Prognostic assessment of diffuse large B-cell lymphoma using ultrasound imaging nomogram and PET/CT indicators

Background: Diffuse large B-cell lymphoma (DLBCL) is the most common aggressive subtype of non-Hodgkin lymphoma (NHL). Existing prognostic models, such as the International Prognostic Index (IPI), have limitations. PET/CT with 18F-FDG provides metabolic data for prognosis, but integrating ultrasound imaging remains underexplored. This study aims to develop and validate an ultrasound imaging nomogram combined with PET/CT indicators for DLBCL prognosis, improving risk stratification.

Methods: A retrospective analysis of 200 DLBCL patients was conducted. Ultrasound imaging and PET/CT indicators were used to assess the prognosis of the patients. Clinical data, ultrasound imaging parameters, and PET/CT metabolic indicators (SUVmax, MTV, TMTV, TLG, SPD) were collected. ACR-Rad nomogram was developed and validated using Cox regression, ROC curves, and calibration analysis.

Results: The ultrasound imaging parameters predicted the prognosis of the patients, with SUVmax, MTV, TMTV, TLG, and SPD showing significant correlations with progression-free survival (PFS) and overall survival (OS). Multivariate Cox regression analysis revealed that age, international prognostic index, pathological subtype, and number of sites involved outside the node were independent risk factors for PFS and OS. The ACR-Rad nomogram showed improved predictive accuracy compared to the International Prognosis Index (IPI) alone.

Conclusions: The combination of ultrasound imaging and PET/CT indicators improves prognostic accuracy in DLBCL and supports the implementation of individualized clinical decision-making.

1 Introduction

Diffuse large B-cell lymphoma (DLBCL) is the most common form of aggressive non-Hodgkin lymphoma (NHL) in adults (1, 2). Current prognostic scoring systems, such as the International Prognosis Index (IPI), have limitations in accurately identifying patients at high risk of disease progression or relapse (3). Therefore, it is important to develop precise prognostic indicators to identify individuals who may benefit from early intervention and targeted treatments for improved outcomes. Positron emission tomography/computed tomography (PET/CT) using ¹⁸F-fluorodeoxyglucose (FDG) has been shown to provide valuable prognostic information for DLBCL, particularly through the assessment of quantitative parameters such as the maximum standardized uptake value (SUVmax) and metabolic tumor volume (MTV) (4–6). However, these measurements do not capture the phenotypic characteristics of the disease, including spatial arrangement, diversity, and lesion morphology.

Recently, radiomics-based quantitative image features that capture the texture and shape of lesions have emerged as promising tools for improving the initial prognosis of DLBCL (7–9). These radiomics features provide detailed and quantitative information about the biological characteristics of the disease and have shown prognostic significance, including for predicting progression-free survival (PFS) and overall survival (OS) in DLBCL patients. Ultrasound imaging is another non-invasive imaging modality that can provide valuable information about the characteristics of DLBCL lesions (10). However, the prognostic value of ultrasound imaging parameters in DLBCL has not explored. In this study, we aim to establish and evaluate an ultrasound imaging nomogram combined with PET/CT indicators for the prognostic assessment of DLBCL.

To address the current gap between conventional imaging evaluation and radiomics-based prognostic prediction, we developed an integrated model termed the ACR-Rad nomogram. This nomogram combines standardized ultrasound descriptors established by the American College of Radiology (ACR) with PET/CT-derived quantitative radiomics features (Rad-Score) (11). By incorporating both qualitative ultrasound characteristics and quantitative metabolic indices, the ACR-Rad model aims to achieve a more comprehensive and noninvasive assessment of tumor biology. Through this approach, the present study seeks to enhance prognostic accuracy and facilitate individualized risk stratification in patients with DLBCL.

2 Materials and methods

2.1 General information

A total of 200 individuals diagnosed with DLBCL were chosen as the subjects for our study, and they were admitted to our hospital between August 2021 and August 2023. The research participants consisted of 98 males and 102 females, with an average age of (50.31 ± 8.92) years. All patients exhibited no substantial disparity in the overall clinical data (Table 1, P>0.05).

Table 1

Table 1. Statistical analysis of general clinical data of patients (n=200).

2.1.1 Entry criteria

The DLBCL was diagnosed pathologically. During the initial diagnosis, CT, PET-CT, and other imaging records indicated that a large mass was defined as having a Maximum tumor diameter (MTD) of ≤ 7.5 cm. The patient received systematic treatment, and the follow-up data were comprehensive.

