage, and inflammatory status. Receiver-operating characteristic (ROC) curves of AmygA for discriminating distant metastasis stratified by (A) primary tumor site (colon vs. rectum), (B) tumor stage (T1–3 vs. T4), and (C) inflammatory status (low vs. high spleen SUV_max). Spleen SUV_max was dichotomized at 2.9 (cohort mean).

Supplementary Table 1 | Assessment of normality for continuous variables. BMI, body mass index; CRP, C-reactive protein; WBC, white blood cell count; CEA, carcinoembryonic antigen; CA 19-9, carbohydrate antigen 19-9; BM, bone marrow; SUV_max, maximum standardized uptake value; Amyg, amygdala.

Supplementary Table 2 | Univariable and multivariable analysis results for the prediction of distant metastasis OR, odds ratio; CI, confidence interval; BMI, body mass index; HTN, hypertension; DM, diabetes mellitus; CRP, C-reactive protein; WBC, white blood cell count; T, tumor; Amyg, amygdala; AmygA, amygdala metabolic activity; SUV_max, maximum standardized uptake value; BM, bone marrow; CEA, carcinoembryonic antigen; CA 19-9, carbohydrate antigen 19-9. ORs for continuous variables are expressed per 1-unit increase. *Statistically significant. **Statistically significant after Bonferroni correction for 6 comparisons (significance threshold: p < 0.0083).

Supplementary Table 3 | Subgroup analyses of the discriminatory performance of AmygA for distant metastasis, stratified by primary site, tumor stage, and inflammatory status AmygA, amygdala metabolic activity; AUC, area under the curve; BCa 95% CI, bias-corrected and accelerated 95% confidence interval; T, tumor; SUV_max, maximum standardized uptake value. Comparison p values indicate between-subgroup differences in AUCs. Spleen SUV_max was categorized into high and low by 2.9, which was the mean of spleen SUV_max in this study.

References

1. Siegel RL, Wagle NS, Cercek A, Smith RA, and Jemal A. Colorectal cancer statistics, 2023. CA Cancer J Clin. (2023) 73:233–54. doi: 10.3322/caac.21772

PubMed Abstract | Crossref Full Text | Google Scholar

2. Roshandel G, Ghasemi-Kebria F, and Malekzadeh R. Colorectal cancer: epidemiology, risk factors, and prevention. Cancers (Basel). (2024) 16:1530. doi: 10.3390/cancers16081530

PubMed Abstract | Crossref Full Text | Google Scholar

3. Dekker E, Tanis PJ, Vleugels JLA, Kasi PM, and Wallace MB. Colorectal cancer. Lancet. (2019) 394:1467–80. doi: 10.1016/s0140-6736(19)32319-0

PubMed Abstract | Crossref Full Text | Google Scholar

4. Koroukian SM, Booker BD, Vu L, Schumacher FR, Rose J, Cooper GS, et al. Receipt of targeted therapy and survival outcomes in patients with metastatic colorectal cancer. JAMA Netw Open. (2023) 6:e2250030. doi: 10.1001/jamanetworkopen.2022.50030

PubMed Abstract | Crossref Full Text | Google Scholar

5. Ilyas MIM. Epidemiology of stage iv colorectal cancer: trends in the incidence, prevalence, age distribution, and impact on life span. Clin Colon Rectal Surg. (2024) 37:57–61. doi: 10.1055/s-0043-1761447

PubMed Abstract | Crossref Full Text | Google Scholar

6. Cui B, Peng F, Lu J, He B, Su Q, Luo H, et al. Cancer and stress: nextgen strategies. Brain Behav Immun. (2021) 93:368–83. doi: 10.1016/j.bbi.2020.11.005

PubMed Abstract | Crossref Full Text | Google Scholar

7. Baritaki S, de Bree E, Chatzaki E, and Pothoulakis C. Chronic stress, inflammation, and colon cancer: A crh system-driven molecular crosstalk. J Clin Med. (2019) 8:1669. doi: 10.3390/jcm8101669

