Prognostic Utility of Metabolic Parameters and Tumour Markers in Treatment-Naïve Gallbladder Cancer: A Radiometabolic Analysis.

Study design and patient recruitment

This prospective observational study was carried out at a tertiary care teaching hospital between January 2025 and June 2025. Patients with newly diagnosed, treatment-naïve GBC were prospectively recruited after obtaining informed consent. The patients were referred from the Department of Gastroenterology or Surgery after undergoing clinical and biochemical evaluation, as well as imaging, including abdominal ultrasound and CT scan. They underwent pretreatment FDG PET/CT. Inclusion Criteria included an age of more than 18 years, radiologically suspected or biopsy-confirmed GBC, and an ECOG performance level ranging from 0 to 2 to ensure patients are fit enough for advanced imaging and potential treatment, while still including those with moderate impairment. Radiologically suspected GBC were identified as highly suspicious mass-forming lesions, with or without irregular wall thickening, or biliary obstruction patterns. Histopathology of GBC was confirmed from the primary mass or one of the metastatic sites. Patients received no prior therapy (Surgery, chemotherapy, or radiotherapy). Exclusion Criteria included known non-GBC malignancies and patients lost to follow-up. Patients with poorly managed diabetes (fasting blood glucose >180 mg/dL) were excluded to minimise the confounding effect of hyperglycaemia on FDG uptake. Patients receiving anticancer therapy before the PET/CT scan were also excluded. Clinical data were collected from the hospital's information system and outpatient medical records. We collected demographic data, management strategies before FDG PET/CT, histopathology, and recent clinical status data. The institutional ethics committee approved the study. The study protocol adhered to the ethical guidelines of the institutional review board, and all procedures were conducted per the principles outlined in the Declaration of Helsinki.

Imaging protocol

• Conventional Imaging

All patients underwent triple-phase contrast-enhanced CT (CECT) of the abdomen and pelvis as part of standard-of-care conventional imaging. CT images were interpreted by an experienced radiologist who was blinded to the provisional clinical diagnosis and PET/CT findings. The radiologist reviewed the images to assess the primary mass, its extent, lymph nodes, and potential metastasis. The tumour board decided on the preliminary treatment strategy (Plan A) based on conventional imaging.

• FDG PET/CT Acquisition

All patients underwent FDG PET/CT within two weeks of the CECT. Patients fasted for a minimum of 6 hours. Patients with blood glucose levels below 180 mg/dL were injected with approximately 3.7 MBq/kg of FDG. PET/CT imaging was acquired from the vertex to the mid-thigh on the Discovery IQ Gen² PET/CT scanner (GE Healthcare) after a 60-minute uptake interval. A low-dose CT was initially conducted for attenuation correction and anatomical localisation, followed by PET acquisition for 2 minutes per bed position. Quantitative Metabolic Analysis of the primary GBC was conducted using GE Discovery IQ Gen 2 PET/CT software.

PET/CT image interpretation and parameters obtained

Images were reconstructed and analysed by two nuclear medicine experts who were unaware of the clinical details and CECT findings. All disputes were reconciled through a consensus joint review of images. The primary gallbladder lesion was evaluated for its size and maximum standardised uptake value (SUVmax), normalised to lean body mass (≥2.5). Presence of regional and non-regional lymph node involvement with identification of distant metastases (including liver, peritoneum, and omentum). The SUVmax of all metastatic lesions was noted. The staging derived from PET/CT findings was compared with that from conventional imaging. The gall bladder mass on MTV was evaluated by drawing a semi-automated contour to encase the tumour in the axial, coronal, and sagittal sections.

• Metabolic Tumour Volume (MTV): Quantified as the volume of tumour tissue exhibiting FDG uptake exceeding a SUV of ≥2.5.

• Total Lesion Glycolysis (TLG): Determined by multiplying MTV by the mean SUV of the tumour (SUVmean)

Based on FDG PET/CT findings, the tumour board redecided the final treatment strategy (Plan B).

