Development and validation of a nomogram based on three-dimensional quantitative parameters from dual-layer detector spectral computed tomography for predicting treatment response to neoadjuvant chemotherapy in pancreatic ductal adenocarcinoma

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is an aggressive malignancy with a high annual case fatality rate, with a 5-year relative survival rate of 10% (1). Owing to the nonspecific clinical manifestations and the early involvement of major blood vessels, only approximately 10–20% of patients are eligible for surgical intervention (2,3). Neoadjuvant chemotherapy (NAC) can not only increase the rate of a negative margin resection, but it may also potentially control or eradicate occult systemic disease and may increase the disease-free survival and the overall survival (4,5). However, not all patients respond positively to NAC in clinical practice. Some patients may lack an objective response, develop local or biochemical progression, or develop significant treatment-related toxicities that hinder the continuation of chemotherapy regimens (6,7). As a result, an accurate prediction of the NAC response in PDAC patients has substantial therapeutic and prognostic significance.

In clinical practice, the response of NAC for PDAC is typically assessed solely by changes in imaging or biomarkers such as carbohydrate antigen 19-9 (CA19-9) measured before and after NAC, which limits the utility of these assessments for guiding clinical decision-making prior to treatment initiation (8). Previous studies have reported that functional parameters derived from magnetic resonance imaging (MRI)-based extracellular volume (ECV) and Fluorodeoxyglucose positron emission tomography (PET) are useful for predicting treatment response prior to therapy initiation, with AUCs of 0.918 and 0.876, respectively (9,10). However, the clinical applicability of MRI-based ECV is limited because MRI is not a routine preoperative examination for PDAC, as it cannot clearly delineate tumor involvement with surrounding vessels. Similarly, PET has drawbacks including high radiation exposure, limited availability, high cost, and a relatively slow metabolic response to therapy. Solely dependent on attenuation values, conventional computed tomography (CT) imaging poorly predicts treatment response, as attenuation value measurements are highly variable depending on scanner settings and patient characteristics (11). Therefore, there is an urgent need for an objective and accurate method to predict treatment response of NAC to improve clinical decision-making.

Dual-layer detector spectral computed tomography (DLCT) overcomes the limitations of conventional CT by providing quantification of material-specific parameters such as iodine concentration (IC) and the effective atomic number (Zeff), thereby offering indirect, noninvasive information on tumor vascularization, cellular density, and metabolic status (12). Fractional ECV measured from equilibrium-phase contrast-enhanced CT correlates strongly with the amount of desmoplastic stroma, which is abundant in PDAC and critically drives tumor progression, metastasis, and chemoresistance (13-16). In contrast to SECT, which needs both pre- and post-contrast scans (increasing radiation dose and requiring registration), DLCT computes ECV from equilibrium-phase iodine density maps alone. Emerging evidence indicates that ECV derived from DLCT correlates with treatment response to NAC in PDAC, yielding an AUC of 0.798 (17). However, their study was confined to evaluating only the ECV parameter derived from DLCT to predict NAC response in PDAC, without incorporating other DLCT-based quantitative parameters or clinical features. Furthermore, in most prior studies, the ROI was drawn on the largest lesion slices, leading to potential measurement bias and overlooking information about the entire lesion’s tissue. In contrast, three-dimensional (3D) volumetric analysis of the entire tumor provides a more comprehensive and reproducible assessment of tumor heterogeneity (18,19). 3D quantitative parameters from DLCT showed predictive potential for Ki-67 proliferation index in PDAC (20). Current research on the combined evaluation of DLCT 3D parameters with other clinical indicators for predicting treatment response of NAC in PDAC remains relatively limited.

Therefore, the study aimed to incorporate DLCT-based 3D quantitative parameters and clinical features to predict the NAC response in PDAC patients, which could early verify patients who may benefit from NAC. We present this article in accordance with the TRIPOD reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-1-0194/rc).

