The prognostic significance of intratumoral heterogeneity of 18F-FDG uptake in patients with oral cavity squamous cell carcinoma

The value of a nomogram based on 18F-FDG PET/CT metabolic parameters and metabolic heterogeneity in predicting distant metastasis in gastric cancer

OBJECTIVE

To investigate the value of metabolic parameters and metabolic heterogeneity from pretreatment deoxy-2-[fluorine-18]-fluoro-D-glucose positron emission tomography/computed tomography (18F-FDG PET/CT) in predicting distant metastasis in gastric cancer.

METHODS

Eighty-six patients with pathologically confirmed gastric adenocarcinoma were included in this study. All patients underwent a whole-body 18F-FDG PET/CT scan before treatment. Clinicopathologic and imaging data were collected, including metabolic parameters such as maximum standardized uptake value (SUVmax), mean standardized uptake value (SUVmean), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) of the primary gastric cancer lesions. Heterogeneity index (HI)-1 was expressed as the absolute value of the linear regression slopes between the MTVs at different SUVmax thresholds (40% × SUVmax, 80% × SUVmax), while HI-2 was expressed as the difference between SUVmax and SUVmean. Patients were randomly divided into training and validation cohorts at a 7:3 ratio. The correlation between the above parameters and distant metastasis in gastric cancer was analyzed using the training cohort. A nomogram prediction model was then established and later verified with the validation cohort. Finally, decision curve analysis was used to evaluate the clinical utility of the model.

RESULTS

This study included 86 patients with gastric cancer, with 60 (69.8%) in the training cohort and 26 (30.2%) in the validation cohort. There was no significant difference in the balanced comparison between both cohorts (all P > .05). Among all patients, 31 (36.0%) developed distant metastasis, while 55 (64.0%) did not. In patients who developed distant tumor metastasis, carcinoembryonic antigen, carbohydrate antigen (CA)12-5, CA19-9, CA72-4, MTV, TLG, and HI-1 were significantly higher than in patients without distant metastasis (all P < .05). Multivariate logistic regression analysis identified CA72-4 (OR: 1.151, 95% CI: 1.020-1.300, P = .023) and HI-1 (OR: 1.647, 95% CI: 1.063-2.553, P = .026) as independent risk factors for predicting distant metastasis in gastric cancer. The nomogram constructed from this analysis exhibited high predictive efficacy in the training (AUC: 0.874, 95% CI: 0.766-0.983) and validation (AUC: 0.915, 95% CI: 0.790-1.000) cohorts, providing a net clinical benefit for patients.

CONCLUSION

HI-1 is an independent risk factor for predicting distant metastasis in gastric cancer. A comprehensive prediction model combining HI-1 with the tumor marker CA72-4 can increase the net clinical benefit for patients.

[18F]FDG PET/CT Imaging and Hematological Parameters Can Help Predict HPV Status in Head and Neck Cancer

To determine whether [18F]FDG PET/CT and hematological parameters provide supportive data to determine HPV status in HNSCC patients.Retrospective analysis of clinical and diagnostic data from 106 patients with HNSCC: 26.4% HPV-positive and 73.6% HPV-negative was performed. The following semiquantitative PET/CT parameters for the primary tumor and hottest lymph node and liver were evaluated: SUVmax, SUVmean, TotalSUV, MTV, TLG, maximum, mean and TLG tumor-to-liver ratio (TLRmax, TLRmean,TLRTLG) and heterogeneity index (HI). Following hematological variables were assessed: white blood cell (WBC); lymphocyte (LYMPH); neutrophil (NEU),monocyte (MON); platelet (PLT); neutrophil-to-lymphocyte ratio (NRL); lymphocyte-to-monocyte ratio (LMR); platelet-to lymphocyte ratio (PLR) and monocyte-to-lymphocyte ratio (MLR). Conventional statistical analyses were performed in parallel with an artificial neural network analysis (Neural Analyzer, v. 2.9.5).Significant between-group differences were observed for two of the semiquantitative PET/CT parameters, with higher values in the HPV-negative group: primary tumor MTV (22.2 vs 9.65; p=0.023), and TLRmax (3.50 vs 2.46; p=0.05). The HPV-negative group also had a significantly higher NEU count (4.84 vs. 6.04; p=0.04), NEU% (58.2 vs. 66.2; p=0.007), and NRL% (2.69 vs. 3.94; p=0.038). Based on ROC analysis (sensitivity 50%, specificity 80%, AUC 0.5), the following variables were independent predictors of HPV-negativity: primary tumor with SUVmax >10; TotalSUV >2800; MTV >23.5; TLG >180; TLRmax >3.7; TLRTLG >5.7; and oropharyngeal localization.Several semiquantitative parameters derived from [18F]FDG PET/CT imaging of the primary tumor (SUVmax, TotalSUV, MTV, TLG, TLRmax and TLRTLG) were independent predictors of HPV-negativity.

