Cancer remains one of the leading global health challenges and the second major cause of mortality worldwide, responsible for approximately 9.6 million deaths annually-nearly one in every six deaths. It encompasses a broad spectrum of diseases characterized by uncontrolled cell growth, invasion of surrounding tissues, and the potential to metastasize to distant organs. Among the most common cancers are those of the lung, breast, colon, prostate, stomach, and liver, each contributing significantly to global morbidity and mortality. Understanding tumor metabolism and its systemic effects has become a cornerstone in both oncologic diagnostics and prognostication.[1] The introduction of positron emission tomography (PET) using fluorine-18-labeled fluoro-2-deoxy-D-glucose (18F-FDG) has revolutionized oncologic imaging. PET/CT combines the anatomical precision of computed tomography with the functional insights of PET, enabling clinicians to evaluate both structural and metabolic aspects of malignancies in a single session. The tracer 18F-FDG, a glucose analogue, is preferentially taken up by metabolically active cells through glucose transporters (mainly GLUT-1), phosphorylated by hexokinase, and subsequently trapped intracellularly as FDG-6-phosphate. This accumulation correlates with glycolytic activity-making FDG-PET/CT an invaluable tool for staging, restaging, treatment response assessment, and prognostication of cancers.[2] The human brain, despite constituting only about 2% of body mass, consumes nearly 20% of the body’s glucose. Therefore, under physiological conditions, it demonstrates one of the highest levels of FDG uptake on PET imaging. Brain glucose metabolism is regulated through a finely tuned neurovascular coupling and involves a complex interaction of astrocytes, neurons, and endothelial cells. However, reduced brain FDG uptake-even in the absence of overt brain pathology-has been observed in various systemic conditions such as hyperglycemia, anemia, renal dysfunction, and neurodegenerative diseases like Alzheimer’s disease. In oncology, a distinct but intriguing observation has emerged: concurrent low brain FDG uptake with high tumor uptake, sometimes referred to as the “cold brain phenomenon.” This finding has raised hypotheses regarding the redistribution of glucose and radiotracer from the brain to hypermetabolic tumors, reflecting an overall metabolic reprogramming associated with advanced or aggressive disease states.[3] et al. (20)[4] first reported this phenomenon in Diffuse Large B-Cell Lymphoma (DLBCL), showing that patients with reduced brain FDG uptake had more advanced disease (stage III/IV), higher International Prognostic Index (IPI) scores, lower treatment response rates, and shorter progression-free survival. While this association has been well characterized in hematologic malignancies, its relevance across solid tumors remains largely unexplored. Given that FDG-PET/CT is routinely performed for initial staging of a wide array of malignancies, analyzing incidental findings such as reduced cerebral metabolism could provide valuable prognostic insights without additional cost, radiation, or patient burden. From a pathophysiological perspective, the inverse relationship between cerebral and tumoral FDG uptake may reflect metabolic competition for glucose between the tumor and other vital organs. Malignant cells, through the “Warburg effect,” demonstrate high glycolytic flux even under normoxic conditions, diverting systemic glucose utilization toward tumor metabolism. This metabolic hijacking may result in diminished cerebral uptake visible on PET imaging. Additionally, systemic inflammation, cytokine release, cachexia, and altered neuroendocrine signaling may further suppress neuronal glucose consumption. Consequently, reduced brain FDG uptake could serve as an imaging biomarker of systemic metabolic stress, correlating with tumor aggressiveness and poor outcomes.[5] Aim To assess the prognostic role of concurrent reduced brain FDG uptake in comparison with primary tumor uptake on initial 18F-FDG PET/CT in various malignancies. Objectives 1. To determine the prognostic value of decreased cerebral FDG uptake in cancer patients with high primary tumor metabolism at baseline PET/CT. 2. To correlate the extent of reduced brain FDG uptake with overall survival and disease progression. 3. To evaluate the relationship between brain FDG uptake ratios (primary-to-brain SUVmax) and quantitative metabolic parameters (MTV, TLG).