Methods

Patients

This study obtained ethical approval from the Institutional Review Board of Chongqing General Hospital (No. KY S2024-069-01), and patient consent was waived for this retrospective study. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This was a retrospective prediction model development and validation study conducted at a single tertiary referral center. This study enrolled patients who had been pathologically diagnosed with PDAC between December 2019 and July 2025. The inclusion criteria were as follows: (I) histopathologically confirmed PDAC; (II) no prior treatment before initiation of NAC, and (III) the completion of 2–3 cycles of NAC. The exclusion criteria were as follows: (I) absence of baseline DLCT; (II) concurrent presence of other primary malignancies; (III) absence of follow-up imaging during the 8–12-week period after the start of treatment; or (IV) poor image quality precluding accurate evaluation. Finally, 150 patients were included in the study (Figure 1).

Figure 1 Flowchart of patient selection. DLCT, dual-layer detector spectral computed tomography; NAC, neoadjuvant chemotherapy; PDAC, pancreatic ductal adenocarcinoma.

Clinical data collection and NAC procedure

Clinical data and serum markers were collected from the hospital’s electronic medical records system for all patients. The collected data included age, gender, body mass index, carcinoembryonic antigen (CEA), CA19-9, carbohydrate antigen 125, albumin (ALB), globulin, total bilirubin, platelet-to-lymphocyte ratio, lymphocyte-to-monocyte ratio, neutrophil-to-lymphocyte ratio, and platelet-white blood cell ratio.

The NAC treatments administered included gemcitabine plus nabpaclitaxel, gemcitabine plus tegafur/gimeracil/oteracil, and modified FOLFIRINOX (mFFX, fluorouracil, leucovorin, irinotecan, and oxaliplatin). Each regimen was based on a previously reported protocol (21,22). After initiating NAC, contrast-enhanced CT was performed after 2–3 months.

Image acquisition

All patients acquired unenhanced and contrast-enhanced abdominal images using the DLCT (IQon spectral CT, Philips Healthcare, The Netherlands) after fasting for at least 4 h in this study. The following acquisition parameters were conducted for scanning from the diaphragmatic dome to the inferior liver border: tube voltage, 120 kVp; tube current, automatic modulation; detector collimation, 64×0.625; helical pitch, 0.798; and rotation time, 0.5 s. A non-ionic contrast agent (Ultravist 370) with a 1.5 mL/kg dose was injected at a flow rate of 3–3.5 mL/s through a peripheral vein of the forearm or a central line. The bolus tracking technique enabled initiation of arterial phase (AP) imaging with a 12 s delay after the descending aorta reached a threshold of 150 HU. Portal venous phase (PVP) scans commenced 23 s following the AP, and equilibrium phases (EP) were obtained at 180 s post-contrast injection.

Image analysis

Acquired DLCT images were transferred to prototype software (IntelliSpace Discovery, Philips Healthcare, The Netherlands) and reconstructed to generate virtual monoenergetic images at 40 and 100 keV, as well as Zeff maps, IC maps, and electron density (ED) maps for subsequent quantitative analysis. Tumor contours were manually delineated slice-by-slice on the AP, PVP, and EP images using ITK-SNAP software (v.3.8) by two radiologists (R1 with 7 years and R2 with over 11 years of experience in CT interpretation), carefully avoiding adjacent vasculature, necrotic regions, and pancreatic ducts. A schematic diagram of the volume of interest (VOI) placement and measurement was shown in Figure 2. Then the software automatically yielded the following DLCT-based 3D quantitative parameters: 3D-40 keV_AP, 3D-40 keV_PVP, 3D-100 keV_AP, 3D-100 keV_PVP, 3D-IC_AP, 3D-IC_PVP, 3D-Zeff_AP, 3D-Zeff_PVP, 3D-ED_AP, 3D-ED_PVP and 3D-IC_EP. The formulas for quantitative parameter calculation are displayed in Table 1. Inter-observer agreement for the DLCT-derived 3D quantitative parameters obtained by two radiologists was assessed using intraclass correlation coefficient analysis.

Figure 2 Schematic diagram of delineating regions of interest.

