Factors associated with missed and misinterpreted cases of pancreatic ductal adenocarcinoma

Pancreatic Cancer Surveillance and Survival of High-Risk Individuals

Importance

Pancreatic ductal adenocarcinoma (PDAC) is a deadly disease with increasing incidence. The majority of PDACs are incurable at presentation, but population-based screening is not recommended. Surveillance of high-risk individuals for PDAC may lead to early detection, but the survival benefit is unproven.

Objective

To compare the survival of patients with surveillance-detected PDAC with US national data.

Design, Setting, and Participants

This comparative cohort study was conducted in multiple US academic medical centers participating in the Cancer of the Pancreas Screening program, which screens high-risk individuals with a familial or genetic predisposition for PDAC. The comparison cohort comprised patients with PDAC matched for age, sex, and year of diagnosis from the Surveillance, Epidemiology, and End Results (SEER) program. The Cancer of the Pancreas Screening program originated in 1998, and data collection was done through 2021. The data analysis was performed from April 29, 2022, through April 10, 2023.

Exposures

Endoscopic ultrasonography or magnetic resonance imaging performed annually and standard-of-care surgical and/or oncologic treatment.

Main Outcomes and Measures

Stage of PDAC at diagnosis, overall survival (OS), and PDAC mortality were compared using descriptive statistics and conditional logistic regression, Cox proportional hazards regression, and competing risk regression models. Sensitivity analyses and adjustment for lead-time bias were also conducted.

Results

A total of 26 high-risk individuals (mean [SD] age at diagnosis, 65.8 [9.5] years; 15 female [57.7%]) with PDAC were compared with 1504 SEER control patients with PDAC (mean [SD] age at diagnosis, 66.8 [7.9] years; 771 female [51.3%]). The median primary tumor diameter of the 26 high-risk individuals was smaller than in the control patients (2.5 [range, 0.6-5.0] vs 3.6 [range, 0.2-8.0] cm, respectively; P < .001). The high-risk individuals were more likely to be diagnosed with a lower stage (stage I, 10 [38.5%]; stage II, 8 [30.8%]) than matched control patients (stage I, 155 [10.3%]; stage II, 377 [25.1%]; P < .001). The PDAC mortality rate at 5 years was lower for high-risk individuals than control patients (43% vs 86%; hazard ratio, 3.58; 95% CI, 2.01-6.39; P < .001), and high-risk individuals lived longer than matched control patients (median OS, 61.7 [range, 1.9-147.3] vs 8.0 [range, 1.0-131.0] months; 5-year OS rate, 50% [95% CI, 32%-80%] vs 9% [95% CI, 7%-11%]).

Conclusions and Relevance

These findings suggest that surveillance of high-risk individuals may lead to detection of smaller, lower-stage PDACs and improved survival.

Factors associated with radiological misstaging of pancreatic ductal adenocarcinoma: A retrospective observational study

PURPOSE

Accurate staging of disease is vital in determining appropriate care for patients with pancreatic ductal adenocarcinoma (PDAC). It has been shown that the quality of scans and the experience of a radiologist can impact computed tomography (CT) based assessment of disease. The aim of the current study was to evaluate the impact of the rereading of outside hospital (OH) CT by an expert radiologist and a repeat pancreatic protocol CT (PPCT) on staging of disease.

METHODS

Patients evaluated at the our institute's pancreatic multidisciplinary clinic (2006 to 2014) with OH scan and repeat PPCT performed within 30 days were included. In-house radiologists staged disease using OH scans and repeat PPCT, and factors associated with misstaging were determined.

RESULTS

The study included 100 patients, with a median time between OH scan and PPCT of 19 days (IQR: 13-23 days.) Stage migration was mostly accounted for by upstaging of disease (58.8 % to 83.3 %) in all comparison groups. When OH scans were rereviewed, 21.5 % of the misstaging was due to missed metastases, however, when rereads were compared to the PPCT, occult metastases accounted for the majority of misstaged patients (62.5 %). Potential factors associated with misstaging were primarily related to imaging technique.

CONCLUSION

A repeat PPCT results in increased detection of metastatic disease that rereviews of OH scans may otherwise miss. Accessible insurance coverage for repeat PPCT imaging even within 30 days of an OH scan could help optimize delivery of care and alleviate burdens associated with misstaging.

Improved Pancreatic Cancer Detection and Localization on CT Scans: A Computer-Aided Detection Model Utilizing Secondary Features

The early detection of pancreatic ductal adenocarcinoma (PDAC) is essential for optimal treatment of pancreatic cancer patients. We propose a tumor detection framework to improve the detection of pancreatic head tumors on CT scans. In this retrospective research study, CT images of 99 patients with pancreatic head cancer and 98 control cases from the Catharina Hospital Eindhoven were collected. A multi-stage 3D U-Net-based approach was used for PDAC detection including clinically significant secondary features such as pancreatic duct and common bile duct dilation. The developed algorithm was evaluated using a local test set comprising 59 CT scans. The model was externally validated in 28 pancreatic cancer cases of a publicly available medical decathlon dataset. The tumor detection framework achieved a sensitivity of 0.97 and a specificity of 1.00, with an area under the receiver operating curve (AUROC) of 0.99, in detecting pancreatic head cancer in the local test set. In the external test set, we obtained similar results, with a sensitivity of 1.00. The model provided the tumor location with acceptable accuracy obtaining a DICE Similarity Coefficient (DSC) of 0.37. This study shows that a tumor detection framework utilizing CT scans and secondary signs of pancreatic cancer can detect pancreatic tumors with high accuracy.

Imaging of pancreatic ductal adenocarcinoma - An update for all stages of patient management

Background

Pancreatic ductal adenocarcinoma (PDAC) is a common and lethal cancer. From diagnosis to disease staging, response to neoadjuvant therapy assessment and patient surveillance after resection, imaging plays a central role, guiding the multidisciplinary team in decision-planning.

