Search

Purpose: To explore the predictive potential of serial computed tomography (CT) radiology reports for pancreatic cancer survival using natural language processing (NLP)., Methods: Deep-transfer-learning-based NLP models were retrospectively trained and tested with serial, free-text CT reports, and survival information of consecutive patients diagnosed with pancreatic cancer in a Korean tertiary hospital was extracted. Randomly selected patients with pancreatic cancer and their serial CT reports from an independent tertiary hospital in the United States were included in the external testing data set. The concordance index (c-index) of predicted survival and actual survival, and area under the receiver operating characteristic curve (AUROC) for predicting 1-year survival were calculated., Results: Between January 2004 and June 2021, 2,677 patients with 12,255 CT reports and 670 patients with 3,058 CT reports were allocated to training and internal testing data sets, respectively. ClinicalBERT (Bidirectional Encoder Representations from Transformers) model trained on the single, first CT reports showed a c-index of 0.653 and AUROC of 0.722 in predicting the overall survival of patients with pancreatic cancer. ClinicalBERT trained on up to 15 consecutive reports from the initial report showed an improved c-index of 0.811 and AUROC of 0.911. On the external testing set with 273 patients with 1,947 CT reports, the AUROC was 0.888, indicating the generalizability of our model. Further analyses showed our model's contextual interpretation beyond specific phrases., Conclusion: Deep-transfer-learning-based NLP model of serial CT reports can predict the survival of patients with pancreatic cancer. Clinical decisions can be supported by the developed model, with survival information extracted solely from serial radiology reports.

The objective of this study was to evaluate the role of regional lymph node (LN) metastasis detected on preoperative CT and/or 18 F-fluoro-2-deoxyglucose-positron emission tomography (FDG-PET) scans in the prediction of early tumor recurrence after curative surgical resection of pancreatic ductal adenocarcinoma (PDAC). This retrospective study included 137 patients who underwent upfront surgery with R0 resection of PDAC between 2013 and 2016. Regional LN metastasis was identified using two criteria: positive findings for regional LN metastasis on either preoperative CT or FDG-PET scans (LN OR ), or on both preoperative CT and FDG-PET scans (LN AND ). A total of 55 patients had early tumor recurrence within 12 months after curative resection. Univariable and multivariable Cox proportional hazard regression analysis showed that preoperative carbohydrate antigen 19-9 (CA19-9) levels, preoperative locally advanced status, and regional LN metastasis (both LN OR and LN AND criteria) were significant risk factors for early recurrence. Positive LN OR and LN AND showed significantly poorer recurrence-free survival compared to negative regional LN metastasis groups (p = 0.048 and p = 0.020, respectively). Compared with the LN AND criteria, the LN OR criteria provided higher sensitivity (22.4% vs. 15.5%, p = 0.046) and a higher negative predictive value (61.9% vs. 59.8%, p = 0.046). The LN OR definition provided more sensitive and accurate performance in diagnosing preoperative regional LN metastasis., (© 2022. The Author(s).)

Background Imaging studies have limitations in evaluating pancreatic ductal adenocarcinoma (PDAC) treatment response. Purpose To investigate the effectiveness of combined CT and carbohydrate antigen 19-9 (CA 19-9) evaluation at 8 weeks after first-line treatment to predict overall survival (OS) of patients with nonmetastatic PDAC. Materials and Methods Patients with nonmetastatic PDAC who received first-line treatment with either chemotherapy or concurrent chemoradiation in a single-center PDAC cohort registry were retrospectively enrolled in the study between January 2013 and December 2016. Follow-up CT images obtained 8 weeks after treatment were evaluated according to Response Evaluation Criteria in Solid Tumors. Patients with partial response (PR) or stable disease (SD) were defined as CT responders, and those with progressive disease (PD) were defined as CT nonresponders. Patients with a normalized CA 19-9 level at 8-week follow-up were defined as CA 19-9 responders, and those with a nonnormalized or nonelevated CA 19-9 level were defined as CA 19-9 nonresponders. OS was compared using the Kaplan-Meier method with Breslow analysis. Results A total of 197 patients (mean age ± standard deviation, 65 years ± 10; 107 men) were evaluated. Patients with PD ( n = 17) showed shorter OS than those with SD ( n = 147; P < .001) or PR ( n = 33; P = .003). OS did not differ between the patients with PR and those with SD ( P = .60). When the CT and CA 19-9 responses were integrated, OS was longest in CT and CA 19-9 responders (group 1, n = 27; median OS, 26.6 months [95% CI: 9.0, 44.1]), followed by CT responders but CA 19-9 nonresponders (group 2, n = 153; median OS, 15.9 months [95% CI: 13.3, 18.5]; P = .007 vs group 1) and CT and CA 19-9 nonresponders (group 3, n = 17; median OS, 6.5 months [95% CI: 0.8, 12.2]; P < .001 vs group 2). Conclusion Integrated evaluation with CT and carbohydrate antigen 19-9 response allowed more accurate stratification of survival in patients with pancreatic ductal adenocarcinoma in the early treatment period than did evaluation according to Response Evaluation Criteria in Solid Tumors. © RSNA, 2022 Online supplemental material is available for this article.