Tumor response evaluation

The baseline and follow-up (8–12 weeks) CT images were independently interpreted by two abdominal radiologists (R1, with 7 years of experience; R2, with over 11 years). They were informed of the pathological diagnosis of PDAC for all patients but remained blinded to clinical, laboratory, and follow-up information. The reviewers assessed treatment response by comparing baseline CT images with those obtained at 8–12-week follow-up. Based on RECIST version 1.1 criteria, responses were categorized as complete remission (CR), partial response (PR), stable disease (SD), or progressive disease (PD) (23). Complete response (CR) was defined as disappearance of all primary tumor. PR was defined as a decrease in the largest dimension of the primary tumor by at least 30%. PD was defined as the presence of metastatic lesions or an increase of 20% in the primary tumor’s largest dimension (with a minimum increase of 5 mm). SD was defined as neither sufficient shrinkage to qualify for PR nor sufficient increase to qualify for PD, taking as reference the smallest sum diameters during the study. Patients achieving CR or partial PR were classified into the response group, whereas those with SD or PD were assigned to the non-response group. Discrepancies between reviewers were resolved through a consensus review, with the final consensus data used for analysis.

Predictive models development and evaluation

Univariate analysis was initially conducted in the training set to compare the differences in DLCT-based 3D quantitative parameters and clinical features between the response group and the non-response group. Independent predictors were identified through multivariate logistic regression analysis, which incorporated variables found to be significant (P<0.05) in the univariate logistic regression analysis. A clinical model, a DLCT model, and a combined nomogram were subsequently developed via logistic regression based on the identified predictors. Comparisons of the area under the curve (AUC) between models were performed using the DeLong test. The Hosmer-Lemeshow test and calibration curves were employed to assess calibration performance, and decision curve analysis (DCA) was utilized to evaluate the model’s clinical utility.

Statistical analysis

All statistical analyses were performed using R (version 4.3.0), SPSS (version 26.0), and MedCalc (version 18.2.1). Continuous variables following a normal distribution were presented as mean ± standard deviation, while those not following a normal distribution were expressed as median [interquartile range (IQR)]. Comparison of continuous data between two independent samples was performed using the independent samples t-test or Mann-Whitney U test. The chi-square test was used for categorical variables. Variables with significant differences in univariate analyses were included in backward stepwise binary logistic regression analyses to determine independent predictors of treatment response in patients with PDAC. The variance inflation factor (VIF) was used to check the multi-collinearity, which existed when VIF >5. All statistical tests were two-sided, and a P value <0.05 was considered statistically significant.

Results

Clinical features and DLCT-based 3D quantitative parameters

All 150 PDAC patients were randomized in a 7:3 ratio into a training set (n=105, 56 men and 49 women, with a mean age of 60.87±9.97 years) and a validation set (n=45, 28 men and 17 women, with a mean age of 58.93±9.58 years) using simple randomization. After randomization, the response rates were 34.3% (36/105) in the training set and 35.6% (16/45) in the validation set, with no statistically significant difference between the two groups (P=0.88). Chemotherapy response rates were observed to be 34.3% and 35.6% in the training and validation sets, respectively. The DLCT-based 3D quantitative parameters and clinical features of PDAC in the training and validation sets are listed in Table 2. These parameters and features in the responder and non-responder groups are displayed in Table 3. The response group showed higher ALB and 3D-nZeff_PVP, and lower CA19-9, CEA, and ECV values compared to the non-response group in both sets (P<0.05). Additionally, the response group had higher 3D-nIC_AP and AEF in the training set, while the response group had lower CA125 in the validation set (P<0.05). No significant differences were detected between the two groups regarding other clinical characteristics and CT parameters (P>0.05).