Review aims and findings

This review discusses the most up-to-date imaging recommendations, typical and atypical findings, and issues related to each step of patient management. Example cases for each relevant condition are presented, and a structured report for disease staging is suggested.

Conclusion

Despite current issues in PDAC imaging at different stages of patient management, the radiologist is essential in the multidisciplinary team, as the conveyor of relevant imaging findings crucial for patient care.

Hidden in plain sight: commonly missed early signs of pancreatic cancer on CT

Pancreatic ductal adenocarcinoma (PDAC) has poor prognosis mostly due to the advanced stage at which disease is diagnosed. Early detection of disease at a resectable stage is, therefore, critical for improving outcomes of patients. Prior studies have demonstrated that pancreatic abnormalities may be detected on CT in up to 38% of CT studies 5 years before clinical diagnosis of PDAC. In this review, we highlight commonly missed signs of early PDAC on CT. Broadly, these commonly missed signs consist of small isoattenuating PDAC without contour deformity, isolated pancreatic duct dilatation and cutoff, focal pancreatic enhancement and focal parenchymal atrophy, pancreatitis with underlying PDAC, and vascular encasement. Through providing commentary on demonstrative examples of these signs, we demonstrate how to reduce the risk of missing or misinterpreting radiological features of early PDAC.

Pancreatic cancer detection with dual-energy CT: diagnostic performance of 40 keV and 70 keV virtual monoenergetic images

PURPOSE

To compare the diagnostic performance of 40 keV and 70 keV virtual monoenergetic images (VMIs) generated from dual-energy CT in the detection of pancreatic cancer.

METHODS

This retrospective study included patients who underwent pancreatic protocol dual-energy CT from January 2019 to August 2022. Four radiologists (1-11 years of experience), who were blinded to the final diagnosis, independently and randomly interpreted 40 keV and 70 keV VMIs and graded the presence or absence of pancreatic cancer. For each image set (40 keV and 70 keV VMIs), the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated. The diagnostic performance of each image set was compared using generalized estimating equations.

RESULTS

Overall, 137 patients (median age, 71 years; interquartile range, 63-78 years; 77 men) were included. Among them, 62 patients (45%) had pathologically proven pancreatic cancer. The 40 keV VMIs had higher specificity (75% vs. 67%; P < .001), PPV (76% vs. 71%; P < .001), and accuracy (85% vs. 81%; P = .001) than the 70 keV VMIs. On the contrary, 40 keV VMIs had lower sensitivity (96% vs. 98%; P = .02) and NPV (96% vs. 98%; P = .004) than 70 keV VMIs. However, the diagnostic confidence in patients with (P < .001) and without (P = .001) pancreatic cancer was improved in 40 keV VMIs than in 70 keV VMIs.

CONCLUSIONS

The 40 keV VMIs showed better diagnostic performance in diagnosing pancreatic cancer than the 70 keV VMIs, along with higher reader confidence.

Reinterpretation of Hepatopancreaticobiliary Imaging Exams: Assessment of Clinical Impact, Peer Learning, and Physician Satisfaction

OBJECTIVES

To assess the impact on clinical management, potential for peer learning, and referring physician satisfaction with subspecialist reinterpretations of hepatopancreaticobiliary (HPB) imaging examinations.

MATERIALS AND METHODS

HPB CTs and MRIs from outside hospitals were reinterpreted by two subspecialty radiologists between March 2021 and August 2022. Reinterpretation reports were mailed to radiologists that issued primary reports. The electronic record was reviewed to assess for changes in clinical management based on the reinterpretations (yes/no/unavailable). To assess the potential for peer learning, a survey using a 5-point Likert scale was sent to radiologists who issued primary reports. A separate survey was sent to referring physicians to assess satisfaction with reinterpretations.

RESULTS

Two hundred fifty imaging examinations (122 CT, 128 MRI) were reinterpreted at the request of 19 referring physicians. Ninety-six radiologists issued primary reports. RADPEER scores 1-3 were assigned to 131/250 (52%), 86/250 (34%), and 33/250 (13%) examinations, respectively. Of 213 reinterpretations with adequate records for assessment, 75/213 (35%) were associated with a change in management; of these, 71/75 (95%) were classified as RADPEER 2 or 3. Most radiologists agreed or strongly agreed with the following: prefer to receive reinterpretations (34/36, 94%); reinterpretations changed practice of reporting HPB imaging examinations (23/36, 64%); and reinterpretations offer opportunities for peer learning (34/36, 94%). Referring physicians agreed or strongly agreed (7/7, 100%) that reinterpretations are valuable and often change or clarify management of patients with complex HPB disease, and offer an opportunity for peer learning.

CONCLUSION

Radiologists and referring physicians strongly agree that HPB imaging reinterpretations help support peer learning and patient management, respectively.

Assessing the robustness of a machine-learning model for early detection of pancreatic adenocarcinoma (PDA): evaluating resilience to variations in image acquisition and radiomics workflow using image perturbation methods

PURPOSE

To evaluate robustness of a radiomics-based support vector machine (SVM) model for detection of visually occult PDA on pre-diagnostic CTs by simulating common variations in image acquisition and radiomics workflow using image perturbation methods.

METHODS

Eighteen algorithmically generated-perturbations, which simulated variations in image noise levels (σ, 2σ, 3σ, 5σ), image rotation [both CT image and the corresponding pancreas segmentation mask by 45° and 90° in axial plane], voxel resampling (isotropic and anisotropic), gray-level discretization [bin width (BW) 32 and 64)], and pancreas segmentation (sequential erosions by 3, 4, 6, and 8 pixels and dilations by 3, 4, and 6 pixels from the boundary), were introduced to the original (unperturbed) test subset (n = 128; 45 pre-diagnostic CTs, 83 control CTs with normal pancreas). Radiomic features were extracted from pancreas masks of these additional test subsets, and the model's performance was compared vis-a-vis the unperturbed test subset.