Background/aims: : Controversy regarding the effectiveness of neoadjuvant therapy for resectable pancreatic ductal adenocarcinoma (PDAC) still exists. Here, we aimed to identify the potential benefits of neoadjuvant therapy followed by surgery for resectable PDAC., Methods: We reviewed radiologically resectable PDAC patients who received resection with curative intent at a tertiary hospital in South Korea between January 2012 and August 2019. A total of 202 patients underwent curative resection for resectable PDAC: 167 underwent surgical resection first during this period, and 35 received neoadjuvant chemotherapy/chemoradiation therapy followed by surgery. Resectable PDAC patients were subdivided, and 1:3 propensity score matching (PSM) was performed to reduce selection bias., Results: Compared with the group that received surgery first, the group that received neoadjuvant treatment followed by surgery had significantly smaller tumors (22.0 mm vs 27.0 mm, p=0.004), a smaller proportion of patients with postoperative pathologic T stage (p=0.026), a smaller proportion of patients with lymphovascular invasion (20.0% vs 40.7%, p=0.022), and a larger proportion of patients with negative resection margins (74.3% vs 51.5%, p=0.049). After PSM, the group that received neoadjuvant therapy had a significantly longer progression-free survival than those in the group that underwent surgery first (29.6 months vs 15.1 months, p=0.002). Overall survival was not significantly different between the two groups after PSM analysis., Conclusions: We observed significantly better surgical outcomes and progression-free survival with the addition of neoadjuvant therapy to the management of resectable PDAC. However, despite PSM, there was still selection bias due to the use of different regimens between the groups receiving surgery first and neoadjuvant therapy. Large homogeneous samples are needed in the future prospective studies.

Purpose: To identify common and unique pre-treatment prognostic factors in patients with borderline resectable (BR) or locally advanced (LA) pancreatic ductal adenocarcinoma (PDAC), treated with chemotherapy (CTx) or concurrent chemoradiotherapy (CRT)., Methods: We enrolled 215 patients with BR/LA PDAC, who were treated with either CTx (n = 82) or CRT (n = 133) as a first-line treatment between 2013 and 2016. Clinical data and CT imaging findings for predicting overall survival (OS) and progression-free survival (PFS) were analyzed using Cox regression analysis., Results: Carbohydrate antigen (CA) 19-9 > 1000 U/mL (hazard ratio [HR] 1.91; p = 0.001) and non-homogeneous enhancement (HR 1.95; p < 0.001) were associated with shorter OS in all study populations. There was no significant difference in median OS (15.3 vs 16.8 months, p = 0.297) and PFS (10.0 vs 11.7 months, p = 0.321) between the CTx and CRT groups. Non-homogeneous enhancement (HR 2.04; p = 0.006) and presence of positive lymph node on CT (HR 2.38; p = 0.036) were associated with poor OS in the CTx group, while CA 19-9 > 1000 U/mL (HR 2.38; p = 0.001) and non-homogeneous enhancement (HR 1.73; p = 0.006) were independent predictors for poor OS in the CRT group., Conclusion: Enhancement pattern on CT was a common prognostic factor for patients with PDAC treated with either CTx or CRT. Presence of positive lymph nodes on CT was a poor prognostic factor for the CTx group only, whereas CA 19-9 > 1000 U/mL was a poor prognostic factor for the CRT group only.