PubMed Abstract | Crossref Full Text | Google Scholar

8. Bondar T and Medzhitov R. The origins of tumor-promoting inflammation. Cancer Cell. (2013) 24:143–4. doi: 10.1016/j.ccr.2013.07.016

PubMed Abstract | Crossref Full Text | Google Scholar

9. He XY, Gao Y, Ng D, Michalopoulou E, George S, Adrover JM, et al. Chronic stress increases metastasis via neutrophil-mediated changes to the microenvironment. Cancer Cell. (2024) 42:474–86.e12. doi: 10.1016/j.ccell.2024.01.013

PubMed Abstract | Crossref Full Text | Google Scholar

10. Zheng Y, Wang N, Wang S, Zhang J, Yang B, and Wang Z. Chronic psychological stress promotes breast cancer pre-metastatic niche formation by mobilizing splenic mdscs via tam/cxcl1 signaling. J Exp Clin Cancer Res. (2023) 42:129. doi: 10.1186/s13046-023-02696-z

PubMed Abstract | Crossref Full Text | Google Scholar

11. Zhi X, Li B, Li Z, Zhang J, Yu J, Zhang L, et al. Adrenergic modulation of ampk−Dependent autophagy by chronic stress enhances cell proliferation and survival in gastric cancer. Int J Oncol. (2019) 54:1625–38. doi: 10.3892/ijo.2019.4753

PubMed Abstract | Crossref Full Text | Google Scholar

12. Zhang X, Zhang Y, He Z, Yin K, Li B, Zhang L, et al. Chronic stress promotes gastric cancer progression and metastasis: an essential role for adrb2. Cell Death Dis. (2019) 10:788. doi: 10.1038/s41419-019-2030-2

PubMed Abstract | Crossref Full Text | Google Scholar

13. Jang HJ, Boo HJ, Lee HJ, Min HY, and Lee HY. Chronic stress facilitates lung tumorigenesis by promoting exocytosis of igf2 in lung epithelial cells. Cancer Res. (2016) 76:6607–19. doi: 10.1158/0008-5472.Can-16-0990

PubMed Abstract | Crossref Full Text | Google Scholar

14. Chen H, Liu D, Guo L, Cheng X, Guo N, and Shi M. Chronic psychological stress promotes lung metastatic colonization of circulating breast cancer cells by decorating a pre-metastatic niche through activating B-adrenergic signaling. J Pathol. (2018) 244:49–60. doi: 10.1002/path.4988

PubMed Abstract | Crossref Full Text | Google Scholar

15. Androulidaki A, Dermitzaki E, Venihaki M, Karagianni E, Rassouli O, Andreakou E, et al. Corticotropin releasing factor promotes breast cancer cell motility and invasiveness. Mol Cancer. (2009) 8:30. doi: 10.1186/1476-4598-8-30

PubMed Abstract | Crossref Full Text | Google Scholar

16. Chang A, Le CP, Walker AK, Creed SJ, Pon CK, Albold S, et al. B2-adrenoceptors on tumor cells play a critical role in stress-enhanced metastasis in a mouse model of breast cancer. Brain Behav Immun. (2016) 57:106–15. doi: 10.1016/j.bbi.2016.06.011

PubMed Abstract | Crossref Full Text | Google Scholar

17. Cui B, Luo Y, Tian P, Peng F, Lu J, Yang Y, et al. Stress-induced epinephrine enhances lactate dehydrogenase a and promotes breast cancer stem-like cells. J Clin Invest. (2019) 129:1030–46. doi: 10.1172/jci121685

PubMed Abstract | Crossref Full Text | Google Scholar

18. Ryu KJ, Kim S, Joung C, Lee S, Park H, Song JY, et al. Association between chronic stress-related amygdala metabolic activity and lymph node metastasis in endometrial cancer. Sci Rep. (2024) 14:28418. doi: 10.1038/s41598-024-79987-8