RESULTS

The model correctly classified 43 out of 45 pre-diagnostic CTs and 75 out of 83 control CTs in the unperturbed test subset, achieving 92.2% accuracy and 0.98 AUC. Model's performance was unaffected by a three-fold increase in noise level except for sensitivity declining to 80% at 3σ (p = 0.02). Performance remained comparable vis-a-vis the unperturbed test subset despite variations in image rotation (p = 0.99), voxel resampling (p = 0.25-0.31), change in gray-level BW to 32 (p = 0.31-0.99), and erosions/dilations up to 4 pixels from the pancreas boundary (p = 0.12-0.34).

CONCLUSION

The model's high performance for detection of visually occult PDA was robust within a broad range of clinically relevant variations in image acquisition and radiomics workflow.

[Early detection of sporadic pancreatic cancer]

The incidence of pancreatic cancer is rising. At present, pancreatic cancer is the third most common cancer-causing death in Germany, but it is expected to become the second in 2030 and finally the leading cause of cancer death in 2050. Pancreatic ductal adenocarcinoma (PC) is generally diagnosed at advanced stages, and 5-year-survival has remained poor. Early detection of sporadic PC at stage IA, however, can yield a 5-year-survival rate of about 80%. Early detection initiatives aim at identifying persons at high risk. People with new-onset diabetes at age 50 or older have attracted much interest. Novel strategies regarding how to detect sporadic PC at an early stage are being discussed.

Automated Artificial Intelligence Model Trained on a Large Data Set Can Detect Pancreas Cancer on Diagnostic Computed Tomography Scans As Well As Visually Occult Preinvasive Cancer on Prediagnostic Computed Tomography Scans

BACKGROUND & AIMS

The aims of our case-control study were (1) to develop an automated 3-dimensional (3D) Convolutional Neural Network (CNN) for detection of pancreatic ductal adenocarcinoma (PDA) on diagnostic computed tomography scans (CTs), (2) evaluate its generalizability on multi-institutional public data sets, (3) its utility as a potential screening tool using a simulated cohort with high pretest probability, and (4) its ability to detect visually occult preinvasive cancer on prediagnostic CTs.

METHODS

A 3D-CNN classification system was trained using algorithmically generated bounding boxes and pancreatic masks on a curated data set of 696 portal phase diagnostic CTs with PDA and 1080 control images with a nonneoplastic pancreas. The model was evaluated on (1) an intramural hold-out test subset (409 CTs with PDA, 829 controls); (2) a simulated cohort with a case-control distribution that matched the risk of PDA in glycemically defined new-onset diabetes, and Enriching New-Onset Diabetes for Pancreatic Cancer score ≥3; (3) multi-institutional public data sets (194 CTs with PDA, 80 controls), and (4) a cohort of 100 prediagnostic CTs (i.e., CTs incidentally acquired 3-36 months before clinical diagnosis of PDA) without a focal mass, and 134 controls.

RESULTS

Of the CTs in the intramural test subset, 798 (64%) were from other hospitals. The model correctly classified 360 CTs (88%) with PDA and 783 control CTs (94%), with a mean accuracy 0.92 (95% CI, 0.91-0.94), area under the receiver operating characteristic (AUROC) curve of 0.97 (95% CI, 0.96-0.98), sensitivity of 0.88 (95% CI, 0.85-0.91), and specificity of 0.95 (95% CI, 0.93-0.96). Activation areas on heat maps overlapped with the tumor in 350 of 360 CTs (97%). Performance was high across tumor stages (sensitivity of 0.80, 0.87, 0.95, and 1.0 on T1 through T4 stages, respectively), comparable for hypodense vs isodense tumors (sensitivity: 0.90 vs 0.82), different age, sex, CT slice thicknesses, and vendors (all P > .05), and generalizable on both the simulated cohort (accuracy, 0.95 [95% 0.94-0.95]; AUROC curve, 0.97 [95% CI, 0.94-0.99]) and public data sets (accuracy, 0.86 [95% CI, 0.82-0.90]; AUROC curve, 0.90 [95% CI, 0.86-0.95]). Despite being exclusively trained on diagnostic CTs with larger tumors, the model could detect occult PDA on prediagnostic CTs (accuracy, 0.84 [95% CI, 0.79-0.88]; AUROC curve, 0.91 [95% CI, 0.86-0.94]; sensitivity, 0.75 [95% CI, 0.67-0.84]; and specificity, 0.90 [95% CI, 0.85-0.95]) at a median 475 days (range, 93-1082 days) before clinical diagnosis.

CONCLUSIONS

This automated artificial intelligence model trained on a large and diverse data set shows high accuracy and generalizable performance for detection of PDA on diagnostic CTs as well as for visually occult PDA on prediagnostic CTs. Prospective validation with blood-based biomarkers is warranted to assess the potential for early detection of sporadic PDA in high-risk individuals.

Can we rely on contrast-enhanced CT to identify pancreatic ductal adenocarcinoma? A population-based study in sensitivity and factors associated with false negatives

OBJECTIVES

To determine the sensitivity of contrast-enhanced computed tomography (CECT) in detecting pancreatic ductal adenocarcinoma (PDAC) and identify factors associated with false negatives (FNs).

METHODS

Patients diagnosed with PDAC in 2014-2015 were retrospectively identified by a cancer registry. CECTs performed during the diagnostic interval were retrospectively classified as true positive (TP), indeterminate, or FN. Sensitivity TP/(TP+FN) was calculated for all CECTs and the following subgroups: protocol (uniphasic vs. biphasic); tumor size (≤ 2 cm vs. > 2 cm); and resectability (potentially resectable vs. unresectable). Multivariate logistic regression was performed to assess which of the following factors were associated with FN: clinical suspicion of PDAC; size >2 cm; presence of metastases; protocol; isoattenuating tumor; and potentially resectable disease on imaging.