Source of Data The study was conducted on patients referred to the Department of Nuclear Medicine and PET/CT, Seth G.S. Medical College and KEM Hospital, Mumbai, India, for baseline staging of histologically confirmed malignancies. Study Design A prospective observational study was carried out over a 12-month period, analyzing patients undergoing routine 18F-FDG PET/CT scans who incidentally demonstrated reduced brain uptake concurrent with high tumoral FDG uptake. Study Duration From July 2018 to June 2019. Sample Size A total of 33 patients were included. Given the novelty of this evaluation in the Indian population and the absence of prior regional data, sample size was determined pragmatically based on daily PET/CT case volume and inclusion feasibility. Inclusion Criteria • Biopsy-proven malignancies referred for initial staging with 18F-FDG PET/CT. • High FDG uptake in primary tumor with concurrent visually reduced brain uptake on MIP images. • No prior oncologic treatment before imaging. • Patients registered at the study institution who provided informed consent. Exclusion Criteria • Primary brain tumors or brain metastases. • Prior chemotherapy, radiotherapy, or surgery. • Hyperglycemia > 180 mg/dL, severe anemia (Hb < 8 g/dL), or renal failure (Sr. Creatinine > 2 mg/dL). • Neurological disorders (dementia, neurodegenerative disease). • Pregnancy or lactation. • Inability to tolerate supine position or claustrophobia. Procedure and Methodology Each participant provided written informed consent after ethics committee approval. Patients fasted for at least six hours, and blood glucose was confirmed below 180 mg/dL. Following intravenous administration of 18F-FDG at 3.7 MBq/kg, imaging was performed after a 60-minute uptake period using a Discovery 710 PET/CT scanner (GE Healthcare, USA). Low-dose CT (120 kVp, 30–180 mAs) was used for attenuation correction and anatomical localization, followed by 3D PET emission scans from skull vertex to mid-thigh (2 minutes per bed position). Images were reconstructed with iterative algorithms and interpreted on GE Advantage Workstation using PET VCAR software. The following parameters were measured: • SUVmax and SUVmean for both primary tumor and brain. • Metabolic Tumor Volume (MTV): volume of tumor with FDG uptake > 40% of SUVmax. • Total Lesion Glycolysis (TLG): MTV × SUVmean. • Whole-Body MTV (MTVWB) and Whole-Body TLG (TLGWB) were computed as sums of all lesions. Patients were categorized into groups based on brain SUVmax values: • Group A: < 5 • Group B: 5–7 • Group C: > 7 All scans were reviewed independently by two nuclear medicine physicians; consensus resolved discrepancies. Patients were followed for at least 3 months or until death, whichever occurred earlier. Sample Processing and Data Collection No additional blood sampling was performed specifically for this study. Routine pre-PET investigations (Hb, serum creatinine, FBS) were utilized. Demographic, clinical, and imaging data-including SUVmax, MTV, TLG, and survival outcomes-were recorded in a standardized proforma. Statistical Methods Data were analyzed using SPSS v28.0 and Microsoft Excel 2019. Quantitative variables were expressed as mean ± SD. Group comparisons employed t-tests, ANOVA, and Fisher’s exact test where applicable. Survival outcomes were evaluated using Kaplan–Meier survival curves, with intergroup differences assessed by the log-rank (Mantel-Cox) test. A p-value < 0.05 was considered statistically significant. Radiation Considerations Each standard PET/CT scan delivered an effective dose of approximately 9 mSv for a 370 MBq FDG administration, consistent with diagnostic safety standards.

Table 1: Baseline PET metrics (overall, n=33): primary vs brain uptake Metric Mean (SD) 95% CI for mean Test of significance SUVmax – Primary tumor 19.59 (9.15) 16.35 to 22.84 vs Brain, paired t-test p=0.0238 SUVmax – Brain 7.90 (4.11) 6.44 to 9.36 - SUVmax Primary/Brain ratio 2.84 (1.46) 2.32 to 3.36 - Whole-body MTV (ml) 653.41 (359.36) 525.80 to 781.02 - Whole-body TLG 2855.73 (2921.28) 1818.33 to 3893.13 - Table 1. Baseline PET Metrics (n = 33): Primary vs. Brain Uptake The baseline quantitative PET parameters demonstrated a clear metabolic disparity between the primary tumor and the brain. The mean SUVmax of the primary tumor was markedly higher (19.59 ± 9.15) than that of the brain (7.90 ± 4.11), and this difference reached statistical significance (paired t-test, p = 0.0238). The 95% confidence intervals for these means (16.35–22.84 vs 6.44–9.36) further confirmed that the difference was unlikely to be due to random variation. The mean primary-to-brain SUVmax ratio was 2.84 ± 1.46, reflecting the higher metabolic activity of malignant tissue compared to the cerebral parenchyma. The whole-body metabolic tumor volume (MTV) averaged 653.41 ± 359.36 mL, while the total lesion glycolysis (TLG) was 2855.73 ± 2921.28. These parameters, representing the volumetric and metabolic burden of disease respectively, showed wide confidence intervals (MTV 95% CI 