Artificial intelligence-based predictive model for guidance on treatment strategy selection in oral and maxillofacial surgery

Application of deep learning (DL) and machine learning (ML) is rapidly increasing in the medical field. DL is gaining significance for medical image analysis, particularly, in oral and maxillofacial surgeries. Owing to the ability to accurately identify and categorize both diseased and normal soft- and hard-tissue structures, DL has high application potential in the diagnosis and treatment of tumors and in orthognathic surgeries. Moreover, DL and ML can be used to develop prediction models that can aid surgeons to assess prognosis by analyzing the patient's medical history, imaging data, and surgical records, develop more effective treatment strategies, select appropriate surgical modalities, and evaluate the risk of postoperative complications. Such prediction models can play a crucial role in the selection of treatment strategies for oral and maxillofacial surgeries. Their practical application can improve the utilization of medical staff, increase the treatment accuracy and efficiency, reduce surgical risks, and provide an enhanced treatment experience to patients. However, DL and ML face limitations, such as data drift, unstable model results, and vulnerable social trust. With the advancement of social concepts and technologies, the use of these models in oral and maxillofacial surgery is anticipated to become more comprehensive and extensive.

Intratumoral metabolic heterogeneity by 18F-FDG PET/CT to predict prognosis for patients with thymic epithelial tumors

BACKGROUND

The aim of the present study was to evaluate the impact of intratumoral metabolic heterogeneity and quantitative 18F-FDG PET/CT imaging parameters in predicting patient outcomes in thymic epithelial tumors (TETs).

METHODS

This retrospective study included 100 patients diagnosed with TETs who underwent pretreatment 18F-FDG PET/CT. The maximum and mean standardized uptake values (SUVmax and SUVmean), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) on PET/CT were measured. Heterogeneity index-1 (HI-1; standard deviation [SD] divided by SUVmean) and heterogeneity index-2 (HI-2; linear regression slopes of the MTV according with different SUV thresholds), were evaluated as heterogeneity indices. Associations between these parameters and patient survival outcomes were analyzed.

RESULTS

The univariate analysis showed that Masaoka stage, TNM stage, WHO classification, SUVmax, SUVmean, TLG, and HI-1 were significant prognostic factors for progression-free survival (PFS), while MTV, HI-2, age, gender, presence of myasthenia gravis, and maximum tumor diameter were not. Subsequently, multivariate analyses showed that HI-1 (p < 0.001) and TNM stage (p = 0.002) were independent prognostic factors for PFS. For the overall survival analysis, TNM stage, WHO classification, SUVmax, and HI-1 were significant prognostic factors in the univariate analysis, while TNM stage remained an independent prognostic factor in multivariate analyses (p = 0.024). The Kaplan Meier survival analyses showed worse prognoses for patients with TNM stages III and IV and HI-1 ≥ 0.16 compared to those with stages I and II and HI-1 < 0.16 (log-rank p < 0.001).

CONCLUSION

HI-1 and TNM stage were independent prognostic factors for progression-free survival in TETs. HI-1 generated from baseline 18F-FDG PET/CT might be promising to identify patients with poor prognosis.

Predictive value of 18F-FDG PET/CT multi-metabolic parameters and tumor metabolic heterogeneity in the prognosis of gastric cancer

OBJECTIVE

We aimed to investigate the predictive value of pre-treatment 18F-FDG PET/CT multi-metabolic parameters and tumor metabolic heterogeneity for gastric cancer prognosis.