Impact on management

The impact of PET/CT on management was assessed by comparing the pre-PET/CT treatment plan (Plan A) and post-PET/CT treatment plan (Plan B). Any modifications in treatment intent (e.g., curative to non-curative), modality (e.g., surgery to chemotherapy or radiotherapy), or extent (e.g., extended surgery) were documented. Management change was categorised as Major (changing treatment intent) or minor (shifting method without altering intent). Major changes include a shift from curative to palliative intent or the introduction of neoadjuvant chemoradiotherapy. Minor changes include an increased extent of surgery.

Eighteen patients (36%) who were initially considered for simple cholecystectomy (Plan A) were reconsidered for radical cholecystectomy or neoadjuvant chemotherapy followed by reassessment (Plan B), due to nodal or regional infiltration. Ten patients (20%) were found to have unresectable or metastatic disease, leading to a shift from radical surgery to palliative chemotherapy. In this subgroup, futile surgical intervention would not have conferred a survival benefit and might have introduced unwarranted morbidity. This categorisation adhered to a modified framework from Hillner et al., commonly used in PET/CT impact research.^[22]

Statistical analysis

Normally distributed continuous variables were expressed as mean ± standard deviation, while non-normally distributed variables were expressed as median (range). Categorical variables are presented as frequencies and percentages. The McNemar test or Chi-square test was employed to assess the significance of management alterations following PET/CT. The Mann-Whitney U test or independent t-test was used to evaluate MTV and TLG among patients classified into curative and palliative strategy groups. An examination of the receiver operating characteristic (ROC) curve was conducted to determine cutoff values for MTV and TLG, identifying optimal diagnostic thresholds for predicting noncurative GBC based on metabolic parameters, along with their sensitivity, specificity, and area under the curve (AUC). A p-value less than 0.05 was deemed statistically significant. The Statistical Package for the Social Sciences, version 23 (SPSS, version 23.0; Armonk, NY: IBM Corp.), and MedCalc software were used for data analysis.

Baseline parameters and initial staging

The study included 50 treatment-naïve patients with GBC, with a female predominance (31, 62%). The mean age of the patients was 52.3 years. The study included 50 treatment-naïve patients with gallbladder carcinoma. The metabolic parameters derived from FDG PET/CT revealed that the mean metabolic tumour volume (MTV) was 23.55 cm³ (SD (standard deviation): 13.54), and the mean total lesion glycolysis (TLG) was 151.76 (SD: 13.04). Serum tumour markers showed a mean CA^19-9 level of 1594.87 U/mL (SD: 59.20) and a CEA level of 10.23 ng/mL (SD: 8.92) [Table 1]. CA^19-9 levels were elevated in 40 patients (80%), with a mean and median level of 1,594.87±5,992.80 and 216.50 (54.50, 1250.00)U/mL, respectively. CEA was elevated in 36 patients (72%) with a mean level of 10.23 ± 8.9 ng/mL. Both tumour markers, CA^19-9 and CEA, were elevated in 30 patients (60%). Spearman's rank correlation showed a significant positive correlation between TLG and CA^19-9 levels (r = 0.528, p < 0.001). Conversely, no statistically significant correlations were observed between TLG and CEA (p = 0.187), MTV and CA^19-9 (p = 0.093), or MTV and CEA (p = 0.938)[Table 2].

FDG PET/CT findings

The primary GBC mass was FDG-avid in all patients, with a mean SUVmax of 11.3±6.28. The MTV and TLG were 23.55 ± 13.54 [median 23.40 (13.40, 33.50)] cm³ and 151.76 ± 163.04 [median 112.95 (54.20, 208.00)], respectively. Fig 1 Illustrating Gall Bladder lesion in the fundus. The imaging revealed significant heterogeneity in patterns of liver infiltration and lymphatic dissemination. Liver involvement was seen in 38 (76%) patients. Among these, direct infiltration was observed in 19 patients, while metastatic liver involvement was noted in 19 patients. Patients with liver infiltration had significantly higher MTV (26.11 ± 13.09 vs 13.30 ± 10.48; p = 0.003) and TLG (178.58 ± 171.85 vs 44.51 ± 27.77; p < 0.001). Liver infiltration was also associated with higher CA 19-9 levels (1973.53 ± 6662.12 vs. 80.22 ± 115.89; p < 0.001). Patients with unresectable disease had significantly higher CA 19-9 levels (2259.30 ± 7088.20 vs. 44.53 ± 35.34; p < 0.001) compared to resectable cases. Fig 2 shows FDG avid Gall bladder mass with lymph node involvement and metastasis in left lung. However, no significant difference was noted in the CEA levels.