2.1.2 Exclusion criteria

The exclusion criteria were as follows (1): patients with DLBCL involving the central nervous system (2); patients with primary mediastinal large B-cell lymphoma (3); patients who were concurrently diagnosed with other malignant tumors or had relapsed DLBCL forming a sizable mass (4); individuals with severe mental disorders or impaired consciousness; and (5) participants who were unable or unwilling to actively participate in this research.

2.2 Test methods

2.2.1 Statistical analysis of general clinical data of patients

Data on the clinicopathological characteristics of individuals were gathered from medical documents, encompassing factors such as age, gender, Ann Arbor stages (I, II, III, and IV), IPI, pathological subtypes (GCB and non-GCB), and eastern cooperative oncology group. The Eastern Cooperative Oncology Group Performance Status (ECOG PS) score, the presence or absence of symptoms B, the level of lactate dehydrogenase (LDH), the number of extranodal involved sites (≧2, < 2), the neutrophil count, and the Neutrophil-to-lymphocyte ratio (NLR) were determined by dividing the absolute neutrophil count by the absolute lymphocyte count (12–14).

2.2.2 Evaluation of therapeutic effect

The study followed all participants for a period of 5 years. Throughout the follow-up period, it is important to gather various details about the patients, such as their condition after treatment, any advancements made, instances of disease reoccurrence, factors leading to mortality, and the timing of their demise. The evaluation of lymphoma efficacy, based on international standards, can be categorized into complete remission, partial remission, disease stability, and disease progression. Patients’ five-year PFS and OS durations are calculated, with PFS representing the interval between the diagnosis of diffuse large B-cell lymphoma through puncture or surgical resection of nearby lymph nodes and the onset of disease progression, relapse, or death from any cause, or until the end of the follow-up period. OS is defined as the period starting from the pathological diagnosis of diffuse large B-cell lymphoma through either lymph node puncture or surgical resection, until death from any cause or the end of the follow-up period (15).

2.2.3 Exploration of ultrasonic imaging images

The GE Logic E8 color Doppler ultrasound machine is used. The ultrasound examination is reviewed by two associate chief physicians or chief physicians together. The frequency of the ultrasound probe can be adjusted from 5 to 14 MHz. During the examination, a comprehensive scan will be performed, and the maximum diameter, aspect ratio, and blood flow grading will be recorded. The blood flow grading is divided into levels 0 to III based on the abundance of blood flow signals. Level 0 indicates no blood flow signals observed; level I indicates a small amount of blood flow signals, usually 1 or 2 dot-like or strip-like signals; level II indicates a moderate amount of blood flow signals, usually 3 or 4 dot-like or strip-like signals; level III indicates abundant blood flow, usually more than 4 dot-like or strip-like signals.

The GE Discovery MI PET/CT imaging system was used, and ¹⁸F-FDG was produced by the PET Trace cyclotron. After fasting for 4–6 hours (no oral or intravenous administration of glucose-containing liquids), 185–370 MBq of ¹⁸F-FDG (4.44 MBq/kg) was intravenously injected. Blood glucose levels were immediately checked before administration (<11.1mmol/L). The weight of each patient was measured before the scan for the calculation of standardized uptake values (SUV). Whole-body PET/CT scans were performed 60 minutes after the injection of the radiopharmaceutical (from the base of the skull to the upper thigh). Data were collected for 1-1.5 minutes at each bed position, and CT data were used for attenuation correction. The corrected PET images were reconstructed using an ordered subset expectation maximum algorithm with a matrix size of 192×192. Image registration and fusion were performed using AW software for the images obtained from PET and CT scans (16).

For the GE Discovery MI PET/CT system, the tube voltage was set to 120 kV, the tube current was set to automatic (260–350 mAs), and the gantry rotation time was set to 0.7s. The matrix size was set to 512×512, the slice thickness was set to 3.75 mm, and the pitch was set to 0.984. PET images were acquired in 3D mode with a matrix size of 192×192 and a slice thickness of 2.79 mm. A total of 6–10 bed positions were acquired, with an acquisition time of 1-1.5 minutes per bed position. The PET images were reconstructed using an ordered subset expectation maximization algorithm and attenuation correction was performed using the CT images. The fused PET and CT images were used to obtain cross-sectional, coronal, and sagittal images. Finally, the images were uploaded to the workstation and saved to the PACS storage server (12, 17).

Using the Grow Cut algorithm in 3D slicer software (version 4.8.0; https://www.slicer.org), PET images are uploaded for image analysis, enabling the semi-automatic generation of Volume of interest (VOI).