PubMed Abstract | Crossref Full Text | Google Scholar

19. Hassan MZO, Tawakol A, Wang Y, Alvi RM, Awadalla M, Jones-O'Connor M, et al. Amygdalar activity measured using fdg-pet/ct at head and neck cancer staging independently predicts survival. PloS One. (2023) 18:e0279235. doi: 10.1371/journal.pone.0279235

PubMed Abstract | Crossref Full Text | Google Scholar

20. Saul AN, Oberyszyn TM, Daugherty C, Kusewitt D, Jones S, Jewell S, et al. Chronic stress and susceptibility to skin cancer. J Natl Cancer Inst. (2005) 97:1760–7. doi: 10.1093/jnci/dji401

PubMed Abstract | Crossref Full Text | Google Scholar

21. Parker J, Klein SL, McClintock MK, Morison WL, Ye X, Conti CJ, et al. Chronic stress accelerates ultraviolet-induced cutaneous carcinogenesis. J Am Acad Dermatol. (2004) 51:919–22. doi: 10.1016/j.jaad.2004.08.042

PubMed Abstract | Crossref Full Text | Google Scholar

22. Pahk K, Kwon HW, Joung C, and Kim S. Stress-related amygdala metabolic activity is associated with low bone mineral density in postmenopausal women: A pilot (18)F-fdg pet/ct study. Front Endocrinol (Lausanne). (2021) 12:719265. doi: 10.3389/fendo.2021.719265

PubMed Abstract | Crossref Full Text | Google Scholar

23. Tawakol A, Ishai A, Takx RA, Figueroa AL, Ali A, Kaiser Y, et al. Relation between resting amygdalar activity and cardiovascular events: A longitudinal and cohort study. Lancet. (2017) 389:834–45. doi: 10.1016/s0140-6736(16)31714-7

PubMed Abstract | Crossref Full Text | Google Scholar

24. Kikuchi N, Nishiyama T, Sawada T, Wang C, Lin Y, Watanabe Y, et al. Perceived stress and colorectal cancer incidence: the Japan collaborative cohort study. Sci Rep. (2017) 7:40363. doi: 10.1038/srep40363

PubMed Abstract | Crossref Full Text | Google Scholar

25. Guan Y, Yao W, Yu H, Feng Y, Zhao Y, Zhan X, et al. Chronic stress promotes colorectal cancer progression by enhancing glycolysis through B2-ar/creb1 signal pathway. Int J Biol Sci. (2023) 19:2006–19. doi: 10.7150/ijbs.79583

PubMed Abstract | Crossref Full Text | Google Scholar

26. McCollum SE and Shah YM. Stressing out cancer: chronic stress induces dysbiosis and enhances colon cancer growth. Cancer Res. (2024) 84:645–7. doi: 10.1158/0008-5472.Can-23-3871

PubMed Abstract | Crossref Full Text | Google Scholar

27. Weiser MR. Ajcc 8th edition: colorectal cancer. Ann Surg Oncol. (2018) 25:1454–5. doi: 10.1245/s10434-018-6462-1

PubMed Abstract | Crossref Full Text | Google Scholar

28. Okagawa Y, Seto K, Yoshida K, Hanada K, Hirokawa S, Tomita Y, et al. Clinicopathological features of early-onset colorectal cancer in Japanese patients: A single-center retrospective study. BMC Gastroenterol. (2025) 25:156. doi: 10.1186/s12876-025-03725-1

PubMed Abstract | Crossref Full Text | Google Scholar

29. Burt JR, Qaqish N, Stoddard G, Jridi A, Anderson PS, Woods L, et al. Non-small cell lung cancer in ever-smokers vs never-smokers. BMC Med. (2025) 23:3. doi: 10.1186/s12916-024-03844-8