525.8–781.0; TLG 95% CI 1818.3–3893.1), consistent with heterogeneity in tumor type, size, and stage across the cohort. Table 2: Prognostic value of decreased brain uptake (grouped by brain SUVmax) Grouping (as per thesis):Group A: Brain SUVmax <5 (n=6) • Group B: 5–7 (n=12) • Group C: >7 (n=15). Outcome Group A (<5) Group B (5–7) Group C (>7) Overall Deaths, n (%) 6 (100.0) 10 (83.3) 5 (33.3) 21 (63.6) Survivors, n (%) 0 (0) 2 (16.7) 10 (66.7) 12 (36.4) Mean OS (days) ± SE 60.7 ± 24.7 76.0 ± 18.6 124.6 ± 48.6 83.2 ± 16.1 95% CI for mean OS (days) 12.2 to 109.1 39.5 to 112.5 29.4 to 219.8 51.6 to 114.7 Survival comparison across groups: Log-rank (Mantel-Cox) χ²=3.312, df=2, p=0.191 (as reported). The mortality pattern exhibited a strong gradient with brain uptake. Group A showed 100% mortality, Group B 83.3%, and Group C 33.3%, giving an overall death rate of 63.6%. Conversely, survival improved progressively with higher cerebral SUV: 0%, 16.7%, and 66.7% across Groups A, B, and C, respectively. The mean overall survival (OS) increased in parallel-from 60.7 days in Group A to 124.6 days in Group C-with standard errors reflecting wider variability in the higher-uptake group. Although the log-rank (Mantel–Cox) test for survival comparison yielded χ² = 3.31, df = 2, p = 0.191, indicating that survival differences did not reach statistical significance at the 0.05 level, the numerical trend clearly demonstrated poorer prognosis with decreasing brain FDG uptake. The 95% confidence intervals for mean OS (12.2–109.1 days for Group A vs 29.4–219.8 days for Group C) suggested clinically relevant though statistically nonsignificant separation, likely limited by small subgroup sizes. Table 3: Extent of reduced brain uptake vs overall survival & disease progression Measure Group A (<5) Group B (5–7) Group C (>7) Comparison Brain SUVmax, mean (SD) 3.81 (0.89) 5.69 (0.51) 11.30 (3.79) One-way ANOVA for brain SUV across groups: p=0.0384 (reported). Mean OS (days) [95% CI]* 60.7 [12.2–109.1] 76.0 [39.5–112.5] 124.6 [29.4–219.8] Log-rank p=0.191 (reported). Deaths n/N (%) 6/6 (100%) 10/12 (83.3%) 5/15 (33.3%) Exploratory χ² (3×2) from counts = 11.39; df=2; p=0.00336 Further evaluation of the extent of cerebral hypometabolism revealed a stepwise increase in mean brain SUVmax from 3.81 ± 0.89 in Group A to 11.30 ± 3.79 in Group C. One-way ANOVA (p = 0.0384) confirmed significant inter-group differences, indicating true variation in the degree of brain metabolic suppression. Corresponding survival outcomes mirrored this gradient: mean OS was 60.7 days in Group A, 76.0 days in Group B, and 124.6 days in Group C. While the global survival comparison using the log-rank test (p = 0.191) remained nonsignificant, an exploratory chi-square analysis of mortality counts yielded χ² = 11.39, df = 2, p = 0.003, suggesting that lower brain SUVmax was significantly associated with higher death rates. These findings imply that patients with the most pronounced cerebral hypometabolism were at substantially greater risk of early mortality and disease progression, supporting the concept that reduced brain FDG uptake reflects aggressive tumor biology and systemic metabolic stress. Table 4: Brain uptake ratio vs metabolic burden (MTV, TLG) (a) Overall descriptors (n=33): (from Table 3) SUVmax Primary/Brain ratio: 2.84 ± 1.46 (95% CI 2.32–3.36) Whole-body MTV: 653.41 ± 359.36 ml (95% CI 525.80–781.02) Whole-body TLG: 2855.73 ± 2921.28 (95% CI 1818.33–3893.13) (b) MTV/TLG by brain-SUV groups (proxy for “extent of reduction”): (from Table 4) Group Mean MTV (SD), ml 95% CI Mean TLG (SD) 95% CI A (SUV<5), n=6 773.2 (537.86) 208.7 to 1337.7 6794.2 (3420.76) 3203.8 to 10,384.6 B (5–7), n=12 237.2 (261.23) 71.2 to 403.2 1409.1 (1629.16) 374.0 to 2444.2 C (>7), n=15 217.9 (182.34) 116.9 to 318.9 2437.6 (2147.63) 1248.2 to 3627.0 The relationship between global tumor burden and brain metabolic suppression was explored through SUV ratios and volumetric parameters. Across the total cohort (n = 33), the mean primary-to-brain SUVmax ratio was 2.84 ± 1.46 (95% CI 2.32–3.36), while mean MTV and TLG were 653.41 ± 359.36 mL and 2855.73 ± 2921.28, respectively. When analyzed by brain-SUV categories, patients in Group A (most reduced brain uptake) had the highest mean MTV (773.2 mL) and TLG (6794.2), compared to Group B (MTV 237.2 mL; TLG 1409.1) and Group C (MTV 217.9 mL; TLG 2437.6). The marked elevation of MTV and TLG in Group A indicates a direct association between low brain FDG activity and high systemic metabolic tumor burden. Although formal significance testing for MTV/TLG differences was not reported, the magnitude of disparity and non-overlapping confidence intervals (e.g., TLG 95% CI 3203.8–10,384.6 for Group A vs 374.0–2444.2 for Group B) suggest a robust inverse correlation. These results reinforce the hypothesis that greater tumor glycolytic activity is accompanied by proportionally reduced cerebral glucose utilization, consistent with the metabolic “steal” or diversion phenomenon observed in aggressive malignancies.