Besides radiologic characteristics, traditional metabolic measurements like SUVmax, MTV, total lesion glycolysis (TLG), tumor TLG (tTLG) are also computed using volume viewer software (CompassView 5.0, Philips Corp) on dedicated workstations. The determination of MTV involves creating a circular ROI on axial, coronal, and sagittal PET images. The voxel boundary is then automatically generated using the 41% SUVmax threshold method, which is endorsed by the European Nuclear Medicine Association. SUVmax is determined as the maximum SUV among all hypermetabolic tumor lesions in each PET dataset, while the total metabolic tumor volume (TMTV) is calculated by summing the MTV of all lesions. The total lesion glycolysis is determined by multiplying the mean standardized uptake value (SUV) with the MTV (TLG = sum of [SUV mean × MTV]) (18).

To ensure measurement consistency, all ultrasound and PET/CT examinations were performed and interpreted by two senior radiologists with more than 10 years of experience in lymphoma imaging. Both operators underwent standardized training prior to the study, and all measurements were independently performed and then averaged. In cases of discrepancy greater than 10%, a third senior reviewer adjudicated the final value. Inter-observer agreement for key quantitative parameters (SUVmax, MTV, TLG, and ultrasound size measurements) was evaluated in a random subset of 30 patients, showing good reproducibility (intraclass correlation coefficient > 0.85).

2.2.4 Establishment and evaluation of ACR-rad nomogram

The training queue was analyzed using Cox regression analysis to identify potential independent predictors. The initial step involved establishing the clinical model by utilizing the clinical risk factors that were deemed significant in the multivariate analysis. Next, the significant PET indexes (SUVmax, TMTV, TLG, radiological features) identified in the Cox regression analysis were combined with the clinical model to create a unified model. The combined model is visualized in the form of a graph (19).

1. Univariate and multivariate Logistic regression were used to analyze the clinical, pathological, and ultrasonic characteristics of patients with tumors. Independent risk predictors of highly malignant tumors were identified based on a significance level of P < 0.05. The mean values of ACR-Score1 and ACR-Score2, as well as ACR-Score and Rad-Score, were utilized as continuous variables for constructing a merged model through Logistic regression and subsequently transforming it into a nomogram. The ACR-Rad nomogram was constructed as a joint nomogram in this particular study. ACR-Score developed an ACR-Score model using univariate Logistic regression for comparison.

2. The evaluation of the ACR-Rad nomogram included comparing its information criterion (AIC) with the BIC to assess fitting quality. Additionally, the calibration of the nomogram was evaluated using the Hosmer-Lemeshow test and calibration curve, while the fitness of the nested Logistic regression model was assessed through the likelihood ratio test (LR).

3. The diagnostic efficiency was assessed using AUC, and the Delong test was used to compare the AUC values.

4. Verify the accuracy and clinical value of the model by comparing the follow-up outcomes with the prognosis prediction model for diffuse large B-cell lymphoma, which is based on the ultrasound imaging nomogram.

2.2.5 Statistical analysis

SPSS 22.0 (IBM Corp) and R statistical software (version 4.0.2) were used for all statistical tests. P < 0.05 was considered statistically significant. To assess the outlook of individuals with diffuse large B-cell lymphoma, the endpoints chosen were PFS and OS. The calculation of PFS involved determining the duration between the diagnosis date and the first occurrence of recurrence, progression, death from any cause, or the most recent follow-up. Similarly, OS was determined by calculating the duration from the diagnosis date to the occurrence of death from any cause or the most recent follow-up. Different clinical information between the training group and the verification group was assessed using independent sample t-test, χ²-test, or Mann-Whitney U-test. The optimal threshold of radiologic characteristics was determined based on the Receiver operating characteristic (ROC) curve of the subjects. The Cox regression analysis was employed to examine the significance of potential predictors that could act independently and construct the model. The Kaplan-Meier method was used to assess the living conditions and compare them using the Log-rank test. Additionally, the calculation of the time-varying area (tdAUC) under the ROC curve was performed. The model was assessed using the calibration curve.

3 Results

3.1 Evaluation of therapeutic efficacy

The majority of patients underwent 6 rounds of chemotherapy, while a few patients underwent 7 rounds, and a minority of patients underwent 8 rounds. During the follow-up period, 28.0% of patients experienced disease recurrence or progression, and 11.0% died. The estimated five-year PFS rate was 57.3%, and the OS rate reached 84.0%, indicating generally favorable treatment outcomes among the included DLBCL patients (Table 2). Representative baseline imaging findings are shown in Figure 1. On B-mode (2D) ultrasound, multiple fused hypoechoic nodal masses with indistinct margins and an absent hilum were observed (Figure 1A). Color Doppler flow imaging (CDFI) revealed abundant intranodal vascularity, corresponding to Adler grade III blood-flow patterns (Figure 1B). PET/CT imaging demonstrated extensive hypermetabolic lymphadenopathy in the cervical, supraclavicular, and mediastinal regions with markedly elevated SUVmax values (Figure 1C). These multimodal images illustrate the typical morphological and metabolic characteristics of aggressive DLBCL lesions used for the development of the ACR-Rad prognostic model.