PubMed Abstract | Crossref Full Text | Google Scholar

30. Zeinomar N, Qin B, Amin S, Lin Y, Xu B, Chanumolu D, et al. Association of cigarette smoking and alcohol consumption with subsequent mortality among black breast cancer survivors in new Jersey. JAMA Netw Open. (2023) 6:e2252371. doi: 10.1001/jamanetworkopen.2022.52371

PubMed Abstract | Crossref Full Text | Google Scholar

31. Britz-Cunningham SH, Millstine JW, and Gerbaudo VH. Improved discrimination of benign and Malignant lesions on fdg pet/ct, using comparative activity ratios to brain, basal ganglia, or cerebellum. Clin Nucl Med. (2008) 33:681–7. doi: 10.1097/RLU.0b013e318184b435

PubMed Abstract | Crossref Full Text | Google Scholar

32. Ishibashi M, Norikane T, Yamamoto Y, Imajo M, Takami Y, Mitamura K, et al. Correlation of bone marrow 2-deoxy-2-[ 18 F]Fluoro-D-glucose uptake with systemic inflammation in patients with newly diagnosed endometrial cancer. Nucl Med Commun. (2022) 43:916–21. doi: 10.1097/mnm.0000000000001583

PubMed Abstract | Crossref Full Text | Google Scholar

33. Devesa A, Lobo-González M, Martínez-Milla J, Oliva B, García-Lunar I, Mastrangelo A, et al. Bone marrow activation in response to metabolic syndrome and early atherosclerosis. Eur Heart J. (2022) 43:1809–28. doi: 10.1093/eurheartj/ehac102

PubMed Abstract | Crossref Full Text | Google Scholar

34. Pahk K, Kim EJ, Joung C, Seo HS, and Kim S. Association of glucose uptake of visceral fat and acute myocardial infarction: A pilot (18)F-fdg pet/ct study. Cardiovasc Diabetol. (2020) 19:145. doi: 10.1186/s12933-020-01115-3

PubMed Abstract | Crossref Full Text | Google Scholar

35. Pahk K, Joung C, Kwon HW, and Kim S. Chronic physical exercise alleviates stress-associated amygdala metabolic activity in obese women: A prospective serial (18)F-fdg pet/ct study. Front Endocrinol (Lausanne). (2022) 13:1046838. doi: 10.3389/fendo.2022.1046838

PubMed Abstract | Crossref Full Text | Google Scholar

36. Schmitt M and Greten FR. The inflammatory pathogenesis of colorectal cancer. Nat Rev Immunol. (2021) 21:653–67. doi: 10.1038/s41577-021-00534-x

PubMed Abstract | Crossref Full Text | Google Scholar

37. Dhabhar FS, McEwen BS, and Spencer RL. Adaptation to prolonged or repeated stress–comparison between rat strains showing intrinsic differences in reactivity to acute stress. Neuroendocrinology. (1997) 65:360–8. doi: 10.1159/000127196

PubMed Abstract | Crossref Full Text | Google Scholar

38. Chrousos GP. The hypothalamic-pituitary-adrenal axis and immune-mediated inflammation. N Engl J Med. (1995) 332:1351–62. doi: 10.1056/nejm199505183322008

PubMed Abstract | Crossref Full Text | Google Scholar

39. Heidt T, Sager HB, Courties G, Dutta P, Iwamoto Y, Zaltsman A, et al. Chronic variable stress activates hematopoietic stem cells. Nat Med. (2014) 20:754–8. doi: 10.1038/nm.3589

PubMed Abstract | Crossref Full Text | Google Scholar

40. Yang S, Li Y, Zhang Y, and Wang Y. Impact of chronic stress on intestinal mucosal immunity in colorectal cancer progression. Cytokine Growth Factor Rev. (2024) 80:24–36. doi: 10.1016/j.cytogfr.2024.10.007

PubMed Abstract | Crossref Full Text | Google Scholar

41. Emami H, Singh P, MacNabb M, Vucic E, Lavender Z, Rudd JH, et al. Splenic metabolic activity predicts risk of future cardiovascular events: demonstration of a cardiosplenic axis in humans. JACC Cardiovasc Imaging. (2015) 8:121–30. doi: 10.1016/j.jcmg.2014.10.009