RESULTS

In total, 176 CECTs (127 uniphasic; 49 biphasic) in 154 patients (90 men, mean age 72 ± 11 years) were included. Sensitivity was 125/149 (83.9%) overall and 87/106 (82.1%) and 38/43 (88.4%) for uniphasic and biphasic protocols, respectively. Sensitivity was decreased for tumors ≤ 2 cm (45.4% vs. 90.6%), no liver metastases (78.0% vs. 95.9%), and potentially resectable disease (65.3% vs. 93.0%). Factors significantly associated with FN were clinical suspicion (OR, 0.24, 95% CI: 0.07-0.75), size>2 cm (OR, 0.10, 95% CI: 0.02-0.44), absence of liver metastases (OR, 4.94, 95% CI: 1.29-22.99), and potentially resectable disease (OR, 4.13, 95% CI: 1.07-16.65).

CONCLUSIONS

In our population, the overall sensitivity of CECT to detect PDAC is 83.9%; however, this is substantially lower in several scenarios, including patients with potentially resectable disease. This finding has important implications for patient outcomes and efforts to maximize CECT sensitivity should be sought.

CLINICAL RELEVANCE STATEMENT

The sensitivity of CECT to detect PDAC is significantly decreased in the setting of sub-2 cm tumors and potentially resectable disease. A dedicated biphasic pancreatic CECT protocol has higher sensitivity and should be applied in patients with suspected pancreatic disease.

KEY POINTS

• The sensitivities of contrast-enhanced CT for the detection of PDAC were 87/106 (82.1%) and 38/43 (88.4%) for uniphasic and biphasic protocols, respectively. • The sensitivity of contrast-enhanced CT was decreased for small tumors ≤ 2 cm (45.4% vs. 90.6%), if there were no liver metastases (78.0% vs. 95.9%), and with potentially resectable disease (65.3% vs. 93.0%). • Absence of liver metastases (OR, 4.94, 95% CI: 1.29-22.99) and potentially resectable disease (OR, 4.13, 95% CI: 1.07-16.65) were associated with a false--negative (FN) CT result; suspicion of malignancy on the imaging requisition (OR, 0.24, 95% CI: 0.07-0.75) and size > 2 cm (OR, 0.10, 95% CI: 0.02-0.44) were negatively associated with FN.

Radiologists' Expectations of Artificial Intelligence in Pancreatic Cancer Imaging: How Good Is Good Enough?

BACKGROUND

Existing (artificial intelligence [AI]) tools in radiology are modeled without necessarily considering the expectations and experience of the end user-the radiologist. The literature is scarce on the tangible parameters that AI capabilities need to meet for radiologists to consider them useful tools.

OBJECTIVE

The purpose of this study is to explore radiologists' attitudes toward AI tools in pancreatic cancer imaging and to quantitatively assess their expectations of these tools.

METHODS

A link to the survey was posted on the www.ctisus.com website, advertised in the www.ctisus.com email newsletter, and publicized on LinkedIn, Facebook, and Twitter accounts. This survey asked participants about their demographics, practice, and current attitudes toward AI. They were also asked about their expectations of what constitutes a clinically useful AI tool. The survey consisted of 17 questions, which included 9 multiple choice questions, 2 Likert scale questions, 4 binary (yes/no) questions, 1 rank order question, and 1 free text question.

RESULTS

A total of 161 respondents completed the survey, yielding a response rate of 46.3% of the total 348 clicks on the survey link. The minimum acceptable sensitivity of an AI program for the detection of pancreatic cancer chosen by most respondents was either 90% or 95% at a specificity of 95%. The minimum size of pancreatic cancer that most respondents would find an AI useful at detecting was 5 mm. Respondents preferred AI tools that demonstrated greater sensitivity over those with greater specificity. Over half of respondents anticipated incorporating AI tools into their clinical practice within the next 5 years.

CONCLUSION

Radiologists are open to the idea of integrating AI-based tools and have high expectations regarding the performance of these tools. Consideration of radiologists' input is important to contextualize expectations and optimize clinical adoption of existing and future AI tools.

Annotation-Efficient Deep Learning Model for Pancreatic Cancer Diagnosis and Classification Using CT Images: A Retrospective Diagnostic Study

The aim of this study was to develop a novel deep learning (DL) model without requiring large-annotated training datasets for detecting pancreatic cancer (PC) using computed tomography (CT) images. This retrospective diagnostic study was conducted using CT images collected from 2004 and 2019 from 4287 patients diagnosed with PC. We proposed a self-supervised learning algorithm (pseudo-lesion segmentation (PS)) for PC classification, which was trained with and without PS and validated on randomly divided training and validation sets. We further performed cross-racial external validation using open-access CT images from 361 patients. For internal validation, the accuracy and sensitivity for PC classification were 94.3% (92.8-95.4%) and 92.5% (90.0-94.4%), and 95.7% (94.5-96.7%) and 99.3 (98.4-99.7%) for the convolutional neural network (CNN) and transformer-based DL models (both with PS), respectively. Implementing PS on a small-sized training dataset (randomly sampled 10%) increased accuracy by 20.5% and sensitivity by 37.0%. For external validation, the accuracy and sensitivity were 82.5% (78.3-86.1%) and 81.7% (77.3-85.4%) and 87.8% (84.0-90.8%) and 86.5% (82.3-89.8%) for the CNN and transformer-based DL models (both with PS), respectively. PS self-supervised learning can increase DL-based PC classification performance, reliability, and robustness of the model for unseen, and even small, datasets. The proposed DL model is potentially useful for PC diagnosis.

Computer-Aided Detection for Pancreatic Cancer Diagnosis: Radiological Challenges and Future Directions

Radiological imaging plays a crucial role in the detection and treatment of pancreatic ductal adenocarcinoma (PDAC). However, there are several challenges associated with the use of these techniques in daily clinical practice. Determination of the presence or absence of cancer using radiological imaging is difficult and requires specific expertise, especially after neoadjuvant therapy. Early detection and characterization of tumors would potentially increase the number of patients who are eligible for curative treatment. Over the last decades, artificial intelligence (AI)-based computer-aided detection (CAD) has rapidly evolved as a means for improving the radiological detection of cancer and the assessment of the extent of disease. Although the results of AI applications seem promising, widespread adoption in clinical practice has not taken place. This narrative review provides an overview of current radiological CAD systems in pancreatic cancer, highlights challenges that are pertinent to clinical practice, and discusses potential solutions for these challenges.