Table 2

Table 2. Evaluation of therapeutic effect. [n (%)].

Figure 1

Figure 1. Representative baseline imaging features of DLBCL. (A) B-mode (2D ultrasound) image showing multiple fused hypoechoic nodal lesions with indistinct margins and absence of a visible hilum. (B) Color Doppler flow imaging (CDFI) demonstrating multiple strip-like intranodal blood flow signals, corresponding to Adler grade III vascularity. (C) PET/CT image revealing multiple hypermetabolic lymph nodes in the bilateral cervical, supraclavicular, and mediastinal regions, with a maximum standardized uptake value (SUVmax) of 24.5.

3.2 Prediction of PFS and OS boundary values by parameters of ultrasonic imaging images

The ultrasound imaging parameters predicted the boundary values of PFS and OS, and they were associated with the patients’ prognosis. ROC curve analysis was conducted to determine the optimal cut-off values of ultrasound and PET/CT parameters for predicting PFS and OS. As shown in Table 3, all parameters demonstrated significant predictive value for PFS (P < 0.05). Among these, selective peripheral denervation (SPD) and SUVmax achieved the highest diagnostic performance, with AUC values of 0.746 and 0.724, respectively, suggesting that increased metabolic activity and lesion extent were strongly associated with shorter PFS. Similarly, all imaging parameters were significantly correlated with OS (Table 4). SPD, TLG, and MTV exhibited the best discrimination for survival outcomes (AUC = 0.756, 0.746, and 0.728, respectively). These findings indicate that higher tumor burden and metabolic activity are linked to an increased risk of mortality in patients with DLBCL, underscoring the prognostic utility of combined ultrasound and PET/CT metrics in survival prediction.

Table 3

Table 3. Prediction of PFS boundary value by parameters of ultrasonic imaging images.

Table 4

Table 4. Prediction of OS boundary value by parameters of ultrasonic imaging images.

3.3 The influence of clinical factors of patients with DLBCL on PFS and OS

Kaplan-Meier survival analysis was used to explore the association between clinical characteristics and patient outcomes (Tables 5, 6). Among all variables analyzed, age, IPI, pathological subtype, and ECOG PS were significantly correlated with progression-free survival (P < 0.05). Patients older than 60 years, with higher IPI scores, non-GCB subtype, or ECOG PS ≥ 2 had markedly shorter PFS, indicating poorer disease control. Similarly, age, IPI, pathological subtype, LDH level, and number of extranodal sites were significantly associated with overall survival. In particular, older patients and those with elevated LDH or multiple extranodal involvements exhibited inferior OS outcomes. These results highlight that both systemic disease burden and impaired functional status are major determinants of prognosis in DLBCL (Table 5, 6).

Table 5

Table 5. The influence of clinical factors on PFS in patients with DLBCL.

Table 6

Table 6. The influence of clinical factors of DLBCL patients on OS.

3.4 Influence of parameters of ultrasonic imaging image on PFS and OS

The Kaplan-Meier analysis revealed that all ultrasound and PET/CT-derived quantitative parameters were significantly associated with PFS (P < 0.05, Table 7). Patients in the high-value groups of SUVmax, MTV, TMTV, TLG, TTLG, and SPD exhibited significantly shorter PFS compared with those in the low-value groups. These findings indicate that higher metabolic activity and greater tumor burden are strongly linked to early disease progression and inferior treatment response. Likewise, all imaging parameters were significantly correlated with OS (P < 0.05, Table 8). Elevated values of SPD, TLG, and MTV in particular were associated with poorer OS outcomes, suggesting that patients with higher baseline metabolic tumor activity tend to have a higher mortality risk. Collectively, these results confirm that both metabolic and volumetric imaging indices are robust prognostic markers for disease progression and survival in DLBCL.

Table 7

Table 7. Influence of parameters of ultrasonic imaging image on PFS.

Table 8

Table 8. Influence of parameters of ultrasonic imaging image on OS.

