Differentiating autoimmune pancreatitis from pancreatic ductal adenocarcinoma with CT radiomics features

Park, S.; Chu, Linda C.; Hruban, Ralph H.; Vogelstein, Bert; Kinzler, Kenneth W.; Yuille, Alan; Fouladi, Daniel Fadaei; Shayesteh, Shahab; Ghandili, Saeed; Wolfgang, C.L.; Burkhart, Richard A.; He, Jin; Fishman, Elliot K.; Kawamoto, Satomi

doi:10.1016/j.diii.2020.03.002

Cited by 73 publications

(42 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Radiomics, Texture Analysis) in the differentiation of focal autoimmune pancreatitis and isolated IRC from pancreatic cancer and cholangiocarcinoma, respectively. 223 • as there have been no randomized controlled studies on the treatment of IgG4-related digestive disease, the best evidence-based treatment of this disorder is still unknown. The use of certain medications for treatment of IgG4-related digestive disease varies in different countries, among different specialties, and different organ involvement; therefore, multicentre clinical trials with large numbers of patients are needed to define optimal treatment protocols.…”

Section: Areas Of Uncertainty and Future Researchmentioning

confidence: 99%

“…The major topics are the following: incidence of cancer in patients with IgG4‐related digestive disease compared with age‐, sex‐ and risk factor‐matched control subjects, and any risk factors for its development should be explored in prospective studies. prospective studies to evaluate the accuracy of imaging modalities (CT, MR, PET‐CT) even with the aid of innovative post‐processing methods (i.e. Radiomics, Texture Analysis) in the differentiation of focal autoimmune pancreatitis and isolated IRC from pancreatic cancer and cholangiocarcinoma, respectively 223 as there have been no randomized controlled studies on the treatment of IgG4‐related digestive disease, the best evidence‐based treatment of this disorder is still unknown.…”

Section: Areas Of Uncertainty and Future Researchmentioning

confidence: 99%

See 1 more Smart Citation

European Guideline on IgG4‐related digestive disease – UEG and SGF evidence‐based recommendations

et al. 2020

View full text Add to dashboard Cite

The overall objective of these guidelines is to provide evidence-based recommendations for the diagnosis and management of immunoglobulin G4 (IgG4)-related digestive disease in adults and children. IgG4-related digestive disease can be diagnosed only with a comprehensive work-up that includes histology, organ morphology at imaging, serology, search for other organ involvement, and response to glucocorticoid treatment. Indications for treatment are symptomatic patients with obstructive jaundice, abdominal pain, posterior pancreatic pain, and involvement of extra-pancreatic digestive organs, including IgG4-related cholangitis. Treatment with glucocorticoids should be weight-based and initiated at a dose of 0.6–0.8 mg/kg body weight/day orally (typical starting dose 30-40 mg/day prednisone equivalent) for 1 month to induce remission and then be tapered within two additional months. Response to initial treatment should be assessed at week 2–4 with clinical, biochemical and morphological markers. Maintenance treatment with glucocorticoids should be considered in multi-organ disease or history of relapse. If there is no change in disease activity and burden within 3 months, the diagnosis should be reconsidered. If the disease relapsed during the 3 months of treatment, immunosuppressive drugs should be added.

show abstract

Section: Areas Of Uncertainty and Future Researchmentioning

confidence: 99%

Section: Areas Of Uncertainty and Future Researchmentioning

confidence: 99%

European Guideline on IgG4‐related digestive disease – UEG and SGF evidence‐based recommendations

et al. 2020

View full text Add to dashboard Cite

show abstract

“…These hand-crafted features have been shown to be prognostic for survival and recurrence in different cancer sites (Kumar et al, 2012;Balagurunathan et al, 2014;Haider et al, 2017). In the PDAC radiomics field, more than forty features have been found to be significantly associated with tissue classification or overall survival for PDAC patients (e.g., sum entropy, cluster tendency, dissimilarity, uniformity, and busyness) (Cassinotto et al, 2017;Chakraborty et al, 2017;Attiyeh et al, 2018;Yun et al, 2018;Chu et al, 2019;Sandrasegaran et al, 2019;Li et al, 2020;Park et al, 2020). Furthermore, a few radiomics features have been found to be associated with tumor heterogeneity and genomics profile (Lambin et al, 2012;Itakura et al, 2015;Rizzo et al, 2016;Li et al, 2018).…”

Section: Discussionmentioning

confidence: 99%

Prognostic Value of Transfer Learning Based Features in Resectable Pancreatic Ductal Adenocarcinoma

Zhang

Lobo-Mueller

Karanicolas

et al. 2020

Front. Artif. Intell.