Multiparametric detection and outcome prediction of pancreatic cancer involving dual-energy CT, diffusion-weighted MRI, and radiomics

BACKGROUND

The advent of next-generation computed tomography (CT)- and magnetic resonance imaging (MRI) opened many new perspectives in the evaluation of tumor characteristics. An increasing body of evidence suggests the incorporation of quantitative imaging biomarkers into clinical decision-making to provide mineable tissue information. The present study sought to evaluate the diagnostic and predictive value of a multiparametric approach involving radiomics texture analysis, dual-energy CT-derived iodine concentration (DECT-IC), and diffusion-weighted MRI (DWI) in participants with histologically proven pancreatic cancer.

METHODS

In this study, a total of 143 participants (63 years ± 13, 48 females) who underwent third-generation dual-source DECT and DWI between November 2014 and October 2022 were included. Among these, 83 received a final diagnosis of pancreatic cancer, 20 had pancreatitis, and 40 had no evidence of pancreatic pathologies. Data comparisons were performed using chi-square statistic tests, one-way ANOVA, or two-tailed Student's t-test. For the assessment of the association of texture features with overall survival, receiver operating characteristics analysis and Cox regression tests were used.

RESULTS

Malignant pancreatic tissue differed significantly from normal or inflamed tissue regarding radiomics features (overall P < .001, respectively) and iodine uptake (overall P < .001, respectively). The performance for the distinction of malignant from normal or inflamed pancreatic tissue ranged between an AUC of ≥ 0.995 (95% CI, 0.955-1.0; P < .001) for radiomics features, ≥ 0.852 (95% CI, 0.767-0.914; P < .001) for DECT-IC, and ≥ 0.690 (95% CI, 0.587-0.780; P = .01) for DWI, respectively. During a follow-up of 14 ± 12 months (range, 10-44 months), the multiparametric approach showed a moderate prognostic power to predict all-cause mortality (c-index = 0.778 [95% CI, 0.697-0.864], P = .01).

CONCLUSIONS

Our reported multiparametric approach allowed for accurate discrimination of pancreatic cancer and revealed great potential to provide independent prognostic information on all-cause mortality.

Evaluation of Perfusion Change According to Pancreatic Cancer and Pancreatic Duct Dilatation Using Free-Breathing Golden-Angle Radial Sparse Parallel (GRASP) Magnetic Resonance Imaging

PURPOSE

To evaluate perfusion changes in the pancreas with pancreatic cancer and pancreatic duct dilatation using dynamic contrast-enhanced MRI (DCE-MRI).

METHOD

We evaluate the pancreas DCE-MRI of 75 patients. The qualitative analysis includes pancreas edge sharpness, motion artifacts, streak artifacts, noise, and overall image quality. The quantitative analysis includes measuring the pancreatic duct diameter and drawing six regions of interest (ROIs) in the three areas of the pancreas (head, body, and tail) and three vessels (aorta, celiac axis, and superior mesenteric artery) to measure the peak-enhancement time, delay time, and peak concentration. We evaluate the differences in three quantitative parameters among the ROIs and between patients with and without pancreatic cancer. The correlations between pancreatic duct diameter and delay time are also analyzed.

RESULTS

The pancreas DCE-MRI demonstrates good image quality, and respiratory motion artifacts show the highest score. The peak-enhancement time does not differ among the three vessels or among the three pancreas areas. The peak-enhancement time and concentrations in the pancreas body and tail and the delay time in the three pancreas areas are significantly longer (p < 0.05) in patients with pancreatic cancer than in those without pancreatic cancer. The delay time was significantly correlated with the pancreatic duct diameters in the head (p < 0.02) and body (p < 0.001).

CONCLUSION

DCE-MRI can display the perfusion change in the pancreas with pancreatic cancer. A perfusion parameter in the pancreas is correlated with the pancreatic duct diameter reflecting a morphological change in the pancreas.

Pancreatic Mass Characterization Using IVIM-DKI MRI and Machine Learning-Based Multi-Parametric Texture Analysis

Non-invasive characterization of pancreatic masses aids in the management of pancreatic lesions. Intravoxel incoherent motion-diffusion kurtosis imaging (IVIM-DKI) and machine learning-based texture analysis was used to differentiate pancreatic masses such as pancreatic ductal adenocarcinoma (PDAC), pancreatic neuroendocrine tumor (pNET), solid pseudopapillary epithelial neoplasm (SPEN), and mass-forming chronic pancreatitis (MFCP). A total of forty-eight biopsy-proven patients with pancreatic masses were recruited and classified into pNET (n = 13), MFCP (n = 6), SPEN (n = 4), and PDAC (n = 25) groups. All patients were scanned for IVIM-DKI sequences acquired with 14 b-values (0 to 2500 s/mm2) on a 1.5T MRI. An IVIM-DKI model with a 3D total variation (TV) penalty function was implemented to estimate the precise IVIM-DKI parametric maps. Texture analysis (TA) of the apparent diffusion coefficient (ADC) and IVIM-DKI parametric map was performed and reduced using the chi-square test. These features were fed to an artificial neural network (ANN) for characterization of pancreatic mass subtypes and validated by 5-fold cross-validation. Receiver operator characteristics (ROC) analyses were used to compute the area under curve (AUC). Perfusion fraction (f) was significantly higher (p < 0.05) in pNET than PDAC. The f showed better diagnostic performance for PDAC vs. MFCP with AUC:0.77. Both pseudo-diffusion coefficient (D*) and f for PDAC vs. pNET showed an AUC of 0.73. ADC and diffusion coefficient (D) showed good diagnostic performance for pNET vs. MFCP with AUC: 0.79 and 0.76, respectively. In the TA of PDAC vs. non-PDAC, f and combined IVIM-DKI parameters showed high accuracy ≥ 84.3% and AUC ≥ 0.84. Mean f and combined IVIM-DKI parameters estimated that the IVIM-DKI model with TV texture features has the potential to be helpful in characterizing pancreatic masses.