3.5 Multivariate Cox regression analysis of patients with DLBCL

To identify independent prognostic factors, all variables with statistical significance in the univariate analysis were entered into the multivariate Cox regression model (Table 9). The results demonstrated that age, IPI, pathological subtype, and number of extranodal sites were independent predictors of survival (P < 0.05). Specifically, older age (>60 years), higher IPI scores, non-GCB subtype, and multiple extranodal involvements were associated with increased mortality risk and shorter survival durations. In contrast, ECOG performance status and LDH level did not retain statistical significance after adjustment, suggesting that their effects were largely mediated through the IPI composite score. These findings highlight that disease aggressiveness, systemic spread, and biological subtype jointly determine the prognosis of DLBCL patients.

Table 9

Table 9. Multivariate Cox regression analysis of patients with DLBCL.

3.6 Establishment and evaluation of ACR-Rad nomogram of DLBCL prognosis model

Based on the multivariate logistic regression analysis, both the ACR-Score and Rad-Score were identified as independent prognostic indicators for DLBCL (P < 0.05). These two indices were integrated to construct the ACR-Rad nomogram, which combines ultrasound imaging features, PET/CT metabolic parameters, and key clinical variables to predict individual patient outcomes. The calibration curve showed good agreement between the predicted and observed probabilities of survival, indicating excellent model fitting. The predictive accuracy of the ACR-Rad model was high, with forecast accuracies of 96.78% and 98.41% for the favorable- and poor-prognosis groups, respectively (Table 10; Figure 2). These results demonstrate that the integrated nomogram significantly improves prognostic discrimination compared with conventional clinical scoring systems such as the IPI, supporting its potential value for individualized risk stratification and treatment decision-making in DLBCL.

Table 10

Table 10. Prediction accuracy of DLBCL model.

Figure 2

Figure 2. ACR-Rad nomogram of DLBCL prognosis.

4 Discussion

DLBCL is a common and aggressive form of non-Hodgkin lymphoma (NHL) in adults (1, 20). Accurate prognostic assessment is crucial for determining the appropriate treatment strategy and predicting patient outcomes. Recent prognostic studies based on transcriptomic and metabolic profiles have demonstrated that gene expression signatures can effectively predict overall survival in DLBCL (21). These molecular models provide valuable biological insights of DLBCL prediction. To further enhance clinical prediction and facilitate risk assessment using routinely available methods, the present study established an ACR-Rad nomogram that integrates ultrasound morphological features with PET/CT metabolic parameters. This imaging-based approach provides a noninvasive and accessible tool for prognostic evaluation.

Our results showed that ultrasound imaging parameters, including SUVmax, MTV, TMTV, TLG, and SPD, were significantly correlated with PFS and OS in patients with DLBCL. These findings suggest that ultrasound imaging can provide valuable prognostic information for DLBCL patients. Previous studies have mainly focused on the use of PET/CT parameters, such as SUVmax and MTV, for prognostic assessment in DLBCL. For example, a study by Xie et al. found that high SUVmax was associated with inferior PFS and OS in DLBCL patients (22). Similarly, a study by Oh et al. demonstrated that high MTV was predictive of poor prognosis in DLBCL patients (23). However, our study highlights the potential of ultrasound imaging as an additional tool for prognostic evaluation in DLBCL.

Several studies have shown that ultrasound imaging parameters can be valuable in predicting the prognosis of DLBCL patients. For example, one study by Oiwa et al. (2023) found that lymph node size and the presence of extranodal involvement were significantly associated with poor prognosis in DLBCL patients (24). Another study by Kariya et al. (2023) found that the presence of a hypoechogenic rim around the lymph nodes was associated with worse survival outcomes (10). In our study, we confirmed the importance of ultrasound imaging parameters in prognostic assessment of DLBCL. We found that several ultrasound imaging parameters, including lymph node size, presence of extranodal involvement, and hypoechogenic rim, were significantly associated with both progression-free survival (PFS) and overall survival (OS) in DLBCL patients. These findings are consistent with previous studies and highlight the importance of ultrasound imaging in predicting prognosis in DLBCL.

Furthermore, we identified several clinical factors that were independent risk factors for PFS and OS in DLBCL patients. These factors included age, international prognostic index (IPI), pathological subtype, and the number of sites involved outside the lymph nodes. These findings are consistent with previous studies that have identified these factors as important predictors of prognosis in DLBCL (25–28). To further improve the accuracy of prognostic assessment, we developed an ACR-Rad nomogram that combines ultrasound imaging parameters and clinical factors. The ACR-Rad nomogram showed improved predictive accuracy compared to the International Prognosis Index (IPI) alone. This suggests that the integration of ultrasound imaging parameters into the prognostic model can enhance the risk assessment and help guide personalized treatment strategies for DLBCL patients.