Variables associated with response to NAC

Univariable logistic regression analysis in the training set showed that ALB [odds ratio (OR), 1.142; 95% confidence interval (CI): 1.027–1.270; P=0.01], CA19-9 (OR, 0.999; 95% CI: 0.998–1.000; P=0.003), 3D-nIC_AP (OR, 1.226; 95% CI: 1.080–1.484; P=0.004), 3D-nZeff_PVP (OR, 1.166; 95% CI: 1.019–1.335; P=0.03), AEF (OR, 1.009; 95% CI: 1.001–1.016; P=0.03), and ECV (OR, 0.961; 95% CI: 0.924–0.999; P<0.05) were significantly correlated with response to NAC. The analysis showed no significant collinearity among the all-independent predictors with all VIF values <3 (Table S1). Variables with P<0.05 on univariable analysis were entered into multivariable logistic regression analysis. The results demonstrated that CA19-9 (OR, 0.999; 95% CI: 0.998–1.000; P=0.006), 3D-nZeff_PVP (OR, 1.497; 95% CI: 1.173–1.910; P=0.001), and ECV (OR, 0.896; 95% CI: 0.837–0.959; P=0.002) were significant independent factors of NAC response (Table 4).

Construction and evaluation of the models

Three predictive models for NAC response in patients with PDAC were constructed using the identified independent predictors. The clinical model included CA19-9, the DLCT model included 3D-nZeff_VP and ECV, and a nomogram was then generated by integrating all three independent predictors (Figure 3). Table 5 and Figure 4 summarize the performance of the three models. The clinical model showed AUCs of 0.782 (95% CI: 0.691–0.857) and 0.735 (95% CI: 0.582–0.855) for predicting NAC response in the training and validation sets, respectively. The DLCT model produced AUCs of 0.775 (95% CI: 0.684–0.851) and 0.784 (95% CI: 0.637–0.893), respectively. The nomogram exhibited the highest predictive performance, yielding AUCs of 0.839 (95% CI: 0.755–0.904) and 0.849 (95% CI: 0.711–0.938) for the training and validation sets, respectively. When using the optimal cutoff value of the nomogram (<46.8%) in the training set, the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for predicting the response group were 69.4% (95% CI: 51.9–83.7%), 87.0% (95% CI: 76.7–93.9%), 73.5% (95% CI: 59.3–84.1%), and 84.5% (95% CI: 76.8–90.0%), respectively. In the validation set, using the optimal cutoff value of <29.9%, the corresponding values were 93.8% (95% CI: 69.8–99.8%), 69.0% (95% CI: 49.2–84.7%), 62.5% (95% CI: 48.8–74.4%), and 95.2% (95% CI: 74.7–99.2%). Calibration curves indicated satisfactory calibration performance for all models (Figure 5A,5B). DCA demonstrated that the nomogram provided superior net benefit compared to the clinical and DLCT models across most threshold probability ranges (Figure 5C,5D). Figure 6 presents representative images from one responder and one non-responder.

Figure 3 The nomogram integrating the CA19-9, nZeff_PVP, and ECV was developed in the training set to predict NAC response in PDAC patients. CA19-9, carbohydrate antigen 19-9; ECV, extracellular volume; NAC, neoadjuvant chemotherapy; nZeff_PVP, normalized effective atomic number in the portal venous phase; PDAC, pancreatic ductal adenocarcinoma.

Figure 4 ROC curve depicting the predictive performance of the clinical model, DLCT model, and nomogram for predicting NAC response in PDAC in the training set (A) and the validation set (B). AUC, area under the curve; DLCT, dual-layer detector spectral computed tomography; NAC, neoadjuvant chemotherapy; PDAC, pancreatic ductal adenocarcinoma; ROC, receiver operating characteristic.

Figure 5 Calibration curves (A,B) and DCA (C,D) for the clinical model, DLCT model, and nomogram. DCA, decision curve analysis; DLCT, dual-layer detector spectral computed tomography.

Figure 6 Representative images from one responder and one non-responder. (A) A 79-year-old male with PDAC who received NAC was assessed as having a partial response and was assigned to the responder group. (a-c) PVP imaging reveals a 3D-VOI. (d) The PVP effective atomic number map shows 3D-nZeff value of 0.84. (B) A 61-year-old female with PDAC who received NAC was assessed as having a stable disease and was assigned to the non-response group. (a-c) PVP imaging reveals a 3D-VOI. (d) The PVP effective atomic number map shows 3D-nZeff value of 0.78. 3D, three-dimensional; NAC, neoadjuvant chemotherapy; nZeff, normalized effective atomic number; PDAC, pancreatic ductal adenocarcinoma; PVP, portal venous phase; VOI, volume of interest.