Pancreatic Cancer Detection on CT Scans with Deep Learning: A Nationwide Population-based Study

Background Approximately 40% of pancreatic tumors smaller than 2 cm are missed at abdominal CT. Purpose To develop and to validate a deep learning (DL)-based tool able to detect pancreatic cancer at CT. Materials and Methods Retrospectively collected contrast-enhanced CT studies in patients diagnosed with pancreatic cancer between January 2006 and July 2018 were compared with CT studies of individuals with a normal pancreas (control group) obtained between January 2004 and December 2019. An end-to-end tool comprising a segmentation convolutional neural network (CNN) and a classifier ensembling five CNNs was developed and validated in the internal test set and a nationwide real-world validation set. The sensitivities of the computer-aided detection (CAD) tool and radiologist interpretation were compared using the McNemar test. Results A total of 546 patients with pancreatic cancer (mean age, 65 years ± 12 [SD], 297 men) and 733 control subjects were randomly divided into training, validation, and test sets. In the internal test set, the DL tool achieved 89.9% (98 of 109; 95% CI: 82.7, 94.9) sensitivity and 95.9% (141 of 147; 95% CI: 91.3, 98.5) specificity (area under the receiver operating characteristic curve [AUC], 0.96; 95% CI: 0.94, 0.99), without a significant difference (P = .11) in sensitivity compared with the original radiologist report (96.1% [98 of 102]; 95% CI: 90.3, 98.9). In a test set of 1473 real-world CT studies (669 malignant, 804 control) from institutions throughout Taiwan, the DL tool distinguished between CT malignant and control studies with 89.7% (600 of 669; 95% CI: 87.1, 91.9) sensitivity and 92.8% specificity (746 of 804; 95% CI: 90.8, 94.5) (AUC, 0.95; 95% CI: 0.94, 0.96), with 74.7% (68 of 91; 95% CI: 64.5, 83.3) sensitivity for malignancies smaller than 2 cm. Conclusion The deep learning-based tool enabled accurate detection of pancreatic cancer on CT scans, with reasonable sensitivity for tumors smaller than 2 cm. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Aisen and Rodrigues in this issue.

Assessment of main pancreatic duct cutoff with dilatation, but without visible pancreatic focal lesion on MDCT: a novel diagnostic approach for malignant stricture using a CT-based nomogram

OBJECTIVES

To identify useful features to predict hidden pancreatic malignancies in patients with main pancreatic duct (MPD) abrupt cutoff and dilatation, but without visible focal pancreatic lesions on CT.

METHODS

This retrospective study included 92 patients (mean age, 63.4 ± 10.6 years, 63 men and 29 women) with MPD abrupt cutoff and dilatation, but without visible focal pancreatic lesion on contrast-enhanced CT between 2009 and 2021. Two radiologists independently evaluated the CT imaging features. Multivariable logistic regression analysis was performed to identify clinical and CT imaging features for hidden pancreatic malignancies. A nomogram was developed based on these results and assessed its performance.

RESULTS

Thirty-eight (41.3%) and 54 (58.7%) were classified into the malignant and benign groups, respectively. In the multivariable analysis, CA19-9 elevation (odds ratio [OR] 7.5, p = 0.003), duct cutoff site at the head/neck (OR 7.6, p = 0.006), parenchymal contour abnormality at the duct cutoff site (OR 13.7, p < 0.001), and presence of acute pancreatitis (OR 11.5, p = 0.005) were independent predictors of pancreatic malignancy. A combination of any two significant features showed an accuracy of 77.2%, and a combination of any three features exhibited a specificity of 100%. The CT-based nomogram showed an area under the curve (AUC) of 0.84 (95% confidence interval, 0.77-0.90).

CONCLUSIONS

The three CT imaging features and CA19-9 elevation translated into a nomogram permit a reliable estimation of hidden pancreatic malignancies in patients with MPD abrupt cutoff without visible focal pancreatic lesion. It may facilitate determining whether to proceed to further diagnostic tests.

KEY POINTS

• Isoattenuating pancreatic ductal adenocarcinoma can manifest only as an isolated main pancreatic duct (MPD) dilatation with abrupt cutoff, making it difficult to distinguish from benign strictures. • Along with the serum CA 19-9 elevation, MPD cutoff site at the pancreas head or neck, parenchymal contour abnormality at the duct cutoff site, and associated acute pancreatitis indicated a higher probability of the malignant MPD strictures. • The CT-based nomogram provided excellent diagnostic performance (AUC of 0.84) for hidden pancreatic malignancies in patients with MPD abrupt cutoff and dilatation.

Radiomics-based Machine-learning Models Can Detect Pancreatic Cancer on Prediagnostic Computed Tomography Scans at a Substantial Lead Time Before Clinical Diagnosis

BACKGROUND & AIMS

Our purpose was to detect pancreatic ductal adenocarcinoma (PDAC) at the prediagnostic stage (3-36 months before clinical diagnosis) using radiomics-based machine-learning (ML) models, and to compare performance against radiologists in a case-control study.

METHODS

Volumetric pancreas segmentation was performed on prediagnostic computed tomography scans (CTs) (median interval between CT and PDAC diagnosis: 398 days) of 155 patients and an age-matched cohort of 265 subjects with normal pancreas. A total of 88 first-order and gray-level radiomic features were extracted and 34 features were selected through the least absolute shrinkage and selection operator-based feature selection method. The dataset was randomly divided into training (292 CTs: 110 prediagnostic and 182 controls) and test subsets (128 CTs: 45 prediagnostic and 83 controls). Four ML classifiers, k-nearest neighbor (KNN), support vector machine (SVM), random forest (RM), and extreme gradient boosting (XGBoost), were evaluated. Specificity of model with highest accuracy was further validated on an independent internal dataset (n = 176) and the public National Institutes of Health dataset (n = 80). Two radiologists (R4 and R5) independently evaluated the pancreas on a 5-point diagnostic scale.