Close

Fig 1:

61-year-old Female with adenocarcinoma of the gallbladder fundus. (A) Whole-body maximum intensity projection (MIP) image showing focal uptake in the right upper abdomen. (B) Coronal fused PET/CT image demonstrating an FDG-avid lesion in the fundus of the gallbladder. (C) Axial fused PET/CT image showing the FDG-avid gallbladder fundal mass. Red arrows indicate the focal FDG-avid lesion in the fundus of the gallbladder mass in images A, B, and C . (SUV max 4.2, SUV mean 2.4, MTV 31.1, TLG 74.1; serum CA19-9: 24.3 ng/ml). PET: Positron emission tomography; CT: Computed tomography; FDG: Fluorodeoxyglucose; MIP: Maximum intensity projection; SUV: Standardised uptake value; MTV: Metabolic tumor volume; TLG: Total lesion glycolysis.

Export to PPT

Close

Fig 2:

71-year-old man with biopsy-proven gallbladder adenocarcinoma. (A) Whole-body maximum intensity projection (MIP) image showing primary lesion and metastatic sites. (B) Coronal fused PET/CT image demonstrating the FDG-avid gallbladder mass. (C) Axial fused PET/CT image showing FDG-avid gallbladder mass with locoregional lymph node involvement (SUVmax 16.4, SUVmean 9.4, MTV 47.2, TLG 403; serum CA 19-9 >1200 U/ml). (D) Axial fused PET/CT image showing FDG-avid metastatic nodule in the left lung. (E) Additional axial fused PET/CT image highlighting the extent of gallbladder mass and adjacent FDG-avid deposits. Red arrows denote the primary gallbladder lesion (in A-E.), metastatic sites, and the metastatic nodule in the left lung (in D) PET: Positron emission tomography; CT: Computed tomography; FDG: Fluorodeoxyglucose; MIP: Maximum intensity projection; SUV: Standardised uptake value; MTV: Metabolic tumor volume; TLG: Total lesion glycolysis.

Export to PPT

In our cohort of 50 patients, FDG PET/CT demonstrated nodal involvement in a substantial proportion of cases, with multistation nodal disease being the predominant pattern. Periportal nodes were the most frequently involved (48%), followed by pericystic (24%), preaortic (20%), peripancreatic (16%), and retroperitoneal nodal stations (paraaortic, precaval, and aortocaval nodes together accounting for >30%). Importantly, distant nodal involvement (mediastinal, supraclavicular, cervical, and internal mammary lymph nodes) was also identified, indicating systemic dissemination [Fig 3]. Fig 3 showing distribution of PET-positive nodal stations.

Close

Fig 3:

Distribution of PET-positive nodal stations (n=50). (PET: Positron emission tomography)

Export to PPT

In such scenarios, PET-positive nodal findings were considered clinically definitive based on a composite reference standard, incorporating:

1. characteristic anatomical distribution along known gallbladder lymphatic drainage pathways,

2. concordant findings on contrast-enhanced CT,

3. intraoperative assessment where surgery or diagnostic laparoscopy was performed.

Similarly, peritoneal involvement on FDG PET/CT was frequently associated with serosal deposits, ascites, omental involvement, hepatic infiltration, skeletal lesions, or distant metastases, making peritoneal biopsy unnecessary for management decisions in many patients. Diagnostic laparoscopy with biopsy was used where clinically indicated, while in others, PET-positive peritoneal disease was validated by operative findings, imaging correlation, and follow-up.