Baseline metabolism (Table 1). Cohort shows a strong primary-vs-brain metabolic split-SUVmax 19.59 ± 9.15 in tumors vs 7.90 ± 4.11 in brain, significant on paired t-test (p = 0.0238)-with substantial global burden (mean MTV ≈ 653 mL; TLG ≈ 2856). These numbers mirror the principle that malignant glycolysis outpaces normal cerebral glucose use, providing the quantitative substrate for a “cold brain” phenotype. Physiologically, high cerebral glucose demand is the norm; reduced brain FDG can occur from metabolic competition and systemic factors, but here it coexists with high tumoral avidity, consistent with a diversion/“steal” hypothesis (neuro-metabolic coupling vs. tumor Warburg drive). Kalantari F et al. (2023)[6] Prognostic gradient by brain SUVmax (Table 2). Stratifying by brain SUVmax (<5, 5–7, >7) yields a stepwise mortality gradient (100%, 83.3%, 33.3%) and longer mean OS with higher brain uptake (60.7 → 124.6 days). Although the log-rank across groups is not statistically significant (χ² = 3.31; p = 0.191)-likely a function of small strata (n=6/12/15) and wide CIs-the directionality is clinically coherent and recapitulates lymphoma data from Choi EK et al. (2021)[7], who found that “cold brain” at baseline predicted higher-risk disease and worse R-CHOP responses with shorter PFS (trend) in DLBCL. Extent of brain reduction, survival, and events (Table 3). Group means for brain SUVmax (3.81 vs 5.69 vs 11.30) differ significantly (ANOVA p = 0.0384), and mortality proportions scale inversely with brain SUV; exploratory 3×2 χ² on deaths is significant (p = 0.003), even if time-to-event log-rank is not-again pointing to power/precision constraints rather than absence of effect. Conceptually, this echoes extensive PET prognostic work where volumetric burden (MTV/TLG) rather than a single SUV voxel often best captures risk. Meta-analyses across NSCLC show high MTV and high TLG double the hazard for events/death Damuka N et al. (2022)[8]; large single-center series similarly report TLG_WB and MTV_WB as independent predictors after adjustment for stage and treatment Lee JW et al. (2020)[9]. Brain uptake ratios vs MTV/TLG (Table 4). The lowest brain-SUV group (SUV<5) carries the largest MTV and TLG (MTV 773 mL; TLG 6794), whereas higher brain uptake is associated with substantially lower MTV/TLG. Even without formal p-values in the thesis for these between-group comparisons, the magnitude and non-overlapping intervals (e.g., TLG: 3204–10,385 vs 374–2444) argue for a robust inverse relationship. This dovetails with multi-tumor and lung cancer literature where MTV_WB/TLG_WB outperform SUVmax for survival prediction, with proposed cut-points stratifying 5-year OS and median OS by two- to fourfold differences Lee S et al. (2021)[10], Wang D et al. (2024)[11], Vaz SC et al. (2024)[12]. Put together, data suggest a practical composite: (low brain SUV) + (high MTV/TLG) = adverse biology.