Discussion

In this retrospective study, we developed and evaluated a nomogram based on clinical and DLCT-based 3D quantitative parameters to predict the NAC responses in PDAC patients. By integrating 3D-nZeffPVP, ECV, and CA19-9, the nomogram exhibited superior predictive performance. By integrating easily accessible clinical and imaging parameters, this model enables accurate identification of PDAC patients suitable for NAC and facilitates informed therapeutic decisions. The nomogram enables risk stratification for inadequate NAC response prior to the start of therapy. For patients identified as non-response groups, individualized therapeutic strategies may be formulated, including upfront surgical management or a comprehensive multimodal approach that combines chemotherapy with immunotherapy, radiotherapy, or targeted agents, thereby enabling the optimization of personalized treatment protocols.

Our study found that the clinical feature CA19-9 could predict the NAC responses in PDAC. The response group demonstrated lower CA19-9 levels in comparison to the non-response group. Serum CA19-9 is widely recognized as a crucial tumor marker in pancreatic cancer and is routinely employed in clinical practice. The diagnostic, prognostic, and predictive roles of CA19-9 have been extensively studied and validated (24,25). Numerous studies have consistently reported the prognostic significance of CA19-9 in nonmetastatic PDAC, with widespread agreement that it is an effective indicator of treatment response (8,26). Shi et al. found that higher CA19-9 levels are associated with worse biologic aggressiveness (27). These findings underscore the critical importance of incorporating biomarkers such as CA19-9 into the preoperative evaluation and stratification of PDAC patients undergoing NAC.

Zeff characterizes both the atomic makeup of inorganic components and intratumoral heterogeneity, thereby exerting a key role in tumor identification and prediction of metastatic potential in malignancies (28,29). We observed that 3D-nZeff was significantly elevated in treatment responders relative to non-responders; this phenomenon may be explained by enhanced tumor heterogeneity and the abundance of high-atomic-number constituents. The present study revealed a significant difference in nZeff exclusively during the PVP, potentially attributable to enhanced tumor angiogenesis among responders and the resultant greater influence of iodine-based contrast on the Zeff of tumor tissue. Pre-chemotherapy 3D-nZeff_VP could be important for predicting treatment response.

We also found that PDAC patients with lower ECV values tended to achieve a more favorable response to NAC than those with higher ECV values. Potential mechanisms may involve an abundance of fibroblasts—especially cancer-associated fibroblasts—in the tumor microenvironment, potentially contributing to tumor advancement and chemoresistance (30-32). The desmoplastic stroma, robustly correlated with ECV, has been implicated in facilitating cancer progression and chemotherapy resistance. Additionally, remodeling of the extracellular matrix can lead to tumor hypoxia, which may impede drug delivery and reduce treatment effectiveness, as evidenced in studies of colon cancer (33). The optimal time point for iodine quantification in ECV assessment has not been fully established. In the current study, equilibrium-phase imaging was acquired at 180 s, whereas Fujita et al. adopted a delay time of 240 s (17). The use of late-delay phases may disrupt routine clinical workflow. The technical feasibility of using 180-s EP imaging to predict PDAC prognosis via ECV fraction has been reported in multiple studies (15,34,35). Accordingly, our proposed method holds substantial clinical utility, since additional quantification of the ECV fraction can be readily accomplished within the framework of routine clinical imaging protocols.

To achieve a more precise evaluation of Zeff and ECV and their association with NAC response, we utilized 3D quantitative analysis. 3D analysis offers superior precision and reproducibility relative to traditional two-dimensional (2D) methods, reducing Zeff and ECV variability and enhancing the ability to predict clinical outcomes (36). In contrast to 2D analysis, which samples only a portion of the tumor, 3D analysis encompasses the entire lesion, enabling more comprehensive evaluation and reducing observer bias (37,38). Consequently, studies employing 3D analysis are more prone to generate accurate and reproducible findings than those relying on conventional 2D methods.