Change in management

Out of the 50 patients, FDG PET/CT resulted in upstaging in 32 (64%) patients compared with baseline CECT. It was primarily attributed to upstaging, including the detection of previously inconspicuous or equivocal lesions on CECT, to the detection of FDG-avid lymphadenopathy in periportal (n=36), aortocaval (n=18), peri-pancreatic (n=10), and supraclavicular (n=3) areas. Recognition of distant or peritoneal metastases (e.g., omental deposits, serosal nodules) results in upstaging in 12 patients (24%). No downstaging was noted. Based on additional findings from FDG PET/CT, the intended management (Plan B) was revised in 28 out of 50 patients (56%).

Major Management Changes (n = 10; 20%)

A major alteration in therapeutic strategy was observed in 10 patients (20%), in whom FDG PET/CT revealed previously occult metastatic or unresectable diseases that were not detected on conventional cross-sectional imaging modalities. In these cases, the management was revised from a curative intent involving radical cholecystectomy to a palliative systemic chemotherapy. This re-stratification not only prevented non-beneficial futile surgical interventions but also facilitated early transition to palliative care, thereby minimising patient morbidity and enabling more judicious allocation of healthcare resources.

Minor Management Changes (n = 18; 36%)

In 18 patients (36%), FDG PET/CT findings led to a modification in the scope or aggressiveness of the curative strategy. These patients were initially planned for simple or radical cholecystectomy based on conventional imaging (Plan A); however, PET/CT revealed regional nodal involvement or locoregional extension necessitating neoadjuvant chemotherapy protocol followed by reassessment (Plan B). Although the treatment intent remained curative, it was escalated with close reevaluation.

Metabolic parameters and resectability

Among the 50 patients, resectability was influenced significantly by PET metabolic parameters. Patients with unresectable disease had significantly higher mean TLG of gall bladder mass (180.05 ± 18.11 vs 85.77 ± 17.08; p = 0.004) and CA 19-9 levels (2259.30 ± 26.20 vs 44.53 ± 5.34; p < 0.001) compared to resectable cases. Mean CEA levels were also higher in unresectable cases (12.06 ± 9.89 vs. 5.97 ± 3.58; p = 0.006). MTV was slightly higher in unresectable disease but did not reach statistical significance (p = 0.315) [Table 3]. The resectability was influenced significantly by PET metabolic parameters. Patients with unresectable disease had significantly higher mean TLG (180.05 ± 180.11 vs 85.77 ± 87.08; p = 0.004) and CA 19-9 levels (2259.30 ± 7088.20 vs 44.53 ± 35.34; p < 0.001) compared to resectable cases. The area under the curve (AUC) for TLG was 0.744 (p = 0.004). [Fig 4]. Fig 4 shows the sensitivity analysis for MTV and TLG stations. MTV was slightly higher in unresectable disease, but did not reach statistical significance (p = 0.315)

Close

Fig 4:

Sensitivity analysis for MTV and TLG (MTV: Metabolic tumor volume; TLG: Total lesion glycolysis)

Export to PPT

Liver infiltration and metastatic spread

Liver infiltration was present in 80% of patients. Patients with liver involvement had significantly elevated MTV (26.11 ± 1.09 vs 13.30 ± 1.18; p = 0.003) and TLG (178.58 ± 11.85 vs 44.51 ± 7.77; p < 0.001). CA 19-9 was also significantly higher in liver-infiltrated patients (1973.53 ± 532.12 vs. 80.22 ± 15.89; p < 0.001), while CEA was not significantly different.

Our findings

This prospective study included 50 treatment-naïve GBC patients from a developing nation, affirming the late presentation and metastatic burden. Patients frequently present with bulky local disease. All primary GBC masses were metabolically active, with a mean SUVmax of 11.3±6.28. Direct infiltration or liver metastases are widespread; 38 (76%) result in extended resection or palliative management. Similar trends were noted for locoregional or distant lymph nodal involvement. Distant metastases to bone and lung are uncommon. FDG PET/CT upstaged 32 (64%) patients compared with baseline CECT. It led to major and minor management changes in 10 (20%) and 18 (36%) patients, respectively. Unresectable disease was associated with a higher mean TLG of the gall bladder mass.