The present prospective observational study demonstrates that concurrent reduction in brain FDG uptake observed on baseline 18F-FDG PET/CT correlates with a higher systemic metabolic burden and poorer survival outcomes in patients with newly diagnosed malignancies. Patients with markedly decreased cerebral FDG activity (SUVmax < 5) exhibited significantly higher whole-body MTV and TLG, along with an increased mortality rate compared to those with preserved brain uptake. Although statistical significance for overall survival was limited by small sample size, the consistent trend across groups supports the hypothesis that brain hypometabolism on PET/CT reflects global metabolic derangement and aggressive tumor biology. The study thus establishes that reduced brain FDG uptake-readily identifiable on routine staging PET/CT-may serve as a potential non-invasive prognostic biomarker, complementing conventional volumetric and metabolic indices. Future large-scale, histology-stratified studies with longer follow-up are warranted to validate these findings and integrate them into clinical prognostic models.

1. Chan SC, Yeh CH, Chang JT, Chang KP, Wang JH, Ng SH. Combing MRI perfusion and 18F-FDG PET/CT metabolic biomarkers helps predict survival in advanced nasopharyngeal carcinoma: a prospective multimodal imaging study. Cancers. 2021 Mar 28;13(7):1550. 2. Binneboese A, Covington MF, Horn KP, Archibald ZG, Boucher KM, Morton KA, Hoffman JM. Correlation between FDG-PET uptake and survival in patients with primary brain tumors. American Journal of Nuclear Medicine and Molecular Imaging. 2021 Jun 15;11(3):196. 3. An YS, Kim SH, Roh TH, Park SH, Kim TG, Kim JH. Correlation between 18F-FDG uptake and immune cell infiltration in metastatic brain lesions. Frontiers in Oncology. 2021 Jun 24;11:618705. 4. Yang X, Yang G, Wang R, Wang Y, Zhang S, Wang J, Yu C, Ren Z. Brain glucose metabolism on [18F]-FDG PET/CT: a dynamic biomarker predicting depression and anxiety in cancer patients. Frontiers in Oncology. 2023 May 25;13:1098943. 5. Shen K, Liu B, Zhou X, Ji Y, Chen L, Wang Q, Xue W. The evolving role of 18F-FDG PET/CT in diagnosis and prognosis prediction in progressive prostate cancer. Frontiers in oncology. 2021 Jul 29;11:683793. 6. Kalantari F, Mirshahvalad SA, Hoellwerth M, Schweighofer-Zwink G, Huber-Schönauer U, Hitzl W, Rendl G, Koelblinger P, Pirich C, Beheshti M. Prognostic Value of Baseline 18F-FDG PET/CT to Predict Brain Metastasis Development in Melanoma Patients. Cancers. 2023 Dec 26;16(1):127. 7. Choi EK, Oh JK, Seo YY, Im JJ, Chung YA. Prognostic value of pretreatment F-18 fluorodeoxyglucose PET/CT in colorectal cancer with unresectable metastasis. Nuclear Medicine Communications. 2021 Jun 1;42(6):639-45. 8. Damuka N, Dodda M, Solingapuram Sai KK. PET use in cancer diagnosis, treatment, and prognosis. Cancer Biomarkers: Methods and Protocols. 2022 Jan 19:23-35. 9. Lee JW, Kim SY, Han SW, Lee JE, Lee HJ, Heo NH, Lee SM. [18F] FDG uptake of bone marrow on PET/CT for predicting distant recurrence in breast cancer patients after surgical resection. EJNMMI research. 2020 Jun 30;10(1):72. 10. Lee S, Choi Y, Park G, Jo S, Lee SS, Park J, Shim HK. 18F-FDG PET/CT parameters for predicting prognosis in esophageal cancer patients treated with concurrent chemoradiotherapy. Technology in Cancer Research & Treatment. 2021 Jul 5;20:15330338211024655. 11. Wang D, Mo Y, Liu F, Zheng S, Liu H, Li H, Guo J, Fan W, Qiu B, Zhang X, Liu H. Repeated dynamic [18F] FDG PET/CT imaging using a high-sensitivity PET/CT scanner for assessing non-small cell lung cancer patients undergoing induction immuno-chemotherapy followed by hypo-fractionated chemoradiotherapy and consolidative immunotherapy: report from a prospective observational study (GASTO-1067). European Journal of Nuclear Medicine and Molecular Imaging. 2024 Nov;51(13):4083-98. 12. Vaz SC, Woll JP, Cardoso F, Groheux D, Cook GJ, Ulaner GA, Jacene H, Rubio IT, Schoones JW, Peeters MJ, Poortmans P. Joint EANM-SNMMI guideline on the role of 2-[18F] FDG PET/CT in no special type breast cancer: (endorsed by the ACR, ESSO, ESTRO, EUSOBI/ESR, and EUSOMA). European Journal of Nuclear Medicine and Molecular Imaging. 2024 Jul;51(9):2706-32.