PubMed Abstract | Crossref Full Text | Google Scholar

42. Pijl JP, Kwee TC, Slart R, Yakar D, Wouthuyzen-Bakker M, and Glaudemans A. Clinical implications of increased uptake in bone marrow and spleen on fdg-pet in patients with bacteremia. Eur J Nucl Med Mol Imaging. (2021) 48:1467–77. doi: 10.1007/s00259-020-05071-8

PubMed Abstract | Crossref Full Text | Google Scholar

43. Patel NH, Osborne MT, Teague H, Parel P, Svirydava M, Sorokin AV, et al. Heightened splenic and bone marrow uptake of (18)F-fdg pet/ct is associated with systemic inflammation and subclinical atherosclerosis by ccta in psoriasis: an observational study. Atherosclerosis. (2021) 339:20–6. doi: 10.1016/j.atherosclerosis.2021.11.008

PubMed Abstract | Crossref Full Text | Google Scholar

44. Pak K, Kim SJ, Kim IJ, Kim DU, Kim K, and Kim H. Impact of cytokines on diffuse splenic 18f-fluorodeoxyglucose uptake during positron emission tomography/computed tomography. Nucl Med Commun. (2013) 34:64–70. doi: 10.1097/MNM.0b013e3283595cac

PubMed Abstract | Crossref Full Text | Google Scholar

45. Gosain R, Gage-Bouchard E, Ambrosone C, Repasky E, and Gandhi S. Stress reduction strategies in breast cancer: review of pharmacologic and non-pharmacologic based strategies. Semin Immunopathol. (2020) 42:719–34. doi: 10.1007/s00281-020-00815-y

PubMed Abstract | Crossref Full Text | Google Scholar

46. Pace TW, Negi LT, Adame DD, Cole SP, Sivilli TI, Brown TD, et al. Effect of compassion meditation on neuroendocrine, innate immune and behavioral responses to psychosocial stress. Psychoneuroendocrinology. (2009) 34:87–98. doi: 10.1016/j.psyneuen.2008.08.011

PubMed Abstract | Crossref Full Text | Google Scholar

47. Carlson LE, Speca M, Patel KD, and Goodey E. Mindfulness-based stress reduction in relation to quality of life, mood, symptoms of stress and levels of cortisol, dehydroepiandrosterone sulfate (Dheas) and melatonin in breast and prostate cancer outpatients. Psychoneuroendocrinology. (2004) 29:448–74. doi: 10.1016/s0306-4530(03)00054-4

PubMed Abstract | Crossref Full Text | Google Scholar

48. Rosenkranz MA, Davidson RJ, Maccoon DG, Sheridan JF, Kalin NH, and Lutz A. A comparison of mindfulness-based stress reduction and an active control in modulation of neurogenic inflammation. Brain Behav Immun. (2013) 27:174–84. doi: 10.1016/j.bbi.2012.10.013

PubMed Abstract | Crossref Full Text | Google Scholar

49. Moore SC, Lee IM, Weiderpass E, Campbell PT, Sampson JN, Kitahara CM, et al. Association of leisure-time physical activity with risk of 26 types of cancer in 1.44 million adults. JAMA Intern Med. (2016) 176:816–25. doi: 10.1001/jamainternmed.2016.1548

PubMed Abstract | Crossref Full Text | Google Scholar

50. Courneya KS, Vardy JL, O'Callaghan CJ, Friedenreich CM, Campbell KL, Prapavessis H, et al. Effects of a structured exercise program on physical activity and fitness in colon cancer survivors: one year feasibility results from the challenge trial. Cancer Epidemiol Biomarkers Prev. (2016) 25:969–77. doi: 10.1158/1055-9965.Epi-15-1267

PubMed Abstract | Crossref Full Text | Google Scholar