Limitations

This study had several limitations. Firstly, this single-center retrospective study has a relatively limited sample size and, more critically, lacks an independent external validation cohort. Although internal validation via random splitting provides initial evidence of model stability, it does not guarantee generalizability to other institutions, diverse geographic regions, or CT scanners of different types. Further investigations across multiple centers are necessary to establish the external validity of our findings. Secondly, the measurement of ECV using the EP at 180 s postcontrast administration may not allow sufficient time for contrast equilibration. However, nevertheless, extending image acquisition to the late EP could disrupt standard clinical practice, and the required scan delay for accurate ECV measurement remains to be determined. Thirdly, this study did not address patient survival, which represents an important direction for subsequent investigations. Additionally, this study relied solely on tumor size changes to distinguish responders from non-responders, which limits the assessment of minor responses or symptomatic progression without radiographic tumor progression. Future investigations should address this issue in greater depth.

Conclusions

In conclusion, this study established a non-invasive, effective nomogram incorporating DLCT-based 3D quantitative parameters and clinical features, which exhibited favorable predictive performance for NAC response in patients with PDAC. Following large-scale, multicenter external validation, the nomogram may serve as adjunctive tools to inform clinical decision-making and risk stratification.

Acknowledgments

The authors thank all volunteers who participated in the study and the staff of the Department of Radiology, Chongqing General Hospital, China, for their selfless and valuable assistance.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-1-0194/rc

Data Sharing Statement: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-1-0194/dss

Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-1-0194/prf

Funding: This study was supported by the Medical Research Program of the Combination of Chongqing National Health Commission and Chongqing Science and Technology Bureau, China (No. 2024MSXM094), Chongqing Natural Science Foundation General Program (No. CSTB2025NSCQ-GPX0311), and the Key Special Program of Technological Innovation and Application Development in Chongqing, China (No. CSTB2023TIAD-KPX0059-2).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-1-0194/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Ethical approval for this study was granted by the Institutional Review Board of Chongqing General Hospital (No. KY S2024-069-01), which waived the requirement for informed consent.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Park W, Chawla A, O’Reilly EM. Pancreatic Cancer: A Review. JAMA 2021;326:851-62. [Crossref] [PubMed]
2. Kleeff J, Korc M, Apte M, et al. Pancreatic cancer. Nat Rev Dis Primers 2016;2:16022. [Crossref] [PubMed]
3. Conlon KC, Klimstra DS, Brennan MF. Long-term survival after curative resection for pancreatic ductal adenocarcinoma. Clinicopathologic analysis of 5-year survivors. Ann Surg 1996;223:273-9.
4. Versteijne E, van Dam JL, Suker M, et al. Neoadjuvant Chemoradiotherapy Versus Upfront Surgery for Resectable and Borderline Resectable Pancreatic Cancer: Long-Term Results of the Dutch Randomized PREOPANC Trial. J Clin Oncol 2022;40:1220-30. [Crossref] [PubMed]
5. Hu Q, Wang D, Chen Y, et al. Network meta-analysis comparing neoadjuvant chemoradiation, neoadjuvant chemotherapy and upfront surgery in patients with resectable, borderline resectable, and locally advanced pancreatic ductal adenocarcinoma. Radiat Oncol 2019;14:120. [Crossref] [PubMed]
6. Sherman MH, Beatty GL. Tumor Microenvironment in Pancreatic Cancer Pathogenesis and Therapeutic Resistance. Annu Rev Pathol 2023;18:123-48. [Crossref] [PubMed]
7. Sohal DPS, Duong M, Ahmad SA, et al. Efficacy of Perioperative Chemotherapy for Resectable Pancreatic Adenocarcinoma: A Phase 2 Randomized Clinical Trial. JAMA Oncol 2021;7:421-7. [Crossref] [PubMed]
8. Kim SS, Lee S, Lee HS, et al. Retrospective Evaluation of Treatment Response in Patients with Nonmetastatic Pancreatic Cancer Using CT and CA 19-9. Radiology 2022;303:548-56. [Crossref] [PubMed]
9. Fukukura Y, Kumagae Y, Fujisaki Y, et al. Extracellular volume fraction with MRI: As an alternative predictive biomarker to dynamic contrast-enhanced MRI for chemotherapy response of pancreatic ductal adenocarcinoma. Eur J Radiol 2021;145:110036. [Crossref] [PubMed]
10. Abdelrahman AM, Goenka AH, Alva-Ruiz R, et al. FDG-PET Predicts Neoadjuvant Therapy Response and Survival in Borderline Resectable/Locally Advanced Pancreatic Adenocarcinoma. J Natl Compr Canc Netw 2022;20:1023-1032.e3. [Crossref] [PubMed]
11. Zhang H, Capaldi D, Zeng D, et al. Prior-image-based CT reconstruction using attenuation-mismatched priors. Phys Med Biol 2021;66:064007. [Crossref] [PubMed]
12. Rassouli N, Etesami M, Dhanantwari A, et al. Detector-based spectral CT with a novel dual-layer technology: principles and applications. Insights Imaging 2017;8:589-98. [Crossref] [PubMed]
13. Sofue K, Tsurusaki M, Mileto A, et al. Dual-energy computed tomography for non-invasive staging of liver fibrosis: Accuracy of iodine density measurements from contrast-enhanced data. Hepatol Res 2018;48:1008-19. [Crossref] [PubMed]
14. Zissen MH, Wang ZJ, Yee J, et al. Contrast-enhanced CT quantification of the hepatic fractional extracellular space: correlation with diffuse liver disease severity. AJR Am J Roentgenol 2013;201:1204-10. [Crossref] [PubMed]
15. Fukukura Y, Kumagae Y, Higashi R, et al. Extracellular volume fraction determined by equilibrium contrast-enhanced multidetector computed tomography as a prognostic factor in unresectable pancreatic adenocarcinoma treated with chemotherapy. Eur Radiol 2019;29:353-61. [Crossref] [PubMed]
16. Carvalho TMA, Di Molfetta D, Greco MR, et al. Tumor Microenvironment Features and Chemoresistance in Pancreatic Ductal Adenocarcinoma: Insights into Targeting Physicochemical Barriers and Metabolism as Therapeutic Approaches. Cancers (Basel) 2021;13:6135. [Crossref] [PubMed]
17. Fujita N, Ushijima Y, Itoyama M, et al. Extracellular volume fraction determined by dual-layer spectral detector CT: Possible role in predicting the efficacy of preoperative neoadjuvant chemotherapy in pancreatic ductal adenocarcinoma. Eur J Radiol 2023;162:110756. [Crossref] [PubMed]