METHODS

Seventy-one patients with gastric cancer were included. All patients underwent 18F-FDG PET/CT whole-body scans prior to treatment and had pathologically confirmed gastric adenocarcinomas. Each metabolic parameter, including SUVmax, SUVmean, MTV, and TLG, was collected from the primary lesions of gastric cancer in all patients, and the slope of the linear regression between the MTV corresponding to different SUVmax thresholds (40% × SUVmax, 80% × SUVmax) of the primary lesions was calculated. The absolute value of the slope was regarded as the metabolic heterogeneity of the primary lesions, expressed as the heterogeneity index HI-1, and the coefficient of variance of the SUVmean of the primary lesions was regarded as HI-2. Patient prognosis was assessed by PFS and OS, and a nomogram of the prognostic prediction model was constructed, after which the clinical utility of the model was assessed using DCA.

RESULTS

A total of 71 patients with gastric cancer, including 57 (80.3%) males and 14 (19.7%) females, had a mean age of 61 ± 10 years; disease progression occurred in 27 (38.0%) patients and death occurred in 24 (33.8%) patients. Multivariate Cox regression analysis showed that HI-1 alone was a common independent risk factor for PFS (HR: 1.183; 95% CI: 1.010-1.387, P < 0.05) and OS (HR: 1.214; 95% CI: 1.016-1.450, P < 0.05) in patients with gastric cancer. A nomogram created based on the results of Cox regression analysis increased the net clinical benefit for patients. Considering disease progression as a positive event, patients were divided into low-, intermediate-, and high-risk groups, and Kaplan-Meier survival analysis showed that there were significant differences in PFS among the three groups. When death was considered a positive event and patients were included in the low- and high-risk groups, there were significant differences in OS between the two groups.

CONCLUSION

The heterogeneity index HI-1 of primary gastric cancer lesions is an independent risk factor for patient prognosis. A nomogram of prognostic prediction models constructed for each independent factor can increase the net clinical benefit and stratify the risk level of patients, providing a reference for guiding individualized patient treatment.

Predictive value of intratumoral-metabolic heterogeneity derived from 18F-FDG PET/CT in distinguishing microsatellite instability status of colorectal carcinoma

Purpose/background

Microsatellite instability (MSI) status is a significant biomarker for the response to immune checkpoint inhibitors, response to 5-fluorouracil-based adjuvant chemotherapy, and prognosis in colorectal carcinoma (CRC). This study investigated the predictive value of intratumoral-metabolic heterogeneity (IMH) and conventional metabolic parameters derived from 18F-FDG PET/CT for MSI in patients with stage I-III CRC.

Methods

This study was a retrospective analysis of 152 CRC patients with pathologically proven MSI who underwent 18F-FDG PET/CT examination from January 2016 to May 2022. Intratumoral-metabolic heterogeneity (including heterogeneity index [HI] and heterogeneity factor [HF]) and conventional metabolic parameters (standardized uptake value [SUV], metabolic tumor volume [MTV], and total lesion glycolysis [TLG]) of the primary lesions were determined. MTV and SUVmean were calculated on the basis of the percentage threshold of SUVs at 30%-70%. TLG, HI, and HF were obtained on the basis of the above corresponding thresholds. MSI was determined by immunohistochemical evaluation. Differences in clinicopathologic and various metabolic parameters between MSI-High (MSI-H) and microsatellite stability (MSS) groups were assessed. Potential risk factors for MSI were assessed by logistic regression analyses and used for construction of the mathematical model. Area under the curve (AUC) were used to evaluate the predictive ability of factors for MSI.

Results

This study included 88 patients with CRC in stages I-III, including 19 (21.6%) patients with MSI-H and 69 (78.4%) patients with MSS. Poor differentiation, mucinous component, and various metabolic parameters including MTV30%, MTV40%, MTV50%, and MTV60%, as well as HI50%, HI60%, HI70%, and HF in the MSI-H group were significantly higher than those in the MSS group (all P < 0.05). In multivariate logistic regression analyses, post-standardized HI60% by Z-score (P = 0.037, OR: 2.107) and mucinous component (P < 0.001, OR:11.394) were independently correlated with MSI. AUC of HI60% and our model of the HI60% + mucinous component was 0.685 and 0.850, respectively (P = 0.019), and the AUC of HI30% in predicting the mucinous component was 0.663.