The use of ultrasound imaging in DLBCL prognostic assessment is a novel approach that has not been extensively explored in previous studies. Ultrasound imaging provides detailed information about the characteristics of DLBCL lesions, including spatial arrangement, diversity, and lesion morphology (29). This additional information can complement the quantitative measurements obtained from PET/CT imaging and improve the accuracy of prognostic assessment.

There are several limitations to our study. First, the sample size was relatively small, and the study was conducted at a single center. Further studies with larger sample sizes and multi-center collaborations are needed to validate our findings. Second, the follow-up period was relatively short, and longer follow-up is necessary to assess the long-term prognostic value of ultrasound imaging parameters.

5 Conclusion

In conclusion, our study demonstrates that ultrasound imaging parameters, in combination with PET/CT indicators, can accurately predict the prognosis of patients with DLBCL. The ACR-Rad nomogram provides a more precise risk assessment and can assist in personalized treatment strategies. The integration of ultrasound imaging into the prognostic model represents a promising approach for improving the prognostic assessment of DLBCL patients. Further research is needed to validate these findings and explore the potential clinical applications of ultrasound imaging in DLBCL management.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement

The studies involving humans were approved by Medical Ethics Committee of Hunan Cancer Hospital (approval number: KYJJ - 2023 - 367). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

Author contributions

WL: Writing – review & editing, Project administration, Conceptualization, Supervision. LH: Data curation, Formal Analysis, Methodology, Investigation, Conceptualization, Writing – original draft. MW: Investigation, Writing – original draft, Methodology, Formal Analysis, Software, Data curation. RZ: Software, Investigation, Writing – original draft, Formal Analysis, Methodology. PL: Methodology, Data curation, Investigation, Software, Writing – original draft, Formal Analysis.

Funding

The author(s) declare that no financial support was received for the research and/or publication of this article.

Acknowledgments

The authors would like to express their gratitude to all the staff who were involved in this study.

Conflict of interest

The authors declare that the research 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) declare that no Generative AI was 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.

References

1. Skulimowski A, Hogendorf P, Poznańska G, Smolewski P, Strzelczyk J, and Durczynski A. Successful hemihepatectomy following chemotherapy for primary liver lymphoma: case report and review of literature. Pol Przegl Chir. (2017) . 89:54–8. doi: 10.5604/01.3001.0010.5609

PubMed Abstract | Crossref Full Text | Google Scholar

2. Shao R, Liu C, Xue R, Deng X, Liu L, Song C, et al. Tumor-derived exosomal ENO2 modulates polarization of tumor-associated macrophages through reprogramming glycolysis to promote progression of diffuse large B-cell lymphoma. Int J Biol Sci. (2024) . 20:848–63. doi: 10.7150/ijbs.91154

PubMed Abstract | Crossref Full Text | Google Scholar

3. Liu T, Chen L, Pan J, Pan L, Hu J, and Ji Z. Retrospective analysis of a new prognostic score for diffuse large B-cell lymphoma based on interim positron emission tomography-computed tomography. Acta Haematol. (2018) . 139:148–57. doi: 10.1159/000479486

PubMed Abstract | Crossref Full Text | Google Scholar

4. Jain Y, Agrawal A, Joshi A, Menon S, Prakash G, Murthy V, et al. Can 18F FDG PET/CT metabolic parameters be used to noninvasively differentiate between different histopathological subtypes and Fuhrman grades of renal cell cancer? Nucl Med Commun. (2024) 45(7):601–11. doi: 10.1097/mnm.0000000000001844

PubMed Abstract | Crossref Full Text | Google Scholar

5. Na SJ, o JH, Park JM, Lee HH, Lee SH, Song KY, et al. Prognostic value of metabolic parameters on preoperative 18F-fluorodeoxyglucose positron emission tomography/computed tomography in patients with stage III gastric cancer. Oncotarget. (2016) 7:63968–80. doi: 10.18632/oncotarget.11574

PubMed Abstract | Crossref Full Text | Google Scholar

6. Cengiz A, Yersal Ö, Ömürlü İK, and Yürekli Y. Prognostic role of metabolic 18F-FDG PET/CT parameters and hematological prognostic indicators in patients with colorectal cancer. Indian J Cancer. (2023) . 60:224–9. doi: 10.4103/ijc.IJC_1029_20