RESULTS

Median (range) time between prediagnostic CTs of the test subset and PDAC diagnosis was 386 (97-1092) days. SVM had the highest sensitivity (mean; 95% confidence interval) (95.5; 85.5-100.0), specificity (90.3; 84.3-91.5), F1-score (89.5; 82.3-91.7), area under the curve (AUC) (0.98; 0.94-0.98), and accuracy (92.2%; 86.7-93.7) for classification of CTs into prediagnostic versus normal. All 3 other ML models, KNN, RF, and XGBoost, had comparable AUCs (0.95, 0.95, and 0.96, respectively). The high specificity of SVM was generalizable to both the independent internal (92.6%) and the National Institutes of Health dataset (96.2%). In contrast, interreader radiologist agreement was only fair (Cohen's kappa 0.3) and their mean AUC (0.66; 0.46-0.86) was lower than each of the 4 ML models (AUCs: 0.95-0.98) (P < .001). Radiologists also recorded false positive indirect findings of PDAC in control subjects (n = 83) (7% R4, 18% R5).

CONCLUSIONS

Radiomics-based ML models can detect PDAC from normal pancreas when it is beyond human interrogation capability at a substantial lead time before clinical diagnosis. Prospective validation and integration of such models with complementary fluid-based biomarkers has the potential for PDAC detection at a stage when surgical cure is a possibility.

Transabdominal ultrasound of pancreatic ductal adenocarcinoma: A multi-centered population-based study in sensitivity, associated diagnostic intervals, and survival

OBJECTIVES

To determine the sensitivity of ultrasound (US) in detecting pancreatic ductal adenocarcinoma in our region, to identify factors associated with US test result, and assess the impact on the diagnostic interval and survival.

METHODS

Patients diagnosed between January 1, 2014 and December 31, 2015 in Nova Scotia, Canada were identified by a cancer registry. US performed prior to diagnosis were retrospectively graded as true positive (TP), indeterminate or false negative (FN). Amongst US results, differences in age, weight and tumor size were assessed [one-way analysis of variance (ANOVA)]. Associations between result and sex, tumor location (proximal/distal), clinical suspicion of malignancy, and visualization of the pancreas, tumor, secondary signs and liver metastases were assessed (Chi-square). Mean follow-up imaging, diagnostic, and survival intervals were assessed (one-way ANOVA).

RESULTS

One hundred thirteen US of 107 patients (54 women; mean 70 ± 13 years) were graded as follows: 48/113 (42.5%) TPs; 42/113 (37.2%) indeterminates; and 23/113 (20.4%) FNs. Sensitivity was 48/71(67.6%). There was no difference in age, weight or tumor size amongst US result (P > 0.5). FNs had proportionally more men (P = 0.011) and lacked clinical suspicion of malignancy (P = 0.0006); TPs had proportionally more proximal tumors (P = 0.017). US result was associated with visualization of the pancreas, tumor, secondary signs and liver metastases (P < 0.005). FNs had longer mean follow-up imaging (P < 0.0001) and diagnostic (P = 0.0007) intervals, and worse mean survival (P = 0.034).

CONCLUSIONS

In our region, the sensitivity of US in detecting pancreatic ductal adenocarcinoma is 67.6%. A false negative US is associated with delayed diagnostic work-up and worse mean survival.

Artificial Intelligence-Driven Diagnosis of Pancreatic Cancer

Pancreatic cancer is among the most challenging forms of cancer to treat, owing to its late diagnosis and aggressive nature that reduces the survival rate drastically. Pancreatic cancer diagnosis has been primarily based on imaging, but the current state-of-the-art imaging provides a poor prognosis, thus limiting clinicians' treatment options. The advancement of a cancer diagnosis has been enhanced through the integration of artificial intelligence and imaging modalities to make better clinical decisions. In this review, we examine how AI models can improve the diagnosis of pancreatic cancer using different imaging modalities along with a discussion on the emerging trends in an AI-driven diagnosis, based on cytopathology and serological markers. Ethical concerns regarding the use of these tools have also been discussed.

The diagnostic performance of contrast-enhanced ultrasound versus contrast-enhanced computed tomography for pancreatic carcinoma: a systematic review and meta-analysis

BACKGROUND

Pancreatic carcinoma is a highly fatal disease, and early diagnosis is of vital importance. This meta-analysis aimed to compare the diagnostic performances of contrast-enhanced ultrasonography (CEUS) against contrast-enhanced computed tomography (CECT) for pancreatic carcinoma, using pathological results or alternative imaging modality as the gold standard.

METHODS

A thorough search was conducted in PubMed, EMBASE, Web of Science and Cochrane Library databases. Two investigators selected the studies and extracted the data independently. A bivariate mixed-effects regression model was used to calculate the pooled data. Subgroup analysis and meta-regression were performed to explore the causes of heterogeneity.

RESULTS

A total of 1,227 records were identified, of which 7 articles with 588 patients were assessed for eligibility. The overall sensitivity, specificity of CEUS and CECT with their 95% confidential intervals (95% CI) were 0.91 (0.85-0.94) and 0.88 (0.81-0.92), 0.83 (0.70-0.91) and 0.87 (0.73-0.94), respectively. The area under curve (AUC) of CEUS and CECT were 0.94 and 0.93. Subgroup analysis showed CEUS may be good at diagnosing lesions with diameters less than 2 cm. Tumor features, region and study type were the main causes of heterogeneity.

CONCLUSIONS

CEUS has a satisfying diagnostic performance for pancreatic carcinoma and it has high sensitivity for small pancreatic carcinomas (≤2 cm); besides, it performs well in discriminating pancreatic cancer from chronic pancreatitis. Therefore, CEUS can be a useful supplement to CECT.