Conclusions

Intratumoral-metabolic heterogeneity derived from 18F-FDG PET/CT was higher in MSI-H CRC and predicted MSI in stage I-III CRC patients preoperatively. HI60% and mucinous component were independent risk factors for MSI. These findings provide new methods to predict the MSI and mucinous component for patients with CRC.

Predictive value of intratumor metabolic and heterogeneity parameters on [18F]FDG PET/CT for EGFR mutations in patients with lung adenocarcinoma

PURPOSE

This study aimed to investigate the value of metabolic and heterogeneity parameters of 2-deoxy-2[¹⁸F]fluoro-D-glucose ([¹⁸F]FDG) positron emission tomography/computed tomography (PET/CT) in predicting epidermal growth factor receptor (EGFR) mutations in patients with lung adenocarcinoma (ADC).

MATERIALS AND METHODS

A retrospective analysis was performed on 157 patients with lung ADC between September 2015 and June 2021, who had undergone both EGFR mutation testing and [¹⁸F]FDG PET/CT examination. Metabolic and heterogeneity parameters were measured and calculated, including maximum diameter (Dmax), maximum standardized uptake value (SUVmax), mean standardized uptake value (SUVmean), metabolic tumor volume (MTV), total lesion glycolysis (TLG), and heterogeneity factor (HF). Relationships between PET/CT parameters and EGFR mutation status were evaluated and a multivariate logistic regression analysis was analyzed to establish a combined prediction model.

RESULTS

108 (68.8%) patients exhibited EGFR mutations. EGFR mutations were more likely to occur in females (51.9% vs. 48.1%, P = 0.007), non-smokers (83.3% vs. 16.7%, P < 0.001) and right lobes (55.6% vs. 44.4%, P = 0.017). High Dmax, MTV and HF and low SUVmean were significantly correlated with EGFR mutations, and the areas under the ROC curve (AUCs) measuring 0.647, 0.701, 0.757, and 0.661, respectively. Multivariate logistic regression analysis suggested that non-smokers (OR = 0.30, P = 0.034), low SUVmean (≤ 7.75, OR = 0.63, P < 0.001) and high HF (≥ 4.21, OR = 1.80, P = 0.027) were independent predictors of EGFR mutations. The AUC of the combined prediction model measured up to 0.863, significantly higher than that of a single parameter.

CONCLUSIONS

EGFR mutant in lung ADC patients showed more intratumor heterogeneity (HF) than EGFR wild type, which was combined clinical feature (non-smokers), and metabolic parameter (SUVmean) may be helpful in predicting EGFR mutation status, thus playing a guiding role in EGFR-tyrosine kinase inhibitors (EGFR-TKIs) targeted therapies.

Prediction value of 18F-FDG PET/CT intratumor metabolic heterogeneity parameters for recurrence after radical surgery of stage II/III colorectal cancer

Purpose

We explored the predictive effect of intratumor metabolic heterogeneity indices extracted from ¹⁸F-FDG PET/CT on recurrence in stage II/III colorectal cancer after radical surgery.

Methods

A total of 140 stage II/III colorectal cancer patients who received preoperative ¹⁸F-FDG PET/CT and radical resection were enrolled. ¹⁸F-FDG traditional parameters including the maximum standardized uptake value (SUVmax), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) under different thresholds; heterogeneity indices including the coefficient of variation with SUV 2.5 as a threshold (CV2.5), CV40%, heterogeneity index-1 (HI-1) calculated by the fixed-threshold method, and HI-2 calculated by the percentage threshold method; and clinicopathological information were collected. We concluded that relationships exist between these data and patients’ disease-free survival (DFS).