Metastatic pattern of GBC

The pattern of metastatic dissemination observed in our cohort was consistent with the established natural history of GBC. The liver was the predominant site of infiltration and metastasis. In our series, 80% of patients had liver infiltration or hepatic metastasis. Liver remains the most common site of dissemination in GBC, with involvement in nearly two-thirds of cases.^[23] Late presentation of GBC is well known, as it remains asymptomatic for an extended period. A large study including 326 patients found metastatic disease in 71.4%.^[24]

FDG PET/CT has demonstrated good sensitivity and specificity in the evaluation of GBC. A meta-analysis pooling 21 studies with 495 patients found sensitivities and specificities for primary tumour detection of 87% and 78%, respectively (AUC 0.88).^[13] In our study, the primary SUVmax of the mass was significant (11.3±6.28). For nodal disease, PET/CT has a sensitivity and specificity of 75% and 91% and is especially useful for detecting non-enlarged lymph nodal metastases. It has an unparalleled role in identifying metastatic disease, with a pooled sensitivity and specificity of 95% and 97%, respectively.^[25]

Our findings show that these patients had significantly higher MTV and TLG, suggesting that higher metabolic tumour burden mirrors invasive potential. Strong association between liver infiltration and CA^19-9 levels reinforces the biological aggressiveness. Lymphatic dissemination was highly prevalent, including both locoregional and distant nodes. These findings align closely with the established lymphatic drainage pathways of GBC.^[26] Distant lymph nodal involvement, including mediastinal and supraclavicular nodes, in 6 patients (12%), highlights the value of whole-body imaging in these patients before radical treatments. Extra-abdominal metastases and vascular dissemination are rare. The most common sites of systemic metastases are the lung, followed by the central nervous system, skin, and bone. Distant metastases adversely affect prognosis and survival outcomes.^[27,28] In our study, patients with liver infiltration had significantly higher MTV and TLG. The findings correlate with the research conducted by Chun Y J et al., which indicated that PET/CT parameters were elevated in hepatobiliary malignancies exhibiting vascular invasion or liver metastases, underscoring their role as indicators of local aggressiveness.^[29]

Change in management

The most critical clinical impact of whole-body FDG PET/CT in our study was to disclose metastases in patients considered resectable based on conventional imaging. The upstaging (32, 64%) led to a change in management in more than half of the patients, avoiding futile surgery. The efficacy of FDG PET/CT stems from its ability to identify clinically silent lymph nodes and systemic metastases compared to other imaging modalities. Based on additional findings from FDG PET/CT, the management (Plan B) was revised in more than 50% patients. This observation aligns with the findings of Kumar R et al., who modified therapy in 32% of patients, identifying distant metastases in 25% that were overlooked by CT or MRI alone.^[30]

Metabolic markers and clinical significance

Our research investigated the prognostic significance of disease burden and metabolic parameters, including MTV and TLG. TLG, integrating metabolic activity and tumour load, exhibited a more robust correlation with unresectability than MTV. TLG reveals bulky local disease with high metabolism. This corroborates findings by Mao Z Y et al., who determined that TLG was a superior predictor of progression-free survival and recurrence in cholangiocarcinoma patients compared to SUVmax or MTV.^[31] SUV max is ubiquitously reported in oncologic PET/CT; however, its limitations are apparent, particularly its failure to account for tumour heterogeneity or disease volume. Our results demonstrated a statistically significant association between TLG and CA^19-9 levels. ROC curve analysis demonstrated that TLG had superior predictive value for resectability compared to MTV. The findings are corroborated by a study by Yoo et al., which revealed that an elevated TLG value correlated with diminished overall survival and an increased probability of inoperability in pancreaticobiliary cancer.^[18] This research substantiates the use of TLG as a stratification instrument in treatment decision-making. In contrast to CA^19-9, CEA failed to reveal significant relationships. This corresponds with the findings of Sachan A et al., who indicated that CEA exhibited inadequate sensitivity and specificity for clinical application.^[32]