View full text Add to dashboard Cite

Background: Pancreatic Ductal Adenocarcinoma (PDAC) is one of the most aggressive cancers with an extremely poor prognosis. Radiomics has shown prognostic ability in multiple types of cancer including PDAC. However, the prognostic value of traditional radiomics pipelines, which are based on hand-crafted radiomic features alone is limited. Methods: Convolutional neural networks (CNNs) have been shown to outperform radiomics models in computer vision tasks. However, training a CNN from scratch requires a large sample size which is not feasible in most medical imaging studies. As an alternative solution, CNN-based transfer learning models have shown the potential for achieving reasonable performance using small datasets. In this work, we developed and validated a CNN-based transfer learning model for prognostication of overall survival in PDAC patients using two independent resectable PDAC cohorts. Results: The proposed transfer learning-based prognostication model for overall survival achieved the area under the receiver operating characteristic curve of 0.81 on the test cohort, which was significantly higher than that of the traditional radiomics model (0.54). To further assess the prognostic value of the models, the predicted probabilities of death generated from the two models were used as risk scores in a univariate Cox Proportional Hazard model and while the risk score from the traditional radiomics model was not associated with overall survival, the proposed transfer learning-based risk score had significant prognostic value with hazard ratio of 1.86 (95% Confidence Interval: 1.15-3.53, p-value: 0.04). Conclusions: This result suggests that transfer learning-based models may significantly improve prognostic performance in typical small sample size medical imaging studies.

show abstract

“…Different clinical and radiological modalities, including serum IgG4, CA19‐9, tumour‐specific antigens, gene expression, endoscopic ultrasonography, computerised tomography (CT) and magnetic resonance imaging (MRI), have been explored for differentiating AIP from pancreatic cancer. Radiological studies can be suggestive, but not diagnostic 5,21,22 . Serum IgG4 levels above twice the upper limit of normal are often seen in type 1 AIP, but the sensitivity is not perfect 23–26 .…”

Section: Discussionmentioning

confidence: 99%

“…Autoimmune pancreatitis (AIP) is a mimic of neoplasia in its clinical, radiological and histological features 1–7 . Current consensus divides AIP into two subtypes 2–8 . Type 1 AIP is a manifestation of immunoglobulin (Ig)G4‐related disease and is also known as lymphoplasmacytic sclerosing pancreatitis.…”

Section: Introductionmentioning

confidence: 99%

Prevalence of histological features resembling autoimmune pancreatitis in neoplastic pancreas resections

2020

View full text Add to dashboard Cite

Prevalence of histological features resembling autoimmune pancreatitis in neoplastic pancreas resections Aims: Types 1 and 2 autoimmune pancreatitis (AIP) can mimic pancreatic neoplasia. Due to the small quantity of tissue in mass-targeted pancreas biopsies, inflammatory features may raise the differential of AIP. However, the frequency of AIP-like histology in neoplastic pancreas is not well characterised. Therefore, the specificity of inflammatory lesions on biopsy with respect to the diagnosis of AIP is uncertain. Methods and results: Neoplastic pancreas resections performed at our institution between 2008 and 2019 were retrospectively reviewed. Features of AIP types 1 and 2 were assessed in the non-neoplastic areas. If features of immunoglobulin (Ig)G4-associated AIP were seen, IgG4 immunohistochemistry was performed. We identified 163 neoplastic pancreas resections. Of these, 34 had one or more types of inflammatory lesions in non-neoplastic pancreatic tissue. Dense lymphoplasmacytic inflammation mimicking type 1 AIP was found in six cases with mild to moderately increased IgG4-positive plasma cells. Neutrophilic infiltrates in small intralobular ducts were found in 20 cases. Mild extralobular ductitis or duct microabscess was found in 10 specimens. Marked neutrophilic duct destruction that resembled granulocytic epithelial lesions was found in 12 cases. Some cases showed multiple features. Conclusion: Approximately 20% of neoplastic pancreas resections showed focal areas that could raise the differential of AIP. More cases showed neutrophilic predominant inflammation as seen in type 2 autoimmune pancreatitis, compared to dense lymphoplasmacytic infiltrates seen in type 1 AIP. Pathologists must be cautious when making a diagnosis of AIP on biopsy tissue based on histological findings alone.

show abstract

Differentiating autoimmune pancreatitis from pancreatic ductal adenocarcinoma with CT radiomics features

Cited by 73 publications

References 32 publications

European Guideline on IgG4‐related digestive disease – UEG and SGF evidence‐based recommendations

European Guideline on IgG4‐related digestive disease – UEG and SGF evidence‐based recommendations

Prognostic Value of Transfer Learning Based Features in Resectable Pancreatic Ductal Adenocarcinoma

Prevalence of histological features resembling autoimmune pancreatitis in neoplastic pancreas resections

Contact Info

Product

Resources

About