Results

Regional lymph node status (P < 0.001), nerve invasion (P = 0.036), tumor thrombus (P = 0.005), and HI-1 (P = 0.010) exhibited significant differences between the relapse and non-relapse groups, while SUVmax, MTV2.5, MTV40%, TLG2.5, TLG40%, CV2.5, CV40%, HI-2, and other clinicopathological factors had no differences between the relapse and non-relapse groups. Multivariate analysis demonstrated that HI-1 (HR = 1.02, 1.00–1.04, P = 0.038), regional lymph node metastasis (HR = 2.95, 1.37–6.38, P = 0.006), and tumor thrombus status (HR = 2.37, 1.13–4.99, P = 0.022) were independent factors significantly related to DFS.

Conclusion

HI-1, tumor thrombus status, and regional lymph node status could predict the recurrence of stage II/III colorectal cancer after radical resection and had an advantage over other ¹⁸F-FDG PET/CT conventional parameters and heterogeneity indices.

Prognostic Value of Molecular Intratumor Heterogeneity in Primary Oral Cancer and Its Lymph Node Metastases Assessed by Mass Spectrometry Imaging

Different aspects of intra-tumor heterogeneity (ITH), which are associated with the development of cancer and its response to treatment, have postulated prognostic value. Here we searched for potential association between phenotypic ITH analyzed by mass spectrometry imaging (MSI) and prognosis of head and neck cancer. The study involved tissue specimens resected from 77 patients with locally advanced oral squamous cell carcinoma, including 37 patients where matched samples of primary tumor and synchronous lymph node metastases were analyzed. A 3-year follow-up was available for all patients which enabled their separation into two groups: with no evidence of disease (NED, n = 41) and with progressive disease (PD, n = 36). After on-tissue trypsin digestion, peptide maps of all cancer regions were segmented using an unsupervised approach to reveal their intrinsic heterogeneity. We found that intra-tumor similarity of spectra was higher in the PD group and diversity of clusters identified during image segmentation was higher in the NED group, which indicated a higher level of ITH in patients with more favorable outcomes. Signature of molecular components that correlated with long-term outcomes could be associated with proteins involved in the immune functions. Furthermore, a positive correlation between ITH and histopathological lymphocytic host response was observed. Hence, we proposed that a higher level of ITH revealed by MSI in cancers with a better prognosis could reflect the presence of heterotypic components of tumor microenvironment such as infiltrating immune cells enhancing the response to the treatment.

Correlation between 18F-FDG PET/CT intra-tumor metabolic heterogeneity parameters and KRAS mutation in colorectal cancer

PURPOSE

The study aimed to evaluate the relationship between intra-tumor metabolic heterogeneity parameters of ¹⁸F-FDG and KRAS mutation status in colorectal cancer (CRC) patients and which threshold heterogeneity parameters could better reflect the heterogeneity characteristics of colorectal cancer.

METHODS

Medical data of 101 CRC patients who underwent ¹⁸F-FDG PET/CT and KRAS mutation analysis were selected. On PET scans, ¹⁸F-FDG traditional indices maximum standardized uptake value (SUVmax), metabolic tumor volume (MTV), total lesion glycolysis (TLG), and heterogeneity parameters coefficient of variation with a threshold of 2.5 (CV_2.5), CV_40%, heterogeneity index-1 (HI-1), and HI-2 of the primary lesions were obtained. We inferred correlations between these ¹⁸F-FDG parameters and KRAS mutation status.

RESULTS

41 patients (40.6%) had KRAS gene mutation. Assessment of FDG parameters showed that SUVmax (19.00 vs. 13.16, p < 0.001), MTV (11.64 vs. 8.83, p = 0.001), and TLG (102.85 vs. 69.76, p < 0.001), CV_2.5 (0.55 vs. 0.46, p = 0.006), and HI-2 (14.03 vs. 7.59, p < 0.001) of KRAS mutation were higher compared to wild-type (WT) KRAS. CV_40% (0.22 vs. 0.24, p = 0.001) was lower in the KRAS mutation group, while HI-1 had no significant difference between the two groups. Multivariate analysis showed that MTV (OR = 4.97, 1.04-23.83, p = 0.045) was the only significant predictor in KRAS mutation, using a cut-off of 7.62 (AUC = 0.695), and MTV showed a sensitivity of 90.2% and specificity of 45.0%. However, the PET parameters were not independent predictors in KRAS mutation.