Prevalence, features, and explanations of missed and misinterpreted pancreatic cancer on imaging: a matched case-control study

PURPOSE

To characterize the prevalence of missed pancreatic masses and pancreatic ductal adenocarcinoma (PDAC)-related findings on CT and MRI between pre-diagnostic patients and healthy individuals.

MATERIALS AND METHODS

Patients diagnosed with PDAC (2010-2016) were retrospectively reviewed for abdominal CT- or MRI-examinations 1 month-3 years prior to their diagnosis, and subsequently matched to controls in a 1:4 ratio. Two blinded radiologists scored each imaging exam on the presence of a pancreatic mass and secondary features of PDAC. Additionally, original radiology reports were graded based on the revised RADPEER criteria.

RESULTS

The cohort of 595 PDAC patients contained 60 patients with a pre-diagnostic CT and 27 with an MRI. A pancreatic mass was suspected in hindsight on CT in 51.7% and 50% of cases and in 1.3% and 0.9% of controls by reviewer 1 (p < .001) and reviewer 2 (p < .001), respectively. On MRI, a mass was suspected in 70.4% and 55.6% of cases and 2.9% and 0% of the controls by reviewer 1 (p < .001) and reviewer 2 (p < .001), respectively. Pancreatic duct dilation, duct interruption, focal atrophy, and features of acute pancreatitis is strongly associated with PDAC (p < .001). In cases, a RADPEER-score of 2 or 3 was assigned to 56.3% of the CT-reports and 71.4% of MRI-reports.

CONCLUSION

Radiological features as pancreatic duct dilation and interruption, and focal atrophy are common first signs of PDAC and are often missed or unrecognized. Further investigation with dedicated pancreas imaging is warranted in patients with PDAC-related radiological findings.

Acute Pancreatitis as the Initial Presentation of Pancreatic Adenocarcinoma does not Impact Short- and Long-term Outcomes of Curative Intent Surgery: A Study of the French Surgical Association

BACKGROUND

Acute pancreatitis (AP) can be one of the earliest clinical presentation of pancreatic ductal adenocarcinoma (PDAC). Information about the impact of AP on postoperative outcomes as well as its influences on PDAC survival is scarce. This study aimed to determine whether AP as initial clinical presentation of PDAC impact the short- and long-term outcomes of curative intent pancreatic resection.

PATIENTS AND METHODS

From 2004 to 2009, 1449 patients with PDAC underwent pancreatic resection in 37 institutions (France, Belgium and Switzerland). We used univariate and multivariate analysis to identify factors associated with severe complications and pancreatic fistula as well as overall and disease-free survivals.

RESULTS

There were 764 males (52,7%), and the median age was 64 years. A total of 781 patients (53.9%) developed at least one complication, among whom 317 (21.8%) were classified as Clavien-Dindo ≥ 3. A total of 114 (8.5%) patients had AP as the initial clinical manifestation of PDAC. This situation was not associated with any increase in the rates of postoperative fistula (21.2% vs 16.4%, P = 0.19), postoperative complications (57% vs 54.2%, P = 0.56), and 30 day mortality (2.6% vs 3.4%, P = 1). In multivariate analysis, AP did not correlate with postoperative complications or pancreatic fistula. The median length of follow-up was 22.4 months. The median overall survival after surgery was 29.9 months in the AP group and 30.5 months in the control group. Overall recurrence rate and local recurrence rate did not differ between groups.

CONCLUSION

AP before PDAC resection did not impact postoperative morbidity and mortality, as well as recurrence rate and survival.

CT Abnormalities of the Pancreas Associated With the Subsequent Diagnosis of Clinical Stage I Pancreatic Ductal Adenocarcinoma More Than One Year Later: A Case-Control Study

Background: Pancreatic ductal adenocarcinoma (PDAC) is highly lethal, partly due to challenges in early diagnosis. However, the prognosis for earlier stages (carcinoma in situ or stage T1a invasive carcinoma) is relatively favorable. Objective: To investigate findings of an earlier diagnosis of PDAC on pre-diagnostic CT examinations performed at least one year before the diagnosis of clinical stage I PDAC. Methods: This retrospective study included 103 patients with clinical stage I PDAC and a pre-diagnostic CT at least one year before the CT that detected PDAC, as well as 103 control patients without PDAC on CT examinations separated by at least 10 years. The frequency and temporal characteristics of focal pancreatic abnormalities (pancreatic mass, main pancreatic duct (MPD) change, parenchymal atrophy, faint parenchymal enhancement, cyst, and parenchymal calcification) on pre-diagnostic CT examinations were determined. Results: A focal pancreatic abnormality was present on the most recent pre-diagnostic CT in 55/103 (53.4%) patients with PDAC versus 21/103 (20.4%) control patients (p<.001). In patients with PDAC, the most common focal abnormalities on pre-diagnostic CT were atrophy (39/103, 37.9%), faint enhancement (17/65, 26.2%), and MPD change (14/103, 13.6%), which were all more frequent in patients with PDAC than in control patients (p<.05). In 54/55 (98.2%) patients with PDAC, the PDAC corresponded with the site of a focal abnormality (exact location or the abnormality's upstream or downstream edge) on pre-diagnostic CT. Frequency of focal abnormalities decreased with increasing time before the CT that detected PDAC (1-2 years before diagnosis, 64.9%; 2-3 years, 49.2%; 3-5 years, 41.8%; 5-7 years, 29.7%; 7-10 years, 18.5%; over 10 years, 0%). Mean duration from the finding's initial appearance to diagnosis of PDAC was 4.6 years for atrophy, 3.3 years for faint enhancement, and 1.1 years for MPD change. Conclusion: Most patients with clinical stage I PDAC demonstrated focal pancreatic abnormalities on pre-diagnostic CT obtained at least one year before diagnosis. Focal MPD change exhibited the shortest duration from its development to subsequent diagnosis, where atrophy and faint enhancement exhibited a relatively prolonged course. Clinical impact: These findings could facilitate earlier PDAC diagnosis and thus improve prognosis.