18. Doi S, Yanagawa M, Matsui T, et al. Usefulness of Three-Dimensional Iodine Mapping Quantified by Dual-Energy CT for Differentiating Thymic Epithelial Tumors. J Clin Med 2023;12:5610. [Crossref] [PubMed]
19. Gierada DS, Rydzak CE, Zei M, et al. Improved Interobserver Agreement on Lung-RADS Classification of Solid Nodules Using Semiautomated CT Volumetry. Radiology 2020;297:675-84. [Crossref] [PubMed]
20. Zeng D, Zhang J, Song Z, et al. Development and validation of a model based on preoperative dual-layer detector spectral computed tomography 3D VOI-based quantitative parameters to predict high Ki-67 proliferation index in pancreatic ductal adenocarcinoma. Insights Imaging 2024;15:291. [Crossref] [PubMed]
21. Motoi F, Kosuge T, Ueno H, et al. Randomized phase II/III trial of neoadjuvant chemotherapy with gemcitabine and S-1 versus upfront surgery for resectable pancreatic cancer (Prep-02/JSAP05). Jpn J Clin Oncol 2019;49:190-4. [Crossref] [PubMed]
22. Ozaka M, Ishii H, Sato T, et al. A phase II study of modified FOLFIRINOX for chemotherapy-naïve patients with metastatic pancreatic cancer. Cancer Chemother Pharmacol 2018;81:1017-23. [Crossref] [PubMed]
23. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009;45:228-47. [Crossref] [PubMed]
24. Distler M, Pilarsky E, Kersting S, et al. Preoperative CEA and CA 19-9 are prognostic markers for survival after curative resection for ductal adenocarcinoma of the pancreas - a retrospective tumor marker prognostic study. Int J Surg 2013;11:1067-72. [Crossref] [PubMed]
25. Sugiura T, Uesaka K, Kanemoto H, et al. Serum CA19-9 is a significant predictor among preoperative parameters for early recurrence after resection of pancreatic adenocarcinoma. J Gastrointest Surg 2012;16:977-85. [Crossref] [PubMed]
26. Tsai S, George B, Wittmann D, et al. Importance of Normalization of CA19-9 Levels Following Neoadjuvant Therapy in Patients With Localized Pancreatic Cancer. Ann Surg 2020;271:740-7. [Crossref] [PubMed]
27. Shi S, Yu X. Selecting chemotherapy for pancreatic cancer: Far away or so close? Semin Oncol 2019;46:39-47. [Crossref] [PubMed]
28. Wu J, Lv Y, Wang N, et al. The value of single-source dual-energy CT imaging for discriminating microsatellite instability from microsatellite stability human colorectal cancer. Eur Radiol 2019;29:3782-90. [Crossref] [PubMed]
29. Chen WB, Shi QQ, Li ZM, et al. Diagnostic value of spiral CT energy spectrum imaging in lymph node metastasis of colorectal cancer. Int J Colorectal Dis 2022;37:2021-9. [Crossref] [PubMed]
30. Ren H, Huang J, Liu D, et al. Evaluation of induction chemotherapy response using multiphasic dual-energy CT in nasopharyngeal carcinoma. Eur Radiol 2025;35:5182-92. [Crossref] [PubMed]
31. Yang C, Zhang Y, Yan M, et al. Exosomes derived from cancer-associated fibroblasts promote tumorigenesis, metastasis and chemoresistance of colorectal cancer by upregulating circ_0067557 to target Lin28. BMC Cancer 2024;24:64. [Crossref] [PubMed]
32. Huang R, Wu D, Zhang K, et al. ARID1A loss induces P4HB to activate fibroblasts to support lung cancer cell growth, invasion, and chemoresistance. Cancer Sci 2024;115:439-51. [Crossref] [PubMed]
33. Malier M, Gharzeddine K, Laverriere MH, et al. Hypoxia Drives Dihydropyrimidine Dehydrogenase Expression in Macrophages and Confers Chemoresistance in Colorectal Cancer. Cancer Res 2021;81:5963-76. [Crossref] [PubMed]
34. Fukukura Y, Kumagae Y, Higashi R, et al. Extracellular volume fraction determined by equilibrium contrast-enhanced dual-energy CT as a prognostic factor in patients with stage IV pancreatic ductal adenocarcinoma. Eur Radiol 2020;30:1679-89. [Crossref] [PubMed]
35. Fukukura Y, Kumagae Y, Higashi R, et al. Estimation of Extracellular Volume Fraction With Routine Multiphasic Pancreatic Computed Tomography to Predict the Survival of Patients With Stage IV Pancreatic Ductal Adenocarcinoma. Pancreas 2019;48:1360-6. [Crossref] [PubMed]
36. Watzenboeck ML, Heidinger BH, Rainer J, et al. Reproducibility of 2D versus 3D radiomics for quantitative assessment of fetal lung development: a retrospective fetal MRI study. Insights Imaging 2023;14:31. [Crossref] [PubMed]
37. Liu W, Rong D, Zhu J, et al. Diagnostic accuracy of 3D magnetic resonance elastography for assessing histologic grade of hepatocellular carcinoma: comparison of three methods for positioning region of interest. Abdom Radiol (NY) 2021;46:4601-9. [Crossref] [PubMed]
38. Zhong LJ, Yu N, Zhou XJ, et al. Differentiating between pulmonary adenocarcinoma and squamous cell carcinoma by spectral CT volumetric quantitative analysis: a comparative study with conventional spectral analysis. J Thorac Dis 2023;15:679-89. [Crossref] [PubMed]