CONCLUSION

KRAS gene mutant CRC patients had more ¹⁸F-FDG uptake (SUVmax, MTV, TLG) and heterogeneity (CV_2.5, HI-2) than WT KRAS. MTV was the only independent predictor of KRAS gene mutation in colorectal cancer patients.

Prognostic Value of Intratumor Metabolic Heterogeneity Parameters on 18F-FDG PET/CT for Patients with Colorectal Cancer

Purpose

Intratumor metabolic heterogeneity parameters on ¹⁸F-2-fluoro-2-deoxy-D-glucose (¹⁸F-FDG) positron emission tomography-computed tomography (PET-CT) have been proven to be predictors of the clinical prognosis of cancer patients. The study aimed to examine the correlation between ¹⁸F-FDG PET-CT-defined heterogeneity parameters and the prognostic significance in patients with colorectal cancer.

Methods

The study included 188 patients with colorectal cancer who received surgery and ¹⁸F-FDG PET/CT examinations. Preoperative ¹⁸F-FDG PET/CT conventional and metabolic heterogeneity parameters were collected, including maximum, peak, and mean standardized uptake value (SUVmax, SUVpeak, and SUVmean), metabolic tumor volume (MTV), total lesion glycolysis (TLG), heterogeneity index-1 (HI-1) and heterogeneity index-2 (HI-2), and clinicopathological information. Correlations between these parameters and patient survival outcomes were inferred.

Results

The associations between ¹⁸F-FDG PET/CT parameters and clinical outcomes were analyzed. Tumor thrombus (P < 0.001), tumor stage (P=0.001), MTV (P=0.003), HI-1 (P=0.032), and HI-2 (P=0.001) differed between the two groups with and without recurrence. Multivariate analysis showed that, in the radical surgery group, HI-2 (HR = 1.10, 95% CI: 1.04–1.17, P=0.001), tumor stage (HR = 20.65, 95% CI: 4.81–88.62, P < 0.001), and regional lymph nodes status (HR = 0.16, 95% CI: 0.04–0.57, P=0.005) were independent variables significantly correlated with progression-free survival (PFS) and HI-2 (HR = 1.16, 95% CI: 1.07–1.26, P < 0.001) was an independent variable affecting overall survival (OS). In the palliative surgery group, HI-2 (HR = 1.03, 95% CI: 1.01–1.06, P=0.020) was an independent variable affecting PFS, and all the parameters were not statistically significant for OS.

Conclusion

HI-2, tumor stage, and regional lymph nodes status might predict the outcomes of colorectal cancer more effectively than other ¹⁸F-FDG PET/CT defined parameters.

Prediction of the treatment outcome using machine learning with FDG-PET image-based multiparametric approach in patients with oral cavity squamous cell carcinoma

AIM

To investigate the value of machine learning-based multiparametric analysis using 2-[¹⁸F]-fluoro-2-deoxy-d-glucose positron-emission tomography (FDG-PET) images to predict treatment outcome in patients with oral cavity squamous cell carcinoma (OCSCC).

MATERIALS AND METHODS

Ninety-nine patients with OCSCC who received pretreatment integrated FDG-PET/computed tomography (CT) were included. They were divided into the training (66 patients) and validation (33 patients) cohorts. The diagnosis of local control or local failure was obtained from patient's medical records. Conventional FDG-PET parameters, including the maximum and mean standardised uptake values (SUVmax and SUVmean), metabolic tumour volume (MTV), and total lesion glycolysis (TLG), quantitative tumour morphological parameters, intratumoural histogram, and texture parameters, as well as T-stage and clinical stage, were evaluated by a machine learning analysis. The diagnostic ability of T-stage, clinical stage, and conventional FDG-PET parameters (SUVmax, SUVmean, MTV, and TLG) was also assessed separately.