Background: Venous invasion is not included in the pancreatic ductal adenocarcinoma (PDAC) staging, and its correlation with prognosis remains unclear. We evaluated the prognostic impact of radiographic portal/superior mesenteric vein (PV/SMV) invasion, and its possibility of complementing T staging., Methods: We identified patients with non-metastatic PDAC using our institutional cohort, and divided them according to PV/SMV invasion at imaging, defined as >180-degree tumor-vessel interface or contour deformity. We conducted Cox proportional hazard regression, and compared survival in the original and 1:1 propensity score matched datasets., Results: We identified 454 patients [PV/SMV(+): 172; PV/SMV(-): 282]. In the multivariate analysis, PV/SMV invasion, age (≥70 years), performance status, tumor size (2-4, >4 cm), lymph nodes >4, and arterial invasion was correlated with prognosis. The PV/SMV(+) group had a shorter overall survival (OS) than the PV/SMV(-) group in the original (14.4 vs. 20.9 months; P < 0.001) and matched datasets (14.3 vs. 17.2 months; P = 0.009). Among patients without arterial invasion (cT1-cT3), the PV/SMV(+) group had a shorter OS (15.9 vs. 21.2 months; P = 0.002). Moreover, their OS did not differ from that of patients with arterial invasion (cT4) (15.9 vs. 14.4 months; P = 0.907). Patients with vessel (artery/vein) invasion had a shorter OS than those without vessel invasion (14.5 vs. 21.2 months; P < 0.001)., Conclusions: Radiographic PV/SMV invasion in non-metastatic PDAC was correlated with a poor prognosis. It could identify a group with shorter OS among patients without arterial invasion (cT1-cT3). It is suggested that inclusion of PV/SMV invasion in clinical T4 criteria should be considered., Competing Interests: Declaration of competing interest All authors have no conflicts of interest to disclose., (Copyright © 2020. Published by Elsevier B.V.)

Simple Summary: The 8th edition of the AJCC staging system for pancreatic cancer has been validated for pathological staging; however, its suitability for clinical staging is still uncertain. Clinical staging is important for pancreatic cancer because surgical resection is not suitable for most patients. We validated the prognostic performance and suitability of the current staging system for clinical staging. In our analysis using a prospectively collected pancreatic cancer cohort database, survival difference was not shown between stage IB and IIA, which were divided by the new size-based T3 criterion. Among the patients who received surgery, the pathological stage was more advanced than the clinical stage in 57.3%, mostly due to a false-negative lymph node in clinical staging. These findings suggest that the 8th edition AJCC staging system is not validated for clinical staging and separate criteria more suitable for clinical staging should be established. The 8th edition of the American Joint Committee on Cancer (AJCC) staging system for pancreatic cancer (PC) has been validated for pathological staging; however, its significance for clinical staging remains uncertain. We validated the prognostic performance and suitability of the current staging system for the clinical staging of PC. We identified 1043 patients from our PC registry who were staged by imaging according to the 8th edition staging system and conducted analysis, including overall survival (OS) comparison. Gradual prognostic stratification according to stage hierarchy yielded significant OS differences between stage groups, except between stage I and II (p = 0.193). A substage comparison revealed no survival differences between IB (T2N0) and IIA (T3N0), which were divided by the T3 criterion only (p = 0.278). A higher N stage had significantly shorter OS than a lower N stage (all pairwise p < 0.05). However, among the 150 patients who received upfront surgery, the pathological stage was more advanced than the clinical stage in 86 (57.3%), mostly due to a false-negative cN0 (70.9%). Our results suggest that the new definition of T3 and the number-based N criteria in the 8th edition AJCC staging system may be not adequate for clinical staging. Establishing separate criteria more suitable for clinical staging should be considered. [ABSTRACT FROM AUTHOR]