PubMed Abstract | Crossref Full Text | Google Scholar

7. Kostakoglu L, Dalmasso F, Berchialla P, Pierce LA, Vitolo U, Martelli M, et al. A prognostic model integrating PET-derived metrics and image texture analyses with clinical risk factors from GOYA. EJHaem. (2022) . 3:406–14. doi: 10.1002/jha2.421

PubMed Abstract | Crossref Full Text | Google Scholar

8. Mazzara S, Travaini L, Botta F, Granata C, Motta G, Melle F, et al. Gene expression profiling and FDG-PET radiomics uncover radiometabolic signatures associated with outcome in DLBCL. Blood Adv. (2023) . 7:630–43. doi: 10.1182/bloodadvances.2022007825

PubMed Abstract | Crossref Full Text | Google Scholar

9. Li M, Yao H, Zhang P, Zhang L, Liu W, Jiang Z, et al. Development and validation of a [(18)F]FDG PET/CT-based radiomics nomogram to predict the prognostic risk of pretreatment diffuse large B cell lymphoma patients. Eur Radiol. (2023) 33(5):3354–65. doi: 10.1007/s00330-022-09301-5

PubMed Abstract | Crossref Full Text | Google Scholar

10. Kariya A, Tachibana T, Hiramatsu Y, Wani Y, Matsumoto JY, Furukawa C, et al. Primary thyroid lymphoma: clinical factors predicting the possibility of diffuse large B-cell lymphoma. Ear Nose Throat J. (2023) 10:1455613231218130. doi: 10.1177/01455613231218130

PubMed Abstract | Crossref Full Text | Google Scholar

11. Huang X, Wu Z, Zhou A, Min X, Qi Q, Zhang C, et al. Nomogram combining radiomics with the american college of radiology thyroid imaging reporting and data system can improve predictive performance for Malignant thyroid nodules. Front Oncol. (2021) 11:737847. doi: 10.3389/fonc.2021.737847

PubMed Abstract | Crossref Full Text | Google Scholar

12. Freeman CL, Savage KJ, Villa DR, Scott DW, Srour L, Gerrie AS, et al. Long-term results of PET-guided radiation in patients with advanced-stage diffuse large B-cell lymphoma treated with R-CHOP. Blood. (2021) 137(7):929–38. doi: 10.1182/blood.2020005846

PubMed Abstract | Crossref Full Text | Google Scholar

13. Karak FE, Bou-Orm IR, Ghosn M, Kattan J, and Haidar MJPO. PET/CT scanner and bone marrow biopsy in detection of bone marrow involvement in diffuse large B-cell lymphoma. PLoS One. (2017) 12(1):e0170299. doi: 10.1371/journal.pone.0170299

PubMed Abstract | Crossref Full Text | Google Scholar

14. Bento L, Díaz-López A, Barranco G, Martín-Moreno AM, Baile M, Martín A, et al. New prognosis score including absolute lymphocyte/monocyte ratio, red blood cell distribution width and beta-2 microglobulin in patients with diffuse large B-cell lymphoma treated with R-CHOP: Spanish Lymphoma Group Experience (GELTAMO) %. J Br J Haematol. (2020) 188:888–97. doi: 10.1111/bjh.16263

PubMed Abstract | Crossref Full Text | Google Scholar

15. Bartlett NL, Wilson WH, Jung SH, Hsi ED, and JPJJoCO L. Dose-adjusted EPOCH-R compared with R-CHOP as frontline therapy for diffuse large B-cell lymphoma: Clinical outcomes of the Phase III intergroup trial alliance/CALGB. J Clin Oncol. (2019) 37(21):1790–9. doi: 10.1200/JCO.18.01994

PubMed Abstract | Crossref Full Text | Google Scholar

16. Cottereau AS, Meignan M, Nioche C, and Capobianco N. Oncology IBJAo. Risk stratification in diffuse large B-cell lymphoma using lesion dissemination and metabolic tumor burden calculated from baseline PET/CT. Ann Oncol. (2021) 32(3):404–11. doi: 10.1016/j.annonc.2020.11.019

PubMed Abstract | Crossref Full Text | Google Scholar

17. Ujjani CS, Hill EM, Wang H, Nassif S, Esposito G, Ozdemirli M, et al. 18F-FDG PET-CT and trephine biopsy assessment of bone marrow involvement in lymphoma. Br J Haematol. (2016) 174(3):410–6. doi: 10.1111/bjh.14071

PubMed Abstract | Crossref Full Text | Google Scholar

18. Zhang X, Chen L, Jiang H, He X, Feng L, Ni M, et al. A novel analytic approach for outcome prediction in diffuse large B-cell lymphoma by [18F]FDG PET/CT. Eur J Nucl Med Mol Imaging. (2022) 49(4):1298–310. doi: 10.1007/s00259-021-05572-0