RESULTS

In support-vector machine analysis of the training dataset, the final selected parameters were T-stage, SUVmax, TLG, morphological irregularity, entropy, and run-length non-uniformity. In the validation dataset, the diagnostic performance of the created algorithm was as follows: sensitivity 0.82, specificity 0.7, positive predictive value 0.86, negative predictive value 0.64, and accuracy 0.79. In a univariate analysis using conventional FDG-PET parameters, T-stage and clinical stage, diagnostic accuracy of each variable was revealed as follows: 0.61 in T-stage, 0.61 in clinical stage, 0.64 in SUVmax, 0.61 in SUVmean, 0.64 in MTV, and 0.7 in TLG.

CONCLUSION

A machine-learning-based approach to analysing FDG-PET images by multiparametric analysis might help predict local control or failure in patients with OCSCC.

Glycolytic pathway candidate markers in the prognosis of oral squamous cell carcinoma: a systematic review with meta-analysis

Molecular changes that affect mitochondrial glycolysis have been associated with the maintenance of tumor cells. Some metabolic factors have already been described as predictors of disease severity and outcomes. This systematic review was conducted to answer the question: Is the glycolytic pathway correlated with the prognosis of oral squamous cell carcinoma (OSCC)? A search strategy was developed to retrieve studies in English from PubMed, Scopus, and ISI Web of Science using keywords related to squamous cell carcinoma, survival, and glycolytic pathway, with no restriction of publication date. The search retrieved 1273 publications. After the titles and abstracts were analyzed, 27 studies met inclusion criteria. Studies were divided into groups according to two subtopics, glycolytic pathways and diagnosis, which describe the glycolytic profile of OSCC tumors. Several components of tumor energy metabolism found in this review are important predictors of survival of patients with OSCC.

Prediction of Response in Head and Neck Tumor: Focus on Main Hot Topics in Research

Radiation therapy is a cornerstone in the treatment of head and neck cancer patients; actually, their management is based on clinical and radiological staging with all patients at the same stage treated in the same way. Recently the increasing knowledge in molecular characterization of head and neck cancer opens the way for a more tailored treatment. Patient outcomes could be improved by a personalized radiotherapy beyond technological and anatomical precision. Several tumor markers are under evaluation to understand their possible prognostic or predictive value. In this paper we discuss those markers specific for evaluate response to radiation therapy in head and neck cancer for a shift toward a biological personalization of radiotherapy.

Texture Analysis Using Preoperative Positron Emission Tomography Images May Predict the Prognosis of Patients With Resectable Oral Squamous Cell Carcinoma

PURPOSE

Texture analysis is a computer-assisted technique used to measure intratumoral heterogeneity, which is known to have important roles in cancer research. This study aimed to assess the potential prognostic values of textural features extracted from preoperative ¹⁸F-fluorodeoxyglucose positron emission tomography images in patients with resectable oral squamous cell carcinoma.

PATIENTS AND METHODS

This retrospective cohort study included patients with oral squamous cell carcinoma who underwent resection surgery. We extracted 31 textural indices from preoperative positron emission tomography images. Overall survival (OS) and disease-free survival (DFS) were chosen as the primary outcome variables, and the primary predictor variables were age, sex, primary tumor location, pathological T and N classification, histologic differentiation, resected margin, perineural and lymphovascular invasion, maximum standardized uptake value, and the 14 textural indices selected in the factor analysis. We analyzed OS and DFS using Kaplan-Meier curves, and the differences between survival curves were determined using a log-rank test. The independent prognostic factors were assessed using the Cox-proportional hazards model.

RESULTS

We enrolled 81 patients (median age, 67.3 years; range, 32 to 88 years). The median follow-up duration was 50.1 months (range, 6.3 to 133.7 months). The univariable and multivariable analyses revealed that higher entropy values (≥1.91) were associated with worse OS (hazard ratio, 21.49; 95% confidence interval, 1.36 to 340.71; P = .03) and DFS (hazard ratio, 50.69; 95% confidence interval, 5.23 to 491.18; P = .001).

CONCLUSIONS

This study showed that entropy is a statistically significant prognostic factor of both OS and DFS. Texture analysis using preoperative positron emission tomography images may contribute to risk stratification.

Differentiation of soft tissue and bone sarcomas from benign lesions utilizing 18F-FDG PET/CT-derived parameters

Background

Accurate differentiation between malignant and benign changes in soft tissue and bone lesions is essential for the prevention of unnecessary biopsies and surgical resection. Nevertheless, it remains a challenge and a standard diagnosis modality is urgently needed. The objective of this study was to evaluate the usefulness of ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) PET/CT-derived parameters to differentiate soft tissue sarcoma (STS) and bone sarcoma (BS) from benign lesions.

Methods

Patients who had undergone pre-treatment ¹⁸F-FDG PET/CT imaging and subsequent pathological diagnoses to confirm malignant (STS and BS, n = 37) and benign (n = 33) soft tissue and bone lesions were retrospectively reviewed. The tumor size, PET and low-dose CT visual characteristics, maximum standardized uptake value (SUVmax), metabolic tumor volume (MTV), total lesion glycolysis (TLG), and heterogeneous factor (HF) of each lesion were measured. Univariate and multivariate logistic regression analyses were conducted to determine the significant risk factors to distinguish sarcoma from benign lesions. To establish a regression model based on independent risk factors, and the receiver operating characteristic curves (ROCs) of individual parameters and their combination were plotted and compared. Conventional imaging scans were re-analyzed, and the diagnostic performance compared with the regression model.

Results

Univariate analysis results revealed that tumor size, SUVmax, MTV, TLG, and HF of ¹⁸F-FDG PET/CT imaging in the STS and BS group were all higher than in the benign lesions group (all P values were < 0.01). The differences in the visual characteristics between the two groups were also all statistically significant (P < 0.05). However, the multivariate regression model only included SUVmax and HF as independent risk factors, for which the odds ratios were 1.135 (95%CI: 1.026 ~ 1.256, P = 0.014) and 7.869 (95%CI: 2.119 ~ 29.230, P = 0.002), respectively. The regression model was constructed using the following expression: Logit (P) = − 2.461 + 0.127SUVmax + 2.063HF. The area under the ROC was 0.860, which was higher than SUVmax (0.744) and HF (0.790). The diagnostic performance of the regression model was superior to those of individual parameters and conventional imaging.

Conclusion

The regression model including SUVmax and HF based on ¹⁸F-FDG PET/CT imaging may be useful for differentiating STS and BS from benign lesions.

Deep learning analysis using FDG-PET to predict treatment outcome in patients with oral cavity squamous cell carcinoma

OBJECTIVE

To assess the utility of deep learning analysis using ¹⁸F-fluorodeoxyglucose (FDG) uptake by positron emission tomography (PET/CT) to predict disease-free survival (DFS) in patients with oral cavity squamous cell carcinoma (OCSCC).

METHODS

One hundred thirteen patients with OCSCC who received pretreatment FDG-PET/CT were included. They were divided into training (83 patients) and test (30 patients) sets. The diagnosis of treatment control/failure and the DFS rate were obtained from patients' medical records. In deep learning analyses, three planes of axial, coronal, and sagittal FDG-PET images were assessed by ResNet-101 architecture. In the training set, image analysis was performed for the diagnostic model creation. The test data set was subsequently analyzed for confirmation of diagnostic accuracy. T-stage, clinical stage, and conventional FDG-PET parameters (the maximum and mean standardized uptake value (SUVmax and SUVmean), heterogeneity index, metabolic tumor volume (MTV), and total lesion glycolysis (TLG) were also assessed with determining the optimal cutoff from training dataset and then validated their diagnostic ability from test dataset.

RESULTS

In dividing into patients with treatment control and failure, the highest diagnostic accuracy of 0.8 was obtained using deep learning classification, with a sensitivity of 0.8, specificity of 0.8, positive predictive value of 0.89, and negative predictive value of 0.67. In the Kaplan-Meier analysis, the DFS rate was significantly different only with the analysis of deep learning-based classification (p < .01).

CONCLUSIONS

Deep learning-based diagnosis with FDG-PET images may predict treatment outcome in patients with OCSCC.

KEY POINTS

• Deep learning-based diagnosis of FDG-PET images showed the highest diagnostic accuracy to predict the treatment outcome in patients with oral cavity squamous cell carcinoma. • Deep learning-based diagnosis was shown to differentiate patients between good and poor disease-free survival more clearly than conventional T-stage, clinical stage, and conventional FDG-PET-based parameters.