PubMed Abstract | Crossref Full Text | Google Scholar

19. Li YH, Zhao YM, Jiang YL, Tang S, Chen MT, Xiao ZZ, et al. The prognostic value of end-of-treatment FDG-PET/CT in diffuse large B cell lymphoma: comparison of visual Deauville criteria and a lesion-to-liver SUVmax ratio-based evaluation system. Eur J Nucl Med Mol Imaging. (2022) 49(4):1311–21. doi: 10.1007/s00259-021-05581-z

PubMed Abstract | Crossref Full Text | Google Scholar

20. Yang W, Tang B, Xu D, and Yang W. Lenalidomide regulates the CCL21/CCR7/ERK1/2 axis to inhibit migration and proliferation in diffuse large B-cell lymphoma. Oncol Res. (2025) 33:199–212. doi: 10.32604/or.2024.050036

PubMed Abstract | Crossref Full Text | Google Scholar

21. Wang H, Shao R, Liu W, Tang H, and Lu Y. Identification of a prognostic metabolic gene signature in diffuse large B-cell lymphoma. J Cell Mol Med. (2021) . 25:7066–77. doi: 10.1111/jcmm.16720

PubMed Abstract | Crossref Full Text | Google Scholar

22. Xie M, Wu K, Liu Y, Jiang Q, and Xie Y. Predictive value of F-18 FDG PET/CT quantization parameters in diffuse large B cell lymphoma: a meta-analysis with 702 participants. Med Oncol. (2015) . 32:446. doi: 10.1007/s12032-014-0446-1

PubMed Abstract | Crossref Full Text | Google Scholar

23. Oh MY, Oh SB, Seoung HG, Kim JH, Kim SM, Kim TK, et al. Clinical significance of standardized uptake value and maximum tumor diameter in patients with primary extranodal diffuse large B cell lymphoma. Korean J Hematol. (2012) . 47:207–12. doi: 10.5045/kjh.2012.47.3.207

PubMed Abstract | Crossref Full Text | Google Scholar

24. Oiwa K, Fujita K, Lee S, Morishita T, Tsujikawa T, Negoro E, et al. Prognostic value of metabolic tumor volume of extranodal involvement in diffuse large B cell lymphoma. . Ann Hematol. (2023) 102(5):1141–8. doi: 10.1007/s00277-023-05165-x

PubMed Abstract | Crossref Full Text | Google Scholar

25. Wen J, Zhou J, Liu Z, Liu T, and Xu C. Analysis of survival and prognosis in 409 newly diagnosed patients with diffuse large B-cell lymphoma. Zhonghua Xue Ye Xue Za Zhi. (2014) . 35:318–24. doi: 10.3760/cma.j.issn.0253-2727.2014.04.015

PubMed Abstract | Crossref Full Text | Google Scholar

26. Gaudio F, Giordano A, Perrone T, Pastore D, Curci P, Delia M, et al. High Ki67 index and bulky disease remain significant adverse prognostic factors in patients with diffuse large B cell lymphoma before and after the introduction of rituximab. Acta Haematol. (2011) . 126:44–51. doi: 10.1159/000324206

PubMed Abstract | Crossref Full Text | Google Scholar

27. Wang M, Zhang PP, Zhu Q, Hu ZL, Bai X, Wu YP, et al. Expression of MCP-1 and CCR2 in newly diagnosed diffuse large B-cell lymphoma and clinical significance. Zhongguo Shi Yan Xue Ye Xue Za Zhi. (2024) . 32:112–9. doi: 10.19746/j.cnki.issn.1009-2137.2024.01.018

PubMed Abstract | Crossref Full Text | Google Scholar

28. Huo YJ, Zhang MC, Shi Q, Qin W, Shi ZY, Wang L, et al. Clinical characteristics and prognosis of primary and secondary diffuse large B-cell lymphoma of the pancreas. Zhonghua Xue Ye Xue Za Zhi. (2023) . 44:55–61. doi: 10.3760/cma.j.issn.0253-2727.2023.01.010

PubMed Abstract | Crossref Full Text | Google Scholar

29. Lee WJ, Won KH, Won CH, Chang SE, Choi JH, Moon KC, et al. Secondary cutaneous diffuse large B-cell lymphoma has a higher international prognostic index score and worse prognosis than diffuse large B-cell lymphoma, leg type. Acta Derm Venereol. (2016) . 96:245–50. doi: 10.2340/00015555-2139

PubMed Abstract | Crossref Full Text | Google Scholar